Datasets:

afg1

/

litscan-epmc-subset

like 0

PMC10002166

Molin Li,Weimin Gong,Jie Chen,Yining Zhang,Yufei Ma,Xiaolin Tu

PPP3R1 Promotes MSCs Senescence by Inducing Plasma Membrane Depolarization and Increasing Ca2+ Influx

23-02-2023

BMSC,aging,PPP3R1,Ca2+,membrane potential

Aging of mesenchymal stem cells(MSCs) has been widely reported to be strongly associated with aging-related diseases, including osteoporosis (OP). In particular, the beneficial functions of mesenchymal stem cells decline with age, limiting their therapeutic efficacy in age-related bone loss diseases. Therefore, how to improve mesenchymal stem cell aging to treat age-related bone loss is the current research focus. However, the underlying mechanism remains unclear. In this study, protein phosphatase 3, regulatory subunit B, alpha isoform, calcineurin B, type I (PPP3R1) was found to accelerate the senescence of mesenchymal stem cells, resulting in reduced osteogenic differentiation and enhanced adipogenic differentiation in vitro. Mechanistically, PPP3R1 induces changes in membrane potential to promote cellular senescence by polarizing to depolarizing, increasing Ca2+ influx and activating downstream NFAT/ATF3/p53 signaling. In conclusion, the results identify a novel pathway of mesenchymal stem cell aging that may lead to novel therapeutic approaches for age-related bone loss.

PPP3R1 Promotes MSCs Senescence by Inducing Plasma Membrane Depolarization and Increasing Ca2+ Influx Aging of mesenchymal stem cells(MSCs) has been widely reported to be strongly associated with aging-related diseases, including osteoporosis (OP). In particular, the beneficial functions of mesenchymal stem cells decline with age, limiting their therapeutic efficacy in age-related bone loss diseases. Therefore, how to improve mesenchymal stem cell aging to treat age-related bone loss is the current research focus. However, the underlying mechanism remains unclear. In this study, protein phosphatase 3, regulatory subunit B, alpha isoform, calcineurin B, type I (PPP3R1) was found to accelerate the senescence of mesenchymal stem cells, resulting in reduced osteogenic differentiation and enhanced adipogenic differentiation in vitro. Mechanistically, PPP3R1 induces changes in membrane potential to promote cellular senescence by polarizing to depolarizing, increasing Ca2+ influx and activating downstream NFAT/ATF3/p53 signaling. In conclusion, the results identify a novel pathway of mesenchymal stem cell aging that may lead to novel therapeutic approaches for age-related bone loss. Age-related bone loss, or osteoporosis, is prevalent in animals and humans, and maintenance of skeletal homeostasis throughout life depends on the process of skeletal remodeling, which continuously replaces aged and damaged bones with new bones to maintain skeletal strength and elasticity. Recent studies suggested that senescent cells play a causal role in bone remodeling and the formation–resorption transition in age-related bone loss [1]. Multiple cell types in the skeletal microenvironment senesce with age [2]. Bone marrow mesenchymal stem cells (BMSCs) are multipotent progenitor cells with regenerative potential in various tissues [3]. Bone marrow stromal cell function dramatically declines with age [3,4]. Cellular senescence leads to a stable arrest of cell proliferation [5] and may affect aging BMSC function through intrinsic and extrinsic mechanisms. Aging is a complex, gradual, and inevitable physiological process, accompanied by the accumulation of damaged macromolecules, leading to organ dysfunction and the senescence of MSCs, with the character of significantly impaired biological properties of MSCs [6,7,8]. Several methods have been tried to modify the senescence of MSCs. For example, BMP2 as well as TGFβ signaling play an important role in the aging process of MSCs [9,10], but the mechanism of MSCs aging remains to be investigated. Calcium homeostasis is undoubtedly an important mechanism of action in various types of studies on cellular and organ aging [11,12,13,14,15]. Calcineurin is a Ca2+-dependent phosphatase that is expressed in many cell types, including osteoblasts. It dephosphorylates nuclear factors of activated T cell (NFAT) proteins, which are first involved in T cell activation by controlling interleukin-2 expression [16]. However, calcineurin–NFAT signaling is also effective in osteoblasts [16]. Calcineurin inhibitors such as cyclosporine A (CsA) or tacrolimus (FK-506) are widely used as immunosuppressive drugs to treat patients after organ transplantation or with severe autoimmune diseases [17,18]. Cells have a transmembrane potential, which is maintained by the balance between ions on both sides of the plasma membrane. Although the underlying mechanism is unknown, alterations in membrane potential have been found to be associated with senescence in fibroblasts and epithelial cells [19]. Since membrane potential as well as calcineurin have been reported to be associated with cellular senescence, we hypothesized that membrane depolarization may play a role in Ca2+-dependent calcineurin PPP3R1 (protein phosphatase 3, regulatory subunit B, α isoform; also known as CNB1) -regulated stromal stem cell senescence and further revealed specific mechanisms. The discovery of this mechanism will provide new clues for a novel electrophysiological pathway to control aging and bone aging in stromal stem cells. By GEO database analysis, PPP3R1 mRNA expression was significantly higher in samples from four older age groups in nine sets of samples from dataset GSE35955 (Figure 1A). Volcano plots showed the number of differentially expressed genes identified from each dataset. Similarly, PPP3R1 mRNA expression was found to be significantly higher in samples from the older age group (Figure 1B). Expression of PPP3R1 was elevated ≥2.7 in the older group compared with the younger group (Figure 1C). In the meantime, to determine the role of PPP3R1 in BMSC cell senescence, we examined the levels of PPP3R1 in BMSCs from mice of different ages and found that the expression of PPP3R1 in the BMSCs increased step by step with mouse age (Figure 1D). To investigate the expression and role of PPP3R1 in the senescence of MSCs in vitro, C3H10T1/2 cells were treated with tert-butyl hydroperoxide (t-BHP) to obtain senescent MSC cell lines [20]. As expected, tert-butyl hydroperoxide treatment increased SA-β-Gal activity in C3H10T1/2 cells (Figure 2A), resulted in proliferation arrest (Figure 2B), increased cellular reactive oxygen species production (Figure 2C), and induced increased expression of the senescence markers p16 and p21 (Figure 2D). In parallel, PPP3R1 mRNA expression in cells was detected, and tert-butyl hydroperoxide treatment was found to increase PPP3R1 expression in C3H10T1/2 cells (Figure 2E). Overall, we successfully generated senescent stromal stem cell lines using the tert-butyl hydroperoxide treatment of C3H10T1/2 cells and found increased PPP3R1 expression. To better characterize the role of PPP3R1 in stromal stem cell senescence, we generated stromal stem cell lines in which PPP3R1 was activated as well as inhibited by treating C3H10T1/2 cells with PPP3R1 recombinant protein and with FK506. Markers of senescence in stromal stem cells were mainly proliferation arrest, reduced osteogenic differentiation ability, and enhanced adipogenic differentiation ability. We also found that C3H10T1/2 cells had reduced osteogenic differentiation ability (Figure 3A–C), substantially enhanced adipogenic differentiation ability (Figure 3D,E), and showed proliferation arrest (Figure 3F) when PPP3R1 was activated. At the same time, C3H10T1/2 cells showed enhanced activity of SA-β-Gal (Figure 3G), increased cellular reactive oxygen species production (Figure 3H), and greatly enhanced expression of the senescence marker P16P21 (Figure 3I). However, the above results were reversed after PPP3R1 was inhibited using FK506 (Figure 3A–I). On the basis of these results, we conclude that PPP3R1 inhibition enables mesenchymal stem cells to escape senescence. Then, we tried to preliminarily clarify the mechanisms responsible for the regulation of MSC senescence by PPP3R1. Recently, membrane depolarization has been reported to be associated with cell senescence [19,21]. We therefore investigated whether changes in plasma membrane potential might play a role in PPP3R1-regulated senescence in C3H10T1/2 cells. To measure their relative plasma membrane potential, the fluorescent dye DiBAC4 was used to incubate senescent C3H10T1/2 cells with increased uptake in cells with depolarized plasma membranes (decreased uptake in hyperpolarized cells). As mentioned above, C3H10T1/2 cells presenting senescence showed higher DiBAC4 fluorescence intensity, whereas control cells showed lower plasma membrane depolarization (Figure 4A). Meanwhile, PPP3R1-activated C3H10T1/2 cells showed high DiBAC4 fluorescence intensity, and C3H10T1/2 cells in PPP3R1 inhibited by FK506 showed low DiBAC4 fluorescence intensity (Figure 4B). We further treated FK506-treated C3H10T1/2 cells with potassium chloride, a well-established plasma membrane depolarizer. Forced depolarization in C3H10T1/2 cells abrogated the favorable effect of PPP3R1 inactivation on cellular senescence (Figure 4C,D). These results suggest that PPP3R1 regulates senescence in stromal stem cells by controlling plasma membrane potential. Finally, we investigated the mechanism by which plasma membrane depolarization promotes senescence in C3H10T1/2 cells. Depolarization of the plasma membrane has been shown to activate voltage-gated calcium channels and increase intracellular calcium [22], whereas this increased calcium promotes cellular senescence [23,24]. Our study found decreased calcium ion levels in senescent C3H10T1/2 cells but increased depolarization induced by potassium chloride (Figure 5A). The increase in intracellular calcium induced cellular senescence by triggering NFAT dephosphorylation and translocating it to the nucleus [25]. NFATc1 inhibits the expression of ATF3, which in turn increases the expression of P53 and other aging-related markers [26]. P53 induces transcription of the Cyclin-dependent kinase inhibitor p21. In turn, p21 inhibits the activity of CDK4/6, which leads to a decrease in Cyclin-D1 levels, a decrease in the degree of Rb phosphorylation, and cell cycle exit [27]. Consistent with these findings, we observed increased P53 expression in senescent C3H10T1/2 cells (Figure 5B), and nuclear translocation of NFATc1 was promoted (Figure 5C), while ATF3 expression was decreased in PPP3R1-activated C3H10T1/2 cells, which further increased P53 expression in cells and decreased Cyclin-D1 levels and Rb phosphorylation in cells (Figure 5D). Meanwhile, after interfering with P53 expression, senescent C3H10T1/2 cells showed increased Cyclin-D1 levels and Rb phosphorylation levels (Figure 5E), and the senescent phenotype was alleviated (Figure 5F,G). These results suggest that plasma membrane depolarization leads to stromal stem cell senescence by increasing Ca2+ influx and activating the NFAT/ATF3/p53 signaling pathway. Mesenchymal stem cells are characterized by their ability to proliferate and maintain an undifferentiated state as well as their ability to differentiate into multiple cell lineages. Bone marrow mesenchymal stem cells (MSCs) are generally considered to be the best source of MSCs. However, the number and differentiation potential of mesenchymal stem cells showed an age-related decline. Understanding the mechanisms of pre-aging of stromal stem cells is critical for developing therapeutic interventions for age-related bone loss. In this study, PPP3R1 was identified as a regulator of senescence in stromal stem cells using tert-butyl hydroperoxide co-treatment with C3H10T1/2 cells, and the further results showed that PPP3R1 promotes plasma membrane depolarization in senescent stromal cells, which promotes cellular senescence by increasing Ca2+ influx and activating downstream NFAT/ATF3/p53 signaling. Therefore, PPP3R1 might be a novel regulator during aging in stromal stem cells. Calcineurin has undoubtedly been an important research direction in previous studies on aging [26,28,29,30]. Calcineurin (protein phosphatase 3, PPP3) is a widely expressed calcium-sensitive serine-threonine phosphatase consisting of the 61 kD calmodulin-binding catalytic subunit A (gene name, PPP3C) and the 19 kD Ca2+-binding regulatory subunit B (gene name, PPP3R) [31]. Three genes encode catalytic subunit A: PPP3CA (isoform a), PPP3CB (isoform b), or PPP3CC (isoform g), and PPP3CA and PPP3C show partially overlapping expression patterns and functions [32]. For regulatory subunit B, two genes (PPP3R1 and PPP3R2) have been described, but PPP3R2 expression appears to be confined to the testis [33]. Calcineurin is activated by intracellular calcium influx and represents a critical signaling node that transmits environmental stimuli into adaptive responses in multiple tissues and organs. In this study, we found that there was a large difference in PPP3R1 expression in mouse BMSCs at different ages, and further studies revealed that activation of PPP3R1 accelerated the aging of stromal stem cells and altered the differentiation ability of stromal stem cells into bone and adipocyte lineages, while the opposite was true when stromal stem cells were treated with FK506, a calcineurin inhibitor. Cells have a transmembrane potential, which is maintained by a balance between ions on either side of the plasma membrane. Although studies of membrane potential have focused on excited cells, recent studies have shown that dynamic membrane potential is also present in most non-excitatory cells, but its mechanism of action is unknown [34]. Changes in plasma membrane potential during cellular senescence were first reported by Lallet-Daher et al. [35]. Since then, senescence in epithelial cells [19] and fibroblasts [21] has also been attributed to changes in membrane potential in several studies, although the specific mechanisms are unknown. In the present study, we show that plasma membrane depolarization induces stromal stem cell senescence by increasing Ca2+ influx and activating downstream NFAT/ATF3/p53 signaling. Voltage-gated calcium (Ca2+) channels (VGCCs) are critical sensors for the conversion of membrane potential changes into intracellular Ca2+ transients [36]. In parallel, Cav1.2 has been found to be constitutively expressed by preosteoblasts and may mediate Ca2+ influx in response to depolarization [37]. In addition, Fei et al. found that Cav1.2 itself promotes osteogenesis of bone marrow-derived mesenchymal stem cells, and upregulation of Cav1.2 expression alleviates osteoporosis in prematurely aging mice [38]. This suggests that a similar mechanism can also exist in our study, which requires further study and confirmation at a later stage. In conclusion, we reveal the mechanism by which PPP3R1/plasma membrane depolarization induces senescence in stromal stem cells, including PPP3R1-promoting plasma membrane depolarization which produces membrane potential changes as well as calcium influx, plasma membrane depolarization, and calcium involvement in the regulation of the NFAT/ATF3/P53 pathway. Thus, this work provides novel insights into the involvement of ion channels and plasma membrane potential in controlling stromal stem cell senescence. Pharmaceutical studies of related pathways and drugs may lead to new potential treatments for the use of stem cells during age-related bone loss. All microarray data were downloaded from the GEO database (http://www.ncbi.nih.gov/geo;GSE35955; accessed on 12 April 2022). The raw data were downloaded as MINiML files which contain the data for all platforms, samples, and GSE records of the GSE. The extracted data were normalized by log2 transformation. The microarray data were normalized by the normalize quantiles function of the preprocessCore package in R software (version 3.4.1). Probes were converted to gene symbols according to the annotation information of the normalized data in the platform. Probes matching multiple genes were removed from these datasets. The average expression value of gene measured by multiple probes was calculated as the final expression value. In cases of the same dataset and platform but in different batches, we used the removeBatchEffect function of the limma package in the R software to remove batch effects. In cases of different datasets or in the same dataset but in different platforms, extracting multiple datasets with common gene symbols, and marking different datasets or different platforms as different batches, we used the removeBatchEffect function of the limma package in the R software to remove batch effects. The result of the data preprocessing was assessmented by boxplot. Mouse embryo-derived mesenchymal stem cells (C3H10T1/2) were purchased from the American Type Culture Collection, while we also isolated primary BMSCs from the long bones of mice [39]. Cells were maintained in α-MEM medium (αMEM, Corning, New York, NY, USA) containing 10% FBS (Gibco, New York, NY, USA), 100 U mL−1 penicillin/streptomycin, and 5% CO2 at 37 °C. We used tert-butyl hydroperoxide (t-BHP) (Sigma-Aldrich, Gaithersburg, MD, USA, 100 µmol·L−1) treatment to induce cellular senescence, and C3H10T1/2 cells were continuously cultured with tert-butyl hydroperoxide for 6 h. To induce osteogenic differentiation, the C3H10T1/2 cells were seeded at a density of 1.5 × 105 cells/well (six-well plates) and incubated with osteoblast differentiation medium containing β-glycerophosphate (Sigma-Aldrich, 5 mmol/L) and ascorbic acid (Sigma-Aldrich, 50 g/mL). To alter PPP3R1 expression, the C3H10T1/2 cells were cultured with PPP3R1 murine recombinant protein (CUSABIO, Houston, TX, USA, CSB-EP737231MO) as well as FK506 (MedChemExpress, Monmouth Junction, NJ, USA, HY-13756). For SA-β-galactosidase staining, cells were washed with PBS, fixed with 4% paraformaldehyde in PBS at room temperature for 20 min and incubated with reagents from a senescence-associated β-galactosidase staining kit (Beyotime Institute of Biotechnology, Shanghai, China, #C0602) according to the manufacturer’s suggestions. For alkaline phosphatase staining: Treated C3H10T1/2 cells were washed with a phosphate buffer, fixed in 4% paraformaldehyde for 30 min at room temperature, and stained with an alkaline phosphatase staining kit (Beyotime Institute of Biotech, Shanghai, China) for 1 h at room temperature in the dark. For alizarin red staining, cells were fixed with paraformaldehyde for 30 min, incubated with 1% alizarin red for 30 min at room temperature and washed with PBS to remove the excess dye. Protein samples were extracted using a cell lysis buffer (P0013, Beyotime, Shanghai, China) supplemented with proteinase (04693159001, Roche, Basel, Switzerland) and phosphatase inhibitors (4906837001, Roche, Basel, Switzerland). Then, a 1/5 volume of loading buffer was added into the cell lysis at 100 °C for 10 min. Protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and then transferred onto polyvinylidene fluoride membranes. Blots were blocked in 5% bovine serum albumin for 2 h at 25 °C, and then the membrane was incubated with specific antibodies against NFATc1 (Cell Signaling Technology, Boston, MA, USA, 1:1000, #8032), p-Rb (Bioss, Beijing, China, 1:1000, bsm-52197R), ATF3 (Abbkine, Beijing, China, 1:1000, #Abp55330) and Cyclin-D1 (Bioss, Beijing, China, 1:1000, bs-0623R) at 4 ℃ overnight. Blots were incubated with a secondary antibody conjugated to horseradish peroxidase (diluted 1:5000) for 2 h at 25 °C. Finally, each membrane was exposed to ECL (SQ202, Epizyme, Shanghai, China). The total RNA of cells was extracted using the TRIzol reagent (15596026, Invitrogen, Shanghai, China). Following evaluation of the RNA concentration, the DNA group was removed and cDNA was obtained through reverse transcription using a reverse transcription kit (RR047A, Takara, Beijing, China) according to the manufacturer’s instructions. The IQ SYBR Green Supermix (RR820, Takara, Beijing, China) was used to conduct 3 biological duplications of real-time quantitative PCR on an iCycler real-time detection system (CFX Connect Optics Module, Bio-Rad, Hercules, CA, USA). The corresponding gene expression level was normalized to that of GAPDH from the same samples. Table 1 provides the primer sequences. To assess the cell cycle, we seeded 3 × 105 treated cells into six-well plates and incubated them at 37 °C for 48 h. For the cell cycle analysis, cells were digested with trypsin (Hyclone, Logan, UT, USA), washed twice with phosphate-buffered saline (PBS), and fixed in 70% ethanol at 4 °C overnight. The cells were centrifuged at 500× g for 5 min, washed twice with cold PBS, and centrifuged. Cell cycle analysis was performed using fluorescence-activated cell sorting after the digested cells were treated with RNase A (0.1 mg/mL) and stained with propidium iodide (0.05 mg/mL; 4A Biotech, Beijing, China) for 30 min at 37 °C. Cultured cells were washed with a HEPES buffer (pH 7.4; 20 mM HEPES, 120 mM NaCl, 2 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 5 mM glucose) and incubated with 5 μmol·L−1 membrane voltage-reporter dye DiBAC4 (Molecular Probes) for 1 h at 37 °C. Then, the cells were observed and photographed with an inverted fluorescence microscope (Olympus, Tokyo, Japan). Fluorescence data were analyzed with Image-J software. Cytosolic calcium levels were detected with a Fura-2/AM (Invitrogen™, Shanghai, China, # F1221) according to the manufacturer’s suggested procedure. C3H10T1/2 cells were loaded with 5 μmol·L−1 Fura-2/AM in Hanks’ balanced salt solution (HBSS) for 1 h at 37 °C. After the cells were washed extensively with HBSS, cytosolic Ca2+ was measured with a calcium imaging system built on an inverted fluorescence microscope (Olympus IX51, Tokyo, Japan). Fluorescence images (filtered at 515 nm ± 25 nm) were captured with a CCD camera (CoolSNAP fx-M) and analyzed in MetaFluor software. Ca2+ levels are shown as the ratio of fluorescence intensity at 340 nm/fluorescence intensity at 380 nm (F340/F380). At least three independent experiments were performed for each condition. All results are presented as the mean ± SD. Curve analysis was performed in Prism (GraphPad). The data in each group were analyzed with an unpaired, two-tailed Student’s t-test. The significance threshold was set at p < 0.05. In summary, this study found that PPP3R1 increases Ca2+ influx by polarization depolarization, activates downstream NFAT/ATF3/p53 signaling, induces membrane potential changes, promotes cell senescence, and leads to decreased osteogenic differentiation and enhanced adipogenic differentiation. Taken together, the results identify a novel mesenchymal stem cell senescence pathway that may lead to new treatments for age-related bone disease.

PMC10002170

Mina Ding,Eunjin Cho,Zhihao Chen,Sang-Wook Park,Tae-Hoon Lee

(S)-2-(Cyclobutylamino)-N-(3-(3,4-dihydroisoquinolin-2(1H)-yl)-2-hydroxypropyl)isonicotinamide Attenuates RANKL-Induced Osteoclast Differentiation by Inhibiting NF-κB Nuclear Translocation

21-02-2023

osteoporosis,osteoclast differentiation,PRMT5,NF-κB,methylation

Osteoporosis is a common skeletal disease; however, effective pharmacological treatments still need to be discovered. This study aimed to identify new drug candidates for the treatment of osteoporosis. Here, we investigated the effect of EPZ compounds, protein arginine methyltransferase 5 (PRMT5) inhibitors, on RANKL-induced osteoclast differentiation via molecular mechanisms by in vitro experiments. EPZ015866 attenuated RANKL-induced osteoclast differentiation, and its inhibitory effect was more significant than EPZ015666. EPZ015866 suppressed the F-actin ring formation and bone resorption during osteoclastogenesis. In addition, EPZ015866 significantly decreased the protein expression of Cathepsin K, NFATc1, and PU.1 compared with the EPZ015666 group. Both EPZ compounds inhibited the nuclear translocation of NF-κB by inhibiting the dimethylation of the p65 subunit, which eventually prevented osteoclast differentiation and bone resorption. Hence, EPZ015866 may be a potential drug candidate for the treatment of osteoporosis.

(S)-2-(Cyclobutylamino)-N-(3-(3,4-dihydroisoquinolin-2(1H)-yl)-2-hydroxypropyl)isonicotinamide Attenuates RANKL-Induced Osteoclast Differentiation by Inhibiting NF-κB Nuclear Translocation Osteoporosis is a common skeletal disease; however, effective pharmacological treatments still need to be discovered. This study aimed to identify new drug candidates for the treatment of osteoporosis. Here, we investigated the effect of EPZ compounds, protein arginine methyltransferase 5 (PRMT5) inhibitors, on RANKL-induced osteoclast differentiation via molecular mechanisms by in vitro experiments. EPZ015866 attenuated RANKL-induced osteoclast differentiation, and its inhibitory effect was more significant than EPZ015666. EPZ015866 suppressed the F-actin ring formation and bone resorption during osteoclastogenesis. In addition, EPZ015866 significantly decreased the protein expression of Cathepsin K, NFATc1, and PU.1 compared with the EPZ015666 group. Both EPZ compounds inhibited the nuclear translocation of NF-κB by inhibiting the dimethylation of the p65 subunit, which eventually prevented osteoclast differentiation and bone resorption. Hence, EPZ015866 may be a potential drug candidate for the treatment of osteoporosis. Osteoporosis is a common skeletal disease that occurs when the bone mineral density (BMD) decreases or the bone structure deteriorates [1]. The development of osteoporosis is primarily caused by the imbalance of bone homeostasis, including bone resorption by osteoclasts and bone formation by osteoblasts [2]. Post-menopause, old age, medications, endocrine disorders, immobilization, inflammatory arthropathy, hematopoietic disorders, and nutrition disorders increase osteoclast activity, leading to osteoporosis [3]. Given that the population’s average age is rapidly increasing worldwide, osteoporosis could become a major health concern that significantly impacts the quality of life of older adults [4]. Currently, the preferred treatment for osteoporosis is pharmacological interventions. Bisphosphonates, denosumab, and strontium ranelate are the main medicines for the treatment of osteoporosis [5]. However, some studies have shown a rapid decrease in BMD and an increased risk of vertebral fractures after the discontinuation of denosumab [6,7]. Additionally, there have been some reported side effects of bisphosphonates in the treatment of osteoporosis [8]. Therefore, this study aimed to find a new pharmacological agent to treat osteoporosis. Osteoclasts are giant multinucleated cells that have critical roles in the regulation of bone development and bone homeostasis [9]. Osteoclasts are derived from cells of the monocyte/macrophage lineage by the activation of receptors by two factors, the macrophage-colony stimulating factor (M-CSF) and the receptor activator of nuclear factor-kappa B ligand (RANKL). M-CSF primarily regulates the proliferation and survival of osteoclast precursors and mature cells [10]. RANKL is the major osteoclast differentiation factor, and its interaction with RANK recruits tumor necrosis factor receptor-related factors and activates downstream signaling pathways, thereby inducing the nuclear factor of activated T cells 1 (NFATc1) [11,12]. NFATc1 has a major role in regulating several osteoclast-specific genes including matrix metallopeptidase 9 (Mmp9), Cathepsin K (Ctsk), and acid phosphatase 5, tartrate resistant (Acp5) [13,14]. Histone methylation is the modification of certain amino acids in histones, such as lysine, arginine, and histidine, by the addition of one to three methyl groups. Histone methylation is a dynamic process, and methyl groups can be added or removed by histone methyltransferases and histone demethylases [15,16]. These enzymes have been shown to be involved in tumorigenesis [17], angiogenesis [18], and the development of acute myeloid leukemia (AML) [19]. Studies have shown that histone methylation is regulated in bone cell differentiation [20]. The protein arginine N-methyltransferase (PRMT) family is a group of methyltransferases. There are two types of PRMTs: PRMT1, 3, 4, 6, and 8 are type I PRMTs that asymmetrically demethylate arginine, while PRMT5 and PRMT7 are type II PRMTs that symmetrically demethylate arginine [21,22]. PRMT5 is known to play important roles in gene transcriptional regulation and signal transduction [23,24]. Previous research has indicated that PRMT5 protein increases during osteoclastogenesis, and the reduction of PRMT5 via (S)-N-(3-(3,4-Dihydroisoquinolin-2(1H)-yl)-2-hydroxypropyl)-6-(oxetan-3-ylamino)pyrimidine-4-carboxamide (EPZ015666) that inhibits RANKL induced osteoclast differentiation [25]. (S)-2-(Cyclobutylamino)-N-(3-(3,4-dihydroisoquinolin-2(1H)-yl)-2-hydroxypropyl)isonicotinamide (EPZ015866) is another PRMT5 specific inhibitor, which blocks the enzyme activity of PRMT5 in the proliferation and cell cycle progression of human colorectal cancer cells [26]. This study investigates the underlying molecular mechanisms of EPZ015866 on osteoclast differentiation. Nuclear factor-κB (NF-κB) is a transcription factor that has an important role in the survival, formation, and functions of osteoclasts [27,28]. The inhibition of NF-κB has been shown to be an efficient method to suppress osteoclast formation and bone resorption [29,30]. Therefore, many studies have focused on NF-κB as a target for the treatment of osteoporosis [28,31]. Studies have confirmed that methylation of lysine and arginine residues in the p65 subunit of NF-κB regulates its activity [32,33]. Additionally, there is evidence that PRMT5 can regulate NF-κB activity through the methylation of p65 [34,35]. The present study demonstrates that EPZ015866, a derivative of EPZ015666, has a better therapeutic effect on osteoporosis than EPZ015666. EPZ015666 is a known PRMT5 inhibitor that suppresses osteoclast differentiation [25]. Since there is a structural similarity (Figure 1A), we compared EPZ015866 with EPZ015666 in a dose-dependent manner to investigate the effect of EPZ015866 as an inhibitor of osteoclast formation. Bone marrow-derived macrophages (BMMs) isolated from the femur and tibia of a mouse were stimulated with RANKL and M-CSF in the absence or presence of EPZ015866 or EPZ015666 at the indicated concentrations for four days. EPZ015866 significantly reduced RANKL-induced tartrate-resistant acid phosphatase (TRAP) positive multinucleated giant cell formation at a low dose, 20 nM, whereas EZP015666 inhibited it at a high dose, 1000 nM (Figure 1B). When we calculated the area of TRAP-positive cells and the number of mature osteoclasts containing more than three nuclei, EPZ015866 dramatically decreased the area and number of osteoclasts at a concentration of 20 nM, the same concentration at which EZP015666 did not significantly work (Figure 1C,D). Therefore, we suggest that EPZ015866 is an effective compound for inhibiting osteoclastogenesis and is better than EPZ015666. Both EPZ compounds were not cytotoxic when the concentration of the compounds was equal to or less than 1000 nM (Figure 1E). These results suggest that the EPZ compounds suppress the RANKL-induced osteoclastogenesis without causing cytotoxicity. Bone remodeling is regulated by the homeostasis between osteoclasts and osteoblasts [36]. Therefore, we indicated whether the EPZ compounds affected osteoblast differentiation. The osteoblast differentiation was analyzed by alkaline phosphatase (ALP) staining after bone morphogenetic protein 2 (BMP2) stimulation. Neither EPZ compound affected BMP2-induced osteoblastogenesis compared with the control (Figure S1A,B). These results suggest that EPZ compounds only suppress osteoclast differentiation and have no effect on osteoblast formation. To determine whether the EPZ compounds inhibited F-actin ring formation, BMMs were treated with or without the EPZ compounds. F-actin ring formation was observed on day 4 after RANKL stimulation in the control group (Figure 2A). However, treatment with EPZ015866 remarkably reduced the size of F-actin ring structures, starting at 20 nM in a dose-dependent manner. Although the EPZ015666 treatment suppressed F-actin ring structures at high doses (500–1000 nM), it did not significantly inhibit them at low doses (Figure 2B). Additionally, we confirmed whether the EPZ compounds suppressed the bone-resorbing activity of osteoclasts. In the control group, bone resorption pits were detected after RANKL stimulation. However, bone resorption pits were decreased by the EPZ015866 treatment starting at 20 nM (Figure 2C). The area of the bone resorption pits was quantified according to the bone resorption assay results (Figure 2D). In the EPZ015666 treatment group, the inhibitory effect of the resorption pits was weaker than EPZ015866. These data indicated that EPZ015866 suppressed the formation of mature osteoclasts and the bone resorption ability better than EPZ015666. To investigate the effect of the EPZ compounds on osteoclastogenesis-associated gene expression, mRNA expression levels were examined by RT-PCR. We found that the mRNA expression of osteoclast-specific genes, including Acp5, Ctsk, Dendritic cell-specific transmembrane protein (Dc-stamp), Osteoclast stimulatory transmembrane protein (Oc-stamp), Atp6v0d2, and Mmp9 were suppressed in a dose-dependent manner by EPZ015866 (Figure 3A–F). However, EPZ015666 only significantly inhibited Acp5 and Atp6v0d2 expression at 1000 nM. These data reveal that EPZ015866 prevents osteoclast differentiation via inhibiting the expression of osteoclast-mediated genes better than EPZ015666 in vitro. To demonstrate the molecular mechanisms by which the EPZ compounds may regulate osteoclastogenesis, we examined the expression levels of osteoclast-associated proteins. The protein levels of NFATc1, PU.1, and Ctsk were suppressed in a dose-dependent manner by EPZ015866 (Figure 4A,B). However, NFATc1 expression levels and PU.1 levels were decreased only at 1000 nM EPZ015666. The NF-κB signaling pathway plays a key role in osteoclast differentiation [37]. NF-κB expression levels were not significantly altered by EPZ compound treatment. However, p-NF-κB levels were slightly maintained by EPZ015866 until day 4, although its expression was decreased in the control and by EZP015666 (Figure 4C,D). As a regulator of NF-κB, p-IκBα and IκBα expression levels were not altered between the control and the EPZ compounds (Figure 4C,D). These data indicate that EPZ015866 suppresses osteoclast differentiation by reducing the transcription factors, PU.1 or NFATc1, and altering the expression of p-NF-κB. In addition, the NF-κB gene level in osteoclast was checked after treatment with the EPZ compounds by RT-PCR. The result suggests that the NF-κB gene is not regulated by the EPZ compounds (Figure S2A). Luciferase assay was performed on the NF-κB luciferase activity in the NF-κB reporter HEK293 cell line after treatment with the EPZ compounds. The result showed the NF-κB transcriptional activity was not altered by the EPZ compounds (Figure S2B). These results indicate that the NF-κB transcriptional is not regulated by the EPZ compounds. To confirm whether the expression and activity of PRMT5 were regulated by the EPZ compounds during RANKL-induced osteoclastogenesis, BMMs were treated with various concentrations of the EPZ compounds (Figure 5). Because PRMT5 regulates the dimethylation of arginine symmetrically [23], we examined the symmetric demethylation of histone H3 arginine 8 (H3R8me2s) and the symmetric demethylation of histone H4 arginine 3 (H4R3me2s) levels to determine the activity of PRMT5 (Figure 5A,B). PRMT5 expression was suppressed in a dose-dependent manner in the cells treated with EPZ015866 or EPZ015666 compared with the control group. Furthermore, the methylation level of H4R3me2s was significantly increased when stimulated with RANKL in the control, but its expression was significantly inhibited in EPZ015866- or EPZ015666-treated groups. However, the methylation level of H3R8me2s was only suppressed by EPZ015866. Interestingly, the Prmt5 mRNA levels were not altered during osteoclastogenesis by the EPZ compounds (Figure 5C). These data indicate that the EPZ compounds suppress PRMT5 expression and suggest their activity in RANKL-induced osteoclast differentiation. To demonstrate PRMT5 activation of NF-κB via the demethylation of the p65 subunit of NF-κB [32], we examined the expression levels of symmetric dimethylarginine (SDMA) in Raw 264.7 cells transfected with a pcDNA3-HA-p65 plasmid vector (Figure 6A). p-p65 expression was suppressed by the EPZ compounds compared to the control. Although non-transfected Raw 264.7 cells expressed p-p65 in response to RANKL stimulation, the expression levels were lower than in the transfected cells. Interestingly, SDMA levels were slightly reduced by the EPZ compounds even though we detected whole SDMA levels. To examine whether the EPZ compounds regulate the activity of p65 by blocking SDMA, immunoprecipitation (IP) was performed with anti-HA−Agarose antibody and then analyzed with anti-SDMA antibodies (Figure 6B). The data show that the SDMA of p65 had a high expression in the control group, which was reduced after EPZ compound treatment. To further investigate the role of methylation in the NF-κB pathway, we examined the translocation of NF-κB. Interestingly, the EPZ compounds inhibited the p65 expression in the nucleus compared to the control. Immunofluorescence was performed to investigate the nuclear translocation of p65 (Figure 6C). p65 was detected in the nucleus by RANKL stimulation on day 2 in control; however, the EPZ compounds effectively inhibited the nuclear translocation of p65. The quantitative analysis results of Figure 6C were shown in Figure 6D. The number of p65 in the cell nucleus was significantly decreased in the EPZ compound treatment groups compared with the control. The nuclear and cytoplasmic fractionation were also separated after RANKL stimulation to determine the expression levels of p65 (Figure 6E). Although nuclear p65 levels were decreased by the EPZ compounds, p65 levels in the cytoplasm were not significantly altered. These results indicate that the inhibitory effect of EPZ015866 on osteoclast differentiation is mediated by reducing the nuclear translocation of NF-κB. These results also indicate that the EPZ compounds inhibit the nuclear translocation of NF-κB by blocking the dimethylation of p65. In the present study, we demonstrated that EPZ015866 and EPZ015666, known inhibitors of PRMT5, inhibit osteoclast differentiation as promising anti-osteoclastogenesis agents. The inhibition of osteoclastogenesis by EPZ015666 has previously been studied [25], but our current study found a more effective compound, EPZ015866, that could be used at a lower concentration in osteoclastogenesis. Our results indicated that, in the same concentration, EPZ015866 had a more significant inhibitory effect on osteoclast differentiation than EPZ01566 (Figure 1). Based on the area of TRAP-positive cells, the half maximal inhibitory concentration (IC50) values of EPZ015866 and EPZ015666 were around 30 nM and 600 nM, respectively. EPZ015866 reduced RANKL-induced osteoclast differentiation significantly better than EPZ015666 in vitro. Mature osteoclasts firmly attach themselves to the bone surface using specialized actin rings through cytoskeletal reorganization and cell polarization, ultimately leading to bone resorption [9]. The F-actin ring indicates the fusion state of osteoclasts and is required for osteoclast formation and activation [38]. It has been reported that actin ring formation is a structural factor essential for bone resorption [39]. In our study, EPZ015866 and EPZ015666 inhibited the formation of actin rings in a dose-dependent manner, resulting in reduced formation of mature osteoclasts and marked inhibition of bone resorption. Osteoclast differentiation is controlled by the transcriptional activation or repression of target genes by transcription factors. NFATc1 and PU.1 play important roles as transcriptional activators in osteoclast differentiation [40,41]. NFATc1 is a major regulator of osteoclast differentiation [42]. NFATc1 regulates the transcription of osteoclast-specific markers, including Acp5, Atp6v0d2, and Ctsk, which are important for the activation of mature osteoclasts [43]. We observed that Acp5, Ctsk, Oc-stamp, Dc-stamp, Atp6v0d2, and Mmp9 mRNA expression levels were significantly inhibited in EPZ015866 treatment, although only Acp5 and Atp6v0d2 levels were reduced at high doses of EPZ015666 (Figure 3). EPZ015866 showed more effective suppression via inhibiting all osteoclast-associated genes at low doses. Some studies have shown that the expression of osteoclast-specific markers is regulated by the transcription factors, NFATc1, PU.1, and NF-κB [44]. Moreover, PU.1 has been reported as a transcriptional activator of NFATc1 involved in the expression of osteoclast-specific genes, including Ctsk, Acp5, and Itgb3 [45]. Our results demonstrated that EPZ015866 more effectively inhibited the protein expression of NFATc1 and PU.1 than EPZ015666, although their regulatory mechanism is unknown or indirect regulation by EPZ compounds. Moreover, there are p65 independent pathways [46], including the alternative NF-κB pathway and direct regulation of NF-κB subunits in osteoclastogenesis [47,48]. These combined results suggest that EPZ015866 suppresses osteoclastogenesis-related genes through the expression of transcription factors PU.1 and NFATc1. PRMT5 catalyzes the symmetric dimethylation of histone proteins to induce gene silencing by generating repressive histone marks, including H3R8me2s and H4R3me2s [23]. Thus, we checked the expression and activity of PRMT5 during RANKL-induced osteoclast differentiation. Our results suggest that EPZ015866 and EPZ01566 suppress the activation of PRMT5 during osteoclastogenesis. Although EZP015666 did not greatly inhibit H3R8me2s, it did inhibit PRMT5 protein levels. The NF-κB signaling pathway plays an important role in RANKL-stimulated osteoclast differentiation [37,49]. RANKL-mediated NF-κB activation is further transmitted by inducing the transcription factor NFATc1 in BMMs. Furthermore, during the regulation of NF-ĸB activation by the IκB kinase (IKK) complex, the NF-ĸB/Rel dimer proteins are themselves subject to complex regulation through a series of post-translational modification (PTM) events [50]. Numerous studies have confirmed that PTM on the p65 subunit of NF-κB includes methylation [34], acetylation [51], and ubiquitination [52]. Moreover, Levy. et al. confirmed that the methylation and phosphorylation of p65 are mutually regulated, forming a more complex NF-κB regulatory system [53]. In addition, previous research shows that NF-κB is activated by the dimethylation of arginine 30 of the p65 subunit [54]. PRMT5 regulates the methylation of the arginine residues of p65 [33]. Therefore, we investigated whether the EPZ compounds inhibited p65 methylation in osteoclasts. Interestingly, as shown in Figure 6, pcDNA3-HA-p65 transfected Raw 264.7 cells revealed decreased expression of whole SDMA levels via the EPZ compounds. Additionally, p65-specific SDMA levels were decreased in EPZ treatment, assessed by IP analysis, which is correlated with SDMA regulation by the EPZ compounds [55]. We observed whether a reduction in the SDMA of p65 regulates the subcellular localization of p65. The results of immunostaining and Western blot showed that RANKL stimulation enhanced the translocation of p65 into the nucleus, while the EPZ compounds attenuated NF-κB activation by interfering with the nuclear translocation of p65. Taken together, the EPZ compounds repress p65 nuclear translocation via the inhibition of p65 dimethylation, leading to the prevention of osteoclast formation. It has been studied that the PRMT5-mediated methylation of the p65 subunit of NF-κB at R30 is involved in the regulation of p65 activity [34]. Harris, D.P. et al. (2014) indicated that PRMT5 symmetrically methylates R30 and R35 of NF-κB/p65 in TNF-α-activated endothelial cells [35]. In different research, Harris, D.P. et al. (2016) also suggested that the PRMT5-mediated methylation of p65 at R174 is required for the induction of CXCL11 in TNF-α-activated endothelial cells [56]. Therefore, we speculate that the role of PRMT5 in the regulation of p65 activity might be by regulating the methylation of R174, R35, and R30 of the NF-κB/p65 subunit in osteoclast differentiation. We will continue to explore in more detail which methylation sites of p65 mediate the suppression of osteoclast differentiation by PRMT5 inhibitors in the future. Duncan et al. compared the inhibitors of PRMT5 based on their structure and medicinal chemistry optimization [57]. In their results, EPZ015866 shows lower PRMT5 IC50 and lymphoma cell line proliferation IC50 values than EPZ015666, suggesting a greater inhibitory effect on PRMT5. Therefore, EPZ015866 inhibits at lower concentration than EPZ015666 in osteoclast differentiation. However, both EPZ compounds inhibited the methyl status of p65 at a high dose (1000 nM), suggesting that the final inhibitory effect on the substrate may be the same, although further study is necessary to prove their pharmacological inhibitory effect on the substrate. In conclusion, the EPZ compounds significantly reduced RANKL-induced osteoclast differentiation, F-actin ring formation, and bone resorption. The EPZ compounds were identified as potent inhibitors of NF-κB activity and were shown to inhibit osteoclast differentiation through the inhibition of NF-κB nuclear translocation (Figure 7). Furthermore, we indicated that arginine methylation-mediated NF-κB activity has a critical role in RANKL-induced osteoclast differentiation. Therefore, these EPZ compounds may be suitable for drugs for treating bone diseases characterized by excessive osteoclast activity. The alpha modification of Eagle’s minimal essential medium (α-MEM) and fetal bovine serum (FBS) were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Recombinant mouse M-CSF and RANKL were procured from Peprotech (Cranbury, NJ, USA). A tartrate-resistant acid phosphatase staining kit was bought from CosmoBio (Tokyo, Japan). Characterized fetal bovine serum (chFBS) was purchased from Hyclone (Logan, UT, USA). Alpha MEM (αMEM, without ascorbic acid) was purchased from Welgene (Taipei, Taiwan). Recombinant human BMP2 was provided by Sino biological (Wayne, PA, USA) and dissolved in distilled water. Phalloidin was bought from Thermos Fisher Scientific (Waltham, MA, USA). 4′,6-diamidino-2-phenylindole (DAPI) stain was purchased from Sigma–Aldrich (St. Louis, MO, USA). Specific antibodies for NFATc1 (#8032s), Ctsk (#48353), PU.1 (#2266), NF-κB (#4764s), p-NF-κB (#3033), IκBα(#9242s), and p-IκBα (#2859s), and secondary antibodies were all purchased from Cell Signaling Technology (Beverly, MA, USA). PRMT5 (ab109451) was purchased from Abcam (Cambridge, UK), and H3R8me2s (A2374) and H4R3me2s (A3159) were purchased from ABclonal Technology (Woburn, MA, USA). EPZ015866 (PubChem CID: 117072552) and EPZ015666 (PubChem CID: 90241673) were purchased from Chemscene and Selleckchem, respectively. Mouse bone marrow cells were obtained from the femur and tibia of 10-week-old C57BL/6J mice, as described previously [58]. Briefly, red blood cells in bone marrow immune cells were lysed with Ammonium-Chloride-Potassium lysing buffer and then cultured in complete medium (α-MEM containing 10% FBS and 1% P/S) at the 37 °C in humidified air with 5% CO2 for 1 day. Non-adherent cells were harvested and cultured in Petri dishes for BMM selection with the complete medium in the presence of 30 ng/mL M-CSF. After three days, adhesion cells (BMMs) were harvested by Enzyme Free Cell Dissociation Solution Hank’s Based. The harvested cells were cultured further in the induction medium to induce the differentiation of osteoclasts. For the cell viability study, BMMs were cultured at a density of 1 × 104 cells per well in 96-well plates for 24 h. The cells were treated with M-CSF or M-CSF and RANKL (CTRL group) in the presence or absence of the indicated concentrations of EPZ compounds. After 48 h, the cell viability was assessed using an EZ-Cytox Kit. The experiment protocol was conducted following the manufacturer’s manual. Finally, the optical density was measured at 450 nm using a microplate reader (San Jose, CA, USA). For the TRAP staining assay, a TRAP staining kit was obtained from Takara Biotechnology (Shiga, Japan). This kit was used in accordance with the manufacturer’s instructions. BMMs were cultured in 96-well plates in complete α-MEM containing 30 ng/mL M-CSF. After 24 h, the cells were treated with various concentrations of the EPZ compounds that were changed every two days during the experiment period. Afterward, the culture medium was replaced, and cells were fixed in 4% PFA at room temperature for 20 min and then stained for TRAP. Micrographs of cells were captured by the microscope, and the area of TRAP+ multinucleated osteoclasts (≥3 nuclei) was quantified using the Image J software (1.8.0_112 version, National Institutes of Health, Bethesda, MD, USA). Primary calvarial cells were obtained from three-day-old mice by enzyme digestion. Briefly, the calvarias were cut into pieces and incubated in a digestion solution (0.1% type I collagenase with 0.2% Dispase II) at 37 °C for 40 min. After digestion, the calvarias were washed twice with a complete culture medium and cultured in a 10 cm dish at 37 °C in 5% CO2 for 3–4 days. Primary osteoblasts growing out of the bone chips were harvested and seeded into the 96-well plate for further experiments. For in vitro osteoblast differentiation, the cells were treated with or without the EPZ compounds in the presence of BMP2 (100 ng/mL) for seven days. Media were refreshed every two days. At the end of differentiation, cells were washed with PBS, fixed in 70% ice-cold ethanol for 30 min, and rinsed with distilled water two times. The cells were stained with BCIP®/NBT Liquid Substrate System for 20 min at room temperature. Images were captured with a microscope, and the intensity of ALP staining was quantified using the Image J software. BMMs were seeded in 12-well plates at a density of 1.5 × 105 per well. The EPZ compounds were added to the wells co-treated with RANKL for five days. After treatment, the cells were fixed with a 4% paraformaldehyde solution for 20 min at room temperature. The fixed cells were permeabilized with 0.1 % Triton-X 100 and blocked with 2 % BSA for 1 h. For immunofluorescence staining, the cells were incubated with a p65 primary antibody at 4 °C overnight. After the cells were washed with PBS, a goat anti-Mouse IgG (H+L) highly cross-adsorbed secondary antibody was incubated for 3 h at room temperature in the dark. After one day, the cells were washed with PBS. Finally, the cells were stained with 4′,6 diamidino2 phenylindole (DAPI). F-actin ring formation is a critical indicator of the bone resorption activity of osteoclasts and is a characteristic of mature cytoskeletal in osteoclasts [59]. Phalloidin staining was performed on osteoclasts treated with DMSO or EPZ compounds for four days, as described previously [60]. The cells were stained with FITC-conjugated Phalloidin for 45 min. After incubation with Phalloidin, the cells were washed with PBS. Finally, nuclei were visualized with DAPI. Images were captured by fluorescence microscopy. The effect of the EPZ compounds on bone resorption was assessed in accordance with the method of a previous study [61]. To explore the effect of the EPZ compounds on osteoclast-mediated bone resorption, BMMs were seeded into a 48-well bone resorption assay plate (2.5 × 104 cells/well). After 24 h, the cells were treated with DMSO or the EPZ compounds, along with M-CSF and RANKL, for five days further. After cell differentiation, the attached cells were treated with 5% sodium hypochlorite for 5 min. The plates were air-dried at room temperature, and resorption pits were captured using a microscope. The total resorption area was quantified by Image J software. BMMs were seeded on six-well plates (3 × 105/well) and cultured overnight. Then, the cells were stimulated with RANKL and treated with the indicated concentrations of the EPZ compounds. At the end of the differentiation, the cells were harvested using a plastic cell scraper and lysed with radioimmunoprecipitation assay (RIPA) buffer. The supernatant was collected following sonication and centrifugation. The concentration of proteins was detected by a BCA protein assay following the manufacturer’s protocol. The same amounts of protein (10 μg) were separated by polyacrylamide gel electrophoresis (PAGE) and transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories, Hercules, CA, USA). The membranes were blocked with 5% non-fat dry milk for 1 h at room temperature, and then incubated with primary antibodies at 4 °C overnight. The next day, the membranes were incubated for 2 h with secondary antibodies at room temperature, and signals were visualized by the enhanced chemiluminescence (ECL) western blotting detection reagent (Cytivalifescences, Marlborough, MA, USA). Total RNA was extracted from cells with TRIzol reagent (Qiagen Sciences, Valencia, CA, USA), and then reversely transcribed using a PrimeScript RT Reagent Kit following the manufacturer’s instructions (Takara Bio-technology, Shiga, Japan). The cycling conditions were 37 °C for 30 min, 85 °C for 15 s, and storage 4 °C. Quantitative PCR was performed using a QuantStudio 3 real-time PCR system (Applied Biosystems, Foster City, CA, USA) with a Power SYBR Green PCR Master Mix. The mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as the control gene. The primers employed for the amplification are presented in Table S1. The cells’ nuclear and cytoplasmic proteins were extracted using the NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Scientific, Waltham, MA, USA) ac-cording to the manufacturer’s instructions. In brief, the cells were seeded in 10 cm dishes (1.5 × 106 cells per dish). After one day, the cells were stimulated with RANKL and co-treated with EPZ015866 or EPZ015666 for 48 h. The cells were lysed with CER I and CER II buffer and centrifuged at 16,000× g for 5 min at 4 °C, and the supernatant (cytosolic protein) was stored at −80 °C. Nuclear pellets were re-suspended with NER buffer and vortexed to extract the nuclear protein, followed by incubation on ice for 40 min. The samples were then centrifuged at 16,000× g for 10 min at 4 °C. The supernatant (nuclear protein) was transferred to microtubes immediately after centrifugation. Finally, the nuclear and cytoplasmic proteins were analyzed by Western blot. Constructs were transfected into Raw264.7 cells using the NeonTM transfection system. The pcDNA3-HA-p65 plasmid DNA was a kind gift from Professor Park Jun Soo of the Division of Biological Science and Technology of Yonsei University. Raw 264.7 cells were transiently transfected with pcDNA3-HA-p65 plasmid DNA. The following day, the cells were cultured in complete media for an additional 48 h. Transfected Raw 264.7 cells were cultured with RANKL in the presence or absence of the indicated concentrations of EPZ compounds for two days. The cells were rinsed with PBS, harvested, and lysed in IP lysis buffer (20 mM Tris–HCl, pH 7.5, 150 mM NaCl, 10% glycerol, and 1% Triton X-100) containing protease inhibitors. The whole cell lysates were incubated on ice for 30 min, collected by centrifugation, and quantified. After the quantification of protein samples, equal volumes of protein samples were incubated with the Monoclonal Anti-HA−Agarose antibody at 4 °C overnight under gentle shaking. The resin was harvested by centrifugation, washed five times with PBS, and re-suspended in an equal volume of 2 × SDS loading buffer. All samples were heated at 95 °C for 5 min. Finally, the protein samples were separated by 10% SDS-PAGE. Western blot assays were performed as described previously. The results were presented as mean ± SD from three independent experiments in this study. The significant differences between the control and experimental groups were determined by a two-tailed Student’s t-test. Data were analyzed with GraphPad Prism 6.0 (GraphPad Software Inc., San Diego, CA, USA). * p < 0.05 and ** p <0.01 were considered significant.

PMC10002183

Iku Sazaki,Toshihiro Sakurai,Arisa Yamahata,Sumire Mogi,Nao Inoue,Koutaro Ishida,Ami Kikkai,Hana Takeshita,Akiko Sakurai,Yuji Takahashi,Hitoshi Chiba,Shu-Ping Hui

Oxidized Low-Density Lipoproteins Trigger Hepatocellular Oxidative Stress with the Formation of Cholesteryl Ester Hydroperoxide-Enriched Lipid Droplets

21-02-2023

LDL,non-alcoholic steatohepatitis,non-alcoholic fatty liver disease,liquid chromatography-mass spectrometry,lipidomics

Oxidized low-density lipoproteins (oxLDLs) induce oxidative stress in the liver tissue, leading to hepatic steatosis, inflammation, and fibrosis. Precise information on the role of oxLDL in this process is needed to establish strategies for the prevention and management of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH). Here, we report the effects of native LDL (nLDL) and oxLDL on lipid metabolism, lipid droplet formation, and gene expression in a human liver-derived C3A cell line. The results showed that nLDL induced lipid droplets enriched with cholesteryl ester (CE) and promoted triglyceride hydrolysis and inhibited oxidative degeneration of CE in association with the altered expression of LIPE, FASN, SCD1, ATGL, and CAT genes. In contrast, oxLDL showed a striking increase in lipid droplets enriched with CE hydroperoxides (CE-OOH) in association with the altered expression of SREBP1, FASN, and DGAT1. Phosphatidylcholine (PC)-OOH/PC was increased in oxLDL-supplemented cells as compared with other groups, suggesting that oxidative stress increased hepatocellular damage. Thus, intracellular lipid droplets enriched with CE-OOH appear to play a crucial role in NAFLD and NASH, triggered by oxLDL. We propose oxLDL as a novel therapeutic target and candidate biomarker for NAFLD and NASH.

Oxidized Low-Density Lipoproteins Trigger Hepatocellular Oxidative Stress with the Formation of Cholesteryl Ester Hydroperoxide-Enriched Lipid Droplets Oxidized low-density lipoproteins (oxLDLs) induce oxidative stress in the liver tissue, leading to hepatic steatosis, inflammation, and fibrosis. Precise information on the role of oxLDL in this process is needed to establish strategies for the prevention and management of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH). Here, we report the effects of native LDL (nLDL) and oxLDL on lipid metabolism, lipid droplet formation, and gene expression in a human liver-derived C3A cell line. The results showed that nLDL induced lipid droplets enriched with cholesteryl ester (CE) and promoted triglyceride hydrolysis and inhibited oxidative degeneration of CE in association with the altered expression of LIPE, FASN, SCD1, ATGL, and CAT genes. In contrast, oxLDL showed a striking increase in lipid droplets enriched with CE hydroperoxides (CE-OOH) in association with the altered expression of SREBP1, FASN, and DGAT1. Phosphatidylcholine (PC)-OOH/PC was increased in oxLDL-supplemented cells as compared with other groups, suggesting that oxidative stress increased hepatocellular damage. Thus, intracellular lipid droplets enriched with CE-OOH appear to play a crucial role in NAFLD and NASH, triggered by oxLDL. We propose oxLDL as a novel therapeutic target and candidate biomarker for NAFLD and NASH. Non-alcoholic fatty liver disease (NAFLD), characterized by the accumulation of fat stored in liver cells, affects people who consume little to no alcohol [1]. These patients are referred to as having simple steatosis (SS). Simple steatosis can progress to non-alcoholic steatohepatitis (NASH) with hepatitis and liver fibrosis when additional oxidative and cytokine stress occurs [2]. Non-alcoholic steatohepatitis is irreversible and progresses to cirrhosis and hepatic carcinoma [3]. Simple steatosis and NASH are collectively defined as NAFLD, which is often observed in patients with metabolic disorders (such as obesity and type 2 diabetes) [4,5]. The global prevalence of NAFLD and NASH in the general population is estimated to be 10–35% and 3–5%, respectively [2]. In the USA, 34% of the general adult population (or at least 43 million adults) have NAFLD, and 12% have NASH [6,7]. Furthermore, an estimated 20% of patients with NASH develop cirrhosis, and NASH is projected to become the leading indication for liver transplantation in the United States [8,9]. Thus, it is important to prevent the progression of SS to NASH. Currently, the histopathological diagnosis of NASH requires invasive liver biopsies and is clinically problematic because of the lack of useful noninvasive blood testing methods. Therefore, elucidation of NASH pathogenesis is urgently required. In addition to the fatty liver, NAFLD and NASH are also associated with dyslipidemia [10]. Particularly, patients with NASH have increased plasma oxidized low-density lipoproteins (oxLDL), which contain lipid hydroperoxides [11]. Lipid peroxidation is a leading factor in the development and progression of NASH [12]. Therefore, the oxLDL level is considered a risk factor for NASH. Our laboratory previously reported a NASH mouse model that was fed a long-term high-fat diet and administered with oxLDL [13]. This suggests that oxLDL is one of the factors involved in the pathogenesis of NASH. Clarifying the association between oxLDL and lipid droplet formation in hepatocytes can contribute to the understanding of NASH pathogenesis. However, only a few studies have focused on oxLDL levels and hepatocytes. There have been no reports on the involvement of oxLDL compared with native low-density lipoproteins (nLDL) in the mechanism of lipid droplet formation, the components of lipid droplets, and the increase of oxidative stress in hepatocytes. Herein, we clarified that the addition of nLDL or oxLDL to human liver-derived C3A cells causes fat accumulation and analyzed the lipid components using liquid chromatography-tandem mass spectrometry (LC-MS/MS) and transcriptional expression analyses to better understand lipid metabolic changes in the cells. We analyzed the profiles of cholesteryl ester (CE), triacylglycerol (TG), and their hydroperoxides in nLDL and oxLDL (which were added to C3A cells) because the major lipid components in lipid droplets are CE and TG [14]. Six CE molecular species were detected in the LDL particles (Supplementary Figure S1). All molecular species except CE 16:0 decreased with increasing oxidation time. The behavior of the sum of CE was consistent with that of each CE molecular species (Supplementary Figure S2). CE 20:5 was not detected in oxLDL (24 h). In the detection of CE hydroperoxide (CE-OOH), only CE-OOH 18:2, which increased considerably only in oxLDL (2 h), was detected (Supplementary Figure S3). Furthermore, 41 TG molecular species were detected in the LDL particles (Supplementary Figure S4). Similar to CE, almost all TG molecular species decreased with longer oxidation times. The behavior of the sum of TG was consistent with that of each TG molecular species (Supplementary Figure S5). In the detection of TG hydroperoxide (TG-OOH), 12 TG-OOH molecules were detected (Supplementary Figure S6). Similar to CE-OOH, the sum of the TG-OOH levels was considerably high in oxLDL (2 h) under all conditions (Supplementary Figure S7). Regarding phosphatidylcholine (PC), 22 PC and 10 PC hydroperoxides (PC-OOH) were detected in LDL (Supplementary Figures S8–S11). Additionally, the sum of PC-OOH was the highest in oxLDL (2 h). Taken together, oxLDL (2 h) was enriched in CE-OOH, TG-OOH, and PC-OOH compared with LDL at other oxidation times (8, 24 h). To evaluate the toxicity of LDL in C3A cells, lactate dehydrogenase (LDH) in the culture supernatant was analyzed (Figure 1). Under the present ranges of added LDL concentrations, no reduction in cell toxicity was observed in nLDL and any oxLDL groups (2, 8, and 24 h) compared with the control group at 50, 100, and 200 ng protein of LDL/104 cells. Thus, nLDL/oxLDL concentrations of 200 ng protein/104 cells were used to stimulate nLDL/oxLDL in this study. According to a previous report [15], fluorescence imaging was used for observing nLDL- or oxLDL (2 h)-induced lipid droplets, nuclei (blue), and the accumulation of neutral lipids (red) and lipid hydroperoxides (green) (Figure 2A–C). The locations where neutral lipids and lipid hydroperoxides overlapped are shown in yellow. As a result, their behaviors were different from each other. Native LDL increased the number of non-oxidized lipid droplets (non-oxLDs) and not oxidized lipid droplets (oxLDs), whereas oxLDL increased the number of oxLDs and non-oxLDs compared with the control (Figure 2D,E). Five CE molecular species were detected in LDL-supplemented cells (Figure 3A). The levels of all molecular species were considerably high in the nLDL group. Similarly, the sum of CE was also increased considerably in the nLDL group (Figure 3B). Three types of CE-OOH molecules (CE-OOH 18:1, CE-OOH 18:2, and CE-OOH 22:6) were also detected in LDL-treated cells (Figure 3C). The sum of CE-OOH increased considerably only in oxLDL (2 h)-treated cells (Figure 3D). Regarding the TG profile, 28 TG molecular species were detected in the LDL-supplemented cells (Figure 4A). In addition, 17 TG molecular species showed a considerable decrease in the nLDL group compared with the control. Furthermore, seven TG molecular species were decreased considerably in the oxLDL group compared with the control. The sum of the TG molecular species was reduced considerably in the nLDL and oxLDL groups compared with that in the control group (Figure 4B). Furthermore, three types of TG-OOH molecules (TG-OOH 52:2, TG-OOH 56:7, and TG-OOH 62:12) were detected in LDL-treated cells (Figure 4C). Overall, there was no notable difference in the sum of TG-OOH levels among the three groups (Figure 4D). Seventeen PC molecular species were detected in LDL-supplemented cells (Supplementary Figure S12A). The levels of all molecular species and the sum of PC were considerably higher in the nLDL-supplemented cells than in the oxLDL-supplemented cells and the control (Supplementary Figure S12B). Two types of PC-OOH molecules (36:5 and 36:6) were also detected in LDL-supplemented cells (Supplementary Figure S12C). Although there was no notable difference, the sum of PC-OOH showed an increasing trend in the oxLDL-supplemented cells (Con. vs. oxLDL, p = 0.093; nLDL vs. oxLDL, p = 0.103) (Supplementary Figure S12D). The sum of PC-OOH/PC, used as an index of cellular oxidative stress, increased considerably only in oxLDL-supplemented cells (Figure 5). To investigate the transcriptional changes in lipid metabolism in LDL-supplemented C3A cells, real-time polymerase chain reaction (PCR) was performed under the same conditions as the LC-MS/MS experiments. The expression level of sterol O-acyltransferase 1 (SOAT1), a gene associated with CE biosynthesis, was reduced considerably in the oxLDL group compared with that in the control and nLDL groups (Figure 6A). The expression of lipase E, a hormone-sensitive-type (LIPE) gene associated with the degradation of CE, was markedly reduced in the nLDL- and oxLDL-supplemented cells compared with that in the control (Figure 6A). No considerable differences were observed in the expression of diacylglycerol O-acyltransferase 1 (DGAT1), associated with TG biosynthesis (Figure 6B). The expression level of adipose triglyceride lipase (ATGL), a gene associated with TG degradation, was considerably increased in the nLDL group only (Figure 6B). The expression level of sterol regulatory element-binding protein 1 (SREBP1), a factor that regulates fatty acid biosynthesis, was decreased considerably in the oxLDL group compared with that in the control group (Figure 6C). The expression level of fatty acid synthase (FASN) was considerably reduced in the nLDL and oxLDL groups compared with that in the control group (Figure 6C). The expression of stearoyl-CoA desaturase (SCD1) for fatty acid unsaturation was considerably reduced in the oxLDL group compared with that in the control group and notably reduced in the nLDL group (Figure 6C). The expression of catalase (CAT), a hepatic antioxidant enzyme, was considerably increased in the nLDL group compared with that in the control and oxLDL groups (Figure 6D). From the above results, the lipid metabolic changes in nLDL- (Figure 7A) and oxLDL-supplemented C3A cells (Figure 7B) are summarized. The degree of oxidation varies among oxLDLs in plasma [16]; thus, oxLDL is a heterogeneous particle. Lipid components in LDL become hydroperoxides (-OOH) under early oxidative conditions and aldehydes (-CHO) with increasing degrees of oxidation [17,18]. Mild oxLDL reflects the physiological form of oxLDL and contains lipid hydroperoxides [19]; thus, mild oxLDL is toxic to the body. To determine the optimal oxidative conditions with high levels of lipid hydroperoxides, oxLDL was prepared using different oxidation times (0–24 h). The present experiments showed that the sum of CE decreased with increasing LDL oxidation time and that LDL oxidized for 2 h contained the highest levels of CE-OOH (Supplementary Figures S2 and S3). This suggests that CE may have been reduced by oxidative denaturation and transformed into CE-OOH in LDL, which was oxidized for 2 h. Further oxidation (8 and 24 h) resulted in the reduction of CE-OOH species (Supplementary Figure S3), which may subsequently produce aldehydes and other oxidative compounds. Similar to CE, the TG and PC levels decreased with increasing oxidation time (Supplementary Figures S4, S5, S8 and S9). TG-OOH and PC-OOH were the most abundant in LDL oxidized for 2 h (Supplementary Figures S6, S7, S10 and S11). Based on these results, oxLDL (2 h) with increased CE-OOH, TG-OOH, and PC-OOH was adopted as mildly oxidized LDL for the stimulation condition. Fluorescence imaging analysis showed that both nLDL and oxLDL accumulated neutral lipids in hepatocytes, suggesting that nLDL or oxLDL was incorporated into the cells and excess lipids were stored in lipid droplets. Furthermore, oxLDL-supplemented cells were observed to have lipid hydroperoxides overlapping with neutral lipids as a reference [15]. This suggests that lipid hydroperoxides of added oxLDL may have accumulated in lipid droplets. CE-OOH levels in the liver tissue of patients with NASH are elevated [20], and the presence of lipid hydroperoxides (-OOH) in hepatocytes is closely associated with NASH [21]. Thus, oxLDL uptake may enhance oxidative stress in the hepatocytes. Native LDL is taken up by hepatocytes via LDLR [22] and oxLDL via scavenger receptors, such as CD36 and LOX1 [23,24]. Simultaneous analysis of lipids in hepatocytes revealed that CE increased in the nLDL group. Native LDL is a CE-rich lipoprotein (Supplementary Figure S2). The lipid droplets that were stained in the fluorescence microscopy experiment can be derived from the CE in the incorporated nLDL (Figure 2). CE taken into cells can be degraded by hydrolysis [25,26], or CE synthesis can be suppressed by the downregulation of SOAT1 (a gene associated with CE biosynthesis) [27]. In the present study, the expression level of SOAT1 remained unchanged in nLDL-supplemented cells. In contrast, the expression level of LIPE (a gene involved in CE degradation) showed a distinct decrease (Figure 6A). Therefore, CE degradation can be suppressed, and subsequent CE accumulation occurs in hepatocytes. In contrast, CE was at the same level in the oxLDL group as in the control group (Figure 3A,B). This could be reasonable because oxLDL was markedly poor in CE owing to oxidative modifications of CE. The increase in CE-OOH in oxLDL-supplemented cells was smaller than the increase of CE in nLDL-supplemented cells. This might indicate that CE-OOH reacted with numerous other oxidative species (e.g., CE-CHO) that were not targeted in this study. CE-OOH 18:2 and CE-OOH 22:6 were increased only in oxLDL-supplemented cells (Figure 3C), suggesting a state of increased intracellular oxidative stress. The increase in CE-OOH 18:2 levels in the cells implied that oxLDL (2 h) was rich in CE-OOH 18:2 (Supplementary Figure S3). These acyl chains are polyunsaturated fatty acids (PUFAs). Because PUFAs are susceptible to oxidation, CE with PUFA may be oxidized in oxLDL-supplemented cells. This increase was consistent with the detection of more lipid hydroperoxide using fluorescence staining (Figure 2) and was likely to be attributed to CE-OOH in oxLDL. In contrast, nLDL-supplemented cells showed no increase in CE-OOH (Figure 3), which can be attributed to less CE-OOH in nLDL-supplemented cells. Thus, hepatic antioxidant enzymes may exert inhibitory effects on the excessive oxidation of CE incorporated into the cells. The present transcriptional study revealed an increased expression of the antioxidant enzyme-related gene CAT in nLDL-supplemented cells (Figure 6D). CAT is a family of antioxidant enzymes induced by the activation of the Keap1-Nrf2 pathway [28]. It is most abundant in the liver, kidneys, and erythrocytes and is responsible for degrading most of the hydrogen peroxide [29]. Hydrogen peroxide generates lipid hydroperoxides. Thus, the induction of CAT could eliminate CE-OOH. We predicted that TG derived from LDL might accumulate via LDL uptake. However, TG levels decreased in both nLDL- and oxLDL-treated cells (Figure 4A,B). An increase in ATGL (a gene involved in TG hydrolysis) promotes a decrease in TG accumulation [30]. Thus, in nLDL, the high expression level of ATGL in nLDL-supplemented cells may have promoted TG degradation and prevented TG accumulation in hepatocytes. However, the effects on the reduction of TG species were weaker in oxLDL-supplemented cells than in nLDL-supplemented cells (Figure 4A,B). This might be due to fewer TG species in the added oxLDL (Supplementary Figures S4 and S5) or few changes in TG species due to little induction of ATGL (Figure 4A and Figure 6B). The sum of TG-OOH in the cells did not change among the three groups (Figure 4D), despite the addition of oxLDL, including high levels of TG-OOH. Because the amount of TG-OOH was smaller than that of CE-OOH, even in oxLDL (Supplementary Figures S3 and S7), the changes in the cells caused by TG-OOH of oxLDL may have been neglected. PC-OOH is a primary peroxidative lipid that has been used to monitor hepatocellular damage by lipid peroxidation [31]. Liver PC-OOH levels were higher in NASH model mice than in control mice [21]. Furthermore, PC-OOH/PC has been used as an index of cellular oxidative damage [32]. Thus, we analyzed the PC-OOH and PC-OOH/PC using LC-MS/MS. We found that PC-OOH/PC was increased in oxLDL-supplemented cells compared with that in other groups, which suggests hepatocellular damage due to oxidative stress (Figure 5). In contrast, PC-OOH/PC was unchanged in nLDL, which suggests a protective effect of CAT against oxidation. In the nLDL and oxLDL groups, the expression levels of genes associated with the synthetic pathway of fatty acids induced by acetyl-CoA were suppressed despite the formation of lipid droplets (Figure 6C). This was consistent with previous reports on mice with fatty livers [33]. This might indicate negative feedback against the accumulation of excessive lipids, possibly suppressing TG accumulation. Blood samples were obtained from healthy participants after overnight fasting. Serum samples were separated by centrifugation at 2200× g for 10 min at 4 °C using a CE16RX (Hitachi Koki Co., Ltd., Tokyo, Japan). As previously reported, total lipoproteins were separated by ultracentrifugation [34]. Briefly, 2.0 mL of serum was adjusted to a density of 1.225 kg/L using potassium bromide (Fujifilm Wako Pure Chemical Corporation, Osaka, Japan) and mixed with 6.0 mL of a specific density solution (density = 1.225). The mixed solution was then ultracentrifuged at 50,000 rpm for 20 h at 4 °C in an Optima MAX Ultracentrifuge (Beckman Coulter Inc., Brea, CA, USA) with a near-vertical rotor MLN-80 (Beckman Coulter Inc., Brea, CA, USA). The total lipoprotein fraction was collected from the top layer. Gel filtration chromatography was performed to separate LDL and other lipoproteins (very low-density lipoprotein and high-density lipoprotein) [34]. The total lipoprotein fraction was injected into a high-performance liquid chromatograph (HPLC, Shimadzu Corp., Tokyo, Japan) equipped with a Superose 6 column (GE Healthcare, Little Chalfont, UK). The lipoproteins were then eluted with 50 mM phosphate-buffered saline (PBS) (pH 7.4) at a rate of 0.5 mL/min and monitored at OD 280 nm. The LDL fractions were collected at an elution time of 21–27 min. The protein concentrations of these fractions were determined using the Lowry method [35]. As indicated in previous reports [36], LDL fractions were diluted to a protein concentration of 0.2 mg/mL with phosphate buffer (50 mM PBS, pH 7.4); copper sulfate was added to a final concentration of 0.06 mM and incubated for 2, 8, and 24 h in a thermostatic chamber at 37 °C. Oxidation was stopped by adding ethylenediaminetetraacetic acid (EDTA) to a final concentration of 1.0 mM. To prevent oxidation due to residual copper ions, the solvent was replaced with phosphate buffer (50 mM PBS, pH 7.4) using a 100 kDa filter (Merck Millipore Ltd., Cork, Ireland) [37]. Oxidized LDL solutions were diluted to a protein concentration of 0.2 mg/mL and stored at 4 °C until immediately before use. nLDL was added to equal volumes of water instead of copper sulfate. As with oxLDL, EDTA was added and stored at 4 °C until use. Lipid extraction from each LDL was performed following the procedure described by Folch et al. [38] and Hui et al. [36] (n = 4 for each group). nLDL or oxLDL (600 µL) was added to 1.4 mL of distilled water and stirred (3500 rpm, 1 min) using a Multi-Speed Vortex (BIOSAN Ltd., Riga, Latvia). The mixture was transferred to a screw-tip test tube. Internal standards (IS, SPLASH™LIPIDOMIX® Quantitative Mass Spec Internal Standard, Avanti Polar Lipids, Inc., Alabaster, AL, USA) were diluted 50-fold with methanol. Next, 100 µL of diluted IS, 300 µL of methanol, and 2 mL of chloroform were added sequentially, stirred, and centrifuged (2200× g, 4 °C, and 10 min). The lower chloroform layer was collected. Then, 2 mL of chloroform was added to the remaining upper layer, the above collection procedure was repeated, and the lower layer was collected again. The solution was then dried using an evaporator (centrifugal concentrator CC-105; Tomy Industries, Tokyo, Japan). After drying, 300 µL of methanol was added, and the mixture was stirred to collect the total volume. The samples were centrifuged (18,800× g, 4 °C, 10 min), and the supernatant was collected. The samples were stored at −80 °C until immediately before measurement. For the simultaneous analysis of lipids using Orbitrap LC-MS, cells stimulated with LDL were collected (n = 6 for each group). C3A cell suspensions (10% fetal bovine serum, FBS) were seeded in 24-well plates at 2.0 × 105 cells/mL and 1 mL/well, and preincubation and stimulation were performed under the same conditions as for fluorescence staining. The cells were then washed once with PBS. PBS (500 µL) was added, and the cells were collected using a scraper. Fifty microliters of this solution were used to determine the protein concentration of the cells. The cells in the remaining 450 µL were precipitated by centrifugation, and the supernatant was discarded. Next, the cells were washed with PBS, centrifuged again, and the supernatant discarded. The precipitated cell mass was used as a sample and stored at −80 °C until immediately before lipid extraction from the cell mass. Lipids were extracted from each cell following the procedure described by Hara et al. [39] (n = 6 for each group). Diluted IS (200 µL) and chloroform (400 µL) were added to the collected cell mass and agitated using a Multi-Speed Vortex (BIOSAN, Ltd, Riga, Latvia.). The mixture was then centrifuged (Himac CE15R, Hitachi Koki Co., Ltd., Tokyo, Japan, 18,800× g, 4 °C, 10 min). The supernatant was collected, and the solution was allowed to dry using a centrifugal concentrator (CC-105, Tomy Industries, Tokyo, Japan). Next, 400 µL chloroform was added to the sample before transferring the supernatant and stirring under the same conditions. It was centrifuged, and the supernatant was collected into a sample that had just been allowed to dry. The solution was allowed to dry again, 100 µL of methanol was added, and the mixture was stirred using a Multi-Speed Vortex (BIOSAN Ltd., Riga, Latvia). The mixture was centrifuged, and the supernatant was collected. The samples were stored at −80 °C until immediately before measurement. Orbitrap LC-MS/MS was used for the simultaneous analysis of lipids in LDL and liver culture cell C3A, based on previous reports [40]. The LC section was performed on a Shimadzu Prominence HPLC system (Shimadzu Corp., Kyoto, Japan), and the analytical column was an Atlantis T3 Column (C18, 2.1 × 150 nm, and 3 µm; Waters Corp., Milford, CT, USA). The column temperature was 40 °C, and the sample injection volume was 10 µL. A gradient elution method was used, and the mobile phase consisted of 5 mM ammonium acetate (A), isopropanol (B), and methanol (C). The flow rate was 0.2 mL/min; the MS section was an LTQ Orbitrap mass spectrometer (Thermo Fisher Scientific Inc., Waltham, MA, USA). Depending on the target molecule, the analysis was performed in the electrospray ionization (ESI)-positive ion mode (Supplementary Table S1). Peak areas were calculated using Xcalibur 2.2 (Thermo Fisher Scientific Inc., Waltham, MA, USA). Based on published data, we identified the peak using the LIPIDMAPS database [21]. The identified species were annotated as class abbreviations: lipid class, total number of carbons in fatty acid moieties:total number of double bonds in the acyl chains (e.g., TG 52:2) (Supplementary Tables S2–S5). When calculating the lipid concentrations in LDL, the peak areas of the target molecules were corrected using the peak area of the IS. When calculating intracellular lipid concentrations, the peak area was corrected using the peak area of the IS and protein concentration in the hepatocytes. CE, TG, PC, and their peroxides (CE-OOH, TG-OOH, and PC-OOH) were analyzed. Human liver-derived strain C3A (ATCC) cells were passaged and cultured in MEM supplemented with fetal bovine serum (FBS, Thermo Fisher Scientific Inc., Waltham, MA, USA, final concentration of 10%), penicillin-streptomycin (Thermo Fisher Scientific Inc., Waltham, MA, USA, final concentration 1%), and GlutaMAX supplement (GlutaMAX, Thermo Fisher Scientific Inc., Waltham, MA, USA, final concentration of 1%) at 37 °C and 5% CO2. To confirm the hepatotoxicity of LDL, a cell toxicity test was performed (n = 6 for each group): 1.0 × 104 cells/mL of C3A (10% FBS-containing medium) were seeded in 96-well plates at 100 µL/well and pre-incubated at 37 °C and 5% CO2 for 24 h. For the preparation of stimulants, the supernatant was mixed with Clear MEM (0% FBS, Thermo Fisher Scientific Inc., Waltham, MA, USA) with LDL solution adjusted to 50, 100, and 200 ng protein per 1 × 104 cells in the nLDL and oxLDL groups. An equal volume of PBS was used instead of LDL for the control group. The supernatant was then replaced with stimulants (100 µL/well), and cells were stimulated at 37 °C and 5% CO2 for 22 h. At the end of stimulation, the supernatant was collected. LDH was measured for toxicity testing according to the manufacturer’s instructions (Takara Bio Inc., Shiga, Japan). Absorbance at 490 nm was measured using a microplate reader (xMark™ Microplate Spectrophotometer, Bio-Rad Laboratories, Inc., Hercules, CA, USA). The values are shown as the percentage of cell toxicity, with the control set to 100%. With reference to Tsukui et al. [15], fluorescence staining was performed to quantify the lipid droplets observed when LDL was added. C3A cell suspensions (10% FBS) were seeded with 2.0 × 105 cells of C3A in glass-bottomed dishes and pre-cultured for 24 h at 37 °C and 5% CO2. The cells were washed once with PBS (Fujifilm Wako Pure Chemical Corporation, Osaka, Japan). For the preparation of the stimulant medium, Clear MEM (0% FBS) was mixed with nLDL or oxLDL and adjusted to 200 ng protein/104 cells for nLDL, oxLDL, or phosphate buffer (50 mM PBS, pH 7.4) in the same volume as LDL for the control group. Subsequently, the cells were incubated at 2 mL/dish at 37 °C and 5% CO2 for 24 h. At the end of the stimulation, the cell supernatant was discarded. The staining solution was added at 1 mL/dish at 37 °C, 5% CO2, and light-shielded conditions for 30 min. The staining solution was a mixture of Clear MEM, SRfluor (a fluorescence probe for neutral lipids, Molecular Targeting Technologies, Inc., West Chester, PA, USA), Liperfluo (a fluorescence probe for lipid peroxides, Dojin Chemical Laboratories, Kumamoto, Japan), and Hoechst 33342 (a fluorescence probe for nuclei, Fujifilm Wako Pure Chemical Corporation) at a ratio of 1000:5:5:1. After staining, the cells were washed twice with PBS, and the supernatant was replaced with 2 mL/dish of FluoroBrite™ DMEM (Thermo Fisher Scientific Inc., Waltham, MA, USA). Cells were observed under a fluorescence microscope (HS all-in-one fluorescence microscope BZ-9000, Keyence Co. Ltd., Osaka, Japan). Excitation/emission wavelengths for SRfluor, Liperfluo, and Hoechst 33342 were 620 nm/700 nm, 470 nm/525 nm, and 360 nm/460 nm, respectively. Exposure times for SRfluor, Liperfluo, Hoechst 33342, and bright field observations were unified at 1/1.5, 1/2.5, 1/12, and 1/120 s, respectively. With reference to Piao et al. [41], the images were converted to images suitable for analysis using the HS all-in-one fluorescence microscope BZ-II analysis application (Keyence Co. Ltd., Osaka, Japan) and analyzed for the total number and area of lipid droplets using ImageJ software (NIH, Bethesda, MD, USA) [41,42]. For simplicity, cells in which the entire cell could be identified were included in the analysis (number of counted cells = 53–84 in each group) (Supplementary Figure S13). The images were then converted to 8-bit grayscale images for binarization. A grayscale threshold (0–10) was applied to the images to remove hepatocellular structures that did not exhibit lipid droplet features. All particles with a circularity of 0.00–1.00 and an area of 0.1–50 µm2 were counted. The total number of particles was calculated using this method. The settings of the various analysis parameters were standardized for all images. To confirm the hepatocellular lipid metabolic changes induced by LDL, transcriptional expression analysis was performed under the same conditions as in LC-MS/MS experiments (n = 6–8 for each group) [43]. After stimulation for 24 h in an incubator, the cells were washed with PBS, and RNA was extracted from the cells using an RNA extraction kit (PureLink™ RNA Mini Kit, Thermo Fisher Scientific Inc., Waltham, MA, USA) according to the manufacturer’s instructions. RNA concentrations were measured using NanoDrop One (Thermo Fisher Scientific Inc., Waltham, MA, USA). RNA was converted to complementary DNA (cDNA) using ReverTra Ace® qPCR RT Master Mix with gDNA Remover (TOYOBO, Co., Ltd., Osaka, Japan) in a thermal cycler (GeneAmp® PCR System 9700, Applied Biosystems, Foster City, CA, USA). The cDNA was stored at −80 °C. For the PCR reaction, Thunderbird® Next SYBER® qPCR Mix (TOYOBO, Co., Ltd., Osaka, Japan) was mixed with cDNA samples and primers according to the manufacturer’s instructions. Target genes included SOAT1, LIPE, DGAT1, ATGL, SREBP1, FASN, SCD1, and CAT, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was the housekeeping gene. Gene-specific primers were used to analyze gene expression (Supplementary Table S6). Gene expression levels were analyzed by the 2−(ΔΔCq) method using a real-time PCR analysis system (CFX Connect Real-Times System, Bio-Rad Laboratories, Inc., Hercules, CA, USA). The expression levels of each target gene were corrected for the expression level of the housekeeping gene GAPDH. All data obtained were subjected to statistical analysis using the GraphPad Prism V7.0 software (GraphPad Software Inc., San Diego, CA, USA). Statistical analysis was performed using rejection tests to identify outliers where necessary. Then, we performed one-way analysis of variance (ANOVA), followed by Tukey’s multiple comparisons test, or one-way ANOVA, followed by Dunnett’s multiple comparisons test, or followed by Kruskal-Wallis test, or Student’s t-test. The significance level was set at 5%. All results are expressed as mean ± standard deviation (SD) or box plots. Ethical approval for blood sampling from healthy subjects was obtained from Hokkaido University (approval number: 19-107-3). Informed consent was obtained from all the subjects. The present study demonstrated that nLDL causes the accumulation of CE and the formation of lipid droplets, possibly due to the reduced expression of LIPE in hepatocytes. In contrast, oxLDL appeared to increase lipid hydroperoxide-rich LDs and PC-OOH/PC, mainly CE-OOH and PC-OOH derived from oxLDL. These results suggest that stimulation by oxLDL mediates oxidative stress in the liver and could trigger NASH development. The limitation of the present study is that it is difficult to determine whether this model is closer to NASH or NAFLD because it is a simple cell experiment. Further studies using various cell lines, primary cells, and in vivo experiments are needed to determine the interaction between oxLDL and the liver in detail. However, our demonstration of an association between oxLDL and hepatocytes may lead to new findings in understanding the pathogenicity of oxLDL in NASH. In addition, our data suggest that oxLDL could be established as a novel pharmacological target and candidate biomarker for NAFLD/NASH.

PMC10002205

Merlin G. Butler

Prader–Willi Syndrome and Chromosome 15q11.2 BP1-BP2 Region: A Review

21-02-2023

Prader–Willi syndrome (PWS),PWS molecular genetic classes,typical 15q11-q13 Type I,Type II deletions,15q11.2 BP1-BP2 deletion,clinical findings

Prader–Willi syndrome (PWS) is a complex genetic disorder with three PWS molecular genetic classes and presents as severe hypotonia, failure to thrive, hypogonadism/hypogenitalism and developmental delay during infancy. Hyperphagia, obesity, learning and behavioral problems, short stature with growth and other hormone deficiencies are identified during childhood. Those with the larger 15q11-q13 Type I deletion with the absence of four non-imprinted genes (NIPA1, NIPA2, CYFIP1, TUBGCP5) from the 15q11.2 BP1-BP2 region are more severely affected compared with those with PWS having a smaller Type II deletion. NIPA1 and NIPA2 genes encode magnesium and cation transporters, supporting brain and muscle development and function, glucose and insulin metabolism and neurobehavioral outcomes. Lower magnesium levels are reported in those with Type I deletions. The CYFIP1 gene encodes a protein associated with fragile X syndrome. The TUBGCP5 gene is associated with attention-deficit hyperactivity disorder (ADHD) and compulsions, more commonly seen in PWS with the Type I deletion. When the 15q11.2 BP1-BP2 region alone is deleted, neurodevelopment, motor, learning and behavioral problems including seizures, ADHD, obsessive-compulsive disorder (OCD) and autism may occur with other clinical findings recognized as Burnside–Butler syndrome. The genes in the 15q11.2 BP1-BP2 region may contribute to more clinical involvement and comorbidities in those with PWS and Type I deletions.

Prader–Willi Syndrome and Chromosome 15q11.2 BP1-BP2 Region: A Review Prader–Willi syndrome (PWS) is a complex genetic disorder with three PWS molecular genetic classes and presents as severe hypotonia, failure to thrive, hypogonadism/hypogenitalism and developmental delay during infancy. Hyperphagia, obesity, learning and behavioral problems, short stature with growth and other hormone deficiencies are identified during childhood. Those with the larger 15q11-q13 Type I deletion with the absence of four non-imprinted genes (NIPA1, NIPA2, CYFIP1, TUBGCP5) from the 15q11.2 BP1-BP2 region are more severely affected compared with those with PWS having a smaller Type II deletion. NIPA1 and NIPA2 genes encode magnesium and cation transporters, supporting brain and muscle development and function, glucose and insulin metabolism and neurobehavioral outcomes. Lower magnesium levels are reported in those with Type I deletions. The CYFIP1 gene encodes a protein associated with fragile X syndrome. The TUBGCP5 gene is associated with attention-deficit hyperactivity disorder (ADHD) and compulsions, more commonly seen in PWS with the Type I deletion. When the 15q11.2 BP1-BP2 region alone is deleted, neurodevelopment, motor, learning and behavioral problems including seizures, ADHD, obsessive-compulsive disorder (OCD) and autism may occur with other clinical findings recognized as Burnside–Butler syndrome. The genes in the 15q11.2 BP1-BP2 region may contribute to more clinical involvement and comorbidities in those with PWS and Type I deletions. Prader–Willi syndrome (PWS) is caused by genomic imprinting errors with absence of expression of imprinted genes in the paternally derived PWS/Angelman syndrome (AS) region involving the chromosome 15q11.2-13 region by several genetic mechanisms. The most common cause is a paternal deletion followed by maternal disomy 15 where both 15s are from the mother, imprinting defects or chromosome 15 abnormalities (e.g., [1,2,3,4,5]). PWS affects about one in 15,000–20,000 individuals with an estimated 400,000 cases worldwide. This rare obesity-related genetic disorder has severe infantile hypotonia accompanied by poor suck with swallowing problems, sticky saliva and failure to thrive along with hypogonadism, hypogenitalism and development delay; many of these features compose the consensus diagnostic criteria triggering genetic testing for PWS [1,6,7,8]. Unique facial features are noted in PWS including bifrontal narrowing, almond-shaped eyes and a small chin with a high palate; additionally, small hands and feet with short stature due to growth and other hormone deficiencies involving the endocrine system and sex organs, pancreas, adrenal and thyroid glands occur [1,4,6,7,8,9,10,11,12,13,14,15,16,17]. Obesity, growth anomalies and hypogonadism are due to central and peripheral mechanisms involving the hypothalamus–pituitary–gonadal axis. Nutritional phases previously described in PWS [7,18] display clinical stages of failure to thrive during infancy and excessive eating with hyperphagia in early childhood along with reduced physical activity and a lower metabolic rate leading to severe obesity, if not controlled externally [7,14]. Hyperphagia is an important health problem leading to both mortality and causes of death [19,20] with associated comorbidities typically lasting throughout life. The most common causes of death in a large survey of PWS patients studied were respiratory failure in 31%, followed by cardiac (16%), gastrointestinal (10%), infection (9%), obesity (7%) and pulmonary embolism (7%). Choking (6%) and accidents (6%) were reported more often in childhood or as young adults. The average age of death was 29.5 years [19,20]. The mortality rate for PWS is estimated at 3% per year across an age range of 0 to 47 years and 7% per year for patients aged >30 years. The most reported causes of death in children are respiratory infections and sudden deaths [21]. Mild intellectual disabilities in PWS are noted with an average IQ of 65 and about one-third having a normal IQ but often with delayed language and motor skills (e.g., [22]). Patients with PWS have unique symptoms and associated psychiatric or behavioral problems beginning in early childhood in greater than 70% of PWS patients, including emotional disturbances and obsessive-compulsive disorders, anxiety, depression, controlling and manipulative behavior, violent outbursts, stubbornness and skin picking [7,16,23]. The severity increases with age but diminishes in older patients. A high pain threshold is present along with eating nonfood or inedible items [1,6,7,11,14,16]. They also become easily frustrated with impulsivity, have a quick response to anger and lack of flexibility. Attention deficit hyperactivity with insistence to sameness is often observed in PWS and at an early age. Early diagnosis appears to lead to an improved prognosis and allows for potential treatment approaches to impact quality of life and life expectancy [24] (see Figure 1). An emerging disorder that shares genetic components with PWS is now recognized as the 15q11.2 BP1-BP2 deletion (Burnside–Butler) syndrome. The 15q11.2 BP1-BP2 region contains four genes in common with those with PWS having a typical chromosome 15q11-q13 deletion and will be discussed later in this review. Burnside–Butler syndrome is associated with motor and developmental delays, neurobehavioral problems including dyslexia, autism and psychosis with reported congenital anomalies [7,9]. Several of these findings are common in PWS, more so in those with the larger typical deletion. The most recent studies using advanced genetic testing in the largest PWS cohort to date [2] showed that a 15q11-q13 paternal deletion is found in about 60% of PWS individuals, about 35% with maternal disomy 15, and the remaining individuals with imprinting defects, chromosome 15 translocations or inversions. The 15q11-13 proximal deletion breakpoint is located at two sites (i.e., breakpoint BP1 or breakpoint BP2) at the 15q11.2 band and located within either of two large duplicons predisposing for deletion hotspots at these sites (e.g., [25,26]). The larger Type I deletion at approximately 6 Mb in size involves the proximal BP1 breakpoint near the centromere and a distal 15q11-q13 breakpoint (BP3). The Type II deletion is smaller and involves proximal BP2 site located about 500 kilobases distal to breakpoint BP1 and the more distal breakpoint BP3, a breakpoint that is common in both the typical 15q11-q13 Type I or Type II deletion subtypes (e.g., [1,2,25,26], see Figure 2). Prader–Willi syndrome is recognized as the first example of genomic imprinting in humans with dozens of genes and transcripts identified and located between chromosome 15q11-q13 breakpoints BP1 and BP3 flanked by low copy repeats prone to non-homologous recombination that leads to PWS. Genes in this 15q11-q13 region are both imprinted (NDN, MAGEL2, MKRN3, SNURF-SNRPN, SNORDs, UBE3A, ATP10A) and non-imprinted (NIPA1, NIPA2, CYFIP1, TUBGCP5, GABA receptors, OCA2 albinism). There are 165 recognized human and 197 mouse genes currently known to be imprinted or active depending on the parent of origin. Several genes in the 15q11-q13 region have been implicated in neurodevelopment and function with a role in behavior and learning, ataxia, hyperphagia and obesity, magnesium transportation, hypogonadism and precocious puberty, circadian rhythm, autism and skin pigment production with albinism [1,2,7,17,24,25,26,27,28]. Information about the functional status of chromosome 15 genes, both imprinted and non-imprinted have been characterized. Specifically, the NDN (neurally differentiated EC cell-derived factor) gene which interacts with hundreds of encoded proteins such as brain-derived neurotrophic factor (BDNF) and ubiquitin E3 ligase has been studied which leads to degradation of the proapoptotic or cell cycle apoptosis regulatory protein. Additionally, MAGEL2 or melanoma antigen-like 2 gene is imprinted in the brain and expressed from the paternal chromosome 15 allele. This gene is intron-less and associates with ubiquitin E3 ligase by altering activity, substrate specificity and subcellar location. Nonsense mutations of the MAGEL2 gene are reported in Schaaf–Yang syndrome. At an early age, individuals with this syndrome have overlapping features including hyperphagia seen in PWS (e.g., [7,8]). The MKRN3 or Makorin ring finger protein 3 gene is also imprinted and expressed on the paternal allele. The MKRN3 gene plays a role in puberty and is expressed in the hypothalamus. It blocks transcription of KISS (KiSS-1 metastasis suppressor) and TAC3 (tachykinin precursor 3), which are important for release of GnRH (gonadotrophin-releasing hormone) which initiates puberty [7,17]. SNURF-SNRPN (SNRPN upstream reading frame (SNURF)-small nuclear ribonucleoprotein polypeptide N (SNRPN)), a complex gene locus belonging to the SNRPN SmB/SmN family. The protein plays a role in pre-mRNA processing, tissue specific alternative splicing events and transcript production. SNURF-SNRPN is bi-cistronic in nature with over 100 exons that undergo alternative splicing and encodes two different proteins with exons 1–3 for SNURF producing a polypeptide and exons 4–10 generating a spliceosome protein (SmN) involved in mRNA splicing. The 5′ untranslated region component of this gene is identified as an imprinting center. This gene hosts six snoRNAs which are regulated or under the control by expression of the SNURF-SNRPN complex gene locus. SnoRNAs do not encode or generate protein but can impact the expression of genes and function of related proteins. Errors of paternally expressed genes/transcripts in this region do cause PWS and a second embedded imprinting center controls the maternally expressed UBE3A (ubiquitin-protein ligase E3A) gene causing Angelman syndrome when altered (deleted, mutated or by paternal disomy 15) (e.g., [7,16,17,29]). SNORD116 is one of the snoRNAs in the chromosome 15q11-q13 region and is expressed in the hypothalamic region of the brain which regulates appetite, leading to obesity [17]. It appears to play a key role in the development of features seen in PWS as recognized in those with a deletion of this transcript [7]. In mice with a deletion of this transcript, postnatal growth retardation and increased food intake are noted in research studies. These mice also show dysregulation of diurnally expressed genes such as Mtor and circadian Clock, Cry1 and Per2 genes [7]. Sleep disturbances are also found in individuals with PWS [1,4,7,16]. Studies on prohormone convertase (PC1) are helping to further understand the clinical phenotype in PWS based on this protein involvement in several hormonal pathways and disturbances seen in PWS. These include hyperphagic obesity, hypogonadism, short stature, growth and other hormone deficiencies, hyperghrelinemia and relative hypoinsulinemia often due to impaired prohormone processing [30]. Rare microdeletions of the SNORD116 in humans, Snord116 knock-out mice models and induced pluripotent stem cell-derived neurons from PWS patients support the role of disturbed prohormone convertase (PC1) activity. Burnett et al. [30] in 2016 reported reduced levels and activity of prohormone convertase PC1 in their studies. They proposed and presented data that humans and mice who are deficient in PC1 display hyperphagic-related obesity. They also have high ghrelin levels (key appetite-inducing hormone produced by the stomach), hypogonadism, and decreased growth hormone and insulin levels. This leads to short stature with reduced growth and diabetes occurring as a result of impaired prohormone processing and lack of active or functional hormones required for normal response, affecting multiple organ systems as seen in PWS. For example, POMC (pro-opiomelanocortin) is a large prohormone that requires cleavage from inactive prohormone status to smaller active peptides for normal function, including for appetite regulation and normal eating behavior which is absent in PWS. Hence, PC1, which is the protein (enzyme) encoded by the PCSK1 (proprotein convertase, subtilisin/kexin-type 1) gene located on chromosome 5q15 and is involved in post-translational modification or change of prohormone (inactive) to individual peptides (active) status, may be influential. It is suggested that several major neuroendocrine clinical findings seen in PWS could be due to PC1 deficiency requiring more investigations with the potential to lead to therapeutic agents [17,30]. Butler et al. [31] reported PWS patients and the chromosome 15q11-q13 deletion were more affected than patients with maternal disomy 15. Distinct differences were also reported in those with the two typical 15q11-q13 deletions compared with maternal disomy 15, particularly in phenotype, learning and psychiatric/behavioral parameters [31]. Specifically, Roof et al. [32] reported those with PWS and maternal disomy 15 had higher Verbal IQ scores than those with the 15q11-q13 deletion. Furthermore, PWS individuals with the deletion had more self-injury and severe behavior with lower intellectual ability than those with maternal disomy 15 [1,7,16]. The first clinical differences in individuals with PWS were described by Butler et al. [31] in 2004 when examining Type I or Type II deletions including assessments for intellectual, adaptive and aberrant behavior assessments, reading and math skills and visual-motor integration. Generally, poorer assessment scores were found in PWS individuals with Type I deletions compared to those with the smaller Type II deletions or maternal disomy 15. The larger typical Type I deletion accounts for about 40% of the typical 15q11-q13 deletions in PWS [2,31]. Specifically, those with the larger Type I deletions had more compulsions, poorer adaptive behavior and reduced cognition than those with the smaller Type II deletions. Those with the larger deletion had more severe compulsions related to grooming and bathing and compulsions that were more disruptive to daily living [1,31,33,34,35]. Intellectual ability and academic achievement skills were analyzed, and visual processing was poorer in those with the larger deletion [36,37]. Furthermore, self-injurious behaviors were more commonly observed in those with the larger Type I deletion. Hence, individuals with PWS and the larger 15q11-q13 Type I deletion including the four non-imprinted protein coding genes (NIPA1, NIPA2, CYFIP1, TUBGCP5) in the 15q11.2 BP1-BP2 region are more clinically impaired. Those individuals with 15q11.2 BP1-BP2 deletions are missing the four genes alone and do not have PWS but have Burnside–Butler syndrome (BBS) (e.g., [27,38,39]) with developmental motor and speech delays, congenital findings, behavioral problems including autism and brain imaging abnormalities (e.g., [27]). Individuals with the larger typical 15q11–q13 Type I deletion were found to have more severe neurodevelopmental symptoms when compared to those with PWS or Angelman syndrome, a sister genomic imprinting disorder with loss of maternally expressed genes on the chromosome 15q11-q13 region with the smaller typical Type II deletions (e.g., [30,33,34,35,40,41,42]). Bittel et al. [43] reported on molecular gene expression (messenger-RNA) studies from lymphoblastoid cells obtained from PWS males and females from highly conserved NIPA1, NIPA2, CYFIP1 and TUBGCP5 genes. They found that 24–99% of the phenotypic variability seen in behavioral and academic measures in PWS subjects could be explained by individual gene expression patterns. Levels of messenger-RNA from NIPA1, NIPA2, CYFIP1 and TUBGCP5 were reduced but detectable in individuals with PWS and the Type I deletion, supporting biallelic expression. Generally, messenger-RNA values were positively correlated with assessment measures, indicating a direct relationship between messenger-RNA levels and better assessment scores. The highest correlation was for NIPA2. Negative associations were found between age and behavior in the Type I deletion subtype only and implicating the four genes, specifically CYFIP1 and NIPA2. Disturbed expression of CYFIP1 is seen in other developmental disabilities including those with 15q disorders without PWS [44,45] and fragile X syndrome [46]. NIPA1 and NIPA2 are known to encode magnesium transporters and magnesium levels were recently reported to be lower in those with PWS and the Type I deletion compared to those with Type II deletions [28,47,48]. Clinical findings were reported in the literature from 200 patients with 15q11.2 BP1-BP2 deletion (Burnside–Butler) syndrome grouped into five categories [27]. These categories were (1) developmental (73% of cases), speech (67%) and motor delays (42%); (2) dysmorphic ear (46%) and palatal defects (46%); (3) writing (60%) and reading (57%) difficulties, memory problems (60%) and verbal IQ scores ≤ 75 (50%); (4) general behavioral problems, unspecified (55%); and (5) abnormal brain imaging including white matter disease (43%). Less often seen features were seizures/epilepsy (26%), autism spectrum disorder (ASD) (27%), attention-deficit hyperactivity disorder (ADHD) at 35% of cases and schizophrenia/paranoid psychosis (20%). Furthermore, Davis et al. [49] reported the parent of origin in dozens of families and found a maternal origin of the 15q11.2 BP1-BP2 deletion to be associated with a significantly higher risk for developmental, motor and speech delays, intellectual and learning problems, autism and behavioral/psychiatric diagnoses. Those with paternal chromosome 15q11.2 BP1-BP2 deletions were more prone to poor coordination/ataxia and congenital anomalies. The 15q11.2 BP1-BP2 region is deleted in PWS individuals having the larger Type I deletion. Butler [47] and Butler et al. [48] reported on the role of the four genes found in the 15q11.2 BP1-BP2 region involving magnesium transportation in the clinical presentation and potential treatment of those with Type I deletion. Figure 3 illustrates a frontal view of a non-dysmorphic mother and child having the 15q11.2 BP1-BP2 deletion (Burnside–Butler) syndrome. One allele is missing in each of the four genes (NIPA1, NIPA2, CYFIP1, TUBGCP5) in the 15q11.2 BP1-BP2 region due to a deletion designated as Burnside–Butler syndrome, emerging with variable clinical findings including a neurodevelopmental-autism non-dysmorphic phenotype with low penetrance. This phenotype presents with features including language and/or motor delay, cognitive impairment, aberrant behavior and autism, poor coordination with ataxia, seizures and congenital anomalies [50,51,52,53,54,55,56,57]. This small proximal 15q11.2 deletion characterized by Burnside et al. [39] in 2011 is now recognized as the most common chromosome finding in large cohorts of those presenting with neurodevelopmental problems and/or autism [50]. Ho et al. [50] in 2016 used 2.8 million DNA markers including single nucleotide polymorphic (SNP) and copy number variant (CNV) probes optimized for detection of regions of homozygosity and CNVs associated with neurodevelopmental disorders with high-resolution chromosomal microarrays. Neurodevelopmental disorders include developmental delay and intellectual disabilities with ASD, which affect up to 15% of all children. About 40% of those with ASD also have learning disabilities and approximately 30% show other comorbidities including seizures (e.g., [58,59]). Genetic testing may allow pinpointing the causes critical for clinical management and genetic counseling of at-risk family members. Ho et al. [50] summarized results in a total of 10,351 custom microarrays performed on patients over a period of four years with a male to female ratio of 2.5:1 and a mean age of 7.0 years. This neurodevelopmental patient cohort comprised 55% of cases with a diagnosis of ASD with or without other features (ASD+ and ASD only). Neurologists were the most common referring physician group at 36%, followed by developmental pediatricians at 31%, pediatricians at 16% and medical geneticists at 14%. Psychiatrists referred only 2% of the total cases but had the highest indication of ASD at 72% with or without features. In the study, 74% of ASD cases were referred by pediatric neurologists or developmental/behavioral pediatricians. Potentially abnormal CNVs were observed in 28% of cases with an average 1.2 reportable CNVs per individual. Overall detection rate for individuals with ASD was significant at 24.4%. The detection rate for a pathogenic cause using chromosome microarray analysis varied by indication for testing, age and gender as well as specialty of the ordering physician. The most common genetic defect identified in the combined ASD group (N = 5694 patients) was the 15q11.2 BP1-BP2 deletion, and the 22q11.2 deletion was seen most often in the non-ASD group (N = 4657). The most common cytogenetic finding seen in both the ASD+ group (N = 2844) and ASD only group (N = 2850) was also the 15q11.2 BP1-BP2 deletion. This chromosome 15 defect was the most common finding in both females and males. Of the 85 genetic findings reported by Ho et al. [50], 9% of the patients had the 15q11.2 BP1-BP2 deletion, followed by the 16p11.2 deletion at 5% and 16p11.2 duplication at 5%. Other findings included the 15q13.3 deletion, 16p13.1 duplication and NRXN1 gene deletion all at 4%. Hence, a greater understanding of the 15q11.2 BP1-BP2 deletion may further impact the role in the context of PWS, the most classical chromosome 15 disorder in humans. To further investigate biological pathways involving the 15q11.2 BP1-BP2 region, Rafi and Butler [28] examined STRING protein–protein interactions that encompass the four 15q11.2 BP1-BP2 genes with predicted Gene Ontology (GO) functions and processes. They found a role in magnesium ion transport, regulation of cellular growth and development with production of bone morphogenetic protein (BMP) and signaling pathways, regulation of axonogenesis and axon extension, cellular growth and development, and plasma membrane bounded cell projection and mitotic spindle organization. Using searchable genomic disease websites and tabulating disease data, they found the top ten overlapping significant neurodevelopmental disorders to be Prader–Willi Syndrome (PWS); Angelman Syndrome (AS); 15q11.2 Deletion Syndrome with Attention Deficit Hyperactive Disorder and Learning Disability; Autism Spectrum Disorder (ASD); Schizophrenia; Epilepsy; Down Syndrome; Microcephaly; Developmental Disorder; and Peripheral Nervous System Disease. These were also individually associated with PWS, ASD, ataxia, intellectual disability, schizophrenia, epilepsy and Down syndrome. Of the four non-imprinted biallelic genes and their encoded proteins, NIPA1 protein interacts with 11 other proteins, of which five (45%) are bone morphogenic protein (BMP) superfamily members, three (27%) are BMP receptors and one is the TGFB1 (9%) protein [28]. Therefore, three-fourths of NIPA1 interacting proteins are important for developmental bone morphogenesis or involved in multifunctional proteins controlling proliferation, differentiation and other cellular functions. The NIPA2 gene and protein interact with 19 other proteins and of these three (16%) are involved with the BMP protein superfamily, three (16%) proteins interact with BMP receptors ACVR1 and TGFBR1, and six are members of the SMAD superfamily of proteins (42%). These genes are important as intracellular signal transducers and transcriptional modulators activated by TGFB [28]. NIPA1 and NIPA2 genes also encode magnesium transporter proteins (e.g., [28,54]). Picinelli et al. [55] reported on a small number of patients with 15q11.2 BP1–BP2 deletions or duplications and found an inverse relationship for the NIPA2 gene encoding a magnesium transporter in both central nervous system (CNS) and renal tubules to be directly associated with urinary magnesium levels. Those with this gene deletion had higher urinary magnesium and those with this gene duplicated showed lower urinary levels. PWS patients with the larger Type I deletion had lower plasma magnesium levels [48]. In addition, Xie et al. [56] reported NIPA2 gene mutations that showed incorrect localization of the NIPA2 transporter protein in neurons. This resulted in decreased intracellular magnesium levels apparently due to reduced cross-membrane transport involving renal tubules. Furthermore, Mg2+ is involved in gating and activation of channels and receptors including NMDARs, playing a role in memory processing and altered neural extracellular Mg2+ concentrations caused by NIPA2 deletions in neuronal cells. NIPA1 (non-imprinted in PWS/AS 1) gene defects cause autosomal dominant hereditary spastic paraplegia and postural disturbance [60,61]. NIPA1 is known to mediate Mg2+ transport and is highly expressed in the brain [28]. The NIPA1 protein can also transport other divalent cations such as Fe2+, Sr2+, Ba2+, Mn2+ and Co2+ and possibly impacting their levels. The NIPA2 (non-imprinted in PWS/AS 2) gene encodes a protein acting as a renal Mg2+ transporter. Three reported NIPA2 mutations (p.I178F, p.N244S and p.N334_E335insD) are found in childhood absence epilepsy [54,56]. NIPA2 gene variants and functional studies have shown decreased intracellular magnesium concentrations in neurons, suggesting lower intracellular magnesium levels may enhance N-methyl-d-aspartate receptor (NMDAR) currents impacting neuron excitability and brain function. CYFIP1 (cytoplasmic fragile X mental retardation 1 FMR1 interacting protein 1) is reported to interact with other proteins with functions related to actin filament binding with cell-matrix adhesion, cytoskeleton organization, MAP kinase signal transduction of cell growth, survival and differentiation with stimulation of glucose uptake, intracellular protein breakdown and tissue remodeling and mediation of translational repression. These interactions may impact brain morphology associated with learning and memory impairment. The CYFIP1 gene encodes a protein that also interacts with FMRP, the protein coded by the FMR1 gene. It is associated with fragile X syndrome, the most common cause of intellectual disabilities and autism found in families [45,62]. The TUBGCP5 (tubulin gamma complex associated protein 5) generates an interaction with proteins involved with mitotic spindle formation and assembly along with microtubule organization and production of centrosomal proteins. These are involved in centriole duplication and regulation during cell division. They are also associated with chorioretinopathy and microcephaly (e.g., [28,44,55]) as well as ADHD and obsessive-compulsive disorder (OCD) [28]. Seven genes interact directly with the non-imprinted 15q11.2 BP1-BP2 genes including CFHR1 or complement factor H-related protein 1 interacting with complement regulation; SPAST or Spastin, an ATP-dependent microtubule-severing protein involved in movement; SPG20 or Spartin implicated in endosomal trafficking participating in cytokinesis; CFHR3 or complement factor H-related protein involved with complement regulation; MNS1 or meiosis-specific nuclear structural protein 1 controlling meiotic division and germ cell differentiation; IGFBF2 or insulin-like growth factor-binding protein 2 inhibiting IGF-mediated growth and developmental rates with BMPR2 or bone morphogenetic protein receptor 2 [26]. Clinical and brain imaging data support brain disturbances with global morphology and subcortical volume differences in those with the 15q11.2 BP1-BP2 deletion [63]. Significantly lower nucleus accumbens volume and total surface brain area were found along with thicker cortices reported in those with the deletion when compared to individuals without the deletion, typically across the frontal lobe, anterior cingulate and precentral and postcentral gyri regions using brain magnetic resonance imaging. The investigators also measured cognitive function and found lower performance on all tasks in those with the 15q11.2 BP1-BP2 deletion and larger intracranial volume and total surface area were associated with higher performance on nearly all cognitive tasks. Generally, frontal cortex surface regions were found to be associated with task performance, particularly for fluid intelligence and trail-making tasks. In addition, onset of adverse perinatal events and early life outcomes have been examined in pregnancies with 15q11.2 BP1-BP2 anomalies [64,65]. For example, Chu et al. [64] analyzed 1,337 pregnancies with genetic amniocentesis and found that 0.7% of cases had the 15q11.2 BP1-BP2 deletion and 0.8% had a duplication of the same region. They compared the pregnancies with normal microarray results to those with the 15q11.2 BP1-BP2 deletion and more cases with the deletion received neonatal intensive care, Apgar scores less than 7 (at 1 min) and recorded neonatal deaths. Perinatal findings noted in other studies included development delays and more infantile deaths in the deletion group, specifically related to congenital heart disease [66]. Other birth defects reported at birth in those with this deletion include congenital arthrogryposis [67] and tracheoesophageal fistula with congenital cataracts [68]. The published abnormal results associated with the 15q11.2 BP1-BP2 deletion do not fit one molecular dysfunction, but multiple altered functions of the four genes, suggesting involvement in neuro-plasticity, development and function. One important interactive gene involved in this cytogenetic region and is associated with learning and motor delays, autism and schizophrenia is CYFIP1, which encodes a protein involved with actin cytoskeletal dynamics. It interacts with the fragile X mental retardation protein and when disturbed causes fragile X syndrome with abnormal white matter microstructure and postnatal hippocampal neurogenesis microglial disturbances [69,70,71]. For example, Silva et al. [69] applied a brain-wide voxel-based approach and analyzed diffusion tensor imaging data from healthy individuals and those with 15q11.2 BP1-BP2 deletions or duplications. A reciprocal effect was found for 15q11.2 BP1-BP2 on white matter microstructure suggesting reciprocal chromosomal imbalances may lead to opposite changes in brain structure. For example, findings in the deletion group overlapped with more white matter differences previously reported in the fragile X syndrome suggesting common pathogenic mechanisms derived from disruptions of the cytoplasmic CYFIP1–fragile X mental retardation protein complexes. In addition, other interactive genes including solute carrier (SLC) family composed of at least 43 gene families and hundreds of transporters may impact features seen in PWS and Burnside–Butler syndrome [28,52]. This collection of gene interactions may contribute to comorbidities seen in PWS, particularly those with the larger Type I deletion including glucose and insulin metabolic alterations [48]. Identified components of the 15q11.2 BP1-BP2 phenotype and neurobiological mechanisms should stimulate more studies to test similarities between the 15q11.2 BP1-BP2 disorder and in those with PWS having the larger typical 15q11-q13 Type I deletion with these four genes deleted. Bi-allelic NIPA1, NIPA2, TUBGCP5 and CYFIP1 genes located in the proximal 15q11 chromosomal region between breakpoints BP1 and BP2 are implicated in compulsivity, aberrant behavior and lower intellectual ability in individuals with Prader–Willi syndrome with Type I versus Type II deletions. For example, the coefficient of determination for the deletion type alone explained 5 to 50% of the variation in the clinical parameters that were assessed when examining expression of the four genes [43]. In summary, this chromosome deletion has been unequivocally associated with congenital defects, neurobehavior disturbances including autism, schizophrenia, dyslexia or dyscalculia in the vast majority of individuals studied related to these four genes. Additional studies are needed to identify the function of the four genes and their interaction with gene networks to clarify their potential role and include brain gene expression patterns with involved pathways. Parent of origin and gender-based differences should be studied. These may impact fetal phenotype, prognosis and development/behavior/psychiatry with the need for follow-up and long-term care, including echocardiography due to the higher risk of congenital anomalies and heart defects with developmental assessments from disturbances seen in those with the 15q11.2 BP1-BP2 deletion. The clinical presentation and findings seen in Burnside–Butler syndrome (BBS) are also seen in individuals with PWS, particularly those with the larger Type I deletion coined PWS and BBS, with mounting evidence of more severity than is observed in other genetic defects in PWS such as the smaller Type II deletion or maternal disomy 15. The description and explanation of the clinical findings focused on this report and associations could be recognized as a separate category of PWS, particularly in the 25 to 30% of all PWS patients who have the larger Type I deletion with loss of the four described genes and their interactions. These observations could lead to the potential for earlier intervention, treatment and surveillance when applied at a young age, particularly magnesium surveillance and supplementation, if low, with close medical care. More research is requested to pursue and confirm these potentially associated findings with impact on prognosis, longevity and quality of life.

PMC10002223

Mahinbanu Mammadli,Liye Suo,Jyoti Misra Sen,Mobin Karimi

TCF-1 Is Required for CD4 T Cell Persistence Functions during AlloImmunity

21-02-2023

TCF-1,CD4 T cells stemness,CD4 T cell serum level cytokine production,alloimmunity

The transcription factor T cell factor-1 (TCF-1) is encoded by Tcf7 and plays a significant role in regulating immune responses to cancer and pathogens. TCF-1 plays a central role in CD4 T cell development; however, the biological function of TCF-1 on mature peripheral CD4 T cell-mediated alloimmunity is currently unknown. This report reveals that TCF-1 is critical for mature CD4 T cell stemness and their persistence functions. Our data show that mature CD4 T cells from TCF-1 cKO mice did not cause graft versus host disease (GvHD) during allogeneic CD4 T cell transplantation, and donor CD4 T cells did not cause GvHD damage to target organs. For the first time, we showed that TCF-1 regulates CD4 T cell stemness by regulating CD28 expression, which is required for CD4 stemness. Our data showed that TCF-1 regulates CD4 effector and central memory formation. For the first time, we provide evidence that TCF-1 differentially regulates key chemokine and cytokine receptors critical for CD4 T cell migration and inflammation during alloimmunity. Our transcriptomic data uncovered that TCF-1 regulates critical pathways during normal state and alloimmunity. Knowledge acquired from these discoveries will enable us to develop a target-specific approach for treating CD4 T cell-mediated diseases.

TCF-1 Is Required for CD4 T Cell Persistence Functions during AlloImmunity The transcription factor T cell factor-1 (TCF-1) is encoded by Tcf7 and plays a significant role in regulating immune responses to cancer and pathogens. TCF-1 plays a central role in CD4 T cell development; however, the biological function of TCF-1 on mature peripheral CD4 T cell-mediated alloimmunity is currently unknown. This report reveals that TCF-1 is critical for mature CD4 T cell stemness and their persistence functions. Our data show that mature CD4 T cells from TCF-1 cKO mice did not cause graft versus host disease (GvHD) during allogeneic CD4 T cell transplantation, and donor CD4 T cells did not cause GvHD damage to target organs. For the first time, we showed that TCF-1 regulates CD4 T cell stemness by regulating CD28 expression, which is required for CD4 stemness. Our data showed that TCF-1 regulates CD4 effector and central memory formation. For the first time, we provide evidence that TCF-1 differentially regulates key chemokine and cytokine receptors critical for CD4 T cell migration and inflammation during alloimmunity. Our transcriptomic data uncovered that TCF-1 regulates critical pathways during normal state and alloimmunity. Knowledge acquired from these discoveries will enable us to develop a target-specific approach for treating CD4 T cell-mediated diseases. T cell factor-1 (TCF-1, encoded by Tcf7) regulates T cell development, cell fate specification, and maintenance of tissue homeostasis [1]. TCF-1 plays a critical role in T cell responses to viral infection, cancer, and autoimmunity [1,2,3,4]. TCF-1 is a key mediator of both Th1 and Th17 cytokines, and T-bet expression has been shown to be regulated by TCF-1 [5,6]. Studies from germline TCF-1 knock-out mice and enforced expression models have demonstrated that T-bet recruits the transcriptional repressor Bcl6 to the TCF-1 promoter and inhibits TCF-1 gene expression [7]. These data suggest that T-bet can regulate the role of TCF-1 in Th1 differentiation or effector function. A major limitation of these studies was the use of TCF-1 germline knock-out mice or Wnt pathway inhibitors that are not specific. CD4 T cells with a higher TCF-1 expression have been shown to have self-renewing capacity, while CD4 T cells with a lower TCF-1 expression do not, at least in the context of a viral infection [8]. Studies have also shown that both LEF-1 and TCF-1 play a central role in Th2 cell development by suppressing Th2 specific cytokines and the induction of GATA-3 [9]. TCF-1 has been shown to play a critical role in Tfh cells by regulating IL-4/STAT-6 signaling to control the differentiation of CD4 cytolytic cells [9,10,11,12]. TCF-1 has also been shown to inhibit IL-17 during the early stages of development, limiting peripheral Th17 cells [13]. However, to our knowledge, TCF-1 has not been studied in the context of alloimmunity, which is a different process from canonical T cell activation. To study the role of TCF-1 in CD4 T cells in a clinically relevant model, we utilized an allogeneic hematopoietic stem cell transplantation (allo-HSCT) murine model. Allo-HSCT is a curative option in the treatment of aggressive malignant and non-malignant blood disorders [14,15]. However, the benefits of allo-HSCT can be compromised by graft-versus-host disease (GvHD), a prevalent and morbid complication of allo-HSCT. We used a unique mouse strain that has a deletion of TCF-1 in mature T cells, rather than a global deletion [16,17]. The progeny of a TCF-1 flox/flox mouse was crossbred with a CD4 cre+/+ C57BL/6 mouse experiencing a deletion of TCF-1 in all T cells at the double-positive (DP) phase of development when all T cells express CD4 [18]. This approach allowed us to directly investigate the role of TCF-1 on peripheral CD4 T cells [19]. Our compelling data demonstrate that the loss of TCF-1 in CD4 T cells reduces both the severity and the persistence of GvHD, leading to improved survival of recipient mice following transplantation. We also showed that TCF-1 significantly impacts CD4 T cell activation memory formation. Our data also uncovered that TCF-1 differentially impacts chemokine expression, however, donor CD4 T cells from TCF-1 cKO mice had no impact on donor T cell migration to GvHD target tissues. Several lines of evidence have suggested that CD28 is critical for the CD4 T cell persistence function [20,21]. Our data showed that CD4 T cells from TCF-1 cKO mice expresses significantly less CD28 co-stimulatory molecules. For the first time, we provide evidence that TCF-1 regulates CD28 expression in T cells [17]. The loss of CD28 in CD4 T cells has no impact on CD4 T cell initial cytokine production [22]. We uncovered that mature CD4 T cells from TCF-1 cKO mice produce significantly less Th1 cytokines but show increased production of Th2 cytokines in the in-vivo alloimmune disease model, which was previously unknown. Th1 cytokines have been shown to cause donor T cell proliferation/expansion and to significantly amplify the development of GvHD (more than Th2 cytokines) [23,24,25,26]. Our transcriptomic data uncovered that pre- and post-transplanted CD4 T cells from TCF-1 cKO mice have an altered expression of pathways related to apoptosis and cell death, T cell-mediated processes, cytokine production, and cell adhesion. Overall, our data demonstrate that TCF-1 significantly impacts peripheral CD4 T cell phenotype, cytokine production, chemokine expression, and cell survival, and alters the genetic profile in a clinically relevant murine model. Thus, our findings contribute significantly to understanding the role of TCF-1 in CD4 T cells in alloimmunity. Previous research on TCF-1 focused on T cell development and canonical activation [2,27]. These studies primarily used a global TCF-1 knockout, resulting in limited production of mature T cells. We sought to examine the role of TCF-1 in mature T cells, so to overcome this limitation we employed mice with a T-cell-specific deletion of TCF-1 using TCF-1 flox/flox mice bred with CD4 cre+/+ mice [28,29]. This mouse strain has TCF-1 deleted in all CD4 and CD8 T cells at the DP phase, allowing production of mature T cells with a TCF-1 deletion. The CD4 T cell phenotype has been extensively studied in response to viral infections, cancer, and GvHD [30,31,32]. More specifically, CD4 T cell phenotype plays a central role in autoimmune disorders [32,33]. However, whether TCF-1 regulates CD4 T cell activation, effector, and central memory phenotypes in peripheral CD4 T cells is unclear. To investigate whether TCF-1 regulates these critical functions of CD4 T cells, we examined naïve CD4 T cells from either WT C57Bl/6, CD4 cre+/+, or TCF-1 cKO mice. CD4 T cells were isolated from spleens, and we confirmed by flow cytometry the loss of TCF-1 in CD4 T cells (Figure 1A). Next, we examined the expression of activation markers, such as CD44 and CD122 on CD4 T cells. CD4 T cells expressing these markers have been shown to induce significantly less or no GvHD [32,34]. Our data showed that splenic naïve CD4 T cells from TCF-1 cKO mice expressed significantly more CD44 and CD122 markers, compared to CD4 T cells from control mice including WT and CD4 cre+/+ strains (Figure 1B,C). These data uncovered that TCF-1 plays a suppressive role in CD4 T cell activation and the loss of TCF-1 affects the CD4 T cell phenotype [17,35]. We also examined whether CD4 T cells from TCF-1 cKO mice might impact the T-box transcription factor family members Eomes and T-bet [36], which are important in anti-tumor responses of T cells and play central roles in T cell mediated GvHD and alloimmunity [37]. The key transcription factor T-bet is known to regulate the balance between the effector and central memory phenotype of T cells [38,39]. Our data showed that CD4 T cells from TCF-1 cKO mice expressed significantly more T-bet than CD4 T cells from control WT, and CD4 cre+/+ mice (Figure 1D). The data with higher T-bet expressions in TCF-1 cKO mice correlate with higher expressions of CD122 and CD44 in CD4 T cells from TCF-1 cKO mice [40,41,42]. However, we did not observe any differences in Eomes expression in splenic naïve CD4 T cells from WT, compared to TCF-1 cKO mice (Supplementary Figure S1A). Since both effector (EM) and central (CM) memory cells have been shown to play a significant role in CD4 T cell-mediated diseases, including GvHD [32], we wanted to examine the memory phenotype in CD4 T cells from TCF-1 cKO mice. Effector memory cells were defined as CD44high CD62Llow, central memory cells defined as CD44high CD62Lhigh, and naïve CD4 T cells defined as CD44 low CD62Lhigh subgroups. Surprisingly, our data showed that there was a significant increase in both EM and CM cells and significantly fewer naïve cells in TCF-1 cKO mice, compared to WT, and CD4 cre+/+ mice (Figure 1E). These findings shows that TCF-1 does play a significant role in the EM and CM formation in CD4 T cells [17,43,44]. Because ICOS has been shown to play a critical role in effector memory, CD4 T cell survival, and maintenance [45,46,47], we wanted to examine the expression of ICOS in the effector memory of CD4 T cells from TCF-1-deficient and WT, or control and CD4 cre+/+ mice. Our data uncovered that the splenic EM memory of CD4 T cells from TCF-1 cKO mice express ICOS significantly less frequently, compared to CD4 T cells from other control mice, suggesting that even though TCF-1-deficient mice have more effector memory cells, their survival and maintenance may be negatively affected, providing evidence that TCF-1 is required for CD4 T cell persistence functions (Figure 1F). Published data have shown that CXCR3 expression is directly linked to T-bet expression in EM cells during viral infections, and that TCF-1 controls the chemokine receptor expression in CD4 T cells and their trafficking to GvHD target organs [48,49]. A molecular analysis also showed that TCF-1 has been implicated in T cell migration [50,51]. Our data showed that naïve splenic CD4 T cells from TCF-1 cKO mice express CXCR3 significantly more frequently, which suggests that TCF-1 CD4 T cells might migrate to GvHD target organs more easily than CD4 T cells from WT or CD4cre+/+ control mice (Figure 1G). The differences we observed in the phenotype of the CD4 T cells between WT and TCF-1 cKO mice could be cell-intrinsic (due directly to gene deletion within the cell) or cell-extrinsic (due to different microenvironments caused by gene deletion) [52,53]. To determine whether the above-mentioned differences of CD4 T cell phenotypes from TCF-1 cKO mice are cell-intrinsic or cell-extrinsic, we mixed TCDBM from WT mice with the congeneric marker CD45.1 along with bone marrow from TCF-1 cKO mice with the congeneric marker CD45.2 at a 1:4 ratio. These ratios were determine based on our previous publication [54]. To determine whether the phenotypic effects we observed were cell-intrinsic or cell-extrinsic, we developed a chimeric mouse model. Briefly, we mixed bone marrow from WT and TCF-1 cKO mice at a 1:4 (WT:TCF) ratio for a total of 50 × 106 BM cells, then used this mixture to reconstitute lethally irradiated into mice with the congeneric marker Thy.1. to generate a mixed chimera model. Nine to 10 weeks post transplantation, we bled the recipient Thy1.1 mice to confirm the successful reconstitution and creation of a chimera model [17]. Ten weeks post-chimerization, we euthanized recipient Thy1.1 mice and used a flow cytometry analysis to examine TCF-1 expression in the WT by CD45.1 or in the TCF-1cKO by CD45.2 bone marrow-derived CD4+ T cells [17]. Our data confirmed that the loss of TCF-1 in the TCF-1 cKO mice is cell-intrinsic, compared to control groups (Supplementary Figure S2A). However, when we examined mixed bone marrow chimera derived CD4 T cells from our chimera model from either WT or TCF-1 cKO mice developed in the same thymus, we did not observe any differences in either CD122 or CD44 expression (Supplementary Figure S2B,C). Next, we examined whether mixed bone marrow derived CD4 T cells from WT that develop in the same thymus as bone marrow derived CD4 T cells from TCF-1 cKO bone marrow derived model developed in the same thymus, we observed no differences in the EM and CM phenotypes (Supplementary Figure S2D,E). We also observed no difference in T-bet expression among the mixed bone marrow derived CD4 T cells from WT and TCF-1 cKO mice (Supplementary Figure S2F). The increase in expression of CD122, CD44, EM, CM, and T-bet were significantly higher in naïve CD4 T cells from TCF-1 cKO mice (Figure 1B–E). However, in a mixed chimera model we did not see any differences. These data suggested that these activating marker CD122 and CD44 expression in a mixed bone marrow chimera become similar to that found in TCF-1 cKO mice. EM, CM, and T-bet expression also become similar to TCF-1 cKO mice. Altogether, these data showed that TCF-1 controls a number of activation markers and memory formation in a cell-extrinsic way. To investigate whether TCF-1 in mature CD4 T cells contributes to GvHD after allo-HSCT, we employed a murine model of MHC-mismatched allotransplantation. The MHC haplotype mismatch (H2Kb in donors, H2Kd in recipients) results in the alloactivation of the donor T cells, leading to GvHD [35,54,55]. A group of irradiated BALB/c mice were transplanted with 10 × 106 TCDBM cells alone. This group of mice will be considered a control that will not develop GvHD due to the lack of mature T cells. A second group of irradiated BALB/c mice were transplanted with 10 × 106 TCDBM along with 1 × 106 CD4 T cells from WT mice. These groups will develop lethal GvHD, as shown previously [35,43,54,55]. To determine whether CD4cre might contribute to the development of CD4 T cell-mediated GvHD, a third group of irradiated BALB/c mice were transplanted with 10 × 106 TCDBM along with 1 × 106 CD4 T cells from CD4cre mice. Next, we asked whether CD4 T cells from TCF-1Flox/Flox mice with or without CD4cre develop CD4 T cell- mediated GvHD. A fourth group of irradiated BALB/c mice were transplanted with 10 × 106 TCDBM along with 1 × 106 CD4 T cells from TCF-1Flox/Flox mice. Finally, we examined whether CD4 T cells from mice lacking TCF-1 expression specifically on mature T cells develop GvHD. A fifth group of irradiated BALB/c mice were transplanted with 10 × 106 TCDBM along with 1 × 106 CD4 T cells from TCF-1 cKO mice. Recipient BALB/c mice were monitored for survival (Figure 2A) for up to about 70 days. Recipient mice were examined, weighed, and given a GvHD score (Figure 2B,C) to identify the severity of GVHD for up to about 70 days post-transplant. Recipient mice were scored based on weight loss, fur texture, posture, activity, skin condition, and diarrhea, as previously described [17,35,54,55,56,57]. Recipient BALB/c mice receiving donor CD4 T cells from WT C57Bl/6, CD4 cre+/+ or TCF-1Flox/Flox control donor’s cells experienced a rapid increase in GvHD symptoms and peaked at a very high score, indicating very severe disease (Figure 2C). CD4 T cells are known to cause very severe GvHD symptoms, so this finding was expected [35,53,58]. Over time, these recipient mice continued to show severe symptoms, with consistent scores until death from disease (Figure 2C). Survival was also poor, with most mice in this group dying within 25 days of transplantation (Figure 2A). These mice lost weight due to GvHD and died before they were able to regain much weight (Figure 2B). In contrast, recipient BALB/c mice receiving CD4 T cells from TCF-1 cKO C57Bl/6 mice had significantly better survival (Figure 2A), weight gain following an initial weight loss (Figure 2B) and reduced GvHD scores (Figure 2C). Interestingly, recipient BALB/c mice in the TCF-1 cKO CD4 T cell showed a peak in GvHD score early on around day 5, as was seen in the other control group, but peaked at a much lower score. Additionally, this peak score did not persist over time, as the scores for these mice quickly reduced to the level seen in bone marrow-only transplanted controls (Figure 2C). In addition, this low score remained for an extended period, suggesting that the disease had resolved rather than being delayed (Figure 2C). These data indicated that GvHD symptoms were not only less severe in these mice, but also less persistent over time. GvHD involves early migration of alloreactive donor T cells into the target organs, followed by T cell expansion and tissue destruction [23,24]. Modulation of alloreactive T cell trafficking has been suggested to play a significant role in ameliorating experimental GvHD [59]. Therefore, we examined the trafficking of donor T cells to GvHD target tissues, as previously described [60]. To determine whether TCF-1 regulates CD4 T cells trafficking to GvHD target organs, we repeated the short-term experiments, as described. Irradiated BALB/c recipient mice were transplanted with 10 × 106 TCDBM. We mixed CD4 T cells from WT mice with WT B6LY5 (CD45.1+) in a C57Bl/6 background with CD4 T cells from TCF-1 cKO mice with a (CD45.2+) C57Bl/6 background at a 1:1 ratio. Seven days post-transplantation, recipient mice were examined for the presence of donor CD4 T cells in the spleen, lymph nodes, liver, and small intestines. We observed no differences in trafficking of donor CD4 T cells from either WT or TCF-1 cKO mice (Supplementary Figure S2D,E). Chemokines direct cellular infiltration to tissues, and their receptors and signaling pathways represent targets for therapy in multiple disease models, including autoimmunity, cancer, and T cell responses to viral infections [61,62,63,64,65,66]. To determine whether TCF-1 regulates specific chemokine receptors expression, we sorted back donor CD4 T cells from allotransplanted BALB/c mice (using H2Kb to identify donor cells) and performed a qPCR using a 96-well mouse chemokine/chemokine receptor array plate (Thermo Fisher liver pool NY USA). We found that the expression of chemokines and chemokine receptors were upregulated following alloactivation, as expected. However, expression of these markers was consistently higher in TCF-1 cKO CD4 T cells from spleen, both pre- and post-transplant (Figure 3A,B), while these markers were downregulated in TCF-1 cKO CD4 T cells from post-transplanted liver (Figure 3C). These changes also confirmed our observation of CXCR3 expression as increased in splenic cells from TCF-1-deficient mice, compared to WT mice. Our data uncovered that TCF-1 regulates chemokine receptor expression before and after transplantation, with tissue-specific changes. During GvHD, host tissues are damaged by the activity of alloactivated CD4 T cells [53,67,68]. To determine whether damage to the target organs of GvHD (skin, liver, and small intestine) was altered by loss of TCF-1 in donor CD4 T cells, we collected organs from mice allotransplanted, as described above [35,54,55,69]. To induce GvHD, we used MHC-mismatched donors and recipients, with T cell-depleted bone marrow (TCDBM) from WT mice, donor CD4 T cells from either C57BL/6 (B6) WT or TCF-1 cKO mice (MHC haplotype b), and lethally irradiated BALB/c (MHC haplotype d) mice as recipients. Recipient mice were injected intravenously with 10 × 106 wild-type (WT) TCDBM cells along with purified 2 × 106 donor CD4 T cells. To examine the pathological damage to target organs including the liver, small intestine, and skin, tissues from the recipient BALB/c mice and these organs were collected for histology at day 7 and day 21 post-transplantation. Collected organs were fixed, sectioned, hematoxylin and eosin (H&E) stained, and analyzed by a pathologist (L.S.) In the liver, (magnification ×400), much less inflammatory infiltrates in the bile ductal epithelium of the portal triad (black arrows showing the inflammatory cells around interlobular bile ducts) was seen in the recipient mice transplanted with CD4 T cells from TCF-1 cKO cells, compared with recipient mice transplanted with CD4 T cells from WT C57Bl/6 mice, and both euthanized at day 7r: WT (Figure 4A) and TCF-1 cKO (Figure 4B); euthanized at day 21: WT (Figure 4C) and TCF-1 cKO (Figure 4D) post-transplant. In the small intestine (magnification ×400), no apoptotic bodies were seen in the crypts of the small intestine in the recipient mice transplanted with CD4 T cells from TCF-1 cKO C57Bl/6, while many apoptotic bodies with micro abscesses (black arrows and red circle) were present in the small intestine of the recipient mice transplanted with CD4 T cells from WT C57Bl/6 mice and euthanized at day 7: WT (Figure 4E) and TCF-1 cKO (Figure 4F). We observed that fewer apoptotic bodies were present in the small intestine of recipient mice that were transplanted with CD4 T cells from WT C57Bl/6 mice and euthanized at day 21, compared to recipient mice transplanted with CD4 T cells from TCF-1 cKO mice at day 21: WT (Figure 4G) and TCF-1 cKO (Figure 4H). In the skin (magnification ×200), a mild increase of inflammatory cells (red circle) was observed in the dermis of the recipient mice transplanted CD4 T cells from WT C57Bl/6 mice and euthanized on day 7: WT (Figure 4I) and TCF-1 cKO (Figure 4J), and a marked increase of inflammatory cells (red circle) with frequent apoptotic bodies involving both epidermic and dermis was observed in the dermis of the recipient mice transplanted with CD4 T cells from WT C57Bl/6 mice and euthanized at day 21 (Figure 4K), while the dermis of the recipient mice transplanted with CD4 T cells from TCF-1 cKO mice appeared normal at both timepoints (Figure 4L). These findings further support the idea that disease resolves over time and does not persist when recipient mice transplanted with donor CD4 cells from TCF-1 cKO mice. Together, these results indicate that TCF-1 normally contributes to and is indispensable for GvHD damage by T cells, and the loss of TCF-1 reduces its severity and persistence of GvHD. CD4 T cell survival and persistence functions are critical in both health and disease [70]. The importance of CD4 T cell persistence has been shown in autoimmunity, cancer, viral infection, and several cardiovascular diseases [71,72,73]. Therefore, we sought to examine whether TCF-1 is critical for the survival of peripheral CD4 T cells and for their function. We isolated splenocytes from either control mice, including WT C57Bl/6 or TCF-1 cKO mice and performed an in vitro death and apoptosis assay. These CD4 T cells were either stimulated with 2.5 μg/mL anti-CD3 and 2.5 μg/mL anti-CD28 antibodies for 6, 24, 48, or 72 h in culture or left unstimulated, then were stained for apoptosis and death markers. We did not observe any differences in apoptosis, live cell, or dead cell percentages at 0 h between the strains of mice (Figure 5A). CD4 T cells from TCF-1cKO mice that were stimulated for 6 h or for 24 h had more dead cells (annexin V+, near-IR+) and fewer live cells (annexin V-, near-IR-) than cells from the control WT mice (Figure 5B,C). When stimulated for 48 h, CD4 T cells from TCF-1 cKO mice showed more dead (annexin V+, near-IR+) and apoptotic cells (Annexin+, near-IR-) and fewer live cells (annexin V-, near-IR-), compared to CD4 T cells from WT C57Bl/6 mice (Figure 5D). By 72 h post-stimulation, the frequencies of live, apoptotic, and dead cells from the control mice, WT C57Bl/6, and TCF-1 cKO mice were the same (Figure 5E). These data suggested that TCF-1 is critical for early survival and the persistence function of CD4 T cells. CD4 T cell exhaustion has been well documented in CD4 T cell responses to viral infections, and the role of TCF-1 in regulating T cell exhaustion during viral infections is also clear [74,75,76]. PD-1 has also been shown to be critical for CD4 T cell function [77]. Thus, we examined whether TCF-1 regulates PD-1 expression on naïve splenic CD4 T cells from WT C57Bl/6 or TCF-1 cKO mice. Splenocytes from TCF-1 cKO or control mice and WT C57Bl/6 mice were isolated and were either stimulated with 2.5 μg/mL anti-CD3 and 2.5 μg/mL anti-CD28 antibodies for 24, 48, or 72 h in culture or were left unstimulated. These cells were stained for PD-1, Ki-67, and TOX. We did not observe any differences in PD-1 expression before or after 24 or 48 h of stimulation (data not shown), but at 72 h post-stimulation, CD4 T cells from TCF-1 cKO mice expressed more PD-1, compared to CD4 T cells from WT C57Bl/6 mice (Figure 5F). This finding suggested that TCF-1 is a critical regulator of PD-1 expression during the late stages of in vitro activation. To examine whether CD4 T cells become exhausted due to the lack of TCF-1 in alloimmunity, as is shown during T cell responses to viral infection, we examined the Ki-67 expression and TOX expression [78,79], in CD4 T cells from WT C57Bl/6 and TCF-1 cKO mice before and after stimulation. Our data uncovered that there were no differences at any timepoint in expression of Ki-67 and TOX among CD4 T cells from WT and TCF-1 cKO mice (Figure 5H,I). CD28 receptor provides a critical second signal alongside T cell receptor (TCR) ligation for naïve T cell activation. We and others have shown that TCF-1 is critical for TCR stemness [2,17,80]. Published data have also demonstrated that the lack of CD28 significantly weakens TCR stemness [81]. Thus, we examined whether the CD4 T cells from TCF-1 cKO mice also have reduced CD28 expression. We isolated CD4 T cells from either WT or TCF-1 cKO mice. These freshly isolated CD4 T cells were examined for CD28 expression by flow cytometry. Our data uncovered that CD4 T cells from TCF-1 cKO mice have no CD28 expression, compared to that seen in the WT mice (Figure 5J). These findings highlight that TCF-1 has minimal impact on CD4 T cell exhaustion and proliferation in in vivo studies. These findings suggested that TCF-1 regulates CD28 expression required for CD4 stemness. To understand whether TCF-1 might regulate CD4 T cells differently in vitro than in vivo, we transplanted 1 × 106 purified CD4 T cells from either WT or TCF-1 cKO mice into irradiated BALB/c mice to establish an allo-HSCT model, as described above, to assess these changes in vivo. At day 7 post-transplant, recipient BALB/c mice were euthanized and donor H2Kb+ donor CD4 T cells from the liver and spleen were stained for annexin V and near-IR (apoptotic and dead cell markers). Our data uncovered that there were no differences between the strains in live, apoptotic, or dead CD4 T cells coming from the liver or spleen of the recipients (Figure 6A,B). However, our data demonstrated that there were significantly fewer donor H2kb+ CD4 T cells in the spleen and liver of recipients that were transplanted with CD4 T cells from TCF-1 cKO mice, compared to in those given WT cells (Figure 6C,D). We also wanted to determine the expression of Ki-67 and TOX in the in vivo alloactivated CD4 T cells from TCF-1-deficient and WT mice. Again, we did not find any differences in Ki-67 and TOX expression in CD4 T cells from the liver or spleen between the recipients of the two donor strains (Figure 6E–H). These findings highlight that TCF-1 have minimal impact on CD4 T cells exhaustion and proliferation in in vivo studies. These findings suggested that TCF-1 regulates CD28 expression required for CD4 stemness. Production of inflammatory cytokines, eventually culminating in a cytokine storm, is considered a hallmark of CD4 T cell-mediated alloimmunity [25,31,82,83]. Th1 cytokines and cytotoxic mediators are essential for T cells to maintain the GVL effect and kill tumor cells, yet they also lead to the damage of healthy host tissues [84,85,86,87], More specifically INF-γ and TNF-α secretion by donor CD4 T cells are the hall mark of persistence of GvHD mediators [88]. To examine cytokine production by TCF-1 cKO CD4 T cells, we allotransplanted recipient mice, as described above. At day 7 post-transplantation, we also took blood from these recipient mice and obtained serum, which we tested for various cytokines using a LEGEND plex ELISA kit (Biolegend). We uncovered that recipient BALB/c mice transplanted with CD4 T cells from TCF-1 cKO mice had higher serum levels of IFN-γ and TNF-α at day 7 post-transplant than recipient mice transplanted with CD4 T cells from WT mice. However, the level of these proinflammatory cytokines were significantly decreased at day 14 post transplantation (Figure 7A–D). IL-5 and IL-2 have been shown to play critical roles in GvHD [89,90,91]. Thus, we examined whether the levels of IL-2 or IL-5 are impacted by the loss of TCF-1 on mature CD4 T cells. Our data uncovered that recipient BALB/c mice transplanted with CD4 T cells from TCF-1 cKO mice showed increased expression of IL-5 at day 7, compared to recipient mice transplanted with CD4 T cells from WT C57Bl/6 mice. However, at day 14, we did not see significant differences in the serum level IL-5 in recipient mice either transplanted with donor CD4 T cells from WT or TCF-1 cKO mice. Similarly, recipient BALB/c mice transplanted with donor CD4 T cells from TCF-1 cKO mice produced higher serum levels of IL-2 at day 7 post transplantation, however the serum levels dropped at day 14 post transplantation (Figure 7E–H). We observed that recipient mice transplanted with CD4 T cells from TCF-1 cKO mice showed higher serum levels of IL-6 after 7 days post-transplant than mice given WT cells, but the IL-6 expression in these mice later dropped at day 14, such that there was no difference compared to mice given WT cells (Figure 7I,J). These findings provide evidence that recipient BALB/c mice transplanted with CD4 T cells from TCF-1 cKO mice exhibit increased levels of serum cytokines linked to GvHD severity from day 1 to day 14, but these levels reduce to WT levels at day 14 onward. This supports the pattern of resolving the disease severity seen in the in vivo models (Figure 2). Published data have shown that TCF-1 plays a significant role in CD8 T cell-mediated cytokine expression in viral infections [86]. However, the role of TCF-1 in CD4 T cell-mediated Th1 and Th2 cytokine production in alloimmunity has not been defined. Thus, we examined whether TCF-1 regulates CD4 T cell-mediated Th2 cytokines in an allogeneic transplant model. Our data showed that recipient BALB/c mice transplanted with CD4 T cells from TCF-1 cKO mice expressed increased levels of IL-4 and IL-13 at day 7 post transplantation, compared to mice given WT cells. However, levels of both IL-4 and IL-13 in mice given TCF-1 cKO cells decreased at day 14, compared at day 7, while we did not observe any difference in IL-4 and IL-13 levels in between day 7 and day 14 in WT CD4 T cell recipients (Figure 7K–N). This was correlated with the GvHD scores of the same recipient mice as well, suggesting less severe and less persistent GvHD in TCF-1-deficient CD4 T cell transplanted mice. Recipient BALB/c mice transplanted with either TCF-1 cKO or WT cells did not show any differences in IL-10, IL-9, IL-17A, and IL-17F expression (Supplementary Figure S4A–H). These data suggest that allotransplanted TCF-1 cKO CD4 T cells are more activated early in the response but are less active (or less present) later, suggesting a unique mechanism for how TCF-1 modulates cytokine responses during alloimmunity. We also wanted to determine the cellular cytokine production from the donor cells to correlate with the cellular and serum levels of cytokines at day 7 post-transplant. Splenocytes were obtained from recipient mice and were restimulated to induce cytokine production. Cells were restimulated in culture with anti-CD3/anti-CD28 or left unstimulated for 6 h at 37 °C, and Golgiplug was included in the culture media. Then, cells were stained for H2Kb, CD3, CD4, TNF-α, and IFN-γ markers. Our data showed that the production of TNF-α or IFN-γ did not appear to be affected by the loss of TCF-1 (Figure 7P). The difference between cellular cytokine production and serum levels could be due to the lower numbers of CD4+ H2kb+ donor cells present at day 7 post-transplant in the liver and spleen for mice given TCF-1 cKO cells (Figure 6C,D). To understand the molecular mechanisms behind the changes we saw in the TCF-1 cKO donor CD4 T cells, and to understand the role of TCF-1 in regulating gene expression in these cells, we employed RNA sequencing. We allotransplanted recipient BALB/c mice with WT or TCF-1 cKO CD3 T cells, as described above. FACS-sorted pre-transplant samples (Pre-Tx) of CD4+ donor cells from WT and TCF-1 cKO mice were taken and stored in TRizol. At day 7 post-transplant, donor T cells were sorted back from the spleen of recipients using H2Kb, CD3, CD4, and CD8 (Post-Tx samples). A principal component analysis (PCA) of pre-transplant and post-transplant samples showed two clusters of samples, WT and TCF-1 cKO, that were clearly separated by principal component 1 (PC1 46% for Pre-Tx, 53% for post-Tx), which suggests that the transcriptomic profile of the TCF-1 cKO CD4 T cells differs from the CD4 T cells from WT mice (Figure 8A,B). Further analysis of pre-transplanted samples identified 812 differentially expressed genes (DEGs, defined by FDR < 0.05 and Log FC = 1) in TCF-1 cKO cells, compared to WT cells, of which 220 were downregulated and 592 were upregulated in CD4 T cells from TCF-1 cKO mice (Figure 8C). We identified 839 DEGs (defined by FDR < 0.05) in post-transplanted CD4 T cells from TCF-1 cKO cells, compared to CD4 T cells from WT mice, of which 356 were downregulated and 483 were upregulated in CD4 T cells from TCF-1 cKO mice (Figure 8D). All DEGs were plotted in the heatmap for pre- and post-transplanted samples, and by using the Spearman correlation method, which is associated with hierarchical clustering, the pre-transplanted and post-transplanted samples were categorized into two clusters (WT and TCF-1 cKO). Gene co-regulation was determined by hierarchical clustering by using the Pearson correlation method with a grouping cutoff (k) of two (Figure 8E,F). Module 2 shows all of the upregulated DEGs and module 1 shows all of the downregulated DEGs in pre-transplanted and post-transplanted samples (Figure 8E,F). A gene ontology (GO) analysis of the pre-transplanted and post-transplanted samples revealed that all of the identified DEGs are involved in a number of biological processes, such as cell death, apoptotic processes, T cell-mediated processes, T cell functions, cytokine production, and response to cytokines, among others. We clustered the pathways that related to the cell death and apoptotic processes in a group and listed them based on the adjusted p-value (FDR) for both pre- and post-transplanted samples (Supplementary Figures S2 and S3). Genes that were involved in each pathway are also listed in the tables (Supplementary Figures S2 and S3). The majority of genes involved in cell death and apoptotic processes were upregulated in the pre- and post-transplanted samples in CD4 T cells from TCF-1 cKO samples, compared to CD4 T cells from WT samples (Figure 8G,H). Interestingly, the β-catenin gene (encoded by Ctnnb1) was significantly upregulated in pre-transplanted samples from donor CD4 T cells from TCF-1 cKO mice, and the WNT-4 gene was significantly upregulated in post-transplanted samples in CD4 T cells from TCF-1 cKO mice, suggesting a compensatory mechanism of upregulation of the β-catenin pathway in the absence of TCF-1 (Figure 8G,H). Even though most of the genes shared were in the cell death and apoptosis-related pathways, some of the genes were unique for each pathway (Supplementary Figures S2 and S3). We also clustered the pathways that related to T cell function and signaling in a group and listed them based on the adjusted p-value (FDR) for both pre- and post-transplanted samples (Supplementary Figures S2 and S3). For pre-transplant samples, even though the majority of genes involved in T cell function and signaling were upregulated in CD4 T cells from TCF-1 cKO mice, compared to CD4 T cells from WT mice, interestingly, LAT, LCK, ZAP70, and CD3e genes (which are downstream of the TCR) were downregulated in CD4 T cells from TCF-1 cKO mice, compared to CD4 T cells from WT mice (Figure 8I). When we analyzed the T cell function and signaling-related genes in post-transplanted samples, we observed that most of the genes were downregulated in CD4 T cells from TCF-1 cKO mice than in CD4 T cells from WT mice (Figure 8J), which suggests that alloactivated CD4 T cells from TCF-1 cKO mice having attenuated T cell signaling and T cell responses, compared to T cells from WT mice. Even though most of the genes were shared between the T cell function and signaling-related pathways, some of the genes were unique for each pathway (Supplementary Figures S2 and S3). Pre-transplanted CD4 T cells from TCF-1 cKO mice showed upregulation of the genes involved in cytokine production and cell response to cytokines (Figure 8K, Supplementary Figure S2) compared to CD4 T cells from WT mice. Post-transplanted CD4 T cells from TCF-1 cKO mice showed downregulation of a number of genes that were involved in cytokine production and responses, such as Ifitm1, JAk3, CD4, CD28, Iκβ, IκG, CD3e, and others, which supported the observed decrease in cytokine production from CD4 T cells from TCF-1 cKO mice (Figure 8L). We also observed that a number of genes involved in chemokine receptor signaling and cell adhesion, such as CCL5, CCL3, CCL4 CXCR5, and CXCR3, are downregulated, and Slit2 is upregulated in CD4 T cells TCF-1 cKO mice, compared to CD4 T cells from WT mice (Figure 8L, Supplementary Figure S3). It has been shown that Slit2 blocks CXCL12/CXCR4-mediated functional effects in T cells, which is important for HIV infection and viral replication. Altogether, the transcriptomic analysis revealed that TCF-1 regulates the CD4 T cell genetic profile, with a loss of TCF-1 directing the cell towards decreased T cell signaling, decreased cytokine and chemokine signaling, and increased apoptosis and cell death, specifically after allotransplantation. T cell factor-1 (TCF-1) is a T cell transcription factor that is known to be critical for T cell development, activation, and in some cases, responses to pathogens [1,92,93]. The functional and development role of TCF-1 has been extensively studied in CD8 T cell responses to viral infections [2,17,18,76,94,95]. To some extent, the role of TCF-1 has been examined in CD4 T cell development [5,8,11], however, it is unclear whether TCF-1 may regulate alloactivated CD4 T cells during responses to alloantigens. The main significance of our results is that we utilized a clinically relevant model of allo-HSCT, enabling us to study all of the major CD4 T cell functions, as well as phenotypes, clinical outcomes, and gene expression, in a single model. Several publications, including our own, have shown that CD4 T cells with higher CD44, CD122, and Eomes, or T-bet referred to as the innate memory phenotype (IMP) [54,96] T cells with the IMP significantly delayed the development of GvHD, but were able to clear tumors [35,54,55,97]. Our data also showed that CD4 T cells from TCF-1 cKO mice have a significantly higher expression of a IMP phenotype, suggesting the TCF-1 might regulate the IMP phenotype and TCF-1 is considered a repressor factor for IMP cells. Memory phenotypes have been reported to play a significant role in the induction (or lack thereof) of GvHD [44]. Our data showed that CD4 T cells from TCF-1 cKO mice upregulate the effector or central memory phenotypes, and these mice show a decreased naïve cell population, which suggests that TCF-1 regulates the memory formation of CD4 T cells. We and others have shown that the upregulation of EM and CM plays a central role in GvHD development. Studies have shown that T cells with a higher EM and CM do not cause GvHD. [35,35,44,54]. Thus, our studies are of great importance because they show that TCF-1 regulates both EM and CM on CD4 T cells. Another molecule that is critically important for effector memory cell survival and maintenance is ICOS [44,45,98]. The lower expression in ICOS on effector memory CD4 T cells from TCF-1 cKO mice suggest that even though TCF-1-deficient CD4 T cells have more effector memory cells, their survival and homeostasis may be affected by the loss of TCF-1. Modulation of alloreactive CD4 T cell trafficking has been suggested to play a significant role in ameliorating experimental GvHD [99]. However, we did not observe any differences in the donor CD4 T cell migration to GvHD target organs, including the liver and small intestine. Pro-inflammatory conditioning treatment may promote T cell migration into GvHD target tissues [100,101]. Donor CD4 T cells upregulate the chemokine receptor expression upon alloactivation, which mediates donor T cells migration to the site of inflammation [102]. Since chemokine receptor expression in T cells is central to several T cell-mediated diseases [43,103], determining whether TCF-1 regulates chemokine expression either positively or negatively is critically important. We uncovered that mature splenic CD4 T cells from TCF-1 cKO mice expressed higher levels of chemokine receptors than CD4 T cells from WT mice, both pre- and post-allo-HSCT. However, our data showed that CD4 T cells from the liver in mice given TCF-1 cKO cells showed reduced chemokine receptor expression post-transplantation. Our data showed that CD4 T cells from TCF-1 cKO mice caused significantly less tissue damage. All of our data suggest that CD4 T cells from TCF-1 cKO mice can migrate to GvHD target organs, but also provided stronger evidence that TCF-1 is critical for CD4 T cell stemness, however due to the loss of TCF-1 on CD4 T cells, these CD4 T cells are unable to cause persistence of GvHD symptoms. TCF-1 has been shown to be significantly important in T cell development and survival [104], so we also examined whether TCF-1 is critical for CD4 T cell survival in both in vitro and in vivo models. Our data showed that CD4 T cells from TCF-1 cKO mice developed more rapid cell death and apoptosis in vitro within the first 48 h. We observed increased PD-1 expression on TCF-1 cKO cells versus WT cells only at 72 h of in vitro culture, indicating that the cells that survived after 72 h might cause less severe CD4 T cell-mediated diseases [77,105]. Even though we did not observe any differences in dead, apoptotic, or live cell percentages in in vivo alloactivated CD4 T cells from TCF-1 cKO or WT mice, the frequency of donor CD4+ H2kb+ T cells from TCF-1 cKO mice was significantly less in GvHD target organs, thus supporting our central hypothesis that TCF-1 is indispensable for CD4 T cell stemness. However, our recent findings suggested that TCF-1 is dispensable for anti-tumor response [17]. TCF-1 has been shown to play a critical role in CD4 T cell exhaustion and activation in responding to viral infections [75,76]. Our data, both in vivo and in vitro, showed that activated CD4 T cells from TCF-1 cKO mice had no change in exhaustion. These findings demonstrate that alloactivated CD4 T cells are functioning significantly different to how CD4 T cells from TCF-1 cKO mice function in response to viral infections. Pro-inflammatory cytokine production by donor cells and host tissues causes damage to nearby healthy host cells [25,84,87]. Our data showed that during initial activation, donor CD4 T cells from TCF-1 cKO mice produce more serum level cytokines, but these drop significantly over time. These findings suggest that despite early increased activation, cytokine production by donor CD4 T cells from TCF-1 cKO mice donor cells quickly reduce post-transplant, allowing the disease to resolve, further confirming that TCF-1 is required for CD4 T cell persistence functions, including cytokine production. This supports our hypothesis that donor CD4 T cells from TCF-1 cKO mice become exhausted and stop proliferating and producing cytokines, allowing for the resolution of the usual persistent disease state. Even though, we initially (day7) observed significant increases in proinflammatory cytokines, including TNF-a and IFN-g from CD4 T cells from TCF-1 cKO mice, this upregulation of TNF-a and IFN-g fails to recruit other inflammatory cells, such as macrophages to the site of inflammation to induce GvHD. These findings suggest that the loss of TCF-1 significantly weakens CD4 T cell persistence during GvHD by the loss of CD28 on CD4 T cells. These findings are supported by our recent publication that the loss of TCF-1 significantly weakens TCR signaling on CD8 T cells [17]. Our transcriptomic data uncovered that TCF-1 regulates several pathways that are critical for CD4 T cell-mediated diseases. Cell death pathways play a central role in CD4 T cell-mediated diseases, including autoimmunity and cancer [17]. We uncovered that TCF-1 significantly and differentially regulates cell death and apoptotic process-related pathways before and after transplantation, which is critical to understand the role of TCF-1 in alloimmunity and fighting infection and cancer. Our data showed that TCF-1 significantly regulates genes in CD4 T cell programmed cell death, including pathways for apoptotic signaling, necrotic signaling and mitochondrial fragmentation. CD4 T cell activation, signaling proliferation, and Th1/Th2/Th17 differentiation are central to both CD4 T cell function and CD4 T cell-mediated diseases, including CD4 T cell responses to viral infection, autoimmunity, cancer, and aging [106]. We also observed that TCF-1 differentially regulates sets of genes in the I-κβ and NF-κβ pathways. This information will enable us to develop target specific approaches to design therapeutic interventions. CD4 T cells primarily function as regulators of other immune cells either through secreted cytokines or by direct cell–cell contact. Inflammatory and anti-inflammatory cytokines production are central to T cell responses to viral infections, autoimmune disorders, cancer, and GvHD [30]. Our data uncovered that CD4 T cells from TCF-1 cKO mice have significantly higher β-catenin expression before transplantation and higher WNT4 expression after transplantation, compared to cells from WT mice. Another key finding of this report is that TCF-1 regulates CD28 expression. CD28 is a key co-stimulatory receptor that plays a central role in T cell receptor stemness [2,80]. Published data has also demonstrated that the lack of CD28 significantly weakens TCR stemness [81]. Therefore, both our in vivo and in vitro data demonstrated that CD4 T cells from TCF-1 cKO mice are prone to activation and cell death. These findings are consisting with transcriptomic data and the development of apoptosis. These findings are also supported our GvHD studies that CD4 T cells from TCF-1 cKO mice showed peak GvHD clinical scores, but that this significantly diminished over time. Overall, our data uncovered several novel discoveries regarding how TCF-1 differentially regulates CD4 T cell functions, at baseline and during alloactivation. More significantly, how TCF-1 functions during T cell development and on mature peripheral CD4 T cells was not previously known for an alloactivation context. These discoveries will enable us to design target specific approaches in treating CD4 T cell-mediated diseases and alloimmunity. Limitations of the Study: Currently, the limitation of this study is the use of a mouse model. We are working with structural and medicinal chemists to make specific activators for Wnt/β-catenin pathways. Currently available reagents are Wnt3 ligands [107] or GSK3β-inhibitors. The primary problems with these activators are that either T cells become over activated or there is non-specific activation of several other signaling proteins. Therefore, we are currently working to develop our own specific activators. Thy1.1 (B6.PL-Thy1a/CyJ, 000406), B6-Ly5 (CD45.1+, AKA “WT” or B6.SJL-Ptprca Pepcb/BoyJ, 002014), and BALB/c mice (CR:028) were purchased from Charles River or Jackson Laboratory. TCF-1 cKO mice (Tcf7 flox/flox cross bred with CD4cre) [108] were obtained from Dr. Jyoti Misra Sen at the NIH and bred in our facilities. CD4cre (022071). Using genomic PCR, we confirmed that our newly generated mice are TCF-1 cKO. Eight–12-week-old and sex-matched mice were used for all experiments. Recipient mice for transplant experiments were female BALB/c mice (CR:028 from Charles River, age 8 weeks or older). Recipient mice for the chimera experiments were Thy1.1 mice (B6.PL-Thy1a/CyJ, 000406 from Charles River). Animal maintenance and experimentation were approved by the Upstate Medical University IACUC committee with IACUC #433. All mice used for transplants were female, and flow cytometry experiments were carried out with both male and female mice. Donor mice were genotyped using PCR. Ear punches were taken from each mouse at 4 weeks of age, DNA was extracted, and run in a PCR reaction using the Accustart II mouse genotyping kit (95135-500 from Quanta Biosciences). Standard PCR reaction conditions and primer sequences from Jackson Laboratories were used for Eomes, T-bet, and CD4 cre+/+. For TCF-1, primer sequences and reaction conditions were obtained from Dr. Jyoti Misra Sen of NIH. Primers used for CD4 cre+/+ genotyping are: Common primer: 5′-GTTCTTTGTATATATTGAATGTTAGCC; WT reverse primer: 5′-TATGCTAGGACAAGAATTGACA; and Mutant reverse primer: 5′-CTTTGCAGAGGGCTAACAGC. PCR conditions: Step 1. 94 °C for 2:00 min; Step 2. 94 °C, 20 s; Step 3. 65 °C, 15 s; Step 4. 68 °C, 10 s; Step 5. Go to Step 2, 10×; Step 6. 94 °C, 15 s; Step 7. 60 °C, 15 s; Step 8. 72 °C, 10 s; Step 9. Go to Step 6, repeat 28×; Step 10. 72 °C, 2 min; Step 11. 10 °C, infinite hold. Primers used for TCF-7 genotyping: Forward primer: 5′-AGCTGAGCCCCTGTTGTAGA, Reverse primer #1: 5′-TTCTTTGACCCCTGACTTGG, Reverse primer #2: 5′-CAACGA GCTGGGTAGAGGAG. PCR conditions for TCF-7 are: Step 1. 94 °C, 2 min; Step 2. 55 °C, 30 s; Step 3. 72 °C, 1 min; Step 4. Go to Step 2. repeat 38×; Step 5. 72 °C, 10 min, 12 °C infinite hold. For phenotyping experiments, splenocytes were obtained from WT C57Bl/6 and CD4 cre+/+ C57Bl/6 control mice and TCF-1 cKO mice. For all other experiments, cells were obtained from transplanted recipients. Cells were incubated with RBC lysis buffer (00-4333-57 from eBioscience) to remove red blood cells when necessary. Following processing, cells were stained in MACS buffer (1× PBS with EDTA and 4 g/L BSA) with extracellular markers and were incubated for 30 min on ice. Cells were then washed and run on a BD LSRFortessa flow cytometer to collect data. If intracellular markers were used, cells were washed after extracellular staining, then fixed overnight using buffers from the Fix/Perm Concentrate and Fixation Diluent from FOXP3 transcription factor staining buffer set (eBioscience cat. No. 00-5523-00). The following day, cells were washed in Perm buffer from the same kit and were stained with intracellular markers in Perm buffer for 40 min at room temperature. Stained cells were resuspended in FACS buffer (eBioscience cat. No. 00-4222-26) and transferred to flow tubes. All antibodies were used at 1:100 dilution and were purchased from Biolegend or eBioscience. The cells were then washed and run on a BD LSRFortessa. For cell sorting, cells were stained in the same manner and run on a BD FACSAria, equipped with cold-sorting blocks. Cells were sorted into sorting media (50% FBS in RPMI) for maximum viability, or TRizol for RNAseq/qPCR experiments. All flow cytometry data was analyzed using FlowJo software v9 (BD). Depending on the experiment, the antibodies used were: anti-CD4 (FITC, PE, BV785, BV510, BV421), APC, PerCP, Pacific Blue, PE/Cy7), anti-CD3 (BV605 or APC/Cy7), anti-H2Kb-Pacific Blue, anti-H2Kd-PE/Cy7, anti-CD122 (FITC or APC), anti-CD44 (APC, PercP or Pacific Blue), anti-CD62L-APC/Cy7, anti-ICOS-PE, anti-CXCR3-Percp-Cy5.5, anti-TNF-α-FITC, anti-IFN-γ-APC, anti-Eomes (AF488 or PE/Cy7), anti-T-bet-BV421, anti-CD45.2-PE/Dazzle594, anti-CD45.1-APC, anti-Ki67 (PE or BV421), anti-PD-1-BV785, anti-annexin V-FITC, LIVE/DEAD near IR, and anti-TOX-APC. For GvHD experiments, we utilized the MHC-mismatch mouse model of allo-HSCT (WT C57Bl/6 ➔ BALB/c, i.e., H2Kb+ ➔ H2Kd+). BALB/c recipient mice were irradiated twice with 400 cGy x-rays (total dose 800 cGy), with a rest period of at least 12 h between doses, and 4 h of rest prior to transplantation. Lethally irradiated BALB/c mice were transplanted with 10 × 106 T cell-depleted bone marrow (TCDBM) cells. Briefly, bone marrow T-cells were depleted using CD90.2 MACS beads (130-121-278 from Miltenyi) and LD columns (130-042-901 from Miltenyi) [35,54]. T cells CD4+ were separated from donor mice spleens using CD90.2 or CD4 microbeads and LS columns (Miltenyi, CD4: 130-117-043, CD90.2: 130-121-278, LS: 130-042-401) [17,35]. We used purified donor CD4 T cells from WT, CD4cre, or Tcf7 flox/flox mice as controls and TCF-1 cKO mice as the experimental group. The cells were then IV injected into the tail vein in PBS. The recipient mice received 1 × 106 T cells per mouse, along with 10 × 106 T cell-depleted bone marrow cells (TCDBM) collected from WT C57BL/6J mice. Recipient mice were evaluated for clinical signs of GvHD and weight loss for more than 70 days [17,35]. Clinical presentation of the mice for each experiment was assessed 2–3 times per week by a scoring system that summed the changes in 6 clinical parameters: diarrhea, posture, activity, fur texture, weight loss, and skin integrity (Cooke et al., 1996). Mice were euthanized when they lost ≥ 30% of their initial body weight [17,35,56]. Lethally irradiated BALB/c mice were transplanted, as described above. At day 7 or day 14, the recipients were euthanized, and serum, spleen, small intestine, skin, or liver was collected, depending on the experiment. For the mixed bone marrow chimera experiments, a 1:4 ratio of WT (CD45.1) to TCF-1 cKO (CD45.2) bone marrow was injected into Thy1.1 mice, and the mice were rested for 9 weeks. This ratio was chosen based on our previous publication [54]. At 9 weeks post-transplant, tail vein blood was collected and stained with anti-CD45.1 and anti-CD45.2 to detect the two donor cell types. At 10 weeks, these chimeras were euthanized, and their spleens were processed and stained for phenotyping by flow cytometry. Mice were short-term transplanted, as described above (1 × 106 donor CD4 T cells and 10 × 106 TCDBM) cells, and at day 7, recipient mice were humanely euthanized. Splenocytes were obtained from pre- and post-transplanted mice, and FACS sorted, as described above to obtain donor cells. These cells were all sorted into TRizol, then RNA was extracted using chloroform phase separation protocols. The extracted RNA was eluted using the Qiagen RNAEasy Minikit (74104 from Qiagen Germantown MD USA) and run on a spectrophotometer to determine concentration. RNA was then converted to cDNA with the Invitrogen Super-Script IV First Strand Synthesis System kit (18091050 from Invitrogen) and run on a spectrophotometer to determine the concentration. The master cocktail, including 10 ng/μL cDNA and Taqman Fast Advanced Master Mix (4444557 from Invitrogen), was prepared for each sample, and 20μL was added to each well of a 96 well custom TaqMan Array plate with chemokine/chemokine receptor primers (Thermo Fisher, Mouse Chemokines & Receptors Array plate, 4391524). The plates were run on a Quantstudio 3 thermocycler, according to manufacturer’s instructions for the TaqMan assay, and data were analyzed using the Design and Analysis software v2.4 (provided by Thermo Fisher). Five separate recipient mice were sorted, and cells were combined to make one sample for qPCR testing per condition/organ. Lethally irradiated recipient mice were transplanted with 10 × 106 TCDBM) cells, 1 × 106 WT CD4+ T donor cells and at day 7 and day 21 post-transplant, the organs were removed from WT or TCF-1 cKO T cell-transplanted recipient mice. The spleen, liver, small intestine, and skin from the back and ear were removed and fixed in 10% neutral buffered formalin. Tissue pieces were sectioned and stained with hematoxylin and eosin (H&E) by the Histology Core at Cornell University. Stained slides were then imaged and analyzed by a pathologist at SUNY Upstate (L.S.) who was blinded to the study conditions and slide identity. To isolate the lymphocytes from the liver, they were perfused with 5–10 mL of ice-cold PBS to remove red blood cells (RBCs) before removal. Livers were then ground through a 70 mm filter with PBS, centrifuged to remove debris, and lymphocytes were isolated by a 22-min spin in 40% Percoll in RPMI/PBS (22 °C, 2200 rpm, no brake, no acceleration). Isolated lymphocyte pellets were washed, cells were briefly incubated with red lysis buffer to remove remaining RBCs, and resuspended with PBS or MACS buffer (BSA in PBS). To isolate lymphocytes from the small intestine, the intestine was removed and put into an ice cold MACS buffer, opened lengthwise, washed with MACS, and epithelial cells were stripped off by a 30 min shaking incubation (37 °C) in strip buffer (1× PBS, FBS, EDTA 0.5 M, and DTT 1 M). The guts were then cut into small pieces and digested by a 30 min shaking incubation (37 °C) in digestion buffer (collagenase, DNAse, and RPMI). The tubes were then vortexed, and liquid and solid gut pieces were filtered through a 70 mm filter to obtain a cell suspension. Percoll was then used to isolate lymphocytes, as was carried out for liver, with no RBC lysis afterwards. The gut cells were then placed in MACS buffer for further use. Mice were short-term transplanted, as described above (1 × 106 donor CD4 T cells), and at day 7 post-transplant, recipient mice were humanely euthanized and splenocytes were obtained. The splenocytes were cultured for 6 h at 37 °C and 7% CO2 with GolgiPlug (1:1000) and PBS or anti-CD3 (1 μg/mL)/anti-CD28 (2 μg/mL) to restimulate them. Then, after 6 h, the cells were removed from the culture, stained for surface markers, fixed and permeabilized, then stained for the cytokines IFN-g and TNF-a using the BD Cytokine Staining kit (555028), and run on a flow cytometer. Mice were short-term transplanted, as described above with 10 × 106 TCDBM) cells, along with 1 × 106 WT CD4+ T donor T cells. At day 7, the recipient mice were euthanized, and serum was obtained from cardiac blood. The serum was collected from recipient mice in the cytokine experiment and analyzed using the Biolegend LEGENDplex Assay Mouse Th Cytokine Panel kit (741043) [35,54]. This kit quantifies serum concentrations of: IL-2 (T cell proliferation), IFN-g and TNF-a (Th1 cells, inflammatory), IL-4, IL-5, and IL-13 (Th2 cells), IL-10 (Treg cells, suppressive), IL-17A/F (Th17 cells), IL-21 (Tfh cells), IL-22 (Th22 cells), IL-6 (acute/chronic inflammation/T cell survival factor), and IL-9 (Th2, Th17, iTreg, Th9—skin/allergic/intestinal inflammation). Data were collected on a BD LSR Fortessa cytometer, and data were analyzed using the LEGENDplex software (provided with kit via Biolegend). We obtained splenocytes from WT and TCF-1 cKO mice and either activated them with 2.5 μg/mL anti-CD3 (Biolegend #100202) and anti-CD28 antibodies (Biolegend #102115) for 6, 24, 48, or 72 h in culture, or left them unstimulated. Annexin V-FITC (V13242 from Invitrogen) and LIVE/DEAD near IR (L34976 from Invitrogen) were used to identify dead (Ann.V+NIR+), live (Ann.V-NIR-), and apoptotic (Ann.V+NIR-) T cells Mice were short-term transplanted, as described above (1 × 106 CD4 donor T cells), and at day 7 post-transplant, cells from the spleen and liver were stained with annexin V-FITC (V13242 from Invitrogen) and LIVE/DEAD near IR (L34976 from Invitrogen). Annexin V and NIR were used to identify dead (Ann.V+NIR+), live (Ann.V-NIR-), and apoptotic (Ann.V+NIR-) cells. Donor T cells were identified by H2Kb, CD3, and donor CD4 T cells. Freshly isolated CD4T cells were FACS sorted in TRizol for pre-transplanted (Pre-Tx) samples. For post-transplanted (post-Tx) samples, recipient mice were transplanted with 1 × 106 CD3 donor T cells, and at day 7 post-transplant, donor CD8 T cells were FACS-sorted back from the recipient spleen of TCF-1 cKO and WT transplanted mice and sorted into TRizol. RNA was extracted from all of the pre- and post-transplanted samples and prepped by the Molecular Analysis Core (SUNY Upstate, https://www.upstate.edu/research/facilities/molecular-analysis.php (accessed on 21 November 2022)). Paired end sequencing was carried out with an Illumina NovaSeq 6000 system at the University at Buffalo Genomics Core (http://ubnextgencore.buffalo.edu (accessed on 21 November 2022)). The statistical computing environment R (v4.0.4), the Bioconductor suite of packages for R, and Rstudio (v1.4.1106) were used for transcriptomic analysis. Kallisto (version 0.46.2) was used for transcript abundance determination and performing the pseudoalignment. Calculated transcript per million (TPM) values were normalized and fitted to a linear model by the empirical Bayes method with the Voom and Limma R packages to identify differentially expressed genes (DEGs) for both pre- and post-transplanted samples. For pre-transplant samples, DEGs filtered by adjusted p-value (FDR) < 0.05, log fold change = 1, and for post-transplant samples by adjusted p-value (FDR) < 0.05. DEG’s were used for hierarchical clustering and heatmap generation in R. A gene ontology enrichment analysis was conducted using the g: Profiler toolset; g:GOSt tool. Data will be deposited in the Gene Expression Omnibus (GEO) database for public access (https://www.ncbi.nlm.nih.gov/geo (accessed on 21 November 2022)). With accession number GSE204747. All numerical data are reported as means with standard deviation unless otherwise noted. Data were analyzed for significance with GraphPad Prism v9. Differences were determined using one-way ANOVA and Tukey’s multiple comparisons tests for three or more groups, or with a Student’s t-test when only two groups were used. A Kaplan–Meier survival analysis was used for survival experiments. All tests were two-sided. p-values less than or equal to 0.05 were considered significant. All transplant experiments were carried out with N = 3–5 mice per group and repeated at least twice. Ex vivo experiments were carried out two to three times unless otherwise noted with at least three replicates per condition each time. RNAseq was carried out once with three replicates per group. qPCR was completed once with one sample per condition, and 5 mice combined to make the one sample. Sorting was carried out once for each of these two experiments, and data were recorded for Figure 3 and Figure 8. Data in the figures are presented as mean and SD unless otherwise noted.

PMC10002226

Patricia Reboucas,Carine Fillebeen,Amy Botta,Riley Cleverdon,Alexandra P. Steele,Vincent Richard,René P. Zahedi,Christoph H. Borchers,Yan Burelle,Thomas J. Hawke,Kostas Pantopoulos,Gary Sweeney

Discovery-Based Proteomics Identify Skeletal Muscle Mitochondrial Alterations as an Early Metabolic Defect in a Mouse Model of β-Thalassemia

23-02-2023

thalassemia,iron,skeletal muscle,proteomics,mitochondria

Although metabolic complications are common in thalassemia patients, there is still an unmet need to better understand underlying mechanisms. We used unbiased global proteomics to reveal molecular differences between the th3/+ mouse model of thalassemia and wild-type control animals focusing on skeletal muscles at 8 weeks of age. Our data point toward a significantly impaired mitochondrial oxidative phosphorylation. Furthermore, we observed a shift from oxidative fibre types toward more glycolytic fibre types in these animals, which was further supported by larger fibre-type cross-sectional areas in the more oxidative type fibres (type I/type IIa/type IIax hybrid). We also observed an increase in capillary density in th3/+ mice, indicative of a compensatory response. Western blotting for mitochondrial oxidative phosphorylation complex proteins and PCR analysis of mitochondrial genes indicated reduced mitochondrial content in the skeletal muscle but not the hearts of th3/+ mice. The phenotypic manifestation of these alterations was a small but significant reduction in glucose handling capacity. Overall, this study identified many important alterations in the proteome of th3/+ mice, amongst which mitochondrial defects leading to skeletal muscle remodelling and metabolic dysfunction were paramount.

Discovery-Based Proteomics Identify Skeletal Muscle Mitochondrial Alterations as an Early Metabolic Defect in a Mouse Model of β-Thalassemia Although metabolic complications are common in thalassemia patients, there is still an unmet need to better understand underlying mechanisms. We used unbiased global proteomics to reveal molecular differences between the th3/+ mouse model of thalassemia and wild-type control animals focusing on skeletal muscles at 8 weeks of age. Our data point toward a significantly impaired mitochondrial oxidative phosphorylation. Furthermore, we observed a shift from oxidative fibre types toward more glycolytic fibre types in these animals, which was further supported by larger fibre-type cross-sectional areas in the more oxidative type fibres (type I/type IIa/type IIax hybrid). We also observed an increase in capillary density in th3/+ mice, indicative of a compensatory response. Western blotting for mitochondrial oxidative phosphorylation complex proteins and PCR analysis of mitochondrial genes indicated reduced mitochondrial content in the skeletal muscle but not the hearts of th3/+ mice. The phenotypic manifestation of these alterations was a small but significant reduction in glucose handling capacity. Overall, this study identified many important alterations in the proteome of th3/+ mice, amongst which mitochondrial defects leading to skeletal muscle remodelling and metabolic dysfunction were paramount. Thalassemia comprises a group of monogenetic disorders that result from the impairment of α- or β-globin expression due to mutations in the α- or β-globin genes [1,2]. The imbalance of α/β-globin chains in the hemoglobin tetramer leads to the formation of hemichromes, reduction in the lifespan of red blood cells and the increased apoptosis of erythroid progenitor cells, which, in turn, triggers extramedullary erythropoiesis and hepatosplenomegaly. The degree of anemia and the overall clinical phenotype vary among patients carrying homozygous or compound heterozygous globin gene mutations. In the most severe form of the disease, known as thalassemia major (TM), frequent blood transfusions are required for sufficient tissue oxygenation and survival [3]. This treatment corrects anemia and attenuates extramedullary erythropoiesis. However, red blood cells contain high amounts of iron (200–250 mg per unit), and, therefore, blood transfusions lead to iron overload. This is further aggravated by increased dietary iron absorption due to the suppression of the iron hormone hepcidin, mainly by erythroferrone and possibly also by additional bone marrow-derived factors, which are induced in response to ineffective erythropoiesis [4]. Indeed, hereditary hemochromatosis is a frequent genetic disorder of iron overload that is caused by the inactivation of hepcidin regulators [5]. Large increases in iron levels cannot be fully controlled by iron chelation therapy [6]; therefore, this imbalance can cause cardiomyopathy, diabetes and liver diseases [7]. As many as 50% of TM patients die before the age of 35 [8]. Heart failure, arrhythmia and myocardial infarction are responsible for ~70% of all thalassemic patient deaths [9]. Diabetes, a known driver of cardiovascular disease, also occurs frequently [10,11,12], with one meta-analysis finding the prevalence among TM patients to be 9%, with around 12% having impaired fasting glucose and glucose tolerance [13]. In the early stages of the condition, patients are asymptomatic, and there is often no correlation between serum ferritin levels and the development of cardiac and metabolic dysfunction. Thus, understanding early events that act as drivers of complications in thalassemia is an important research question. Skeletal muscle is the major site of fatty acid catabolism, plays a key role in mediating whole-body glucose homeostasis and is a major determinant of insulin sensitivity. Hence, it is important to consider that impairments to muscle health/metabolism could expedite the progression of complications in persons with thalassemia. Due to iron overload conditions, the tissues of TM patients experience increased oxidative damage to lipid membranes, proteins and DNA [14]. They have also been shown to exhibit mitochondrial damage, which results in decreased mitochondrial potential [15] and a reduced mitochondrial copy number compared with normal healthy controls [16,17]. Additionally, studies in animal models of thalassemia have demonstrated significant cardiac mitochondrial dysfunction, a contributor to cardiomyopathy [18,19]. Previous studies investigating thalassemia-related alterations in skeletal muscle have documented numerous findings relevant to this work. For example, a study examining muscle biopsies from a small number of α-thalassemia patients indicated a higher capillary tortuosity and unchanged capillary density and diameter [20]. It was concluded that the increased capillary tortuosity would promote oxygen supply to muscle tissues in order to compensate for the lower hemoglobin in those subjects [20]. Ultrastructural changes in the heart of a mouse model and patients with thalassemia included mitochondrial swelling, loss of myofilaments and the presence of lipofuscin, related to the high tissue iron content [21]. Thus, there is increasing evidence that mitochondrial dysfunction is a major manifestation of iron overload in thalassemia, although the temporal and mechanistic aspects of these changes remain to be resolved. In this study, we used a thalassemic mouse model (th3/+) and wild-type control animals at 8 weeks of age. We performed a proteomic analysis of skeletal muscle to discover differentially expressed proteins. Data input into an Ingenuity pathway analysis allowed for the identification of key pathways, which differed between genotypes. String analysis was used to identify major alterations in protein interactomes. Key data were further verified via standard histological, Western blotting and qPCR approaches. The impact on metabolic genotype was determined using glucose and insulin tolerance tests. Hematological analysis validated the thalassemia phenotype of th3/+ mice at the age of 8 weeks. When compared with the wild-type controls, the th3/+ animals exhibited reduced hemoglobin (HGB) content (Figure 1A), hematocrit (HCT) (Figure 1B), red blood cell (RBC) count (Figure 1C), mean corpuscular volume (MCV) (Figure 1D) and mean corpuscular hemoglobin (MCH) (Figure 1E). Additionally, they manifested increased red cell distribution width (RDW) (Figure 1F), white blood cell (WBC) count (Figure 1G), platelet (PLT) count (Figure 1H) and mean platelet volume (MPV) (Figure 1I). This typical hematological phenotype of thalassemia in th3/+ mice was accompanied by systemic iron overload, as indicated by the increased liver iron content (LIC) (Figure 1J) and serum ferritin (Figure 1K), a reflection of LIC. There was no genotype-specific difference in the body weight of the female animals, which was lower compared with that of the males; however, the body weight of the th3/+ males was significantly reduced vs. the wild-type (Figure 1L). In order to investigate potential alterations to the skeletal muscle proteome of th3/+ mice, we conducted an unbiased quantitative mass spectrometry-based proteomics study using label-free quantitation. We measured differences in protein expression relative to control mice. Using this approach, we were able to quantify 891 proteins with a minimum of 1 protein-group-unique peptide (80% having ≥2 protein-unique peptides). Of these quantified proteins, 97 showed statistically significant differential expression based on an FDR-adjusted p-value of <0.05 and a fold change cut-off of twofold. Based on these proteins, we performed hierarchical clustering, which revealed several clusters of regulated networks (Figure 2A). Clusters 4–10, generally representing proteins with decreased expression in the thalassemic group vs. controls (Figure 2B,C), and clusters 1–3, representing proteins with generally increased expression (Figure 2D). Ingenuity pathway analysis (IPA) demonstrated functional enrichment for components involved in cardiomyopathy signalling, protein ubiquitination and mitochondrial dysfunction, among others (Figure 3A). The further functional enrichment of protein–protein interaction networks using StringDB indicated that regulated components were enriched for members of the proteasome complex, ribonucleoproteins, mitochondrial proteins involved in oxidative phosphorylation and electron transport, as well as myofibril proteins and components of the troponin complex (Figure 3B). Gastrocnemius muscles from th3/+ mice had a significantly lower proportion of type IIA fibres vs. WT (p = 0.0003) and a significant increase in the proportion of type IIB fibres vs. WT (p = 0.0066) (Figure 4A,B). In gastrocnemius muscles, cross-sectional area (CSA) differences between fibre types were present. A trend toward a greater CSA for type I fibres existed in th3/+ compared with the WT mice (p = 0.68). The CSA of the type IIa fibres was significantly greater in th3/+ compared with WT mice (p < 0.0001). The CSA of the type II a/x hybrid fibres was also significantly greater in th3/+ vs. WT mice (p = 0.021) (Figure 4C). To observe the influence of the thalassemia phenotype on muscle capillarization, we investigated the proportion of capillaries in regions of interest with an alkaline phosphatase stain. There was an increase in capillary density in gastrocnemius muscles from th3/+ mice shown by a significant increase in alkaline phosphatase-stained areas compared with WT mice (p = 0.0003) (Figure 4D,E). Figure 4F shows representative images of oxidative and glycolytic muscle samples (the red-highlighted area shown at higher magnification), although no overt differences were observed between genotypes. Mitochondrial DNA (mtDNA) quantification has been used as a reliable indicator of mitochondrial quantity, as mtDNA levels remain almost constant in healthy organisms. Analysis of the 8-week-old tissue data showed significantly reduced mitochondrial content in the skeletal muscle of the th3/+ mice compared with the controls (Figure 5A). Skeletal muscle from the th3/+ mice had significantly decreased content for mitochondrial markers, 16S rRNA and ND1, whereas, in heart tissue, there was no significant difference in mtDNA markers (Figure 5A). Similarly, when we examined the content of OXPHOS complex I-V, there was a significant decrease in the skeletal muscle, but not the hearts, from th3/+ mice (Figure 5B). In addition, we observed a decrease in the expression of the mitochondrial proteins TOM20, MitoNEET, DRP1 and OPA1 (Figure 5C–F) in skeletal muscle from th3/+ mice. Interestingly, there was no significant difference in ferritin or mitoFerritin expression between genotypes when adjusted to the GAPDH and TOM20 controls, respectively (Figure 5G,H). To determine how these changes in skeletal muscle are linked to diabetes, a glucose tolerance test (GTT) was performed. No significant difference between wild-type and thalassemia mice was observed regardless of sex (Figure 6A or B). A significant increase in blood glucose was, however, observed during the first 5 min of the GTT in both male and female th3/+ mice (Figure 6C). There was a higher insulin level (indicative of reduced insulin sensitivity) in 8-week-old th3+/− mice at the first 5 min post-glucose injection timepoint (Figure 6D). Studies with thalassemia patients have demonstrated an increased risk of diabetes, heart disease and metabolic syndrome [22,23]. Interestingly, approximately 30% of metabolic syndrome patients exhibit iron overload, and the term dysmetabolic iron overload syndrome (DIOS) has been coined to describe this population [24]. Interventions to reduce iron excess, such as via venesection or the use of chelators, have been shown to improve insulin sensitivity and delay the onset of type 2 diabetes and heart failure [25,26], although this approach has not always been successful [27]. Based on the overall knowledge derived from studies to date, we postulated that iron overload itself may be a primary driver of impairments in the metabolic health of skeletal muscle. This work is of clinical significance, as additional insight into mechanisms underlying the development of metabolic complications in thalassemia is needed, particularly with a view to early intervention. Thus, we used an established mouse model of thalassemia and a proteomics-driven approach, which directed our analyses to mitochondrial alterations in skeletal muscle. Here, we show that the thalassemia phenotype is associated with the substantial remodelling of the skeletal muscle proteome. Based upon our standard analytical criteria, in this study, there were 97 quantified proteins that showed statistically significant differential expression. Hierarchical clustering revealed several clusters of regulated networks. Broadly speaking, the most prominent changes were observed in proteins related to mitochondrial pathways and protein ubiquitination. Accordingly, additional functional enrichment of protein–protein interaction networks using StringDB indicated that regulated components were enriched for members of the proteasome complex and mitochondrial proteins involved in oxidative phosphorylation and electron transport. In this study, we focused mainly on further investigating the proteomics signature for mitochondrial dysfunction. The association between mitochondrial defects and thalassemia has been indicated by various previous studies [28,29]. In erythroblasts from thalassemia patients versus controls, decreased mitochondrial oxidative phosphorylation was observed [17]. Impaired fatty acid oxidation in mitochondria was suggested to be related to decreased carnitine levels found in the circulation of individuals with thalassemia [30]. A study using reticulocytes showed the increased expression of mitochondrial ferritin in patients with α-thalassemia, suggesting that iron excess in mitochondria occurs, and this may be important in mitochondrial dysfunction [31]. There may be many contributors to skeletal muscle mitochondrial remodelling in thalassemia, one of which may be the direct effects of labile iron excess, in particular, intramitochondrial iron overload [28]. This is known to attenuate mitochondrial respiration by causing a decrease in cytochrome C oxidase [32], a finding we observed here in our study. Interestingly, targeting improved cytochrome c oxidase content via in vitro-transcribed (IVT)-mRNA delivery has been proposed as a therapeutic approach applicable to thalassemia [33]. In addition to mitochondrial metabolic dysfunction, our data also indicated potential defects in antioxidative mechanisms, thus exacerbating the impact of mitochondrial damage. For example, predominant in our dataset was decreased NADPH dehydrogenase content. It has been suggested that the peroxiredoxin-2-mediated induction of NADPH dehydrogenase quinone-1 is an important adaptive response to counteract oxidative stress [34], and it is likely that lack of this in the muscle of th3/+ mice correlates with the development of metabolic dysfunction. It should be noted that, while the proteomics analysis presented here indicates many interesting findings, one potential limitation is that the number of animals used is relatively small. One of the primary drivers for the changes in proteome observed here may be the need for skeletal muscles to adapt to reduced (due to anemia) oxygen delivery. Indeed, the idea that anemia may be modifying the morphology of skeletal muscle is consistent with the present results (increased capillary density) and previous work showing an increased capillary tortuosity in the muscles of thalassemia patients, a phenomenon the authors speculated would promote an increased oxygen supply to muscle tissue [20]. In this study, we observed a shift away from oxidative and mitochondrial-dependent fibre types toward more glycolytic and less mitochondrial-dependent fibre types. This response is consistent with the compensation of reduced oxygen delivery to skeletal muscles with glycolytic metabolism, yet it comes at the cost of inefficiency. This increase in the glycolytic phenotype of the muscle is also in agreeance with a reduction in the mitochondrial capacity in thalassemia mice, which we observed based on the proteomics profile; reduced mitochondrial DNA markers and Western blotting showing reduced OXPHOS expression. Based on glucose tolerance tests, we observed a mild insulin resistance phenotype in th3/+ mice at 8 weeks of age. The development of insulin resistance in muscle is expected to occur following mitochondrial dysfunction [35] or iron overload [36,37]. It is likely that as the mice age this defect would become more prominent [38]. Another aspect of the proteomic signature we found in this study that is likely of great significance in the context of the related literature is altered protein homeostasis pathways, both proteasome- and lysosome-mediated. It is well established that the ubiquitin–proteasome system and autophagy both play an important role in thalassemia via the excess amounts of free α-globin being processed via these protein quality control mechanisms [39,40]. Elevated proteasome activity was found in a previous study using red blood cell units from fourteen β-thalassemia donors versus sex- and aged-matched controls [41]. A study of the platelet proteome of X-linked thrombocytopenia with thalassemia patient pathway analysis revealed protein ubiquitination as a principal alteration [42]. Many and varied connections of autophagy with thalassemia have been reported, particularly the regulation of β-thalassemia erythropoiesis [43]. However, it is also likely that altered autophagy at the tissue level may have an important role in disease pathogenesis. Specifically, altered skeletal muscle autophagy, particularly mitophagy, can certainly contribute to muscle mitochondrial damage and metabolic dysfunction [44]. Excess labile iron is also likely to be a direct driver of the changes seen here, as persistent high levels of iron attenuate skeletal muscle autophagy by inhibiting autophagosome lysosome regeneration (ALR) [36]. Muscle can normally initiate endogenous mechanisms to protect itself from slightly elevated iron levels, and autophagy is a critical part of this response [45]. However, our data indicated this may not occur in th3/+ mice muscle, rendering them more susceptible to cellular damage. Excess iron has also been shown to attenuate the ubiquitin–proteasome protein quality control system in various cell types [46,47]. As far as we are aware, this is the first study of the skeletal muscle proteome in either human patients or a mouse model of thalassemia. Nevertheless, other proteomics-based studies have been conducted [48]. Analysis of the plasma proteome of patients with β-thalassemia versus healthy controls identified 13 potential biomarkers [49]. Interestingly global correlation analysis identified some pathways that can be cross-referenced with findings from our study; for instance hypertrophic/dilated cardiomyopathy signature being prominent and altered lysosomal function, especially cathepsins. In keeping with a cardiomyopathy signature, serum lipidomics analysis indicated that transfusion-dependent thalassemia patients had elevated triacylglycerols and long-chain acylcarnitines, with lower ether phospholipids or plasmalogens, sphingomyelins and cholesterol esters, reminiscent of what has been previously characterized in cardiometabolic diseases [5]. Untargeted metabolomics studies have also begun to characterize the metabolic defects in thalassemic individuals and how this can be altered upon intervention with hydroxyurea [50,51]. In summary, various types of omics-driven studies have established defective metabolic pathways in thalassemia that must now be further investigated. The most appropriate targets need to be defined and interventions tested. Overall, in 8-week-old th3/+ mice, we identified alterations in the proteome that point toward early mitochondrial alterations and dysfunction. A lack of hemoglobin-mediated oxygen delivery to muscle, in concert with this reduced mitochondrial capacity, correlated with a switch toward glycolytic muscle fibres and an increase in capillarization. These remodelling events are logical in terms of maximizing the use of available oxygen. Thus, we identified many important alterations in the proteome of th3/+ mice, among which, mitochondrial defects associated with skeletal muscle remodelling and metabolic dysfunction were paramount. From this and various other proteomics studies related to β-thalassemia, an important insight into disease pathogenesis is beginning to emerge, with mitochondrial function and protein degradation (lysosome and proteasome) being two prominent cellular processes that are perturbed. Eight-week-old mice heterozygous for the β-globin gene deletion (th3/+; also known as Hbbth3/+) on C57BL/6 background were used as a model of thalassemia and compared with wild-type littermates [52]. This mouse model is most akin to human thalassemia intermedia, as these mice do not require blood transfusions. The animals, both male and female, were housed in macrolone cages (up to 5 mice/cage, 12:12 h light–dark cycle: 7 am–7 pm; 22 ± 1 °C, 60 ± 5% humidity) and were allowed ad libitum access to chow and drinking water. Experimental procedures were approved by the Animal Care Committee of McGill University (protocol 4966). At the endpoint, blood was collected via cardiac puncture following anesthesia under isoflurane; tissues were then rapidly collected and snap-frozen in liquid nitrogen followed by storage at −80 °C until further use. Hematological parameters were determined with the Scil Vet-ABC hematology analyzer. Serum was prepared by using micro-Z-gel tubes with a clotting activator (Sarstedt) and kept frozen at −20 °C until analysis. Serum ferritin was determined at the Biochemistry Department of the Montreal Jewish General Hospital using a Roche Hitachi 917 Chemistry Analyzer. Tissue iron was quantified with a ferrozine assay, as previously described [53]. Gastrocnemius muscle tissue samples from wild-type and th3/+ mice (total, n = 10) were lysed in buffer containing 5% sodium dodecyl sulphate (SDS) and 100 mM TRIS pH 7.8. Samples were subsequently heated to 99 °C for 10 min and subjected to probe-based sonication using a Thermo Sonic Dismembrator at 25% amplitude for 3 cycles × 5 s. Remaining debris was pelleted by centrifugation at 20,000× g for 5 min. An aliquot of the supernatant was diluted to <1% SDS and used for estimation of protein concentration via bicinchoninic acid assay (BCA) (Pierce, cat# 23225). Lysates were clarified by centrifugation at 14,000× g for 5 min and transferred to a new reaction tube; disulphide bonds were reduced by the addition of tris(2-carboxyethyl)phosphine (TCEP) to a final concentration of 20 mM and incubated at 60 °C for 30 min. Free cysteines were alkylated using iodoacetamide at a final concentration of 30 mM and subsequent incubation at 37 °C for 30 min in the dark. An equivalent of 200 µg of total protein was used for proteolytic digestion via suspension trapping (S-TRAP) [54]. Proteins were acidified by adding phosphoric acid to a final concentration of 1.3% v/v. Samples were then diluted 6-fold in STRAP loading buffer (9:1 methanol:water in 100 mM TRIS, pH 7.8) and loaded onto an S-TRAP Mini cartridge (Protifi LLC, Huntington NY) prior to centrifugation at 2000× g for 2 min. Samples were washed three times with 350 µL of STRAP loading buffer and proteolytically digested using trypsin (Sigma, Toronto, Canada) at a 1:10 enzyme-to-substrate ratio for 16 h at 37 °C. Peptides were sequentially eluted in 50 mM ammonium bicarbonate, 0.1% formic acid in water and 50% acetonitrile. Peptide-containing samples underwent solid phase extraction using Oasis HLB, 30 mg, 1CC cartridges (Waters). Peptide samples were dried and reconstituted in 0.1% trifluoro acetic acid (TFA) prior to analysis using mass spectrometry. Samples were analyzed via data-dependent acquisition (DDA) using an Easy-nLC 1200 online coupled to a Q Exactive Plus (both Thermo Fisher Scientific). Samples were first loaded onto a pre-column (Acclaim PepMap 100 C18, 3 µm particle size, 75 µm inner diameter × 2 cm length) in 0.1% formic acid (buffer A). Peptides were then separated using a 100 min binary gradient ranging from 3–40% B (84% acetonitrile, 0.1% formic acid) on the analytical column (Acclaim PepMap 100 C18, 2 µm particle size, 75 µm inner diameter × 25 cm length) at 300 nL/min. MS spectra were acquired from m/z 350–1500 at a resolution of 70,000, with an automatic gain control (AGC) target of 1 × 106 ions and a maximum injection time of 50 ms. The 15 most intense ions (charge states +2 to +4) were isolated with a window of m/z 1.2, an AGC target of 2 × 104 and a maximum injection time of 64 ms and fragmented using a normalized higher-energy collisional dissociation (HCD) energy of 28. MS/MS spectra were acquired at a resolution of 17,500, and the dynamic exclusion was set to 40 s. DDA MS raw data were processed with Proteome Discoverer 2.5 (Thermo Scientific) and searched using Sequest HT against the canonical mouse SwissProt FASTA database downloaded from UniProt. The enzyme specificity was set to trypsin with a maximum of 2 missed cleavages. Carbamidomethylation of cysteine was set as the static modification and methionine oxidation as the variable modification. The precursor ion mass tolerance was set to 10 ppm, and the product ion mass tolerance was set to 0.02 Da. The percolator node was used, and the data were filtered using a false discovery rate (FDR) cut-off of 1% at both the peptide and protein level. The Minora feature detector node of Proteome Discoverer was used for precursor-based label-free quantitation. Proteins quantified by at least one protein-unique peptide were further filtered based on a minimum of >50% valid values in at least one of the two sample groups. Remaining missing values were imputed by low abundance sampling within Proteome Discoverer 2.5. LFQ abundances were scaled (normalized) based on the total amount of quantified peptides, and abundance ratios were calculated as the ratio of grouped protein abundances. Statistical significance was determined using background-adjusted t-tests and adjusted for false discovery rate (FDR) using the Benjamini–Hochberg method within Proteome Discoverer 2.5. Regulation was defined on the basis of having an adjusted p-value of less than 0.05 and an expression ratio cut-off of 2-fold. Dimensional reduction via principal component analysis and hierarchical clustering was conducted using the normalized protein and phosphopeptide LFQ abundances in Instant Clue (http://www.instantclue.uni-koeln.de/). Functional enrichment analysis of protein–protein interaction networks was performed using STRINGDB analysis [55] with the StringApp for visualization within Cytoscape (v 3.9.1) [56]. Muscle fibre typing was performed according to the protocol by Bloemberg and Quadrilatero [57]. Briefly, 10 μm of gastrocnemius muscle sections were cut with the Leica CM1850 Cryostat, maintained at −20 °C. The muscle sections were blocked with 10% goat serum (Vector Laboratories, S-1000) for 60 min followed by incubation with a primary antibody cocktail (Developmental Springs Hybridoma Bank, Iowa City, IO; BA-F8 1:50, SC-71 1:600, BF-F3 1:100) against three myosin heavy chain (MHC) isoforms (type I, type IIA and type IIB) for 120 min. Sections were washed with PBS and then incubated with the appropriate secondary antibodies (Alexa Fluor 350 IgG2b 1:500, Alexa Fluor 488 IgG1 1:500, Alexa Fluor 555 IgM 1:500) for 60 min. Gastrocnemius (10 μm) sections were cut and stained for alkaline phosphatase with SIGMAFAST (BCIP/NBT (B5655-25TAB)) dissolved in 10 mL of dH₂O to determine capillary density. Muscle sections were incubated with the alkaline phosphatase solution for 15 min at 37 °C, whereafter, they were rinsed with dH₂O and counterstained with eosin (0.5% v/w in dH₂O). Gastrocnemius muscle was dissected in 1 mm3 for analysis via TEM. After two washes with ice-cold 0.2 M sodium cacodylate buffer containing 0.1% calcium chloride, pH 7.4, samples were fixed overnight at 4 °C in 2.5% glutaraldehyde and washed 3× with washing buffer. Pellets were post-fixed with 1% aqueous OsO4 + 1.5% aqueous potassium ferrocyanide for 1 h and washed 3× with washing buffer. Specimens were dehydrated in a graded alcohol series, infiltrated with graded epon:alcohol and embedded in epon. Sections were polymerized at 58 °C for 48 h. Ultrathin sections (90–100 nm thick) were prepared with a diamond knife using a Reichert Ultracut E-ultramicrotome, placed on 200 mesh copper grids and stained with 2% uranyl acetate for 6 min and Reynold’s lead for 5 min. Grids were then examined with transmission electron microscopy. Samples were prepared and processed following the methods and protocol, as described before [58]. The primers used were 16S rRNA: Forward: 5′-CCGCAAGGGAAAGATGAAAGAC-3′ Reverse: 5′-TCGTTTGGTTTCGGGGTTTC-3′, ND1: Forward: 5′-CTAGCAGAAACAAACCGGGC-3′ Reverse: 5′-CCGGCTGCGTATTCTACGTT-3′, HK2: Forward: 5′-GCCAGCCTCTCCTGATTTTAGTGT-3′ Reverse: 5′-GGGAACACAAAAGACCTCTTCTGG-3′. Skeletal muscle, gastrocnemius and heart tissues were chop-frozen or pulverized with mortar and pestle in liquid nitrogen. The powdered tissue was then suspended in lysis buffer and homogenized with beads and then incubated on a rotating rocker for 1 h at 4 °C, followed by centrifuging samples at 1000 rpm for 10 min at 4 °C. The supernatant was transferred to a new Eppendorf tube, concentration was measured using BCA and samples were kept at −80 °C. Samples were then centrifuged at 10,000 rpm for 5 min at 4 °C and denatured at 95 °C for 5 min. Samples were run on 8%, 12% and 15% SDS-PAGE gels conducted at approximately 90 V for 2 h, followed by transfer to a polyvinylidene difluoride (PVDF) membrane at 120 V for 1.5 h. Membranes were blocked in 3% bovine serum albumin (BSA) blocking solution for 1 h at room temperature, followed by incubation in a 1:1000 dilution of the primary antibody overnight. The next day, membranes were washed and incubated with a secondary antibody in 1:5000 dilution for 1 h at room temperature. Membranes were activated using Clarity Western ECL Substrate solution and visualized using X-ray film development techniques. Western blot (WB) band intensity was quantified using ImageJ software and normalized to specific loading control. The following primary antibodies were used in this study: TOM20 (Cat#42406), CISD1/mitoNEET (Cat#83775), total DRP1 (Cat#8570), OPA1 (Cat#80471), GAPDH (Cat#2118). They were purchased from Cell Signalling Technology, Beverley, MA. Ferritin (Cat#PA1-29381) was purchased from Invitrogen, Toronto, Canada. Mitochondrial ferritin (Cat#ab66111) and total OXPHOS (Cat#ab110413) were purchased from Abcam. The following secondary antibodies were used: anti-rabbit immunoglobulin G horseradish peroxidase-linked antibody (Cat#7074) and anti-mouse immunoglobulin G horseradish peroxidase-linked antibody (Cat#7076), from Cell Signalling Technology. Wild-type and th3/+ mice littermates fasted for 5 h before the experiment. Subsequently, the animals were injected with 100 g/kg glucose ip. Blood glucose levels were measured by using the OneTouch Verio Flex glucose meter. Data are presented as mean ±SEM. Statistical significance between treatment groups was calculated using an unpaired Student’s t-test when comparing two groups. One-way ANOVA and two-way ANOVA were used for the comparison of more than two groups. A p-value of <0.05 was considered statistically significant, and it was plotted with GraphPad Prism 9.

PMC10002231

Jessica Calo,Sara Comesaña,Ángel L. Alonso-Gómez,José L. Soengas,Ayelén M. Blanco

Fatty Acid Sensing in the Gastrointestinal Tract of Rainbow Trout: Different to Mammalian Model?

21-02-2023

gut sensing,fatty acids,gut-brain axis,feed intake,fish

It is well established in mammals that the gastrointestinal tract (GIT) senses the luminal presence of nutrients and responds to such information by releasing signaling molecules that ultimately regulate feeding. However, gut nutrient sensing mechanisms are poorly known in fish. This research characterized fatty acid (FA) sensing mechanisms in the GIT of a fish species with great interest in aquaculture: the rainbow trout (Oncorhynchus mykiss). Main results showed that: (i) the trout GIT has mRNAs encoding numerous key FA transporters characterized in mammals (FA transporter CD36 -FAT/CD36-, FA transport protein 4 -FATP4-, and monocarboxylate transporter isoform-1 -MCT-1-) and receptors (several free FA receptor -Ffar- isoforms, and G protein-coupled receptors 84 and 119 -Gpr84 and Gpr119-), and (ii) intragastrically-administered FAs differing in their length and degree of unsaturation (i.e., medium-chain (octanoate), long-chain (oleate), long-chain polyunsaturated (α-linolenate), and short-chain (butyrate) FAs) exert a differential modulation of the gastrointestinal abundance of mRNAs encoding the identified transporters and receptors and intracellular signaling elements, as well as gastrointestinal appetite-regulatory hormone mRNAs and proteins. Together, results from this study offer the first set of evidence supporting the existence of FA sensing mechanisms n the fish GIT. Additionally, we detected several differences in FA sensing mechanisms of rainbow trout vs. mammals, which may suggest evolutionary divergence between fish and mammals.

Fatty Acid Sensing in the Gastrointestinal Tract of Rainbow Trout: Different to Mammalian Model? It is well established in mammals that the gastrointestinal tract (GIT) senses the luminal presence of nutrients and responds to such information by releasing signaling molecules that ultimately regulate feeding. However, gut nutrient sensing mechanisms are poorly known in fish. This research characterized fatty acid (FA) sensing mechanisms in the GIT of a fish species with great interest in aquaculture: the rainbow trout (Oncorhynchus mykiss). Main results showed that: (i) the trout GIT has mRNAs encoding numerous key FA transporters characterized in mammals (FA transporter CD36 -FAT/CD36-, FA transport protein 4 -FATP4-, and monocarboxylate transporter isoform-1 -MCT-1-) and receptors (several free FA receptor -Ffar- isoforms, and G protein-coupled receptors 84 and 119 -Gpr84 and Gpr119-), and (ii) intragastrically-administered FAs differing in their length and degree of unsaturation (i.e., medium-chain (octanoate), long-chain (oleate), long-chain polyunsaturated (α-linolenate), and short-chain (butyrate) FAs) exert a differential modulation of the gastrointestinal abundance of mRNAs encoding the identified transporters and receptors and intracellular signaling elements, as well as gastrointestinal appetite-regulatory hormone mRNAs and proteins. Together, results from this study offer the first set of evidence supporting the existence of FA sensing mechanisms n the fish GIT. Additionally, we detected several differences in FA sensing mechanisms of rainbow trout vs. mammals, which may suggest evolutionary divergence between fish and mammals. In mammals, there is unequivocal evidence that the gastrointestinal tract (GIT) is critically involved in the homeostatic control of feeding and energy balance through the so-called gut-brain axis [1]. For this, the GIT contains intestinal cells able to sense the presence of nutrients (carbohydrates, fatty acids/lipids, and amino acids/proteins) in the lumen by specific “taste” receptors or transporters and respond to such information by releasing signaling molecules [2]. While three types of intestinal cells (enterocytes, brush cells, and enteroendocrine cells (EECs)) have been associated with nutrient sensing, the main chemosensory cells within the GIT are EECs [3]. When EECs sense nutrients, multiple regulatory peptides, mainly ghrelin (GHRL), cholecystokinin (CCK), peptide tyrosine-tyrosine (PYY), and glucagon-like peptide-1 (GLP-1), are released. These peptides can act paracrinally on neighboring cells, but their main role is to serve as signaling molecules for gut-brain communication, which can take place either by transmission through the vagus nerve or systemic circulation [2,3,4]. Information derived from the GIT reaches the central nervous system (CNS), where it is integrated, ultimately resulting in changes in the production of key hypothalamic factors that govern food intake [5]. Over the last decades, the mechanisms underlying nutrient sensing in the GIT have become an area of increasing scientific interest, and, to date, several carbohydrate, fatty acid, and amino acid sensing systems have been described in the mammalian EECs [2,3,4]. For the purpose of the present research, only sensing mechanisms involving lipids/fatty acids will be further described. The GIT is exposed to high levels of lipids derived from diet (mainly triglycerides, TGs), which, after lipase digestion in the small intestine, are cleaved to release free fatty acids (FFAs) that are sensed by different G protein-coupled receptors (GPCRs). The main GPCRs sensing FFAs are referred to as free fatty acid receptors (FFARs), and they respond to FFAs depending on the length of their aliphatic chain [6]. Thus, medium-chain (6–12 carbons) and long-chain (13–21 carbons) fatty acids (MCFAs and LCFAs, respectively) are detected by FFAR1 (previously termed as GPR40) and FFAR4 (or GPR120), both primarily located in I- and L-cells [6,7,8]. In contrast, FFAR2 and FFAR3 (previously named GPR43 and GPR41, respectively) are responsive to short-chain fatty acids (SCFAs; <6 carbons) such as butyrate, propionate or acetate, which can be acquired from food but predominantly derive from the metabolism of non-digestible carbohydrates by gut microbiota in the distal intestine; because of this, FFAR2 and FFAR3 are expressed at large amounts in colonic L-cells [6,9]. Apart from these major receptors, GPR84 has been later discovered to bind MCFAs [10], although available evidence indicates that this receptor is not very abundant in the mammalian GIT and that it is not expressed in EECs. Thus, its role as a fatty acid sensor appears to be secondary; instead, its major role seems to be to enhance pro-inflammatory signaling and macrophage effector functions [11]. Finally, the receptor GPR119 has been considered an intestinal lipid sensor, although its natural ligands are not typically FFAs but endogenous lipid derivatives such as oleoylethanolamide (OEA) [12]. Nevertheless, a recent study has reported GPR119 activation in response to FFAs such as palmitoleic acid in human islet EndoC-betaH1 cells [13]. GPR 119 is also activated by dietary TG-derived 2-monoacylglycerols (2-MAG) [14]. Indeed, GPR119 is at least as important as FFAR1 in mediating the TG-induced secretion of gastrointestinal incretins in the small intestine, co-acting in synergy with FFAR1 [14]. Finally, in the colon, GPR119 is activated by microbiota-derived metabolites [15]. Besides GPCRs, several carriers have been associated with fatty acid sensing in the mammalian GIT. These include the fatty acid transporter CD36 (FAT/CD36), the fatty acid transport protein 4 (FATP4), and the monocarboxylate transporter isoform-1 (MCT-1) [2,3,4]. Fatty acid carriers are typically located on the apical membrane of enterocytes, where they facilitate FFA uptake. FAT/CD36 and FATP4 appear to be involved in LCFA translocation along the intestine, while MCT-1 participated in the absorption of SCFAs in the colon. MCFAs are absorbed by passive diffusion [2,4]. Despite its predominant enterocyte location, some studies have demonstrated the presence of some fatty acid transporters (at least FAT/CD36) in EECs, where they contribute to lipid-derived gastrointestinal hormonal release [16]. Gut nutrient sensing mechanisms remain almost unexplored in fish. In a previous recent study from our research group, we identified that the rainbow trout (Oncorhynchus mykiss) genome contains 10 different isoforms of ffar genes, of which only ffar1 seems clearly homologous to its mammalian counterpart. By contrast, the remaining isoforms identified appear to have evolved independently and it is not clear whether they are homologous to mammalian genes. In addition, we observed that a gene encoding a Ffar4 receptor subtype is missing in the rainbow trout. These observations allow us to suggest functional differences in gut fatty acid sensing between mammals and rainbow trout. However, as far as we are aware, there is no information in the literature regarding gut fatty acid sensing mechanisms in fish. Besides all the general roles of lipids in vertebrates [17], this nutrient type is particularly relevant for fish because the major aerobic fuel source for energy metabolism of fish muscle is FFAs derived from triglycerides (as those in diet) [18], and the main source of energy in aquaculture nutrition is lipids [19]. Additionally, it is important to note that fish and mammals differ importantly in terms of lipid metabolism (e.g., fish have the ability to produce long-chain polyunsaturated fatty acids (PUFAs), essential for multiple physiological processes, endogenously, while most mammals have a very low capacity for PUFA synthesis [20]), thus being of enormous interest to study whether evolutionary variations in terms of sensing mechanisms may exist between two groups derived by such differences. With this background, the present study aimed to identify and characterize fatty acid sensing mechanisms in the GIT of a fish model with a great interest in aquaculture, the rainbow trout. These carnivorous species have a better ability to digest lipids compared to herbivorous and omnivorous species, which appears to be attributed to their more specific and higher lipase activity and/or their genetic potential to store lipids [21]. Therefore, the comparison between rainbow trout and the known mammalian models provide two different frames, mostly unknown: evolutionary trends within vertebrates and putative differences between carnivore species and the omnivore models assessed so far in mammals. As shown in Figure 1, mRNAs encoding different Ffar isoforms, Gpr84, Gpr119, Fat/cd36, Fatp4, and Mct-1 (a/b), are found, at different abundance levels, in almost all of the regions of the rainbow trout GIT studied, i.e., stomach, pyloric caeca, proximal intestine, middle intestine, and distal intestine. Specifically, ffar1 mRNAs were more abundantly expressed in the pyloric caeca and proximal intestine, followed by the rest of the intestinal sections, with undetected expression in the stomach (Figure 1A). The abundance of ffar2b1.1 mRNAs was higher in the pyloric caeca compared with the rest of the tissues analyzed, but quantifiable levels were also observed in the proximal and middle intestine; however, expression levels were extremely low in the stomach and hindgut, thus hampering gene expression quantification in these regions (Figure 1B). ffar2b1.2 mRNAs were more abundant in the pyloric caeca, followed by the proximal and distal intestine, although low levels were detected in all gastrointestinal regions (Figure 1C). mRNAs encoding Ffar2b2a and Ffar2b2b were the most abundant of all receptor mRNAs studied (Ct values ≈28) and were detected along the entire GIT, with the highest levels found in distal intestine (Figure 1D,E). ffar2a1b mRNAs were higher in the intestine compared to the stomach, and pyloric caeca, with the highest levels detected in the proximal and distal regions, (Figure 1F). The expression of ffar2a2 mRNAs was high in the distal intestine, low in the rest of the intestinal regions, and almost undetected in the stomach (Figure 1G). Both gpr84 and gpr119 mRNAs were predominantly found in the distal intestine, although quantifiable expression levels were observed in all gastrointestinal regions (Figure 1H,I). It should be noted that the abundance of all receptors mRNAs throughout the gastrointestinal, yet quantifiable, was rather low, as indicated by high Ct values in the real-time PCR runs (≈28–34). In contrast, lower Ct values (≈25–29), and therefore greater expression levels, were observed for mRNAs encoding the fatty acid transporters Fat/cd36, Fatp4, and Mct-1a. All three were abundantly expressed throughout the entire GIT, although some differences in expression levels were detected among regions for fatp4 (lower relative expression in the stomach) and slc16a1a (higher relative expression in the distal intestine) (Figure 1J–M). The expression of slc16a1b was high in the stomach, with levels compared to the rest of transporters throughout the GIT (Ct values ≈25), but very low in pyloric caeca and all intestinal regions (Ct values ≈30–31) (Figure 1M). Fish fasted for 48 h were intragastrically administered with octanoate, oleate, ALA, or butyrate, and samples of different regions of the GIT were collected at 20 min and 2 h post-administration to assess different parameters related to fatty acid sensing and appetite regulation (Figure 2A). Figure 2B–M shows the effects of intragastrically administered fatty acids on the mRNA expression of fatty acid receptors and transporters along the rainbow trout GIT at 20 min and 2 h post-administration. In a short time, treatment with octanoate led to a significant upregulation of ffar1 and ffar2b1.2 in the middle intestine (Figure 2B,D), ffar2a1b in the proximal and middle intestines (Figure 2G), ffar2a2 in the stomach and middle intestine (Figure 2H), and slc16a1a (encoding Mct-1a) in the middle intestine (Figure 1M). Oleate induced ffar2b1.1, ffar2b1.2, ffar2a1b, ffar2a2, gpr84 and gpr119 in proximal intestine (Figure 2C,D,G–J), and also increased ffar1 in middle intestine (Figure 2B), and ffar2b1.2 in stomach (Figure 2D), while it decreased gpr84 in the distal intestine (Figure 2I) and fatp4 in the proximal intestine (Figure 2L). In addition, both octanoate and oleate significantly increased gpr119, cd36, and fatp4 mRNAs in the distal intestine (Figure 2J–L). Administration of ALA resulted in increased levels of ffar2b1.1 and ffar2b2a in the proximal intestine (Figure 2C,E), ffar1 and ffar2b1.2 in the proximal and middle intestines (Figure 2B,D), cd36 and fatp4 in the distal intestine (Figure 2K,L), and gpr119 in all regions of the GIT analyzed, except for the stomach (Figure 2J). On the contrary, significant ALA-induced downregulations of stomach slc16a1a (Figure 2M) and distal intestine ffar2a1b and gpr84 (Figure 2G,I) was detected. Finally, significantly higher levels of gpr84 and gpr119 in the stomach (Figure 2I,J), ffar2a1b, gpr84, cd36, fatp4, and slc16a1a in the proximal intestine (Figure 2G,I,K–M), gpr84, gpr119, and cd36 in the middle intestine (Figure 2I–K), and ffar2b2b and slc16a1a in the distal intestine (Figure 2F,M) were observed in fish administered with butyrate compared to control fish. Butyrate treatment also led to decreased slc16a1b mRNAs in the stomach (Figure 2N), ffar2b1.1 in the middle intestine (Figure 2C), and fatp4 mRNAs in the stomach (Figure 2L). At 2 h post-administration, a significant increase in the mRNA levels of ffar2b2a, ffar2a1b, and gpr84 in the stomach (Figure 2E,G,I), cd36 in the proximal and middle intestine Figure 2K, and fatp4 in the stomach and proximal intestine (Figure 2L), was observed in response to octanoate. Oleate up-regulated the expression of ffar2b2a, gpr84, and cd36 in the stomach (Figure 2E,I,K), of ffar1 in the proximal intestine (Figure 2B), and of ffar1, ffar2b1.2, and gpr84 in the distal intestine (Figure 2B,D,I). On the contrary, it down-regulated the mRNA abundance of gpr119 in the distal intestine (Figure 2J). Treatment with ALA resulted in significantly higher levels of ffar1 and ffar2b2a in the stomach (Figure 2B), ffar2b1.2 in the proximal, middle, and distal intestine (Figure 2D), ffar2a1b in the distal intestine (Figure 2G), gpr84 in the stomach, proximal intestine, and middle intestine (Figure 2I), fatp4 in the stomach and proximal intestine (Figure 2L), and slc16a1a in the distal intestine (Figure 2M). Lastly, increased levels of ffar2b1.2 and ffar2b2b in the middle and distal intestine (Figure 2D,F), ffar2a1b in the stomach and proximal intestine (Figure 2G), ffar2a2 and slc16a1a in proximal intestine (Figure 2H,M), gpr119 in the distal intestine (Figure 2J), fatp4 and slc16a1b in the proximal and middle intestine (Figure 2L,N), and cd36 in all gastrointestinal regions analyzed (Figure 2K), were detected upon butyrate administration. Considering that gustducin is the major G protein activated in response to FFAR activation in the mammalian GIT [2] and that the phospholipase C (PLC)- inositol triphosphate (IP3) and adenylate cyclase (AC)-cAMP-protein kinase A (PKA) pathways are the main intracellular signaling cascades triggered as a consequence [3,4,5,6], in this study we measured the gastrointestinal mRNA expression of a putative G protein involved in nutrient signaling in fish (Gnai1) as well as the mRNA levels of key elements of both the PLC-IP3 and AC-cAMP-PKA pathways in response to fatty acid administration to study whether the same mechanisms may operate in rainbow trout. The changes in mRNA abundance of such parameters at 20 min and 2 h after intragastric administration of fatty acids are shown in Figure 3 and Supplementary Table S1. Increased gnai1 mRNAs were observed upon octanoate treatment at 20 min in the middle and distal intestines, upon oleate treatment at 2 h in proximal and middle intestines, upon ALA treatment at 20 min in proximal and distal intestines, and at 2 h in the stomach and all intestinal regions, and upon butyrate treatment at 20 min in the stomach. Expression of plcβ1 was found to be up-regulated by butyrate in the stomach, proximal and middle intestines at 20 min, while at 2 h, only a significant upregulation was detected in the stomach. Oleate and ALA also caused significant increases in plcβ1, as well as plcβ3, mRNAs, especially in the proximal and middle intestines. On the contrary, the expression of both genes remained unaltered or even down-regulated in response to octanoate. Except for the middle intestine, all fatty acids tested led to significantly lower levels of plcβ4 mRNAs compared to the control group in all or most regions tested and at 20 min and/or 2 h. As for itpr1, we found increased mRNA levels at 20 min in the proximal intestine in response to butyrate, in the middle intestine in response to all fatty acids, and in the distal intestine in response to octanoate and ALA, as well as at 2 h in the proximal intestine in response to ALA and butyrate, middle intestine in response to ALA and distal intestine in response to butyrate. The expression of itpr3 was, in general, down-regulated in response to the luminal presence of fatty acids, with the major changes found in the stomach. Finally, ac mRNA levels were observed to be downregulated by butyrate treatment in the proximal intestine at 20 min and in the middle intestine at 20 min and 2 h. However, increased ac mRNAs were found in the middle intestine 2 h after oleate administration and in the distal intestine 20 min after ALA administration. The luminal presence of fatty acids modulates mRNA and protein levels of gastrointestinal hormones, as shown in Figure 4. The abundance of ghrl mRNAs was observed to be up-regulated by ALA and butyrate in the stomach, proximal intestine (only butyrate), and middle intestine at 20 min post-treatment and by oleate in the stomach and ALA in the proximal and middle intestine at 2 h. No significant differences in stomach Ghrl levels were detected in response to any of the fatty acids (Figure 4B,H). Levels of cck/Cck were unaltered by luminal fatty acids at 20 min (Figure 4C,D,H). At 2 h, oleate and ALA led to a significant upregulation of cck and/or Cck levels in the proximal intestine, while the opposite effect was observed for butyrate (Figure 4C,D,H). In addition, oleate and butyrate led to increased cck mRNAs in the distal intestine (Figure 4C). Treatment with octanoate, oleate, and ALA resulted in a general tendency to increase the abundance of pyy/Pyy, especially at 2 h (Figure 4E,F,H). Butyrate, on the other hand, reduced pyy mRNA levels in both the proximal and middle intestine at 20 min and 2 h (Figure 4E), although a significant increase in Pyy protein levels was observed in the proximal intestine at 2 h (Figure 4F,H). Finally, gcg (proglucagon, gene encoding Glp-1) mRNAs were found to be up-regulated by octanoate, oleate, and ALA treatment at 20 min in the proximal intestine, while down-regulated by the former two at the same time point in the distal intestine (Figure 4G). Due to technical difficulties with finding a suitable Glp-1 antibody, we were not able to detect levels of this protein by Western blot in the present study. Great interest in elucidating the mechanisms by which the gut senses luminal nutrients and how this sensing impacts the homeostatic control of feeding has been taking place over the last few years. Gut nutrient sensing relies on the presence of specific receptors and transporters located mainly in the luminal surface of enteroendocrine cells and enterocytes, respectively, which are able to respond to variations in the luminal levels of nutrients [7,8,9]. Studies on this topic, however, have focused mainly on mammalian models, and whether equivalent mechanisms operate in other vertebrate groups remains practically unknown, particularly in fish. This research aims to address this scarcity of information in the fish literature and provides the first evidence on the presence and functioning of fatty acid sensing mechanisms in the GIT of a carnivore fish species, the rainbow trout. We focused on lipids because of three reasons: (i) they are the main energy source in aquaculture nutrition [11], (ii) they are the major aerobic fuel source for energy metabolism of the fish muscle [10], and (iii) there are several key differences between fish and mammalian lipid metabolism [12]. The first objective of our research was to study whether the main fatty acid receptors and transporters described in the mammalian GIT (i.e., FFAR1/2/3/4, GPR84/119, FAT/CD36, FATP4 and MCT-1; [7,8,9]) are present in rainbow trout. A previous in silico study from our research group, together with a study by Roy and coworkers (Roy et al., under review), described that some genes encoding such mammalian receptors (particularly Ffars) are not present within the rainbow trout genome (ffar4), some are duplicated (as expected due to whole-genome duplications events during evolution), and some appear not to be orthologous to mammalians. In the same previous study, we showed that most of these ffar genes identified within the rainbow trout genome (specifically, ffar1, ffar2b1.1, ffar2b1.2, ffar2b2a, ffar2b2b, ffar2a1b, and ffar2a2) are expressed at a smaller or greater extent in the stomach, anterior intestine and/or posterior intestine. In the present study, we carried out PCRs targeting these genes to confirm previous observations. However, the putative presence of the rest of the fatty acid receptors and fatty transporters remains unknown. Fatty acid transporter genes are pretty well conserved throughout evolution, and sequences encoding Fat/cd36, Fatp4, and Mct-1 (with two copies for the latter) can all be found within the rainbow trout genome [14,15]. A high degree of conservation is also observed for genes encoding Gpr84 and Gpr119 [16]; thus, their presence in the rainbow trout genome is also evident. PCR and RT-qPCR analyses targeting all these genes indicated the expression of mRNAs encoding Gpr84, Gpr119, Fat/cd36, Fatp4, and Mct-1 (a and b isoforms) in the GIT of rainbow trout. In the mammalian gut, fatty acid receptors (except for GPR84, [17]) and transporters are located in the apical membrane of different cell types, with receptors being typically found in enteroendocrine cells while transporters in enterocytes [7,8,9]. Experimental approaches used in the present research do not allow us to discriminate the cell type location of receptors or transporters, so we will discuss obtained results considering gastrointestinal cells in general. However, the fact that the mRNA abundance of receptors was low (very high Ct values) and that of transporters considerably high might be an indirect indicator of their cell type location. Thus, considering that enterocytes are the most abundant epithelial cells in the GIT, and EECs represent only 1% of them [18], we might suggest receptor presence in EECs and transporter in enterocytes. Future lines of research will focus on the sorting of rainbow trout intestinal epithelial cells by type using flow cytometry and the study of nutrient-sensing mechanisms taking place in each individual cell type. Based on the GIT distribution study, we observed that except for ffar1 and ffar2b1.1, whose transcripts were not quantifiable in the stomach, and also hindgut in the case of the latter, all fatty acid receptors and transporters detected are found at quantifiable levels in the stomach, pyloric caeca, and along the entire intestine, thus showing a widespread distribution within the GIT. This differs from the mammalian model for some receptors/transporters. For instance, any receptor was observed to be almost exclusively expressed in the distal intestine of the rainbow trout, as FFAR2 and FFAR3 are in the case of mammals [19,20]. It is also worth pointing out the case of MCT-1. Two isoforms of this transporter (a and b) have been described in rainbow trout [21]. These forms were here observed to show a very different expression profile along the GIT, with slc16a1a (encoding Mct-1a) mRNAs expressed in the whole GIT but most importantly in the hindgut, and slc16a1b (encoding Mct-1b) almost exclusively detected in the stomach. In mammals, studies in mice and rats have shown that the single MCT-1 isoform found in these vertebrates is poorly expressed in the stomach but abundant in the colon [22], as expected considering that MCT-1 is involved in SCFA uptake and that the colon is the predominant location for SCFA synthesis. However, interestingly, a high expression of this transporter was reported in both the caprine stomach and large intestine [23], which appears to be related to the fact that ruminants also produce large amounts of SCFAs in the rumen. Indeed, SCFAs constitute the major fuel source in ruminants, providing up to 80% of their energy requirements [24]. The physiological significance of the distinct expression profile of the two Mct-1 isoforms along the rainbow trout GIT observed in this study requires additional investigation. However, we could hypothesize that each isoform, predominant at each end of the GIT, might be involved in the uptake of different SCFAs. This could relate to the previous report that the microbiota of the rainbow trout stomach and intestine shows considerable differences [25]. The next step of our study was to characterize the response of the identified receptors to the luminal presence of fatty acids of different lengths and degrees of unsaturation [i.e., octanoate (8-carbon saturated FA), oleate (18-carbon monounsaturated FA), ALA (18-carbon PUFA), and butyrate (4-carbon saturated FA)]. For this, we intragastrically administered fatty acids into fasted rainbow trout and assessed the abundance of mRNA encoding the target receptors in the GIT. The most important changes include increases in the mRNA abundance of ffar2a1b and ffar2a2 in response to octanoate, ffar1, ffar2b1.1 and ffar2b1.2 in response to oleate, ffar1, ffar2b1.2 and gpr119 in response to ALA, and ffar2b2b, and gpr84 in response to butyrate, particularly in anterior regions of the GIT (importantly involved in nutrient sensing in mammals). Some of the fatty acids tested, mainly oleate and ALA, also led to increased expression of other types of fatty acid receptors (e.g., ALA up-regulated the expression of ffar2b1.1, ffar2a and gpr119, while oleate that of the ffar2a); however, these increases were, in general, less pronounced than the formers. This suggests that the different fatty acid receptors appear to be more responsive to specific ligand/s (i.e., Ffar1:oleate and ALA, Ffar2b1 (1 and 2): oleate and ALA, Ffar2b2b: butyrate, Ffar2a1b: octanoate and butyrate, Ffar2a2: octanoate, Gpr84: butyrate, Gpr119: ALA), although they may also be activated by other types of fatty acids. While we only measured mRNA abundance here, and experiments testing the ligand affinity of each receptor are required, the activation profile of fatty acid receptors in the rainbow trout GIT that can be suggested from our experiment points out important putative differences with regard to gut fatty acid sensing mechanisms in mammals, in which FFAR1 is only activated by MCFAs and LCFAs, FFAR2 and FFAR3 are only activated by SCFAs, GPR84 mainly by MCFAs, GPR119 by lipid derivatives (e.g., OEA) [7,8,9]. These differences strengthen our hypothesis of rainbow trout not having clear orthologous receptors to mammalian FFFAR2 and FFAR3. Although a deeper understanding of the mechanisms underlying fatty acid sensing in the rainbow trout GIT is required, present observations establish a basis in favor of the existence of major functional (maybe evolutionary) differences between gut nutrient sensing mechanisms between fish and mammals. An interesting observation from our intragastric experiment is that there is a clear differentiation in the activation of receptor mRNA abundance in response to the luminal presence of fatty acids depending on the region of the GIT and time. Such a differentiation also applies to the mRNA abundance of the fatty acid transporters tested, i.e., Fat/cd36, Fatp4, and Mct-1a/b (Figure 5). In general terms, we can distinguish between one type of response in the anterior region of the GIT (including stomach, proximal intestine, and likely middle intestine) and another type of response in the distal intestine. In addition, results obtained point towards comparable mechanisms of action for octanoate, oleate, and ALA, while butyrate displayed clear differences. With this in mind, major results from our study allow us to hypothesize that the presence of octanoate, oleate, ALA, or butyrate in the intestinal lumen would be first sensed by specific membrane receptors located in the anterior regions of the GIT. As mentioned earlier, although functional studies on ligand affinity are needed, we propose that Ffar2a (1b and 2) could be more activated by octanoate, Ffar1 and Ffar2b1 (1 and 2) by oleate, Ffar1, Ffar2b1 (1 and 2) and Gpr119 by ALA, and Ffar2a1b and Gpr84 by butyrate, although less pronounced activations with other fatty acids may occur. Besides receptors, the butyrate-induced increased expression of cd36, fatp4 and slc16a1a in the proximal intestine suggests the transporters Fat/cd36, Fat/p4, and Mct-1a as additional important sensors of butyrate in the rainbow trout GIT at a short-time. This observation differs from the mammalian model, in which only MCT-1, and not FAT/CD36 or FATP4, plays a role in the intestinal transport of SCFAs like butyrate [26]. Interestingly, our results demonstrated the increased abundance of mRNAs encoding Mct-1a not only in response to butyrate but also to octanoate in the proximal and middle intestine, which points towards this fatty acid as an additional activator of Mct-1a in the rainbow trout GIT. Except for this, treatment with octanoate, oleate, and ALA led to a general inhibition of the transporter’s mRNA abundance in the stomach and proximal intestine. Since we measured mRNA abundance only, this downregulation does not discard Fat/cd36, Fatp4, and Mct-1a/b as putative sensors for octanoate, oleate, and ALA, but could be the result of another response (e.g., negative feedback), although further studies are required for a certain explanation. In the distal intestine, unlike what has just been stated, we observed that octanoate, oleate, and ALA increased cd36 and fatp4 mRNA abundance. This response can be attributed to the fact that the number of fatty acids in the distal vs. proximal intestine is likely considerably lower and/or that the mRNA levels of both cd36 and fatp4 are lower in the distal vs. proximal intestine/stomach (as observed from the GIT distribution study). Therefore, an increase in transporter expression in response to fatty acids may be related to transporter sensitivity increase. In any case, all three fatty acids (not only oleate) increased cd36 and fatp4 mRNA abundance (and considering that this observation might be an indicator of increased transport activity) is different from the mammalian model, in which both FAT/CD36 and FATP4 are in charge of LCFA translocation, whereas MCFAs are absorbed by passive diffusion [7,9]. However, again, this observation is just based on mRNA abundance data, and further research devoted to the study of transporter activity in response to different fatty acids is needed to confirm that both FAT/CD36 and FATP4 would be translocating fatty acids of different lengths (LCFAs, MCFAs, and PUFAs) in the rainbow trout distal intestine. The translocation model for SCFAs would likely occur according to the mammalian model [26], with MCT-1 (specifically, Mct-1a isoform in rainbow trout) being responsible for such an action, as suggested for the increased slc16a1a mRNA abundance and unaltered/decreased cd36 and fatp4 mRNA abundance in the distal intestine in response to butyrate. As for the receptors, Gpr119 and Ffar2b2b appear to be the only receptor types detecting the luminal presence of fatty acids (octanoate, oleate, and ALA in the case of the former, and butyrate in the latter) in the distal intestine at a short-term, as depicted by increased gpr119 (and not other receptors) expression in response to octanoate, oleate, and ALA, and increased ffar2b2b in response to butyrate. Over a long-time, major results demonstrated that all fatty acids up-regulated the mRNA abundance of cd36 and/or fatp4 in proximal areas of the GIT. This result seems controversial when compared with the octanoate/oleate/ALA-induced down-regulation of the mRNA abundance of both transporters in the proximal intestine at 20 min. However, as discussed for the distal intestine, such up-regulation after 2 h could increase the sensitivity of the transporters in response to low luminal levels of fatty acids (as there would likely be compared to 20 min). In any case, these results support the wider affinity of FAT/CD36 and FATP4 to fatty acids of different lengths in rainbow trout vs. the restricted affinity in mammals [7,9]. In contrast, the induced expression of slc16a1a and slc16a1b in the proximal and/or middle intestine in response to butyrate (and not to other fatty acids) argues in favor of Mct-1 being more devoted to the translocation of SCFAs rather to other fatty acid types. Unlike proximal gastrointestinal regions, no major fatty acid-induced changes in the transporter mRNA abundance (except for a butyrate-induced increase in cd36 mRNAs) were detected in the distal intestine, which may indicate that transport of at least octanoate, oleate, and ALA into distal intestinal cells occur at a shorter time. Regarding receptors, we observed a general attenuation in expression activation of fatty acid receptors mRNA abundance compared to 20 min, as observed, for instance, in the cases of ffar2b1.1 (unaltered upon all treatments) and gpr119 (only activated in response to butyrate in the distal intestine). Other receptors, such as ffar1, showed a similar induction in expression to that observed at 20 min, i.e., mainly in response to oleate and ALA in the proximal and middle intestine. Expression of ffar2b1.2 was also mainly induced by the same ligands (oleate and ALA) but at more distal regions of the GIT. Finally, we can highlight the case of gpr84, which was observed to be increased in the stomach 2 h after intragastric fatty acid administration regardless of the fatty acid assessed. It might be possible that these observations respond to an increase in receptor sensitivity over time. Altogether, results from the present research clearly suggest a differential activation profile of fatty acid receptors along the rainbow trout GIT depending on the time after nutrient administration. Fatty acid receptors, as classical GPRs, respond to fatty acid binding with structural changes that lead to the activation of intracellular guanine nucleotide-binding proteins (G proteins) and the subsequent triggering of diverse signaling pathways. The major G protein coupling FFAR activation to hormonal release in the mammalian GIT appears to be gustducin (initially found in taste cells) [2]. Nevertheless, it appears that no ortholog of the mammalian gustducin gene (gnat3, guanine nucleotide-binding protein g (t) subunit alpha-3) is present in teleost fish; instead, other G proteins (e.g., Gnai1) appear to participate in the signaling of gut sensing [27,28]. The general up-regulation of gnai1 mRNA abundance in response to intragastrically administered fatty acids observed in this study argues in favor of this G protein being activated as a consequence of fatty acid binding to FFARs in the rainbow trout GIT. In mammals, different signaling pathways appear to be triggered upon G protein activation depending on the receptor. Thus, the major effector for FFAR2 and FFAR3 seems to be PLC, whose activation results in increased production of IP3, which in turn binds to its receptor (ITPR3) located at the endoplasmic reticulum, releasing Ca2+ into the cytoplasm [3]. GPR119 operates mainly through the AC-Camp-PKA pathway: its activation results in the activation of AC, responsible for converting ATP to the second messenger Camp, thus leading to Camp accumulation and, thereby, activation of PKA [4]. In the case of GPR84, signaling pathways downstream of its activation have been well studied regarding its pro-inflammatory nature. Considering that elevated intracellular Camp levels suppress innate immune functions, it has been proposed that GPR84 exerts its pro-inflammatory actions by inhibiting AC and thereby suppressing intracellular Camp [5,6]. Additionally, other signaling pathways, such as the ERK cascade, have been associated with GPR84 signaling in immune functions [29]. Nevertheless, no information is available on the specific intracellular cascades in charge of coupling GPR84 and gastrointestinal hormone release. With this background, and considering that no evidence in this respect is available in fish literature, we investigated in the present study whether the gastrointestinal abundance of mRNAs encoding key elements within the PLC-IP3 and AC-Camp-PKA pathways is affected by the luminal presence of fatty acids. The results demonstrated increased levels of plcβ1, plcβ3, and itpr1 mRNAs in anterior regions of the rainbow trout GIT in response to oleate and ALA, which may indicate that these two fatty acids could possibly signal through the PLC-IP3 pathway, although some important differences, such as the involvement of the Plcβ1 and 3 (instead of PLCβ2) and Itpr1 (instead of ITPR3), may exist with respect to the mammalian model. The PLC-IP3 pathway (specifically involving the isoforms Plcβ1 and Itpr1) may also participate in mediating butyrate actions in anterior regions of the rainbow trout GIT. As for the AC-cAMP-PKA signaling cascade, it might mediate at least some ALA responses in the distal intestine, maybe by acting through GPR119, as suggested by increased ac mRNA levels upon treatment with this fatty acid in the mentioned region. We also observed an interesting down-regulation of ac mRNAs in the proximal and middle intestine upon butyrate treatment, which, considering the role herein proposed for GPR84 in the mediation of butyrate responses, may match the intracellular signaling cascade proposed to this receptor in mammals [5,6]. Notably, no major changes occurred in the mRNA abundance of the intracellular signaling elements tested in response to octanoate, suggesting that other signaling pathways different from PLC-IP3 and AC-cAMP-PKA likely mediate octanoate actions. It has to be taken into consideration, however, that all these observations are based on gene expression data only, and future studies measuring the levels of second messengers should be performed in order to confirm present results. In mammals, the triggering of intracellular signaling cascades in response to the sensing of luminal fatty acids and leads to the release of gastrointestinal hormones [8,9]. In the case of FFAR2 and 3, such a release occurs as a consequence of the rise in intracellular Ca2+, which activates the fusion machinery of the secretory granules containing hormones, thus triggering their release by exocytosis. Ca2+-triggered exocytosis likely operates for GPR119 as well, with PKA acting as a regulator of such a process [30]. Experimental approaches included in this study do not allow to describe the triggering process underlying hormonal release, but both qPCR and Western blot analysis demonstrated increased mRNA/protein levels of major gastrointestinal hormones (Ghrl, Cck, Pyy, and/or Glp-1) in the rainbow trout GIT upon fatty acid intragastric treatment, suggesting that hormone release is a consequence of gut fatty acid sensing in rainbow trout, as is the case in mammals [8,9]. Major increases in gastrointestinal hormone levels occurred in anterior regions of the GIT (stomach, proximal, and, to a lesser extent, middle intestine), suggesting these regions are primarily involved in appetite regulation. We observed a differential modulation of the Ghrl, Cck, and Pyy mRNA and/or protein level abundance depending on the fatty acid. In general, both octanoate and oleate led to increased Glp-1 levels at a short time-post administration, while at a longer time, they led to increased Pyy and also Cck in the case of oleate. All these hormones are of anorexigenic nature [31,32]; thus, their release in response to octanoate and oleate would be in agreement with an inhibitory role in feed intake for these two fatty acids. In the case of oleate, present observations regarding the hormonal release are in accordance with mammalian studies, which reported that LCFAs trigger CCK, GLP-1, and PYY secretion and suppress ghrelin release [33,34]. However, this response was not seen with fatty acids of 11 carbon atoms or fewer [35]; thus, results here observed for octanoate support a different model than that known in mammals. In mammals, both octanoate and oleate are primarily sensed by FFAR1 and FFAR4, and these two receptors are, therefore, related to hormone release. MCFAs are also detected by GPR84, but this receptor in mammals appears not to be expressed in EECs; thus, it would not be involved in the release of gastrointestinal hormones. In rainbow trout, we proposed that Ffar4 is absent, and thus the receptor binding these two fatty acids in this species (apparently n-Ffar5b (1b and 2a) in the case of octanoate and Ffar1 and n-Ffar2a (1 and 2) in the case of oleate) would be associated with octanoate- and oleate-evoked hormone release. FFAR2 and FFAR3 in mammals are responsive to SCFAs (such as butyrate), and they are believed to induce the release of PYY [36,37] and GLP-1 [38]. However, a later study using isolated rat colons suggested that the release of colonic PYY/GLP-1 in response to the presence of luminal SCFAs does not involve FFAR2/FFAR3; it rather occurs in response to the metabolization of SCFAs and subsequent function as a colonocyte energy source [39]. Results from the present study using butyrate demonstrated, in general lines, increased levels of Ghrl and a decrease in those of Cck and Pyy in response to this SCFA, hormonal responses that would presumably occur upon activation of n-FFar2b2b, n-Ffar5b1b, and/or GPR84. Contrary to octanoate and oleate, increased levels of Ghrl (orexigen; [31,32]) and decreased levels of Cck and Pyy (anorexigens; [31,32]) would suggest a stimulatory role in feed intake. Finally, the release of Ghrl and Glp-1 at a short time and of Cck and Pyy at a longer time would likely be responses occurring upon activation of gastrointestinal sensors of luminal PUFAs, such as ALA. In this case, hormonal release (especially at a short time) would suggest a contradictory effect on feed intake. It must be highlighted that the observation of hormonal release in response to ALA administration in rainbow trout indicates an important difference compared to mammals, in which n-3 PUFAs (such as ALA) do not seem to activate fatty acid sensors [40]. Future studies should focus on the determination of feed intake levels upon fatty acid intragastric administration to confirm changes in the abundance of gastrointestinal hormones observed in the present study. In summary, the present study offers the first set of evidence supporting the presence of mechanisms able to sense fatty acids in the GIT lumen of rainbow trout. The data presented here show clear similarities to the widely accepted mammalian model of fatty acid gut sensing and its involvement in food intake regulation but also suggest several important differences (Figure 6). The most notorious of such differences is probably the lack within the rainbow trout genome of one of the main sensors of MCFAs and LCFAs in mammals, i.e., FFAR4, which appears to be compensated by other receptors binding and responding to these types of fatty acids. Another important difference lies in the ALA-induced modulation of fatty acid sensors and putative response triggered; this observation differs from mammals, in which no activation of fatty acid sensors seems to occur in response to n-3 PUFAs [41,42,43]. The differences between rainbow trout and mammalian fatty acid gut sensing mechanisms may be due to phylogenetical reasons (divergence between mammals and fish) and/or to the different dietary habits between carnivore (rainbow trout) and omnivore (mammalian models assessed so far) species of vertebrates. Further studies are required to study the basis for these differences. Rainbow trout (body weight (bw) = 90 ± 20 g) were obtained from a local fish farm (A Estrada, Spain) and maintained in 100 L tanks (n = 40 fish/tank) with dechlorinated and aerated tap water (15 ± 1 °C) in an open circuit. The photoperiod was set to 12 h light:12 h darkness (12L:12D, lights on at 08:00 h). Fish were fed with a commercial dry pellet diet (proximate analysis: 44% crude protein, 21% crude fat, 2.5% carbohydrates, and 17% ash; 20.2 MJ kg−1 of feed; Biomar, Dueñas, Spain) daily at 11:00 until apparent visual satiety. All studies adhered to the ARRIVE Guidelines, were performed following guidelines of the European Union Council (2010/63/UE) and the Spanish Government (RD 53/2013) for the use of animals in research and were approved by the Ethics Committee of the Universidade de Vigo (00013-19JLSF). Three 48 h-fasted fish were anesthetized in water containing 2-phenoxyethanol (0.02% v/v; Sigma-Aldrich, St. Louis, Missouri, USA) and euthanized by decapitation. Samples from the stomach, pyloric caeca, and intestine (proximal, anterior middle, intermediate middle, posterior middle, and distal; see Supplementary Figure S2 for graphical details) were collected, snap-frozen in dry ice and stored at −80 °C until quantification of the mRNA abundance of fatty acid receptors and transporters by RT-qPCR as described in Section 4.5. This experiment was repeated twice. Following RT-qPCR, representative samples of each tissue were run on 1.5% agarose gels, and single bands for each PCR were purified using QIAquick Gel Extraction Kit (Qiagen, Hilden, Germany) and sent for sequencing (CACTI, University of Vigo, Vigo, Spain). The specificity of the nucleotide-deduced sequences was analyzed using the BLAST tool (https://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastn&PAGE_TYPE=BlastSearch&LINK_LOC=blasthome; accessed on 3 February 2023). This experiment was performed on two consecutive days. For both days, fish scheduled for use in the experiment (maintained in acclimation tanks) fasted for 48 h so that intestinal emptying and basal levels of hormones involved in the metabolic control of food intake were achieved. On day 1, 30 fish were captured in batches of 6 (n = 6 per treatment) and slightly anesthetized with 2-phenoxyethanol (0.02% v/v). Then, intragastric administration of 1 mL. 100 g−1 bw of vehicle (distilled water containing 5% EtOH) alone (control) or containing 50 μmol.mL−1 of octanoate/octanoic acid (Sigma-Aldrich, Cat # C-2875), oleate/oleic acid (Sigma-Aldrich, Cat # O-1008), α-linolenate (ALA, Sigma-Aldrich, Cat # L2376) or sodium butyrate (Sigma-Aldrich, Cat # B5887) was performed. We selected octanoate, oleate, and ALA as representative MCFA, LCFA, and PUFA, respectively, because previous experiments from our research group demonstrated their effectiveness as feed intake modulators and/or modulators of related parameters in rainbow trout or Senegalese sole [44,45]. No available previous studies show a role for SCFAs in the control of feed intake in fish; butyrate was selected as representative in this study among the main SCFAs. To calculate the dose of fatty acid, we based on a typical amount of oleate (selected because we previously reported important effects of this fatty acid on feed intake in rainbow trout [44,46,47]) ingested daily by a trout fed with a standard commercial diet [48]). We then used an equimolar dose for the remaining fatty acids. Administration of treatments was carried out with a 13 cm-long cannula attached to a blunt-tip syringe. Putative regurgitation was checked visually, and we did not observe any during treatment administration. After intragastric treatments, fish from each experimental group were placed in individual tanks for recovery. After 20 min, they were again anesthetized to collect blood samples and, subsequently, plasma, which was used to determine the circulating levels of glucose, lactate, triglyceride, and free fatty acid (see Section 4.4). Then, fish were sacrificed by decapitation, and stomach and intestine (proximal, middle, and distal) samples were collected (see Supplementary Figure S2 for a graphical description of the regions sampled) for RT-qPCR or Western blot analysis; see below). We selected 20 min as sampling time based on preliminary experiments demonstrating this time to be adequate for a dye-containing saline solution to reach the middle/distal intestine after intragastric administration. On day 2, 30 fish per day were captured and intragastrically administered as described above, but sample collection was carried out 2 h post-administration. Plasma levels of lactate, glucose, triglyceride, and free (non-esterified) fatty acid were assessed as indicators of the metabolic status of fish during experiments. Levels of all metabolites were assessed enzymatically using commercial kits adapted to a microplate format (For glucose, lactate, and triglyceride: Spinreact, Barcelona, Spain; for fatty acid: Fuji, Neuss, Germany). Results from these analyses are included in Supplementary Table S2. Levels of all parameters tested showed values comparable to those previously detected in healthy, unstressed individuals of the same species, with no considerable significant differences observed among groups, allowing us to consider that fish used for experiments have an adequate metabolic status and that fish were not exposed to major stress during experiments. Isolation of total RNA from tissues and DNase treatment (n = 6 fish) were carried out using Trizol reagent (Life Technologies, Grand Island, Nebraska, USA) and RQ1-DNAse (Promega, Madison, Wisconsin, USA), respectively, as directed by the manufacturers. Optical density (OD) absorption ratio (OD 260 nm/280 nm) was used as an indicator of RNA purity, and it was determined using a NanoDrop 2000c (Thermo, Vantaa, Finland); only samples with an OD 260 nm/280 nm ratio > 1.8 were used for analysis. Following DNase treatment, 2 μg of total RNA was reverse transcribed into cDNA using Superscript II reverse transcriptase (Promega) and random hexamers (Promega) in a final volumen reaction of 20 μL, following manufacturer’s guidelines. Finally, using specific forward and reverse primers, mRNA abundance was quantified by RT-qPCR using MAXIMA SYBR Green qPCR Mastermix (Life Technologies). Specific primers to ffar1, ffar2b1.1, ffar2b1.2, ffar2b2a, ffar2b2b, ffar2a1b, and ffar2a2 were designed based on rainbow trout cDNA sequences obtained in a previous study of our research group. Among the 10 ffar isoforms described in such a study to be present in the rainbow trout genome, we selected the 7 mentioned above because they are the most abundantly expressed in the trout intestine. Primers to fatp4 were designed from the nucleotide sequence of Salmo salar (GenBank ID: XM_014138749.1) and positively checked for specificity within the rainbow trout genome using Genoscope (https://www.genoscope.cns.fr/trout/; accessed on 3 February 2023). Primers to gpr84, gpr119, cd36, and slc16a1 (gene encoding Mct-1; two isoforms, a and b), as well as those to intracellular signaling elements and gastrointestinal hormones, were designed from rainbow trout nucleotide sequences available on GenBank, using Primer-BLAST online tool (https://www.ncbi.nlm.nih.gov/tools/primer-blast/; accessed on 3 February 2023). All primers used are included in Table 1 and were ordered from IDT (Leuven, Belgium). PCRs were performed in 96-well plates using 1 µL cDNA (replaced by water and RNA for controls) and 500 nm of forward and reverse primers in a final volume of 10 µL. Each sample was run in duplicate wells. All qPCRs were carried out in an iCycler iQ (Bio-Rad, Hercules, California, USA). Cycling conditions for qPCRs consisted of an initial step at 95 °C for 10 min, followed by 40 cycles at 95 °C for 30 s and 60 °C (except for gcg and itpr3, whose annealing temperature is 59 °C, and ffar2b1.1 and ffar2a1b, with an annealing temperature of 62 °C; see Table 1) for 30 s. We included a melting curve (temperature gradient at 0.5 °C/5 s from 65–95 °C) at the end of each run to ensure that a single amplicon was being amplified. R2 of all reactions was 0.97–1, and efficiency was 95–100%. Following PCRs, resulting products were run on 1.5% agarose gels to confirm that a single product of the expected size was being amplified. The relative abundance of target transcripts was calculated using the 2-ΔΔCt method [49], using actb (gene encoding β-actin) and ef1a (gene encoding elongation factor 1α) as reference genes. These two genes were both stably expressed in this experiment. Western blot analysis was performed from tissue samples from 6 fish. Extraction and quantification of protein were carried out as previously described [27]. Then, 50 µg protein was mixed with 4x Laemmli buffer containing 0.2% 2-mercaptoethanol (Bio-Rad) and denatured at 95 °C for 10 min. Then, samples were electrophoresed in Stain-Free 20% acrylamide gels (Bio-Rad) and transferred to a nitrocellulose membrane (0.2 µm pore-size; Bio-Rad) with the use of the Trans-Blot Turbo transfer system (Bio-Rad). After 60 min-blocking using Pierce Protein-Free T20 (PBS) Blocking Buffer (ThermoFisher), a specific primary antibody was added to the membrane and allowed to incubate overnight. Primary antibodies used for detecting gastrointestinal hormones in the stomach and intestine were custom synthesized as rabbit-raised polyclonal antibodies against synthetic peptide synthesized based on rainbow trout sequences (GenScript, Piscataway, NJ, USA). The exact antigen peptide sequences used are as follows: Ghrl: SQKPQVRQGKGKPPC (UniProtKB: Q76IQ4), Cck: CRPSHSQDEDKPEPP (UniProtKB: Q9YGE3), and Pyy: YPPKPENPGEDAPPC (UniProtKB:A0A060X2J5). All antibodies were diluted 1:500. After washing, membranes were incubated with secondary antibody (goat anti-rabbit IgG (H + L) HRP conjugate; Cat # ab205718, Abcam, Cambridge, United Kingdom) diluted to 1:5000. Clarity Western ECL substrate (Bio-Rad) was used to visualize proteins in a ChemiDoc Touch imaging system (Bio-Rad). We quantified protein bands by densitometry using Image Lab software and expressed results relative to the amount of total protein. All data were first checked for homogeneity of variance and normality, and, in case of failure of any of these requirements, they were log-transformed and re-assessed. Then, statistical differences among groups were assessed by one-way ANOVA followed by Dunnett’s test (for the in vivo intragastric experiment) or the Student-Newman-Keuls test (for plasma metabolite levels). Significance was considered when p < 0.05. SigmaPlot version 12.0 (Systat Software Inc., San Jose, CA, USA) statistical package was used to carry out all analyses.

PMC10002233

Catherine Cerutti,Jing-Ru Shi,Jean-Marc Vanacker

Multifaceted Transcriptional Network of Estrogen-Related Receptor Alpha in Health and Disease

21-02-2023

nuclear receptor,transcriptional regulator,target gene,regulatory network,metabolism,cancer

Estrogen-related receptors (ERRα, β and γ in mammals) are orphan members of the nuclear receptor superfamily acting as transcription factors. ERRs are expressed in several cell types and they display various functions in normal and pathological contexts. Amongst others, they are notably involved in bone homeostasis, energy metabolism and cancer progression. In contrast to other nuclear receptors, the activities of the ERRs are apparently not controlled by a natural ligand but they rely on other means such as the availability of transcriptional co-regulators. Here we focus on ERRα and review the variety of co-regulators that have been identified by various means for this receptor and their reported target genes. ERRα cooperates with distinct co-regulators to control the expression of distinct sets of target genes. This exemplifies the combinatorial specificity of transcriptional regulation that induces discrete cellular phenotypes depending on the selected coregulator. We finally propose an integrated view of the ERRα transcriptional network.

Multifaceted Transcriptional Network of Estrogen-Related Receptor Alpha in Health and Disease Estrogen-related receptors (ERRα, β and γ in mammals) are orphan members of the nuclear receptor superfamily acting as transcription factors. ERRs are expressed in several cell types and they display various functions in normal and pathological contexts. Amongst others, they are notably involved in bone homeostasis, energy metabolism and cancer progression. In contrast to other nuclear receptors, the activities of the ERRs are apparently not controlled by a natural ligand but they rely on other means such as the availability of transcriptional co-regulators. Here we focus on ERRα and review the variety of co-regulators that have been identified by various means for this receptor and their reported target genes. ERRα cooperates with distinct co-regulators to control the expression of distinct sets of target genes. This exemplifies the combinatorial specificity of transcriptional regulation that induces discrete cellular phenotypes depending on the selected coregulator. We finally propose an integrated view of the ERRα transcriptional network. In eukaryotes, regulation of gene expression relies on a combinatorial interplay between DNA-binding transcription factors (TFs) and non-DNA binding coactivators or corepressors. Among non-DNA-binding co-regulators, those involved in histone modifications are of importance to control chromatin accessibility and the dynamics of the transcriptional process [1]. The coordinated activity of all these cooperating components results in specific spatiotemporal effects on target gene expression [2,3,4,5]. Pairwise interactions between TFs or between TF and non-DNA binding coactivators can be demonstrated at the protein level [6,7,8]. However, simultaneous cooperative recruitment of more than two transcriptional partners may occur and is currently difficult to demonstrate experimentally. Nuclear receptors (NRs) form a family of transcription factors whose activities are generally controlled by the recruitment of specific, endogenous ligands. NRs are present in all animals and 21 of them have been identified in D. melanogaster vs 48 in H. sapiens. NR proteins are organized in a similar manner. They comprise an N-terminal domain that can contribute to ligand-independent transcriptional activities, a centrally located DNA-binding domain (DBD) containing two Zn fingers, a hinge domain and a C-terminally located Ligand Binding Domain (LBD). The LBD is also involved in receptor homo- or heterodimerization. Furthermore, ligand recruitment induces a conformational change in the LBD that allows interactions with transcriptional cofactors, leading to the modulation of target gene expression. The DBD and, to a lesser extent, the LBD are the most conserved domains of NRs across evolution. The transcriptional activities exerted by NRs also require a large set of proteins to modulate chromatin structure and to recruit the basal transcription machinery. As for most of the TFs, the involvement of cofactors is both dynamic and hierarchical. Primary cofactors have been proposed as those directly binding to NRs to enhance their functions. Secondary cofactors could be those recruited to the promoter through contact with a primary coactivator or corepressor, thus enhancing or inhibiting NR functions, respectively [9]. The Estrogen-Related Receptors (ERRα, β and γ in mammals) form a subfamily of orphan (i.e., lacking an identified natural ligand) NRs. They are expressed in several tissues during embryologic development and in the adult, and display various physiological and pathological functions [10,11,12,13]. The ligand-independent transcriptional activity of ERRs has been noted for several years. This is apparently due to the presence of particular amino acid side chains in the putative ligand binding pocket that lock the LBD in an active conformation and allows constitutive contacts with co-regulators [14]. Due to the lack of natural ligands that directly regulate the activities of ERRs, it is thought that their transcriptional activity is mediated by the recruitment of coactivators and corepressors [14,15,16]. The necessary presence of these cofactors makes their participation in diverse ERR-centered networks instrumental in the cellular effects of ERRs. It should however be mentioned that synthetic compounds have been identified that promote or restrict the transcriptional activities of ERRs. Some crystallographic studies have suggested that these compounds may alter the conformation of the receptors, compromising the recruitment of co-regulators. The most-studied member of the ERR family, ERRα, is involved in various functions related to energy metabolism in tissues displaying high energy demands, such as liver, muscles, adipose tissues or heart [11,17,18]. It has also important roles in osteogenesis, immunity, brain functions and tumorigenesis [19,20,21]. In cancer, ERRα has been shown to control proliferation, metabolism, resistance to hypoxia, angiogenesis and cell migration and invasion [12,22,23,24,25]. Work from various laboratories suggests that these specific ERRα activities may be controlled by interaction with dedicated cofactors expressed in given tissues and that thereby control the phenotypic output of ERRα activities. The main purpose of this review is to provide a detailed understanding of the ERRα transcriptional network derived from work over the past decades. Special emphasis is given to the relationship between the ERRα co-regulators and the cellular functions that they specifically modulate through the receptor. Many efforts have been devoted these last 25 years to dissect the mechanisms underlying the transcriptional activities of NRs. Co-regulatory proteins have been identified, and in general they are components of multi-protein complexes that contain associated chromatin remodeling and/or histone modifying proteins [26]. Co-regulators generally possess LXXLL motifs (where L is leucine and X is any amino-acid residue), also known as nuclear receptor boxes, which enable their interaction with NRs. For most of them, these interactions have been shown to be ligand dependent. For orphan NRs such as HNF-4, Nur77, RORs, TLX or ERRs, a number of papers each focusing on one member of this subfamily have been published suggesting several interacting cofactors, including those that interact with ligand-dependent NRs [26]. Most NRs bind DNA as dimers, either as hetero- or as homodimers, on DNA sequences organized as two half-sites (with AGGTCA as a consensus sequence) with specific orientation and spacing. For instance, the thyroid hormone receptor-RXR heterodimer mostly binds to the so-called DR4 (Direct Repeat 4), with AGGTCAnnnnAGGTCA as a consensus sequence (n is any nucleotide). On another hand, the Estrogen Receptor α (ERα) homodimer binds to AGGTCAnnnTGACCT, an inverse repeat of half sites separated by three nucleotides. In contrast, ERR-response elements (ERREs) are composed of a single half-site generally preceded by a CA-containing sequence. Yet, the sequence-dependent DNA shape of the binding site influences the recruitment of homodimers of ERRα through their DBD. This is achieved by the promotion of the right conformation of the two subunits on DNA for cooperative interaction and dimer stabilization [27]. Furthermore, interaction with a co-activator may induce an allosteric change in ERRα that allows stable dimerization and DNA binding of both receptor DBDs [28]. ERREs are often located at a distance from the Transcriptional Start Site (TSS) of their target genes (about 75% at >1 kb upstream) in introns or in distal intergenic regions where coactivators can be directly recruited by ERRα [29]. Recent literature has suggested that a cooperating TF recruited to the TSS could be important to bridge the ERRE-bound ERRα-cofactor complex to the TSS where their molecular effect is exerted [30] (Figure 1; see below). The p160 coactivator family members (SRC-1/NCoA1, SRC-2/GRIP1/NCoA2, and SRC-3/pCIP/AIB1/ACTR/NCoA3) have been early associated with NRs and the understanding of their importance has grown over time [31]. The three members of this family interact with the activation region of the LBD of NRs and ERRs have been shown to be co-activated by all p160 proteins [32,33]. For instance, in breast carcinoma, NCoA3 has been suggested as a major coactivator of ERRα with binding to the ERREs of ERRα target genes and transcriptional regulation of transfected ERRα-responsive promoters [34]. The p160 coactivators can bind to two other types of coactivators, CREB-binding protein (CBP) and p300, two closely homologous proteins known as p300/CBP family, as well as to p/CAF (p300/CBP-associated factor) [35,36,37]. These proteins act as co-regulators for a wide variety of TFs and also as major lysine acetyltransferases [38,39]. Their various biological functions and how disruption of these functions by mutations and alterations in expression or subcellular localization contributes to cancer phenotype has been more recently reviewed [40]. Although no interaction between ERRs and p300/CBP has been reported, an interaction between ERRα and p/CAF has been revealed in vitro and in mouse liver [41]. This results in the regulation of ERRα transcriptional capacities through unusual mechanisms. Indeed, p/CAF acetylates four Lys residues in the DBD of ERRα. As a consequence, the DNA binding capacity and transcriptional activities of ERRα are strongly reduced. In contrast, SIRT1 and HDAC8 deacetylate the p/CAF-acetylated Lys residues of ERRα resulting in increased DNA binding. The authors also suggest that ERRα acetylation is likely to act in combination with other post-translational modifications, such as phosphorylation or sumoylation, to fine-tune the receptor’s activities. Altogether, this provides an efficient on and off mechanism to regulate the activities of ERRα. Proline, glutamic acid, and leucine-rich protein 1 (PELP1) is a scaffolding protein with several motifs commonly found in co-regulators that has been ascribed to a large number of cellular functions, including regulation of NR signaling and cross-talk [42]. PELP1 has been shown to interact with ERRα and proline-rich nuclear receptor co-regulatory protein 2 (PNRC2) in the transcriptional activation of aromatase in breast cancer cells [43]. Interestingly, PNRC2 has been reported to modulate the transcriptional activation of other NRs, in particular SF1 and ERRγ, suggesting a general feature exerted by this factor [44,45]. Several corepressors have also been described for ligand-regulated, as well as orphan, NRs. Nuclear receptor corepressor (NCoR1) and the highly similar silencing mediator of retinoic and thyroid receptor (NCoR2/SMRT) were the first identified ones, based on their ability to mediate transcriptional repression of thyroid hormone receptor and RARs. These cofactors mediate transcriptional repression by bridging histone deacetylases (HDACs), in particular HDAC3, to NRs in the absence of their corresponding ligands [46,47]. In breast cancer cells, NCoR1 represses a number of negative ERRα –LSD1 targets, but does not act on other (i.e., non-LSD1 dependent) ERRα targets, suggesting a contribution to the regulation of a specific subset of targets [48]. NCoR1 has been identified as a key physiological regulator of muscle mass and function, through its association with MEF2, PPARβ/δ, and ERRα [49]. The receptor-interacting protein 140 (RIP140), encoded by NRIP1 gene, is one of the first proteins that have been identified as recruited by hormone-bound NRs. Strikingly, this protein mainly acts as a direct corepressor of NRs [50] but may alternatively regulate their activity by competing the recruitment of coactivators such as SRC-1 as demonstrated in mammalian cells [51]. RIP140′s activity as a corepressor of ERRs was established as depending on the regulatory elements present in the target promoters [52]. However, this study also suggested that RIP140 can increase the activation exerted by ERRα and ERRγ on SP1-binding sites through a mechanism that involves histone deacetylases. Although this remains to be documented in vivo, this suggests that RIP140 can act on ERRs as a coactivator or as a corepressor, depending on the DNA context. Fifty-eight proteins interacting with ERRα are reported in the BioGRID database (https://thebiogrid.org) that collects protein–protein interactions from a number of experimental studies. The BioGRID-built network includes transcriptional regulators as well as enzymes, structural or RNA-binding proteins (Figure 2). Among the 30 transcriptional regulators, 9 are part of the pre-cited co-regulators. PNRC2 and NCoR1/2 do not appear in the network suggesting that their co-regulatory role is indirect and occurs through another factor. Effects on bone status have been shown for ERRα and ERRγ. Using complete knock-out (KO) mouse models, ERRα was shown as an activator of bone loss during ageing [53,54]. This observation was extended to bone loss resulting from ovariectomy that is also induced by ERRα. This bone loss is indeed abolished in female mice in which ERRα is specifically knocked out in osteoblasts (bone forming cells) [20]. Consistently, it appears that ERRα is an inhibitor of osteoblast differentiation, as observed in vivo and in cell cultures [53,54]. However, the effects of ERRα on osteoblast differentiation are complex and could depend on the presence of the co-regulators PGC-1α and PGC-1β [21]. How ERRα acts in the absence of PGC-1 proteins (as is the case in the early stages of osteoblast differentiation) and through which co-regulators is currently undocumented. Similarly, ERRγ has been identified as anti-osteogenic in bone and pro-osteogenic in the vasculature and no co-regulator was identified that modulates these activities [55,56]. The transcriptional activity of ERRα was also highlighted in the brain with an important regulatory role in response to social challenge in mice [57]. In addition, cross-talk between ERs and ERRs is documented in the brain, as well as a potential beneficial role of ERRα in Alzheimer disease [58,59]. However, no ERRα coactivator has yet been proposed that could document the mechanism of the receptor’s effect in this context. Several studies have shown that ERRα promotes innate host defense. The receptor was first identified as a transcriptional regulator of effector T lymphocytes metabolism [60] and as a transcriptional and post-translational activator of autophagy-related genes via a feed-forward loop with the deacetylase SIRT1 [61]. In this context, ERRα was identified as a target of the NR Nur77 (encoded by the NR4A1 gene) that represses several TFs known to regulate T cell metabolism following activation [62]. ERRα also represses Toll-like receptor (TLR)-induced inflammation [63], mostly via fine-tuning of metabolic reprogramming in macrophages. Again, no data about any transcriptional coactivator of ERRα in this field have currently been published. The first coactivator of ERRα that has been identified is PGC-1α (PPARgamma Co-activator 1α) in the frame of its involvement in mitochondrial energy metabolism [64,65]. PGC-1α has been extensively studied in humans, in health and disease situations. It has been characterized as a master regulator of cellular energy metabolism, including adaptive thermogenesis mediated by multiple transcription factors, such as the NR PPARγ [66]. Work by different laboratories has next shown that ERRα is also instrumental in the activities of PGC-1α in tissues with high energy demand, such as skeletal muscle, heart, liver, or brown adipose tissue [64,67,68]. For instance, ERRα is important for the PGC-1α driven regulation of energy metabolism in cardiac and skeletal muscle [69] and is required for the induction of Ucp1 expression by PGC-1α in the brown adipose tissue [70]. Furthermore, mice lacking ERRα are impaired for thermogenic adaption [71,72]. In addition, PGC-1α also positively regulates ERRα expression thus forming a feed-forward loop [65,73]. PGC-1β has been shown to interact with ERRα and NRF1 to induce several key genes of mitochondrial biogenesis and respiration during differentiation of C2C12 mouse myoblast cells in skeletal myotubes [74]. In mouse heart, ERRα/γ-responsive promoters of metabolic target genes are also enriched for NRF1 as well as for CREB or STAT3 binding sites [75]. Both coactivators PGC-1α and PGC-1β can also mediate the transcriptional activities of ERRα and ERRγ in cancer [76,77]. As such, involvement of the receptors and PGC-1α or β has been documented in the metabolic shift from oxidative to aerobic glycolysis, known as the Warburg effect [78,79]. Several factors have been shown to modulate the transcriptional activity of the ERRα/PGC-1 complex, such as NCoR1, an important modulator of energy metabolism in several tissues. In skeletal muscle, competition between NCoR1 and PGC-1α in the antagonistic regulation of ERRα activity has been proposed for adaptation of oxidative metabolism to physical activity or caloric restriction [80]. The prospero homeobox PROX1 is a TF previously known to regulate the activity of several nuclear receptors, mostly NR5 family members or HNF4A in the liver [81]. PROX1 has been identified as a negative modulator of ERRα/PGC-1α energetic functions in mouse liver [82]. As a TF, PROX1 shares targets with ERRα and interacts directly with PGC-1α [83]. This study also showed a cross-talk between ERRα, PROX1, and BMAL1 (an instrumental factor in the establishment of the circadian cycle) in the rhythmic control of metabolic genes. It should also be reminded that the expression of ERRα is itself under the control of the circadian clock [84]. Together with the association with BMAL, this may provide a reinforcement of the circadian control over gene expression. A direct interaction of the SP1 TF with the three ERRs has been identified in vitro and in human cancer cells. ERRs are able to activate transcription of some targets through SP1 binding sites as pointed above [54]. In addition, the recruitment of the SP1 protein adjacent to ERRα-binding element was identified in muscle cells as one of the conditions preventing the interaction between PGC-1α and ERRα [85]. In cancers from different tissues (breast, ovary, prostate, colon, etc.), a strong ERRα expression has been correlated with a poor prognosis [reviewed in 12,23]. Consistently, important roles of ERRα in the promotion of cancer progression have been documented in the past decades, suggesting that inactivation of the receptor may be beneficial against cancers. For some of these functions, transcriptional co-regulators have been identified. ERRα is involved in the regulation of the proliferation of cell lines derived from breast cancer (such as MCF7 or MDA-MB231) or prostate cancer (LNCaP or PC3) as well as of mammary cancer cells xenografted on Nude mice [86]. Co-regulators of ERRα in this cellular effect have not been investigated. The receptor is also involved in the adaptation of prostate cancer cell lines (LNCaP or PC3) to hypoxia and in the induction of angiogenesis [87]. HIF-1α plays a major role in the regulation of cancer cell metabolism that is reconfigured towards lactate production following aerobic glycolysis through the Warburg effect [88]. ERRs have been shown as essential cofactors of HIF-1α mediating the response to hypoxia [89]. In addition, ERRα physically interacts with HIF-1α in prostate cancer cells [86]. Together, these two factors activate the expression of genes regulating metabolism (such as LDHA or PKM2) and blood vessel growth (such as VEGF or EPO). As developed above, ERRs cooperate with PGC-1 factors in the (dys)regulation of cancer cell metabolism. It has been shown that the ERRα-PGC-1α complex regulates the expression of VEGF in cooperation with HIF-1α. Other activities of ERRα in cancer cells have been documented that do not depend on PGC-1 proteins. This is for instance the case of phenomena contributing to cell migration. These activities have first been identified in physiological situations. This is the case of zebrafish embryonic development where ERRα inactivation reduces cell motility [90], as well as of the capacity of activated macrophages to invade the peritoneal cavity in vivo which is reduced in ERRαKO mice [24]. In cell cultures, ERRα regulates the dynamics of actin network and of focal adhesions which anchor cells to their substrate [25]. These phenomena need to be synchronized for cell migration which is promoted by ERRα [22,24]. In addition, the receptor increases the capacity of cells to invade the extracellular matrix [30,48]. Recent work has shown that ERRα cooperates with different transcriptional coactivators to regulate cell migration and invasion. Our team has identified the histone lysine specific demethylase 1 (LSD1) as an ERRα coactivator that promotes the migration of breast cancer cells [48]. As is the case for ERRα, high expression of LSD1 has been identified by others as a poor-prognosis marker in breast cancers [91,92]. LSD1 can act as a transcriptional repressor by demethylating H3K4me2 or as a transcriptional activator by demethylating H3K9me2 [93,94,95]. The conditions governing the balance between these two activities are unclear to date. However, in vitro experiments have shown that ERRα switches the activities of LSD1 from H3K4me2 to H3K9me2 demethylation [48]. In breast cancer cells, this biochemical activity occurs at the TSS of target genes commonly activated by ERRα and LSD1 and which are involved in the promotion of cell migration. However, the ERRα-LSD1 complex is bound to DNA on enhancer-localized ERREs. Additional work has shown that the NRF1 TF recruits the ERRα-LSD1 complex to the TSS of target genes involved in the regulation of cell invasion [30]. This is for instance the case of MMP1 (Matrix Metalloprotease 1), whose TSS displays an increase in H3K9 methylation in the absence of ERRα or LSD1. As a consequence, inactivation of ERRα, LSD1 or NRF1 leads to reduced degradation of the extracellular matrix and decreased cell invasion, a defect that can be rescued by MMP1 re-expression [24,30,48]. Recent results of the team identified possible transcriptional regulators associated with ERRα using unbiased statistical expression models of ERRα-activated genes across various breast cancer cells [29]. Gene expression modeling is a suitable computational way to propose potential transcriptional regulators, even non-DNA-binding ones, selected from a large set. Indeed, using RNA expression data, those regulators contributing to the expression of some genes in association with a specific TF gene, as ERRα, are potential co-regulators of this TF. Interestingly, this approach can unveil several potential co-regulators in the same experimental context. Among those identified, DDX21, MYBBP1A, NFKB1, and SETD7 were validated in breast cancer cells as modulators of some ERRα-activated genes. Moreover, SET7 was further confirmed as a transcriptional partner of ERRα, with which it physically interacts and regulates the expression of target genes involved in cell motility. Consistently, both factors are necessary to induce orientated cell migration. It is likely that all transcriptional co-regulators of ERRα have not been characterized to date. However, the identified co-regulators are very diverse as are the cellular functions that are regulated by ERRα in association with a particular cofactor (see Figure 3). The enhancer of zeste homolog 2, EZH2, is a subunit of the Polycomb repressor complex 2 that acts as a histone methyltransferase. Several lines of evidence have implicated EZH2 in the development and progression of a variety of cancers and it has become a potential therapeutic target [96,97]. Functional interaction of EZH2 with ERRs was evidenced in gastric and breast cancers [98,99]. Indeed, EZH2 binds to all ERR promoters and represses their expression. Combined treatment of an EZH2 inhibitor and ERRγ agonist displays a synergistic suppressive effect on gastric cancer progression [98]. In breast cancer, EZH2 was identified as a regulator of ERRγ activities in a methyltransferase-dependent manner [99]. Various post-transcriptional or post-translational modifications exerted by diverse compounds can affect the stability and transcriptional activity of ERRα. Detailed knowledge on this topic is summarized in a recent review on the regulation of the expression of ERRs [100]. Reduction of ERRα transcript abundance induced by various miRNAs (miR-125a, miR-137, miR-135a, miR-497) has been shown during adipocyte differentiation or cancer cell migration [101,102,103,104]. In contrast, stabilization of the ERRα protein by high levels of LSD1 has been shown in breast cancer cell lines [105]. LSD1 protects ERRα from proteasome-dependent degradation, independently of its demethylase activity, and without any effect on ERRα mRNA. In addition, some synthetic compounds modulate the transcriptional activities of ERRs such as bisphenol A, diethylstilbestrol, and 4-hydroxytamoxifen that act mostly on ERRγ and ERRβ as agonists or antagonists [106]. However, these compounds all display minimal activity on ERRα. Proteasome-dependent degradation of ERRα can be induced by its synthetic inverse agonist XCT790 that also blocks the transcriptional activity of the receptor [107]. This degradation effect is potentiated by the ERα antagonist ICI182,780 (also known as Fulvestrant), that is used in the treatment of certain breast cancers. The transcriptional activity of ERRα relies on the recruitment of a number of co-regulators that have been identified in various tissues or cells and in health or disease states. In addition, various in vitro and in vivo experiments were used to disclose the interaction between ERRα and its co-regulators as well as their effects on ERRα target genes. The identified target genes mostly differ across studies, possibly reflecting the different tissues/cells that were used to characterize these targets. This suggests that each ERRα-co-regulator complex modulates the expression of specific sets of genes and thus exerts selective phenotypic effects. PGC-1s are the main ERRα coactivators turning the effects of ERRα towards cellular energy metabolism in healthy tissues with high-energy demand or in tumors. The role of ERRα in cancer progression appears mediated by different coactivators to control other specific target genes. Figure 4 recapitulates the target genes identified from PGC-1 and LSD1 studies (see all the co-regulators and their identified target genes in Table 1 and Figure S1). In summary, orphan nuclear receptors ERRs represent a complex model of transcriptional regulation. In particular for ERRα, the diversity of co-regulators reflects the diversity of target genes as part of a diversity of cellular functions. However, co-regulators have been identified one at a time in various conditions and we are still lacking a global view for a given condition. Notably, our knowledge on the multiple ERRα coactivators and their physical interaction or cooperation needs to be improved. Because multiple-protein interactions cannot yet be simultaneously investigated with current molecular or cellular biology approaches, this challenging issue could be first addressed by computational approaches [108,109]. In addition, the recruitment of numerous transcriptional coactivators occurs through a specific dynamics that remains to be explored for ERRs [110]. Such new data could lead to a better understanding of the different mechanisms underlying the effects of ERRα for designing new context-specific compounds in the treatment of diseases in which ERRα is expressed and involved.

PMC10002278

Lulu Liu,Lu Qiu,Yaqian Zhu,Lei Luo,Xinpei Han,Mingwu Man,Fuguang Li,Maozhi Ren,Yadi Xing

Comparisons between Plant and Animal Stem Cells Regarding Regeneration Potential and Application

23-02-2023

plant regeneration,animal regeneration,stem cell,molecular mechanism,regeneration applications

Regeneration refers to the process by which organisms repair and replace lost tissues and organs. Regeneration is widespread in plants and animals; however, the regeneration capabilities of different species vary greatly. Stem cells form the basis for animal and plant regeneration. The essential developmental processes of animals and plants involve totipotent stem cells (fertilized eggs), which develop into pluripotent stem cells and unipotent stem cells. Stem cells and their metabolites are widely used in agriculture, animal husbandry, environmental protection, and regenerative medicine. In this review, we discuss the similarities and differences in animal and plant tissue regeneration, as well as the signaling pathways and key genes involved in the regulation of regeneration, to provide ideas for practical applications in agriculture and human organ regeneration and to expand the application of regeneration technology in the future.

Comparisons between Plant and Animal Stem Cells Regarding Regeneration Potential and Application Regeneration refers to the process by which organisms repair and replace lost tissues and organs. Regeneration is widespread in plants and animals; however, the regeneration capabilities of different species vary greatly. Stem cells form the basis for animal and plant regeneration. The essential developmental processes of animals and plants involve totipotent stem cells (fertilized eggs), which develop into pluripotent stem cells and unipotent stem cells. Stem cells and their metabolites are widely used in agriculture, animal husbandry, environmental protection, and regenerative medicine. In this review, we discuss the similarities and differences in animal and plant tissue regeneration, as well as the signaling pathways and key genes involved in the regulation of regeneration, to provide ideas for practical applications in agriculture and human organ regeneration and to expand the application of regeneration technology in the future. Animals and plants are subjected to a variety of stimuli during their life span that can cause tissue damage. Both animals and plants promote tissue regeneration through adult stem cells or by the induction of stem cell differentiation to maintain their lives [1]. Tissue regeneration refers to the continuous renewal of biological tissues, the re-differentiation of existing adult tissues to produce new organs, or the repair process after tissue damage. It is one of the phenomena of biological life [2,3]. As plants are sessile, they face various challenges in the external environment. Both lower and higher plants have dramatic regenerative capacities. The super-regenerative capacity of plants is important for maintaining their survival [4]. The regenerative capacity of animals is species-specific. For example, planarians can regenerate whole bodies from tissue fragments of almost any part of the body [5,6,7]. Amphibians such as salamanders can also completely regenerate lost organs and limbs, such as the legs, gills, tail, retina, spinal cord, and heart [8,9]. Although the zebrafish is a vertebrate, it has dramatic regenerative capacity and is, therefore, often used as a model of organ regeneration. Zebrafish can regenerate their hearts, livers, spinal cords, and caudal fins [10,11,12]. Humans, however, can only regenerate intestinal cells, skin, and bones, either continuously or periodically [13,14]. Regeneration of animals and plants is dependent upon stem cells. Stem cells undergo differentiation and division to form the tissues or organs required by animals and plants. Plant stem cells mainly exist in the meristem, upon which the formation of plant organs is reliant [15,16,17]. The existence of meristems ensures plasticity in the growth and development of plants [18]. Plant regeneration is mainly regulated by auxin and cytokinin signaling [19]. In animals, Wnt/β-catenin, Hedgehog (Hh), Hippo, Notch, Bone Morphogenetic Protein (BMP), Transforming growth factor-beta (TGF-β), and other signaling pathways regulate animal tissue regeneration [20,21]. Interestingly, the target of rapamycin (TOR) plays an important regulatory role in both animal and plant regeneration. In plants, TOR is involved in the regulation of roots, stem growth, and callus formation [22,23,24], and in animals, TOR is a central hub for integrating nutrients, energy, hormones, and environmental signals [25,26]. Cell growth and cell cycle progression are generally tightly connected, allowing cells to proliferate continuously while maintaining their size. TOR is an evolutionarily conserved kinase that regulates both cell growth and cell cycle progression coordinately [27]. Stem cells and their metabolites have great application value in agriculture and regenerative medicine. The advancement in regenerative medicine benefits human health, and it has great prospects in the medical field [28]. Stem cells can be regarded as ideal seed cells for genetic engineering, able to the repair damaged tissues and organs and to overcome immune rejection. In this review, we discuss the regeneration mechanisms of animals and plants, highlighting the similarities and differences between these biological processes. Additionally, we summarize the main recent findings on animal and plant stem cells in the field of regeneration, and provide new ideas and directions for the protection of endangered species and the development of regenerative medicine. Plants have the remarkable ability to drive cellular dedifferentiation and regeneration [19]. However, the regenerative capacity of animals varies greatly across different species. Invertebrates and amphibians generally have a high regenerative capacity [29]. In contrast, the regeneration capacity of vertebrates, such as mice, is relatively weak [30,31,32]. Whether the research subject is a planarian with strong regenerative capacity or a human with weak regeneration capacity, the fundamental mechanism of regeneration is the differentiation of stem cells into the damaged/missing tissues. The regeneration processes of animals and plants have certain similarities. Firstly, they can be divided into the same levels of regeneration, including cell, tissue, structural, organ, and systemic regeneration [33]. Secondly, in both plants and animals, injury is the main stimulus for the formation of specialized wound tissue that initiates regeneration. A regenerative response from these organisms can be elicited by environmental insults, such as pathogens or even predatory attacks. Amputation in animals is usually, but not always, followed by the formation of a specialized structure known as a regeneration blastema. This structure consists of an outer epithelial layer that covers mesodermally derived cells, inducing a canonical epithelial/mesenchymal interaction, a conserved tissue relationship central to the development of complex structures in animals [34]. In plants, a frequent, but not universal, feature of regeneration is the formation of a callus, a mass of growing cells that has lost the differentiated characteristics of the tissue from which it arose. A callus is typically a disorganized growth, arising on wound stumps and in response to certain pathogens. One common mode of regeneration is the appearance of new meristems within callus tissue. Therefore, the plant callus and animal blastema share the characteristics of being specialized yet undifferentiated structures capable of regenerating new tissues [4]. Moreover, the process of stem cell regeneration induced by somatic cells in plants is similar to that induced by animal pluripotent stem cells. In animals, the production of induced pluripotent stem cells (iPSC) depends on the expression of many key transcription factors. Similar to animal cells, the induction and maintenance of stem cells in plants also depend on the induction and expression of several key transcription factors, such as class B-ARR, WUSCHEL (WUS), or WUSCHEL RELATED HOMEOBOX5 (WOX5). Therefore, the stem cells induced in plants that express the pluripotent genes such as WUS or WOX5 can also be called plant iPSC [35]. In addition, the regeneration of animals and plants requires the participation of stem cells. The regenerative capacity of animals and plants varies greatly. Generally speaking, the regenerative capacity is weak in higher animals, and varies greatly between body parts (Figure 1). The skin, as well as other microorgans and tissues of animals, have relatively fast renewal speeds and strong regeneration capacities [36]. The regeneration capacity of the heart, stomach, and other organs is weak, whereas that of the liver is relatively strong [37]. Unlike certain nerve tissues that still retain axonic connections, animal nerve cells have almost no regenerative capacity; therefore, certain types of brain cell damage and senile dementia are irreversible and can only be repaired via stem cell treatment [38]. The regenerative ability of plants is generally stronger than that of animals, but also vary greatly between species. For example, the regenerative capacities of Taxus chinensis, Metasequoia glyptostroboides, and Ginkgo biloba are relatively weak, whereas those of lower plants, such as Ficus virens, Laminaria japonica, and Undaria pinnatifida, are relatively strong [39]. Stem cells are divided into totipotent stem cells, pluripotent stem cells, and unipotent stem cells [40]. The distribution of animal and plant stem cells is also quite different. In plants, stem cells existing in the shoot apical meristem (SAM) and root apical meristem (RAM) are pluripotent, and plant stem cells mainly exist in the meristem of plants for a long time [41]. Meristems can differentiate into vegetative tissues, protective tissues, conducting tissues, mechanical tissues, secretory tissues, and other plant cell populations with identical physiological functions and morphological structures to form vegetative and reproductive organs of plants [42,43]. In addition, plants can also produce calluses, which are similar to stem cells, and are the tissue formed by somatic cells in response to injury and dedifferentiation [19,44,45]. There is often a lack of stem cell aggregation in animal tissues; however, they are widely distributed in various tissues and organs, though in small numbers [46]. In addition, due to the differences in evolution, there are significant differences in the signal pathways and regulators regulating plant and animal regeneration (Table 1 and Table 2). In plants, a feedback regulation pathway is formed between WUS and CLAVATA (CLV), which regulates the steady state of stem cells in stem tips [47]. The SHORTROOT (SHR)-SCARECROW (SCR) signaling pathway plays a key role in maintaining apical meristems [48,49]. In animals, the Wnt and Notch classical signaling pathways regulate self-renewal of hematopoietic, intestinal epithelial, skin, and neural stem cells [50]. There are great differences in the regenerative capacities of animals and plants, and the involved signaling pathways are also different. Even in plants, the transcription factors and signal pathways regulating SAM and RAM regeneration vary [80]. SAM is formed in the early stage of embryonic development and is structurally divided into the central zone (CZ), rib zone (RZ), and peripheral zone (PZ). The CZ region is composed of pluripotent stem cells in an undifferentiated state, with a long cell division cycle; the RZ region provides cell support for the vascular meristem; and the PZ region is the core region for further cell division, differentiation, and development into lateral organs [81]. In SAM, STM and WUS are essential for stem cells to remain undifferentiated [57]. STM can inhibit the differentiation while maintaining the proliferation of meristem cells, and can also integrate mechanical signals that play a role in the formation of lateral organs [82]. Plant stem cells require induction niches. In SAM, this role is played by cells located in the organizing center (OC). At the molecular level, the OC is defined by highly localized expression of the homeodomain transcription factor WUS [83]. WUS fluidity is highly directional, but its specific mechanism has not yet been elucidated. CLV3, as a major stem cell-derived signal, connects WUS with STM. In Arabidopsis, WUS and STM form heterodimers and combine with the promoter region of CLV3, ensuring a stable number of stem cells [56]. CLV3 is a short secretory peptide modified after processing and translation. CLV3 peptides diffuse in the interstitial space and act by binding with a group of related leucine rich repeat (LRR) receptor complexes found on the plasma membrane [84]. The joint action of these receptors is to combine with CLV3 to activate intracellular signaling cascades. The net effect of CLV signaling is reduced WUS expression, defining a local negative feedback loop to induce WUS migration from the OC to stem cells in order to maintain their fate [85]. In addition, STM gene expression depends on WUS, and WUS-activated STM expression enhances WUS-mediated stem cell activity (Figure 1) [47,56]. In addition, in SAM, the local regulatory system appears insufficient to synchronize stem cell behavior without developmental or environmental input. Communication between peripheral developmental organs and central stem cells in SAM is mainly controlled by phytohormones, among which auxin and cytokinin have the greatest impact [86]. Cytokinin acts as a cell cycle inducer and is important for WUS activation, while auxin mainly triggers peripheral differentiation [87]. Interestingly, auxin also enhances the output of cell division proteins by directly inhibiting the expression of negative feedback regulators of cytokinin signal transduction [86]. Recent studies have found that TOR kinases play a central role in metabolism, light-dependent activation of WUS, and stem cell activation in SAM [23]. RAM is mainly regulated by the auxin-dependent PLT pathway and the auxin-independent SHR/SCR pathway [88,89]. Key transcription factors such as SHR, SCR, and PLT1/2/3/4 play a crucial role in the organization and maintenance of RAM. SCR is expressed in the static center and endothelium, and SHR is expressed in the periapical stele cells. Both are necessary to maintain static center function and jointly provide signals for the stem cell microenvironment [48,49]. In addition, PLTs strongly affect the characteristics, cell expansion, and differentiation of stem cells and RAM by forming gradients which depend on the stability and movement of PLT proteins [90,91]. PLTs and auxin gradients are correlated, but also partially independent (Figure 1A) [92,93]. The regeneration process includes tissue repair, de novo organ regeneration, the formation of wound-induced calluses, and somatic embryogenesis. Root tip repair involves a wounding response, redistribution of auxin and cytokinin, reconstruction of the quiescent center (QC), and stem cell niche re-establishment [94]. Studies have found that damage-induced jasmonic acid (JA) signaling can also activate stem cells to promote regeneration, and JA signaling regulates the expression of the RETINOBLASTOMA-RELATED (RBR)-SCR molecular network and stress response gene ERF115 to activate the root stem cell tissue center, thereby promoting root regeneration. Auxin activates WUSCHEL RELATED HOMEOBOX11/12 (WOX11/12) to transform root-initiating cells into the root primordium. During this process, the expression level of WOX11/12 decreases, whereas that of WOX5/7 increases. The WOX11/12 protein directly binds to the WOX5/7 promoter to activate its transcription, whereas WOX5/7 mutation leads to defects in primordium formation [65]. At the genetic level, the highly specific and QC-expressed gene WOX5 delineates QC identity and maintenance [95]. WOX5 activity most likely occurs through direct effect on cell cycle regulators. Plants with disrupted expression levels of WOX5 show aberrant differentiation rates of the distal stem cells, indicating the role of WOX5 in preventing stem cell differentiation [96]. In contrast to SAM, where auxin triggers differentiation, hormones need to specify niches and maintain cell proliferation in RAM. Cytokinin mainly acts far away from the root tip and promotes differentiation through mutual inhibition with auxin [97]. However, cytokinins have also been shown to counteract the unique properties of QC cells by reducing auxin input from the surrounding environment and inducing cell division [98]. Maintaining stem cell homeostasis in the stem and root niches is essential to ensure that sufficient numbers of new cells are generated to replace removed cells, as well as the proper differentiation and growth and formation of new tissues and organs. It is worth noting that RBR protein is a plant homologue of RB (a tumor suppressor protein) and plays a crucial role in SAM and RAM [99,100]. Like in animals, RBR in plants inhibits cell cycle progression by interacting with E2F transcription factor homologues. In addition, decreased RBR levels lead to increased numbers of stem cells, while increased RBR levels lead to stem cell differentiation, indicating that RBR plays an important role in stem cell maintenance. At present, RBR is a protein known to be involved in stem cell function, and is conserved between the animal and plant kingdoms [1]. Interestingly, TOR not only plays a role in SAM stem cell activation, but also promotes QC cell division in RAM (Figure 1A) [101]. De novo root regeneration is the process by which adventitious roots form from wounded or detached plant organs. Auxin is the key hormone that controls root organogenesis, and it activates many key genes involved in cell fate transition during root primordium establishment [102]. The detached leaves of Arabidopsis thaliana can regenerate adventitious roots on hormone-free medium [103]. From 10 min to 2 h after leaf detachment, a wave of JA is rapidly produced in detached leaves in response to wounding, but this wave disappears by 4 h after wounding [104]. JA activates the expression of transcription factor gene ERF109 through its signaling pathway, which, in turn, up-regulates the expression of ANTHRANILATE SYNTHASE α1 (ASA1). ASA1 is involved in the biosynthesis of tryptophan, a precursor of auxin production. After 2 h, the concentration of JA decreased, resulting in the accumulation of JAZ protein, which could directly interact with ERF109 and inhibit ERF109, thus turning off the wound signal. In general, the post-injury JA peak promotes auxin production and, thus, promotes root regeneration from the cuttings. Root organogenesis also requires a strict turning-off of the JA signal [105]. Callus formation is one of the most important methods of plant regeneration. Studies have analyzed why calluses have regenerative capacity. Through single cell sequencing of Arabidopsis hypocotyl calluses, researchers confirmed that calluses are similar to the root primordium or root tip meristem, and can be roughly divided into three layers: the outer cells are similar to the epidermis and root cap of the root tip, the middle layer cells to the quiescent center (QC), and the inner cells to root tip initial vascular cells. It was found that middle layer cells of calluses had highly similar transcriptome characteristics to the QCs of root tip resting centers, and were also source stem cells for root and bud regeneration [59]. AAR12, of the cytokinin signal transduction pathway, is the main enhancer of callus formation [62]. APETALA2/ETHYLENE RESPONSE FACTOR (AP2/ERF) transcription factors, such as WIND1, ERF113/RELATED TO AP2 L (RAP2.6L), ESR1, and ERF115, in Arabidopsis thaliana are key regulators of rapid post-traumatic-induced regeneration when wounded. Wounding upregulates cytokinin biosynthesis and signal transduction, thereby promoting cell proliferation and callus formation [60,106,107,108]. WIND1 can promote callus formation and shoot regeneration by upregulating ESR1 (Figure 1A) [45]. Plants can undergo multiple regenerative processes after wounding to repair wounded tissues, form new organs, and produce somatic embryos [109]. Plant somatic embryogenesis refers to the process by which somatic cells produce embryoids through in vitro culture [110]. This process can occur directly from the epidermis, sub-epidermis, cells in suspension, protoplasts of explants, or from the outside or inside of a callus formed from dedifferentiated explants. The transformation from somatic cells to embryogenic cells is the premise of somatic embryogenesis. In this process, the isolated plant cells undergo dedifferentiation to form a callus. The callus and cells undergo redifferentiation into different types of cells, tissues, and organs, and finally generate complete plants [111]. This process involves cell reprogramming, cell differentiation, and organ development, and is regulated by several transcription factors and hormones [112]. For example, the WUS gene regulates the transformation of auxin-dependent vegetative tissues to embryonic tissues during somatic embryogenesis [113,114]. Overexpression of WUS can induce somatic embryogenesis and shoot and root organogenesis. Ectopic expression of the WUS gene can dedifferentiate recalcitrant materials that do not undergo somatic embryogenesis easily to produce adventitious buds and somatic embryos [115]. Additionally, LEAFY COTYLEDON 1 (LEC1), highly expressed in embryogenic cells, somatic embryos, and immature seeds, can promote somatic cell development into embryogenic cells. Furthermore, LEC1 can maintain the fate of embryogenic cells at the early stage of somatic embryogenesis. At present, LEC1 is used as a marker gene for somatic embryogenesis in several species [116]. Unlike LEC1, LEC2 can directly induce the formation of somatic embryos, which may activate different regulatory pathways [117]. In recent years, through research on animals with strong regeneration capacities, such as planarians, leeches, and salamanders, it was found that the early stages of regeneration are jointly regulated by cell death/apoptosis-related genes, MAPK signal-related genes, and EGR [118]. In plants, programmed cell death (PCD) plays crucial roles in vegetative and reproductive development (dPCD), as well as in the response to environmental stresses (ePCD) [119,120]. Sexual reproduction in plants is important for population survival and for increasing genetic diversity. During gametophyte formation, fertilization, and seed development, there are numerous instances of developmentally regulated cell elimination, several of which are forms of dPCD essential for successful plant reproduction [121]. In the late stages of regeneration, many signal pathways participate in cell proliferation and regulation of various responses. The Wnt signaling pathway is widely distributed in invertebrates and vertebrates, and is a highly conserved pathway during evolution. Wnt signaling plays an important role in early embryonic development, organ formation, tissue regeneration, and other physiological processes [69,122,123]. Wnt proteins are a family of 19 highly conserved secretory glycoproteins that act as ligands for several receptor-mediated signaling pathways, including those that regulate processes throughout development [123]. The classic Wnt signaling pathway is mainly mediated by β-catenin. β-catenin is a multifunctional protein which helps cells respond to extracellular signals and influences by interacting with the cytoskeleton [124]. When Wnt binds to its membrane receptor, Frizzled (FZD), it activates the intracellular protein Dvl. Dvl receives upstream signals in the cytoplasm and is the core regulator of the Wnt signaling pathway. Wnt inhibits the function of the β-catenin degradation complex formed by APC, AXIN, CK1, glycogen synthase kinase 3β (GSK3β), and other proteins, thus stabilizing β-catenin in the cytoplasm. Stably accumulated β-catenin in the cytoplasm enters the nucleus and binds to the TCF/LEF transcription factor family to initiate the transcription of downstream target genes, such as c-Myc and cyclin D1, in order to promote regeneration. TCF/LEF transcription factor’s association with β-catenin initiates the expression of key genes in the multiple Wnt signaling pathways [50,125]. The Wnt signaling pathway is important for human development and the maintenance and regulation of adult stem cells, but improper Wnt activation can lead to carcinogenesis [126]. For example, in the differentiation of mouse embryonic stem cells (mESC), Wnt activation of β-catenin signaling inhibits myocardial differentiation and promotes endothelial and hematopoietic lineage differentiation. During vertebrate embryonic development, Wnt activation induces ESCs to enter the anterior and posterior lamellar mesoderm (LPM). In pre-LPM, Dickkopf (Dkk) is secreted from the endoderm, preventing Wnt from binding to its receptor and leading to the induction of the cardiogenic mesoderm and the formation of cardiac progenitor cells (CPC) (Figure 1B) [123]. Similar to Wnt signaling, Notch signaling is a highly conserved signaling pathway that is widely involved in various regeneration processes in different organs, such as the tail fin, liver, retina, spinal cord, and brain [127]. Notch signaling also plays an important role in the self-renewal and differentiation regulation of stem cells. In stem cell biology, Notch signal transduction is highly environmentally dependent, and the biological consequences of pathway activation vary from maintaining or expanding stem cells to promoting stem cell differentiation [128]. Researchers found that Notch receptors and ligand expression were up-regulated during zebrafish fin regeneration in 2003, and many studies have also shown that Notch signaling plays a key role in fin repair, regulating venous arterialization, and cell proliferation and differentiation [129]. Notch signaling can also regulate duct cell accumulation and biliary tract differentiation, promote the expansion and differentiation of liver progenitor cells, and antagonize Wnt signaling during liver regeneration. However, different Notch receptors have different effects on hepatocytes, confirming the complex functions of Notch signaling in the treatment of liver diseases [130]. Notch signaling is mediated by the interaction between Notch ligands and receptors in adjacent cells. There are four kinds of Notch receptors (Notch1-4) in mammals, which are composed of three parts: the extracellular domain (NEC), transmembrane domain (TM), and intracellular domain (NICD). The Notch protein is cleaved three times, and its NICD is released into the cytoplasm and enters the nucleus to bind to the transcription factor CBF-1, suppressor of hairless, Lag (CSL) to form a transcriptional activation complex. The CSL protein is a key transcriptional regulator in the Notch signaling pathway, which is also known as the classical Notch signaling pathway or the CSL-dependent pathway. It activates the Hairy Enhancer of Split (HES), Hairy, and Enhancer of split-related genes with the YRPW motif (HEY), homocysteine-induced ER protein, and other basic helix–loop–helix (bHLH) transcription factor families of the target genes [131,132]. For example, Notch signaling can enhance bone regeneration in the mandibles of zebrafish, and is reactivated after valvular damage in zebrafish larvae and adults, which is necessary in the initial stage of heart valve regeneration (Figure 1B) [133]. In addition, the more conserved Hh pathway also plays a key role in adult tissue maintenance, renewal, and regeneration [134]. The Hh protein has been identified in many animals, from jellyfish to humans. Drosophila has only one Hh gene, while vertebrates have 3–5. All Hh proteins are composed of the N-terminal “Hedge” domain and the C-terminal “Hog” domain. The Hedge domain mediates protein signaling activity. The Hog domain can be further subdivided into the N-terminal Hint domain and the C-terminal sterol recognition region (SRR). The N-terminal Hint domain is sequentially similar to the self-splicing intron, and the C-terminal SRR binds to cholesterol [135]. Hh signal transmission is mediated by two receptors on the target cell membrane, Patched (Ptc) and Smoothened (Smo). The receptor Smo is encoded by the proto-oncogene Smoothened and is homologous to the G-protein-coupled receptor. It is composed of a single peptide chain with seven transmembrane regions. The N-terminal is located outside the cell, and the C-terminal is located inside the cell. The amino acid sequence of the transmembrane region is highly conserved [136]. The serine and threonine residues at the C-terminal are phosphorylated sites. When protein kinase catalyzes, it binds phosphate groups. The members of this protein family have the function of a transcription promoter only when they maintain their full length and start the transcription of downstream target genes. When the carboxyl end is hydrolyzed by the proteasome, a transcription inhibitor is formed to inhibit the transcription of downstream target genes. Smo is a necessary receptor for Hh signal transmission. Glioma-associated oncogene transcription factors (GLI) are transcriptional effectors of the Hh pathway. Stimulated by Hh signal transduction activation, GLI proteins are differentially phosphorylated and processed into transcriptional activators that induce the expression of Hh target genes to initiate a series of cellular responses, such as cell survival and proliferation, cell fate specification, and cell differentiation [137,138]. A previous study found that Hh signaling mediates liver regeneration by regulating DNA replication and cell division. Treatment of mice with Hh inhibitors caused a slowing of cell proliferation and mitotic arrest, which led to the inhibition of liver regeneration. Mice treated with the Hh inhibitor vismodegib showed inhibited liver regeneration, accompanied by significant decreases in the expression of Hh-inducible factors GLI1 and GLI2 (Figure 1B) [139]. The Hippo signaling pathway is a major regulator of cell proliferation, tissue regeneration, and organ size control [132]. Hippo is highly conserved in mammals, controlling development and tissue organ homeostasis; imbalances can lead to human diseases such as cancer [140]. The core of the Hippo pathway is the kinase cascade; that is, mammalian STE20-like1/2 (Mst1/2) (Hippo homolog) and Salvador 1 protein (SAV1) form a complex that phosphorylates and activates large tumor-suppressing kinases (LATS1/2). LATS1/2 phosphorylates and inhibits transcription coactivators such as Yes-associated proteins (YAP) and transcriptional coactivators with PDZ-binding motifs (TAZ) [141]. LATS1/2 is a protein kinase that plays an important role in the Hippo signaling pathway, and exhibits anticarcinogenic activity. LATS1/2 deletion enhances TAZ/YAP activity and directly activates oncogene expression [142]. During tissue damage, the activity of YAP, the main effector of the Hippo pathway, is instantaneously induced, which in turn promotes the expansion of tissue-resident progenitor cells and promotes tissue regeneration [143]. Recent animal model studies have shown that the induction of endogenous cardiomyocyte proliferation is crucial for cardiac regeneration, and inhibition of Hippo signaling can stimulate cardiomyocyte proliferation and cardiac regeneration [144]. TGF-β superfamily signal transduction plays an important role in regulating cell growth, differentiation, and development in many biological systems [145]. TGF-β signaling phosphorylates Smad proteins and transports them to the nucleus. Activated Smad proteins regulate a variety of biological processes by binding to transcription factors, leading to cell state-specific transcriptional regulation [146]. For example, TGF-β signaling in zebrafish promotes cardiac valve regeneration by enhancing progenitor cell proliferation and valve cell differentiation. In addition, TGF-β superfamily members also play important roles in the steady renewal and regeneration of the adult intestine (Figure 1B) [147,148]. TOR signaling pathways are present in both animals and plants, and are also associated with regeneration. Plant growth is affected by light and glucose, which are known activators of the TOR pathway [149]. The TOR signaling pathway is involved in root and stem growth and callus formation, and TOR phosphorylates downstream cell cycle factor E2Fa to promote these processes [22,24]. Moderate expansion of the Akt gene in animals activates the mTOR signaling pathway and promotes cell proliferation [150]. GSK3β is a direct substrate of Akt and is inhibited by Akt during animal regeneration [151]. BR-INSENSITIVE 2 (BIN2) was the first plant GSK3-like kinase to be characterized by genetic screening. The kinase domain of the GSK3-like kinase found in Arabidopsis and rice has 65–72% sequence homology to human GSK3β [25,152]. Biochemical and genetic analyses have confirmed that BIN2 plays a negative role in BR signal transduction and the regulation of cell growth. However, in plants, it was found that TOR can regulate the phosphorylation level of the neglected ribosomal protein S6 kinase beta 2 (S6K2), and S6K2 can interact with BIN2 to directly phosphorylate BIN2 and regulate plant growth [153]. The conserved characteristics of TOR signaling in the normal physiology and regeneration of animals and plants suggest its important role in maintaining normal physiological homeostasis of animals and plants. The growth and development of animals and plants is a process of differentiation from pluripotent stem cells (fertilized eggs) to pluripotent stem cells, and then to specialized stem cells [154]. On the contrary, the terminally differentiated cells of animals and plants carrying complete genetic material also have the potential to transform into stem cells. In plants, somatic cells can restore their totipotency through dedifferentiation and regenerate intact plants. Consistent, in the study of animal cell dryness, it was also found that four transcription factors, octamer binding transfer factor 4 (Oct4), SRY box transfer factor 2 (Sox2), Kruppel like factor 4 (Klf4), and c-Myc, were transferred into mouse fibroblast cells, which can cause them to become iPSC. This discovery indicates that immature cells can develop into all types of cells [155,156]. Stem cells and their metabolites, from both plants and animals, are widely used in agriculture, animal husbandry, and regenerative medicine (Figure 2). Plant totipotent stem cells have good application potential in crop breeding. The totipotent stem cells of animals in the placenta can be cryopreserved to treat some diseases after adulthood. Plant pluripotent stem cells and their metabolites can be used in the development of drugs, health foods, and cosmetics. For animals, iPSC can produce various necessary organs, but at present, due to ethical constraints, artificial organs have not been allowed [154]. Artificial meat that can be made from animal multipotent stem cells can also be used for pet disease treatment. The unipotent stem cells of plants are also used for the extraction of some pigment substances. In addition, purple shirt stem cells in a suspension culture can produce anti-cancer substances such as Taxamairin A and B, and the unipotent stem cells in milk have therapeutic potential in treating some animal diseases [157]. In agriculture, plant genetic transformation and callus culture are key processes in crop gene editing and breeding [158]. A previous study found that overexpression of the wheat WUSCHEL family gene TaWOX5 can significantly improve transformation efficiency, and that callus culture can aid wheat transgenics [159]. In Arabidopsis, the injury-inducing factor WIND1 can promote callus formation and bud regeneration by upregulating Arabidopsis ESR1 expression, and the esr1 mutant shows defects in callus formation and bud regeneration [61]. This finding is of great significance for in vitro plant tissue culture. Regenerating adventitious roots from cuttings is a common plant clonal reproduction biotechnology in the forestry and horticulture industries. Plant somatic embryogenesis also has broad application prospects in artificial seeds, haploid breeding, asexual reproduction, and germplasm conservation [160]. Plant viral diseases are serious agricultural diseases, significantly affecting the yield/quality of crops and leading to crop failure. Stem tip virus-free technology is the only effective biotechnology to be found thus far that can remove viruses from plants. It has been widely used in agricultural production to obtain virus-free seedlings, and has also been applied in potatoes, fruit trees, flowers, and other crops. Stem cells and their daughter cells of SAM from Arabidopsis thaliana can inhibit infection with the cucumber mosaic virus (CMV). The mechanism study found that viruses cause local WUS protein induction and accumulation in stem cells, as well as subsequent migration to surrounding compartments. By directly inhibiting protein synthesis in cells, the replication and transmission of viruses can be restricted, which can protect stem cells and their differentiated daughter cells from viral infections [161]. The WUS protein has anti-viral characteristics in plant stem cells, and can help plants resist viral invasion. With growth in the global population and meat demand, the harmful effects of animal husbandry on the environment and climate will increase [162]. Moreover, animal-borne diseases and antibiotic resistance are harmful to humans [163]. A suggested method to reduce the consumption of animal meat is to increase the production of artificial meat through species iPSCs, which can also eliminate many environmental and ethical issues which occur with traditional meat production [164]. In 2013, Dutch biologist Mark Post produced the first piece of artificial meat in history by using the animal cell tissue culture method, which attracted widespread attention [165]. Animal cell culture artificial meat is mainly composed of skeletal muscle containing different cells. These skeletal muscle fibers are formed by the proliferation, differentiation, and fusion of embryonic stem cells or muscle satellite cells. They first isolated the growth-differentiable primitive stem cells. By adding a culture medium rich in amino acids, lipids, and vitamins, they accelerated cell proliferation and differentiation and obtained a large number of bovine muscle tissue cells [166]. The production of cultured meat requires robust cell sources and types. In order to achieve the scale required for the commercial production and sales of cultured meat products, it is necessary to further develop immortal special cell lines. In addition to technical challenges, the relationship between cultured meat and social/cultural phenomena and social systems must also be considered [167]. In the racing industry, tendon and ligament injuries are common problems that can end the careers of racehorses. Therefore, stem cell therapy has received attention in this field. Common clinical applications include the use of stem cells to treat tendon and ligament strains in the joints of horses [168]. Stem cell technology also has applications in the medical beauty industry. Some plants contain raw materials needed in cosmetics, and stem cell culture can overcome barriers such as low endogenous content and difficult extraction methods [169]. For example, plant cell culture technology can be used to derive certain mint-based hair care products [170,171]. Plants containing antioxidant substances, such as grapes and cloves, can be used in anti-ultraviolet light protection skincare products. Plant stem cells can be used to obtain these antioxidant components at a more efficient rate [172]. Although plant stem cells are widely used in the medical beauty field, their full potential remains to be explored due to the lack of scientific evidence and the large variety of flora that may have potential for stem cell culture. In addition, Taxus chinensis and Catharanthus roseus suspension cell cultures can also be used to produce taxol- and vinblastine-based anticancer substances [173,174]. Although promising advances have been made in the field of plant stem cells and their various applications, it is unclear whether plant-derived extracts and stem cell extracts have race-specific effects in humans. Regenerative medicine is a new research area in the field of medicine. It uses biological and engineering methods to create lost or damaged tissues and organs so that they mimic the structure and function of normal tissues and organs [175]. At present, stem cell therapy is a widely used type of regenerative medicine therapy, and plays an important role in the treatment of chronic diseases, including autoimmune diseases, leukemia, heart disease, and urinary system problems [28,176]. Autoimmune Addison’s disease (AAD) is an inevitably fatal disease in the absence of treatment. Affected patients must receive steroid replacement for life to survive. Studies have found that AAD can be improved by manipulating endogenous adrenal cortical stem cells to enhance adrenal steroid production [177]. Hematopoietic stem cell transplantation can be used to treat leukemia, and around 80–90% of leukemia patients show improvement after hematopoietic stem cell transplantation, of which 60–70% enter remission [178]. The cardiac regenerative medicine field is currently facing challenges due to the lack of cardiac stem cells in adults, low turnover rate of mature myocardial cells, and difficulty in providing treatment for injured hearts. At present, cell reprogramming technology has been applied to generate patient-specific myocardial cells through both direct and indirect methods [179]. Stem cell therapy can also be used to treat stress-induced urinary incontinence, and preclinical studies have made advances in regenerating the urethral sphincter by using secretory group cells or chemokines that can return repair cells to the injured site [180]. In addition, regenerative medicine is closely related to tissue engineering. At present, organ transplantation is still widely used to replace failed tissues and organs. However, with substantial increases in the demand for organ transplantation in recent decades, it is difficult to maintain an adequate supply of available organs [181]. The emergence of 3D biological printing technology has made up for the lack of supply of tissues and organs. Compared with traditional tissue engineering methods, 3D bioprinting utilizes a more automatic process and can create more advanced scaffolds with accurate anatomical characteristics, allowing the precise co-deposition of cells and biomaterials [182]. 3D biological printing technology is also used in cancer research, drug development, and even clinician/patient education [183]. However, there are still some issues with 3D biological printing technology, such as limitations with biological inks and printers, as well as the size of the end product. At present, bioprinted tissues are often small and composed of only a few cell types, resulting in limited function and scalability [184,185]. In addition, the cost of 3D biological printing is high, and the resolution requires further improvement. Although stem cell therapy has good outcomes, it also has safety risks. For example, pluripotent stem cells have the ability to form teratomas themselves [186]. The IPS cells established using retroviral vectors are used to introduce exogenous genes, and their expression may be retained or reactivated during differentiation. This may have impacts on the directivity and carcinogenicity of differentiation [187]. To fully realize the benefits of regenerative medicine, the real and imaginary boundaries of social, ethical, political, and religious views must be addressed [188,189]. We must carefully measure the potential therapeutic benefits of the clinical application of stem cells and weigh them according to the possible side effects in each patient and disease indication, because the clinical use of stem cells can lead to overly high expectations. Our decision-making process regarding disease management should continue to firmly follow the conservative principles of evidence-based medicine. At present, it is not uncommon to utilize stem cells in both medicine and agriculture, such as for the effective repair of damaged tissues and organs and to treat cardiovascular and metabolic ailments, as well as diseases of the nervous system, blood system, and others [161,190,191]. Recently, some research has revealed the “switch” mechanism underlying the brain regeneration of salamanders, and constructed a space–time map of brain development and regeneration of single salamander cells [192]. The next step in this field is to achieve brain regeneration in mammals, including humans, which would involve the activation of brain “seed cells” and the introduction of key factors, thus turning on the “switch” of human brain regeneration. It is expected that new treatment methods will soon be developed to improve the clinical rehabilitation of patients with brain diseases. In addition, the potential value of stem cells in anti-viral applications is of great interest. Plant stem cells can resist viruses, and animal stem cells can also use antiviral Dicer (AviD) to resist the invasion of multiple RNA viruses [193]. The antiviral mechanism of stem cells may be of great value for future medical and pharmaceutical research on human viral infection resistance. Plant regeneration is mainly carried out through somatic embryogenesis or organogenesis; however, plant regeneration can be promoted by transferring plant-related genes [194]. With the rapid development of synthetic biology, this concept has been applied to the regeneration of animals and plants [195]. The concept of “build-to-understand” synthetic biology is instructive in the field of tissue regeneration, where more extensive and flexible research can be achieved by building genetic circuits. Using synthetic biology, we can import genes with strong regenerative abilities into rare and precious plants to increase their yield. CRISPR-Cas9 technology enables genome-wide epigenetic modifications to modify plant regeneration pathways or affect specific gene loci to regulate plant regeneration [196,197,198]. In this review, we have made a more detailed and systematic summary of the research of animal and plant stem cells in the field of regeneration in recent years, and described the regeneration mechanisms of animals and plants. In addition, we also proposed the application prospects of stem cells in agriculture, animal husbandry, and regenerative medicine, which would provide new ideas and directions for the protection of endangered species and the development of regenerative medicine. However, due to the lack of existing genetic information on higher animals and plants, current research is mainly focused on simpler species, such as Arabidopsis, planarians, etc. There is still a long way to go before applications in advanced endangered plants and regenerative medicine can be fully realized. However, with rapid developments in synthetic biology, single cell sequencing, and other technologies, research on higher animals and plants is becoming more feasible. It is believed that with more research, the mystery of regeneration will eventually be solved.

PMC10002314

Evida Poopedi,Tanusha Singh,Annancietar Gomba

Potential Exposure to Respiratory and Enteric Bacterial Pathogens among Wastewater Treatment Plant Workers, South Africa

28-02-2023

domestic wastewater,sanitation workers,biological hazards,occupational exposure,waterborne diseases

Wastewater handling has been associated with an increased risk of developing adverse health effects, including respiratory and gastrointestinal illnesses. However, there is a paucity of information in the literature, and occupational health risks are not well quantified. Grab influent samples were analysed using Illumina Miseq 16S amplicon sequencing to assess potential worker exposure to bacterial pathogens occurring in five municipal wastewater treatment plants (WWTPs). The most predominant phyla were Bacteroidota, Campilobacterota, Proteobacteria, Firmicutes, and Desulfobacterota, accounting for 85.4% of the total bacterial community. Taxonomic analysis showed a relatively low diversity of bacterial composition of the predominant genera across all WWTPs, indicating a high degree of bacterial community stability in the influent source. Pathogenic bacterial genera of human health concern included Mycobacterium, Coxiella, Escherichia/Shigella, Arcobacter, Acinetobacter, Streptococcus, Treponema, and Aeromonas. Furthermore, WHO-listed inherently resistant opportunistic bacterial genera were identified. These results suggest that WWTP workers may be occupationally exposed to several bacterial genera classified as hazardous biological agents for humans. Therefore, there is a need for comprehensive risk assessments to ascertain the actual risks and health outcomes among WWTP workers and inform effective intervention strategies to reduce worker exposure.

Potential Exposure to Respiratory and Enteric Bacterial Pathogens among Wastewater Treatment Plant Workers, South Africa Wastewater handling has been associated with an increased risk of developing adverse health effects, including respiratory and gastrointestinal illnesses. However, there is a paucity of information in the literature, and occupational health risks are not well quantified. Grab influent samples were analysed using Illumina Miseq 16S amplicon sequencing to assess potential worker exposure to bacterial pathogens occurring in five municipal wastewater treatment plants (WWTPs). The most predominant phyla were Bacteroidota, Campilobacterota, Proteobacteria, Firmicutes, and Desulfobacterota, accounting for 85.4% of the total bacterial community. Taxonomic analysis showed a relatively low diversity of bacterial composition of the predominant genera across all WWTPs, indicating a high degree of bacterial community stability in the influent source. Pathogenic bacterial genera of human health concern included Mycobacterium, Coxiella, Escherichia/Shigella, Arcobacter, Acinetobacter, Streptococcus, Treponema, and Aeromonas. Furthermore, WHO-listed inherently resistant opportunistic bacterial genera were identified. These results suggest that WWTP workers may be occupationally exposed to several bacterial genera classified as hazardous biological agents for humans. Therefore, there is a need for comprehensive risk assessments to ascertain the actual risks and health outcomes among WWTP workers and inform effective intervention strategies to reduce worker exposure. Municipal wastewater naturally contains hazardous biological agents, including bacteria, fungi, and viruses [1]. A significant amount of municipal wastewater is generated by households and contains human excreta such as faeces and urine, including that from disease-carrying individuals. Waterborne infections and diseases caused by ingestion, inhalation, or dermal contact with hazardous biological agents remain a global public health concern [1]. Daily work activities performed by WWTP workers such as manual screening and desludging, plant maintenance, and emergency repairs increase the risk of exposure to preventable waterborne diseases [2]. Although the World Health Organization (WHO) recognises the occupational health risks of WWTP workers [2], specific guidelines to protect wastewater workers from workplace hazards remain elusive, particularly in low to middle-income countries (LMICs). Consequentially, workers at wastewater treatment plants (WWTPs) are at perpetual risk of exposure to waterborne pathogens through different exposure routes [3]. Higher prevalence of gastrointestinal [4,5,6] and respiratory [6,7,8,9] symptoms have been reported in WWTP workers compared to controls. For instance, in a study to investigate the impact of inhalable particles and gas exposure on the respiratory system of WWTP workers from urban and rural sewage plants and the sewer net system in Norway, Heldal and co-workers (2019) [9] observed a lower lung function and a higher prevalence of airway symptoms (33 and 11%, respectively) among WWTP workers compared to the control group [8]. Furthermore, Van Hooste et al. (2010) [4] reported a higher prevalence (37%) of gastrointestinal symptoms such as stomach ache, abdominal pain, indigestion, heartburn, and burping compared to the control group (14.7%) among WWTP workers in Belgium [4]. However, the precise cause of the reported symptoms and illnesses is not well studied. Among the most commonly detected microbial contaminants in municipal wastewater with potential to cause health issues in exposed populations, bacteria and their components are a major concern, particularly for immunocompromised individuals [10]. While a lot of studies on wastewater bacteriology employ traditional culture tests and/or targeted molecular assays to provide absence/presence and quantified measurements of a few selected bacteria, the advent of advanced high throughput DNA sequencing technologies dramatically increases the scope to comprehensively identify even the fastidious microorganisms. Therefore, advanced sequencing technologies provide a more accurate representation of the microbial population in a given sample with high taxonomic resolution [11]. In recent years, researchers have employed high throughput molecular approaches such as targeted amplicon sequencing and shotgun metagenomics to profile bacterial communities in wastewater and sludge. These studies have revealed the presence of several pathogenic bacteria at community level, including Streptococcus pneumoniae, Vibrio cholerae, Salmonella spp., Mycobacterium tuberculosis, Legionella pneumophila, Pseudomonas aeruginosa, Arcobacter spp., and Acinetobacter spp., among others [10,12,13,14,15]. Considering that WWTP workers are potentially exposed to a plethora of harmful bacterial genera at any given time, knowledge of occupational exposure risks at microbial community level is critical. Moreover, microorganisms have evolved together in communities, and certain health effects may be due to synergistic or antagonistic behaviors of specific pathogens [16]. Additionally, understanding microbial composition at community level enables microbial source tracking and comprehensive risk assessment to minimise exposure to harmful biological agents. Microbial communities in municipal wastewater and the extent of worker exposure may differ depending on geographic location, season, time of the day, population served, facility treatment capacity, the technology used, and performed activities [17,18,19]. Yet there is a general lack of research in LMICs to characterise bacterial communities in untreated wastewater using high throughput sequencing technologies [15,20,21], with little to no data documented in most regions [22,23]. In South Africa, for example, previous studies in wastewater environments have mainly focused on pathogen removal and effluent quality, essential to protect public health and the environment. Consequently, the likelihood of occupational exposure to untreated or partially treated wastewater at WWTPs and the associated health risks in LMICs remain unclear. Hence, in this study we used targeted 16 rRNA gene amplicon sequencing to describe bacterial diversity and identify bacterial pathogens that could pose an occupational health risk from repeated exposure to untreated municipal wastewater. The study was an experimental design conducted at the National Institute for Occupational Health, Immunology and Microbiology Laboratory, South Africa. Grab influent samples were obtained from five participating WWTPs on 20 October (WWTP2, 3 and 4) and 21 October (WWTP1 and 5), 2020. Sampling was conducted at five municipal WWTPs in the City of Tshwane Metropolitan Municipality, South Africa. The characteristics of the sampling sites are summarised in Table 1. The selected WWTPs all receive wastewater from healthcare facilities; however, the quantities are not measured. Non-sewered informal settlements were observed nearby WWTP1, 2, and 4 but no intensive animal (cattle or battery chicken) production or roaming domestic animals were evident near any of the sampling sites. Grab influent samples were collected in 1 litre sterile polypropylene bottles (Sigma-Aldrich, Saint Louis, MO, USA). Samples were transported on ice and stored at 2–8 °C for not more than 24 h until analysis. Sample temperature and pH were measured during sampling using a digital water-resistant thermometer 125 mm (−50 to +200 °C) (Lasec, Cape Town, South Africa). Metadata such as environmental temperature, relative humidity, and CO2 were measured using IAQ-CALC Indoor Air Quality Meter, Model 7545 (TSI Instruments Ltd., High Wycombe, UK) Microbial cells were concentrated following the protocol by Kumar et al. (2020) [24]. Briefly, each 1 L grab sample was manually shaken by hand to allow adequate mixing and an aliquot of 200 mL in four batches of 50 mL was centrifuged at 5000 rpm for 45 min at 4 °C. The supernatant was carefully removed and each pellet was washed in 1 mL of sterile distilled water to remove deposited salts and other impurities. The four pellets were pooled and stored at −20 °C until further analysis. Total genomic DNA (gDNA) was extracted from the pelleted biomass of influent samples using DNeasy Powersoil Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The DNA was quantified using a Nanodrop 2000c spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The gDNA was amplified with primers targeting bacterial V3 and V4 regions, namely, 16S Amplicon PCR forward primer (5′-TCG TCG GCA GCG TCA GAT GTG TAT AAG AGA CAG CCT ACG GGN GGC WGC AG) and 16S Amplicon PCR reverse primer (5′-GTC TCG TGG GCT CGG AGA TGT GTA TAA GAG ACA GGA CTA CHV GGG TAT CTA ATC C) [25]. The PCR reaction contained 12.5 µL of 2x KAPA HiFi HotStart ready mix, 5 ng/µL gDNA template, and 5 µL of 1 µM each primer. Cycling conditions included initial denaturation at 95 °C for 3 min, 25 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 sec, extension at 72 °C for 30 sec, and final extension at 72 °C for 5 min. The 16S amplicons were purified using AMPure XP beads (Beckman Coulter, Brea, CA, USA), and paired-end libraries were attached using the Nextera XT Index Kit (Illumina, San Diego, CA, USA) and Illumina sequencing adapters (Illumina, USA). Indexed amplicons were cleansed with AMPure beads and normalised with PhiX Control v3 (Illumina, USA). Purified PCR amplicons were then sequenced using the Miseq platform (Illumina, USA). Raw reads were quality controlled and filtered (Q > 20 and length > 50 bp) using fastqc (v0.11.8) and trimGalore (v0.6.4_dev; https://github.com/FelixKrueger/TrimGalore, accessed on 13 July 2022), respectively. TrimGalore was also used to remove adapters. Krona charts for interactive data visualisation were generated using Kraken2 [26] and Krona [27]. All downstream analyses, including classification, abundance estimations, statistical analysis, and visualisation, were carried out in R (v3.6.1 Dada2 package (v1.12.1) [28] which was used to pre-process clean reads, including quality inspection, trimming, de-replication, merging paired-end reads, and removal of chimeric sequences. The obtained amplicon sequence variants (ASVs) were taxonomically classified, and ASV abundance estimates were determined using training sequence sets based on the SILVA reference database (v138; https://zenodo.org/record/1172783#.XvCmtkUzY2w, accessed on 14 July 2022). Ordinations for beta diversity, abundance bar plots, alpha diversity, and richness estimates were generated using the phyloseq (v1.28.0) [29], ggplot2 (v3.2.1), and AmpVis2 (v2.6.4) [30] packages. To compare alpha diversity between groups, Wilcoxon and/or Kruskal–Wallis rank-sum tests were used. The UpSet plots were generated using UpsetR (v1.4.0) [31]. DESeq2 (v1.24.0) was used to perform differential abundance analysis between sample groups [32]. A total of 397,766 quality raw reads were generated from all the samples collected (Table 2). The alpha diversity indices revealed comparably similar community diversity across the sites, except for WWTP4, which showed the highest diversity and species richness. The values for species richness (Chao1 and Abundance-based coverage estimators) ranged between 534–957 and for evenness (Shannon) ranged between 5.85–6.32. These results were confirmed by the rarefaction curve estimates, with WWTP4 presenting considerably higher values than the other sites, followed by relatively similar low values between WWTP1 and 2, and WWTP 3 and 5. The rarefaction curves for the observed operational taxonomic units (OTUs) for all samples plateaued at around 5000 reads (Figure 1), indicating adequate sequencing depth; hence, a good representation of the bacterial community was achieved as most of the abundant species and some rare species are represented. A total of 24 phyla were identified, with fourteen (Acidobacteriota, Actinobacteriota, Bacteroidota, Campilobacterota, Cyanobacteria, Desulfobacterota, Elusimicrobiota, Firmicutes, Fusobacteriota, Patescibacteria, Proteobacteria, Spirochaetota, Synergistota, Verrucomicrobiota) appearing across the five WWTPs (Figure 2). Of these, the five most predominant phyla were Bacteroidota (31.7%), Campilobacterota (18.4%), Proteobacteria (20.2%), Firmicutes (8.6%), and Desulfobacterota (6.5%). A total of 122 families were identified across all samples. Families of the top three dominant phyla contributing at least 1% to total abundance in the five WWTPs influent samples are shown in Figure 3. For the phylum Bacteroidota, the predominant families were Bacteroidaceae, Paludibacteraceae, Prolixibacteraceae, Tannerellaceae, and Williamwhitmaniaceae. The phylum Proteobacteria was dominated by Pseudomonadaceae, and Rhodocyclaceae and Campilobacterota were dominated by Arcobacteraceae and Sulfurospirillaceae. At the genus level, 253 genera were identified from all samples accounting for the majority of the classified sequences, with approximately 25% of the total communities remaining unidentified. Genera that were detected at all sites with a relative abundance of at least 1% are presented in Figure 4. Sulfurospirillum, Macellibacteroides, and Bacteroides were the three most abundant genera identified from the influent samples, accounting for 12.8, 5.2, and 5.1% of total genera, respectively. Shared bacterial genera (those that appeared in two or more WWTPs) and distinct genera (those that occurred at one WWTP) are presented in Figure 5. Of the 253 genera identified, 40 were shared by all five sites, 24 shared by four sites, 32 by three sites, and 29 by two sites. WWTP4 had the highest bacterial richness while WWTP3 showed the least richness in the number of genera detected. We also observed genera that were unique to specific sites with 62 distinct genera observed at WWTP4, 29 at WWTP1, 15 at WWTP2, and 14 at WWTP5. The least diverse bacterial genera, with only eight distinct genera, were observed in WWTP3. A total of 36 genera (approximately 20% of total abundance) of medical importance to human health were identified (Table 3) and belonged to the major phyla Bacteroidota, Campilobacterota, Proteobacteria, Firmicutes, Synergistota, Fusobacteriota, and Actinobacteriota. The dominant pathogenic genera were Bacteroides (5.1%), Pseudomonas (2.9%), Aeromonas (2.8%), Arcobacter (2.6%), Leptotrichia (1.4%), Treponema (0.9%), Streptococcus (0.6%), Enterobacter (0.5%), Shewanella (0.5%), and Acinetobacter (0.4%). Table 3 shows the relative abundance of pathogenic genera from the bacterial community as represented by influent samples of the five WWTPs. Potentially pathogenic bacterial genera were classified into three main groups according to the type of infection they cause in humans: (1) respiratory pathogens, (2) enteric pathogens, and (3) opportunistic pathogens commonly associated with nosocomial infections and multidrug resistance. Respiratory:Mycobacterium and Coxiella were the most abundant respiratory tract-associated pathogens contributing up to 0.1% and 0.05% of the total bacterial community of influents, respectively. Mycobacterium was detected only at WWTP1 and WWTP2, with a relative abundance of 0.2% at each plant, while Coxiella was only detected at WWTP4, accounting for approximately 0.2% of the total bacterial community at this site. Enteric: Enteric pathogens detected were Escherichia/Shigella, Laribacter, Arcobacter, and Aeromonas with total relative abundances of 0.1%, 0.2%, 2.6%, and 2.8%, respectively. Aeromonas was the most prevalent enteric pathogen, with a relative abundance ranging between 0.7% and 5.3%, followed by Arcobacter with the highest abundance (9.7%) at WWTP5 and the lowest at WWTP3 (0.9%). It is also important to mention that Aeromonas and Arcobacter were detected at all sites. Escherichia/Shigella were rare genera detected only at WWTP1 and WWTP2, with abundances of 0.3% and 0.1% at the two sites, respectively. Opportunistic: Thirty opportunistic bacterial genera were identified in the present study, with an overall abundance of 14.4% in the total bacterial community (Table 3). The top three opportunistic pathogens were Bacteroides, Pseudomonas, and Leptotrichia contributing (5.1%, 2.9%, and 1.4%, respectively, of the total bacterial community. The other opportunistic genera were ˂1% in relative abundance. A total of 13 opportunistic genera (Acinetobacter, Aeromonas, Arcobacter, Bacteroides, Comamonas, Dysgonomonas, Enterobacter, Leptotrichia, Pseudomonas, Pseudoxanthomonas, Shewanella, Streptococcus, and Treponema) were detected in at least four of the five WWTPs. The observed pathogenic genera were classified into different risk groups according to the revised South African Regulation for Hazardous Biological Agents, 2022 [33] (Table 3). Thirteen of the 36 genera (36%) belong to HBA Risk Group 2 (may cause disease, are unlikely to spread to the community, and effective treatment is available), and three genera belong to HBA Group 2 or 3 (HBA Risk Group 3, may cause severe disease, present a risk of spreading to the community but effective treatment is available) depending on the species type (Table 3). Coxiella belongs to HBA Risk Group 3 and has only one member, C. burnetii. In summary, close to 50% (17/36) of pathogenic genera identified were classified as hazardous biological agents, indicating that these organisms can cause human diseases and may pose a health risk to WWTP workers. Municipal WWTPs receive wastewater from different sources but mostly households, surface runoff, and industrial activities, contributing to the complexity of bacterial communities in wastewater [34]. The phyla Bacteroidota, Campilobacterota, Proteobacteria, Firmicutes, and Desulfobacterota were predominant in the present study. Except for Campilobacterota, the top dominant phyla in the current study have previously been reported in high abundance in influent wastewater [35,36]. Wu and colleagues (2019) [37] recently provided a comprehensive wastewater analysis on a global scale, representing 23 countries from six continents including Africa, Asia, Australia, Europe, North America, and South America. Their findings revealed that despite the considerable diversity in bacterial communities between samples, a core global community (28 OTUs) exists. A majority of these members belonged to the phyla Proteobacteria and Bacteroidota, implying some degree of bacterial community conservation in municipal wastewater at higher taxa rank [37]. In addition, the phyla Bacteroidota, Proteobacteria, Firmicutes, and Actinobacteria have consistently been reported as the predominant bacterial community members in the human microbiome, suggesting that a large proportion of bacterial members in the samples analysed originated from human faecal material [38]. When comparing bacterial communities in other wastewater types, in particular, activated sludge, previous studies highlighted a distinct microbial ecosystem with high bacterial diversity and a high concentration of biomass in activated sludge samples [39,40]. Begmatov and co-workers [40] recently reported a dominance of Proteobacteria, Chlorofexi, Myxococcota, Firmicutes, Patescibacteria, and Nitrospirota in activated sludge [40]. A comparison between the activated sludge community identified in Moscow with that identified globally by the Global Water Microbiome Consortium show clustering, emphasising that influent characteristics, which are largely influenced by cultural, social, and environmental factors in each region, are more important than WWTP operating conditions [40]. It is noteworthy that at the genus level in the current study, bacterial composition exhibited some degree of diversity across the five WWTPs. The highest bacterial richness was recorded at WWTP4, whereas WWTP3 showed the least richness in the number of genera detected. The lowest bacterial richness was found at WWTP3, which could be explained by the fact that this plant also treated industrial wastewater, whereas the other four WWTPs received municipal wastewater primarily from households. Chemical substances in industrial wastewater have been shown to negatively impact microbial community structures, resulting in reduced bacterial richness and diversity compared to municipal wastewater [41,42]. Interestingly, WWTP4 contained many distinct genera compared to the other plants, which may be attributed to it being the largest treatment plant in the area, serving approximately seven different communities. Overall, these findings suggest that while some WWTPs harboured exclusive genera, the predominant genera did not vary considerably regardless of plant location, indicating a high degree of bacterial community stability in the influent source. This study grouped bacterial genera with known disease-causing species into three major infection categories: respiratory, enteric, and opportunistic pathogens. Coxiella and Mycobacterium were the only medically important genera identified in the present study with the potential of causing respiratory tract infections. Although the total relative abundance of these genera was much lower than that of the majority of the identified genera, C. burnetii and M. tuberculosis have extremely low infectious doses, requiring less than 10 living organisms to cause an infection [43,44]. Respiratory pathogens cause infections of the upper and lower respiratory tract. By far, the most serious respiratory infections involve the lower respiratory tract such as bronchitis and pneumonia, and are the leading cause of high mortality rates worldwide [45]. The abundance of Mycobacterium in influent samples was comparable to previous studies [10,12,46]. For instance, studies conducted in Germany and Australia found that the overall abundance Mycobacterium in influent and effluent was less than 0.02% [12,46]. In comparison to other types of wastewater, studies on Mycobacterium in influent are not common. However, Mycobacterium has primarily been studied in effluents to assess the efficacy of wastewater treatment processes in removing biological agents and the safety of treated effluents [21,36,47]. Humans become infected with Coxiella by inhaling aerosols from contaminated animal waste, soil, or food products, and veterinarians, slaughterhouse, and farm workers are generally considered to be at increased risk of occupational exposure to Coxiella [43]. Given that WWTPs receive wastewater from various institutions and farms, it raises the question of whether WWTP workers are at risk of occupational exposure to this pathogen. Five enteric genera, namely, Escherichia/Shigella, Laribacter, Arcobacter, and Aeromonas were identified in the study, accounting for 5.7% relative abundance of the total bacterial community. Enteric pathogens normally reside in the intestines of humans and can utilise their pathogenic mechanisms to cause gastrointestinal tract infections [48]. Enteric organisms are typically transmitted via the faecal–oral route, and illness symptoms are caused by consuming contaminated food or water [48]. Using metagenomics analysis, previous studies have reported the prevalence of Aeromonas to be noticeably high in wastewater with counts similar to those of faecal coliforms [47,49,50]. Ye and co-workers [47] analysed potentially pathogenic bacteria in sixteen samples comprising of influent, activated sludge, and effluent from 14 municipal WWTPs across, China, Canada, United States, and Singapore, and reported that Aeromonas were among the most dominant genera occurring at least in ten samples across four countries [47]. Other metagenomics studies [17,51,52,53] have also reported a high abundance of Aeromonas in various wastewater types. Aeromonas genus has been ranked third as the leading cause of diarrhoea after Camplylobacter and Salmonella [54], and incidences of gastroenteritis linked to Aeromonas have been reported across the globe, with cases more common in developing countries [54]. Two Arcobacter species, A. butzleri and A. cryaerophilus, are considered emerging pathogens threatening human health [55]. Human-associated Arcobacter species have been consistently isolated from human sewage systems [35,55]. In the present study, the genus Arcobacter was observed in high abundance exclusively at WWTP5 (9.7%), compared to the other WWTPs that had an abundance ranging from 0.9% to 2.5%. This inconsistency could not be explained further. In the present study, the genus Laribacter was 0.2% relative abundance, which is comparable with that reported in municipal influent (≤0.1% relative abundance) from a study conducted at three WWTPs in Western Australia [56]. The genus Laribacter has one species, L. hongkongensis, which is associated with traveller gastroenteritis and diarrhoea [57]. It should be noted that, even though Laribacter has not been previously reported among the most prevalent genus in wastewater, if ingested, this bacterium can cause gastroenteritis in humans [56]. The present study identified 32 genera that could cause opportunistic infections in humans. Although opportunistic pathogens pose little risk to healthy WWTP workers, they can cause serious illnesses in individuals with weakened immune systems and the elderly. The main concern with opportunistic organisms is that they are typically resistant to commonly used antimicrobial treatments, posing a serious problem for public health [58]. Five pathogenic genera (Enterococcus, Klebsiella, Acinetobacter, Pseudomonas, and Enterobacter) that could contain species belonging to the group of ESKAPE pathogens were identified in the present study with Pseudomonas topping the list. Members of the ESKAPE pathogens are well known for their ability to develop multidrug resistance and account for most nosocomial infections [59]. Moreover, Acinetobacter baumannii, Pseudomonas aeruginosa, and Streptococcus pneumonia are on the WHO priority list of antibiotic-resistant pathogens as they can cause fatal infections [60]. Resistance to commonly used antibiotics is a serious global problem. Current efforts to monitor the development of antimicrobial resistance have mainly focused on clinical settings; however, interest has grown in recognising the importance of antibiotic resistance in the environment and water supply [57]. In fact, WWTPs have been implicated as key reservoirs, incubators, and source for disseminating antimicrobial resistance and virulence genes [61]. In this study, 17 genera were classified as potentially hazardous biological agents to human health. Overall, the risk characterisation exercise revealed that workers at WWTP may be exposed to genera that cause airway obstruction, gastrointestinal problems, and opportunistic infections in the workplace. Except for Mycobacterium and a few Gram-positives such as Streptococcus, Enterococcus, Leuconostoc, Erysipelothrix, Atopobium, Brachybacterium, Finegoldia, Gordonia, and Actinomyces, a majority (28/36 genera) of the classified pathogenic genera identified at WWTPs were Gram-negative. Gram-negative bacteria express endotoxins as their main component of the outer membrane [62], and the presence of numerous and diverse Gram-negative bacteria at WWTPs may be a major contributor to workers’ exposure to elevated levels of endotoxins. Long-term exposure to inhalable endotoxins has been linked to inflammatory responses in the lungs, leading to symptoms such as chronic bronchitis, organic toxic dust syndrome, or asthma [63]. Therefore, the findings of this study suggest that workers at the selected five WWTPs may be exposed to pathogens, including endotoxins from Gram-negative bacteria, which could compromise their respiratory health. Waterborne diseases are expected to rise with climate change and a growing global population [64]. Consequently, WWTP workers remain at risk from waterborne diseases and outbreaks irrespective of the country’s economic status (i.e., developed or developing) or whether in the tropics or temperate, highlighting the importance of regular monitoring of wastewater microbiomes using advanced but cost-effective detection techniques, precise disinfectant procedures, and proper management of operations to minimise worker exposure. The morbidity and mortality of waterborne diseases are enormous [64], and they can only be controlled by providing workers with a microbiologically safe environment. WHO has established a list of priority waterborne pathogens with moderate to high health significance [65]. In the present study, Mycobacterium and Escherichia/Shigella genera were detected in wastewater and these organisms are listed on the WHO priority pathogens. Additionally, emerging waterborne pathogens of importance Aeromonas and Leptotrichia were detected in the study. Therefore, the findings of this study warrant further investigation into incidences of infections in WWTP workers associated with exposure to these pathogens. Workers at WWTP play an important role in urban communities, making sure that the treatment plants function optimally. However, the working conditions for sanitation workers, including WWTP workers, expose them to significant health risks such as waterborne diseases, injuries, and even death [22]. Data presented in this study are from a single type of wastewater (influent), and samples were collected once over one month from five municipal treatment plants in South Africa. Hence, the small sample size is a limitation. Furthermore, sample collection during and outside the respiratory infection season was impractical as some COVID-19 symptoms were similar to respiratory influenza and occurred throughout the year; hence, defining the respiratory infection season was challenging and thus omitted. Therefore, more research is needed to monitor trends and changes in the diversity of bacterial pathogens in various types of wastewater on a large scale over a long period to assess temporal and spatial variation. Although the current study provides local baseline data on potential pathogens circulating in WWTP environments, further research to provide impact of exposure on workers’ health is needed. Therefore, a follow-up study assessing the associations between work activities and the incidence of gastrointestinal and respiratory infections among WWTPs in South Africa is underway. This information is currently non-existent in South Africa, making it difficult to reform occupational health and safety policies and practices. The study found evidence of several potentially pathogenic bacteria in untreated municipal wastewater, which may pose a health risk to WWTP workers. The major phyla identified in the study were Bacteroidota, Proteobacteria, Campilobacterota, Firmicutes, and Desulfobacterota and the dominant pathogenic genera were Bacteroides, Pseudomonas, and Aeromonas. Risk characterisation of the identified pathogenic genera revealed that the assigned genera are capable of causing gastrointestinal illnesses, airway obstruction, and some are intrinsically resistant to commonly used antibiotics. However, further studies are imperative to establish an association between the identified pathogenic bacterial pathogens and the commonly reported symptoms among WWTP workers. Despite the preliminary nature of the results, intervention strategies should focus on raising awareness of bacterial contaminants present at WWTP and improving personal protective equipment (PPE) compliance in workers to mitigate health risks. WWTP workers are vulnerable to increased occupational and environmental health hazards, barriers to healthcare, access to personal protective equipment, legal protection, and other safeguards. Molecular profiling of bacterial communities generates scientific-based evidence on the abundance of potential pathogenic bacteria in untreated wastewater that WWTP workers may be exposed to. With a lack of such information for workers in LMICs, comparison with high income countries is impracticable. Therefore, the findings of this study contribute to the body of knowledge, the relevance of region-specific data for a low-income countries given that bacterial communities in wastewater differs between geographical locations and are influenced by factors such as cultural, social, and environmental conditions. This may lead to the identification of region-specific diseases thus tailoring interventions that are specific and targeted to local settings.

PMC10002321

José Vicente Gil,Esperanza Such,Claudia Sargas,Javier Simarro,Alberto Miralles,Gema Pérez,Inmaculada de Juan,Sarai Palanca,Gayane Avetisyan,Marta Santiago,Carolina Fuentes,José María Fernández,Ana Isabel Vicente,Samuel Romero,Marta Llop,Eva Barragán

Design and Validation of a Custom Next-Generation Sequencing Panel in Pediatric Acute Lymphoblastic Leukemia

23-02-2023

next-generation sequencing,NGS,molecular characterization,childhood acute lymphoblastic leukemia

The molecular landscape of acute lymphoblastic leukemia (ALL) is highly heterogeneous, and genetic lesions are clinically relevant for diagnosis, risk stratification, and treatment guidance. Next-generation sequencing (NGS) has become an essential tool for clinical laboratories, where disease-targeted panels are able to capture the most relevant alterations in a cost-effective and fast way. However, comprehensive ALL panels assessing all relevant alterations are scarce. Here, we design and validate an NGS panel including single-nucleotide variants (SNVs), insertion–deletions (indels), copy number variations (CNVs), fusions, and gene expression (ALLseq). ALLseq sequencing metrics were acceptable for clinical use and showed 100% sensitivity and specificity for virtually all types of alterations. The limit of detection was established at a 2% variant allele frequency for SNVs and indels, and at a 0.5 copy number ratio for CNVs. Overall, ALLseq is able to provide clinically relevant information to more than 83% of pediatric patients, making it an attractive tool for the molecular characterization of ALL in clinical settings.

Design and Validation of a Custom Next-Generation Sequencing Panel in Pediatric Acute Lymphoblastic Leukemia The molecular landscape of acute lymphoblastic leukemia (ALL) is highly heterogeneous, and genetic lesions are clinically relevant for diagnosis, risk stratification, and treatment guidance. Next-generation sequencing (NGS) has become an essential tool for clinical laboratories, where disease-targeted panels are able to capture the most relevant alterations in a cost-effective and fast way. However, comprehensive ALL panels assessing all relevant alterations are scarce. Here, we design and validate an NGS panel including single-nucleotide variants (SNVs), insertion–deletions (indels), copy number variations (CNVs), fusions, and gene expression (ALLseq). ALLseq sequencing metrics were acceptable for clinical use and showed 100% sensitivity and specificity for virtually all types of alterations. The limit of detection was established at a 2% variant allele frequency for SNVs and indels, and at a 0.5 copy number ratio for CNVs. Overall, ALLseq is able to provide clinically relevant information to more than 83% of pediatric patients, making it an attractive tool for the molecular characterization of ALL in clinical settings. Acute lymphoblastic leukemia (ALL) is the most common type of pediatric cancer. The current cure rate reaches 80–90%, but it decreases with age, and the prognosis of relapsed patients is very poor [1,2]. ALL is caused by the clonal proliferation of immature lymphocytes due to the accumulation of genetic alterations, which gives rise to different biological and clinical subtypes of leukemia. The presence of genetic lesions that drive distinct subtypes of leukemia has set the bases of ALL classification, in which novel categories defined by point mutations (such as PAX5 p.P80R or IKZF1 p.N159N) and additional gene fusions (i.e., ZNF384 or MEF2D rearrangements) have been recently acknowledged [3,4]. Some of these variants, as well as copy number variation (CNV) affecting other genes, are used for risk stratification, which drives treatment intensity [5,6]. Genetic lesions can further be utilized for targeted therapy selection, against either the affected gene or the altered signaling pathway [7,8]. Therefore, as molecular variants are used as diagnostic, prognostic, and predictive biomarkers, their identification is being increasingly demanded by pediatricians and hematologists. Next-generation sequencing (NGS) allows the simultaneous assessment of a broad number of targets in several samples and has therefore become an indispensable tool for clinical laboratories. Different approaches have been developed in order to optimize resources and shorten the turnaround time, with pathology-directed panels being the preferred option in most centers [9]. However, the availability of commercial ALL NGS panels is scarce, as pan-hematological assays usually lack relevant genes or omit CNV identification. The purpose of this work was to design and validate an ALL-targeted NGS panel (ALLseq), including clinically relevant point mutations, insertion–deletions (indels), CNVs, fusions, and gene expression. Our results show that ALLseq constitutes a useful tool for patient characterization, identifying driver and secondary alterations in a single experiment, thus allowing accurate diagnosis, patient risk stratification, and (in some cases) treatment selection. The ALLseq design included the targets listed in Table 1 and Supplemental Table S1, for which point mutations, indels, CNVs, fusions, and/or gene expression can be assessed. The maximum chip efficiency was obtained when 200 ng of the library pool (4:1 DNA:RNA) was used for template preparation. Under these conditions, the mean chip load was 88.3% (35.88% polyclonal; 55.6% usable reads), which yielded a mean of 18,451,079 total reads. The mean quality metrics per sample showed a read depth of 1903×, on-target and uniformity percentages > 95%, and a median absolute pairwise difference (MAPD) mean value of 0.19. The mean read depth < 100× was found on 9/1138 amplicons (0.8%). We compared the ALLseq main sequencing metrics with those obtained with the Oncomine Childhood Research Assay (OCCRA). The OCCRA is a commercial pan-pediatric cancer panel, where some ALL targets are present. No significant differences were found for the on-target % or the percentage of amplicons showing a low mean read depth (<100×) (Figure 1A,B). However, the median uniformity % and MAPD were lower in ALLseq (uniformity: 96.46 vs. 97.77, p < 0.05; MAPD 0.16 vs. 0.25) (Figure 1C,D). Focusing on RNA results, the mean total number of RNA mapped reads per sample was 358,240. All of the analyzed samples passed the quality control test (>20,000 total mapped reads). For technical validation, 25 molecularly characterized samples were selected. Nineteen carried SNVs or indels in DNM2, EP300, FBXW7, FLT3, H3F3A, IKZF1, JAK1, JAK3, KMT2D, NOTCH1, NRAS, PHF6, PTPN11, or TP53. Eighteen samples harbored CDKN2A, CDKN2B, IKZF1, JAK2, RB1, PAX5, and/or ETV6 deletions. Six samples were used as DNA negative controls for SNVs/indels, and seven for CNVs. Seventeen RNA samples carried one of the following fusions: TCF3::ZNF384, KMT2A::MLLT10, BCR::ABL1, TCF3::PBX1, ARID1B::ZNF384, PICALM::MLLT10, TERF::JAK2, IGH::CRLF2 (does not generate a fusion transcript but overexpresses CRLF2), ETV6::RUNX1, KMT2A::AFF1, STIL::TAL1 (overexpresses TAL1), or EBF1::PDGFRB. Eight samples were used as negative RNA controls. ALLseq detected all of the expected SNVs and indels; KMT2D c.8743C>T and H3F3A c.82A>G were not detected, as these genes were not included in the ALLseq design. ALLseq identified two additional variants: DNM2 c.2080G>T (not included in OCCRA’s design) and KRAS c.34G>C (confirmed by direct sequencing and Minor Variant Finder analysis) (Supplemental Table S2). The VAF of the 34 overlapping variants showed a high correlation between ALLseq and OCCRA (R2 = 0.93), and the Bland-Altman plot showed that 93.9% of the VAF values were in agreement within the 95% confidence intervals (Figure 2). ALLseq detected at least one CNV in 18/25 (72%) samples and a total of 39 CNVs: CDKN2A—9/39 (23.1%), CDKN2B—9/39 (23.1%), ETV6—6/39 (15.4%), PAX5—6/39 (15.4%), IKZF1—4/39 (10.3%), JAK2—2/39 (5.1%), RB1—2/39 (5.1%), and EBF1—1/39 (2.6%). Eight discrepancies were observed between ALLseq and MLPA (three for IKZF1, two for PAX5, one for ETV6, one for EBF1, and one for RB1) (Supplemental Table S3). Altogether, the CNV Cohen’s kappa coefficient was 0.88. ALLseq detected fusions and/or high gene expression in 18/25 samples (72%). No false positives were observed (Supplemental Table S4). RNA 10 showed CRLF2 expression 229-fold higher than the median, consistent with the presence of a CRLF2 translocation at the IGH locus in this sample. Similarly, RNA 18, which harbored a STIL::TAL1 fusion (also identified by ALLseq), expressed TAL1 25-fold compared to the median (Supplemental Table S5). The expression values of CRLF2 and TAL1 measured by ALLseq and RT-qPCR showed a high correlation (R2 = 0.98 and 0.90, respectively) (Supplemental Figure S1). As expected, TLX1, TLX3, NKX2-1, and HOXAA expressions were undetectable in all patients, as these genes are not expressed in bone marrow or peripheral blood unless they are deregulated. LMO2 showed a stable basal expression in all samples. Overall, RNA results were in 100% agreement with the expected results. A summary of ALLseq performance can be found in Table 2 and Supplemental Table S4. The limit of detection (LoD) for SNVs and indels was established at 2% of VAF. Intra-experiment repeatability and inter-experiment reproducibility showed a 100% concordance above this VAF, and a high correlation (R2 ≥ 0.98) was observed between inter- and intra-sequencing runs (Figure 3, Supplemental Table S6). The CNV LoD was established at a copy number ratio of 0.5, corresponding with a heterozygous deletion, when, at least, confidence > 20 and precision > 10 were reached (Table 3). Different analysis rounds showed 100% repeatability and reproducibility for CNV analysis. Regarding fusion expression quantitation, serial dilutions yielded a linear range up to the 10−4 dilution (Supplemental Figure S2). In total, 43 correlative patients were prospectively analyzed with ALLseq, whose main characteristics are shown in Supplemental Table S7. DNA sequencing identified 54 SNVs or indels, resulting in a mean of 1.26 variants per sample. Twenty-five cases (58.1%) harbored at least one mutation. The genes with the highest mutational frequency were KRAS (18.52%) (10/54), NRAS (14.81%) (8/54), PTPN11 (14.81%) (8/54), and NOTCH1 (9.26%) (5/54). All of the SNVs and indels were confirmed by direct Sanger sequencing (Supplemental Table S8, Supplemental Figure S3A). Sixty-eight CNVs were found in 28 out of 43 samples (65.12%). The mean number of affected genes was 1.58 (range 0–6) per patient. The most frequently deleted gene was CDKN2A (19.12%; 13/68), followed by CDKN2B (16.18%; 11/68) and ETV6 (8.82%; 6/68). Notably, ALLseq detected additional copies of RUNX1 and TP53 in patients carrying chromosome 17 and 21 gains (including a patient harboring intrachromosomic amplification of chromosome 21, iAMP21), respectively (Figure 4, Supplemental Figure S3B). Results from 5/43 (11.63%) patients were discordant with MLPA (Supplemental Table S8). Overall, a total of 122 SNVs, indels, and CNVs were detected in the DNA. Genes affected by these alterations were grouped according to the signaling pathway in which they have a role. The most frequently altered were the TP53-cell cycle (30.33%; 37/122), followed by the RAS (18.03%; 22/122) and lymphoid differentiation (17.21%; 21/122) pathways. Interestingly, the RAS pathway was only affected by SNVs or indels and represented 41.07% (22/54) of these alterations (Supplemental Figure S4). ALLseq detected fusions in 11/43 samples (25.6%). The most frequent fusion was ETV6::RUNX1, identified in 5/43 patients (11.6%), followed by KMT2A rearrangements and STIL::TAL1, each detected in 2/43 patients (4.6%) (Supplemental Table S8). All the fusions were confirmed by orthogonal methods. Regarding gene expression, 3 out of 43 (6.9%) patients overexpressed CRLF2. One harbored the t(X;14)(p22;q32) translocation (detected by cytogenetics and FISH); another carried the CRLF2::CSF2RA fusion, also detected by ALLseq and FISH; and the third patient showed a pseudoautosomic region 1 (PAR1) amplification, which contains CRLF2. Furthermore, the two samples harboring STIL::TAL1 (detected by ALLseq and FISH) overexpressed TAL1 (Figure 5). Additionally, two patients with translocated TLX3 (confirmed by FISH) met the overexpression criteria for this gene (Supplemental Table S8). Of note, we detected ectopic TLX1 and NKX2-1 expression in two patients. These genes are not usually expressed in hematopoietic tissue unless deregulated, but TLX1 was expressed in one T-cell patient harboring the t(7;17)(q31;q12) translocation (identified by conventional cytogenetics), and NKX2-1 was expressed in a T-cortical patient for which no molecular mechanism was found. Overall, the combined DNA and RNA results showed a total of 142 alterations. Comprehensive molecular and basic clinical data are shown in Figure 6. Next, we tested the ALLseq clinical yield by exclusively classifying prospective patients according to these NGS results. Only pathogenic variants complying with at least one of these criteria were considered as clinically relevant, i.e., (a) variants that define World Health Organization (WHO 2022) and/or the International Consensus Classification of myeloid neoplasms and acute leukemias (ICC 2022) categories [3,4] (of note, ALL classification was updated during the development of this project so we used the latest classifications, although the design was based on the previous versions); (b) variants defining genetic risk groups considered by the ALLTogether treatment protocol; and (c) variants allowing patient selection for targeted therapy according to the ALLTogether protocol or active clinical trials. Under these premises, 63/142 (44.37%) of the pathogenic variants were considered as clinically relevant. Subsequently, ALL patients were allocated into three groups, depending on the clinical utility of the alteration(s) they carried: diagnosis, risk stratification, and/or targeted therapy. Moreover, 12/43 (27.91%) patients carried entity-defining alterations, 32/43 (74.42%) harbored risk-associated lesions, and 10/43 (23.26%) were suitable for targeted therapy (Figure 7A). Co-occurrence among these categories is shown in Figure 7B. Overall, 36/43 (83.72%) patients could benefit from molecular findings derived from ALLseq. Among the seven remaining patients, four harbored aneuploidies, and three did not show any additional molecular lesion. Finally, we analyzed the clinical utility of ALLseq and cytogenetics as independent or combined techniques. When used as a stand-alone method, ALLseq provided information related to prognosis or treatment to more patients than cytogenetics, while slightly more patients benefited from cytogenetics than ALLseq for the identification of ALL entities. By combining both techniques, most patients (40/43; 93.02%) could be diagnosed, classified into risk groups, and/or benefit from targeted therapies (Figure 7C). In the present study, an ALL-targeted NGS panel was designed and validated. ALLseq allows clinically relevant SNVs, indels, CNVs, fusions, and gene expression alterations to be detected, which, to the best of our knowledge, makes our panel unique. ALLseq was conceived specifically for somatic analysis. However, recent research points to germline variants as a driving mechanism in familiar cases [10]. In suspected cases, germline origin must be confirmed in culture skin fibroblasts according to current recommendations. If confirmed, the patient should be assigned to a specialized unit [11,12]. ALLseq sequencing metrics were equivalent to those reported by commercial panels such as the OCCRA, which were also in line with its Illumina counterpart [13]. The overall performance of ALLseq, as assessed by sensitivity, specificity, PPV, NPV, and accuracy, was 100% on SNVs, indels, and fusions, and only CNV identification was slightly poorer. The reliable identification of CNVs using NGS continues to be a challenge due to unequal target coverage [14]; however, ALLseq CNV reliability was acceptable according to Cohen’s criterion. The LoD was set at 2% VAF for SNVs and indels, which is acceptable in the clinical context, as most protocols establish 5% as the cut-off value for considering variants as clinically relevant [15,16]. Although the LoD is sufficient for SNV and indel detection, special attention to low blast % samples is required when analyzing CNV and gene expression with NGS, as recommended by Jennings et al. [17]. The main limitation of ALLseq, like most targeted panels, is its inability to identify aneuploidies, which are present in up to 30% of pediatric ALL cases. However, these alterations are easily identified by cytogenetic techniques, which are routinely performed in all laboratories. It cannot detect DUX4 deregulation (which represents 7% of B-ALL), which has been considered an independent entity by the WHO 2022 and ICC 2022 classifications and confers favorable prognosis [18]. DUX4 deregulation is technically difficult to detect, as it is located within a repetitive region on chromosome 4q, with an almost identical locus on 10q. Thus, primers can bind to multiple loci on both 4q and 10q. Additionally, DUX4 fusions show great variability in breakpoints [19]. ALLseq allows 634 fusions to be identified (including ABL-class translocations and virtually all of the class-defining fusions), as well as the aberrant expression of seven genes including CRLF2. Notably, up to 50% of Ph-like ALLs overexpress this gene due to fusions, mutations, and alterations in the JAK-STAT pathway [20]. Therefore, with the ALLseq design, most Ph-like patients can be diagnosed, allowing the identification of candidates for targeted therapies with tyrosine kinase or JAK-STAT inhibitors (NCT03571321). In fact, we were able to define the ALL subtype of around 30% of patients, among whom 7% were Ph-like cases overexpressing CRLF2. The recent update of the WHO 2022 and new ICC ALL classifications included, for the first time in this disease, categories defined by point mutations. Moreover, the potential use of targeted therapies in patients carrying SNVs or indels highlighted the clinical relevance of these alterations in ALL. In fact, ALLseq identified point mutations in FLT3, the NOTCH1 pathway, or JAK family genes, for which targeted inhibitors have been developed [21], in 16% of patients. Regarding CNV identification, IKZF1 deletions have been classically recognized as conferring poor prognosis [22]. More recently, several European groups have developed different CNV profiles that are significant for risk stratification. In particular, the COALL has proposed the IKZF1plus group, defined by the deletion of IKZF1 co-occurring with at least one additional deletion in CDKN2A/Bhomo, PAX5, or the pseudo autosomic region 1 (PAR1) in the absence of ERG deletion, which distinguishes high-risk ALL patients who benefit from treatment intensification [23]. Similarly, the British group proposes a CNV profile (UKALL-CNA) involving IKZF1, CDKN2A/B, PAR1, BTG1, EBF1, PAX5, ETV6, and RB1 to refine risk groups [24]. The ALLTogether (NCT04307576) treatment protocol incorporates the UKALL-CNA risk stratification, making CNV assessment mandatory. With ALLseq, we were able to correctly risk-stratify around 90% of patients. Cytogenetic approaches have been the main diagnostic tool in ALL, given the exclusive importance of aneuploidies and a few translocations in this disease just a decade ago [25]. However, the development of “omic” technologies has substantially broadened the spectrum of molecular lesions that explain the onset of ALL. In this context, NGS has been incorporated as a complementary tool into most clinical laboratories [26]. Our results show that the combination of ALLseq and cytogenetics provide clinical information to virtually all patients, reducing the number and type of assays necessary for ALL characterization (RT-PCR, MLPA, SNP arrays, etc.). It is worth mentioning emerging technologies, such as optical genome mapping, which will surely be useful for the analysis of ALL patients. This technique, unlike NGS targeted panels, is not restricted to a list of genes and therefore is able to detect novel alterations [27]. Moreover, it can detect numerical and structural chromosome alterations as well as gene-level gains or losses, which makes it very useful for ALL characterization. However, it is unable to detect SNVs and indels which, as discussed above, are currently needed for ALL diagnoses [3,4]. The continuous availability of novel genomic methodologies makes it difficult to define the optimal technology(ies) for ALL characterization. The molecular knowledge of ALL is an evolving field; therefore, the best diagnostic approach has to be flexible and adapt to conform to the latest guidelines. In this context, interdisciplinary groups carrying out an integrated diagnosis become essential in configuring the diagnostic workflow of ALL [28]. In conclusion, ALLseq allows the most frequent alterations in ALL to be identified. Although there are certain limitations to be considered when interpreting the results, the panel constitutes a useful tool for patient characterization and management, as it allows the identification of driver and secondary alterations in a single experiment, thus permitting accurate diagnosis, patient risk stratification, and (in some cases) treatment selection. The study included pediatric and adolescent (≤18 years old) ALL patients diagnosed at Hospital Universitari i Politècnic La Fe (Valencia, Spain). Inclusion criteria were as follows: availability of high-quality DNA and RNA from bone marrow or peripheral blood, and written informed consent in accordance with the recommendations of the Declaration of Human Rights and the Conference of Helsinki. The Institutional Ethics Committee for Clinical Research approved this study (approval numbers 2021-045-1 and 2022-09-04). A custom panel targeting ALL (ALLseq) was designed using the White Gloves Service from Thermo Fisher Scientific. Target selection was based on its potential clinical utility according to 2 levels of evidence: Level 1: clinical guidelines and clinical trials: (a) alterations included in the WHO classification of hematolymphoid tumors in force at the time of the start of the study [29]; (b) alterations defining genetic ALL subtypes [30]; (c) alterations used for risk stratification by international cooperative groups [31], NCT04307576]; (d) alterations used for potential targeted therapy [32]. Level 2: other pathogenic alterations described in large cohorts: (a) variants that cluster into specific subtypes of ALL [15,33]; (b) variants associated with good or bad prognosis but not currently used for patient risk stratification [34]; (c) variants that confer resistance to specific drugs in vitro/in vivo experiments [35]. The sequencing workflow was carried out on Ion Torrent platforms (Thermo Fisher Scientific, San Francisco, CA, USA). DNA libraries were generated from 10 ng of DNA, with an initial PCR consisting of 17 cycles and 4 min of extension time; for RNA libraries, cDNA was generated with the SuperScript™ IV VILO™ kit (Thermo Fisher Scientific) from 10 ng of total RNA, and PCR was performed with 20 cycles and 4 min of extension time. Library and template preparation was carried out automatically on the Ion Chef™ Instrument (Thermo Fisher Scientific) using the Ion AmpliSeq™ Kit for Chef DL8 (Thermo Fisher Scientific) and the Ion 510™ & Ion 520™ & Ion 530™ Kit-Chef (Thermo Fisher Scientific), respectively. Libraries from eight samples were loaded onto an Ion 530™ Chip (Thermo Fisher Scientific) and sequenced on an Ion S5 sequencer (Thermo Fisher Scientific). Human genome build 19 was used as the reference genome. Base calling was performed on Torrent Suite software version 5.10.0 (Thermo Fisher Scientific). Variant identification was accomplished with the Variant Caller Plugin (Thermo Fisher Scientific), and variant annotation was performed using Ion Reporter (IR) software version 5.10.3.0 (Thermo Fisher Scientific). For CNV assessment, the normalized read depth of each sample was compared with that of the reference baseline (generated by sequencing 20 healthy controls). The IR software applied an algorithm based on a hidden Markov model, which predicts the copy number or the ploidy state. A copy number of 2 was considered as normal, values ≥ 3 were considered as amplifications, and a copy number ratio of one or zero suggested heterozygous or homozygous deletions, respectively. For technical validation, 25 retrospective patients harboring a >90% blast count in bone marrow or peripheral blood and a complete molecular characterization were selected. A second validation round was carried out by sequencing sequential unbiased ALL samples to assess the clinical utility of the panel. All the ALL samples were analyzed at diagnosis by the following methods: conventional cytogenetics, an ALL FISH custom panel (Cytocell Ltd., Cambridge, UK), RT-PCR to asses ETV6::RUNX1 and BCR::ABL1 fusions [36], and MLPA SALSA P335 ALL-IKZF1 (MRC Holland, Amsterdam, NL). The Oncomine Childhood Research Assay (OCCRA; Thermo Fisher Scientific) was used retrospectively to further characterize ALL samples following the manufacturer’s instructions. After sequencing, variant filtering was performed on IR software version 5.10.3.0 (Thermo Fisher Scientific). For ALLseq technical validation, intronic and synonym variants were filtered out, whereas pathogenic and likely pathogenic variants, as well as variants of unknown significance (VUS), were retained. Variants detected by NGS (OCCRA and/or ALLseq) were confirmed by Sanger sequencing and Minor Variant Finder software (Thermofisher Scientific) (SNVs and indels) or qRT-PCR (fusions and gene expression alterations). Fusion characterization and gene expression were assessed by qRT-PCR on a LightCycler 480 II (Roche Diagnostics, Switzerland, AG) using Sybr green and ABL1 as the control gene. Primers and PCR parameters are described in Supplemental Tables S9 and S10. For selected samples, optical genome mapping (OGM) was used following the manufacturer’s instructions in order to confirm CNVs not included in the MLPA SALSA P335 ALL-IKZF1. A DNA result was considered evaluable if it met the following requirements: mean read depth ≥ 1500× per sample; uniformity and on-target reads ≥ 80%; and MAPD < 0.5, confidence > 20, and precision > 10 for CNV analysis. Regarding RNA, a minimum number of 20,000 mapped reads was established; gene and fusion expression levels were calculated as [(target reads × 1000)/total RNA reads]. The mean on-target and uniformity percentages, depth of coverage, and MAPD yielded by ALLseq were compared with those from the OCCRA. Sensitivity [true-positive (TP)/(TP + false-negative (FN))], specificity [true negative (TN)/(TN + false-positive (FP))], precision [(TP + TN)/n], positive predictive values (PPVs) [TP/(TP + FP)], and negative predictive values (NPVs) [FN/(FN + TN)] were assessed by comparing ALLseq results with data from orthogonal techniques described above. The Bland–Altman method was used to assess the variant allele frequency (VAF) agreement between ALLseq and OCCRA panels. CNV detection reliability was further evaluated with Cohen’s kappa coefficient, where a value > 0.8 indicates a high agreement with the gold-standard method. Fusion expression linearity was assessed by diluting an ETV6::RUNX1-positive sample at 1:10, 1:100, and 1:1000 ratios into a negative control. Gene expression was quantified by ALLseq and qRT-PCR and compared. The overexpression cutoff was established at 105 expression units (target reads × 104/total mapped reads). In order to calculate analytical performance for gene expression, expression units were dichotomized (overexpression vs. no overexpression) according to the cutoff criteria. To assess the LoD, repeatability, and reproducibility, a DNA pool from samples harboring NOTCH1 (p.Phe2509fs, VAF 6.55%; p.Phe1606_Lys1607insAspSerPro, VAF 7.25%), NRAS (p.Gly12Cys, VAF 17.47%), KRAS (p.Leu19Phe, VAF 16.37%), and/or DNM2 (p.Arg123Ter, VAF 21.16%) was created. Two serial VAF dilutions (ratios of 1:2 and 1:4) were prepared using a wild-type control sample. For each dilution, two independent libraries were sequenced twice in back-to-back experiments. In these experiments, a coefficient of variation (CV) ≤ 20% was considered acceptable for VAF values. In order to obtain the CNV LoD, two samples harboring CDKN2A/B homozygous deletion (CDKN2A/Bhomo) and IKZF1 heterozygous deletion (IKZF1hetero), respectively, were combined at different ratios (CDKN2A/Bhomo:IKZF1hetero; 3:1, 1:1, and 1:2) creating different copy number ratios. A sample pool was analyzed in three consecutive experiments to test the reproducibility and repeatability. A total of 43 correlative patients were prospectively analyzed. In these patients, pathogenic and likely pathogenic variants were confirmed with complementary methods, as described above. Medians of quantitative variants were compared with Mann–Whitney’s U test. Qualitative parameters were compared with the chi-square’s test and Cohen’s kappa coefficient. A Bland–Altman plot was used to assess the agreement between VAF values obtained by ALLseq and OCCRA panels. Box plots were generated with BoxPlot R (http://shiny.chemgrid.org/boxplotr, accessed on 16 December 2022). A Circos plot was generated as described by Krzywinski et al. [37]. A landscape diagram was created using Oviz-bio, a free web-based platform for interactive data visualization [38].

PMC10002328

Yu-Lei Chen,Xin-Xin Xie,Ning Zhong,Le-Chang Sun,Duanquan Lin,Ling-Jing Zhang,Ling Weng,Tengchuan Jin,Min-Jie Cao

Research Progresses and Applications of Fluorescent Protein Antibodies: A Review Focusing on Nanobodies

21-02-2023

fluorescent protein,monoclonal antibody,nanobody,research progress,application

Since the discovery of fluorescent proteins (FPs), their rich fluorescence spectra and photochemical properties have promoted widespread biological research applications. FPs can be classified into green fluorescent protein (GFP) and its derivates, red fluorescent protein (RFP) and its derivates, and near-infrared FPs. With the continuous development of FPs, antibodies targeting FPs have emerged. The antibody, a class of immunoglobulin, is the main component of humoral immunity that explicitly recognizes and binds antigens. Monoclonal antibody, originating from a single B cell, has been widely applied in immunoassay, in vitro diagnostics, and drug development. The nanobody is a new type of antibody entirely composed of the variable domain of a heavy-chain antibody. Compared with conventional antibodies, these small and stable nanobodies can be expressed and functional in living cells. In addition, they can easily access grooves, seams, or hidden antigenic epitopes on the surface of the target. This review provides an overview of various FPs, the research progress of their antibodies, particularly nanobodies, and advanced applications of nanobodies targeting FPs. This review will be helpful for further research on nanobodies targeting FPs, making FPs more valuable in biological research.

Research Progresses and Applications of Fluorescent Protein Antibodies: A Review Focusing on Nanobodies Since the discovery of fluorescent proteins (FPs), their rich fluorescence spectra and photochemical properties have promoted widespread biological research applications. FPs can be classified into green fluorescent protein (GFP) and its derivates, red fluorescent protein (RFP) and its derivates, and near-infrared FPs. With the continuous development of FPs, antibodies targeting FPs have emerged. The antibody, a class of immunoglobulin, is the main component of humoral immunity that explicitly recognizes and binds antigens. Monoclonal antibody, originating from a single B cell, has been widely applied in immunoassay, in vitro diagnostics, and drug development. The nanobody is a new type of antibody entirely composed of the variable domain of a heavy-chain antibody. Compared with conventional antibodies, these small and stable nanobodies can be expressed and functional in living cells. In addition, they can easily access grooves, seams, or hidden antigenic epitopes on the surface of the target. This review provides an overview of various FPs, the research progress of their antibodies, particularly nanobodies, and advanced applications of nanobodies targeting FPs. This review will be helpful for further research on nanobodies targeting FPs, making FPs more valuable in biological research. Fluorescent protein (FP) is the most commonly used tool protein in biomedical science research. Following the increased use of FP, FP antibodies have also been developed. Compared to the traditional antibody (Figure 1A), heavy-chain antibodies (HCAb) from camelid (Figure 1B) and immunoglobulin new antigen receptors (IgNAR) from sharks (Figure 1C) naturally lack light chains. Single-domain antibody (sdAb) is an antibody entirely composed of variable domains of HCAb or IgNAR. Its molecular weight is 12–15 kDa, which is about 1/10 the size of a monoclonal antibody; hence, it is also referred to as nanobody. The nanobody, with a smaller size, high stability against acid and heat, strong penetration capability, ease of genetic engineering and expression, and high affinity, has been applied in various fields by scientific researchers. It can easily access grooves, seams, or hidden antigenic epitopes on the surface of the target, recognizing many antigens that conventional antibodies cannot recognize [1]. Hence, nanobodies have been used as biomolecules in organisms and cell culture systems in developmental biology, crystalline chaperones for the conformational state in stable biology, regulators or inhibitors of enzyme activity, immunohistochemical reagents for biochemical analysis, and secondary antibodies in immunoblotting and immunofluorescence [2,3,4,5,6]. Nanobodies that specifically bind FP are widely used in subcellular localization, intracellular signaling pathway studies, live cell imaging, and targeted nanomaterials [7,8,9]. In particular, nanobodies have unique advantages over traditional IgG antibodies in super high-resolution imaging. Luminescence is a common phenomenon in marine invertebrates. Coelenterates, including jellyfish, hydra, and coral, emit green fluorescence under ultraviolet (UV) or blue light, while ctenophores emit blue fluorescence. Green fluorescent protein (GFP) was first identified in Aequorea victoria by Shimomura et al. in 1962 [10], and then cloned and expressed in 1985 [11]. The unique properties and stable structure of wild-type GFP (wtGFP) have, accordingly, attracted many researchers to study it. Since then, based on the existing knowledge and its crystal structure, Tisen and his colleagues generated various GFP proteins with different fluorescent properties by mutation and elaborated the luminescence mechanism of GFP [12]. Concerning GFP-related research, three scientists (Shimomura, Chalfie, and Tsien) won the 2008 Nobel Prize in Chemistry for discovering GFP and their outstanding contributions to its application. Ancestral red fluorescent protein (RFP), namely, DsRed, was cloned from non-bioluminescent reef corals in 1999 [13]. RFP can be used together with GFP to solve scientific problems that GFP alone cannot solve. Most importantly, due to the low background of RFP in intracellular imaging, it is more suitable for applications in bioscience research [14,15]. With the efforts of many scientists, the fluorescence spectra of FPs have been reported to cover the entire visible region and even the near-infrared region, providing a wealth of tools for visualizing and quantifying proteins in living cells [16]. Figure 2 summarizes the development of FPs. GFP is a naturally occurring, globular, and soluble acidic FP. The primary structure of GFP consists of a monomer composed of 238 amino acid residues, with a molecular weight of 27–30 kDa. Its crystal structure displays a center formed by 11 anti-parallel β chains, forming a cylindrical, tightly packed β-barrel structure. The β-barrel center is protected by a light-emitting group of 4-(p-hydroxybenzylidene)-5-imidazolinone connected to its α-helix, with both ends sealed by short α-helix segments [17,18]. The maturation of the chromophore requires no cofactor other than O2 [19]. GFP absorbs blue light or UV light and emits green fluorescence, with the main excitation peak at 395 nm, the lowest excitation peak at 475 nm, and the emission peak at 509 nm. GFP is extremely stable and can easily withstand high-temperature treatment. Formaldehyde fixation or paraffin embedding does not affect its fluorescence characteristics. Through the in-depth investigation of the structure and maturation process of FPs, more and more researchers redesign the structure of FPs to provide them new functions and properties, further promoting their development and application. The limiting factors in GFP applications are pH, chloride ion sensitivity, and poor photostability. Targeted modifications of GFP correct these flaws, boosting its quantum yield, brightness, molar extinction coefficient, and photostability. In addition, several GFP derivatives have been established to broaden the range of colors emitted, including blue, ultramarine, cyan, and yellow. In response to the low fluorescence intensity of wtGFP under blue light excitation and its unstable expression in mammalian cells, Zhang et al. engineered an amino acid substitution of GFP (F64L and S65T), namely, enhanced GFP (EGFP), whose fluorescence intensity is increased by 35-fold [20]. EGFP remains the most commonly used FP mutant in the green light region, given its good photostability and high brightness. In addition, the folding properties of GFP have been optimized, obtaining a superfolder GFP (sfGFP) with super folding ability. The sfGFP has six new mutations, including S30R, Y39N, N105T, Y145F, I171V, and A206V. sfGFP can efficiently fold even when expressed in fusion with insoluble proteins, ensuring the brightness of its fluorescence [21]. GFPuv, another modified form of GFP, is optimized for fluorescence under UV light. It generates brighter green fluorescence in the presence of UV light with an excitation peak at 395 nm and an emission peak at 509 nm and can be utilized to pinpoint the location of different intracellular proteins [22,23]. Three amino acid substitutions occur in GFPuv, including F99S, M153T, and V163A. These mutations induce an 18-fold stronger expression of GFPuv than wtGFP in E. coli. The GFPuv gene is a synthetic gene where the five low-frequency Arg codons are replaced by codons suitable for E. coli to ensure effective expression of GFPuv. The site-directed mutation of Y66H constructs the blue fluorescent protein (BFP) with an excitation peak at 384 nm and an emission peak at ~445 nm. Its excitation spectrum is close to UV light, which damages cells during operation, and its short emission wavelength induces autofluorescence in cells. Enhanced BFP (EBFP) is weakly luminescent and poorly resistant to photobleaching and acid, with a high background signal for cellular imaging. Three brighter EBFP mutants have been developed, including Azurite [24], EBFP2 [25], and mTagBFP [26], which are 1.6-, 2.0-, and 3.7-fold brighter than EBFP, respectively. mTagBFP, the brightest BFP, is derived from a mutant of the red fluorescent protein (RFP) TagRFP [27], with an extinction coefficient of 52,000 M−1 cm−1 and a quantum yield of 0.63, which significantly improves its resistance to photobleaching. Mutation of Y66F in wtGFP results in an ultramarine fluorescent protein with an emission wavelength of 442 nm [28]. However, the low fluorescence quantum yield of the GFP-Y66F variant severely limits its applicability in imaging. Further mutation of GFP-Y66F with amino acid substitutions of T65Q, Y145G, H148S, and T203V results in the ultramarine fluorescent protein Sirius, which is 25 times brighter than GFP-Y66F [29]. Cyan fluorescent protein (CFP) has an excitation peak at 449 nm and an emission peak at ~482 nm, with spectral properties ranging between BFP and EGFP. Similarly to BFP, a site-directed mutation of Y66W is used to construct CFP. CFP has wide applications in multicolor imaging and localization research due to its outstanding photostability. For example, the gene encoding CFP has been optimized for human codon preference and is now commercially available under the trade name AmCyan1 (Clontech). Additionally, there are several mutants brighter than enhanced CFP (ECFP), such as Cerulean, mTFP (twice as bright as Cerulean), and mTurquoise (about twice as bright as ECFP) [21,30]. Mutations in the yellow fluorescent protein (YFP) are not restricted in the core motif of the chromophore Y66 but in residues structurally adjacent to Y66. The excitation peak of YFP is ~518 nm, with an emission peak of ~531 nm. Compared with GFP, the fluorescence of YFP shifts toward the red spectrum, which is mainly due to the substitution of Thr with Tyr at position 203. Enhanced YFP (EYFP) is one of the most fluorescent and extensively utilized FP biosensors to detect intracellular pH and chloride ion concentrations. However, it is very susceptible to acid and chloride ions and is less photostable than many jellyfish-derived FPs [31]. mCitrine and mVenus, improved versions of EYFP, are currently the most used YFP [32,33]. The photophysical properties of GFP and its derivatives are summarized in Table 1. RFP is another widely used FP. DsRed is a tetramer that tends to form multimers when performing protein fusions and can be toxic when expressed intracellularly, so it has been engineered to obtain monomers. The most notable RFP is the “mFruit” family, including mCherry, mBanana, mOrange, dTomato, mTangerine, and mStrawberry (Table 1), of which mCherry, with mutations of K163Q and K83L, is the most preferred. mCherry has fast maturation, good monomeric properties, and better photostability, although it is less bright [34]. However, TagRFP in its monomer form is about three to four times brighter than mCherry, making it a relatively bright monomeric RFP that is currently available [27]. mKate is a novel monomeric far-red fluorescent protein derived from a four-site mutation of TagRFP with an excitation peak at 588 nm and an emission peak at 635 nm, which is only 45% as bright as EGFP; its dimerization protein, Katushka, is 67% as bright as EGFP [35]. mKate2, a mutant of mKate with amino acid substitutions of V38A, S165A, and K238R, has approximately double the brightness of mKate [36]. Compared to GFP and RFP, mKate and Katushka exhibit better imaging depth and richer optical signals for intra-object fluorescence imaging [35]. Hence, developing far-red fluorescent proteins is more valuable for in vivo bioimaging. Near-infrared (NIR) FPs are highly desired as protein tags in imaging applications. Most NIR FPs are designed from bacterial phytochrome photoreceptors (BphPs) [37]. The first NIR FP used in live animal imaging is IFP1.4, a fluorescent protein derived from BphP that fluoresces by binding to a biliverdin (BV) chromophore [38]. By DNA shuffling and random mutagenesis, a brighter IFP2.0 is then developed [39]. Although monomeric IFP1.4 and IFP2.0 can be achieved by breaking the dimerization interface of BphP, they still tend to dimerize at high concentrations. A natural monomeric infrared fluorescent protein (IFP), mIFP, was designed in 2015 [40]. mIFP has been proven to have sound imaging effects in Drosophila larvae and neurons. Shcherbakova et al. designed three bright monomeric NIR FPs with distinct spectra, namely, miRFP670, miRFP703, and miRFP709 [41]. The BphP-derived NIR FPs minimally require two domains, PAS (Per-ARNT-Sim) and GAF (cGMP phosphodiesterase-adenylate cyclase-FhlA), to covalently attach a BV chromophore and also possess a complex “figure-of-eight knot” structure topologically linking the GAF and PAS domains, which affects their folding. Hence, another class of bacterial photoreceptors, allophycocyanins (APCs), is used to engineer NIR FPs, such as smURFP [42]. Although APC-based NIR FPs are smaller, they bind BV less efficiently, leading to significantly lower brightness in mammalian cells than BphP-derived NIR FPs. To overcome the defects of BphP- and APC-based NIR FPs, a single-domain NIR FP named miRFP670nano was developed from cyanobacteriochrome (CBCR), representing the first CBCR-derived NIR FP that can efficiently bind endogenous BV chromophore and emit bright fluorescence in mammalian cells [43]. An essential advantage of miRFP670nano over BphP-derived NIR FPs is high photostability. The enhanced miRFP670nano3, with 14 mutations relative to the parental miRFP670nano, exhibits similar photostability to miRFP670nano [44]. Table 2 summarizes the photophysical properties of NIR FPs. The discovery and application of FPs have provided a powerful research tool for modern biology. Nowadays, FPs have become one of the common tools scientists use to extend their applications to many research areas, such as gene expression regulation, organelle labeling, signal transmission, drug screening, and biomolecular interactions. Most cloning vectors expressing FPs have been commercialized and are available through commercial companies. Antibody is a kind of immunoglobulin secreted by B lymphocytes. Monoclonal antibodies consist of two heavy chains and two light chains, with the heavy chain consisting of one variable region (VH) and three constant regions (CH), and the light chain consisting of one variable region (VL) and one constant region (CL). The heavy chains are covalently linked by disulfide bonds, and the CL region of the light chain is non-covalently linked to the CH1 domain of the heavy chain to form a stable antibody molecule [45]. Due to their specific receptor binding capacity, monoclonal antibodies produce a variety of biological activities, such as classical blocking, neutralization, complement activation, the killing of target cells through the Fc receptor, and the regulation of immune activity. They are critical biological macromolecules widely applied in immunoassay, in vitro diagnostics, and drug development. Polyclonal antibodies are a mixture of heterotypic antibodies derived from the immune response process of multiple B cells, and each antibody recognizes a different epitope of the same antigen [45]. Compared to polyclonal antibodies, monoclonal antibodies specifically detect an epitope on the antigen and are less likely to cross-react with other proteins, thus producing low background staining signals. Moreover, the reproducibility of results concerning monoclonal antibodies is higher than that of polyclonal antibodies under the same experimental conditions. GFP is a unique in vivo reporter that can be analyzed for gene expression in many species. Gengyoando and Mitani used the glutathione-S-transferase (GST) fusion protein, which contains the full-length GFP coding region and a synthetic peptide corresponding to residues Ser208-His217 of GFP, as an immunogen to create the monoclonal antibody 65B12 against GFP [46]. Immunoblot analysis demonstrates that 65B12 specifically bound the GFP fusion protein. Furthermore, it can recognize the fluorescent cells in transgenic animals expressing the uric-86-gfp reporter construct; hence, it can be applied in immunohistochemistry. Zhuang et al. developed an improved method to purify GFP protein [47]. The monoclonal antibody against GFP, FMU-GFP.5, was prepared by immunizing mice using purified GFP as an antigen. GFP with high purity (>97% homogeneity) and sample yield (>90%) is purified using a straightforward 2-step technique using mAb FMU-GFP.5-coupled Sepharose 4B resin. In addition, all the functional recombinant target proteins coupled to GFP can be easily and directly isolated from cells, owing to the GFP epitope. These data suggest that this method is more effective in purifying GFP than any available method and resolves the low yield and purity challenge in most GFP purification methods. The inherent properties of monoclonal antibodies, such as their large molecular weight, complex structure, and limited biological activity, have increasingly restricted their further applications. Therefore, it is urgent to develop alternatives to monoclonal antibodies. In 1989, Ward et al. prepared a variable domain of heavy-chain only antibody with weak binding ability toward lysozyme, spiking interest in sdAb research [48]. In 1993, Hamers-Casterman et al. discovered a new type of antibody in camel serum for the first time, which is entirely different from the traditional mammalian antibody [49]. This type of antibody is named HCAb due to its natural lack of light chains. HCAb consists of CH2, CH3, the hinge region, and the variable domain of HCAb (VHH), but it still has full antigen-binding capacity. In 1995, Greenberg et al. discovered heavy-chain only antibodies, known as IgNAR, in nurse shark [50]. It exists in both secretory and membrane-bound forms, and consists of a variable region, named variable new antigen receptor (VNAR), and several constant regions. Subsequently, IgNAR has been found in wobbegong shark, spiny dogfish, horn shark, and white-spotted bamboo shark, supplementing the sdAbs repertoires. A nanobody possesses the characteristics of small molecular weight, high affinity, strong stability, good solubility, strong tissue penetration, and recognition of hidden antigen epitopes. It has attracted increased attention in disease diagnosis, immune reagents, pathogen detection, and drug development [51]. With the development of molecular biology techniques and the improvement of genetically engineered antibody preparation technology, nanobody has become a popular research field in immunoassay. Figure 3 summarizes the development of nanobody to date. Fluorescent proteins have changed cell biology and biochemistry by offering simple-to-use gene-encoded fluorescent protein markers. Several tight binding agents for research and drug targets have been developed through the synthesis and selection of protein scaffold libraries during parallel development [52]. An example is the development of nanobody, which is easier to select with improved stability, solubility, and yield [53]. In contrast to traditional antibodies, these small and stable nanobodies are functional in living cells. Therefore, nanobody with specific binding activity to FP is a potent tool for FP fusion, separation, and cellular engineering in various areas of biological research. Camelidae-derived nanobodies targeting GFP Various proteins with excellent subcellular localization characteristics have been fused to GFP, creating visual antigens to detect proteins directly in various subcellular compartments. In 2006, Rothbauer et al. screened a GFP-specific antibody fragment (cAbGFP4) by immunizing alpaca with GFP [54]. The surface plasmon resonance assay (SPR) of the interaction between cAbGFP4 and the GFP antigen showed a high affinity (KD = 0.23 nM). In addition, the anti-GFP nanobody was combined with a monomeric RFP, producing the visible GFP-binding antibody employed to examine the distribution of the anti-GFP nanobody in living cells. By gel filtration, immunoblotting, and confocal microscopy assays, GFP-specific nanobody was demonstrated to be stably distributed in mammalian cells without detectable protein degradation or aggregation commonly found with single-chain variable fragments. Furthermore, Kubala et al. characterized the GFP:cAbGFP4 complex (Figure 4A) by X-ray crystallography and isothermal titration calorimetry (ITC) [55], revealing the basis for high affinity and specificity of nanobodies in protein binding. In 2020, four distinct anti-GFP nanobodies, termed A12, B9, D5, and E6, were identified using phage display [56]. Native PAGE and immunoprecipitation assays revealed that these nanobodies could bind GFP in vitro and in vivo. Protein conformation is closely related to function and is usually controlled by regulatory factors. Axel et al. identified seven GFP-specific binders, namely GFP-binding proteins (GBPs) 1–7 [57]. Among them, GBP1 increased GFP fluorescence by 10-fold, while GBP4 induced a 5-fold decrease in GFP fluorescence; hence, these were referred to as “enhancer” and “minimizer”, respectively. Structural analyses of the GFP–nanobody complex (Figure 4B,C) revealed that the two nanobodies caused modest opposing changes in the chromophore milieu, altering its absorption characteristics [58]. Furthermore, 25 GFP-specific nanobodies (LaGs) were identified, with KD values ranging from 0.5 nM to over 20 μM [59]. A bivalent nanobody (LaG-16-LaG-2) showed the highest affinity, with a KD value of 36 pM. Moreover, the maximum fluorescence intensity of GFP increased by ~60% when incubated with excess LaG protein. Zhang et al. reported the crystal structure of GFPuv complexed with LaG16 (Figure 4D) at 1.67 Å resolution [60]. The binding site of LaG16 on the GFP β-barrel was located on the other side of the GFP enhancer. Hence, LaG16 and GFP-enhancer were fused with a (GGGGS)4 linker. The bivalent nanobody had an affinity of 0.5 nM, demonstrating the feasibility of designing ultra-high-affinity target protein binders by dimerization of 2 nanobodies binding with different epitopes. In 2014, Twair et al. prepared sfGFP and immunized an adult one-humped camel. Seven anti-sfGFP nanobodies targeting three epitopes (NbsfGFP01, 02, 03, 04, 06, 07, and 08) were isolated [61]. Based on ELISA and immune-blotting assays, these nanobodies recognized sfGFP labeled as free or fused to growth hormone. In addition, the crystal structure of nanobody NbsfGFP02 complexed with sfGFP (Figure 4E) was established at a resolution of 2.2 Å. The affinity between NbsfGFP02 and sfGFP was determined to be 15.8 nM by biolayer interferometry (BLI). The melting temperature was 75.6 ℃ for NbsfGFP02 [62]; hence, it is a prospective GFP nanobody candidate in applications that demand harsh testing conditions. Most nanobodies can be functionally expressed in vivo via plasmid transfection into eukaryotic cells. Hence, nanobodies are an excellent tool for identifying structural or dynamic features in living cells. In 2020, Zhou et al. proposed a novel method of expressing the constructed GFP-binding nanobodies (cAbGFP4) as in vitro transcription (IVT) mRNA, referred to as nanobody-mCherry [63]. The mRNA with untranslated regions and reverse cap analogues capped with chemically modified nucleotides and poly(A) tail was prepared in vitro and used for transfection. In contrast to the nanobody expressed using the plasmid DNA, the anti-GFP nanobody expressed using IVT mRNA was identified within 3 h of transfection and degraded within 48 h. Therefore, expressing the encoded mRNA of nanobody in living cells allows efficient delivery of the nanobody. Shark-derived nanobodies targeting GFP. Wei et al. demonstrated that bamboo sharks produced an effective immune response against GFP immunization, characterized by elevated lymphocyte counts and antigen-specific IgNARs [64]. In total, 7 anti-GFP nanobodies, including BsG3, 73, 80, 89, 93, 98, and 105, with an affinity of up to 0.3 nM, were isolated from immunized bamboo sharks, implying that bamboo sharks manufacture high-affinity IgNARs. In addition, the bi-paratopic VNARs with the highest affinity to GFP (20.7 pM) were constructed, and the character of anti-GFP nanobodies as intrabodies was validated in mammalian cells. These findings will speed up the research and progress of bamboo shark sdAbs to provide low-cost and easy-to-use nanobodies for the biomedical industry. Nanobodies targeting other fluorescent proteins. mWasabi is a bright monomeric green fluorescent protein. Li et al. successfully constructed an antibody library of 4 × 107 transformants by immunizing camels with mWasabi as an antigen, and screened 3 high-affinity mWasabi-specific nanobodies, termed Nb4, Nb6, and Nb27 [65]. These nanobodies recognized mWasabi alone or when combined with programmed death 1 (PD-1). In total, 6 nanobodies (LaMs) targeting mCherry with high specificity were identified, with KD values ranging from 0.18 nM to 63 nM [59]. In addition, three iRFP713-specific nanobodies (BSR1, BSR3, and BSR4) with nanomolar binding affinities were isolated [64]. These data suggest that immunization of bamboo sharks can produce high-affinity nanobodies. Overall, the ability of nanobodies to bind cellular proteins and attract FP fusion proteins enables precise control of cellular processes and structures in living cells. This multifunctional FP-nano trap enables microscopic, biochemical, and functional analyses with a unique combination of the same protein. Nanobodies targeting FPs are summarized in Table 3. Understanding protein function requires reliable and quantifiable high-resolution protein localization. Although the use of antibodies to label target proteins has been well established in molecular biology, this technique is constrained by the size and multivalency of conventional antibodies. Given the small size of nanobodies, they can be used as tracers for intracellular imaging after ligation with fluorescent molecules, enzymes, peptides, receptors, biotin, and other drugs. For instance, nanobodies against FP were used in super-resolution microscopy imaging when tagged with organic dyes [66,67]. Ptk2 cells that continuously express tubulin-YFP were imaged using this approach, and the resolution was improved to 269 ± 37 Å, as opposed to approximately 450 Å with conventional antibodies. Nanobody-mediated labeling offers a quick and versatile method to label almost any commonly accessible FP-derived fusion structure for sophisticated single-molecule localization microscopy (SMLM) imaging. Specifically, two-color SMLM can investigate the subcellular localization of any functional GFP and RFP fusion constructs when nanobodies against GFP and RFP are used simultaneously [68]. Thus, numerous biological problems can be promptly addressed using two-color SMLM imaging. Ariotti et al. proposed a modular approach for enzyme-based protein labeling, allowing for improved speed and sampling for analyzing subcellular protein distributions to EM-resolution [7]. By designing GBP4 directly to the modified soybean ascorbate peroxidase (APEX) tag, it was shown that APEX could be directed to any GFP-labeled protein of interest. APEX-GBP4 fusion provides notable high-resolution protein localization to the organelle subdomains and significantly shortens the time for characterizing subcellular protein distributions. Furthermore, it permits EM-resolution of GFP-labeled proteins expressed at endogenous levels. A toolbox of FP-specific nanobody-encoding plasmids was generated and fused into functional modules. This toolbox enables the visualization and manipulation of intracellular signaling pathways in living cells, significantly expanding its uses in vivo [9]. These include fluorescent sensors for dynamic visualization of Ca2+, H+, and ATP/ADP, and oligomeric or heterodimeric modules that allow protein recruitment or isolation and recognition of membrane contact sites between organelles. In 2017, Herce et al. used cell-penetrating peptide (CPP) to transport antibodies directly into cells for immunolabeling and antigen manipulation [69]. A system that took advantage of the high affinity of cyclic arginine-rich CPP toward RNA in the nucleolus was constructed. By linking the cyclic arginine-rich CPP to the GFP nanobody, the re-localization of GFP in the nucleolus can be directly observed in cells (Figure 5A). Therefore, this visualized system can be used to track the protein’s location and to compare the efficiency of different CPPs quantitatively. The fluorescence resonance energy transfer (FRET) technique is widely used in life science research because it enables dynamic real-time detection of signaling molecules under physiological conditions in living cells. Due to its exceptional sensitivity and specificity, time-resolved Förster resonance energy transfer (TR-FRET)-based analysis is becoming increasingly popular in biomedical research. The nanobody-based TR-FRET method allows easy quantification of fluorescent (fusion) proteins in lysates with much higher sensitivity than conventional fluorescence intensity readouts [70]. The bacterial surface display is a promising technology for producing cell-anchored proteins and designing whole-cell catalysts. Although various outer membrane proteins are used for surface display, no simple, universal, and high-throughput compatible methods are available to evaluate and develop surface display systems. In addition, it is challenging to distinguish between intracellular and surface-displayed proteins. Wendel et al. constructed a fluorescence-based surface display detection system by fusing GFP-nanobody to outer membrane anchors [71]. Two commonly used outer membrane proteins were chosen as anchors: outer membrane protein A (OmpA) and autotransporter (C-IgAP). Hence, two different display modules are constructed by fusing OmpA or C-IgAP with GFP-specific nanobodies, visualized by adding purified GFP externally. Although GFP itself can be displayed on the cell surface, this new method avoids the problem of false positives since only if GFP binds to the nanobody presented on the cell surface can the cells produce a fluorescent signal. The assay is compatible with many fluorescence detection methods, including whole-cell fluorescence detection in plate, in-gel fluorescence, microscopy, and flow cytometry. This inexpensive and easy-to-read surface display method will help to demonstrate the transport mechanism of proteins onto the surface of living cells, enabling the rational development of bacterial surface display systems and robust whole-cell biocatalysts in the future. The release of neurotransmitters requires exocytosis, endocytosis, and the formation of new fusion vesicles. What happens to vesicle proteins after exocytosis, when left on the plasma membrane, is poorly understood. These proteins are frequently conjugated to pH-sensitive GFP moieties (pHluorins). As pHluorin imaging is usually limited by the diffraction of spots several times larger than vesicles, using anti-GFP nanobodies to selectively label exocytosed vesicles is valuable [72]. By linking anti-GFP nanobodies to chemical fluorophores suitable for super-resolution imaging, the size and intensity of pHluorin-labeled proteins under various conditions can be detected in ways not possible with pHluorin alone. Upon stimulation of exocytosis, new vesicle proteins are exposed to the plasma membrane, and then the fluorescently labeled nanobodies will bind with pHluorin (Figure 5B). Thus, nanobody-based pHluorin detection is a promising tool for studying post-exocytosis events in neurons. Compared to gene editing and RNA interference, direct manipulation of biomolecules at the protein level is a more efficient route for protein function studies, overcoming limitations such as potential off-target effects, gene inactivation, and loss of essential gene phenotypic function. Targeted protein degradation is currently a major research strategy to achieve the loss of function and proteolysis of proteins of interest. A nanobody-based protein degrader can help achieve rapid degradation and reversible regulation of proteins of interest. Caussinus et al. developed a protein degradation method for the direct and rapid depletion of target GFP fusion proteins in any eukaryotic system [73]. Briefly, to knock down the GFP fusion protein, an anti-GFP nanobody is fused to the F-box protein (FBP) in the SKP1-CUL1-F-box protein (SCF) E3 ligase complexes to recognize the GFP fusion protein. The ubiquitin-conjugating enzyme (E2) covalently links multiple ubiquitin molecules to the target GFP fusion protein. Subsequently, the SCF complex degrades the polyubiquitinated protein; thus, removing the target protein is easy to monitor (Figure 5C). This technique, termed degrade green fluorescent protein (deGradFP), has been used for the degradation of GFP and its fusions in mammalian cells, zebrafish embryos [74], and plants [75]. In 2016, Croucher et al. developed a bimolecular complementation affinity purification system (BiCAP) which can effectively distinguish human epidermal growth factor receptor (HER) dimer (homologous and heterologous) from monomer [76]. In this system, two complementary fragments of an FP molecule are fused with two HER proteins (HER1, HER2, or HER3). These two fragments cannot be spontaneously assembled into active fluorescent proteins. However, suppose the two HER proteins interact with each other. In that case, the two fragments will be spatially close to each other and complementary, reconstructing them into a complete and active fluorescent protein (Figure 5D). The HER dimer and its interacting proteins can be enriched using a GFP nanobody, which had no affinity for the GFP fragments, thus enabling the isolation and enrichment of the dimer. By analyzing the identified proteins, it was revealed that the three dimers (HER2:HER2, HER2:HER3, and HER1:HER2) have common interacting proteins and their specific interactome. FAM59A, a protein that specifically interacted with HER1:HER3 dimer, mediated the activation of the extracellular signal-regulated kinase pathway, providing a new target for breast cancer therapy. Early in vitro diagnosis of disease and monitoring of therapeutic effects is a major problem in disease diagnosis. An ideal contrast agent should have good tissue penetration, high antigen affinity, and rapid clearance capability with minimal damage to normal tissues. Nanobodies have the potential as ideal imaging agents that can cross blood vessels and enter tissues for better imaging and therapeutic applications. FPs not only illuminate cells and biological processes but also make excellent scaffolds due to their apparent lack of linkage to numerous host protein networks. Using GBPs and GFP as a scaffold to drive biologically active complex formation, Tang et al. created a library of hybrid transcription factors that exclusively control gene expression in the presence of GFP and its derivatives [77]. The production of GFP controls the expression of cell-specific genes (Figure 5E) and promotes the dysfunction of the mouse retina and brain. In addition, the GFP transgenic mice and zebrafish strains are modified to achieve GFP-dependent transcription for the photogenetic monitoring of neural circuits. This work establishes the position of GFP as a versatile scaffold and opens the door to selectively manipulating various GFP-tagged cells in transgenic strains. Based on this, other intracellular products can also be developed as cell-specific scaffolds in multicellular organisms. DNA nanostructures have become an essential and effective tool for studying enzyme activity and protein function. However, developing universal strategies for forming protein complexes on DNA nanostructures is difficult. One of the difficulties is the attachment of proteins of interest to DNA nanostructures. Sommese et al. proposed a novel approach to labeling DNA nanostructures [78]. By functionalizing them with a GFP nanobody, the ability of protein attachment can be precisely controlled (Figure 5F). Compared with GFP-specific DNA aptamers, nanobodies exhibit higher specificity, stability, and affinity toward GFP. Therefore, the application of DNA nanostructures as a programmable scaffold in biological research has been dramatically simplified by connecting the DNA nanostructures with FPs commonly found in cells, developmental biology, and protein biochemistry. Compared with traditional antibodies, nanobodies have better physical and chemical properties and are easier to express and screen. Nanobodies are much simpler in structure than conventional antibodies. They are encoded by a single gene and can be easily produced by microorganisms, significantly reducing the production cost. Although nanobodies provide a breakthrough for antibody research, some problems still need to be solved. On the one hand, they possess the disadvantage of inconvenient operation and high cost of animal immunization. Furthermore, as HCAb and IgNAR are abundant in peripheral blood monocytes, nanobodies targeting FP antigens are usually screened from specific Camelidae or shark immune libraries. With the maturation of library construction and nanobody development technologies, screening FP-specific nanobodies from synthetic libraries is a new direction which will overcome the inconvenience of immunizing animals, shorten the experimental period, and significantly reduce the cost. On the other hand, although the screening and expression of nanobodies is relatively simple, obtaining FP nanobodies with potential application values is a very complicated process. The screening of phage nanobody libraries cannot avoid the false positives caused by the binding of filamentous phage surface with antigen, while the small capacity of yeast and bacterial nanobody libraries is another difficult problem in nanobody screening. The applications of nanobodies against FP will continue to increase in the future. (1) With the development of intracellular antibody technology, detecting intracellular molecules in living cells has become accessible. Nanobodies have unique advantages in this respect, in that (i) the nanobody is small, and the efficiency of cell entry is much higher than that of traditional antibodies; and (ii) the nanobody can be expressed and functional in living cells. (2) Bispecific nanobody is a kind of artificially modified antibody that can specifically bind two different antigens simultaneously. It can be easily constructed from monovalent nanobodies and expressed in microorganisms. Hence, nanobodies against FP provide a broad application prospect for bispecific nanobody development. This paper reviews the origin, structure, and properties of FPs. Monoclonal antibodies and nanobodies targeting FPs and their applications are also summarized. Although antibodies are valuable tools for displaying biological components in immobilized cells, the use of traditional antibodies in living cells is constrained by the ineffective folding and assembly of their variable heavy and light chains. Direct microinjection of antibodies is the primary method used in antibody intracellular applications, which is technically challenging and stressful to cells. Camel or shark-derived single-domain antibodies recognize antigens through their variable domains of heavy chains. These small and stable nanobodies can be expressed and functional in living cells. Therefore, generating nanobodies targeting FPs will make FPs more valuable in biological research.

PMC10002329

Ligang Cao,Di Wu,Lin Qin,Daopeng Tan,Qingjie Fan,Xiaohuan Jia,Mengting Yang,Tingting Zhou,Chengcheng Feng,Yanliu Lu,Yuqi He

Single-Cell RNA Transcriptome Profiling of Liver Cells of Short-Term Alcoholic Liver Injury in Mice

22-02-2023

short-term alcoholic liver disease,scRNA-seq,hepatocytes,endothelial cells,Kupffer cells,transcription factor

Alcoholic liver disease (ALD) is currently considered a global healthcare problem with limited pharmacological treatment options. There are abundant cell types in the liver, such as hepatocytes, endothelial cells, Kupffer cells and so on, but little is known about which kind of liver cells play the most important role in the process of ALD. To obtain a cellular resolution of alcoholic liver injury pathogenesis, 51,619 liver single-cell transcriptomes (scRNA-seq) with different alcohol consumption durations were investigated, 12 liver cell types were identified, and the cellular and molecular mechanisms of the alcoholic liver injury were revealed. We found that more aberrantly differential expressed genes (DEGs) were present in hepatocytes, endothelial cells, and Kupffer cells than in other cell types in alcoholic treatment mice. Alcohol promoted the pathological processes of liver injury; the specific mechanisms involved: lipid metabolism, oxidative stress, hypoxia, complementation and anticoagulation, and hepatocyte energy metabolism on hepatocytes; NO production, immune regulation, epithelial and cell migration on endothelial cells; antigen presentation and energy metabolism on Kupffer cells, based on the GO analysis. In addition, our results showed that some transcription factors (TFs) are activated in alcohol-treated mice. In conclusion, our study improves the understanding of liver cell heterogeneity in alcohol-fed mice at the single-cell level. It has potential value for understanding key molecular mechanisms and improving current prevention and treatment strategies for short-term alcoholic liver injury.

Single-Cell RNA Transcriptome Profiling of Liver Cells of Short-Term Alcoholic Liver Injury in Mice Alcoholic liver disease (ALD) is currently considered a global healthcare problem with limited pharmacological treatment options. There are abundant cell types in the liver, such as hepatocytes, endothelial cells, Kupffer cells and so on, but little is known about which kind of liver cells play the most important role in the process of ALD. To obtain a cellular resolution of alcoholic liver injury pathogenesis, 51,619 liver single-cell transcriptomes (scRNA-seq) with different alcohol consumption durations were investigated, 12 liver cell types were identified, and the cellular and molecular mechanisms of the alcoholic liver injury were revealed. We found that more aberrantly differential expressed genes (DEGs) were present in hepatocytes, endothelial cells, and Kupffer cells than in other cell types in alcoholic treatment mice. Alcohol promoted the pathological processes of liver injury; the specific mechanisms involved: lipid metabolism, oxidative stress, hypoxia, complementation and anticoagulation, and hepatocyte energy metabolism on hepatocytes; NO production, immune regulation, epithelial and cell migration on endothelial cells; antigen presentation and energy metabolism on Kupffer cells, based on the GO analysis. In addition, our results showed that some transcription factors (TFs) are activated in alcohol-treated mice. In conclusion, our study improves the understanding of liver cell heterogeneity in alcohol-fed mice at the single-cell level. It has potential value for understanding key molecular mechanisms and improving current prevention and treatment strategies for short-term alcoholic liver injury. Liver is a combination of liver parenchyma cells and non-parenchymal cells (NPCs). Hepatocytes constitute the liver parenchyma and approximately account for 78% of liver volume [1]. NPCs include liver sinusoidal endothelial cells, macrophages, B cells, T cells, hepatic stellate cells, Kupffer cells, and bile duct epithelial cells [2,3]. All liver cells play an important role in maintaining the liver’s physiological homeostasis. When the cell function is impaired or its composition becomes abnormal, this can cause diseases such as fatty liver disease, cirrhosis and hepatocellular carcinoma (HCC). Alcoholic liver disease (ALD) is a disease caused by the decline and failure of a variety of liver functions due to short-term or long-term alcohol intake, including alcoholic fatty liver (AFL), alcoholic hepatitis (AH), alcoholic cirrhosis and liver cancer [4]. ALD is the leading cause of death with alcohol consumption, and more than 50% of deaths related to liver cirrhosis in the world can be attributed to alcohol [5]. A large number of studies has demonstrated that alcoholic liver injury involves a variety of biological processes, including changes in alcohol metabolic enzymes, liver steatosis, inhibition of AMPK signaling pathway, oxidant/antioxidant imbalance, hepatocyte hypoxia, NF-κB and TLR4 signaling pathway activation, hepatocyte apoptosis, and activation of hepatic stellate cells [6,7,8]. However, due to the liver tissue containing heterogeneous cell mixtures, the pathogenesis of alcoholic liver injury in various liver cell types remains unclear. Bulk RNA sequencing and proteome of the liver tissues cannot distinguish the gene expression of different cell types and do not provide information about cell-cell interaction and microenvironment composition. Single-cell RNA-sequencing (scRNA-seq) has been developing rapidly and has been applied in many research fields in recent years, such as the construction of cell maps, as well as research on the development process of embryo and liver, and the pathogenesis of diseases [9,10,11,12]. Compared with bulk transcriptome sequencing, scRNA-seq also enables us to identify normal and pathogenic cell populations in the liver [13]. However, the change process of various cells during the progression of alcoholic liver injury is still unclear and there are few scRNA-seq studies on liver development in alcoholic liver injury mice. Therefore, comprehensive studies on cell type-specific composition and function are required to assess the detailed molecular mechanisms behind alcoholic liver injury and provide a comprehensive understanding of disease pathogenesis. To extend our understanding of subsets of each cell type involved in the development and progression of short-term alcoholic liver injury and cell subsets primarily affected by alcohol, we performed short-term pathological models of alcoholic liver injury. ScRNA-seq was used to analyze mice’s hepatocytes and NPCs at four critical stages of liver development in short-term alcoholic liver injury. Our analysis reveals hepatocyte and NPCs gene expression landscapes in the liver during the pathogenesis of alcoholic liver injury, as well as gene regulation and transformation occurring at the onset. In addition, we identify several cell populations that respond most to alcohol and their aberrantly activated transcription factors during the progression of alcoholic liver injury. This study reveals the heterogeneity, complexity and gene expression changes of liver cells and provides novel insights into the fundamental biology and pathology of alcoholic liver injury. The study confirmed that short-term consumption of alcohol caused liver injury in mice. Excessive alcohol consumption for 1 day could increase serum levels of ALT, AST [14]. Liver steatosis could be induced by excessive alcohol consumption for 3 to 7 days [15]. Continuous excessive alcohol consumption for 14 days could cause liver cell edema and necrosis in mice [16]. To elucidate dynamic changes during alcoholic liver injury pathogenesis, we performed mouse models from four key time points of alcoholic liver injury development. The results showed that the serum ALT was significantly upregulated at AG1, AG3, AG7 and AG14; serum AST was significantly upregulated at AG3 and AG7; serum HDL-C levels were significantly downregulated at AG1, AG3, AG7 and AG14; serum LDL-C levels were significantly downregulated at AG1, AG3 and AG7; serum TC levels of mice were downregulated at AG1 and upregulated at AG3, AG7, and AG14 (Figure 1A). In addition, alcohol exposure significantly increased liver and stomach coefficients in alcohol-fed mice (Figure S1). H&E staining results of liver tissue (Figure 1B) show that distilled water (BG group) has no significant effect on the pathological characteristics of liver tissue. Compared to the BG group, the paraffin sections AG group showed obvious hepatocyte necrosis, edema and nucleus pyknosis. These results suggested that short-term administration of 53% alcohol (10 mL/kg) could cause liver damage in mice. To elucidate liver cell complexity, heterogeneity and their dynamic changes in the pathogenesis of alcoholic liver injury, scRNA-seq was performed on liver cells from healthy and alcoholic liver injury mice at different times. Each cell had on average 33,918 reads and the exon reads took up 70.94% of the total reads. After removing low-quality cells, 50,274 single-cell transcriptomes were reserved and analyzed, including 10,439 from BG group, 10,164 from AG1 group, 9459 from AG3 group, 9260 from AG7 group, and 10,952 from AG14 group (Table S1). Subsequently, the expression matrix of each cell was created and analyzed using the Seurat R package. t-SNE plot showed that the liver cells were evenly distributed in each group, no significant intergroup batch effect was observed among the 5 groups (Figure S2A). A total of 38 clusters were identified with a resolution of 0.8 (Figure S2B). All liver cells could be assigned to 12 major liver cell types based on the expression of marker genemarker gene expression and SingleR package (version 2.0.0) (Figure 2A,B). They were B cells (B, marked with Ms4a1), Cycling (marked with Birc5), dendritic cells (DCs, marked with Slglech), endothelial cells (Endo, marked with Kdr), granulocyte (Gran, marked with S100a9), hepatocytes (Hep, marked with Alb), hepatic stellate cells (HSCs, marked with Dcn), monocyte or monocyte-derived macrophages (Mo/MoMF, marked with Ccr2), natural killer cells (NK, marked with Klrb1c), plasma cell (Plasma, marked with Jchain), T cells (T, marked with Cd3d) and Kupffer cells (KCs, marked with Clec4f). The proportions of cells in each sample are shown in Figure 2D and Table S2. The top three differentially expressed genes for each identified cell type are listed in Figure 2C; enrichment analysis further confirmed the cell identity (Figure S3). The molecular and biochemical mechanisms of alcoholic liver injury pathogenesis and the exact triggers of disease progression are not completely understood. Many mechanisms have been postulated to be involved in the pathology of alcoholic liver injury, such as mitochondrial damage, oxidative stress, endoplasmic reticulum stress, inflammatory pathway activation and dysfunctional lipid metabolism [4,17,18]. A better understanding of these mechanisms and the role of different cell types in this process is essential for the prevention and treatment of alcoholic liver injury. Therefore, this study has collected 713 genes related to the pathogenesis of ALD and investigated their changes in different liver cell types. These ALD-associated genes were matched by all liver cell types to a scRNA-seq gene expression matrix, and DEGs (p < 0.01 vs. BG group) in each cell type were screened. We observed the expression of these DEGs in different cell types of the liver of alcoholic liver injury mice at different time points (Figure 3A). The heatmap showed that only a few genes in each cell type were continuously up-regulated or down-regulated with the prolongation of the alcohol infusion, and these genes may continue to play a role in the process of alcoholic liver injury. The R package cluster was used to screen and show the DEGs that vary continuously in different cells and these DEGs were mainly distributed in Hep, Endo and KCs (Figure 3B, Tables S3 and S4). In addition, we also performed a heatmap display of the DEGs (p < 0.01 vs. BG) of each cell type that was not collected (Figure 3C). The results were consistent with previous studies. Only some of the DEGs continued to change with the prolongation of alcohol and they were mainly present in Hep, KCs and Endo; there were 591 DEGs in Hep, 596 DEGs in KCs and 217 DEGs in Endo (Figure 3D). These results indicated that sustained alcohol stimulation predominantly affects hepatocyte, endothelial cell, and Kupffer cell gene expression in mouse liver. Hepatocytes are the predominant cell in the liver, comprising about 60% of liver cells, and play an important role in detoxification, lipid metabolism, protein metabolism and glycogenolysis [19]. In the present study, 7739 hepatocytes from the livers of healthy and alcoholic liver injury mice were analyzed, and the result showed that hepatocyte markers Alb, Apoa1, Apoa2 and Ass1 were enriched (Figure 4A). Hepatocytes were generally less proliferative cells. Most hepatocytes were assigned to the G1 phase, and G1 phase cells increased in AG7 and AG14 compared with BG (Figure 4B). These results indicated that alcohol could inhibit the proliferation of hepatocytes. Alcohol abuse causes an imbalance in the oxidant/antioxidant status of individuals and reduces their ability to regulate oxidative stress [20]. Alcohol induces Cyp2e1 to induce oxidative stress, while Cyp2a5 can be induced to inhibit alcohol-induced oxidative stress [21,22]. In this study, 5 oxidative stress-related genes were mainly expressed in hepatocytes (Figure S2A). Gpx1 were significantly down-regulated, while Cyp2a5, Cyp2e1, Mt2 and Sod were significantly up-regulated in the hepatocytes of alcohol-fed mice (Figure 4C). This suggested that alcohol reduced the detoxification capacity of hepatocytes, and enhanced oxidative stress. Iron overload in the liver can aggravate liver damage by promoting lipid peroxidation, oxidative stress and iron death [23]. Three iron-related genes were mainly expressed in hepatocytes (Figure S2B). The Trf and Tmprss6 were significantly up-regulated in hepatocytes while Hamp was significantly down-regulated in all cell types in alcoholic liver injury mice (Figure 3B and Figure 4D). Steatosis is an early manifestation of alcoholic liver injury and may increase the susceptibility of the liver to secondary injury. Lipidomic analysis also showed that alcohol could promote the accumulation of lipids in hepatocytes [8]. Many genes of fatty acid synthesis and metabolism were mainly expressed in hepatocytes in this study (Figure S2C,D). Fatty acid synthesis genes Scd1, Acsl1, and Acsl5 were significantly up-regulated in hepatocytes of alcohol-fed mice (Figure 4E). Fatty acid degradation genes Acad1, Acadm, Acat1, Acat3, Eci1, Gcdh, Hadn, and Hadhhb were significantly down-regulated, but Cpt1a, Cyp4a10, and Cypa14 were significantly up-regulated in hepatocytes of alcohol-fed mice (Figure 4F). In addition, some genes of cholesterol metabolism were mainly expressed in hepatocytes (Figure S2E). Acaa2, Angptl3, and Apoc1 were significantly down-regulated, while Angptl4, Apoa1, Apoa4, Apob, Apoe, Apoh, Cyp27a1, Cyp39a1, Cyp4a31, Cyp8b1, Ehhadh, and Lcat were significantly up-regulated compared with the BG group in hepatocytes (Figure 4G). These results indicated that alcohol could increase fat synthesis, decrease degradation and disorder cholesterol metabolism in hepatocytes, resulting in the accumulation of fat in hepatocytes causing alcoholic fatty liver. The complement system is an important part of the innate immune defense, and activation of complement through classical and alternative pathways was detected in the livers of patients with alcohol-associated hepatitis [24]. In our study, it was found that complement and coagulation genes were mainly expressed in hepatocytes (Figure S2F), and C1s1, C3, C4b, C4bp, Cfh, Cfi, Cfhr2, F12, F2, F5, F7, Fga, Fgb, Fgg, Hc, Kng1, Kng2, Plg, Serpinc1, Serpingl and Vtm were significantly upregulated in hepatocytes of alcoholic liver injury mice (Figure 4H). Hence, the complement system and anticoagulant system in hepatocytes might be activated during alcoholic liver injury. In addition, we annotated the GO function of the DEGs that continuously changed with ethanol in hepatocytes and analyzed their biological processes (Figure 4I). The 190 up-regulated DEGs were mainly enriched in the regulation of lipid metabolism, alcohol metabolism, acylglycerol metabolism, triglyceride metabolism and fatty acid biosynthesis. The 401 downregulated DEGs were mainly enriched in aerobic respiration, oxidative phosphorylation, cellular respiration, ATP metabolic, energy generation, respiratory electron transport chain, and mitochondrial ATP synthesis coupled electron transport. These results suggest that short-term alcohol injury to mouse hepatocytes mainly involves lipid metabolism, oxidative stress, iron overload, complement and coagulation, and energy metabolism. SCENIC analysis was performed to assess changes in transcription factors (TFs) in alcoholic liver injury mice. In this way, we predicted specific TFs in alcoholic liver injury hepatocytes (Figure 4J). SCENIC analysis of hepatocytes revealed that some TFs were significantly activated in hepatocytes from alcohol-fed mice, including Xbp1, Stat3, Rxra, Nfic, Nfia, Hif, Hnf4a, Sf1, Nfat5. However, the activity of Nr1i2 and Ppara decreased after 1-day of alcohol and then increased with continued stimulation by alcohol. Functional enrichment of TFs target genes revealed that Xbp1 regulates the unfolded protein response (UPR) associated with endoplasmic reticulum stress [25]. Stat3 and Rxra are involved in the acute phase response and coagulation. Nfic, Nfia, Nr1i2, Hlf, Hnf4a, Rora, and Ppara are involved in lipid synthesis/metabolism processes (Figure S5A). These results indicated that continuous alcohol consumption could significantly activate ER stress, acute phase response proteins, coagulation system and lipid synthesis/metabolism processes in hepatocytes. Liver endothelial cells, including sinusoidal endothelial cells (LSEC), vascular endothelial cells and lymphatic endothelial cells (LyECs), play a key role in liver homeostasis, regulating intrahepatic vascular pressure and immune cell function [26]. Traditional immunofluorescence, flow cytometry, isolation of endothelial cells for RNA-seq and other methods are still limited by antigens and immune reagents [27]. scRNA-seq can provide abundant cell markers and cell function profiles and has been used to reveal the region specificity and function of mouse and human liver endothelial cells [27,28]. We analyzed 8399 endothelial cells in total, which highly express the endothelial cell markers Kdr, Pecam1, Lyvel and Oit3 (Figure 5A). The cells of all groups were less proliferative, indicating that alcohol did not significantly affect the proliferation of hepatic endothelial cells (Figure 5B). Some evidence supports that LSEC injury is increased when Nos3 is inhibited [29]; down-regulation of Klf2 and Nos3 can reduce NO production and lead to LSEC dysfunction [26]. We found that Klf2 was down-regulated at AG1 and AG3 and Nos3 were downregulated at AG14 compared to the BG group (Figure 5C). In addition, we enriched Go functions of DEGs that continuously changed with alcohol in liver endothelial cells and analyzed their biological processes (Figure 5D). The 79 up-regulated DEGs were enriched in blood pressure regulation, active regulation of inflammatory response, regulation of leukocyte differentiation, regulation of T cell activation, gliogenesis, hematopoiesis regulation and other functions. The 138 down-regulated DEGs were enriched in amebic cell migration, epithelial cell migration, epithelial migration, tissue migration, endothelial cell migration and other functions. These results indicated that the effects of alcohol on mouse endothelial cells involve the reduction of NO production, blood pressure regulation, inflammatory reaction, and epithelial cell migration. SCENIC analysis showed the changes of TFs in endothelial cells (Figure 5E). The activities of TFs Bcl3 and Klf6 were significantly enhanced at AG1 and subsequently returned to normal levels, but Nfe2l1 decreased at AG1 and recovered under continuous alcohol stimulation. TFs target genes enrichment analysis showed that Bcl3 was enriched in immune cell differentiation and cytokine pathways, while Klf6 was enriched in glucose synthesis or metabolism (Figure S5B). Kupffer cells are resident macrophages in the liver and can participate in the development of alcoholic liver injury by activating cytokines and chemokines [30]. Because the Kupffer cell is difficult to isolate from the human liver and has a complex developmental process, less is known about it. We detected 344 Kupffer cells, which highly expressed the Kupffer cells markers Clec4f, Timd4 and Vsig4 (Figure 6A). Cell cycle analysis showed that the proportion of G2M cells in Kupffer cells decreased at AG1, AG7 and AG14 (Figure 6B). In addition, the GO enrichment of persistently changing DEGs in Kupffer cells of the liver of alcoholic liver injury mice showed that 304 up-regulated genes were enriched in intracellular receptor signaling pathways, regulation of mRNA metabolic processes, mRNA processing, RNA splicing and translation regulation, while 292 down-regulated genes were enriched in the processes of ATP metabolism, antigen processing and presentation of exogenous antigens, aerobic respiration, cellular respiration, antigen processing and presentation (Figure 6C). The above results indicate that the effect of alcohol on Kupffer cells in mice involves processes such as antigen presentation and cellular energy metabolism. SCENIC analyzed the changes of TFs activity in Kupffer cells (Figure 6D). The activities of TFs Spi1, Spic and Elf4 on AG1 increased and decreased with continuous alcohol stimulation. The activities of Zmiz1, Z1b1, Gata4 and Sox18 decreased in AG and increased with the stimulation of alcohol. Functional enrichment of TFs target genes showed that Spic, Elf4, Sox18, Spi1 were involved in immune function, and Zeb1 and Gata4 were enriched in relation to cell migration (Figure S5C). Liver cells are mainly divided into hepatocytes and NPCs, and they maintain the microenvironment to keep homeostasis or break the balance under a pathologic environment [31]. The occurrence of ALD involves liver steatosis, hepatocyte necrosis and apoptosis, oxidative stress, immunity and inflammation [32]. Studying the changes in these processes in different cell types is necessary for the treatment and prevention of alcoholic liver injury. Liver transcriptome research based on scRNA-seq can obtain information on different liver cell types, which is conducive to understanding the changes of different liver cells. Therefore, we performed large-scale unbiased scRNA-seq to accurately and systematically profile mice livers with healthy and alcohol-induced liver injury. The large-scale dataset and deep analysis of scRNA-seq truly recognize the heterogeneity and complexity of the alcoholic liver injury progression. This will be beneficial to understand the alcoholic liver injury mechanism and identify new potential therapeutic targets. In this study, 12 major liver cell types were identified and the changes in gene expression in those cell types were investigated during the progression of alcoholic liver injury. Hepatocytes, endothelial cells and Kupffer cells showed more abnormal DEGs than other types of cells (Figure 3B,D, Table S3). We found that fatty acid synthesis and coagulation genes of alcoholic liver injury mice were significantly upregulated in hepatocytes (Figure 4C,H), which was consistent with the study of Michael Schonfeld [33]. Cpt1a is a rate-limiting enzyme of fatty acid β-oxidation (FAO); the change of its expression or activity will affect liver fat accumulation, and alcohol can reduce its activity and expression [34,35]. In our study, however, Cpt1a was upregulated in hepatocytes of alcoholic liver injury mice (Figure 4F). Cyp4a10 and Cyp4a14 are known to metabolize arachidonic acid and are significantly increased in ALD patients, promotes lipid accumulation and oxidative stress [36]. For this study, Cyp4a10 and Cypa14 were mainly expressed in hepatocytes and significantly increased in alcohol-fed mice (Figure 4F), indicating that liver damage caused by Cyp4a10 and Cypa14 mainly occurred in hepatocytes. In this study, the complement genes C1s1, C3, C4b, C4bp, Cfh, Cfi, Cfhr2, Hc and Serping1 were significantly increased in hepatocytes of alcohol-fed mice (Figure 4H), suggesting that hepatocytes can produce a large number of complements to establish inflammatory response and fight against alcohol-induced damage. In addition, the coagulation-related genes F12, F2, F5, F7, Fga, Fgb, Fgg, Kng1, Kng2, Plg, Serpinc1 and Vtn were significantly up-regulated in hepatocytes of alcohol-fed mice (Figure 4H), indicating that alcohol may enhance the coagulation process of mice. Studies have shown that alcohol can prolong the prothrombin time in male mice [33]. Ethanol is oxidized to acetaldehyde through hepatic alcohol dehydrogenase (ADH) and the microsomal ethanol oxidation system (MEOS), and the oxidation process is dependent on cytochrome P450 2E1 (CYP2E1) [37,38]. MEOS produces reactive oxygen species (ROS) through CYP2E1; this is significantly increased in acute or chronic alcoholic liver injury and facilitates liver injury [39]. Clinical studies have shown that consumption of 40 g ethanol per day for one week in humans leads to increased expression of CYP2E1 [40,41,42]. In addition, recent clinical studies have shown that the CYP2E1 inhibitor chlormethiazole reduces serum AST and ALT levels, and improves steatosis in patients with ALD [43]. In our study, Cyp2e1 and ADH were mainly expressed in hepatocytes (Figure S4A,G), and alcohol consumption for 3 and 14 days induced the up-regulation of Cyp2e1 expression (Figure 4C), indicating that short-term alcohol consumption induced liver injury through Cyp2e1 in hepatocytes. ADH is involved in the oxidative metabolism of ethanol to acetaldehyde, and chronic alcohol consumption leads to a decrease in ADH and further leading to liver injury [44,45]. ADH4 and AHD5 have higher Km (Michaelis—Menten constant) values for alcohol than 30 mM and 100 mM, respectively, while ADH1 has only 0.5–1.0 mM Km for alcohol [41]. In our study, Adh4 and Adh5 expression decreased after alcohol treatment, whereas Aldh1 expression increased at 7 and 14 days of alcohol consumption (Figure S4H), suggesting that alcohol decreased the level of ADH in hepatocytes, leading to alcohol accumulation and liver injury. Considering the activity of TFs, we analyzed the activity of TFs in normal and alcohol groups using SCENIC. Several TFs related to alcoholic liver injury were identified. Xbp1 is involved in ER stress and is increased in the liver of non-alcoholic steatohepatitis (NASH) patients, increasing fat accumulation and hepatic inflammation [46]. Xbp1 activity is increased in hepatocytes from mice with alcoholic liver injury (Figure 4J), but the role of alcoholic liver injury in humans requires further investigation. Studies have shown that increased expression of Stat3 in human alcoholic liver disease patients can further reduce alcoholic liver injury and inflammation [47,48,49]. These studies suggest that increased Stat3 activity in hepatocytes from alcoholic liver injury plays an essential role in counteracting alcohol-induced liver injury. Strikingly, transcription factors Spi1, Spic, and Elf4 in Kupffer cells showed significant increases in activity only at one day of alcohol consumption. Spi1 is a transcriptional activator that may specifically participate in the differentiation and activation of macrophages or B cells [50,51]. Spic inhibits inflammation and participates in macrophage development associated with iron homeostasis [52]. Elf4 maintains anti-inflammatory genes and inhibits anti-inflammatory gene expression and suppresses inflammatory responses [53]. This implies that Kupffer cells were involved in the hepatic immune reactions in the early stage of alcoholic liver injury (AG1). In addition, transcription factors Zeb1, Sox18, Gata4 and Zmiz1 were activated in AG14 samples, which target genes involved in ameboid-type cell migration, epithelial cell proliferation and response to transforming growth factor beta, implying that these transcription factors are activated in the later stages of alcoholic liver injury (AG14). Whether they are activated in response to longer alcohol stimulation still needs further investigation, however. In our study, 50,274 liver-single-cell transcriptome data were analyzed. Thanks to scRNA-seq studies, we analyzed and identified cell types that predominantly change during alcoholic liver injury and transcription factors with abnormal activity. This study revealed a small proportion of liver injury in mouse models following short-term ethanol application. However, fewer Kupffer cells and hepatic stellate cells were identified by sequencing the entire cells of the liver, and the key role of these two types of cells remains unclear and warrants further investigation. What is more, the long-term application of ethanol is not mentioned in this study, which will lead to more serious alcoholic liver diseases, such as AFL, AH and HCC. In summary, this study reveals liver heterogeneity, describes gene expression in liver cell types and provides a comprehensive single-cell transcriptional atlas in alcoholic liver injury. The results showed that continuous alcohol consumption mainly affected gene expression in hepatocytes, endothelial cells and Kupffer cells in mouse liver. In addition, we revealed the important transcription factors for alcoholic liver injury development. These findings help to understand the key molecular mechanisms of alcoholic liver injury pathogenesis and progression and also provide some directions for its prevention and treatment. Adult male C57BL/6 mice (23–25 g) were purchased from Hunan Slaughter Jindo Laboratory Animal Co Ltd. (license: SCXK (Xiang) 2019-0004). The mice were fed in an SPF (specific pathogen-free) class environment and under 12 h of light per day in a temperature-controlled environment (22 ± 1 °C, 60–70% humidity). Animals had free access to drink and food during the feeding period. We established alcoholic liver injury model by intragastric administration of 53% alcohol at different times. The experimental alcoholic liver injury model was established by comprehensive reference to other relevant studies and modification [14,16]. In this study, the alcohol used in the model group was 53% alcohol (prepared with absolute ethanol) at a dose volume of 10 mL/kg (equivalent to 5.3 g/kg alcohol), which is the concentration of commonly used commercial liquor, and We established a mouse model of alcoholic liver injury by gavage. The control group was given the same volume of distilled water (10 mL/kg) by continuous intragastric administration. In the study, mice were randomly grouped after one week of acclimatization feeding, with 15 mice in each group as follows: blank control group (BG, continuous gavage of water for 14 days), 1-day alcohol model group (AG1, gavage with water for 13 days and 53% alcohol for 1 day), 3 days alcohol model group (AG3, gavage with water for 11 days and 53% alcohol for 3 days), 7 days alcohol model group (AG7, gavage with water for 7 days and 53% alcohol for 7 days), 14 days alcohol model group (AG14, continuous gavage of 53% alcohol for 14 days). At the end time point of each group, mice were anesthetized with 20% urethane (0.075 mg/g) and blood was taken from the eyes. Blood was centrifuged at 4500× g rpm for 15 min at 4 °C to obtain serum after incubating at room temperature for 30 min. About 100 mg of the liver was fixed in 10% formaldehyde solution. All experiments in this study were approved by the Animal Experiment Ethics Committee of Zunyi Medical University. The mouse serum total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), aspartate transaminase (AST), and alanine transaminase (ALT) reagent kits were purchased from Nanjing Jiancheng Institute of Biological Engineering. Measurement of serum indicators according to the manufacturing instructions of the kits. HE staining (hematoxylin and eosin staining) was used to analyze liver tissue for histopathology. The fixed liver tissues were dehydrated in alcohol and xylene sequentially. The dehydrated liver samples were embedded in paraffin, sectioned in the sagittal plane in 6–8 µm sections, and stained with hematoxylin and eosin (H&E). The stained sections were sequentially dehydrated in 80%, 90%, and 100% alcohol, and the coverslips were sealed and observed under the microscope (Olympus BX43, Tokyo, Japan). At the end time point of each group, three mice in each group were randomly selected. Mice were anesthetized using urethane and perfused with a two-step liver perfusion protocol [54,55]. After perfusion, the cells were pushed through a sterile 40 µm filter, and separated into individual cells. Hepatocytes were spun and collected at 50 g for 1 min at 4 °C. The suspension was centrifuged at 350× g for 5 min at 4 °C to collect NPCs. Hepatocytes and NPCs were resuspended using DMEM complete medium. Hepatocytes were added to these NPCs to give a final hepatocyte concentration of approximately 10% of the total cell number. The viability of the mixed cells should be higher than 85%, and the cell concentration was adjusted to 1000 cells/μL [56]. Cells not immediately sequenced were stored frozen at −80 °C. The single-cell RNA-seq libraries were prepared with Chromium Next GEM Single Cell 3′ Reagent Kits v3. Briefly, the prepared single-cell suspension was combined with the barcoded mRNA capture beads, droplet generation oil and the mixture of enzymes. Then it was encapsulated in the “double cross” droplets of microfluid to form gel bead-in-Emulsions (GEMs). Cell lysis and reverse transcription reactions were performed in GEM. The GEMs were broken up and collected by the bead filter, and PCR amplification was performed using cDNA as the template. The quality of amplification products was checked (the size of amplification fragments and the output of amplification products). After the amplification products were qualified, the Chromium 3’v3 kit (10× Genomics) was used to construct the sequencing library. Finally, after the library was completed, we checked the database, and used Illumina HiSeq sequencing platform for sequencing to obtain the sequencing data and subsequent data analysis. The raw data were compared to the mouse reference genome (mm10-3.0.0) using Cell Ranger (v 3.1.0) provided by 10×Genomics. Next, cellranger was used to generate the UMI matrix. The Seurat (https://github.com/satijalab/Seurat/ (accessed on 14 March 2022)) R package was used for quality control, dimensionality reduction, and clustering analyses. The subset function screened and filtered low-quality and abnormally expressed cells. First, cells expressing less than 200 and over 6000 genes were excluded. Second, dead cells identified as cells with more than 25% reads coming from mitochondrial genes were excluded. Third, erythrocytes identified as cells with more than 1% of reads coming from mitochondrial genes were removed. Fourth, cells with a UMI number less than 500 and greater than 40,000 per cell were excluded from the analysis. After data cleaning, the resulting filtered UMI matrices were transformed into Seurat objects with the function CreateSeuratObject with the UMI matrix as counts, min.cells = 3, min.features = 300. Cell cycle was assessed with the R function CellCycleScoring, with s.features and g2m.features provided by Seurat in the R object cc.genes after being transformed into mice genes. The data were normalized, the highly variable genes were identified and scaled with the function SCTransform with parameters vars.to.regress = “percent.mt”, considering all the cells. All five Seurat objects were integrated using the CCA algorithm used by FindIntegrationAnchors and IntegrateData functions in Seurat. We performed a Principal Component Analysis (PCA) of the Seurat object with the R function RunPCA for all the cells. Euclidean distance K-nearest neighbors (KNN) were constructed using the FindNeighbor function to refine the boundary weights between cells and delineate cells with similar gene expression patterns, and then FindClusters function (resolution = 0.8) was used to identify cell subpopulations. An unsupervised clustering with a t-distributed stochastic neighbor embedding (t-SNE) analysis was performed on the transcriptomes using the R function RunTSNE with parameters dims = 1:20 (All cells). The cell type-specific genes of each cluster were identified with the FindAllMarkers function with parameters min.pct = 0.25, logfc.threshold = 0.25, test.use = ‘MAST’. Top-ranked genes were ordered by fold change under a threshold of expression of at least 25% of cells, fold change greater than 1.5-fold and adjusted p value < 0.01. Enrichment analysis was performed using the R packages ReactomePA on top-ranked genes [57]. Finally, according to cell type-specific genes, combined with marker genes from literature reports, SingleR and CellMarker databases, cell type identification was carried out to annotate cell subsets. DEGs analysis between alcoholic liver injury mice and control mice was performed using the function FindMarker in Seuart, using a MAST test [58]. Genes with a p value less than 0.05, expressed in at least 25% of cells were considered to be differentially expressed. Functional enrichment analysis of all DEGs was performed using the enrichGO function in the R package clusterProfiler [59]. We applied single-cell regulatory network inference and clustering (SCENIC) analysis to identify transcription factors (TFs) for some cell types (hepatocyte, endothelial cells and Kupffer cells) [60]. For each cell type, we performed SCENIC analysis for five groups, respectively. The regulon s and TF activity for each cell were calculated with motif collection version mc9nr (10 kb up and down, and 500 bp up and 100 bp down) from the cisTarget (https://resources.aertslab.org/cistarget/ (accessed on 19 October 2022)). Gene regulation of cells was constructed using the R package GENI, RcisTarget and AUCell. When focusing on the activity of TFs, transcription factors that activated in at least 70% of cells in at least one group and p value < 0.01 compared to the BG group were analyzed. All quantification data were expressed as mean ± SEM. Statistical analyses were performed utilizing R software (version 4.1.1). All parameters were analyzed by Student’s t-test and one-way ANOVA with R; a threshold of p < 0.05 was considered statistically significant.

PMC10002336

Han Zhang,Shujing Liu,Tianmeng Ren,Mengxue Niu,Xiao Liu,Chao Liu,Houling Wang,Weilun Yin,Xinli Xia

Crucial Abiotic Stress Regulatory Network of NF-Y Transcription Factor in Plants

23-02-2023

Nuclear Factor Y,abiotic stress,transcriptional regulation,functional mechanism

Nuclear Factor-Y (NF-Y), composed of three subunits NF-YA, NF-YB and NF-YC, exists in most of the eukaryotes and is relatively conservative in evolution. As compared to animals and fungi, the number of NF-Y subunits has significantly expanded in higher plants. The NF-Y complex regulates the expression of target genes by directly binding the promoter CCAAT box or by physical interaction and mediating the binding of a transcriptional activator or inhibitor. NF-Y plays an important role at various stages of plant growth and development, especially in response to stress, which attracted many researchers to explore. Herein, we have reviewed the structural characteristics and mechanism of function of NF-Y subunits, summarized the latest research on NF-Y involved in the response to abiotic stresses, including drought, salt, nutrient and temperature, and elaborated the critical role of NF-Y in these different abiotic stresses. Based on the summary above, we have prospected the potential research on NF-Y in response to plant abiotic stresses and discussed the difficulties that may be faced in order to provide a reference for the in-depth analysis of the function of NF-Y transcription factors and an in-depth study of plant responses to abiotic stress.

Crucial Abiotic Stress Regulatory Network of NF-Y Transcription Factor in Plants Nuclear Factor-Y (NF-Y), composed of three subunits NF-YA, NF-YB and NF-YC, exists in most of the eukaryotes and is relatively conservative in evolution. As compared to animals and fungi, the number of NF-Y subunits has significantly expanded in higher plants. The NF-Y complex regulates the expression of target genes by directly binding the promoter CCAAT box or by physical interaction and mediating the binding of a transcriptional activator or inhibitor. NF-Y plays an important role at various stages of plant growth and development, especially in response to stress, which attracted many researchers to explore. Herein, we have reviewed the structural characteristics and mechanism of function of NF-Y subunits, summarized the latest research on NF-Y involved in the response to abiotic stresses, including drought, salt, nutrient and temperature, and elaborated the critical role of NF-Y in these different abiotic stresses. Based on the summary above, we have prospected the potential research on NF-Y in response to plant abiotic stresses and discussed the difficulties that may be faced in order to provide a reference for the in-depth analysis of the function of NF-Y transcription factors and an in-depth study of plant responses to abiotic stress. Nuclear Factor-Y (NF-Y), known as the heme activator factor (HAP) or CCAAT-box binding factor (CBF), is a transcription factor (TF) ubiquitous in eukaryotes and can specifically bind to CCAAT cis-acting elements [1]. These proteins were named NF-Y following the suggestion of a reported study [2]. NF-Y is generally considered to be a heterotrimeric complex composed of NF-YA, NF-YB and NF-YC [3]. The intact NF-Y factor, assembled by NF-YA/B/C, mainly functions in the nucleus by directly binding to the CCAAT cis-acting elements in the promoter region of the target genes or by interacting with other factors to activate or inhibit the expression of the target genes [4]. Studies have shown that the animal NF-Y complex is essential in many life activities, such as cell proliferation and apoptosis, the occurrence of tumors and cancers, the stress response, growth and development [5]. NF-Y was first identified in Brassica napus [6], and has since been identified in other plants, including Arabidopsis thaliana [7], Musa nana L [8], Hordeum vulgare L [9] and Populus [10]. It has been revealed that NF-Y executes different biological functions by multitudinous studies recently, including flowering [11], fruit ripening [12] and response to adversity [13], in the processes of plant growth and development. The NF-Y complex regulates the target gene expression either by directly binding the CCAAT-containing promoters or by physically interacting and mediating the binding of other proteins that may be transcriptional activators or repressors [14]. A combination of research studies on the structural and functional characteristics manifests the powerful and mysterious regulating effect of the NF-Y gene family in many aspects of plant life. NF-Y is critical to the response to adversity stresses, including abiotic stress, high salt concentrations, drought, high temperature, cold, nutrient deficiency, hypoxia, ultraviolet radiation and heavy metal toxicity. However, these stresses have been reported to negatively affect plant growth, development or crop yield [15,16,17]. To this end, plants have evolved various effective stress-resistance capacities to counteract these stresses. It is noteworthy that some researchers have found that the expression of plant NF-Y is significantly affected in response to abiotic stress [18,19], whereas NF-Y should be one of the pivotal regulators in this process. In recent years, progress has been made on the relationship between plant NF-Y TFs and plant abiotic stress. It is urgent to comprehensively summarize and sort these new results for future studies on NF-Y. Therefore, this work reviews the research progress of the plant NF-Y gene family on its structural characteristics, mechanism of function and its involvement in the plant’s response to abiotic stress, proposing the directions of its function in regulating the abiotic stress response in plants (Figure 1). The Nuclear Factor-Y (NF-Y), also known as CCAAT box binding factor (CBF) or Heme Activator Protein (HAP), is a highly conserved trimeric transcription complex that is present in all eukaryotes [20]. Earlier research found that the NF-Y complex is composed of three subunits: NF-YA (also termed CBF-B and HAP2), NF-YB (CBF-A and HAP3) and NF-YC (CBF-C and HAP5). This complex can bind to the CCAAT element on the promoter of the target gene to regulate gene expression [21]. The classical NF-Y complex can bind to a promoter containing CCAAT sequences to form specific DNA–NF-Y complexes. Any mutation in the CCAAT sequence affects the formation of the DNA–NF-Y complex, suggesting that the binding of NF-Y to DNA requires the specific CCAAT sequence [1]. However, not all CCAAT elements can be recognized and bound by the NF-Y complex, suggesting that the nucleotide sequence and chromatin structure on both sides of CCAAT elements may affect the conformation of the NF-Y complex, and thus affect the binding ability of NF-Y complex [22]. Although NF-YA, NF-YB and NF-YC are all required for binding to the CCAAT box, it is NF-YA that binds directly to the CCAAT box. NF-YB and NF-YC subunit possess histone-fold motifs that allow them to form a heterodimer. After the heterodimer is formed, NF-YB/NF-YC is transported from the cytoplasm to the nucleus where NF-YA is recruited. NF-YA must combine with the NF-YB/NF-YC heterodimer to form a triplex in order to bind to the CCAAT box, and then their complex further regulates the transcription of downstream genes (Figure 2A). NF-YB/NF-YC heterodimer binds to the DNA glycophosphate skeleton, which may play a role in stabilizing the protein complex and recognizing specific DNA binding sites [23,24]. However, not all NF-Y regulates downstream gene expression in the form of the NF-YA/NF-YB/NF-YC triplex. In Arabidopsis, flowering time regulator CONSTANS (CO/B-BOX PROTEIN 1 BBX1) can bind to NF-YB2/NF-YC3 to form a new triplex structure and bind to flowering regulatory elements. CONSTANS contains a CCT domain similar to that of NF-YA but still requires the participation of the NF-YB/NF-YC subunit containing histone folding properties; a triplex comprised of NF-YB2, NF-YC3 and CO binds to the promoter of flowering-promoting gene FLOWERING LOCUS T, and the core binding site is the CCACA elements [25,26,27,28]. In rice, NF-YB1 binds with NF-YC12 and bHLH144 in order to form a heterotrimer, the trimer complex protects NF-YB1 from the ubiquitin/26S proteasome-mediated degradation. NF-YB1 activates the expression Wx, a key granule-bound starch synthase gene, by directly binding to the G-box elements in its promoter and, therefore, regulating starch synthesis [29] (Figure 2B). The abscisic acid (ABA) signaling pathway is essential to plants in the response to drought stress [30,31]. In the past decade, one of the most important advances in plant drought response has been the identification of the ABA receptor and the elucidation of ABA signaling pathways [32,33,34,35]. Recent works have shown that NF-Ys play key roles in regulating ABA signaling pathways; NF-Ys can be involved in regulating ABA synthesis or response to ABA signals, which is known as the ABA-dependent response to drought stress [36,37,38]. A root specific transcription factor PdNF-YB21 was isolated from Populus, which could directly interact with transcription factor PdFUS3, and PdFUS3 could directly activate PdNCED3, a key gene for ABA synthesis, resulting in a significant increase in root ABA content. Furthermore, ABA enhanced auxin transport in roots, which finally increased root growth and drought resistance. The results showed that NF-YB21 enhanced the growth and development of poplar under drought by promoting ABA synthesis in roots [16] (Figure 2C). ABA-induced drought responsive transcription factor PdNF-YB7 was isolated from a fast-growing poplar clone NE-19. The overexpression lines showed a decrease in water loss and an increase in instantaneous leaf water use efficiency (WUE) and leaf water potential; these phenotypes lead to enhanced drought resistance [39]. In well-watered, ABA (100 µM) and dehydration treated vermiculite-grown nine-day-old chickpea seedlings for 2 and 5 h, RNA-seq and RT-qPCR revealed 12 of 18 CaNF-Y genes in chickpea in response to ABA in the leaf and root, suggesting that the function of these genes in drought stress may be ABA-dependent [40]. NF-Y regulated plant drought tolerance by regulating the expression of ABA receptor gene PYR1. GmNF-YC14, which formed a heterotrimer with GmNF-YA16 and GmNF-YB2, activated the GmPYR1-mediated ABA signal transduction pathway, thereby regulating the drought response in soybean, which was verified by gene knockout and gene overexpression techniques [41]. NF-Y interacted with ABF, a key transcription factor in the ABA pathway. In Arabidopsis thaliana, ABF3 and ABF4 could interact with NF-YC3/YC4/YC9. SOC1 expression in nf-yc3/yc4/yc9 mutants was significantly reduced by ABA. The response of nf-yc3/yc4/yc9 to drought was not obvious. Under drought stress, SOC1 transcription was induced in order to promote flowering under the co-regulation of ABF3/ABF4/NF-Ys, which responded to adversity by shortening the plant’s life span under drought stress [19]. NF-Y enhances plant tolerance to drought by not only regulating the ABA-dependent pathway but also non-ABA dependent activity, mainly through improving plant photosynthetic efficiency, increasing the antioxidant enzymes activity and reducing the content of active hydrogen peroxide [42,43,44,45,46]. Overexpressing StNF-YC9 in potatoes increased the root length and photosynthetic rate and decreased the water loss rate under short-term drought stress. Under long-term drought stress, the malondialdehyde content decreased, while proline accumulation and the activity of antioxidant enzyme, including superoxide dismutase, catalase and peroxidase, increased. StNF-YC9 reduced the accumulation of malondialdehyde in potato and played an important role in drought resistance [47]. Two-year-old sweet oranges (Citrus sinensis) were used in the drought stress experiment. The experiment was carried out under greenhouse conditions with the plants grown in plastic pots of 45 L, and CsNF-YA5 was discovered to play different roles in the leaf and root under water stress. Overexpressing CsNF-YA5 in tobacco significantly reduced the production of H2O2 under water stress and increased the photosynthetic rate under normal conditions and drought stress. The biochemical and physiological responses of overexpression lines under drought stress maintained the growth advantage in an environment where soil was water deficient [48]. NF-Y family members TaNF-YB2 and TaNF-YC7 were identified in wheat, which could combine with TaNF-YA7-5B to form heterotrimers. The expression of TaNF-YA7-5B was induced by drought, and the oxTaNF-YA7-5B plant could grow and develop normally under dehydration conditions. This was mainly because TaNF-YA7-5B regulated stomatal closure, promoted leaf water retention and maintained cell ROS (reactive oxygen species) homeostasis under drought conditions. The expression levels of key regulatory genes TaCAT1 and TaPOD4 were positively correlated with the expression levels of TaNF-YA7-5B under drought stress, and it was confirmed in previous studies that TaCAT1 and TaPOD4 were involved in proline accumulation and ROS clearance. TaNF-YA7-5B was an important regulator of drought adaptation in plants, which was independent of ABA [49]. The function of NF-Y in coping with drought stress is expected to be applied in production. In previous studies, scientists confirmed that NF-Y increased the yield of crops, such as wheat, soybeans, corn and rice, in areas suffering from drought or chronic water scarcity [11,29,41,50] and increased biomass accumulation in forestry, such as poplar, apple, spruce and other tree species [16,51,52]. Although NF-Y has been studied for more than 30 years, its function related to drought tolerance has only been discovered in the last 15 years, which may have evolved through the diversification of the gene family encoding the NF-YB subunit. Due to the complexity of plant traits and many influencing factors under drought conditions, NF-Y overexpressed plants show enhanced drought tolerance based on a number of stress-related parameters, including chlorophyll content, stomatal conductance, leaf temperature, reduced wilt and photosynthetic maintenance. Adaptation to these stresses contributes to plant growth advantages in water-deficient environments [53]. Salt stress is one of the major adversities for plants. Plants undergo osmotic stress growing in saline–alkali soil [17,54]. Salt stress leads to stunted growth and developmental defects in the plant, including shortened height, obstructed seed germination, hindered reproduction [55,56] and even the demise of the plant [57,58,59]. The function of the NF-Y transcription factor involved in the salt stress response has been discovered through RNA-seq analysis in plants treated with salt [60,61]. The adventitious roots of Jilin ginseng were treated with different concentrations of salt in B5 medium (0, 70, 80, 90 and 100 mM NaCl), and the treated adventitious roots were incubated under dark conditions at 22 °C for 30 days. The expression pattern of NF-Y in Panax ginseng suggested that the PgNF-Y expression level was different not only among lines, but in temporality and space. The weighted gene co-expression network analysis (WGCNA) indicated that the PgNF-Ys function co-ordinately in ginseng. PgNF-YB9, PgNF-YC2 and PgNF-YC7 responded to salt stress in a synergistic manner [60]. The alfalfa grew for 10 days at 22 °C after treatment with 250 mM NaCl for 0 h, 0.5 h, 1 h, 3 h, 6 h, 12 h and 24 h; the root tips were sampled. The results of RNA-seq suggested that MsNF-YB2 responded to salt stress at early stages, while MsNF-YC5 responded to medium term salt stress. Other NF-Ys including MsNF-YB5, MsNF-YB7, MsNF-YB15 and MsNF-YC6 responded to salt stress as well. The WGCNA suggested that MsNF-YB2 co-expressed with DEAD at the early stages of salt stress, and there are ten genes co-expressed with MsNF-YC6 coding serine/threonine protein kinases interacting with CBL, indicating MsNF-YC6 might involve in calcium signalling pathway under salt stress [61]. The petunias growing under 25 °C with daily 14-hour light and 10-h darkness were treated with 500 mM NaCl; the samples were taken after 1 h, 3 h, 6 h and 12 h. The RNA-seq revealed that the PhNF-YA5/6/10 were quickly induced and were highly expressed in the roots. These genes, therefore, might regulate salt tolerance by promoting root development. The expression level of PhNF-YB3 is relatively higher in the leaf and flower under salt stress. PhNF-YB3 might play a part in regulating the flowering time and the salt stress response. 12 NF-Y genes, four for each NF-YA, NF-YB and NF-YC, were induced in the root and stem in 30-day-old barley treated with a Hogland solution containing 300 mM NaCl. NF-YC2 and NF-YC3 were strongly induced by salt stress and took part in the early plant reaction to stress exposure [9,62]. In recent years, the mechanism of NF-Ys in response to salt stress has been reported. NF-YC9 is located in the cytoplasm under stress-free conditions in poplar, while translocation to the nucleus occurred in response to salt or ABA by interacting with SRMT, a MYB transcriptional factor. NF-YC9 promoted the expression of SRMT regulated genes, enhancing the salt tolerance in poplar. It is worth noting that the expression level of NF-YC9 was not induced by salt stress in this process, while the NF-YC9 protein was translocated to the nucleus from the cytoplasm. This phenomenon provides a new way of thinking for the study of the function of NF-Ys [63] (Figure 2D). In maize, it was demonstrated by the technical means of EMSA and yeast that ZmNF-YA1 promotes the expression of multiple genes that play pivotal roles in response to salt stress and plant development, including ZmbHLH116, ZmPOD64, ZmLOX5 and ZmMBF1c, under salt stress. GmNF-YA is a nuclear factor discovered in soybean, which plays an important role in salt tolerance [64]. GmNF-YA interacts with GmFVE and weakens the histone deacetylation by reducing the relevance between GmFVE and GmHDA13. The maintenance of the acetylation level of GmH3K9 leads to the expression of salt-induced genes [65]. Moreover, NF-Y independently regulates the response to salt stress in wheat. TaNF-YA10 was identified in salt-tolerant wheat SR3, and the heterogeneous expression of TaNF-YA10 in Arabidopsis increased the sensitivity to salt. Further investigation unveiled the independent role TaNF-YA10 in response to salt stress [66]. Not all NF-Ys that responded to salt positively regulated the response to salt stress. The overexpression of AtNF-YA1 in Arabidopsis prevented the growth of sapling after seed germination under salt stress. Additionally, ABI3 and ABI5 is significantly up-regulated along with their target genes AtEM1 and AtEM6, which might restrict the growth of plant [43]. Drought stress along with osmotic stress commonly come with salt stress, thus the response to these two adversities shall be considered when searching for genes involved in salt tolerance. For instance, the ABA signalling pathway plays a key role in response to salt stress since under salt stress more tolerant rootstocks accumulate more ABA [67,68]. The functions of NF-Y family proteins in regulating the response to salt stress are diverse, complicated and cannot be fully understood at this current stage. Most of the study of the NF-Ys function stays in the omics level, and the mechanism of the NF-Y complex is yet to be studied, while most research focuses on the function of a single NF-Y protein. The function of the NF-Y complex is believed to be the priority in the research of the NF-Y family. Either insufficient nutrition or overnutrition will cause stress in plants [69,70]. Improving nutrient use efficiency is an important competitive strategy for plants to adapt to barren environments [71]. Studies have shown that NF-Y transcription factors are involved in regulating the absorption and distribution of plant nutrients, including nitrogen absorption, phosphorus absorption, carbon–nutrient balance and coping with low nitrogen and low phosphorus (LNLP) environments [72,73]. QQS (Qua-Quine Starch; At3g30720) regulates metabolic processes affecting carbon and nitrogen partitioning to proteins and carbohydrates. Studies have shown that the QQS protein interacts with the transcriptional regulator AtNF-YC4. Overexpression of AtNF-YC4 in Arabidopsis mimics the QQS-overexpression phenotype, increasing protein and decreasing starch levels; therefore, NF-Y can maintain the homeostasis of plant development by regulating the allocation of carbon and nitrogen in plants [74]. Qu B et al. performed a genome-wide sequence analysis of the A (NF-YA), B (NF-YB) and C (NF-YC) subunits of Nuclear Factor Y (NF-Y) in wheat (Triticum aestivum). It was found that most expressions of NF-YAs were positively responsive to low nitrogen and low phosphorus availability, and overexpressing TaNFYA-B1 significantly increased both nitrogen and phosphorus uptake and grain yield under differing nitrogen and phosphorus supply levels. The increased nitrogen and phosphorus uptake may have resulted from the fact that that overexpressing TaNFYA-B1 stimulated root development and up-regulated the expression of both nitrate and phosphate transporters in roots. Meanwhile, TaNFYA-B1 was negatively regulated by miR169. Thus, the adaptability of wheat to low nitrogen and phosphorus is enhanced [50] (Figure 2E). A total of 108 NF-Y family members were identified in B. napus and categorized into three subfamilies (38 NF-YA, 46 NF-YB and 24 NF-YC). It was found that BnaNF-Ys had different expression patterns under multiple nutrient starvations. Moreover, more BnaNF-YA genes were differentially expressed under nutrient limited environments compared to the BnaNF-YB and BnaNF-YC subfamilies. Among the five rapeseed tissues, 16 BnaNF-Ys genes responded diversely to N deprivation [72]. Transcriptome data and RT-qPCR analysis of poplar showed that the expression of NF-YA responded to different forms of nitrogen treatment, and the expression change in the root system was distinctive. NF-YA regulates the response of poplar roots to different forms of nitrogen, indicating that these genes regulate root growth and development [75]. The AtNF-YA family can be induced by nutrient stress, similarly nitrogen and phosphorus deficiencies strongly induce the expression of the five Arabidopsis NF-YA subfamily members; however, they showed a long-term expression window. It is important that, in its role as a negative regulator, it does so mainly by isolating the recognition of NF-YB/NF-YC heterodimers to other transcription factors that induce the expression of stress response genes [76]. In an analysis of the transcriptomic response to low concentrations of nitrate, the steady-state mRNA levels of NF-YA5 and other members of the NF-YA family were significantly increased [77]. After 5 weeks of Arabidopsis growth, Arabidopsis plants were supplied with N-free nutrient solutions to simulate N-deficiency conditions. Nutrient solutions were renewed daily to ensure pH stability. The expression levels of AtNF-YA2/3/5/8 were significantly up-regulated after 48 h of a nitrogen deficiency culture [78]. The balance of nutrients is essential in plants, especially in short life cycle plants. In field planting, the use of a large number of fertilizers for a long time has led to serious soil pollution, and in forestry production, the long-term nutrient deficiency of plants will lead to tree stuntedness. Therefore, it is particularly important to improve nutrient use efficiency in plants, especially nitrogen and phosphorus [79]. NF-Y TFs play a key role in plant response to nutrient stress, and its function is being continuously explored based on current research, which is expected to make breakthroughs in agroforestry breeding. Drought, cold and high salinity cause osmotic stress, which directly affects the development, growth and productivity of plants resulting in crop yield losses [80,81]. The NF-Y transcription factor has been reported to be involved in the regulation of osmotic stress. In peanuts, to analyze the expression pattern of AhNF-Y genes, two-week-old seedlings were treated with a nutrient solution containing 200 mM NaCl. The leaves and roots of seedlings treated with NaCl were harvested at 0 h, 4 h, 8 h, 12 h and 16 h. The results revealed that the transcript levels of AhNF-YA4 and AhNF-YA8 were down-regulated, and both reached the lowest levels under osmotic stress at approximately 8 h. In contrast, AhNF-YA11, AhNF-YC2 and AhNF-YC8 had similar expression profiles and showed a trend towards up-regulation. Under osmotic stress, the expression pattern of AhNF-YA4 and AhNF-YA11 were different. This difference may be due to the cis-elements in the promoter region [82]. Cultivated with 250 mM mannitol in Arabidopsis, overexpression of NF-YC9 results in osmotic stress hypersensitivity, but the down-regulation of NF-YC9 expression shows no effect on the osmotic response during post-germination growth. The overexpression of NF-YC9 enhances the sensitivity to ABA, salt and osmotic stresses during early seedling growth, though the knockdown mutants of NF-YC9 show wild-type ABA-related phenotypes, suggesting that NF-YC9 may positively regulate ABA signaling but likely with a functional redundancy. NF-YC9 interacts with and improves the activities of an ABA responsive bZIP transcription factor ABI5 and enhances expression of the ABI5 gene in response to ABA [37]. In addition, apple MsNF-YB21 was expressed in Arabidopsis Thaliana, and it was found that drought stress induced the expression of MsNF-YB21, which resulted in root elongation and further increased osmotic-stress tolerance in Arabidopsis thaliana. Physiological analysis of the oxMsNF-YB21 also showed an enhanced antioxidant system. These results provide useful information for further study of the relationship between NF-Ys and osmotic stress in apple [51]. Moreover, after treatment with ABA and exposure to osmotic stress, salt and H2O2, 27 PmNF-Y gene expression profiles were obtained by RT-PCR in Prunus mume. It was found that PmNF-YA1/2/4/5/6, PmNF-YB3/4/8/10/11/13 and PmNF-YC1/2/4/5/6/8 were responsive to ABA and osmotic stress [83]. There are few studies on NF-Ys function regarding osmotic stress, which may be due to the lack of markers on osmotic stress detection. Little is known about the signal perception and signal transduction pathways of osmotic stress in plant cells. The presence of osmotic stress is always accompanied by the presence of other types of stresses, and there are limited studies on the mechanism of osmotic stress. Higher plants maintain their growth and development by responding to abiotic stress in various physiological processes. NF-Y TFs are widely involved in the plant’s stress response as monomers, complexes or in combination with other transcription factors. In Arabidopsis thaliana, the involvement of the NF-Ys complex in the regulation of ER stress has been clearly elucidated and relies on the complex formed by membrane-associated basic domain/leucine zipper (bZIP) transcription factor and NF-Ys: bZIP28 binds to CAGG in the ER stress-response element I (ERSE-I) and then binds to NF-YB3, NY-YC2 and NF-YA4 to form complexes in order to promote the expression of ER stress-induced genes [84]. In studies of the NF-YC family in Arabidopsis thaliana and tobacco, AtNF-YC2 shows an up-regulated expression under oxidative stress conditions. In the study of the necrotizing cell death phenotype that integrates the mechanisms of photooxidative stress accumulation during light exposure, the results showed that reactive oxygen species generated during these photodynamic processes might induce the NF-YC2 expression [85]. AtNF-YA2 directly regulates the expression of AtHSFA3 and AtHSFA7b by binding to the promoters, and AtHSFA7b regulates the transcription of miRNA169. Moreover, SlyNF-YA9/A10 are the best orthologs of AtNF-YA2 in tomato as it mediates downstream reactions by the same mechanism as in Arabidopsis, suggesting that the SlyNF-YA9/10 and SlyHSFA7 have conserved functions in heat tolerance in both plant species [86]. In addition, the Arabidopsis NF-Y family is involved in the pathway of heat response stress regulation. In this research, a trimer consisting of NF-YA2, NF-YB3 and DPB3-1 activated the expression of heat stress-induced genes associated with DREB2A in protoplasts, and the identified trimer enhanced heat stress-induced gene expression under heat [87]. In the study of the NF-Y family genes in Sorghum bicolor L, NF-YA7 appeared to be associated with high temperature (40 °C) stress, while NF-YA8 was triggered by both cold (4 °C) and high-temperature stresses [88]. A highly precise and complex role of NF-YA in promoting temperature-induced flowering has been demonstrated, where high environmental-temperature-mediated down-regulation of miR169hn leads to the activation of AtNF-YA2, which increases the expression of the flowering genes FT and YUCCA2 (YUC2) [89]. In cold conditions, CBF/NF-Y (YZ9)-overexpressed fruits promoted the coloring in strawberries, suggesting that NF-Y family genes also play a role in cold stress in fleshy fruit [90]. The NF-Y family is also widely involved in vegetative stress. The NF-Y family in Brassica napus, which is hyper-sensitive to nitrogen (N) deprivation, was comprehensively identified and systematically characterized 38 NF-YAs, 46 NF-YBs and 24 NF-YCs [72]. Although the NF-Y TFs have been actively studied for decades, the last decade has seen a giant leap in our understanding of their function. The NF-Y family transcription factors have multiple roles in regulating plant growth, development and response to various environmental stresses (Table 1). The functions of the NF-Y family in response to abiotic stress have been extensively studied, mainly focusing on the ABA signaling pathway, particularly related to ABA-mediated drought and salt tolerance. In poplar, the mechanisms of some NF-Ys, including NF-YB21 and NF-YC9, in regulating drought and salt tolerance have been elucidated [16,63]. This is helpful for the application of NF-Ys in production practice, especially in molecular breeding, and has great reference significance. Due to the complexity of the NF-Y complex, the signaling pathways and regulatory models led by the NF-Y complex are still poorly understood. It is still necessary to further investigate the mechanisms of NY-Ys responses to abiotic stress in plant hormones (auxin, cytokinin, ethylene and brassinolide) dependent pathways and the role in other stress responses. It is generally believed that the complex formed with NF-YA is directly bound to the CCAAT box. However, recent studies have demonstarted that NF-Ys interact with other transcription factors. Since NF-YBs/YCs generally cannot bind directly to promoters and the target is determined by the transcription factors they interact with, more potential targets for NF-Ys should exist. Plants will have complex physiological changes in response to stress, and there are many genes involved in both sensing stress signals and responding to stress signals. However, the way in which NF-Ys sense and respond to stress signals is currently unclear. Most studies of NF-Ys have not reported how NF-Ys bind to promoters. Moreover, the majority of them are in response to drought, salt, temperature stress and nutrient stress. There are some general environmental factors, without specific regulatory mechanisms, which also lead to our lack of understanding of NF-Ys. The reason for this may be that NF-Ys have strong interactions among themselves, and NF-Ys interacting with other genes or TFs produce additional complexes, making these questions extremely complicated. There is an urgent need to decipher the role of NF-Ys in diverse physiological processes in plants and a theoretical basis is needed to address critical problems regarding plant production and resistance. Many studies have reported the existence of different abiotic stress responsive cis-elements in NF-Y promoters, and understanding each TF’s specific regulation of these cis-elements is critical for stress mitigation in plant improvement projects, which helps to achieve the agriculture and forestry sustainability and food security for the growing world population.

PMC10002337

Qiuxuan Wu,Qiong Wu,Xiaoxiang Wang,Xuesheng Zhang,Rui Zhang

Polychlorinated Diphenyl Ethers in the Environment: A Review and Future Perspectives

23-02-2023

PCDEs,persistent substances,bioaccumulation,biomagnification,environmental behavior,environmental fate

Polychlorinated diphenyl ethers (PCDEs) are a class of synthetic halogenated aromatic compounds, which have gradually attracted widespread attention due to potential environmental risks to humans and ecosystems. This paper presents a literature review of research on PCDEs using PubMed, Web of Science and Google Scholar as search engines/databases with no constraints on publishing year or number. A total of 98 publications on the sources, environmental levels, environmental behavior and fate, synthesis and analysis and toxicology of PCDEs were retrieved. Existing studies have shown that PCDEs widely exist in the environment with the ability of long-range transport, bioaccumulation and biomagnification, which are almost comparable to polychlorinated biphenyls. They can elicit adverse effects including hepatic oxidative stress, immunosuppression, endocrine disorders, growth retardation, malformations, reduced fertility and increased mortality in organisms, among which some seem to be related to the activation of the aryl hydrocarbon receptor. PCDEs can be metabolized into other organic pollutants, such as hydroxylated and methoxylated PCDEs and even polychlorinated dibenzo-p-dioxins and furans through biotransformation, photolysis and pyrolysis reactions in the environment. Compared with reviews on PCDEs published previously, some new information and findings are summarized in this review, such as new sources, current environmental exposure levels, main metabolism pathways in aquatic organisms, acute toxicity data for more species and relationships between structural parameters and toxicity and bioaccumulation potentials of PCDE congeners. Finally, current research deficiencies and future research perspectives are proposed to facilitate the assessment of health and ecological risks of PCDEs.

Polychlorinated Diphenyl Ethers in the Environment: A Review and Future Perspectives Polychlorinated diphenyl ethers (PCDEs) are a class of synthetic halogenated aromatic compounds, which have gradually attracted widespread attention due to potential environmental risks to humans and ecosystems. This paper presents a literature review of research on PCDEs using PubMed, Web of Science and Google Scholar as search engines/databases with no constraints on publishing year or number. A total of 98 publications on the sources, environmental levels, environmental behavior and fate, synthesis and analysis and toxicology of PCDEs were retrieved. Existing studies have shown that PCDEs widely exist in the environment with the ability of long-range transport, bioaccumulation and biomagnification, which are almost comparable to polychlorinated biphenyls. They can elicit adverse effects including hepatic oxidative stress, immunosuppression, endocrine disorders, growth retardation, malformations, reduced fertility and increased mortality in organisms, among which some seem to be related to the activation of the aryl hydrocarbon receptor. PCDEs can be metabolized into other organic pollutants, such as hydroxylated and methoxylated PCDEs and even polychlorinated dibenzo-p-dioxins and furans through biotransformation, photolysis and pyrolysis reactions in the environment. Compared with reviews on PCDEs published previously, some new information and findings are summarized in this review, such as new sources, current environmental exposure levels, main metabolism pathways in aquatic organisms, acute toxicity data for more species and relationships between structural parameters and toxicity and bioaccumulation potentials of PCDE congeners. Finally, current research deficiencies and future research perspectives are proposed to facilitate the assessment of health and ecological risks of PCDEs. Polychlorinated diphenyl ethers (PCDEs) are a class of synthetic halogenated aromatic compounds comprising 209 possible congeners, which are structurally similar to polychlorinated biphenyls (PCBs) and polychlorinated dibenzo-p-furans (PCDFs) [1]. However, PCDEs are typically more polar than PCBs due to the presence of the oxygen atom and the resultant asymmetry over the horizontal axis [2,3]. The theoretical 209 congeners can be divided into ten congener groups from mono- to deca-CDE and numbered (Table S1) according to the International Union of Pure and Applied Chemistry (IUPAC) system established for PCBs [3]. The structural formula of PCDEs is shown in Figure 1 and their molecular formula is C12H10–nClnO (n = 1–10). PCDEs were widely used as flame retardants, hydraulic fluids, electric insulators, lubricants and plasticizers in the 20th century [4,5]. Currently, congener CDE 13 is used directly as the intermediate in the synthesis of the fungicide difenoconazole [6]. In addition, PCDEs are by-products produced during the synthesis of commercial chlorophenols as important intermediates in the chemical industry [7,8]. PCDEs can also be generated in the incineration of municipal waste [9,10]. Therefore, PCDEs have inevitably leaked into the environment and have been detected in water, sediment, soil, atmosphere and various biological samples at total concentrations of 0.351–1800 ng/L, 0–3,980,000 ng/g dry weight (dw), <38–6800 ng/g dw, 8.75 × 10−3–1.15 × 1032 pg/m3 and 0–50,924 ng/g lipid weight (lw), respectively (Table 1 and Table 2). Since the negative log-transformed values of 298 K supercooled liquid vapor pressure (PL) of most PCDEs range from approximately 2 to 5, PCDEs may thus transport over long distances with the atmosphere [41]. For example, although there were no sources of pollution in the remote Arctic region, PCDEs were also detected in Arctic cod (Arctogadus glacialis) at concentrations of 2–21 ng/g lw [32,42]. Owing to the strong acid and alkali resistance and antioxidant capacity, PCDEs are persistent in various environmental matrices [1,17,18]. Moreover, the biological half-lives of tetra- to hepta-CDEs generally exceed 100 days in rainbow trout (Salmo gairdneri), which are almost equivalent to those of PCBs [43]. In addition, high lipophilicity renders PCDEs susceptible to accumulate in organisms and biomagnify through trophic transfer [44]. Toxicokinetic experiments showed that the absorption rate of PCDEs in fish were high at 2.4–48.9 μg/day, and the bioconcentration factor (BCF) could reach 1001–32,000 [44,45,46,47]. The bioaccumulation capacities of PCDEs are even higher than those of polychlorinated dibenzo-p-dioxins (PCDDs) and PCDFs in oligochaete worm (Lumbriculus variegatus) [48]. In some aquatic food chains, such as oligochaete worm (Lumbriculus variegatus) to white sucker (Catostomus commersoni), the biomagnification factors (BMFs) of PCDEs are 13.7 to 34.6 on a lipid-normalized basis, comparable to those of PCBs [11]. In addition, PCDEs have been detected in daily food and health products of humans [9,40]. Toxicology studies have shown that the toxic effects of PCDEs in organisms are similar to dioxins. During the early life stages of fish, PCDEs may cause embryonic vascular hemorrhage, growth inhibition, deformity and death [49,50]. PCDEs can also cause oxidative stress in the liver of mice (Mus musculus) and disturb the balance of trace elements [51]. Exposure of mice (Mus musculus) to PCDEs during pregnancy resulted in reproductive developmental toxicity, such as reduced survival of fetuses and pups, and disturbed thyroid hormone secretion in maternal and fetal mice [52,53]. In addition, there is evidence that PCDEs may induce immunotoxicity through the mediation of the aryl hydrocarbon receptor (AHR) [54,55,56]. However, toxicological data of PCDEs are still limited probably due to the paucity of commercially available standards of PCDE congeners. It further leads to insufficient attention to the health and ecological risks brought by PCDEs, along with slowly increasing research on their toxic mechanisms, environmental exposure levels and environmental behavior. In this context, a systematic literature search was performed with a query based on the keywords of “Polychlorinated diphenyl ethers”, “PCDEs” and “polyhalogenated diphenyl ether”. PubMed, Web of Science and Google Scholar were used as search engines/databases. Publishing year was not restricted to retrieve as much of the available literature as possible. The relevant publications on PCDEs, focusing on the sources, environmental level, environmental behavior and fate, synthesis and analysis methods and toxicological research (Figure 2) were screened based on the examination of title, abstract and full text. The results and data were manually extracted and cross-checked by two authors. Additional relevant studies were identified from the reference lists of already identified publications. Finally, a total of 98 relevant publications were obtained from the literature search. Compared with reviews on PCDEs published previously [3,7], some new information and findings were summarized in this review: new sources including solid waste incineration [22], intermediates [6] and impurities in drugs, daily necessities and pesticides [57]; current environmental exposure levels [12,18,21]; main metabolism pathways in different aquatic organisms [58]; acute toxicity data for more species and relationships between it and structural parameters [59]; and relationships between bioaccumulation potentials and the number/location of substituting Cl atoms of PCDE congeners [58]. Furthermore, current research deficiencies were further proposed, and future research perspectives were explored to facilitate the environmental chemistry and toxicology research on PCDEs in the future. Over the last several decades, there has been an accumulation of evidence that PCDEs primarily come from the production of chlorophenol preparations as by-products and impurities [8,9]. Commercial chlorophenols and their sodium and potassium salts were widely used as industrial wood preservatives, fungicides, insecticides, antifungal and antibacterial agents from the 1940s to 1980s [60]. At present, chlorophenols are still extensively used in chemical production as important intermediates, such as for the synthesis of clofibrate (CAS 637-07-0; a lipid-lowering drug), triclosan (CAS 3380-34-5; a broad-spectrum antimicrobial agent added in daily necessities, such as toothpaste), bifenox (CAS 42576-02-3; a nitrodiphenyl ether herbicide), 2-chlorophenyl N-methylcarbamate (CAS 3942-54-9; a carbamate insecticide) and triadimefon (CAS 43121-43-3; a triazole fungicide) [57]. Therefore, PCDEs may leak into the environment as impurities during the production, use, handling and disposal of chlorophenols and related products. It was found that the exposure levels of PCDEs in bearded gull (Chlidonias hybrida) eggs and night heron (Nycticorax nycticorax) eggs collected from the Yangtze River Delta were related to the use and discharge of pentachlorophenol and sodium pentachlorophenate [26]. In commercial chlorophenols and related products, the content of PCDEs ranges from 4.4 to 1000 mg/kg wet weight (ww) (Table 3), which is related to the synthesis processes and the ratios of reactants. Moreover, the annual production of global chlorophenols was estimated to be 200,000 tons [61,62]. Thus, according to the estimated annual production of chlorophenols and the average content of PCDEs in them, at least 66 tons of PCDEs per year have been produced as impurities in chlorophenols globally. Municipal waste incineration is another important potential source of PCDEs. The concentration of PCDEs in the stack flue gas of some electric arc furnaces are 0.0115 ng/Nm3 [22]. They were detected at concentrations of 1.48–10.3 ng/Nm3 in the flue gas of a small household waste incinerator, but the level of PCDEs dramatically increased with the addition of chlorine-containing plastic and Cu, reaching 279,000 ng/Nm3 [64]. Seventy-nine PCDE congeners were analyzed in fly ash samples from a municipal waste incineration plant in Germany. The total concentration of all PCDE congeners was 93 µg/kg fly ash, and each PCDE congener had the same level at µg/kg fly ash [9]. In fly ash samples of Finnish municipal waste incineration plants, total PCDE concentrations were detected at 0.1–3.8 µg/kg fly ash [30]. Moreover, chlorinated compounds, such as chlorobenzene, chlorophenol and chlorophenoxy, can condense to PCDEs in the early stages of municipal waste combustion when the combustion chamber temperature is lower than 450 °C, compared to the complete combustion stage when the combustion temperature reaches over 800 °C [10,65]. Unlike the highly chlorinated PCDEs in chlorophenol formulations, the number of substituted Cl atoms of PCDEs from municipal waste combustion is mainly between two and six [9]. This difference may be attributed to the composition of the incineration material. One study suggested that low-chlorine PCDEs were the main congeners in the flue gas of a small household waste incinerator without chlorine-containing plastic, while the percentage of high-chlorine congeners increased by the co-presence of chlorine-containing plastics or Cu [64]. The presence of FeCl3 and CuCl2 in solid waste also increases the formation of highly chlorinated PCDEs on a simulated fly ash surface, whereas Fe2O3 and CuO increase the formation of lower chlorinated PCDEs [66,67]. Furthermore, without oxygen, Fe2O3 catalyzes the formation of PCDEs, whereas CuO reduces PCDE’s formation [67]. The physicochemical properties related to the environment behavior and fate of pollutants have been determined for 106 PCDE congeners by direct chromatographic methods (Table S2) [41,68]. However, given the time- and cost-consuming characteristics to evaluate the physicochemical properties experimentally as well as unavailability of standards for the remaining 103 PCDE congeners, various quantitative structure–property relationship (QSPR) methods have been developed and applied to predict the physicochemical properties based on diverse molecular structural descriptors and regression models. For example, seventeen theoretical molecular structural descriptors and partial least squares (PLS) regression were used to predict the PL and n-octanol/water partition coefficient (KOW) of 209 PCDE congeners [69]. Linear relationships were established between gas-chromatographic relative retention time (RRT), KOW, PL and aqueous solubility (SW,L) of PCDEs and some structural descriptors derived from molecular surface electrostatic potentials by a multiple linear regression (MLR) method and used to predict the physicochemical properties of PCDE congeners not determined experimentally [70]. QSPR models were developed by molecular electronegativity distance vector (MEDV-4) and MLR methods to estimate the PL, KOW and SW,L of 209 PCDE congeners [71]. Based on the number of substituting Cl atoms on the different positions of parent compound diphenyl ether and the number of relative positions for these Cl atoms, a QSPR model was established by the theoretical linear solvation energy relationship (TLSER) method to predict the PL of PCDEs with correlation coefficients R2 of 0.991 [72]. An MLR approach was utilized to develop QSPR models to predict the PL of 106 PCDEs based on calculated molecular descriptors [73]. The SW,L values of five PCDE congeners were predicted using a PLS method [74]. The physicochemical properties predicted from the QSPR models mentioned above are listed in Table S3 [69,71,72,73,74]. The experimental and predicted results show that logPL, logKow and logSw,l of PCDEs range from −5.97 to −0.27, 4.38 to 8.31 and −12.95 to −4.21, respectively. These physicochemical properties indicate that PCDEs tend to accumulate in environments rich in organic matter, such as soils, sediments and organisms. To the best of our knowledge, only three studies are available on the levels of PCDEs in water. Samples from the contaminated area of Whitby Harbor and a bridge near the entrance to Pringle Creek on the north shore of Lake Ontario were analyzed; 45 PCDE congeners were found in the semi-permeable membrane device (SPMD) at total concentrations of 0.68–7.07 ng/L [11]. In China, 15 PCDE congeners were detected in surface water samples from the Nanjing section of the Yangtze River [13]. The total concentration ranged from 1150 to 1800 ng/L and 730 to 1300 ng/L during the low- and high-water periods, respectively, with CDE 30 being the dominant congener. In the next study by the same group, the total concentrations of the PCDE congeners ranged from 0.351 to 2.021 ng/L in surface water samples from Chaohu Lake and its eight main tributaries in China, with CDE 30 (20.63%), CDE 28 (9.78%) and CDE 37 (9.52%) as the major congeners [12]. In general, PCDEs with less substituted Cl atoms have lower logKow and relatively higher water solubility [41]. Therefore, lower chlorinated PCDEs, such as mono-, di- and tri-CDEs are more easily transferred to the aqueous phase than higher chlorinated congeners [12]. The presence of PCDEs in water may be associated with surrounding or upstream industrial production and human activities, such as the production and use of chlorophenols, clofibrate, triclosan, bifenox, 2-chlorophenyl N-methylcarbamate and triadimefon [11,12,13,57]. Studies showed that CDE 37 and 77 could induce severe oxidative damage in green algae (Scenedesmus obliquus), water flea (Daphnia magna), zebrafish (Danio rerio) and crucian carp (Carassius auratus) at environmentally relevant concentrations [59,75]. PCDEs tend to accumulate in the sediment compared to water due to their higher hydrophobicity. The pollution of sediment by PCDEs was first reported for Whitby Harbour on the north shore of Lake Ontario in 1981 [76]. Subsequently, the environmental exposure of PCDEs has gradually received attention. The mean concentrations of total PCDE congeners in sediments of the contaminated area of Whitby Harbour were between 622 and 1929 ng/g dw in 1995 [11]. The average detection concentration of PCDEs in Lake Ontario was 1.30 ng/g dw, which was comparable to that of PCDDs (1.10 ng/g dw) and PCDFs (2.44 ng/g dw) [18]. In the sediment of Kymijoki River in Finland, which was highly contaminated by PCDEs due to the intensive production and use activities nearby of chlorophenol in the 19th century, the total concentration of PCDEs was determined in the range of approximately 130 to 554 ng/g dw (50 congeners tested) in 1993 [15], 8.79 to 606 ng/g dw (40 congeners tested), except for the reference sediment, in 1997 [16], and 85 ng/g dw (nine congeners tested) in 2001 [17]. The types and quantities of the measured compounds were different; therefore, it is difficult to judge the changing trend of PCDEs concentration in sediments of the Kymijoki River year by year. In industrially developed areas of eastern China, sediment samples were collected from Chaohu Lake and the Nanjing section of the Yangtze River, where the total concentrations of 15 PCDE congeners were in the range of 0.279–2.47 ng/g dw and 1.24–3.98 ng/g dw, respectively [12,13]. The level of PCDEs (mean: 1.30 ng/g dw) in the sediments of Chaohu Lake were higher than that of structurally similar polybrominated diphenyl ethers (PBDEs) tested (mean: 0.714 ng/g dw) [77], while lower than that of PCBs (mean: 12.07 ng/g dw) [78]. In addition to sediments, PCDEs in suspended particulate matter (SPM) of Chaohu Lake were also detected. The result showed that the mean total concentration of PCDEs in SPM was comparable to that in the sediment, which was 1.15 ng/g dw, lower than that of PBDEs (mean: 232.5 ng/g dw). In the SPM of the upper Narragansett Bay, the detected concentrations of tri-CDEs and tetra-CDEs were 0.03 ppt dw and 0.06 ppt dw, respectively, which were lower than that of tri-CDF (0.25 ppt dw) [19]. Furthermore, compared with the chlorinated degree of PCDEs in water, PCDEs with more chlorine atoms were more likely to accumulate in sediment and SPM. By contrast, very little information is available on the levels of PCDEs in soils. An earlier study showed that the total concentration of 19 PCDE congeners ranged from <38 to 6800 ng/g dw in soils at 5 contaminated sawmill sites in Sweden [20]. Only one report to our knowledge has recently showed the levels of PCDEs in the atmosphere. That is, the atmospheric occurrence of six PCDE congeners were investigated over the rural area and the Pacific Ocean near Taiwan and the northern Philippines [21]. An elevated mean level of PCDEs was found in the ambient air of the rural area (0.014 pg/m3) compared with that found in the oceanic atmosphere (0.00875 pg/m3). CDE 28 was the predominant congener, accounting for 98.3 and 95.8% of the total PCDEs in the oceanic atmosphere and the ambient air over the land, respectively. Organisms are susceptible to contamination by PCDEs in the environment due to their lipophilic nature. The presence of PCDEs in organisms was first identified in marine organisms, including clam (Mercenaria mercenaria), mussel (Mytilus edulis) and lobster (Honarus americanus), from Narragansett Bay in the United States [19]. PCDEs were also detected in freshwater fish in the North American Great Lakes. The total concentration of 28 monitored PCDE congeners ranged from 24 to 891 ng/g lw in lake trout (Sulvelinus namaycush) and walleye (Stizostedion vitreum vitreum) collected from the Great Lakes on a whole-fish basis [29]. Penta-, hexa- and hepta-chlorinated congeners were the most abundant homologue groups, representing approximately 80 to 90% of the total concentrations. In another study, the occurrence of 15 PCDE congeners was examined in whole fish samples of common carp (Cyprinus carpio) and northern pike (Esox lucius) caught from Whitby Harbour on the north shore of Lake Ontario [28]. The total levels of PCDEs varied from 768 to 14,005 ng/g ww, well above the detected concentrations of PCDFs (58–254 pg/g ww). In a later investigation on 8 fish species, including common shiner (Notropis cornutus), rosyface shiners (Notropis rubellus), spottail shiner (Notropis hudsonius), pumpkinseed (Lepomis gibbosus), yellow perch (Perca flavescens), brown bullhead (Ameiurus nebulosus), white sucker (Catostomus commersoni) and northern pike (Esox lucius), collected also from Whitby Harbour, the total lipid-normalized concentrations of 45 PCDE congeners in muscle samples for each species ranged from 100 to 2857 ng/g, 23231 to 43,231 ng/g, 20,706 to 96,529 ng/g, 30,417 to 68,250 ng/g, 4200 to 130,333 ng/g, 7538 to 213,231 ng/g, 16,714 to 174,571 and 21,000 to 47,000 ng/g, respectively [11]. CDE 99, 153 and 154 were the dominant congeners, and CDE 47, 74, 100, 118, 163, 182 and 184 were also significant. In addition to fish in inland lakes and coastal waters, PCDEs were also indirectly detected in deep sea fish through investigating levels of 106 PCDE congeners in 2 cod liver oils made from North Atlantic deep sea fish [9]. The total PCDE levels were 49 and 659 ng/g lw, respectively. These studies reflect the common presence of PCDEs in organisms in both marine and freshwater environments. PCDEs have also been detected in organisms in other countries and regions. In oligochaete worm (Lumbriculus variegatus), chironomids and northern pike (Esox lucius) collected from sampling sites in the Kymijoki River in Finland, located downstream of an adjacent Ky-5 (which was a chlorophenol wood preservative) production plant, the total concentrations of 40 or 50 PCDE congeners were detected ranging from 215 to 1325 ng/g lw, 0 to 1200 ng/g lw and 677 to 706 ng/g lw, respectively [15,16]. The patterns of PCDE levels in these organisms were similar and resembled that in the sediments collected at the same sampling sites, and these dominant congeners were also abundant in Ky-5 as well. The major PCDE congeners detected in salmon from the Tenojoki Rive, Lake Saimaa and the Simojoki River in Finland were also similar and abundant in Ky-5 too [30]. Moreover, the congener patterns appear to be similar to those detected in Whitby Harbour fish [11]. It indicates that PCDEs contamination in the two regions may be attributed to the production or use of Ky-5 there. PCDEs in organisms are acquired not only by bioconcentration from the ambient environment, but also by biomagnification throughout the food chain. PCDEs have been detected in birds and mammals that eat fish and other aquatic organisms. For example, eggs of fish-eating birds, including common tern (Sterna hirundo), black skimmer (Rynchops niger) and bald eagle (Haliaeetus leucocephalus) from Rhode Island, Louisiana, Michigan and Ohio were examined, and it was found that the total concentrations of three PCDE congeners tested ranged from 11 to 900 ng/g ww [24]. The high concentrations of PCDEs (sum of 7 congeners) were also found in eggs of black-crowned night herons (Nycticorax nycticorax) from Tianmu Lake and whiskered terns (Chlidonias hybrid) from East Tai Lake in China with levels ranging from 11 to 450 ng/g lw and 15 to 700 ng/g lw, respectively [26]. They were well above the detected total concentrations of PCDD/Fs of 0.38–19 and 2.6–33 ng/g lw in the two birds, respectively. In an investigation within the Baltic Sea area as the most polluted brackish water area in the world, the concentrations of individual PCDE congeners were detected ranging from <3 to 79 ng/g lw in eggs of black guillemots (Cepphus grylle L.) and from <5 to 13,000 ng/g lw in breast muscle of white-tailed sea eagles (Hallaeetus albicilla L.) as a top predator of the Baltic food chain [25]. The total concentrations of the 50 tested PCDE congeners varied from 233 to 354 ng/g lw and 1027 to 50,924 ng/g lw, respectively. They were also significantly higher than those of PCDD/Fs, i.e., 3.9–4.0 ng/g lw in black guillemots (Cepphus grylle L.) and 1.6–133 ng/g lw in white-tailed sea eagles (Hallaeetus albicilla L.). In mammals, such as seals, high levels of PCDEs were also detected. The contents of 50 individual congeners ranged from <0.3 to 62 ng/g lw in blubber of ringed seals (Phoca hispida botnica) and grey seals (Halichoerus grypus) from the Gulf of Finland in the Baltic Sea with the total concentrations of 39.9–373.9 ng/g lw [14]. PCDEs were at similar levels in the seal blubber compared to fish captured here and from the Kymijoki River that finally flows into the Gulf of Finland [16,27]. In blubber samples of a Baikal seal (Phoca sibirica) from Lake Baikal in East Siberia of Russia and several ringed seals (Phoca hispida saimensis) from Lake Saimaa in Southeast Finland, the total concentrations of the 50 congeners were found to be 60 ng/g lw and 217–459 ng/g lw, respectively [15]. In blubber samples of harbor seals (Phoca vitulina) captured from the Salish Sea in north–western North America, lower total contents of PCDE congeners (6.5–21 ng/g lw; sum of 46 congeners) were measured, which might be due to light PCDEs contamination in North America [33]. Furthermore, studies have demonstrated the presence of PCDEs in human adipose tissue. Tetra- to deca-CDE congeners in human adipose tissue collected from Canadian municipalities were analyzed. CDE 206 and 209 were found to be in the range of 0.1–2.9 ng/g lw, and the mean level of CDE 206 in males was greater than that in females [34]. Six hexa- to deca-CDE congeners were also detected in human adipose tissue from the USA, where the predominant congener was CDE 206 with concentrations ranging from 0.6 to 1.4 ng/g lw [35]. In addition, it was reported that the concentrations of 50 individual PCDE congeners varied between <0.5 and 7.9 ng/g lw in Finnish human adipose tissue, which were comparable to the levels of PCDD and PCDF congeners (<5 to 7700 pg/g lw) [35]. The main origin of PCDEs found in humans may be contaminated food. Human exposure to PCDEs through the diet was first reported in Catalonia (Spain) in 2004 [36]. PCDEs were detected in a number of foodstuffs available in the local market. The total PCDE concentrations in fresh hake (Rexea solandri), fresh sardine (Sardina pilchardus), mussels and tinned fish were 45.9–707, 400–2707, 59.8–107 and 3.3–71.9 pg/g ww, respectively. Total dietary intake of PCDEs through fish and shellfish was estimated to be 38 ng/day by a standard male adult of 70 kg body weight and aged between 20 and 65 years in Catalonia (Spain), which was slightly higher than PBDEs of approximately 31 ng/day. Moreover, PCDE intake was always higher in males than in females for people under 45 years old due to a greater food intake by males. In a subsequent study by the same research group, the concentrations of PCDEs were determined in 14 edible marine species widely consumed by the population of Catalonia (Spain) [38]. The highest PCDE levels (pg/g ww) were found in red mullet (Mullus barbatus; 7088) followed by sardine (Sardina pilchardus; 1829), anchovy (1606), tuna (Scombridae gen. sp.; 1292) and mackerel (1031). Children aged 4–9 years (boys 0.88 ng/kg/day and girls 0.73 ng/kg/day) showed the highest PCDE intake when judged by the average body weight [79]. Dietary intake of PCDEs in athletes was also evaluated [37]. In general, sportsmen and sportswomen showed a lower daily dietary intake than the general population due to ingesting lower amounts of fish and seafood. In another survey of PCDEs in foodstuffs in Catalonia (Spain) in 2006, the dietary intake of PCDEs was 51.68 ng/day for a standard male adult of 70 kg body weight, increasing by 26% compared to the previous survey (41 ng/day) in 2000, with fish and seafood being the main contributors to this increase [36,39]. In addition, the influence of different cooking processes including frying, grilling, roasting and boiling on the levels of PCDEs in various foodstuffs was evaluated. Studies showed that almost all cooking processes enhanced the total PCDEs levels in fish and meat samples [40,80]. Detailed information about the levels of PCDEs in various environmental media and biota reported previously is provided in Figure 3 and Table 1 and Table 2, respectively. The logKow values of all PCDE congeners were greater than four (Table S2) [41]. Thus, PCDEs are superlipophilic and tend to accumulate in organisms and biomagnify through the food chain [44]. The accumulation rate and extent of PCDEs in fish are generally similar to those of PCBs, and some PCDE congeners are highly bioaccumulative with a logBCF > 3.70 L/kg [44,46]. The bioaccumulation of PCDEs was revealed to be more intense than that of PCDD/Fs in benthic oligochaete worm (Lumbriculus variegatus) following 28-day exposure to contaminated sediments [48]. The logBCF of CDE 47 was measured to be 4.09 in rainbow trout (Salmo gairdneri Richardson) muscle [47]. Recently, the bioaccumulation of 12 PCDE congeners was studied in 3 model aquatic organisms, including green algae (Scenedesmus obliquus), water fleas (Daphnia magna) and zebrafish (Danio rerio) [58]. The logBCF values were found to be in the range of 2.94–3.77, 3.29–4.03 and 2.42–2.89 L/kg ww, respectively. Moreover, the logBCF values increase with the increasing number of substituted Cl atoms, with the exception of CDE 209 maybe due to its large molecular volume preventing it from penetrating through the cell membrane. In addition, similar to PCBs, the number of Cl atoms at para- and meta-position may be another major positive contributing factor for BCFs in the case of the same number of substituted Cl [58]. Some PCDE congeners exhibited biomagnification potentials comparable to some PCDD/Fs, PCBs and PBDEs. For example, as mentioned above, the total concentration of PCDEs detected in breast muscle of white-tailed sea eagles (Hallaeetus albicilla L.), as a top predator of the Baltic food chain, was 1027 to 50,924 ng/g lw, which was significantly higher than that tested in eggs of black guillemots (Cepphus grylle L.) at a lower trophic level ranging from 233 to 354 ng/g lw [25]. Moreover, they were significantly higher than the total concentrations of PCDD/Fs, i.e., 3.9–4.0 and 1.6–133 ng/g lw in black guillemots (Cepphus grylle L.) and white-tailed sea eagles (Hallaeetus albicilla L.), respectively. In a benthic food chain from chironomids to white sucker (Catostomus commersoni) and a pelagic food chain from plankton to pumpkinseed (Lepomis gibbosus) in Whitby Harbour of Lake Ontario, the BMFs (on a ww basis) of six PCDE congeners, including CDE 47, 74, 99, 100, 153 and 154 ranged from 1.4 to 2.3 and 2.7 to 4.7, respectively [11]. The BMFs (on a ww basis) for PCBs were calculated in a food chain from mysids (Mysis relicta) to alewife (Alosa pseudoharengus) in Whitby Harbour in another study [81]. They were 3.5 and 4.5 for PCB 74 and PCB 99, respectively, comparable to those for corresponding PCDE congeners in the pelagic food chain (i.e., 4.3 and 4.0 for CDE 74 and CDE 99, respectively) [11]. In addition, in a simulated aquatic food chain from green algae (Scenedesmus obliquus) to zebrafish (Danio rerio) through water fleas (Daphnia magna), the BMFs for 12 PCDE congeners varied from 0.800 to 3.31 (on a ww basis) or 0.881 to 3.64 (lipid-normalized) [58]. They were also comparable to the lipid-normalized BMFs of 18 PBDEs, which were determined in a highly contaminated freshwater food chain from South China ranging from 0.26 to 4.47 [82]. Furthermore, there is evidence that the BMFs of PCDEs may increase with increasing number of substituted Cl atoms [11,58]. Biological half-lives of tri- to deca-CDEs in rainbow trout (Salmo gairdneri) reached 46 to >300 days, which are almost equivalent to those of PCBs [43]. Moreover, their mean half-life values tend to increase with chlorine content except for CDE 209 with a half-life of 46 days. After being absorbed by brook trout (Salvelinus fontinalis), PCDEs initially enter the blood and liver and then redistribute to adipose tissue and muscle [45]. A study on the tissue distribution of CDE 99 in rats (Rattus norvegicus) showed that the highest concentration was observed in fat, followed by skin, liver, kidney and muscle, and the concentrations declined almost to the background levels in most tissues except for fat on day 21 [83]. Excretion studies of CDE 99 in rats administered at a single oral dose of 10 mg/kg showed that approximately 55% and 1.3% of the oral CDE 99 were excreted in feces and urine, respectively, in 7 days [83]. Thus, fecal excretion may be the main disposition pathway for PCDEs. Studies showed that PCDEs can be metabolized via dechlorination, scission of the ether bond and hydroxylation and methoxylation of aryl nuclei in organisms. It was found that aromatic hydroxylation was the main metabolism pathway in rats (Rattus norvegicus), and it tended to take place ortho and meta to the ether bond [83,84]. Additionally, if PCDEs contain at least one Cl atom at the ortho position of benzene rings relative to the ether bond, predioxins may form via ortho-hydroxylation [84]. Moreover, lower chlorinated PCDEs seem to be metabolized more rapidly than higher chlorinated congeners [45]. The metabolic pathways of CDE 15 were investigated in three aquatic organisms, including green algae (Scenedesmus obliquus), water fleas (Daphnia magna) and zebrafish (Danio rerio) [58]. In green algae and water fleas, dechlorination was found to be the predominant metabolic mode, while methoxylation was the dominant metabolism pathway in the liver of zebrafish, followed by dechlorination and hydroxylation, which was in contrast to the finding by Tulp et al. in rats (Rattus norvegicus) [84]. The partitioning of organic pollutants between gaseous and aerosol or particulate phases is related to the physicochemical property PL. Thus, the PL of organic substances can be used to estimate their distribution, transport and fate in the environment. Pollutants with PL < 10−5 Pa are almost entirely adsorbed on the solid airborne particles, while they preferentially distribute into the gas phase when PL is between 10−5 and <10−2 Pa [85]. The –log PL values of PCDEs range from 0.27 for CDE 1 to 5.80 for CDE 209, increasing with increasing chlorination degree (Table S2) [41]. The range of –log PL values of PCDE are in the same order of magnitude as PCBs (0.56 for PCB 3 to 4.66 for PCB 208) [86]. The –log PL values of tetra- to nona-CDE congeners range from approximately 2 to 5 (Table S2) [41]. It indicates that most PCDE congeners may transport over long distance with the atmosphere in gaseous phase and atmospheric particulate phase [41]. PCDEs have been shown to undergo natural photolysis under environmental conditions and pyrolysis to generate PCDD/Fs, hydroxylated PCDEs (HO-PCDE) and chlorobenzene [87,88]. Among them, PCDD/Fs, especially PCDFs, are the main decomposition products [87,89]. The photolysis pathways of PCDEs include photodechlorination, C-O bond photodissociation and photocyclization to form PCDFs [88]. The photoactivity of PCDEs increases with the increasing chlorination degree [90]. When the pyrolysis temperature is higher than 700 °C, PCDEs are almost completely decomposed [91]. The highest yield of PCDFs is reached at 600 °C [91]. The pyrolysis of PCDEs also has three routes including C-O bond dissociation, ring-closure reaction to form PCDFs, and the addition of a ground state oxygen molecule at an apparent ortho radical site to form PCDDs [89,91]. The formation of PCDD/Fs from PCDEs needs at least one ortho-chlorine for both photolysis and pyrolysis reactions, during which ortho-H2, ortho-HCl and/or ortho-Cl2 are lost [89,91,92]. The degree and pattern of chlorination did not affect the formation pathways of PCDD/Fs [80]. In addition, this cyclisation reaction can be promoted by palladium (II) acetate [93,94]. Therefore, PCDEs should also be noteworthy as an important source of precursors for the formation of PCDD/Fs in the environment. Studies have shown that the coupling reaction of chlorinated diphenyliodonium salt with chlorinated phenols is the common route to synthesize most individual PCDE congeners [95]. However, some iodinated side-products can be formed, such as chlorinated iododiphenyl ethers, which cannot be separated from desired products [95,96]. Moreover, this approach is time-consuming due to inefficient preparation procedure of diphenyliodinium salts. A cuprous iodide-catalyzed Ullman coupling reaction was used later to synthesize ten PCDE congeners with chlorinated iodobenzene and chlorinated phenol in alkaline conditions [13,97]. The advantages of this synthetic route are easy operation and less side-products. The disadvantage is that chlorinated phenols with Cl atom(s) at the ortho position of the hydroxyl group cannot be used as reactants due to the formation of PCDD/Fs as by-products. Mono- and di-CDEs can also be synthesized using copper diacetate-catalyzed Chan–Lam coupling reactions of aryl boronic acids and phenols [98,99]. However, the yield of CDE 4 is low (only 8%) due to the sterically hindered 2-chlorophenylboronic acid. Although gas chromatography combined with mass spectrometry (GC–MS) is the common method for the determination of PCDEs in samples [11,12], high resolution gas chromatography combined with high resolution mass spectrometry (HRGC–HRMS) is preferred due to higher sensitivity and specificity [16,20]. However, several procedures are required before analysis of PCDEs by GC–MS or HRGC–HRMS, depending on the types of environmental samples. For water samples, the main procedures include filtration through 0.45 μm membrane, solid phase extraction and cleanup [12,13]. For solid environmental samples, such as sediment and biota samples, they mainly involve sample pretreatment (e.g., drying, homogenization and acid/base digestion), solvent extraction and cleanup [15,16]. Samples are commonly dried by freeze-drying [11] or sodium sulfate [9]. The dried samples are homogenized using a homogenizer [100] and acid or base digestion [30]. Extraction of PCDEs from water and milk samples have been performed via solid phase extraction [12,13] and Soxhlet extraction [36], respectively, while for solid samples, Soxhlet extraction [16], ultrasonic extraction [15], column extraction [101] and accelerated solvent extraction [32] have been used. Cleanup techniques mainly include bulk matrix removal and adsorption chromatography for removing interference to the subsequent analysis. Bulk matrix removal can be performed by acid treatment [14] or liquid–solid chromatography on Florisil [31], silica gel [19], modified silica gel [40] and modified celite [101] to remove material in biota and sediment extracts, such as lipids, which can disturb final analysis. Adsorption chromatography is used to separate out PCDEs from other compounds, such as PCBs, PCDDs and PCDFs. Column chromatographies on Florisil [16], silica gel [13], alumina [102] and carbon fiber [103] have been successfully used to remove impurities. Despite the concerns about PCDEs as persistent, bioaccumulative and potential toxic substances in the environment, limited studies have been carried out on their toxic effects. Current studies show that PCDEs can cause various adverse effects including lethal toxicity [45,49,50], growth inhibition [49,104], tissue damage [50,105,106,107], reproductive toxicity [52,53], developmental toxicity [49,105,106], immunotoxicity [54], oxidative stress [51,75,104,108,109] and endocrine disorder [53] in mammals, fish and/or plankton (Table S4) [45,49,50,51,52,53,54,75,104,105,106,107,108,109,110]. Subchronic toxicities of several di- to tetra-ortho-chlorinated congeners have been demonstrated to be moderately toxic in rats (Rattus norvegicus), including CDE 99, 100, 132, 139, 153 and 184 [105,106]. Growth inhibition and thyroid gland injury were also detected in these studies [105,106]. In addition, liver injuries were found in CDE 15-exposed mice (Mus musculus) [51]. Hepatic oxidative stress was observed in CDE 37-exposed crucian carp (Carassius auratus) at environmentally relevant concentrations [75]. Subchronic exposure to CDE 3, 15, 37, 77 and 118 caused oxidative stress in the liver and ovary of adult zebrafish (Danio rerio), and CDE 15 could also induce liver nuclei enlargement, necrosis, hepatocyte vacuolation and the developmental inhibition of ovarian cells [107]. Moderate and high acute toxicities, including lethal toxicity or growth inhibition, were observed for CDE 3, 7, 15, 28, 30, 37, 66, 77, 99, 118 and 209 in zebrafish (Danio rerio), water fleas (Daphnia magna) and green algae (Scenedesmus obliquus) [59,104], as well as for CDE 3, 7, 28 and 74 in brook trout (Salvelinus fontinalis) [45], which were generally comparable with those of certain PCBs and PBDEs [59]. Moreover, CDE 77 could also induce oxidative damages in zebrafish (Danio rerio), water fleas (Daphnia magna) and green algae (Scenedesmus obliquus) at environmentally relevant concentrations [59]. The substitution number and pattern of chlorine atoms seem to influence the acute toxicities of the congeners tested, i.e., the non-ortho-substituted may have higher acute toxicities than the ortho-substituted [59]. QSPR models indicate that their acute toxicities may be correlated with their molecular polarizability (α) and the energy of the lowest unoccupied molecular orbital (ELUMO) [59]. The LC50 or EC50 values of PCDE congeners based on acute toxicities from previous studies are listed in Table S5 [45,49,50,59,75]. Reproductive and developmental toxicities were found in PCDEs-administered mice (Mus musculus). For example, the number of pups born per female mated and the number of pups surviving per litter born were both decreased following exposure of parent female mice (Mus musculus) to CDE 71 and 154 from the 6th to 15th day after pregnancy, while CDE 102 and 153 decreased the number of litters born per female mated, without decreasing postnatal survival [52]. Maternal exposure to CDE 71, 102 and 153 also depressed thyroxine levels in both sexes of juvenile rats (Rattus norvegicus) [53]. Moreover, CDE 71 and 102 caused hypothyroidism in pregnant rats (Rattus norvegicus) as well. It was also found that 28-day subchronic exposure of adult rats (Rattus norvegicus) to CDE 99, 100, 132, 139, 153 and 184 caused thyroid damage [105,106]. These findings showed the thyroid-related endocrine-disrupting activity of PCDEs. In addition, developmental toxicities were observed in early life stages of CDE 15-exposed zebrafish (Danio rerio) at 10 mg/L, such as delayed hatching, growth inhibition and malformations, including pericardial edema, yolk sac edema, spine deformation and tail curvature [49]. CDE 15 can also cause the developmental inhibition of ovarian cells in adult female zebrafish (Danio rerio) perhaps due to mitochondrial disappearance [107]. Moreover, CDE 3, 15, 37, 77 and 118 can significantly enhance mRNA expression of the vtg1 gene in the liver of adult male zebrafish (Danio rerio), among which CDE 3 and 15 can also increase vitellogenin content in blood samples, suggesting that some PCDE congeners may be estrogen endocrine disruptors [107]. Among the PCDE congeners tested previously for immunotoxicity, most were found to be immunosuppressive in rodents, including CDE 77, 101, 118, 126, 153, 154, 156, 184, 206, 207, 208 and 209, except for CDE 167, although they were >200 times less immunotoxic than 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) [54,55,106]. For the PCDE congeners, increasing ortho-substitution is less effective in reducing the toxicity of these congeners compared to the well-recognized ortho effects reported for the PCBs [56]. This may be because the ether bridge increased bond length between two phenyl rings, thereby diminishing the effects of ortho substituents on the toxic potencies. A quantitative structure-activity relationship (QSAR) model was developed for PCDEs based on the immunotoxicity values and electronic properties of the 12 PCDE congeners [111]. It showed that congeners with substitutions at positions three and four tended to have higher immunotoxicity and a lower frontier orbital energy gap. However, higher exposure doses (≥25 μmol/kg) were needed to induce immunosuppressive effects for CDE 206, 207, 208 and 209 in less AHR-responsive DBA/2 mice (Mus musculus) compared with the doses of 2.5 to 10 μmol/kg in AHR-responsive C57BL/6 mice (Mus musculus) [54]. Moreover, for CDE 77, 101, 118, 126, 153 and 156, an excellent linear correlation was observed between their immunotoxicity and induced ethoxyresorufin-O-deethylase (EROD) or aryl hydrocarbon hydroxylase (AHH) activity as markers of AHR activation in mice (Mus musculus) [56]. These findings indicate that the immunotoxicity of PCDEs may be mediated by AHR. In addition, it was found that CDE 28, 74, 77, 126, 128, 105, 156, 170, 177, 180, 187, 195, 203 and 209 increased liver monooxygenase activities and/or cytochrome P-450 levels in rats (Rattus norvegicus) [110,112]. Significantly increased induction of EROD activity was also observed in CDE 77-exposed rainbow trout (Oncorhynchus mykiss) by gavage intubation and CDE 74-exposed speckled trout (Salvelinus fontinalis) by intraperitoneal injection [109,112]. Therefore, some PCDE congeners may be potential dioxin-like compounds. Based on the immunotoxicity and AHR-related enzyme activity of the above PCDEs, an interim toxicity equivalence factor (TEF) value of 0.001 relative to 2,3,7,8-TCDD was proposed for non- and mono-ortho-PCDEs in mice (Mus musculus) [55]. TEF values for CDE 77, 118 and 105 in Japanese medaka (Oryzias latipes) were also estimated based on acute mortality data, i.e., 0.00003, 0.00001 and 0.00056, respectively [50]. However, it was found that the immunosuppressive effects of some highly chlorinated congeners might not involve AHR in mice (Mus musculus) [54]. In addition, the induction of EROD activity by 29 PCDEs was tested using a H4IIE rat hepatoma cell bioassay [108]. It was found that all the 29 PCDEs were inactive except for CDE 156 as a weak EROD inducer with a TEF value of approximately 1.2 × 10−5. The controversial results may be attributed to differences in experimental method and species sensitivity. Overall, the mechanisms of toxicities of PCDEs, especially whether the toxicities are mediated by AHR, need further research. Future research may be carried out from the following aspects. First, more PCDE standards without PCDD/Fs impurities should be synthetized as a research basis. Second, the research on the environmental behavior of PCDEs on living organisms in real ecosystems, as well as exposure levels of PCDEs in wild animals and humans is still limited and needs to be strengthened. Third, chronic and multigenerational toxicological and ecotoxicological studies at environmentally relevant concentrations are needed to assess the impacts of long-term low-concentration exposure on individuals and populations. Fourth, previous studies mainly focused on the determination of basic acute toxicity endpoints. Moreover, there are conflicting views on whether the toxic effects are mediated by AHR. Thus, the molecular toxic mechanisms of PCDEs deserve more research by use of computational techniques, such as molecular docking and molecular dynamics simulations [113,114], as well as experimental methods. Fifth, inter-species sensitivity variations to PCDEs and effects of exposure at different developmental stages of organisms on population are also interesting. Finally, toxicity prediction models for identifying PCDE congeners of high priority and assessing health and ecological risks should be built in the future as soon as possible. The existing studies on PCDEs from the perspectives of source, physicochemical property, environmental level and fate, synthesis and analysis and toxicology are reviewed in the present paper. These studies indicate that PCDEs have widely existed in various environmental media and organisms and can bioaccumulate and bioamplify through the food chain due to the long-range transport potential and high lipophilicity. PCDEs can be metabolized into other organic pollutants, such as HO-PCDEs, methoxylated PCDEs (MeO-PCDEs) and even PCDD/Fs through biotransformation, photolysis and pyrolysis reactions in the environment. In addition, the commonly used synthesis and analysis methods are coupling reaction and GC-MS/HRGC-HRMS, respectively. Toxicological studies have shown that PCDEs can cause lethality, teratogenicity, growth inhibition, tissue damage, reproductive and developmental abnormalities, oxidative stress, immunotoxicity and endocrine disorders in organisms. However, due to the lack of exposure and toxicological data, the health and ecological risk assessment of PCDEs has not yet been evaluated. Therefore, future research perspectives are also proposed to facilitate the assessment of health and ecological risks of PCDEs.

PMC10002349

Immacolata Pietraforte,Loredana Frasca

Autoreactive T-Cells in Psoriasis: Are They Spoiled Tregs and Can Therapies Restore Their Functions?

22-02-2023

autoreactivity,Tregs,psoriasis

Psoriasis is a chronic inflammatory skin disease, which affects 2–4% of the population worldwide. T-cell derived factors such as Th17 and Th1 cytokines or cytokines such as IL-23, which favors Th17-expansion/differentiation, dominate in the disease. Therapies targeting these factors have been developed over the years. An autoimmune component is present, as autoreactive T-cells specific for keratins, the antimicrobial peptide LL37 and ADAMTSL5 have been described. Both autoreactive CD4 and CD8 T-cells exist, produce pathogenic cytokines, and correlate with disease activity. Along with the assumption that psoriasis is a T-cell-driven disease, Tregs have been studied extensively over the years, both in the skin and in circulation. This narrative review resumes the main findings about Tregs in psoriasis. We discuss how Tregs increase in psoriasis but are impaired in their regulatory/suppressive function. We debate the possibility that Tregs convert into T-effector cells under inflammatory conditions; for instance, they may turn into Th17-cells. We put particular emphasis on therapies that seem to counteract this conversion. We have enriched this review with an experimental section analyzing T-cells specific for the autoantigen LL37 in a healthy subject, suggesting that a shared specificity may exist between Tregs and autoreactive responder T-cells. This suggests that successful psoriasis treatments may, among other effects, restore Tregs numbers and functions.

Autoreactive T-Cells in Psoriasis: Are They Spoiled Tregs and Can Therapies Restore Their Functions? Psoriasis is a chronic inflammatory skin disease, which affects 2–4% of the population worldwide. T-cell derived factors such as Th17 and Th1 cytokines or cytokines such as IL-23, which favors Th17-expansion/differentiation, dominate in the disease. Therapies targeting these factors have been developed over the years. An autoimmune component is present, as autoreactive T-cells specific for keratins, the antimicrobial peptide LL37 and ADAMTSL5 have been described. Both autoreactive CD4 and CD8 T-cells exist, produce pathogenic cytokines, and correlate with disease activity. Along with the assumption that psoriasis is a T-cell-driven disease, Tregs have been studied extensively over the years, both in the skin and in circulation. This narrative review resumes the main findings about Tregs in psoriasis. We discuss how Tregs increase in psoriasis but are impaired in their regulatory/suppressive function. We debate the possibility that Tregs convert into T-effector cells under inflammatory conditions; for instance, they may turn into Th17-cells. We put particular emphasis on therapies that seem to counteract this conversion. We have enriched this review with an experimental section analyzing T-cells specific for the autoantigen LL37 in a healthy subject, suggesting that a shared specificity may exist between Tregs and autoreactive responder T-cells. This suggests that successful psoriasis treatments may, among other effects, restore Tregs numbers and functions. Psoriasis is a chronic inflammatory skin disease affecting 2–4% of the global population [1]. Psoriasis has a strong genetic component, but environmental factors are also important [2,3]. Psoriasis is characteristic, with thickening and scaling of the epidermis due to hyper-proliferation of keratinocytes (acanthosis). CD4 and CD8 T-cell infiltrate characterizes the skin in psoriasis, a T-cell-driven disease, but other cell types are also present in the lesions, including neutrophils, macrophages, NK cells, dendritic cells [4]. It has been long known that the pathogenesis of psoriasis is driven by T-cell derived factors, produced by subsets including T-helper (Th) 1 cells, Th17, Th22 and regulatory T cells (Tregs). There is evidence in animal models that Tregs could play a role in ameliorating psoriasis [5]. However, these Tregs are considered dysfunctional, as they are altered in frequency, phenotype, and function [6]. It is possible that Th17-cells could be diverted Tregs in psoriasis. If this is true, pharmacological intervention could restore their functions. This is an important goal to achieve in therapy, to stabilize the disease after treatment, which could also give the opportunity to discontinue therapy without unwanted consequences. Tregs are of various types: they can act through secretion of regulatory/suppressive cytokines like IL-10, TGF-β, IL-35 or act by contact. Tregs that express Foxp3 can lose their capacity to suppress in certain circumstances. For instance, Tregs can convert to IL-17-like cells under inflammatory conditions, especially in the presence of IL-6, which is a cytokine highly up-regulated in psoriasis [7]. In the last part of this narrative review, we report that a healthy person can harbor the T-cell specific for LL37, which are able to proliferate and make cytokines in response to LL37 stimulation. We show the existence of a Tregs counter-part of the LL37-specific T-cell pool, endowed with suppressive functions in vitro. Work with animal models has shown that more Tregs subsets exist, which parallel the T-helper cell subsets [8]. Most of these studies were in mice, as analogous subsets in humans are more difficult to study because of lack of accurate identification markers [9,10]. Tregs expressing the Th1 factor T-bet exist, which also express chemokine receptor CXCR3 [11]. The deletion of Foxp3 in T-bet-expressing cells was shown to determine uncontrolled Th1 immune responses at steady state [12]. In addition, Th2-like Tregs have been identified, for instance, in non-lymphoid organs. The Th2-like Tregs can express GATA-3 and have been detected at mucosal surfaces: skin, gut, and lungs [13,14]. Indeed, studies demonstrated that, in mice, GATA-3 expression is required for Tregs stability and maintenance of Foxp3 expression [15]. A pathogenic reprogramming of T-cells can cause disease, and there are several examples of this. For instance, Th2 Tregs can play a role in food allergies [16]. Relevant to psoriasis, Tregs can also express the canonical Th17 transcription factor, RORγt. Cells of these types have been detected in the gut [17,18], but they have been also described in patients with psoriasis [19], as well as in inflammatory bowel diseases (IBD), and arthritis; in some cases, they convert into IL-17 secreting cells [20,21,22,23]. Single cell RNA sequence analysis shows the existence of distinct T-cell subsets in relapsing psoriasis for tissue-resident T-cells. Th17 cells and NF-kB signaling pathways evidenced include Pellino-1 (PELI1). Mice with systemic and conditional depletion of PELI1 were protected from psoriasiform dermatitis and showed reduced IL-17A production and NFkB activation in Th17 cells. The inhibition of PELI1 significantly ameliorated murine psoriasiform dermatitis by reducing IL-17A production. PELI1 belongs to a member of the E3 ubiquitin ligases, mediating immune receptor signaling cascades involving, among others, the NFkB pathway; it also promotes intrinsic activation of skin resident Th17 cells in psoriasis. This points out that this gene inhibition could be a promising therapeutic strategy for psoriasis, limiting skin-resident Th17 cell responses. Both the T-cell receptor (TCR) and IL-23 stimulation upregulate PELI1 expression in Th17 cells from lesional skin. However, it is currently unknown whether PELI1 directly modulates the suppressive function of Tregs [24]. Recently, the regulatory counterpart of follicular helper T-cells (Tfh) cells, which promote germinal center (GC) responses and antibody production, has been described. These cells express TCF1 and LEF1, required for expression of B-cell lymphoma 6 (Bcl-6) [8]. It has been shown that Tregs are part of the TRM (Tissue Resident Memory cells) pool, and it is postulated that these cells can re-circulate. Human psoriatic plaques contain abundant numbers of IL-17A-producing T-cells with a CD69+ and/or CD103+ TRM phenotype. TRM can play a role in chronic inflammatory diseases of the skin, such as vitiligo and psoriasis, and are likely involved in recurrent lesions. A greater number of CD8 TRM cells infiltrate psoriatic lesional and non-lesional skin, as compared to normal skin. These are mainly infiltrating CD8+IL-17A+ TRM cells and correlate with disease duration. Apparently, healed skin contains CD8 TRM cells, which mostly localize in the epidermis (as they express CD103), while CD4 TRM cells localize in the dermis. Moreover, a large pool of TRM cells express IL-17A alone, IL-17A and IL-22, or IL-22 alone, and persist in post-lesional psoriasis skin. Cytokine production is not exclusively oriented toward a Th17 phenotype as TRM also express IFN-γ, showing a Th1 phenotype, which could be induced by IL-15 stimulation [25]. Tregs residing in the skin seem primary localized, in mouse and humans, in the hair follicles (HFs) [22,26,27]; this is due to HF epithelial cells production of CCL20, which recruits CCR6-expressing Tregs into the skin [26]. In human and mice skin, Tregs constitute about 20–40% of the CD4 T-cells [22]. These Tregs can express GATA-3 [15]. Their role is to maintain the cutaneous immune homeostasis, promote wound healing, and repair tissues. Interestingly, they express ST2 [28], the receptor for IL-33, which is released in situation of tissue damage [29]. IL-33 is an alarmin, which can regulate Foxp3 in the mucosal tissue [30]. IL-33 was originally described as a cytokine that activates Th2 cells, but it can affect various cell types, including Th1 cells or innate lymphoid cells (ILC) and CD8 T-cells [31]. Its exact role in psoriasis is not completely clear due to these pleiotropic effects. In support of the idea that Tregs increased because of tissue damage stimuli, a recent study showed that Tregs in the skin expand following UVB irradiation [32]. After injury, Tregs express the AREG receptor, EGFR, suggesting an AREG autocrine role in tissue repair [33]. The peripheral blood of healthy humans contains Tregs, which express cutaneous lymphocyte antigen (CLA) and other skin homing receptors [34], suggesting that Tregs migrate to the skin using these unique receptors. Injury and microbial invasion influence this migration. Similar to human Tregs, mouse Tregs express some homing receptors, which facilitate their recruitment into inflamed skin [35]. Mouse and human studies indicate that, in psoriasis, the IL-23/IL-17 axis of inflammation, together with Tregs dysfunction, determines the Th17/Tregs imbalance implicated in the disease [6,36], although the link between Tregs and disease severity is debated [37]. The conflicting results may depend on several factors: the site of biopsies, psoriasis subtypes, and different disease status. Although Tregs and Th17 cells increase in adult and pediatric psoriasis patients, Tregs were shown to be unable to suppress Th17 cells activation, and their effector T-cell responses and proliferation [6,37]. Several papers support Tregs dysfunction in psoriasis. The Tregs suppressive function seem impaired due to the pro-inflammatory cytokine milieu. For example, the exposure to high level of IL-6 decreases Tregs activity [7,38]. Dysfunctional Tregs in peripheral blood of patients with psoriasis have been reported showing that they have a phosphorylation and an aberrant activation of STAT3, which is due to the effects of pro-inflammatory cytokines, not only IL-6, but also IL-21 and IL-23 [39]. Additionally, inhibition of Foxp3 by up regulation of microRNA (mir-210), in CD4 T-cells, results in decreased levels of IL-10 and TGF-β by these cells, accompanied by increased levels of pro-inflammatory cytokines release, such as TNF-α and IL-17A, as described by Zhao et al. [40]. Tregs can express CD39 and CD73, so they can use the adenosine signaling for exerting suppressive effects. Yan et al. reported that Tregs from patients with psoriasis have a reduced expression of CD73 and a CD73/AMPK pathway deactivation [41]. The adenosine pathway (CD39 and CD73) is widely known to play a crucial role for the Tregs immunosuppressive function. CD39 removes extracellular ATP by hydrolyzing ATP/UTP and ADP/UDP into AMP. Successively, AMP is rapidly degraded into adenosine in the presence of CD73. Then, adenosine, which binds to the A2AR receptor on Tregs, triggers the accumulation of intracellular cAMP, activation of AMPK and inactivation of mTOR, blocking IL-17 and IFN-γ production. This also blocks the differentiation of Tregs into IL-17 secreting Tregs, which have a decreased immune suppressive function. Moreover, adenosine binding to A2AR of the T effector cells, achieves local repression of immune response and down regulation of INF-γ production. Inefficient recruitment of Tregs to inflamed skin can also concur to the inefficiency in restraining inflammation in individuals with psoriasis [42]. Finally, although microbial infection in the skin is a contributing factor in psoriasis, a connection by skin microbiota and Tregs in psoriasis can be hypothesized [43]. All these data in the scientific literature show that the suppressive function of circulating and in skin-resident Tregs is important in psoriasis and support an impairment of Tregs in the psoriatic disease. Tregs and Th17 cells require TGF-β to develop from a common precursor, indicating a constant competition between the two cell types [44], highlighting a special relationship in the development and function of Th17 and Tregs. Several studies revealed that a subset of Tregs in the skin might differentiate into Th17 cells [19,45]. It seems that CD27 and OX40 expression on Tregs plays a role in suppressing Tregs differentiation toward Th17 phenotype, whereas lack of expression of these molecules induces the expression of high levels of IL-17A and the transcription factor RORα [46,47]. Tregs from patients with psoriasis are able to differentiate into IL-17A producing cells after stimulation ex vivo. Tregs may also differentiate into IL-17A producing cells upon activation of the histone deacetylase 1 (HDAC-1), which is elevated in psoriatic lesions. Moreover, the presence of IL-17A+Foxp3+CD4 T-cells was observed in psoriatic skin [19]. All these data show not only a Tregs dysfunction but also a phenotypic alteration of these cells in psoriasis. Many treatments are in use for psoriasis, and others are being explored. An overview on how these treatments appear to influence expansion or functionality of Tregs and the Th17/Tregs balance is reported below. The therapeutic approach to psoriasis comprises two major categories of drugs: biological agents, and immunosuppressive drugs (methotrexate, cyclosporine), and treatments sub-groups are evidenced. Among biological agents, TNF-α antagonists have been extensively used. These include infliximab (a chimeric monoclonal antibody composed of a human IgG1 constant region and a murine variable region), etanercept (a soluble TNFR, made of two extracellular domains of the human TNFR2 fused to the Fc fragment of human IgG1), and adalimumab (a humanized monoclonal antibody). Different studies in humans refer that anti TNF-α agents increase Tregs and decrease the Th17 cells frequency in peripheral blood of psoriasis patients [48,49,50]. Etanercept regimen showed a more significant modification of the T-cell subsets, as compared to the other two drugs [51,52,53]. Moreover, Diluvio et al. reported that infliximab treatment induces a polyclonal expansion of Tregs, sorted from peripheral blood of patients, showing a diverse TCR repertoire [54]. In all of these studies Tregs suppressive function was not addressed and the modification of the subset of Tregs in the skin has not been analyzed; the data refer only to peripheral blood cells. Of note, in a murine psoriasiform model, some data are conflicting, and not confirmed in humans [55]. Classes of biologic agents targeting either IL-17 or IL-23 demonstrated higher rates of response and superiority compared to previous biologic agents. Among these new therapeutic agents are an anti-IL-17A (secukinumab), an anti-IL-23p19, called guselkumab, and an anti-p40, called ustekinumab. In a model of imiquimod-induced psoriasis in mice, it was described an increase of Foxp3+ Tregs in the skin and restoration of their suppressive function following use of IL-17 or IL-23 blocking antibodies but not with an anti-TNF-α treatment [55]. Of note, Kanman et al. described a principal role of IL-23 in regulation of Tregs plasticity and conversion into a Th17 like phenotype [56]. These data are supported by observations in humans that IL-23 inhibitors act as potent disease modifying drug, more than IL-17 antagonists [57,58,59]. Clinical trials testing IL-23 inhibitors showed long-lasting maintenance of the therapeutic response following treatment discontinuation, compared to IL-17 inhibitors [58,60,61]. Another clinical trial, comparing guselkumab to secukinumab, provided relevant insights about the skin compartment. During a 24-week treatment with guselkumab or secukinumab, the number of CD4 and CD8 TRM cells decreased in psoriatic lesions of both treatment arms, but guselkumab reduced memory T-cells, maintaining Tregs whereas the opposite was observed for secukinumab treatment [57]. Secukinumab treatment decreased the number of Tregs in a more pronounced way than guselkumab. Moreover, a greater decrease of LC (Langerhans cells), infiltrating post-lesional skin, was observed after IL-23 blockade. These findings suggest a successful response to either IL-23 or IL-17 inhibitors, with an increased Tregs/CD8 TRM ratio. A superior long-term control of skin inflammation was achieved by inhibiting IL-23 with a reset of the pathogenic inflammatory T-cells and an increase of Tregs. IL-23 acts on the T-cell compartment and stimulates the expression of RORγt, and the production of IL-17A, IL-1F, and IL-22. Moreover, IL-23 drives and maintains the differentiation of Th17 cells. In contrast, IL-17A is an effectors cytokine that induces skin inflammation. It is also expressed by neutrophil and mast cells and produces indirect effect on the T-cell compartment. This may explain why IL-17 inhibitors have lower modulator ability than IL-23 inhibitors. IL-6 is another important cytokine involved in psoriasis. Although IL-6 plays a role in the maturation of Th17 cells, search of the literature and clinical trials in websites did not reveal psoriasis studies with anti-IL-6/IL-6 receptors and role on Tregs frequency and Th1/Tregs balance [62]. A study in vitro reported that IL-6 was necessary and sufficient to reverse human T-cell suppression by Tregs in in vitro models using activated DCs as a source of IL-6 [7]. Although IL-6 may be another potential target for psoriasis treatment, data in the literature show that attempts to treat psoriasis with tocilizumab (TCZ), a humanized anti-interleukin-6 (IL-6) receptor antibody licensed for the treatment of rheumatoid arthritis (RA), have been unsuccessful [63]. On the other hand, the use of new IL-6 inhibitors such as clazakizumab, a monoclonal antibody with high affinity and specificity for IL-6, could be more promising for psoriatic arthritis (PsA) [57,64,65,66]. Many photo therapeutic approaches can treat psoriasis: natural phototherapy, broadband UVB, narrowband UVB, selective UV phototherapy, Xenon chloride excimer laser, Xenon chloride excimer lamp, UVB light emitting diode, flat-type fluorescent UVB lamp, UVA, Mixed UVB/UVA, Psoralen+ UVA photochemotherapy (PUVA), bath water delivery of 8-methoxypsoralen and subsequent UVA-irradiation (bath-PUVA therapy), UVA-1 phototherapy, Pulsed Dye laser, and others [67,68]. The principal immunomodulatory effect of phototherapy is promoting the death of effector cells, such as T-cells, and keratocytes, and inhibition of LC, macrophages, neutrophils and NK cell function. Effect of bath-PUVA therapy was reported on three distinct Foxp3+ subsets: activated Tregs (aTregs), resting Tregs (rTregs), and cytokine-secreting non-suppressive T-cells from peripheral blood of psoriasis patients and healthy controls. Bath-PUVA therapy increased Tregs and restored dysfunctional Tregs activity in patients: in particular aTregs were significantly increased in the early bath-PUVA therapy sessions, and then diminished. RTregs, which were lower in patients than healthy controls, increased during therapy [68]. Takuya Furuhashi et al. confirmed these data by functional assays. CD4 CD25− cells separated from PBMCs of psoriasis patients treated with PUVA and activated with anti-CD3/CD28-bound beads, were cultured with or without CD4 CD25+ T-cells. The ability of Tregs to suppress CD4 CD25− T-cells was calculated by comparing the proliferation rates of CD4 CD25− T-cells in the presence/absence of CD4 CD25+ T-cells [69]. The same conclusions derive from an UV (B) treatment in psoriasis patients with polymorphic light eruption (PLE), in which UV increased the number of Tregs. This might be a compensatory mechanism to counteract the susceptibility to PLE [70]. Tregs from patients with PLE lacked any capacity to suppress effector T-cell proliferation but this capacity improved after therapy, as demonstrated by regulatory T-cell suppression assay. Moreover, after UVB treatment, keratinocytes upregulated the expression of receptor of activated nuclear factor-B ligand (RANKL). This receptor interacts with RANK on DCs, making DCs able to expand the number of Tregs [71,72]. These data were confirmed also in a mouse model of psoriasis [73]. Vitamin A derivatives, retinoids, are also of common use to treat psoriasis. Retinoids, such as etretinate or acitretin, are absorbed in the small intestine and then are metabolized in other organs to the active acid form of retinoid acids (RAs), which interact with retinoid X receptors (RXRs). This heterodimer binds the RA response element on CNS1 of Foxp3, inducing Foxp3 expression and the generation of peripheral Tregs from naive T-cells [74]. Of note, retinoids not only promote Tregs generation but also regulate TGF-β, capable of inhibiting the IL-6. IL-6 is driving activation of pro-inflammatory Th17 cells, acting on RORγt [75,76]. Vitamin D seems to regulate Tregs. Vitamin D status correlates with circulating Tregs in patients affected by psoriasis; a correlation with the severity of the disease, evaluated with Psoriasis Area Severity Index (PASI) score is present. In a clinical study, patients were analyzed for PASI-score, serum levels vitamin D and regulatory T-cells percentage. Using non parametric Spearman coefficient test to assess correlation between serum levels of vitamin D and the single variables of disease, this study found a positive association between vitamin D and Tregs population (p < 0.001), and an inverse correlation between vitamin D and PASI-score (p = 0.04) [77]. The effects of maxacalcitol, a vitamin D3 analogue, and betamethasone valerate (BV) steroid lotion, confirmed the effects of vitamin D on the differentiation of T-cells with suppressive phenotypes in an imiquimod (IMQ)-induced psoriasiform skin inflammation animal model. The authors report that maxacalcitol and BV reduced the MHC Class II+ inflammatory cell infiltrate and down-regulated IL-17A, IL-17F, IL-22, IL-12p40, TNF-α and IL-6 mRNA expression levels in the inflamed mouse ski. Maxacalcitol alone downregulated IL-23p19 expression, and increased Foxp3+ T-cell infiltrations and IL-10 expression. Of note, adoptive transfer of Tregs from maxacalcitol-treated donor mice improved IMQ-induced inflammation more than of Tregs from a BV-treated donor group [78]. Many data in the literature, from humans and mouse models, report that vitamin D induces myeloid dendritic cells with a tolerogenic phenotype responsible for the differentiation of CD4 CD25+ Tregs from naive T cells [79,80,81]. Topical therapies based on glucocorticoids (GC) and calcipotriol are usually sufficient to manage mild and moderate psoriasis [82]. GCs produce anti-inflammatory effects through GC receptors (GR) and by acting on specific target genes, inhibiting several cytokines [83]. Calcipotriol exerts its effect by binding to the nuclear vitamin D3 receptor [84]. The anti-inflammatory effects of calcipotriol are inferior compared with those of GCs, but an incremented effect is seen with a combinatory therapy [85]. Keijsers and co-workers showed that topical calcipotriol/betamethasone treatment for eight weeks decreased the number of Tregs in psoriatic lesions and the expression of Foxp3 in the skin and PBMCs [86]. Minna E. Kubin et al. confirmed that a combination therapy down-regulated the expression of TNF-α, IL-23, IL-17A, S100A7, CCL20 and interferon-γ in the skin and TNF-α, IL-6, IL-23A, T-bet and IFN-γ in PBMCs. Calcipotriol/betamethasone, but not betamethasone alone, down-regulated expression of Foxp3 in both skin and PBMCs [87]. The European Medicines Agency [88] approves this drug for the treatment of psoriasis patients as of 2017, as an oral formulation. Studies in vitro show that DMF promotes oxidative stress reducing vitality of conventional T-cells but not Tregs. An increased expression on Tregs of cell surface-reduced thiols or thioredoxin-1 [89], protect Tregs from oxidative stress, mediated by DMF. The anti-psoriatic effect of DMF favors Tregs survival but not Th17 expansion [90]. In psoriasis patients, DMF treatment increased Tregs frequency and decreased Th17 cells, confirming in vitro data [91]. A clinical study in psoriasis patients, using the pan-protein kinase C (PKC) inhibitor sotrastaurin (AEB071), showed a reduction of psoriasis clinical severity [92]. Currently, sotrastaurin is in phase II clinical trial studies for psoriasis [93]. Sotrastaurin blocks more than one PKC isoform. The latter belongs to a sub family of PKC calcium-independent and is most abundant in T-cells [94]. The activation of T-cells by CD28 and TCR promotes PKC-theta activation and translocation into the membrane at site of immunological synapse (IS), leading to the activation of NF-kB. Inhibition of PKC-theta restored activity of defective Tregs from RA patients and enhanced protection of mice from inflammatory colitis [95]. Moreover, studies in PKC-knockout mice have shown that PKC-theta is required for productive Th2 [96] and Th17 [97] responses but not for Th1 responses. In particular, Xuehui He et al. confirmed that sotrastaurin prevented TCR/CD28-induced T-cell activation and pro-inflammatory cytokine production, and enhanced Tregs response [98]. This class of inhibitors is apparently restoring Tregs activity in psoriasis. The binding of cytokines to their receptors enables the activation of the JAK/STAT signaling pathways. This happens for IL-6, IFN-γ, IL-22, and IL-21, all involved in psoriasis. JAK inhibitors may thus suppress the effects of inflammatory cytokines involved in the disease [99]. The same (JAK) inhibitors are indicated for treatment of PsA, as the Food and Drug Administration approved the inhibitor tofacitinib, whereas the JAK1 inhibitor upadacitinib is approved in Japan. In a model of hepatitis in mice, induced by concanavalinA (ConA), tofacitinib increased the ratio of Tregs/Th17 cells as detected not only in the mouse liver but also in the spleen, which is representative of the situation in the peripheral regions [100]. An in vitro study reported that tofacitinib suppresses T-effector functions but preserves activity of CD4 CD25bright Tregs. This may explain its capacity to increase the Tregs/Th17 ratio [101]. MTX, a folic acid analogue, is another treatment for psoriasis that is able to inhibit the activation of lymphocytes and macrophages, thus modulating cytokines, and inhibiting neutrophil chemotaxis [102]. MTX monotherapy determined, after 15 weeks of treatment, an increase in the percentages of Th2/Treg cells and a concomitant decrease of Th1 and Th17 cells [103]. In a study by K. Yan et al., Tregs and effector T-cells were isolated from blood of patients with psoriasis and healthy controls. In psoriasis patients, Tregs had a decreased immune suppressive function and a reduced expression of CD73, as compared to the healthy controls. Both IL-17 and IFN-γ were significantly upregulated in psoriasis, implying that T effector cells in the tissues possessed an aberrant secretion capacity of Th1/Th17 cytokines. The authors observed that, in patients, MTX treatment induced a significant growth inhibition of T effector cells. The production of IL-17 and IFN-γ by Tregs was also reduced, suggesting that MTX restores the function of Tregs and restrains the proliferation of T effectors in psoriasis patients. The authors analyzed CD73 expression by flow cytometry, and the phosphorylation of AMPK and mTOR by western blot. In all patients, MTX treatment reversed down-regulation of CD73, activated AMPK and inactivated mTOR [41]. In contrast to conventional resting T-cells, Tregs were found to express both CD39 and CD73 at high levels. These surface nucleosidases enzymatically active possess immunosuppressive properties on effector T-cells by negative feedback responses via the adenosine receptor (A2AR) but also via low-affinity receptors (like the A2B-adenosine receptor, A2BR). A2AR is ubiquitously expressed in a wide variety of immune cells including T-cells, B cells, NK cells, NKT cells, macrophages, dendritic cells, and granulocytes; A2BR plays a distinctive role in controlling inflammation, for example, via the induction of a tolerogenic antigen presenting cells (APC), via an alternative activation. Upon interaction with A2AR, adenosine is responsible for the inhibition of T-cell activation. Moreover, immunosuppressive activity may be further enhanced by adenosine, which induces Tregs, promoting tolerogenic antigen-presenting cells (APC) and myeloid-derived suppressor cells (MDSC) activities [104]. Other potential and promising targets, which can be useful in the regulation of Tregs, or to reset the imbalance of T-helper/Tregs in psoriasis, are under investigation; among them, IL-2 at low dose, histone deacetylase inhibitors sodium butyrate, STAT3 inhibitors, probiotics and T-cell based therapies [105,106,107,108,109,110]. The data and experiments illustrated above, about the capacity of diverse treatments options to restore the number of Tregs and/or their effector functions in psoriasis, indicate that Tregs are important to counteract inflammation in this chronic skin disease. Studies on the specificity of Tregs in psoriasis are lacking, although it has long been known that the disease is T-cell driven. The specificity of psoriasis T-cells driving inflammation in the skin has been elusive for a long time. Previous studies demonstrated that specificity was directed towards keratins [111]. Later, we discovered that the antimicrobial peptide (AMP) cathelicidin LL37 is an autoantigen in psoriasis [50], and another study found that ADAMTSL5 is also an autoantigen in psoriasis [112]. In the past, we have identified LL37-specific T-cells by using peptide-MHC-tetramers. This approach, together with epitope mapping and cloning of the T-cells, identified the most immunogenic parts of LL37 and the restriction molecules for presentation to T-cells (there were several HLA-class I and class II alleles involved in the recognition by CD4 and CD8 T-cells, among which were HLA-DR7, HLA-DR11, HLA-DR4, and HLA-Cw6) [50,113]. Usually, healthy donors (HD) do not respond to LL37 or LL37-derived shorter antigenic peptides in T-cell proliferation assays [50,113]. Occasionally, low and rare proliferation can be detected in HD, which is not significant as compared to the psoriasis group. With this in mind we tried to check, in HD with a low LL37 T-cell proliferative response, whether LL37-specific T-cells existed and whether these cells belong to the Tregs compartment (the strategy and methods we used for this analysis is reported in Section 4 (see below). Knowing the HLA-typing of some HD, we managed to obtain a T-cell clone specific for LL37 in a HLA-DR11-positive individual (see Methods). This clone was stained by a peptide-MHC-tetramer of the MHC haplotype HLA-DR11 linked to peptide P6 (that we call DR11-P6 tetramer), but not by a second peptide-MHC-tetramer of HLA-DR11, and loaded with a different LL37-derived peptide (P4), referred to as DR11-P4 tetramer. Next, we isolated Tregs from the PBMCs of the same donor by using magnetic isolation kit and obtained Tregs that were expanded for three weeks in the presence of a high dose of human recombinant (hr) IL-2 (according to previous published protocols, see methods) [114]. Once Tregs were expanded by repeated stimulations (phenotype of the obtained Tregs is reported in Figure S1), in the presence of high dose hrIL-2 (300 U/mL), we obtained T-cells that could be, in part, stained by the same peptide-MHC-tetramer DR11-P6, but not by the control peptide-MHC-tetramer DR11-P4 (Figure 1). Figure 1 shows staining of clone T (Figure 1, Clone T) and of Tregs (Figure 1, Tregs), derived from the same donor. At this stage, we performed a suppression assay to see whether the recovered Tregs were able to suppress activation of the clone T and its production of IFN-γ after over-night culture (Figure 2a). The gating strategy for this assay is shown in Figure S2. To distinguish responder T-cells of the clone (T) from the Tregs, we pretreated the latter with CSFE to exclude these cells from the analysis of IFN-γ production in response to the LL37 antigenic peptide (P6). The data in Figure 2a (and Figure S2) show that clone T responded to LL37-P6 peptide by producing IFN-γ, but the presence of Tregs reduced this production. The Tregs were not able to respond producing IFN-γ in response to P6, as shown when they were cultured with the APC alone presenting the peptide P6, in the absence of the T-cells derived from the clone (T). We also performed similar experiments during a short-term culture (2 h), where we found inhibition of the capacity of clone T to produce IL-2 in the presence of APC and P6 when Tregs were present into the cultures (Figure 2b). IL-2 production from the clone was assessed by catch assay (see methods). This overview of the role of Tregs in psoriasis and our own experimental findings suggest that autoantigen specific, in this case LL37 specific T-cells, can be part of the physiological Tregs pool in HD. A person with T-cells responding to LL37 may have Tregs that are also specific for LL37 (as shown here by using peptide-MHC-tetramers specific for LL37-epitopes). Such T-cells, which in our hands were able to act as regulatory cells in vitro by suppressing cytokine production by their own reactive LL37-specific T-cells, could be natural Tregs or induced Tregs. The evidence that LL37 is expressed not only in various organs but also in the thymus [115] could support the hypothesis that natural Tregs, specific for LL37, are present physiologically in HD. Expression of LL37 in the thymus may allow the deletion of LL37-specific T-cells by negative selection, or the selection of T-cells, which such specificity, endowed with regulatory/suppressive activity, which regulate the immune responses [116]. We cannot exclude that the cells of the T-cell clone T were obtained after an in vitro priming of the LL37-specific T-cells during the cloning procedure. However, these results are reported here to discuss the possible presence of T-cells specific for LL37 in vivo in HD. The limitation of this approach is that we ignore whether this is true for every HD. The experimental data presented here should, therefore, be viewed as an in vitro model of T-cell suppression, and not as a formal demonstration that both responder and regulatory T-cells specific for the autoantigen LL37 exist in vivo in HD. Tregs certainly play an important role in psoriasis, a disease in which they are dysfunctional. Many psoriasis treatments seem to exert an effect on Tregs, which in some cases acquire again their lost regulatory functions, as shown by our review of the literature. The presence of Tregs specific for LL37 and other T-cell autoantigens in psoriasis can be addressed in the same way reported here, by using peptide-MHC-tetramers. The mechanism of immune-suppression could be also addressed. The experiments are not easy to conduct, as one should perform HLA-typing of different HD and use several peptide-MHC-tetramers to stain the Tregs and identify the correct cells. One issue to address, using improved protocols like those presented here, is whether antigen-specific Tregs are altered in psoriasis and whether effectors autoreactive T-cells are derived from existing Tregs or are newly formed. Similar experiments can be useful to address whether the therapies used for psoriasis, mentioned here, can induce recovering of dysfunctional T-cells or imply a de novo induction of Tregs, or elimination of responder autoreactive T-cells. This review represents a narrative review. A search was conducted in the scientific literature on PubMed and Google Scholar, searching by using the following keywords: “Tregs in psoriasis” “Tregs and psoriasis treatments or psoriasis medications”, considering also the “related articles” in PubMed. We placed no time limit on the research performed and we included epidemiological studies, animal models, and in vitro culture models. Human Tregs were enriched from the peripheral blood of a healthy volunteer, whose cells show low proliferation to LL37 and to peptide P6 (see below) of LL37. Cells were obtained with full informed consent and ethical approval. PBMC were separated by Ficoll-Hypaque (Pharmacia Fine Chemicals, Uppsala, Sweden) density gradient centrifugation and total CD4 T cells were isolated using the CD4+CD25+ CD127dim/− Regulatory T Cell Isolation Kit II, human (Miltenyi Biotech, San Diego, CA, USA). Expansion of the cells was performed over three weeks in the presence of a high dose of hrIL-2 (300 U/mL) in complete medium containing RPMI 1640, 10% heat-inactivated human serum, HS, (Gibco -Thermo Fisher Scientific, Waltham, MA, USA), 2 mM L-glutamine, 10 U/mL penicillin and 100 μg/mL streptomycin (Sigma-Adrich, St. Louis, MO, USA), according to a modified protocol previouly published [115]. Tregs were screened for expression of markers CD38, CD25, CD127 by flow cytometry using a Gallios flow cytometer (Beckman Coulter, CA, USA), after isolation and after expansion. For the generation of the T-cell clone T, specific for LL37 P6, PBMC were stimulated (2 × 106 cells/mL) P6 of LL37, for 7 days and re-stimulated after 10 days with LL37-peptide P6 (10 μg/mL, seq: VQRIKDFLRNLVPRT), (synthesized by Biomatik, Kitchener, ON, Canada), and autologous-γ-irradiated-(30Gy)-PBMC in complete medium in the presence of human/recombinant(hr)IL-2 (Boehringer-Mannheim, Indianapolis, IN, USA). T-cell line-specificity was analyzed by using peptide-pulsed (10 μg/mL) autologous-irradiated PBMC (30Greys) or lymphoblastoid-cell lines (B-LCL) (150Greys) and peptide P6 or P4 (P4: IGKEFKRIVQRIKDF also derived from LL37) and control REV LL37 peptide, as previously described [50]. BrdU was added at day 4, T cells were analyzed by flow cytometry as described [50]. Cells were cloned by limiting dilution in Terasaki plates (Nunc Microwell, Sigma-Aldrich, St. Louis, MO, USA) in the presence of allogenic-irradiated (104) PBMCs activated by phytohaemagglutinin (1 μg/mL, PHA, Murex, Cedex, France) at 0.5 cells per well, as previously described [50]. 100 U/mL of hrIL-2 were added. Expanded clones reactivity and HLA restriction were analyzed using HLA-DR-matched-homozygous B-LCLs (ATCC, Virginia, USA), pulsed with peptide antigen or control peptide antigens, as previously described [50]. T-cell clone T (5 × 104 cells per well, in duplicates) was cultured with autologous-irradiated PBMCs (105 cells per well), and without or with peptide P6 at 10 μg/mL, with or without Tregs (ratio 1.1), overnight. The day after, we performed intra-cellular cytokine staining for IFN-γ on gated responder T-cell clone (T) and on Tregs (identified as CSFE-positive T-cells). Fluorescence was analyzed by a Gallios flow-citometer (Beckman Coulter, CA, USA), and FCS files were analyzed with FlowJo 7.5 software (TreeStar Inc., Ashland, OR, USA). For IL-2 detection in suppression assay, the T-cell clone was cultured with autologous-irradiated PBMCs with or without antigen P6 for two hours in the presence/absence of Tregs, as above and stained with the IL-2 Secretion Assay—Detection Kit (PE), human, (Miltenyi Biotech, Gaithersburg, MD, USA) for IL-2 and for anti-surface molecules (anti-CD3, anti-CD4) to detect IL-2 secretion in the presence or absence of Tregs (1:1 ratio). The latter were colored with CSFE (Sigma). For CSFE labeling of T regs (2 × 105 cells) were treated with 5 mM CSFE in the dark for 10 min in a small volume (100 mL) and, at the end of incubation, cold complete medium was added (four volumes), putting the cells in ice for 5 min. Cells were centrifuged and washed twice in complete medium (medium with 15% of HS). The following peptide-MHC-tetramers were used: P6-DR11-tetramer (called P6 tetramer) (P6: VQRIKDFLRNLVPRT) and P4-DR11 tetramer (P4: IGKEFKRIVQRIKDF), called control tetramer, both synthesized by TC Metrix, Epalinges, CH. Staining was performed at 37 °C for 40 min, followed by staining for CD4 (4 °C, 20 min). For the detection of tetramer-positive cells in Tregs and clone T, before flow cytometry acquisition, cells were labeled with 7-AAD to exclude dead cells and avoid unspecific staining. Mabs to CD4, CD3 conjugated with various fluorochromes (FITC, phycoerythrin (PE), peridinin-chlorophyll-protein (PerCp Cy5.5), or allophycocyanin (APC), were from BD Biosciences or eBiosciences (San Diego, CA). PE- or APC-CD38, PE-CD127, FITC-CD25 mabs were purchased from BD Biosciences, eBiosciences, Novus Biologicals (Littleton, CO, USA), R&D (Minneapolis, MN, USA). Appropriate isotype-matched controls were purchased from the same companies. PerCp-7-AAD was from BD Pharmingen. Foxp3 staining on cultured Tregs was done by intracellular staining, using an APC-anti-FoxP3 (PCH101). Staining was performed with an eBioscience Fix/Perm kit under the manufacturer’s directions. Data shown are means ± SEM, where indicated. Statistical comparison in Tregs suppression assays was performed using a two-tailed paired-samples Student’s t tests. Statistical significance was set at p < 0.05.

PMC10002351

Cristina Maxia,Michela Isola,Eleonora Grecu,Alberto Cuccu,Alessandra Scano,Germano Orrù,Nick Di Girolamo,Andrea Diana,Daniela Murtas

Synergic Action of Insulin-like Growth Factor-2 and miRNA-483 in Pterygium Pathogenesis

22-02-2023

pterygium,IGF-2,IGF-1R,miR-483,oxidative stress,IGF2 LOI

Pterygium is a multifactorial disease in which UV-B is speculated to play a key role by inducing oxidative stress and phototoxic DNA damage. In search for candidate molecules that are useful for justifying the intense epithelial proliferation observed in pterygium, our attention has been focused on Insulin-like Growth Factor 2 (IGF-2), mainly detected in embryonic and fetal somatic tissues, which regulate metabolic and mitogenic functions. The binding between IGF-2 and its receptor Insulin-like Growth Factor 1 Receptor (IGF-1R) activates the PI3K-AKT pathway, which leads to the regulation of cell growth, differentiation, and the expression of specific genes. Since IGF2 is regulated by parental imprinting, in different human tumors, the IGF2 Loss of Imprinting (LOI) results in IGF-2- and IGF2-derived intronic miR-483 overexpression. Based on these activities, the purpose of this study was to investigate the overexpression of IGF-2, IGF-1R, and miR-483. Using an immunohistochemical approach, we demonstrated an intense colocalized epithelial overexpression of IGF-2 and IGF-1R in most pterygium samples (Fisher’s exact test, p = 0.021). RT-qPCR gene expression analysis confirmed IGF2 upregulation and demonstrated miR-483 expression in pterygium compared to normal conjunctiva (253.2-fold and 12.47-fold, respectively). Therefore, IGF-2/IGF-1R co-expression could suggest their interplay through the two different paracrine/autocrine IGF-2 routes for signaling transfer, which would activate the PI3K/AKT signaling pathway. In this scenario, miR-483 gene family transcription might synergically reinforce IGF-2 oncogenic function through its boosting pro-proliferative and antiapoptotic activity.

Synergic Action of Insulin-like Growth Factor-2 and miRNA-483 in Pterygium Pathogenesis Pterygium is a multifactorial disease in which UV-B is speculated to play a key role by inducing oxidative stress and phototoxic DNA damage. In search for candidate molecules that are useful for justifying the intense epithelial proliferation observed in pterygium, our attention has been focused on Insulin-like Growth Factor 2 (IGF-2), mainly detected in embryonic and fetal somatic tissues, which regulate metabolic and mitogenic functions. The binding between IGF-2 and its receptor Insulin-like Growth Factor 1 Receptor (IGF-1R) activates the PI3K-AKT pathway, which leads to the regulation of cell growth, differentiation, and the expression of specific genes. Since IGF2 is regulated by parental imprinting, in different human tumors, the IGF2 Loss of Imprinting (LOI) results in IGF-2- and IGF2-derived intronic miR-483 overexpression. Based on these activities, the purpose of this study was to investigate the overexpression of IGF-2, IGF-1R, and miR-483. Using an immunohistochemical approach, we demonstrated an intense colocalized epithelial overexpression of IGF-2 and IGF-1R in most pterygium samples (Fisher’s exact test, p = 0.021). RT-qPCR gene expression analysis confirmed IGF2 upregulation and demonstrated miR-483 expression in pterygium compared to normal conjunctiva (253.2-fold and 12.47-fold, respectively). Therefore, IGF-2/IGF-1R co-expression could suggest their interplay through the two different paracrine/autocrine IGF-2 routes for signaling transfer, which would activate the PI3K/AKT signaling pathway. In this scenario, miR-483 gene family transcription might synergically reinforce IGF-2 oncogenic function through its boosting pro-proliferative and antiapoptotic activity. Pterygium is a common and sporadic disease of the ocular surface, described as a wing-shaped or triangular overgrowth of conjunctival mucosa extending on the cornea. Possible pathological consequences encompass corneal astigmatism and the obstruction of the visual axis, often contributing to visual loss and diplopia [1]. It is a chronic, degenerative, and hyperplastic disorder in which epithelial proliferation, goblet cell hyperplasia, angiogenesis, inflammation, elastosis, stromal plaques, and Bowman’s membrane dissolution are common features. Despite being considered a benign lesion, pterygium has been suggested to be a neoplastic-like growth disorder for its tumor-like traits (aggressive recurrence after removal and local invasiveness) and association with preneoplastic lesions [2,3]. Notably, pterygium is referred to as an “ophthalmic enigma” [4] because it represents a multifactorial disease as a result of a multitude of risk factors, including inflammation [5,6,7], antiapoptotic growth and proliferative mechanisms [8,9], angiogenic factors [10,11], oxidative stress and hypoxic ischemic injuries [12,13], viral infections [14], and extracellular matrix remodeling [15]. However, there is scientific consensus in the perception of pterygium as one of the most common sun-related eye diseases (ophthalmohelioses) [16], with UV-B playing a key role through inducing oxidative stress and is responsible for phototoxic DNA damage [17,18]. In search for candidate molecules that are useful for justifying the intense epithelial proliferation observed in pterygium, our attention has been focused on Insulin-like Growth Factor 2 (IGF-2), also known as Somatomedin A, a single-chain polypeptide hormone belonging to the Insulin-like Growth Factors’ (IGFs) system, which regulates metabolic and mitogenic functions [19,20]. Unlike Insulin-like Growth Factor 1 (IGF-1), preferentially expressed postnatal IGF-2 is mainly detected in embryonic and fetal somatic tissues and in adult liver, meninges, and choroid plexuses [21], even though traces of IGF-2 can be identified in adult cerebral spinal fluid, possibly related to neurogenesis in the subventricular and subgranular zone of the brain [20]. IGF-2 can specifically bind to three distinct receptors, namely, Insulin-like Growth Factor 2 Receptor (IGF-2R), the isoform A of Insulin Receptors (IR-A), and Insulin-like Growth Factor 1 Receptor (IGF-1R). Moreover, it interacts with IGF-2R, a nonsignaling receptor, which acts as a scavenger for circulating IGF-2 and limits its bioavailability via lysosomal internalization and degradation [22]. Furthermore, IGF-2 is a natural ligand for IR-A with affinity properties related to the receptor close to the same insulin, and for IGF-1R, competing with IGF-1 [19,23,24]. IGF-2/IGF-1R binding activates the phosphatidylinositol 3-kinase (PI3K)-AKT/protein kinase B (PKB) pathway that leads to the regulation of cell growth, differentiation, and the expression of specific genes [25]. Indeed, the activation of the PI3K/AKT cascade has been recently demonstrated in pterygium [26]. In humans, the IGF2 gene is localized on chromosome 11p15.5 and is parentally imprinted. The IGF2 epigenetic deregulation known as “Loss of Imprinting” (LOI) results in IGF-2 overexpression, which has been associated with neoplastic progression [24] and in the overexpression of IGF2-derived intronic miR-483 [27] abnormally expressed in different tumors. Moreover, miR-483 overexpression has been correlated with the promotion of cell proliferation in colorectal cancer [28] and in apoptosis inhibition in hepatocellular carcinoma [29]. Notably, in some benign and malignant tumors, IGF2 expression is directly regulated by Pleomorphic adenoma gene 1 (PLAG1) [30], inducing autocrine IGF-1R signaling that leads to the activation of PI3K/AKT downstream cascade, enhancing cell survival and proliferation [31]. Therefore, the main purpose of this study was to detect the coexistence of IGF-2, IGF-1R, and miR-483 and their synergic action in pterygium, aiming to support the proposition that considers pterygium a benign lesion, wherein the uncontrolled epithelial cell proliferation is the result of autocrine/paracrine cell activation and dysregulation of apoptosis. The results for IGF-2 IHC expression in pterygium and conjunctiva are illustrated in Figure 1. In our study, 45 (45/48, 93.6%) primary pterygium samples showed IGF-2 immunoreactivity. The staining was localized only in the cytoplasm of epithelial cells, and immunoreactivity was not detected in any nuclei (Figure 2A). In detail, in all samples, the staining was detectable in the basal and suprabasal layers of the epithelium, as well as in nine (9/45, 20%) pterygium specimens also in the superficial cells. In nine (9/11, 81.8%) of the normal conjunctiva samples, the immunostaining was absent (Figure 2B). No staining was shown in the negative control (Figure 2C). The results for IGF-1R IHC expression in pterygium and conjunctiva are represented in Figure 1. Thirty-four (34/48, 70.8%) pterygia exhibited IGF-1R immunoreactivity in the cytoplasm of epithelial cells located in the basal and middle layers, and, in 17 (17/34, 50%) of these samples, the staining covered a larger area in the superficial layers of the epithelium (Figure 3A). In most normal conjunctiva samples (10/11, 90.9%), the immunostaining for IGF-1R was not discernible (Figure 3B). No staining was present when the primary antibody was omitted or when an isotype control antibody was applied (Figure 3C). The relationship between IGF-2 and IGF-1R is shown in Table 1. In the group of pterygia with IGF-2-positive immunostaining, there were 34 (34/45, 75.6%) samples with IGF-1R expression, while the remaining 11 (11/45, 24.4%) specimens did not show any IGF-1R immunoreactivity. Moreover, only three (3/48, 6.3%) samples were negative for both IGF-2 and IGF-1R, and there were no IGF-2-/IGF-1R+ samples. A significant correlation between IGF-2 expression and IGF-1R staining in primary pterygium was demonstrated using Fisher’s exact test (p = 0.021) (Table 1), further confirmed through double immunofluorescence (IF) staining (Figure 4). All the recruited samples for the molecular analysis showed sufficient RNA quantity (>0.5 ng/mL) for the evaluation of the genetic expression. Gene sequences referring to IGF2 and miR-483 were both overexpressed in pterygium compared with normal conjunctiva patients by 253.2 and 12.47 folds (respectively), assuming that, in healthy subjects, the expression rate of the test gene/housekeeping gene is considered = 1 (Figure 5). Pterygium is often described as a multifactorial disease with UV-rays thought to play a key role in its pathogenesis. With the aim of disclosing information about the factors possibly involved in the occurrence and progression of pterygium, we evaluated the presence of IGF-2, a well-known factor implicated in embryonic development and tumorigenesis, due to its role in regulating cell proliferation, growth, migration, differentiation, and survival as a result of its paracrine/autocrine effect [32]. As a matter of fact, alterations in the autocrine loops IGF-2/IGF-1R often occur in human cancer, leading to the overactivation of the PI3K/AKT signaling pathway, which is responsible for impairment in cell proliferation and apoptosis [20]. Since IGF-2 dysregulation has been already shown in childhood and adult malignancies [33], as well as in benign tumors [34,35,36], it seemed reasonable to investigate the involvement of the IGF-2/IGF-1R pathway in primary pterygium, a benign tumor with premalignant features. In our study, the specific IHC overexpression ofIGF-2 was observed in 93.6% of the examined pterygium samples, with the immunostaining detected in the cytoplasm of epithelial cells being localized in the basal and suprabasal layers; moreover, in nine (9/45, 20%) samples, the IGF-2-positive cells spread across the superficial layers of the epithelium. Only 18.2% of the normal conjunctiva specimens displayed immunostaining for IGF-2. These results were further confirmed by our molecular analysis. IGF-2 upregulation in pterygium might be explained by a possible occurrence of the IGF-2 epigenetic dysregulation, known as “loss of imprinting”, during aging. Indeed, Yang et al. [37] suggested that the activation of the canonical Nuclear Factor Kappa-light-chain-enhancer of activated B cells’ (NF-κB) signaling pathway, induced by oxidative stress, is responsible for IGF2 LOI. In addition, NF-κB acts as a transcription factor for Hypoxia Inducible Factor 1 Subunit Alpha (HIF1A) [38], which, in turn, translocates into the nucleus and targets the IGF2 gene, resulting in enhanced cell proliferation. Since UVB exposure, responsible for irritative stimuli, induces oxidative stress-mediated NF-κB activation [39,40], it is definitely consistent with the notion that LOI-dependant IGF-2 overexpression may contribute to pterygium development. Likewise, in colorectal cancer tissues, it has been demonstrated that the hypomethylation of a “differently methylated region” (DMR), located adjacent to IGF2, might occur concurrently with IGF2 LOI, contributing to IGF-2 upregulation [41]. Several studies have indicated that IGF2 LOI drives the concomitant overexpression of the miR-483 family gene, embedded in the second intron of IGF2, which reinforces IGF-2 oncogenic function [42]. Since the upregulation of miR-483 has been associated to IGF-2 presence, the molecular analysis of qRT-PCR identified a well-defined IGF2 and miR-483 co-overexpression in pterygium in comparison with normal conjunctiva; in detail, the mean fold-increase expression values for IGF2 and miR-483 were 253.2 and 12.47, respectively. A possible explanation for the different expression levels between IGF2 and miR-483 might be offered by the genetic nature of the miR-483 family itself. Indeed, while hepatocellular carcinoma in miR-483 has been demonstrated to be co-expressed with its host gene IGF2 [43], in pterygium, there is a lack of knowledge about this interaction. miR-483 splicing produces miR-483-5p and -3p, which are processed from the same stem-loop precursor miRNA (pre-miR-483) encoded within the intron of the IGF2 locus. Interestingly, miR-483-5p has been shown to bind directly to the 5′ untranslated region (UTR) of the IGF2 primary transcript, thereby upregulating the expression of its host gene [44] and exerting its pro-proliferative effect. This peculiar miR-483-5p behavior might appear to be surprising, since miRNAs generally act by inhibiting their target gene. However, recent studies have shown that several miRNAs located in the nucleus can alternate into both the silent and activated status within the gene transcription [45]. Hence, this positive feedback loop might justify the more evident IGF2 expression than that observed for miR-483. Furthermore, in liver cancer [44], miR-483-3p has been found to act as an oncogenic and antiapoptotic factor by means of the inhibition of the p53 Upregulated Modulator of Apoptosis (PUMA), a pro-apoptotic protein that, under physiological conditions, blocks the antiapoptotic action of B-Cell CLL/Lymphoma 2 (BCL2) [46]. As a matter of fact, this hypothesis in pterygium has been supported by the finding of BCL-2 overexpression [47,48]. Based on the above evidence, we reason that the presence of miR-483 might be consistent with the uncontrolled epithelial cell proliferation and dysregulation of apoptosis associated to pterygium. However, further studies are required to verify the presence of both miRNAs (miR-483-3p, miR-483-5p), which would support the oncogenic function of IGF-2. The importance of the IGF axis in tumors was unambiguously claimed following the observation that embryonic fibroblasts lacking the IGF-1R are not susceptible of transformation [49]. In our study, 70.8% primary pterygia exhibited IGF-1R immunoreactivity in the cytoplasm of epithelial cells located in the basal and middle layers; moreover, in 17 (17/34, 50%) of these samples, the staining was enlarged to the superficial layers of the epithelium. Immunostaining for IGF-1R in normal conjunctiva samples was detected only in one (1/11, 9.09%) sample. Fisher’s exact test demonstrated a significant correlation between IGF-2 and IGF-1R overexpression (p = 0.021), since, within the group of 45 IGF-2+ samples, 34 (34/45, 75.6%) were also IGF-1R+, while the remaining 11 (11/45, 24.4%) specimens were IGF-1R-. As demonstrated by the results of the double IF analysis, IGF-2 and IGF-1R were co-expressed in the epithelial cells of the basal and middle layers. Therefore, we envision that IGF-2/IGF-1R co-expression in most pterygium samples would favor their interplay through the two different paracrine/autocrine IGF-2 routes for signaling transduction. The paracrine modality would activate the PI3K/AKT pathway, carrying out the downstream transcription of HIF1A, whose protein is a crucial factor involved in the occurrence and development of pterygium [11]. As a confirmation of that, HIF-1a stimulates the transcription of IGF2 and Vascular Endothelial Growth Factor A (VEGFA) genes [50] as a tumoral adaptive response to hypoxia [51]. The autocrine pathway would constantly supply the PI3K/AKT signal cascade, contributing to IGF-2 overexpression through a positive feedback loop mediated by PLAG1, as demonstrated in some benign and malignant tumors [30]. However, the finding of 11 IGF-2+/IGF-1R- samples led us to speculate that the IGF-1R-mediated PI3K/AKT pathway would not be the unique or the main pathway activated in pterygium. Indeed, according to Somers et al. [52], the numerical preponderance of the IGF-2-positive samples versus IGF-1R might arise from the existence of a subcellular localization of cytoplasmic IGF-2, whose function has not been well understood. Hence, in our data interpretation (Figure 6), the IGF-2/IGF-1R co-expression in most pterygium samples implies their interplay through the two different paracrine/autocrine IGF-2 routes for signaling transfer, activating the PI3K/AKT signaling pathway, which is probably responsible for the increased epithelial cell proliferation. In this scenario, the miR-483 gene family transcription might synergically reinforce IGF-2 oncogenic function through its pro-proliferative and antiapoptotic activity. Moreover, considering the well-known tangled crosstalk between oxidative stress and hypoxia, the take-home message of the present study might be the existence, in pterygium, of a sort of feed-forward loop started by UV-induced oxidative stress that would activate NF-κB, also stimulated by the IGF-2/IGF-1R-induced PI3K/AKT pathway; this cascade would carry out the downstream transcription of HIF1A, a hypoxic factor involved in the occurrence and development of pterygium, which, in turn, would trigger the expression of IGF-2, and would be no more silenced as a result of the NF-κB-induced LOI (Figure 6). Therefore, even if the molecular mechanisms involved in pterygium pathogenesis are still partly unknown, the results of this study would further support the hypothesis of pterygium as a neoplastic-like multifactorial disease, in which a network of oncogenic pathways join and mutually boost each other. Moreover, based on the findings that IGF-2 and miR-483 are a promising translational research area for tumor therapy, the pharmacological inhibition of mitogenic signaling by targeting these molecules could be a challenge for pterygium treatment. The study was performed on a group of 52 primary pterygium samples obtained from 42 male and 10 female patients, ranging in age from 36 to 80 years (mean 56.5 yrs years), who underwent surgery for pterygium removal at the Department of Surgical Science, Eye Clinic of the University of Cagliari. Most pterygia were located on the nasal side (33, 63%), and their morphology was clinically graded as atrophic (n = 14, 27%), intermediate (n = 29, 56%), or fleshy (n = 9, 17%) by means of pterygium translucency assessment, which included n = 33 (63%) inflamed and n = 19 (37%) quiescent lesions. Moreover, 15 normal epibulbar conjunctiva samples, used as control tissues, were excised from healthy donors (9 males and 6 females, ranging in age from 13 to 77 years), with no signs or symptoms of pterygium or a conjunctival disorder during strabismus or cataract surgery. The study was approved by the Independent Ethic Committee (CEI) of the Azienda Ospedaliero-Universitaria (AOU), Cagliari (Prot. NP/2022/5126), and written consent was obtained from all patients before the beginning of the study, in accordance with the World Medical Association Declaration of Helsinki. Patients did not receive any medication before surgery except for topical anesthetic; demographic and clinical characteristics of patients, available in all cases, are reported in Table 2. Among the 52 pterygium samples, 48 specimens were 10% formalin-fixed and paraffin-embedded; moreover, 11 epibulbar conjunctiva specimens, belonging to a group of 15 samples from healthy donors, were formalin-fixed, processed for paraffin embedding, and used as normal controls. Three microtome histological sections (6–7 µm thick) per sample were subjected to immunohistochemistry (IHC) for the demonstration of IGF-2 and IGF-1R antigens using the streptavidin-biotin alkaline phosphatase method, as previously described [7,53]. These were dewaxed, rehydrated, and rinsed in phosphate-buffered saline (PBS), pH 7.4. A heat-induced IGF-2 epitope retrieval (HIER) procedure was carried out via the immersion of the samples in a water bath-heated TRIS/EDTA buffer (TRIS 10 mM + EDTA 1 mM, pH 9.0) for 30 min at 95 °C, followed by gradual cooling for 20 min at room temperature (RT); conversely, IGF-1R epitope did not require any antigenic retrieval treatment. Furthermore, sections were treated for 45 min at RT with 10% nonimmune serum to block nonspecific binding and then incubated with the primary antibodies 1 h at RT. Table 3 includes reported sources, dilutions, time of incubation, and the details of the primary antibodies used for the IHC staining, according to the Resource Identification Initiative [54]. Biotinylated anti-mouse or anti-rabbit secondary antisera (1:200; Vector Laboratories, Burlingame, CA, USA) were incubated for 30 min at RT; then, the sections were treated with alkaline-phosphatase streptavidin (1:1000, Vector Laboratories, Burlingame, CA, USA) for 30 min at RT and were reacted using the SIGMA FASTTM Fast red substrate–chromogen system (SIGMA, St. Louis, MO, USA). All sections were carefully rinsed in PBS after each step and were finally counterstained using Carazzi haematoxylin and mounted in glycerol gelatin (SIGMA, St. Louis, MO, USA). Positive and negative controls were run simultaneously to evaluate reaction specificity. Archival sections of normal human skin and human seminoma were used as known positive controls for IGF-2 and IGF-1R, respectively. Negative controls were carried out by omitting the primary antibody or by replacing the primary antibody with an isotype-matched antibody. A Zeiss Axioplan2 microscope (Carl Zeiss Vision, Hallbergmoos, Germany) equipped with the following objectives: 20×/0.45 Zeiss Achroplan; 40×/0.75 Zeiss Plan-Neofluar, 63×/1.40 oil immersion Zeiss Plan-Apochromat, was used for the analysis of immunolabelled slides, while image capture was performed using a Lumenera Infinity 3-1URC camera (1.4 megapixels; Lumenera Corporation, Ontario, Canada) and Infinity Capture 6.3.0 software (Lumenera Corporation, Ontario, Canada). The figure panels’ creation and a slight adjustment of brightness and contrast were carried out using Adobe Photoshop CS3 Extended (ver. 10.0, Adobe Systems Incorporated, CA, USA). Three experienced observers (CM, EG, and DM) independently evaluated the immunoreactivity in a triple-blind fashion, as previously described [18]. In each of the three sections per sample, the percentage of epithelial immunoreactive cells was scored in four to six randomly chosen microscopic fields (×40 original magnification), covering almost the whole field of each section per sample. Furthermore, in all samples, epithelial cellular compartments (nucleus, cytoplasm, or both) were analyzed for the localization of the immunoreactivity. IGF-2 and IGF-1R cut-off was set at 10%, which accounts for the percentage of stained epithelial cells per positive sample. Samples were divided into four groups, i.e., IGF-2-positive, IGF-2-negative, IGF-1R-positive, and IGF-1R-negative. A simultaneous procedure was used for the double staining of IGF-2 and IGF-1R, as previously described [55]. Following deparaffinization, rehydration, antigen retrieval, and blocking of nonspecific binding, sections of pterygium and conjunctiva were incubated for 1 h at RT with a mixture of mouse monoclonal anti-human IGF-2 (clone 8H1, 1:100, ThermoFisher Scientific, Waltham, MA USA 02451) and rabbit polyclonal anti-human IGF-1R (1:50; Abcam Cambridge, CB2 0AX, UK) primary antibodies. Alexa Fluor 594 donkey anti-mouse IgG (H+L) and Alexa Fluor 488 donkey anti-rabbit IgG (H+L) (1:200, Invitrogen Life Technologies, Paisley, UK) secondary antibodies were used for immunofluorescence detection. The sections were mounted in VECTASHIELD® Antifade Mounting Medium with 4′,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories) to visualize nuclear details. Immunofluorescence-labelled sections were imaged using a Zeiss Axioplan2 microscope (HBO 100 illuminator; mercury vapor, short arc lamp) equipped with the appropriate filter sets to distinguish the chosen fluorochromes and the above-mentioned objectives and digital camera. The displayed figure panel was set up using Adobe Photoshop. The immunohistochemical results were analyzed using Fisher’s exact test. Data were computed with IBM® SPSS® Statistics 21.0. The tests used were two-tailed. A p value ≤ 0.05 was considered statistically significant. Among the group of 52 pterygium and 15 conjunctiva samples, 4 pterygium and 4 epibulbar conjunctiva specimens, respectively, were addressed in the molecular analysis. Immediately after the removal, each sample was placed inside an Eppendorf® 2-mL tube containing 100 µL of RNAlater® solution and then frozen at −80 °C until RNA extraction. The samples stored in the RNAlater® solution were thawed and then used for RNA extraction by using a RNeasy® mini-Kit (QIAGEN GmbH, Hilden, Germany), following the manufacturer’s instruction. The quality of the RNA extract was evaluated using a NanoDrop® Instrument (ThermoFisher Scientific, Waltham, MA USA), according to the manufacturer’s directions. The optimal requirements for the A260/A280 ratios were 1.8–2.2, while the requirements for the A260/A230 ratios were >1.7. Regarding the concentration required for Real-Time quantitative PCR (RT-qPCR) amplification (threshold cycle CT > 35), the samples showed a concentration > 0.5 ng/mL, and the RNAs were resuspended in 30 μL of nuclease-free H2O and kept at −80 °C until use. Relative expression of miR483 and IGF2 genes was performed through an RT-qPCR procedure using the 2−∆∆Ct method [56]. To avoid oligonucleotide dimers formation and self-complementarity, and to set the correct annealing temperatures of the RT-qPCR, the oligos were designed using the Oligo program vers. 6 (MedProbe, Oslo, Norway) and “UNAFold Web Server” programs (http://www.unafold.org/mfold/applications/dna-folding-form.php) (accessed on 10 October 2022) using the procedures previously described [57]. The theoretic melting temperatures of the different PCR amplicons (Tms) were calculated using module 2 of the UNAFold Web Server. Oligonucleotide sequences and their thermodynamic parameters are described in Table 4; all primer pairs were checked for their efficacy on the serial dilutions of cDNA. The RT-PCRs showed an efficiency range from 0.95 to 0.98 [58]. Retro-transcription was carried out by using the SuperScript® VILO™ cDNA Synthesis Kit (Invitrogen, ThermoFisher Scientific, Waltham, MA USA), following the manufacturer’s instructions. RT-qPCR was performed in a Light Cycler apparatus (Roche, Basel, Switzerland) by using the LightCycler® FastStart DNA Master HybProbe Kit (Roche Molecular Systems, Inc., Rotkreuz, Switzerland). In 20 µL of the PCR reaction mix, a mixture of 2 µL of cDNA extract, 5 pmol of each primer, 10 pmol of TaqMan probe, and 2 µL of SYBR Green solution (Table 4) was present. The PCR amplification conditions consisted of (i) 30 s (s) of denaturation at 95 °C; (ii) 40 cycles of 1 s at 95 °C, 10 s at 52 °C, and 8 s at 72 °C. The temperature transition rate in the denaturation and annealing steps was 20 °C/s, while, during the polymerization step, it was 5 °C/s. Fluorescence was recorded at a wavelength between 510 nm and 550 nm at the end of each PCR cycle. Three distinct biological replicas were obtained, and quantitative data were expressed as mean ± SD. Changes in gene expression above 2 or below 0.5 were considered significant.

PMC10002364

Raj Bose,Stefan Spulber,Sandra Ceccatelli

The Threat Posed by Environmental Contaminants on Neurodevelopment: What Can We Learn from Neural Stem Cells?

22-02-2023

developmental neurotoxicity,neural stem cells (NSCs),neurogenesis,molecular mechanisms,epigenetic modifications,methylmercury,per- and polyfluorinated substances (PFAS)

Exposure to chemicals may pose a greater risk to vulnerable groups, including pregnant women, fetuses, and children, that may lead to diseases linked to the toxicants’ target organs. Among chemical contaminants, methylmercury (MeHg), present in aquatic food, is one of the most harmful to the developing nervous system depending on time and level of exposure. Moreover, certain man-made PFAS, such as PFOS and PFOA, used in commercial and industrial products including liquid repellants for paper, packaging, textile, leather, and carpets, are developmental neurotoxicants. There is vast knowledge about the detrimental neurotoxic effects induced by high levels of exposure to these chemicals. Less is known about the consequences that low-level exposures may have on neurodevelopment, although an increasing number of studies link neurotoxic chemical exposures to neurodevelopmental disorders. Still, the mechanisms of toxicity are not identified. Here we review in vitro mechanistic studies using neural stem cells (NSCs) from rodents and humans to dissect the cellular and molecular processes changed by exposure to environmentally relevant levels of MeHg or PFOS/PFOA. All studies show that even low concentrations dysregulate critical neurodevelopmental steps supporting the idea that neurotoxic chemicals may play a role in the onset of neurodevelopmental disorders.

The Threat Posed by Environmental Contaminants on Neurodevelopment: What Can We Learn from Neural Stem Cells? Exposure to chemicals may pose a greater risk to vulnerable groups, including pregnant women, fetuses, and children, that may lead to diseases linked to the toxicants’ target organs. Among chemical contaminants, methylmercury (MeHg), present in aquatic food, is one of the most harmful to the developing nervous system depending on time and level of exposure. Moreover, certain man-made PFAS, such as PFOS and PFOA, used in commercial and industrial products including liquid repellants for paper, packaging, textile, leather, and carpets, are developmental neurotoxicants. There is vast knowledge about the detrimental neurotoxic effects induced by high levels of exposure to these chemicals. Less is known about the consequences that low-level exposures may have on neurodevelopment, although an increasing number of studies link neurotoxic chemical exposures to neurodevelopmental disorders. Still, the mechanisms of toxicity are not identified. Here we review in vitro mechanistic studies using neural stem cells (NSCs) from rodents and humans to dissect the cellular and molecular processes changed by exposure to environmentally relevant levels of MeHg or PFOS/PFOA. All studies show that even low concentrations dysregulate critical neurodevelopmental steps supporting the idea that neurotoxic chemicals may play a role in the onset of neurodevelopmental disorders. Neurodevelopmental disorders, such as intellectual disabilities, attention-deficit/hyperactivity disorder (ADHD), dyslexia, autism spectrum disorder (ASD), as well as schizophrenia, bipolar disorder, and depression, are increasing globally causing immense suffering and huge costs to society [1]. Several of these pathological conditions may have a genetic origin, but an increasing number of studies suggest that other factors, including maternal stress, infections, malnutrition, low birth weight, and exposure to toxicants, may play a role in their etiopathogenesis [2]. More than 200 chemicals belonging to different classes, including metals, persistent organic pollutants, organic solvents, and pesticides, have been identified as neurotoxicants [2]. Experimental and epidemiological data indicate that developmental exposure to certain environmental contaminants poses a threat to the developing brain and may lead to neurodevelopmental disorders. Specific and well-organized cellular and molecular processes, including proliferation, migration, differentiation, myelination, and synaptic pruning, characterize the development of the nervous system. Any incident disrupting the sequence of developmental steps can lead to permanent or transient structural and functional losses. The impact that harmful stimuli, such as neurotoxicants, exert on the developing nervous system depends on the timing of exposure that may coincide with region-specific windows of susceptibility. Neurodevelopmental damages may not be evident for a long time (silent neurotoxicity) up until different challenges, including aging, trauma, or exposure to toxicants, disclose and even build up on the existing cellular or biochemical damage. In addition, epigenetic modifications may occur, making the alterations heritable (see [3]). We are exposed to a large and increasing number of chemicals and the lack of information regarding their neurodevelopmental toxic potential has generated a growing concern for public health. Experimental animal models and tests have been critical in toxicology to provide fundamental information about potential neurotoxicants. However, animal studies have intrinsic limitations, including variations across species, labor- and time-intensive procedures, as well as ethical issues [4,5]. Therefore, in vitro assays with cell lines and primary cells, especially those of human origin, have turned out to be good alternatives to live animal experiments for developmental neurotoxicity studies [6]. Chemical neurotoxicity investigations using neural stem cells (NSCs) have shown that the use of these cells provides unique information for the identification of neurodevelopmental toxicants and their mechanism of action. Here, we review original articles on mechanistic data generated using NSCs derived from mice, rats, and humans as experimental models. We specifically focus on the well-known neurotoxicant methylmercury (MeHg) and on two compounds belonging to the larger class of per- and poly-fluoroalkyl substances (PFAS), namely perfluorooctanesulfonic acid (PFOS) and perfluorooctanoic acid (PFOA). NSCs are generated from the embryonic neuroectoderm, which eventually generates most cells in the central and peripheral nervous systems. The Sox gene family members and Otx2 are the earliest markers for NSCs, which are also known as neuroepithelial cells. The definitions NSCs and neural progenitor cells (NPCs) refer to undifferentiated cells with specific characteristics. NSCs are multipotent and have the ability to differentiate into neurons, astrocytes, and oligodendrocytes; they are self-renewing and maintain multipotency across an indefinite number of divisions. Instead, the potential of NPC is more restricted [7]. Both NSCs and NPCs are present throughout the development of the nervous system in the ventricular zone (VZ) and the subventricular zone (SVZ), and embryonic neurogenesis is essential for the formation of specific spatial organization, neuronal networking, and maturation of the nervous system. Signaling molecules in the microenvironment allow the maintenance, proliferation, and neuronal fate commitment of local stem cell populations throughout development. For experimental purposes, primary NSCs have been isolated from various regions of the embryonic/fetal nervous system, such as the olfactory bulb, subventricular zone, hippocampus, cerebellum, cerebral cortex, and spinal cord. In the adult rodent brain, NSCs located in the subventricular zone of the lateral ventricle and the subgranular layer of the hippocampal dentate gyrus are active and generate new neurons, astrocytes, and oligodendrocytes continuously throughout life. These newly generated neural cells play a crucial role in the maintenance of learning and memory [8]. The persistence and the ability to produce new neurons have not been fully clarified in the adult human brain. The integration of newly generated neurons into pre-existing neuronal networks is essential for the function and the plasticity of neural circuits. Recent evidence showed that adult brain injuries induce the generation of cells characterized initially as specialized astrocytes. However, when cultured, these cells demonstrate NSC properties [9], such as multipotency and self-renewal. Thus, toxicity studies using NSCs are highly relevant not only for the developing brain but also for the adult nervous system. We briefly discuss below several NSC models that have been used to investigate known and suspected neurotoxicants. The C17.2 cells, from an immortalized NSC line derived from a murine neonatal cerebellum, have been widely used for understanding cell fate and differentiation of neural progenitors. Although these cells are transformed and have restrictions to generate functional neurons, they maintain the capacity to follow developmental cues. For example, they can generate fully functional neurons when transplanted into the mid-embryonic mouse brain during development but not at later developmental stages, when gliogenesis is predominant. Primary cultures of NSCs have been successfully derived from the telencephalon, striatum, and hippocampus of rat embryos and retain multipotential properties [10]. These cells can be cultured as a monolayer (2D) on coated surfaces, or as three-dimensional (3D) on non-coated surfaces; the stemness is maintained by the addition of fibroblast growth factor (FGF), and/or epidermal growth factor (EGF). Upon removing EGF and FGF from the culture medium, the cells can differentiate into the major cell types found in the cerebral cortex, striatum, and hippocampus, including pyramidal and interneurons, astrocytes, and oligodendrocytes, as well as smooth muscle cells. Primary cultures of adult NSCs (aNSCs) can be obtained from the anterior portion of the lateral walls of the lateral ventricles of adult rats [11]. These cells are cultured as 3D on non-coated surfaces in the presence of EGF for generating neurospheres within a week. Thereafter, mechanical or enzymatic dissociation of neurospheres produces single cells or smaller size spheres, which are either passaged for de novo neurosphere propagation or plated onto coated surfaces for monolayer culture, and further differentiated into neurons, astrocytes, and oligodendrocytes in the absence of EGF. Human NPCs (hNPCs) have been produced from the fetal forebrain at gestational weeks 6–12 [11,12], cultured in the presence of FGF and EGF, and maintained in suspension as proliferating neurospheres [13]. When hNPCs are cultured on coated surfaces as a monolayer or as neurospheres without growth factors, they give rise to the major cell types found in the adult brain. Human NSCs generated from umbilical cord blood (HUCB-NSC cell line) can differentiate into mature neurons expressing functional voltage- and ligand-gated ion channels and are capable of establishing functional networks. In addition, HUCB-NSC cells can also generate major glial cell lineages, such as astrocytes and oligodendroglia [14]. Human iPSC-derived neuroepithelial-like stem cells, such as the AF22 cell line and NES cells are cultured on a coated plate as a monolayer in the presence of EGF and FGF for propagation. They are characterized by the long-term self-renewing capacity and the rosette-like growth pattern after 8–12 days in the presence of growth factors. These cells give rise to neurons and glia when they are induced to differentiate by growth factor withdrawal. In addition, they retain neuro- and gliogenic potential even after long-term proliferation [15]. A recent advance in human iPSC technology is the generation of brain organoids that recapitulate complex processes occurring during embryonic development, and express cellular diversity, networking, and compartmentalization. A further step forward is the generation of vascularized brain organoids that offer unprecedented possibilities to understand the complexity of nervous system damage and disease development [16]. In our daily lives, we are all exposed to a variety of chemicals present in products we use, in the food that we eat, or in our inner or outer environment (Figure 1). Exposure to chemicals poses a high risk to vulnerable groups including pregnant women, children, and the elderly. In particular, exposure during critical periods of development in prenatal and early life stages can predispose to disease later in adult life. The three neurotoxic chemical contaminants that we consider in this review are among the ones detected in human blood at levels that vary depending on geographic location, diet, and working environment. They can cross the placenta barrier, thus being of particular concern for developing organisms. The global environmental and food contaminant MeHg is mostly generated from inorganic mercury by the methylating activity of anaerobic bacteria in water sediments [17]. Contaminated aquatic food is the main source of MeHg exposure for human populations [17]. Several mechanisms of toxicity have been identified as responsible for its toxic effects including interactions with sulfhydryl groups of thiol-containing compounds, thereby targeting peptides and proteins containing cysteine and methionine; mitochondrial function impairment; perturbation of intracellular Ca2+ homeostasis; increased generation of reactive oxygen species (ROS) [18]. All these intracellular alterations have harmful effects on the nervous system, particularly during development. The nervous system is known to be the major MeHg target organ in both animals and humans particularly during prenatal life. By crossing the placenta, as well as the blood-brain barrier, MeHg induces adverse effects on different entities depending on the time and duration of exposure [19]. Fetuses can be heavily affected even when mothers do not show signs of toxicity, which can be explained by the fact that MeHg-fetal blood levels are about 1.7–1.9 times higher than those measured in maternal blood [20]. The developmental neurotoxic effects of MeHg were first identified in the 1950s in Japan. Wastewater contaminated with mercury had been discharged by a chemical factory into Minamata Bay for decades. The accumulated mercury entered the aquatic food chain as MeHg and contaminated the local populations that had a fish-based diet. The most severe neurotoxic effects on humans were observed in children of women who had eaten MeHg-contaminated fish during pregnancy. Surviving children exhibited various neurological clinical signs including ataxia, blindness, spasticity, impairment of motor skills, and variable degrees of mental retardation depending on the severity of the prenatal exposure [21]. Thanks to major efforts to reduce the release of mercury in the environment, the MeHg aquatic contamination has been considerably reduced [22]. Nevertheless, the adverse long-term consequences that developmental exposure to low levels of MeHg (via the maternal diet) may have on the nervous system are a matter of great worry (see [23]). Experimental animal studies and epidemiological investigations of seafood-eating populations have identified behavioral alterations and decreased cognitive abilities linked to exposure to prenatal MeHg (see [24]). Currently, the general population gets exposed to MeHg by eating contaminated fish seafood, and marine mammals (Figure 1), and the concentration of MeHg in cord blood directly correlates with maternal fish intake even at low maternal fish consumption [22]. Therefore, in several countries, women who are pregnant or planning to become pregnant, nursing mothers, and young children are recommended to exclude the consumption of sea food containing high levels of MeHg [25]. The concentration of MeHg In the umbilical cord blood at birth provides an estimation of the level of exposure during development. Levels in the range of 1.5–3.5 µM can be detected in populations from areas exposed to heavy industrial pollution, and such high MeHg levels are mostly due to manmade outbreaks [22]. In most populations, the average concentration of MeHg in the cord blood is below 10 nM (~2.2 µg/L) and reaches 100 nM (~21.5 µg/L) if the diet is based primarily on seafood (see [26,27]). A safety maximum limit for cord blood was defined at 5.8 µg/L (~27 nM) [28]. Therefore, there is genuine concern regarding the potential developmental neurotoxic effects of MeHg. PFAS are synthetic chemicals used in commercial and industrial products, including water and oil repellants for paper, packaging, textile and leather goods, industrial surfactants, fire-fighting foams, food packaging, and non-stick cookware, due to their unique physicochemical characteristics [29,30,31,32,33,34]. While PFOS and PFOA are banned in the European Union and the United States, the most common PFAS contaminants are still found in the environment because of the long-term persistence of those chemicals. Currently, the main source of exposure to PFAS, such as PFOS and PFOA in the general population, is via intake of contaminated foods, beverages, drinking water, and inhalation (Figure 1). Infants are exposed to these chemicals via breastfeeding while fetuses are exposed via maternal cord blood since both PFOS and PFOA can pass through the placenta as well as the blood-brain barrier (BBB). For example, evidence suggested that longer periods of breastfeeding resulted in higher blood PFAS levels in the infant [35]. In the USA, at least six million people can be exposed to PFOS and PFOA at a level of 70 ng/L [36], which is higher than the levels of EPA health advisories. A recent cohort study reported that human maternal serum concentration of PFOS (4.4–6.0 ng/mL) and PFOA (11.2–15.6 ng/mL) [37] and prenatal exposure to PFOS was significantly associated with hyperactivity and hyperactive-type ADHD in young school-aged children (mean PFOS concentration in maternal serum 12.8 ng/mL) [37]. Another recent study reported that prenatal exposure to PFOA was associated with an increased risk of ASD and ADHD in children where the concentration of PFOA is in the range of 1.47–2.17 ng/mL in maternal plasma [38]. Several neurotoxic mechanisms have been proposed, but three have received particular attention: disruption of Ca2+ homeostasis; interference with neurotransmitter signaling; and neuroendocrine disruption (reviewed in [39]). Given the developmental neurotoxicity potential and the persistence in the environment, PFAS exposure is a matter of major concern. The neurotoxic outcomes of exposure to environmental contaminants, such as apoptosis or alterations in proliferation and differentiation, can have several mechanisms, some of which are shared by different compounds. The identification of key events at cellular and molecular levels is critical to provide support for plausible adverse nervous system outcomes and also to indicate possible preventive and therapeutic strategies. In the following sections, we will focus on neurotoxic endpoints and underlying mechanisms separately. The main findings of the studies reviewed here are summarized in Table 1 (MeHg) and Table 2 (PFOS or PFOA). Very low toxic effects (IC5) for cell viability have been reported following exposure to 10 nM MeHg for 10 days, and 100 nM for 5 days in human NPCs derived from H9 human embryonic stem cells [40]. The lowest concentration inducing alterations in cell viability (LOAEC) after 24 h exposure to MeHg was 3000 nM for neurospheres generated from hiPSC-derived hNPCs, and 1000 nM in neurospheres derived from primary hNPCs, while EC50 was similar in the two models [41]. In hiPSC-derived organoids, the lowest concentration reducing cell viability following 7 days of exposure to MeHg was 10 µM [42]. Proliferating NSCs of rodent origin exposed to MeHg 25 nM or higher concentrations for >24 h activate caspase and calpain-dependent pathways [43,44,45,46]. Similarly, human NSCs derived from umbilical cord stem cells undergo apoptosis after exposure to 50 nM MeHg for 48 h [14]. Exposure to 25 nM MeHg for 24 h induced cell death in human fetal or embryonic NPCs [47,48,49] and exposure to 50 nM for 24 h was reported to alter mitochondrial biogenesis and increase ROS production in human cortical neural progenitor cells (ReNcell CX) [48]. However, exposure to higher concentrations of MeHg (250 nM for 24 h) [44] was required to induce apoptosis in C17.2 cells (mouse neural progenitor cell line). Interestingly, exposure to either 100 or 500 nM MeHg for 16 h [50] in adult NSCs isolated from male or female mice demonstrated sex-specific effects, namely that cells derived from females were less sensitive than cells from males. Caspase-dependent apoptosis induced by exposure to 100 nM MeHg for 48 h in mouse NSCs can be prevented by antioxidant treatment (NAC or alpha-tocopherol), and promoted by the inhibition of GSH synthesis [51]. Similarly, MeHg-induced cell death can be reduced by caspase [44,52] and calpain [44] inhibitors. Three-dimensional cell culture systems require a considerably higher concentration of MeHg or PFOS/PFOA to elicit similar effects as in the 2D culture of NSCs (see Table 1). Embryoid bodies appear less sensitive to MeHg than a monolayer culture of cells, and cell death has been documented after exposure to >100 nM for 14 days [53], 200 nM for 11 days [54], or 1000 nM for 16 h [55]. Exposure for 48 h to 100 nM PFOS showed a significant increased percentage of apoptotic nuclei only in primary cultures of rat cortical NSCs [56]. The analysis of cytotoxicity in C17.2 cells exposed to PFOS in the range of 25–200 nM for 48 h demonstrated a dose-dependent decrease in the number of cells [57]. In contrast, 3D cultures of mESC did not show a significant morphological change after 7-day PFOS exposure at a concentration of 10 µM [58]. hiPSC exposed to PFOS or PFOA (≥50 nM) for 24 h did not exhibit significant alterations in cell viability, but alterations in cell-specific differentiation, such as pancreatic, endocrine, and cardiomyocyte differentiation [59,60]. Cell viability and ROS generation assays suggested that PFOS did not induce cytotoxicity in mESCs at the concentrations tested (≤10 µM) for 7 days, but disrupted the expression of neural developmental [45,61] markers without affecting the proliferation of the differentiating cells [58]. Similarly, rat hippocampal NSCs exposed to PFOS or PFOA (≥200 nM) for 24 h did not show significant morphological changes and PFOS even increased NSC viability [62], suggesting that cell culture conditions may modulate PFOS or PFOA effects. Exposure to sub-toxic concentrations of MeHg does not reduce cell viability but it decreases cell proliferation. Monolayer cultures of primary NSCs derived from rat embryos (E14.5) undergo cell cycle arrest following 48 h exposure to MeHg at ranges from 2.5 nM to 10 nM [45,50]. Recently, Yuan and colleagues reported that 2D culture of murine NSCs derived from E12 exhibited a significant decreased proliferation after exposure to 0.25 nM, but not at higher doses (0.5–5 nM) [63]. In addition, MeHg-induced reduction of proliferation was associated with an upregulation of GSK-3b and CDK inhibitors p16 and p21; a significant increase in cyclin E degradation; and an alteration of cytoskeleton dynamics [63]. NSCs exposed to MeHg go through cellular senescence, as shown by the alteration of Bmi, Hmga1, and Hp1g gene expression [61]. Similar effects were shown in human NPCs (ReNcell CX) after exposure to 10 nM MeHg for 24 h, which were linked with the upregulation of p16, p21, and p53 [49]. NSC proliferation has also been evaluated after exposure to PFOS by staining with EdU, a thymidine analogue. Cells exposed to 25 or 50 nM PFOS for 48 h showed a significant decrease in the number of EdU-positive cells as compared to control cells [56]. The analysis of cell proliferation using the CCK-8 assay indicated that 50 nM of PFOS impaired the proliferation of C17.2 cells in both a time- and concentration-dependent manner [57]. This study also showed that the down regulation of the GSK-3β/β-catenin axis, and its target genes, cyclin D1, C-myc, Cox-2, and survivin, played a crucial role in the PFOS-induced reduction of cell proliferation. In contrast, the proliferation of NSCs derived from rat embryonic hippocampus has been shown to be increased after exposure to PFOS doses ranging from 1µM to 10µM for 48 h exposure. Conversely, PFOA exposure induced no alteration in NSC viability or proliferation [58,62]. mESC exposed to 10 µM PFOS for 7 days did not show the altered intensity of alkaline phosphatase (AP) staining compared to the control, meaning that self-renewal ability was not affected. A study using differentiating C17.2 cells for mRNA expression profiling using microarrays with genome-wide coverage identified a set of 30 mRNA species strongly associated with neuronal differentiation, out of which 14 displayed significant alterations after exposure to 90 nM MeHg for 10 days [64]. The decrease in the number of neurons and altered neurite outgrowth confirmed the pattern of alterations suggested by the mRNA expression profile [64]. In NSC culture, the absence of growth factors (FGF or EGF) induces spontaneous differentiation. Rat NSCs exposed to MeHg at ranges from 2.5 nM to 10 nM for 48 h exhibited reduced neuronal differentiation [44,63,65,66]. Tian et al. reported that a sub-toxic concentration of MeHg (10 nM) induced alterations of hippocampal neurons and astrocyte differentiation that could be reversed by antioxidant treatment with polysaccharides from Lycium bararum [66]. These results suggest that oxidative stress may regulate NSC differentiation. Exposure to 10 or 25 nM MeHg for up to 12 days in human NSCs was shown to decrease neuronal differentiation [47,67,68]. Interestingly, Yuan et al. demonstrated that NSCs isolated from mouse embryos exposed to 0.25 nM for 3 days had an enhanced neuronal differentiation but a reduced number of precursor cells [63]. Similarly, exposure to 10 nM MeHg for 4 days reduced hNPCs neuronal differentiation that was associated with decreased expression of BDNF [47]. Using embryoid bodies derived from mouse ESC, Theunissen et al. [69] have shown that exposure to a subtoxic concentration of MeHg (25 nM for 8 days) reduced neuronal outgrowth and was accompanied by an upregulation of late neuronal differentiation genes and a downregulation of early differentiation genes. We reported that MeHg-induced alteration of NSC differentiation is mediated by ERK 1/2 dephosphorylation and Notch signaling pathways. In the same model, we also demonstrated that MeHg-induced effects on neuronal differentiation could be rescued by the metalloprotease inhibitor GM-6001, which prevented cleavage of the Notch receptor [65]. In addition to neuronal differentiation, exposure to 10 nM MeHg for 2 days increased astrocytic differentiation in human NES cells, which could be reversed by DAPT, a gamma-secretase inhibitor that blocks extracellular Notch cleavage [68]. Interestingly, the positive correlation of MeHg-induced alterations of astrocyte differentiation with NES cells derived from a patient with a mutation in the NRXN1 gene linked to autism spectrum disorder [68]. Human cells exposed to a range of concentrations from 1 to 50 nM of MeHg show decreased neuronal migration and neurite outgrowth [47,70,71,72]. Remarkably, MeHg-induced alterations were enhanced in cells of male origin [47]. Similarly, exposure to 500–1000 nM MeHg for 48 h reduced neuronal migration in the 3D (neurospheres) culture of human NPCs [6,73], and the effect was associated with a reduction in ERK 1/2 phosphorylation. Neurospheres generated from hNPCs exposed to MeHg for 3 days displayed reduced neuronal migration from 100 nM [74]. These findings are supported by the altered neuronal migration and positioning of cerebrocortical neurons following in vivo administration of MeHg (0.1 or 1 mg/kg/day i.p. GD11-21) [75]. However, no change in the proliferation or differentiation of NSC has been reported in the same experimental model [75]. In rat NSCs we have shown that exposure to PFOS at doses ranging from 25 to 50 nM for 48 h neuronal differentiation increases while the proportion of nestin-positive cells decreased [56]. We also demonstrated that nanomolar concentrations of PFOS increase neurite outgrowth and significantly increase the number of CNPase-positive cells (oligodendrocytes), whereas astrocyte differentiation is not changed [56]. In contrast, Pierozan and colleagues [62] demonstrated that exposure of embryonic hippocampal NSCs to 10 µM of different PFAS for 24 h led to altered neuronal cell body morphology, but had no effects on neurite number and length, or on the number of branches per cell [62]. In addition, they found no alteration in oligodendrocyte and astrocyte differentiation in NSCs exposed to PFAS. This result may depend on cell type and culture conditions. Using monolayer culture of mESC, Yin et al. demonstrated that exposure to 1 nM up to 10 µM PFOS for 9 days exerted a general inhibitory trend in a dose-dependent manner on the expression of the pluripotency marker Nanog and on neural marker genes such as Sox1, Sox3, Nestin, Pax6, and Map2 [58]. Similarly, in 3D cultures of mESCs, exposure to PFOS for 9 days affected the differentiation process [58]. Mature, terminally differentiated neurons display spontaneous electrical activity and engage in the formation of synapses and networks in culture. Altered synaptogenesis as a neurotoxic effect can be illustrated by changes in the number and morphology of synapses based on the presence of postsynaptic proteins, such as PSD95 or neurotransmitter receptors; alternatively, it can be illustrated by presynaptic markers such as vesicular transporters for glutamate, vGLUT-1, or GABA, vGAT. In a recent study, synaptogenesis has been highlighted as the most sensitive endpoint by mathematical modeling of neurotoxicity in hiPSC-derived neurons [72]. Neurite outgrowth and the expression of synaptic markers were decreased following exposure to as low as 0.25 nM MeHg for 3 days or 0.05 nM MeHg for 14 days [72]. Both PFOS and PFOA have been shown to increase neuronal excitability following acute exposure (10 or 100 µM), but inhibit synaptogenesis and synaptic signaling upon chronic exposure (10 µM) in primary hippocampal neurons [76] and differentiated hiPSC [77]. In vivo, administration of the equivalent of 21 micromol/kg bodyweight PFOS or PFOA (11.3 and 8.7 mg/kg bodyweight, respectively) at postnatal day 10 increased the levels of synaptophysin in the cerebral cortex and the hippocampus, 24 h after administration [78]. The long-term effects in vivo, as well as the effects on synapse formation and function in DNT models in vitro remain to be investigated. The evaluation of neuronal function, however, has received less attention, and the functional implications of impaired synaptogenesis are unclear. MeHg and PFOS/PFOA have been shown to alter intracellular calcium homeostasis and neurotransmitter receptor activity (reviewed in [39,79]). In recent years, multielectrode arrays (MEA) have been used to record electrical activity in large populations of neurons, including action potentials (“spikes”) and groups of action potentials (“bursts”) and network bursts. Dingemans and colleagues demonstrated in primary cultures of rat cortical neurons that exposure to 0.1 µM MeHg for 14 days did not affect cell viability but decreased neuronal firing (Spikes) [80]. Similarly, neuronal firing was inhibited by exposure to 1µM MeHg for 30 min, without reducing cell viability. In addition, the mean burst rate (MBR) was decreased in hiPSC-derived Glutaneuron-Astrocyte co-culture exposed to 30 µM MeHg for 30 min [81]. In the same experimental model, exposure to PFOS (0.1 µM) or PFOA (1 µM) for 30 min significantly decreased mean spike rate (MSR) and mean network bursts (MNB), while MBR was decreased by exposure to 100 µM PFOS or PFOA [81]. Cell culture type and conditions may contribute to variability in response to chemical exposure. In a recent experiment, Tukker et al. have shown that 30 min exposure to 100 µM PFOS increased network activity in primary rat cortical neurons, but decreased neuronal activity in hiPSC-derived neurons [77]. Relevant for neuronal differentiation, MEA systems allow longitudinal observations to capture a clear picture of functional alterations induced by exposure to potentially neurotoxic chemicals. Induction of oxidative stress is the most common mechanism and appears to be critical in NSCs undergoing toxic exposures (see Table 1 and Table 2) [12,48,49,66,82]. Compelling evidence shows that MeHg-induced reactive oxygen species (ROS) formation and impaired mitochondrial function, as shown by in vivo and in vitro studies [83]. Mitochondria contain different antioxidants, including glutathione (GSH), thioredoxin (TRX), and the catalase system, which quench ROS and maintain an oxidant-antioxidant balance. MeHg is a threat to the antioxidant defenses, which further alter the REDOX balance essential for proper mitochondrial function. Based on our data from NSCs, exposures to high levels of MeHg (>10 nM for 48 h) damage mitochondrial functions with the release of cytochrome c and activation of the caspase-dependent apoptotic cell death pathway [44]. While subtoxic concentrations of MeHg (<5 nM for 48 h) do not affect cell survival [61], the expression of genes of mitochondrial respiratory chain enzymes of complexes I and III is repressed. Antioxidants protect from MeHg-induced damage, preventing both apoptosis [51] and the alterations in neuronal differentiation [66] in NSCs isolated from embryonal rodent brains exposed to low-dose (10 nM) MeHg. Alterations of different genetics and epigenetics pathways are associated with oxidative stress that generates both free radicals and nonradical oxidants. Free radicals give rise to macromolecular damage and nonradical oxidants (e.g., H2O2, peroxynitrite, lipid hydroperoxide, and disulfides) disrupt redox signaling pathways. Hydroxyl (•OH) is a free radical that can react with guanosine directly, oxidizing it to 8-oxo-7,8-dihydro-2 deoxyguanosine (8-oxo-dG). While 8-oxo-dG is usually repaired by base excision repair (BER) mechanisms, it can give rise to G/T transversions (point mutation) by mispairing with adenine instead of cytosine (see Figure 2). Such mutations in mitochondrial DNA have been demonstrated in human NPCs (ReNcell CX) exposed to 10 or 50 nM MeHg for 48 h [48]. The molecular mechanisms of PFOS-induced neurotoxicity remain largely obscure. PFOS has been shown to decrease cell viability in human-derived neuroblastoma cells (SH-SY5Y) by increasing ROS [84], but it is not clear whether PFOS-induced oxidative stress affects NSC viability. Activation of the JNK pathway and accumulation of ROS have been demonstrated in PFOS-exposed C17.2 cells, which suggests a link between ROS production and JNK signaling that may critically contribute to PFOS-induced neuronal apoptosis [57]. PFOS-induced activation of JNK signaling results in the expression of pro-apoptotic proteins and the initiation of the mitochondrial apoptotic pathway. In addition, JNK may also promote the expression of both pro-apoptotic Bcl-2 family proteins to trigger mitochondrial apoptotic cascades and alter Nrf2 expression that may regulate the expression of antioxidant proteins that protect against oxidative damage triggered by injury and inflammation [84]. ROS are crucial players in neuronal death under various pathological conditions, and may directly execute cell death by inducing mitochondrial permeability and releasing cytochrome c, or initiating multiple signaling pathways, such as p53-p21, JNK, and FOXO to trigger neuronal death [85]. In line with the above evidence, we showed that PFOS induces alterations of the mRNA expression of PPARs (PPARα, PPARδ, and PPARγ) and their downstream targets; the mitochondrial uncoupling proteins (UPC1, UCP2, and UCP3), and the superoxide dismutase (SOD1, SOD2, SOD3) which are important enzymes in the antioxidant defense system of primary culture of embryonic cortical NSCs [56,86]. We also showed that PFOS induces upregulation of PPARγ and UCP2 associated with the accumulation of ROS and oxidative stress. Thus, these results indicate that PFOS-induced ROS production may be a primary neurotoxic mechanism. In contrast, there is no evidence for oxidative stress to contribute to the neurotoxic effects of PFOA in NSCs. Epigenetic alterations are defined as chemical modifications of DNA that are complementary to the genetic code for turning genes on or off via chromatin remodeling without changing the DNA sequence. Epigenetic mechanisms include (1) histone modifications (the modulation of DNA availability via condensation or relaxation of chromatin wrapping in nucleosomes for binding transcription factors); (2) DNA methylation, which modulates the availability of DNA for binding transcription factors in the DNA transcription machinery (gene expression is suppressed when DNA is methylated in the promoter region); and (3) non-coding RNA, such as microRNA (miRNA) strands, which target sequences in the mRNA and repress gene expression post-transcriptionally. Epigenetic alterations describe a variety of reversible modifications across cell types in an organism, which regulates a wide range of physiological and pathological processes from the meiotically and mitotically cell cycle to the function of non-dividing cells—such as neurons. A wide range of exogenous factors, such as the availability of methyl donors in the diet, and environmental contaminants, influence the modification of epigenetic marks. In recent years, the association between oxidative stress and epigenetic alterations has been suggested as a potential mechanism of environmentally relevant exposure to toxicants. DNA is methylated by the enzymatic family of DNA-methyl transferases (Dnmts), which add a methyl group to cytosine in position 5 to generate methylcytosine (5-mC). This 5-mC is oxidized by a family of Ten-eleven translocation methylcytosine dioxygenases (Tets) and generates 5-hydroxymethylation (5-hmC), and the demethylation process is completed during DNA replication. In addition, 5-hmC can be further oxidized by Tets to produce 5-formylcytosine (5-fC) and 5-carboxylcytosine (5-CaC) before demethylation by replication. Oxidative stress may interfere with DNA methylation in two ways. First, it reduces the availability of methyl groups required for DNA methylation by reducing the activity of, e.g., methionine-adenosyltransferase and methionine synthase, enzymes catalyzing the synthesis of S-adenosylmethionine (SAM; the main donor of methyl groups for DN methylation). Second, ROS can oxidize DNA and generate 8-oxo-dG, which activates DNA base repair enzyme OGG1. Thus, activation of OGG1 prevents DNMTs from methylating the DNA, but recruits Tet1 to demethylate the adjacent 5-mC directly or indirectly via deamination followed by Base Excision Repair (BER) enzymes during replication. When this BER cannot repair the specific base, it will be replicated as a point mutation (see Figure 2). Wang and colleagues showed a significant increase in ROS production in hNPC exposed to low levels of MeHg (10 or 50 nM) [48]. Similarly, we demonstrated that in the primary culture of embryonic rat NSCs low levels of MeHg (2.5 or 5 nM for 48 h) induce long-lasting effects which are still present in cells that were not directly exposed to MeHg and had levels of Hg below the detection limit [61]. Interestingly, the Hg concentrations measured in NSCs exposed to 2.5 and 5 nM were 0.4 and 0.7 ppm respectively, which are comparable to the concentrations reported in post-mortem material from infants exposed to MeHg from the maternal diet (up to 0.3 ppm; [87]). The upregulation of p16 and p21 that we detected in NSCs was associated with decreased proliferation (senescence) [61]. We observed an association between global DNA hypomethylation and Dnmt3b downregulation. This is in line with an earlier study showing that inhibition of DNMT decreases cell proliferation and induces cellular senescence in HUBC-NSCs associated with the upregulation of p16 and p21 [88]. Go et al. reported that LUHMES (CRL-2927) cells exposed to 1 nM MeHg for 6 days had increased global DNAA methylation, associated with DNMT1, 3A, and 3B, and following in vivo exposure (3 mg/kg/day between GD12 and GD14) obtained similar results [71]. These apparently contradicting results can be explained by the experimental protocols that lead to the accumulation of Hg to levels relevant for massive and accidental exposure during development. Furthermore, experimental models of aging demonstrated a link between oxidative stress and DNA demethylation [89]. A recent study showed that prenatal exposure to MeHg via maternal diet affected gene-specific methylation, important in brain development, and neuronal signaling in 7-year-old children [90]. CpG sites in the promoter regions of NR3C1 (glucocorticoid receptor), GRIN2B (NMDA-receptor subunit), and BDNF (neurotrophic factor modulating neuronal development and function) were hypermethylated following the developmental exposure to MeHg. For NR3C1, MeHg-induced de novo methylation was found in CpG3 and CpG5 sites, where CpG3 is part of the binding site for transcription factor Hen-1. CpG3 and CpG4 are the binding sites for the transcription factor NGFI-A. Thus, CpG3 methylation downregulated NR3C1 by inhibiting NGF1-A binding. However, MeHg-induced de novo methylation in CpG4 did not affect NR3C1 expression but downregulated GRIN2B expression by inhibiting nuclear respiratory factor 1 (Nrf1). Notably, neurodevelopmental disorders, such as ADHD, ASD, and schizophrenia have been associated with functional alteration of GRIN2B. Similarly, MeHg-induced CpG5 decreased BDNF expression, which is essential for neuronal development, nerve cell survival, and synaptic plasticity [91]. In addition, BDNF is a particularly relevant target of MeHg because BDNF polymorphisms increase the susceptibility to neurotoxicity and MeHg-induced BDNF downregulation has been associated with depression [92,93]. There is currently no evidence that PFAS changes DNA methylation in NSCs at doses relevant to human exposure. Several studies have reported that global and gene-specific methylation alterations are associated with a micromolar concentration of PFOS or PFOA exposure (100–400µM) in various human cell lines and blood samples [94,95,96]. These studies indicate that PFAS may also change the epigenome of NSCs. Further evidence showed that PFOS could decrease global DNA methylation and methylation of the LINE-1 regulatory region, but increase the GSTP promoter region methylation. Therefore, PFOS could lead to the CpG methylation of BDNF mediated by DNMTs and decrease the expression of BDNF. This study explored the mechanism by which PFOS affected BDNF expression via miRNA and methylation regulation. In addition, PFOS exposure decreases the expression of BDNF at mRNA and protein levels, increases the expression of microRNA-16, microRNA-22, and microRNA-30a-5p, decreases the expression of DNMT1 at mRNA and protein levels, but increases the expression of DNMT3b at mRNA and protein levels [97]. It has also been shown that PFOS exposure changes the methylation status of BDNF promoters I and IV. These findings suggest that the downregulation of BDNF along with the upregulation of BDNF-related microRNA might underlie the mechanisms of PFOS-induced neurotoxicity [97]. Histone modifications of epigenetic mechanisms regulate gene transcription via covalent posttranslational modifications (PTMs), such as acetylation, methylation, phosphorylation, ribosylation, ubiquitination, sumoylation, or glycosylation. Both histone acetylation and methylation marks are redox-sensitive and heritable. Approximately 30 histone acetyltransferases (HATs) and histone deacetylases (HDACs) have been found to regulate histone acetylation. Therefore, gene transcription depends on the balance between HATs and HDACs applied to the epigenetic marks. Methylation of histone can either upregulate or downregulate gene expression, which depends on the position of the histone tail, and the number of methyl groups added to a particular amino acid (lysine or arginine) for methylation. More than 40 histone methyltransferases (HMTs) and demethylases regulate this dynamic process. We showed earlier that in utero exposure to 0.5 mg/kg/day MeHg from GD7 until PND7 induces depression-like behavior in male mice [92] as well as decreased granule cell proliferation in the hippocampal dentate gyrus [61]. In this model, the Hg concentration found in the mouse brain was about 0.9 ppm [93], in the same order of magnitude as the one measured in humans [87]. We demonstrated that MeHg increased histone H3-K27 tri-methylation, and decreased H3 acetylation at the BDNF promoter IV region, which resulted in the downregulation of BDNF expression. [61]. Decreased BDNF has been linked to the onset of depression, and MeHg-induced depression in mice is reversed by the antidepressant fluoxetine, which increases the levels of BDNF [92]. These findings may be relevant for the possible effects of MeHg on human mental health. Noncoding RNAs, such as micro-RNAs (miRNAs), long noncoding RNA (lncRNA), and circular RNA (circRNA) are epigenetic processes involved in the post-transcriptional repression of specific genes. Pallocca et al. [98] have used differentiating NT2 cells to evaluate the use of microRNA profiling as a biomarker for developmental neurotoxicity. Exposure to 400 nM MeHg for 5 weeks resulted in the overexpression of a cluster of 5 microRNA species (miR-302b, miR-367, miR-372, miR-196b, and miR-141). The analysis of mRNA expression for genes known to be regulated by these microRNAs identified alterations in line with cognate effects of MeHg, such as decreased neuronal differentiation, and cellular stress response [98]. These findings, however, need to be replicated in NSCs in order to validate the use of microRNA regulation signatures for DNT studies. Recent evidence demonstrated that MeHg-induced alteration of cell viability and decreased cell proliferation were associated with the upregulation of p53R2 expression [49]. In this study, they demonstrated a negative correlation of different miRNAs (miR-1285, miR-30d, and miR-25) with p53R2 expression. While it is not clear whether miR-25 regulates p53 expression directly, overexpression of this small RNA in MeHg-treated ihNPCs significantly reduces the protein expression of p53 [49]. Thus, this study indicates diverse mechanisms of MeHg-induced developmental neurotoxicity. In this section, we have focused on outcomes and mechanisms of direct neurotoxicity. In vivo data point to biologically relevant effects on neurophysiology and behavior due to indirect neurotoxic effects. Endocrine disruption is particularly relevant during development, as acknowledged in the following definition: “Developmental neurotoxicity (DNT) refers to any adverse effect of perinatal exposure to a toxic substance on the normal development of nervous system structure and/or function.” [99]. During prenatal development, indirect neurotoxic effects due to interactions with the placenta or maternal organs should also be taken into consideration (see [100,101]). The toxic effects of exposure to MeHg are largely restricted to the nervous system, but recent evidence points to potential endocrine-disrupting effects on glucocorticoid receptor signaling [102,103]. For PFAS, the investigation of endocrine disruption in relation to developmental neurotoxicity has focused mainly on the thyroid hormone system, but altered signaling via gonadal and stress steroids may also have a biologically relevant contribution (reviewed in [39,101]). In this context, one of the main limitations is that NSC/NPC models can only provide information on direct neurotoxicity, while interactions with other organs and systems remain difficult to investigate. Indirect neurotoxicity, including but not limited to endocrine disruption, as well as compensatory mechanisms acting at later developmental stages may account for discrepancies between neurotoxic outcomes predicted by NSC models and in vivo or epidemiological observations. The development of 2D and 3D systems for neural stem cell cultures has advanced our understanding of neurodevelopmental processes in physiological conditions, as well as the effects of exposure to toxicants (see the evaluation of myelination in BrainSpheres [42]). Organoids offer the possibility to investigate the more complex processes affected by developmental neurotoxicants by recapitulating brain development in a dish, and the availability of cells of human origin increases the relevance of the results. Co-culture systems and, more recently developed, vascularized organoids [16] may further support the extrapolation of findings to human populations. Additional intrinsic technical limitations stem from cell source and culture conditions. The embryonic developmental stage at the time of harvesting may have a significant impact on susceptibility to neurotoxicity. The study by Edoff et al. [47] indicates that human NPCs derived at earlier embryonal development stages are more sensitive to MeHg exposure. The same study also highlights the fact that the sensitivity to neurotoxic insults also differs according to the sex of the embryo, namely a more pronounced neurotoxic effect in male hNPCs [47]. There is limited information on sex-related differences from the reviewed literature (see e.g., [47,50]). For NSCs derived from animal models, the cells are typically harvested from several pups and pooled to increase the initial yield without specifically selecting male or female pups. The main constraint for NSCs of human origin is the availability of original material, and consistently reporting the sex and developmental age of the source of cells is critical for assessing the generalizability of the findings (see e.g., [74]). Even if the impact of exposure to such environmental contaminants has been decreasing thanks to increasing awareness and control of both industrial emissions and main sources of exposure [104], the estimated impact on healthcare expenditure remains considerable. MeHg is generated by anaerobic bacteria in water sediments and undergoes bioaccumulation and bioamplification in the food chain. PFOS and PFOA have relatively low acute toxicity but have quickly gained attention because of their pervasiveness and persistence in the environment. Human populations are continuously exposed through the consumption of contaminated food, water, and beverages. The neurodevelopmental impact of these environmental contaminants has been evaluated based on estimated intellectual disability and IQ points loss due to prenatal exposure. For MeHg, the annual cost has been estimated to be 2.84 billion USD [104]. For PFOS and PFOA, the annual increase in health expenditure attributable to loss of IQ points associated with low birth weight was estimated to be 1.11 billion and 10.7 billion USD for PFOA and PFOS, respectively [105]. The economic impact of other neurodevelopmental disorders associated with prenatal exposure to these toxicants (intellectual disability, ASD, ADHD, and possibly depression, as indicated by experimental data and occupational exposure [92,106,107]) would considerably increase the cost estimates. Therefore, new information on the mechanisms of action obtained from the most relevant models is necessary to identify, prevent and counteract their harmful effects. The revised literature converges on NSCs (of rodent and human origin) being a valuable model for investigating developmental neurotoxicity. In vitro models seem to support observations made in epidemiological studies, such as higher sensitivity to toxicants in developing subjects (as in NSCs) than in more mature subjects (as in differentiated neurons and glia), as well as sex-related differences in susceptibility. Low levels of exposure induce major alteration in critical neurodevelopmental steps, which presumably leads to functional impairments. Nanomolar levels of relevant contaminants, such as MeHg and PFAS, alter proliferation, differentiation, migration, and neurite outgrowth, while micromolar concentrations induce apoptotic cell death. Neural stem cells of human origin (embryonal or iPSCs-derived), cultured in 2D or 3D systems, including organoids, are the best options available to explore normal neurodevelopment and its alteration in a controlled experimental environment. The fact that NSCs illustrate the neurotoxic potential of known neurotoxicants (such as MeHg) at low exposure levels that are relevant for the general population makes them an ideal in vitro model for identifying new neurotoxic chemicals and the effect of mixtures (see also [59,72]). Altogether, the reviewed literature shows that mechanistic studies are important to support epidemiological and animal experimental data regarding the role of chemical contaminant exposures in the etiopathogenesis of neurodevelopmental disorders. The effects of MeHg and PFAS on proliferation and differentiation are similar in single-exposure models. Oxidative stress, Ca2+ homeostasis, and mitochondrial function impairments are common mechanisms behind cell death. Instead, the signaling pathways regulating NSC proliferation and differentiation affected by MeHg and PFOS/PFOA are different. The occurrence of epigenetic modifications enhances the complexity of the processes initiated by toxic exposures and highlights the risk of transgenerational effects. The identification of common and specific intracellular processes activated by exposures to neurotoxicants may enable the development of preventive and therapeutic strategies to benefit the populations at risk. More research is required to fill the knowledge gaps on most of the chemicals we are exposed to.

PMC10002379

Luana Grupioni Lourenço Antonio,Juliana Meola,Ana Carolina Japur de Sá Rosa-e-Silva,Antonio Alberto Nogueira,Francisco José Candido dos Reis,Omero Benedicto Poli-Neto,Julio César Rosa-e-Silva

Altered Differential Expression of Genes and microRNAs Related to Adhesion and Apoptosis Pathways in Patients with Different Phenotypes of Endometriosis

23-02-2023

endometriosis,genes,microRNAs,MAPK1,CAPN2

We aim to investigate the expression of genes (MAPK1 and CAPN2) and microRNAs (miR-30a-5p, miR-7-5p, miR-143-3p, and miR-93-5p) involved in adhesion and apoptosis pathways in superficial peritoneal endometriosis (SE), deep infiltrating endometriosis (DE), and ovarian endometrioma (OE), and to evaluate whether these lesions share the same pathophysiological mechanisms. We used samples of SE (n = 10), DE (n = 10), and OE (n = 10), and endometrial biopsies of these respective patients affected with endometriosis under treatment at a tertiary University Hospital. Endometrial biopsies collected in the tubal ligation procedure from women without endometriosis comprised the control group (n = 10). Quantitative real-time polymerase chain reaction was performed. The expression of MAPK1 (p < 0.0001), miR-93-5p (p = 0.0168), and miR-7-5p (p = 0.0006) was significantly lower in the SE group than in the DE and OE groups. The expression of miR-30a (p = 0.0018) and miR-93 (p = 0.0052) was significantly upregulated in the eutopic endometrium of women with endometriosis compared to the controls. MiR-143 (p = 0.0225) expression also showed a statistical difference between the eutopic endometrium of women with endometriosis and the control group. In summary, SE showed lower pro-survival gene expression and miRNAs involved in this pathway, indicating that this phenotype has a different pathophysiological mechanism compared to DE and OE.

Altered Differential Expression of Genes and microRNAs Related to Adhesion and Apoptosis Pathways in Patients with Different Phenotypes of Endometriosis We aim to investigate the expression of genes (MAPK1 and CAPN2) and microRNAs (miR-30a-5p, miR-7-5p, miR-143-3p, and miR-93-5p) involved in adhesion and apoptosis pathways in superficial peritoneal endometriosis (SE), deep infiltrating endometriosis (DE), and ovarian endometrioma (OE), and to evaluate whether these lesions share the same pathophysiological mechanisms. We used samples of SE (n = 10), DE (n = 10), and OE (n = 10), and endometrial biopsies of these respective patients affected with endometriosis under treatment at a tertiary University Hospital. Endometrial biopsies collected in the tubal ligation procedure from women without endometriosis comprised the control group (n = 10). Quantitative real-time polymerase chain reaction was performed. The expression of MAPK1 (p < 0.0001), miR-93-5p (p = 0.0168), and miR-7-5p (p = 0.0006) was significantly lower in the SE group than in the DE and OE groups. The expression of miR-30a (p = 0.0018) and miR-93 (p = 0.0052) was significantly upregulated in the eutopic endometrium of women with endometriosis compared to the controls. MiR-143 (p = 0.0225) expression also showed a statistical difference between the eutopic endometrium of women with endometriosis and the control group. In summary, SE showed lower pro-survival gene expression and miRNAs involved in this pathway, indicating that this phenotype has a different pathophysiological mechanism compared to DE and OE. Endometriosis is a benign gynecological disease characterized by the presence and growth of endometrial tissue outside the uterine cavity and myometrium, commonly found in the anterior and posterior compartments of the pelvic cavity, in the pelvic peritoneum, ovaries, rectovaginal septum, bladder, and intestine. It mainly affects women of reproductive age (5% to 10% of women in this phase) and is associated with pelvic pain, dysmenorrhea, dyspareunia, constipation, dysuria, changes in bowel function, and, often, infertility [1]. However, clinical presentation is considerably variable, and none of these symptoms are specific for the disease, making its diagnosis difficult [2]. In addition, endometriosis can significantly affect a woman’s personal, as well as intimate and professional aspects of life. Some studies have shown a higher incidence of depressive symptoms, anxiety, stress, and lower quality of life in women with endometriosis compared to women without endometriosis. Furthermore, a correlation was found between depression/infertility and pain/quality of life [3,4]. Although the disease has heterogeneous presentation, the three main different phenotypes that are currently recognized are ovarian endometrioma, superficial peritoneal endometriosis, and deep infiltrating endometriosis [5]. Superficial endometriosis is thus classified as the presence of small lesions, measuring between 1 mm and 3 mm, with foci generally implanted in the peritoneum and rarely affecting organs. Endometrioma, in turn, is characterized by the presence of cysts in the ovary that are filled with a typical chocolatey liquid. Meanwhile, deep endometriosis presents lesions of at least 5 mm in depth and can be located in several organs, including the peritoneum, bladder, and intestine, the latter being the most advanced form of the disease [6]. These phenotypes may represent three clinically separate disease entities with different pathogenesis. Most deep infiltrating endometriosis lesions present with other forms of endometriosis, and about half deep infiltrating endometriosis lesions present with ovarian endometrioma. It is estimated that deep infiltrating endometriosis can affect 20% of women with superficial peritoneal endometriosis [7]. Clinical data suggest a clear distinction in terms of diagnosis, treatment, and follow-up regarding these three types of phenotypes [5]. However, most pathogenetic research on endometriosis has been conducted mixing the three phenotypic manifestations. Therefore, a better molecular understanding of these lesions is needed in order to provide efficient and specific management for each subgroup [8]. The etiology of endometriosis is complex and multifactorial, where several not fully confirmed theories describe its pathogenesis [9]. Retrograde menstruation is considered an important source of endometrial deposits, although other factors are necessary to promote cell survival, proliferation, formation, and maintenance of endometriotic lesions [1]. Today, it is recognized that alterations in the peritoneal microenvironment must also occur; therefore, the following processes are essential: escaping from immune system surveillance [9,10]; changes in local concentrations of hormones [11] and inflammatory mediators [12]; cell adhesion [13]; tissue invasion [14]; evading apoptosis [15]; angiogenesis [16]; and ectopic cell proliferation [17]. More recently, the possibility that endometriosis is an epigenetic disease was proposed, with modifications in the methylation of DNA and histones and alterations in the expression of non-coding RNAs, among the latter, microRNAs (miRNAs) [18]. Specific changes in endometrial and peritoneal cell adhesion molecules seem to facilitate the binding of endometrial menstrual reflux in ectopic sites, namely the integrins α2β1, α3β1, α4β1, α5β1, and E-cadherin [13]. In this context, most studies have evaluated the expression of integrins in the eutopic endometrium [19]. Studies have shown that patients with endometriosis present reduced susceptibility to apoptosis in endometrial cells released during menstruation, thereby facilitating survival and ectopic implantation [15]. miRNAs are important regulators of gene expression, acting at the post-transcriptional level, through the induction of mRNA degradation or by blocking protein synthesis [20]. They seem to be potent regulators of gene expression in endometriosis, participating in important cellular events that trigger the development of the disease [21]. Several studies have shown that the expression of miRNAs is altered in the eutopic endometrium [22,23,24], in ectopic and eutopic endometrial tissues [21,24,25], and in circulating miRNAs in women with endometriosis when compared to healthy subjects [26,27,28,29,30]. Most studies on miRNA and endometriosis have been carried out comparing the endometrium of women with and without endometriosis, encountering, in many cases, differential expression and, therefore, implicating miRNAs as potential biomarkers of this condition [31]. Although obtaining endometrial tissue presents an invasive aspect, the advantage of its use is that it can be accessed by biopsy without the need for anesthesia [32]. However, a panel with good specificity and sensitivity has not been found so far [33,34]. Unraveling the significance of miRNAs in endometriosis will pave the way for a better understanding of the pathophysiology of this disease and new diagnostic tests, as well as identify new therapeutic targets and treatment approaches that have the potential to improve the clinical options for women with this disabling condition. However, despite the increase in research on the subject, to date, the utility of miRNAs for this purpose has not been specifically analyzed [33]. Thus, the aim of the present study was to investigate the expression of the genes (MAPK1 and CAPN2) and microRNAs (miR-30a-5p, miR-7-5p, miR-143-3p, and miR-93-5p) involved in the adhesion and apoptosis pathways in superficial peritoneal endometriosis (SE), deep infiltrating endometriosis (DE), and ovarian endometrioma (OE), and to evaluate whether these lesions share the same pathophysiological mechanisms. The demographic variables of the patients did not differ significantly among the groups (p > 0.05) (Table 1). The expression of genes and microRNAs among the ectopic implants is illustrated in Figure 1. The expression of MAPK1 (p < 0.0001) and miR-7-5p (p = 0.0006) was significantly lower in superficial lesions compared to deep endometriosis and ovarian endometrioma. Meanwhile, the expression of miR-93-5p showed to be different between superficial and deep lesions (p = 0.0168). The relative expression of CAPN2, miR-143-3p, and miR-30a-5p did not differ significantly (p > 0.05) among the different lesion types. Gene and microRNA expression were compared between the control eutopic endometrium and the eutopic endometrium of women with endometriosis. The expression of miR-30a and miR-93 was significantly (p = 0.0018 and p = 0.0052, respectively) upregulated in the eutopic endometrium of women with endometriosis compared to the controls. The expression of miR-143 was also statistically different between groups (p = 0.0225); however, it was downregulated in patients with endometriosis. There was no significant difference regarding the expression of MAPK1, CAPN2, and miR-7 (Figure 2). The expression of genes and miRNAs between the eutopic endometrium of women with different types of lesions and the eutopic endometrium of the controls were compared, as shown in Figure 3. The expression of miR-30a was significantly lower in the control eutopic endometrium than in the endometrium of women with deep infiltrating lesions (p = 0.0194) and ovarian endometrioma (p = 0.0208). Moreover, the eutopic endometrium of the control group showed reduced expression of miR-93 when compared to the eutopic endometrium of patients with superficial peritoneal lesions (p = 0.0139). The expression of genes and microRNAs in the ectopic implants was compared with the expression of the corresponding eutopic endometrium (within the same woman). The expression of MAPK1 (p < 0.0001) and miR-93 (p = 0.0021) was significantly lower in the lesions when compared to the corresponding eutopic endometrium (Figure 4). The different theories that attempt to explain the etiology of endometriosis seem to complement each other, suggesting that the disease has a multifactorial origin. However, the exact mechanisms that promote and favor the survival and implantation of endometriotic foci in ectopic sites have not yet been precisely clarified. Thus, further studies aimed at investigating the etiology of endometriosis are still needed. The knowledge of its etiology will make it easier to develop new diagnostic tests, as well as identify new therapeutic targets and treatment approaches that have the potential to improve the clinical options for women with this disabling condition. Furthermore, it is necessary to distinguish the pathophysiological mechanisms in the different phenotypes of endometriosis in order to contribute to the individualized care of these patients. It was suggested long ago that the phenotypes ovarian endometrioma, superficial peritoneal endometriosis, and deep infiltrating endometriosis may represent three clinically separate disease entities with different pathogenesis; however, different types of endometriosis very often coexist in one patient [7]. In our experiments, only one type of lesion from each patient was collected. The results obtained in the present study demonstrated a significantly lower expression of MAPK1 in superficial lesions compared to deep endometriosis and ovarian endometrioma. Moreover, MAPK1 expression was significantly lower in the lesions when compared to the corresponding eutopic endometrium. MAPK1 are extracellular signal-regulated kinases (ERKs), such as sex hormones and inflammatory factors, that play an important role in many cellular reactions [35]. Several studies have shown this gene directly participating in the regulation of endometriosis pathophysiology, such that the MAPK pathway was activated in ectopic and eutopic endometrial cells of patients with endometriosis [36]. Research using MAPK1 inhibitors has evidenced anti-inflammatory, anti-proliferative, anti-angiogenic, and apoptotic effects and reduced adhesion and migration [36,37,38,39,40,41,42]. Thus, our results demonstrate a likely greater activation of these pathways in deep lesions and ovarian endometrioma, and that the superficial endometriosis phenotype probably has a different pathophysiological mechanism. The signaling cascades of the MAPK pathway participate in cell survival mechanisms, providing signals that fuel cell cycle progression and affect the transcription factors that regulate apoptosis. The latter can activate Bcl-2 or inactivate caspase-9, thus generating an anti-apoptotic signal [36]. Li et al. (2013) showed that the ability of human primary cell cultures from eutopic endometrial stroma to adhere to Collagen IV and Fibronectin is MAPK-dependent. They found that U0126 (a MEK-targeted MAPK inhibitor) affected endometriotic stromal cell adhesion and invasion in vitro [40]. Ngo et al. (2010) compared ectopic and eutopic endometrial cells from biopsies of patients with and without endometriosis, and observed that the MAPK pathway was activated in ectopic and eutopic endometrial cells from patients with endometriosis, as evidenced by a significantly higher pERK/ERK ratio in these patients compared to the control group [43]. Furthermore, the increased proliferation and survival of eutopic endometrial cells from patients with endometriosis, compared to healthy women, has been correlated with abnormal levels of activation of the MAPK signal pathway [44]. However, according to our results, no difference was found in MAPK1 expression in the eutopic endometrium of women with superficial peritoneal endometriosis, deep infiltrating endometriosis, ovarian endometrioma, and control eutopic endometrium. According to a recent meta-analysis, there are consistently upregulated and downregulated miRNAs in ectopic foci compared to the eutopic endometrium of healthy women [34]. MiR-93-5p and miR-143-3p regulate MAPK1 gene expression. Research on endometriosis has already been carried out for both miRNAs. The results of the present study demonstrated an upregulation of miR-93-5p expression in the eutopic endometrium of women with endometriosis when compared to the controls, the difference being between control endometrium versus endometrium in women with superficial peritoneal lesions. Furthermore, the expression of miR-93 was different between superficial and deep lesions, with greater expression in the latter. This result contrasts with a previous study, in which miR-93 showed to be underexpressed in the ectopic endometrium of patients with endometriosis compared to the peritoneal tissue of patients without the disease, causing increased expression of MMP3 and VEGFA, thus stimulating the proliferation, migration, and invasive capacity of endometrial stromal cells [45]. Such difference between results is probably due to the different tissues analyzed in the comparison between endometriosis and the control. MiR-143-3p is involved in cell proliferation, apoptosis, adhesion, invasion, and other cellular processes [46]. The transfection of a miR-143 mimetic into HL-60 myelocytic leukemia cells remarkably suppressed MAPK1 expression, inhibiting cell proliferation and inducing apoptosis [47]. Chang et al. (2017) found that miR-143 inhibited proliferation, migration, and invasion and promoted apoptosis in endometrial cancer cells by suppressing MAPK1 [48]. Other studies have demonstrated the role of miR-143 as a tumor suppressor, reducing proliferation and adhesion and increasing apoptosis [49,50]. Previous studies reported that miR-143-3p was markedly dysregulated in endometriosis and was found to be overexpressed in the serum and tissue of affected women when compared to the controls [51,52,53]. In our study, downregulation of miR-143 expression was observed in the eutopic endometrium of women with endometriosis compared to the control endometrium. Conversely, in a recent study, the expression of miR-143-3p was upregulated in endometriotic stromal cells from women with endometriosis when compared to eutopic endometrial tissues obtained from women without endometriosis. However, functionally, the overexpression of miR-143-3p suppressed the proliferation and invasion of endometriotic stromal cells, thus inhibiting the progression of the disease [53]. Therefore, a lower expression of miR-143 in the endometrium could promote a greater potential to proliferate and invade, corroborating our results. Previous studies showed that miR-143 was overexpressed in ectopic endometrial tissues when compared to eutopic endometrial tissues [21,54]. The same result was obtained herein in the group presenting deep lesions. Similar to our findings, in serous ovarian carcinoma, there was an overexpression of miR-93 and an underexpression of miR-143 compared to benign lesions, thus correlating with lower patient survival [55]. Genes from the CAPN family are calcium-activated neutral proteases that have been reported to regulate focal adhesion, cytoskeletal remodeling, and apoptosis [56]. The CAPN5 gene was found to be underexpressed in eutopic endometrial biopsies of women with endometriosis when compared to controls without the disease [57]. On the other hand, another study reported an increased expression of CAPN7 in the eutopic endometrium and endometrial stromal cells of women diagnosed with endometriosis. The authors proposed, based on functional tests, that this gene promotes the migration and invasion of human endometrial stromal cells through the regulation of matrix metalloproteinase 2 (MMP-2) [58]. However, in the present study, we did not find differences among the different types of lesions, nor between the control and endometriosis groups regarding the expression of the CAPN2 gene. According to the miRwalk program, miR-30a-5p and miR-7-5p are regulators of CAPN2 gene expression. The expression of miR-7 was significantly lower in the superficial lesion group than in the deep endometriosis and ovarian endometrioma groups. MiR-7 has been reported as a regulator of MMP-2 and MMP-9 expression, acting in the invasion and proliferation of human colon cancer by directing the expression of the focal adhesion kinase [59]. Moreover, the expression of MMP-2 and MMP-9 has been shown to be increased in women with endometriosis when compared to controls [60]. Therefore, this pathway may be more activated in the deep endometriosis and ovarian endometrioma phenotypes. MiR-7 was found to be underexpressed in cervical cancer tissues compared to the corresponding normal adjacent cervical tissues. Furthermore, the expression of miR-7 in metastatic cervical cancer was significantly lower compared to cancer without metastasis. The overexpression of miR-7 by transfection inhibited the migration and invasion of cervical cancer cells by suppressing the expression of FAK (focal adhesion kinase), an important adhesion kinase that contributes to extracellular matrix integrin signaling, cell motility, proliferation, and survival [61]. MiR-7 overexpression in cervical cancer cell lines (HeLa and C-33A) also suppressed cell viability and promoted apoptosis, while its inhibition promoted opposite effects [62]. The expression of miR-30a was upregulated in the eutopic endometrium of women with endometriosis compared to the controls, with the difference in its expression occurring between the control versus deep infiltrating lesions and ovarian endometrioma. MiR-30a downregulates the expression of β3-integrin, modulating cell adhesion and invasion, interrupting the MAPK1 pathway in triple-negative breast cancer [63]. Corroborating our results, some studies have reported reduced expression of β3-integrin in the endometrium of women with endometriosis [16]. In the present study, we conducted comparisons between eutopic endometrium and ectopic lesions within cases, assuming that the differential expression of genes and miRNAs in the affected women would reflect disease processes as opposed to individual differences in gene expression and regulation. In this comparison, we found a difference in the expression of miR-93, which, in the group of women with superficial endometriosis, exhibited upregulated expression in the eutopic endometrium compared to the ectopic lesions. Meanwhile, the expression of miR-143 in the group of women with deep endometriosis was downregulated in the eutopic endometrium when compared to the ectopic lesions. It is well-known that the types of endometriotic lesions are biochemically distinct and, therefore, it is hypothesized that miRNAs are differentially expressed in superficial peritoneal endometriosis, deep infiltrating endometriosis, and endometrioma. In this context, a major criticism in studies carried out in this area is the lack of molecular evaluations of endometriosis in the different phenotypes of the disease. So far, few studies have made such distinction. Haikalis et al. (2018) separately analyzed miRNA expression in the three types of lesions. In their study, 15 endometrioma samples, 11 superficial lesion samples, and 10 deep lesion samples were used, and the expression levels of miR-9, miR-21, miR-424, miR-10a, miR-10b, and miR-204 were evaluated by qPCR. The expression of miR-21 and miR-424 was significantly lower in the superficial lesion group than in the endometrioma group. Meanwhile, the expression of miR-10b in the deep lesion group was significantly lower than in the endometrioma group. No differences were observed in the expression of the following miRNAs: miR-9, miR-10a, and miR-204 [64]. Like Haikalis et al. (2018), our study can also lead to the conclusion that the pattern of expression of miRNAs depends on the type of endometriotic lesion analyzed. Regarding the miRNAs and genes selected herein, we observed that the superficial lesion phenotype had lower pro-survival gene expression levels and miRNAs involved in this pathway, indicating that this phenotype has a different pathophysiological mechanism in relation to deep endometriosis and ovarian endometrioma. Our samples were not microdissected; therefore, we cannot exclude the possibility that our tissue samples contain non-endometriotic cells. Saare et al. (2014) showed significantly different gene and miRNA expression in peritoneal endometriotic lesions compared to healthy peritoneal tissues [65]. However, we believe that an endometriosis lesion is not just the ectopic endometrium, but the entire adjacent inflammatory process, so that this would also be part of the lesion. Malysheva et al. (2020) examined the expression of genes in the peritoneum of patients with endometriosis and healthy women, finding equally high level of expression of genes in endometriotic lesions and underlying peritoneum, indicating a probable common origin of these tissues [66]. The limitations of the present study include the predominance of endometriosis in advanced stages (III-IV), therefore hindering the extrapolation of our results to milder stages of the disease. Furthermore, it was not possible to obtain samples exclusively from patients who did not use hormones due to the clinical treatment currently recommended before surgery. Another confounding variable is the phase of the menstrual cycle, which was not controlled in our experiments, as samples were collected during surgery, which could interfere with our results. Challenges for future studies include the standardization of the definitions of cases and controls, the severity of endometriosis, the phase of the menstrual cycle, tissue sampling, and laboratory methods. Such factors demand rigorous attention in the elaboration of the study design. These potentially confounding variables have been neglected and, therefore, make comparisons between studies difficult and may be responsible for the controversial results [67]. Samples of deep lesions (n = 10), superficial lesions (n = 10), and ovarian endometriomas (n = 10), as well as endometrial biopsies of these respective patients, were collected from patients affected by endometriosis undergoing treatment at the Clinics Hospital of Ribeirão Preto (HCRP). Endometrial biopsies were also collected from women without endometriosis during the tubal ligation procedure, comprising the control group (n = 10). The samplings were carried out in 2018 and 2019. In addition, samples from the endometriosis biorepository were also used, approved by the Research Ethics Committee of the Clinics Hospital of Ribeirão Preto (HCRP Process No. 9699/2006). To this end, we requested the integration of the biorepository with this research project, which was approved in 2006. This study was approved by the Ethics Committee of the HCRP tertiary University Hospital (Protocol No. 12514/2017). All experiments were performed in accordance with relevant guidelines and regulations. Patients who met the following inclusion criteria were included in the study: women with endometriosis, diagnosed with any stage of the disease during a surgical procedure (according to the American Society for Reproductive Medicine), able to provide informed consent, and who were of reproductive age (between 18 and 45 years old). Sample collection was performed during surgery, and the tissue was stored in a freezer at −80 °C in RNAlater for further gene expression analysis using the RQ-PCR technique. We reviewed the medical records of each patient, collecting data such as: date of birth, age, contraceptive method, parity, habits, the stage of the disease, medications in use, other associated diseases, and lesion characteristics. All women signed an informed consent form. The tissue samples were submitted to RNA extraction using the AllPrep DNA/RNA/miRNA Universal Kit (Qiagen, Hilden, Germany). To this end, the sample fragments were weighed as not to exceed 30 mg and homogenized using a Polytron® device. From then on, the manufacturer’s protocol was followed and, immediately after RNA extraction, the samples were stored in a freezer at −80 °C. In order to verify the integrity of the obtained RNA, the samples were analyzed in a 4200 TapeStation System (Agilent Technologies, Santa Clara, CA, USA). For the quantification of the total RNA concentration, Thermo Scientific NanoDrop 2000 equipment was used (Thermo Fisher Scientific, Waltham, MA, USA). For the synthesis of miRNA cDNA, we used the TaqMan™ Advanced miRNA cDNA Synthesis Kit (Applied Biosystems, Waltham, MA, USA), while for the genes, we used the High-Capacity RNA cDNA Kit (Applied Biosystems). cDNA synthesis was carried out following the manufacturer’s protocol regarding the quantities of reagents and cycle time. For miRNA, we used 10 ng, and for the genes, 100 ng of total RNA. After synthesis, the samples were stored in a freezer at −20 °C. Before the real-time PCR reaction, the following dilutions of the synthesized cDNA in DEPC water were performed: 1:10 for the miRNA and 1:4 for the genes. The real-time PCR method was used to confirm the differential expression of the genes MAPK1 (Hs01046830_m1) and CAPN2 (Hs00965097_m1) and the microRNAs miR-30a-5p (479448_mir), miR-7-5p (483061_mir), miR-143-3p (477912_mir), and miR-93-5p (478210_mir). The choice of genes was carried out using the DAVID v6.7 (Database for Annotation, Visualization and Integrated Discovery) tool, and of miRNAs, using the miRwalk 2.0 database. In the quantitative expression analysis, the commercially available systems TaqMan Gene Expression Assay (FAM) and TaqMan Advanced miRNA Assay (Applied Biosystems) were used for the genes and miRNA, respectively. For the miRNAs, we used the following as reference miRNA: hsa-miR-361-5p (478056_mir), hsa-miR-186-5p (477940_mir), and hsa-miR-92a-3p (477827_mir). As for the genes, B2M (Hs00187842_m1) and ACTB (Hs01060665_g1) were used as a reference. The Real-time Quantitative Polymerase Chain Reaction (RQ-PCR) amplification reactions were performed in triplicate in a 96-well plate using the TaqMan™ Fast Advanced Master Mix reagent (Applied Biosystems), with a final volume of 10 µL. A 7500 Fast Real-Time PCR System (Applied Biosystems) was used together with the 7500 Sequence Detection System software (Applied Biosystems) to obtain the Ct values. The standard amplification conditions were: 95 °C for 20 s, followed by 40 cycles of 95 °C for 3 s and 60 °C for 30 s (simultaneous annealing and extension). The expression data were then exported and analyzed in the Thermo Fisher Cloud Software (Last Updated 12 December 2018) Scientific, Waltham, MA, USA). In the statistical analysis, the chi-squared test was used for qualitative variables and one-way analysis of variance (ANOVA) and Tukey’s post-hoc test for quantitative variables (when there was significance in the ANOVA tests). The Mann-Whitney test was used to compare gene and miRNA expression in the eutopic endometrium of the cases and the controls, and the paired t-test between ectopic lesions and the eutopic endometrium of the same woman with endometriosis. The GraphPad Prism 6.0 software (GraphPad Prism, Inc, San Diego, CA, USA) was used to generate the graphs. Results were considered significant when p < 0.05.

PMC10002397

Xinping Guo,Ziang Qian,Qiqi Pan,Yuqing Hu,Wangxin Mei,Xiumei Xing,Shaowu Yin,Jie Ji,Kai Zhang

Effects of Florfenicol on Intestinal Histology, Apoptosis and Gut Microbiota of Chinese Mitten Crab (Eriocheir sinensis)

23-02-2023

florfenicol,gut barrier,apoptosis,gut microbiota,crab

Excessive use of antibiotics in aquaculture causes residues in aquatic animal products and harms human health. However, knowledge of florfenicol (FF) toxicology on gut health and microbiota and their resulting relationships in economic freshwater crustaceans is scarce. Here, we first investigated the influence of FF on the intestinal health of Chinese mitten crabs, and then explored the role of bacterial community in FF-induced intestinal antioxidation system and intestinal homeostasis dysbiosis. A total of 120 male crabs (48.5 ± 4.5 g) were experimentally treated in four different concentrations of FF (0, 0.5, 5 and 50 μg/L) for 14 days. Responses of antioxidant defenses and changes of gut microbiota were assessed in the intestine. Results revealed that FF exposure induced significant histological morphology variation. FF exposure also enhanced immune and apoptosis characteristics in the intestine after 7 days. Moreover, antioxidant enzyme catalase activities showed a similar pattern. The intestinal microbiota community was analyzed based on full-length 16S rRNA sequencing. Only the high concentration group showed a marked decrease in microbial diversity and change in its composition after 14 days of exposure. Relative abundance of beneficial genera increased on day 14. These findings illustrate that exposure to FF could cause intestinal dysfunction and gut microbiota dysbiosis in Chinese mitten crabs, which provides new insights into the relationship between gut health and gut microbiota in invertebrates following exposure to persistent antibiotics pollutants.

Effects of Florfenicol on Intestinal Histology, Apoptosis and Gut Microbiota of Chinese Mitten Crab (Eriocheir sinensis) Excessive use of antibiotics in aquaculture causes residues in aquatic animal products and harms human health. However, knowledge of florfenicol (FF) toxicology on gut health and microbiota and their resulting relationships in economic freshwater crustaceans is scarce. Here, we first investigated the influence of FF on the intestinal health of Chinese mitten crabs, and then explored the role of bacterial community in FF-induced intestinal antioxidation system and intestinal homeostasis dysbiosis. A total of 120 male crabs (48.5 ± 4.5 g) were experimentally treated in four different concentrations of FF (0, 0.5, 5 and 50 μg/L) for 14 days. Responses of antioxidant defenses and changes of gut microbiota were assessed in the intestine. Results revealed that FF exposure induced significant histological morphology variation. FF exposure also enhanced immune and apoptosis characteristics in the intestine after 7 days. Moreover, antioxidant enzyme catalase activities showed a similar pattern. The intestinal microbiota community was analyzed based on full-length 16S rRNA sequencing. Only the high concentration group showed a marked decrease in microbial diversity and change in its composition after 14 days of exposure. Relative abundance of beneficial genera increased on day 14. These findings illustrate that exposure to FF could cause intestinal dysfunction and gut microbiota dysbiosis in Chinese mitten crabs, which provides new insights into the relationship between gut health and gut microbiota in invertebrates following exposure to persistent antibiotics pollutants. As a major aquacultural country, commercial aquatic farming in China is predominantly based on intensive cultivation, which increases risks of disease outbreak and antibiotics abuse. According to a WHO report, China consumed nearly 50% of world’s antibiotics in 2013, approximately 162,000 tons, half of which came from animal farming [1]. Considering the durability and poor absorption after medication, overdoses of antibiotics have been confirmed to cause body deformity in aquatic animals, leave residues in food, and influence the emergence of antibiotic resistance genes for human pathogens [2]. Among antibiotics, as a third generation product of chloramphenicol, florfenicol (FF) is widely used in humans, veterinary clinics and aquaculture for its broad-spectrum benefits and lack of side effects [3]. This class of drugs blocks the process of bacteria protein translation by binding with ribosome 50S subunit to inhibit bacterial activity [4]. FF was first created as an alternative to chloramphenicol and thiamphenicol and was added in feed supplement to improve growth rate in aquaculture [5]. In North America, FF was applied in the therapy of enteric septicaemia of channel catfish (Ictalurus punetaus) [6]. However, abuse of FF caused environmental contamination. FF cannot be absorbed and metabolized completely in the gut, leading the antibiotic to be excreted from aquatic animals into water and sediment. There were also changes to the metabolisms of environmental microorganisms and the nitrogen cycle [7]. As an important source of animal protein, economic aquaculture organisms are mainly consumed by humans, which leads to an increase in FF residues and poses significant risks to human health. Previous toxicological studies of FF were mostly carried out in acute exposure. However, although FF was hydrolyzed and excreted quickly from organisms, many aquatic species continuously suffer through long periods of exposure to FF in aquatic environments, or even throughout their entire life cycle. Evidence demonstrated that chronic exposure to FF in high concentrations will result irreversible inhibit the hematopoietic system [8]. In addition, 21 days of FF treatment in adult zebrafish (Danio rerio) resulted in a substantial increase in glycolipid-related genes, leading to hepatic metabolic disorder [9]. Limited information has been found on the potential mechanism of FF toxicity in chronic environments and extreme concentrations, especially in invertebrates. Therefore, it is necessary to examine the chronic influence of FF residues on aquatic crustaceans more carefully. In aquatic animals, as the major organ for immune defense and nutrient absorption, the intestine has direct contact with contacts and forms the first barrier of defense against the external environment. In addition, the intestine was also proven to be a major toxicity and metabolic target organ for antibiotic exposure in teleost [10]. He et al. [11] suggested that antibiotics exposure could disorganize epithelial structure, increase intestinal permeability, and then induce oxidative stress. In spite of this, previous studies have mainly focused on risk evaluation and exploring the potential functions of the liver and kidney when treated with antibiotics [12,13]. Recent research illustrated that intestine homeostasis is essential to organism health. Chronic exposure of sulfamethoxazole caused irreversible oxidative-stress-cascaded damage to gut and led to apoptosis in grass carp (Ctenopharyngodon idella) [14]. Moreover, the gut is known as a reservoir of antibiotic resistant genes under selective pressure from antibiotic exposure [15]. Collectively, these studies revealed that antibiotics can further damage intestinal health by destroying intestinal histological structure, elevating oxidative stress levels and inducing apoptosis. More evidence demonstrated that gut microbiota has a profound relationship with host immunomodulation, physiology and pathogen defense [16]. Commensal gut microbiota regulate host metabolism in multiple ways, including synthesizing and absorbing nutrients, strengthening gut barrier integrity, producing beneficial metabolites and other functions [17]. However, physical and chemical environmental factors (e.g., oxygen, heavy metals and pesticides) can affect bacterial community diversity and richness, impairing the intestinal barrier and triggering a series of metabolic diseases [18,19,20]. Recent studies suggest that antibiotic exposure also adversely affects the intestinal function of aquatic organisms. Qian et al. reported that three veterinary antibiotics (doxycycline, oxytetracycline and FF) which were received widespread use in China caused gut microbiota dysbiosis and dysfunction in zebrafish (D. rerio), leading to metabolic disorders [9]. Additionally, sulfamethazine changed the composition of gut microbial communities, downregulated alpha diversity and induced oxidative stress in marine medaka (Oryzias melastigma) during 30 days of oral exposure [21]. Thus, a functional and stable gut microbiota is critical for host physiology and intestinal homeostasis. These studies were also based mainly on short-read amplicons, producing higher proportion of inaccurate classification, especially at genus and species level. Nowadays, newly developed full-length 16S rRNA gene sequencing methods could cover V1-V9 hypervariable regions and conduct a highly resolved classification of intestinal community composition [22]. Crustaceans are vulnerable to antibiotics. Accumulated FF in crustaceans can cause hepatopancreatic structural damage and oxidative stress [23], immunosuppressive effect [24] and biomolecule damage [25]. The Chinese mitten crab (Eriocheir sinensis) is an economical freshwater aquatic animal widely bred in China with an annual production of more than 775,887 tons in 2021, according to the China Fishery Statistical Yearbook. The crab suffers from various antibiotics via intensive farming patterns, diet and the deterioration of the ecological environment. More importantly, E. sinensis is also reported to be a model organism for monitoring the ecotoxicological effects of antibiotics in aquatic environment [26]. However, to our knowledge, systematical studies of FF-induced toxicity and gut barrier imbalance in the Chinese mitten crab are scarce. Thus, considering the crucial role of gut microbiota in host intestinal homeostasis, the aim of this study was to identify chronic toxic effects of FF on the intestinal health of E. sinensis under environmental and extreme concentrations. Chinese mitten crabs were exposed to 0.5, 5 and 50 μg/L FF for a 14-day period. Toxicological changes in gut histological morphology, oxidative stress and apoptosis characteristics were evaluated. In addition, full-length 16S rRNA sequencing technology was carried out to analyze microbiota composition alterations in sides of exposure time and concentration. Results of this study will facilitate a better understanding of FF-induced toxicology on gut health of crustaceans. Intestinal morphological changes were monitored in crabs with H&E staining between control and FF treatment groups. Clearly damaged characteristics were observed. As shown in Figure 1A, the intestine in the control group had normal morphology, well arranged epithelial cells and peritrophic membrane. However, crabs treated with 50 μg/L FF exhibited markedly reduced villus lengths and significantly thinner muscle layers (p < 0.05) at either day 7 or 14 (Figure 1B,C). Although the villus length shortened slightly, the apical epithelium of intestinal villi was shed in LC and MC groups. Furthermore, the number of vacuole increased as the FF concentration increased at day 14. Transcriptional changes of ZO-1 were detected to indicate the changes of gut permeability. As apparent from Figure 1D, the transcripts of ZO-1 were all augmented in LC and MC groups on day 7 and strongly inhibited on day 14 in the three treated groups (p < 0.01). Compared with the control group, the THC noticeably decreased even in the LC group at day 7 and 14 (Figure 1E). In addition, the amount of THC was inversely proportional to the exposure concentration of FF. After detection of FF by HPLC, the concentration of FF showed a dose-dependent increase in the model of the crab intestine (Figure A1). In details, the FF contents in HC group were dramatically higher than those in LC and MC groups (p < 0.05). No significant changes were observed between LC and MC group. TUNEL assay was used to evaluate the effect of FF exposure on apoptosis (Figure 2A). Obvious apoptosis characteristics were observed in FF treated groups at day 7 and 14, although weak signals were detected in the control group. The apoptosis index was dramatically increased in experimental groups (peak at 33.05 ± 4.71%) compared with the control group (5.46 ± 1.41% and 10.42 ± 2.57%) on both days 7 and 14 (p < 0.05) (Figure 2B). In day 7, the apoptosis index upregulated in a dose-dependent pattern, with the HC group showing the highest rate. Apoptotic cells first appeared in epithelial cells and were then located around the lamina propria. Relative expression of apoptosis-related genes in gut was tested to evaluate the influence of FF exposure on cell death (Figure 3A). At 7 days after exposure, the mRNA levels of Caspase3 and Caspase 8 were significantly elevated as the FF levels increased compared to the control (p < 0.05). Similar results were obtained in Bax. No significant changes exist in p53 gene. In contrast, as an antiapoptotic-related gene, the transcripts of Bcl-2 noticeably decreased (p < 0.01) in the three FF treated groups, indirectly leading to the upregulated ratio of Bax/Bcl-2 (p < 0.01). However, on day 14, the relative expression pattern of Caspase3, Bax and Bcl-2 showed opposite trends. Among them, even in the LC group, the mRNA levels of Caspase3 and Bax were dramatically decreased (p < 0.001). Similarly, the ratio of Bax/Bcl-2 were of 2.9-fold, 9.7-fold and 21.3-fold lower than the control group in the LC, MC and HC groups, respectively. Simultaneously, the relative expression of p53 was strongly inhibited by FF with increased dosage (p < 0.01 or p < 0.001). In the case of Caspase8, significant change only occurred in the LC group compared to control group (p < 0.001), which was different than the profile on day 7. Levels of key genes in immune response were determined via qRT-PCR analysis, which is displayed in Figure 3B. The expression levels of MyD88 were greatly upregulated in three treated groups and peaked in the HC group (p < 0.001), either at day 7 or day 14 postexposure. Regarding the transcription of antimicrobial peptides, ALF1, Relish and Dorsal mRNA downregulated sharply (p < 0.001) on day 14, showing a high degree of consistency. These results revealed the influence of FF on Chinese mitten crab immunity. To access the oxidative stress levels induced by FF exposure, antioxidant enzymes CAT and SOD activities were measured. Compared with the control group, crabs in three FF treatment groups showed lower CAT activities on day 7 (p < 0.05) (Figure 4A). In contrast, CAT contents increased significantly (p < 0.001) in dose-dependent manner on day 14. Moreover, SOD activities stayed suppressed (p < 0.001) from 7 to 14 days compared to group control (Figure 4B). Full-length 16S rRNA gene high-throughput amplification was employed in 12 intestinal content samples via the PacBio platform. A total of 194,629 original CCS sequences (range from 12,873 to 13,096) were produced in the current microbiome analysis. The Good’s coverage of all samples exceeded 99.91%, suggesting that the sequencing production could represent the majority of bacteria in crab intestine of this study. The rarefaction curve (Figure A2) also reflected sufficient sequencing depth. Alpha diversity of the intestinal microbes in control, LC, MC and HC group after 14 days of FF exposure were shown in Table 1. Compared with those in control group, Shannon and Simpson indexes were remarkably downregulated only in the HC group (p < 0.05), while the Chao1 and ACE indexes were statistically the same in all four groups, indicating no changes in bacterial richness. An NMDS plot based on Bray–Curtis dissimilarity was generated to visualize the β-diversity among groups (Figure 5A). Interestingly, compared with MC and HC group, LC group had a tendency toward variation distribution in community composition approach at the control level. The microbiota compositions at the phylum level were further analyzed. As shown in Figure 5B, Spirochaetota was newly detected in the gut microbial communities of MC and HC groups, although the relative abundance was low with the value of 0.02%. The top five most abundant phyla were Bacteroidota (52.82%), Pseudomonadota (23.86%), Bacillota (23.86%), Mycoplasmatota (4.53%) and Campilobacterota (4.73%) in the control group. Metastats analysis suggested that changes of bacterial phyla only occurred in the HC group when the Bacteroidota sharply increased to 71.82% (p < 0.001) and Pseudomonadota significantly decreased to 12.44% (p < 0.01) compared to the control. No significant difference was detected in the LC and MC groups. Eleven OTUs which showed significant differences among the four groups were chosen for heatmap analyses (Figure 5C). Compared to those in the control group, FF exposure decreased the proportions of Flavobacterium, Roseimarinus (out2), Pseudomonas and Shewanella (p < 0.05). Seven OTUs increased significantly in the antibiotic exposed groups, and those OTUs were distributed in Bacteroidota, Bacillota, Pseudomonadota and Mycoplasmatota. Interestingly, OTU2 and OTU4, which both belong to Roseimarinus, had opposite directions of variation. We also noticed that most significant alterations occurred in the HC group, including OTU20 (Flavobacterium), OTU2 (Roseimarinus), OTU4 (Roseimarinus), OTU1 (Parabacteroides), OTU24 (Vagococcus), OTU13 (Lactovum), OTU9 (Pseudomonas) and OTU33 (Shewanella). These results indicated that high concentration exposure to FF influenced the intestinal microbial composition more deeply than other two concentrations as compared to the control on day 14. To compare the alterations of the intestinal microbial community after 0.5 μg/L of FF exposure at 7 and 14 days, full-length 16S rRNA sequencing was conducted. The Venn diagram (Figure 6A) reflects that 78 OTUs were co-owned between day 7 and day 14, and the unique OUTs were eight and six, respectively. Community diversity was determined by Shannon indices and showed no significant difference between two groups (p > 0.05, Figure 6B). However, the bacterial richness markedly downregulated with the prolonging of FF exposure (p < 0.05). The nMDS analysis further confirmed that differential microbiota patterns existed in crabs from those two groups (Figure 6C). The taxa of dominant bacteria were analyzed at the phylum, family and genus levels. At the phylum level (Figure 6D), the dominant phyla in the two groups were Bacteroidota, Pseudomonadota, Bacillota, Mycoplasmatota and Campilobacterota. The abundance of these five detectable phyla did not significantly alter. Notably, Spirochaetota was absent in the 14-day treatment. Moreover, at the family level, Tannerellaceae, Lachnospiraceae, uncultured_bacterium_o_Clostridiales and Pseudomonadaceae were strongly enriched (p < 0.05). In addition, long-term FF exposure noticeably reduced the abundance of Streptococcaceae, Vibrionaceae and Prolixibacteraceae (p < 0.05, Figure 6E). At the genus level, 68 genera were identified from all samples, and the top 21 relative abundances are shown in Figure 6F. Metastats analysis showed that Roseimarinus, Lactovum and Vibrio were all decreased significantly on day 14 (p < 0.05 or p < 0.01), while Parabacteroides and Pseudomonas were more abundant on day 14 compared to day 7 (p < 0.05 or p < 0.001, Figure 6G). LEfSe was employed to indicate the high-dimensional biomarkers in the LC group that was exposed for 7 and 14 days. As described in the cladogram (Figure 7A), two bacterial taxa contributed to the 7-day group, including Bacillota and Vibrionales. LEfSe analysis also revealed that the relative abundance of Bacteroidota and Pseudomonadales were more abundant in long-term FF exposure. At the genus level, LDA analysis with cut-off value of 4.0 identified Lactovum, Vibrio and Chryseobacterium as the indicator bacteria in the intestine of crabs in 7-day exposure, whereas Pseudomonas and Parabacteroides were the signature bacteria in the 14-day group (Figure 7B). To further understand the changes in metabolic functions, PICRUSt was performed in LC group (Figure 8). Observation of metabolic pathway revealed that “secretion system” was the dominant microbial function detected in LC groups during exposure. Interestingly, the proportion of “glycerophospholipid metabolism” and “biosynthesis of unsaturated fatty acids” expanded with the prolonging of FF treatment, suggesting that FF could affect host lipid metabolism. Additionally, pathways in glycose metabolism, including “starch and sucrose metabolism”, “galactose metabolism”, and “fructose and mannose metabolism”, were enriched in the early exposure stage. After combining the results of the metastats analysis and the genus classification level in LEfSe analysis, Spearman correlations between gut microbiota and biochemical indices of the Chinese mitten crab were conducted. As described in Figure 9, apoptosis-related genes and antimicrobial peptides were positively correlated with Lactovum, Vibrio and Roseimarinus, and negatively associated with Parabacteroides (p < 0.05 or p < 0.01). In addition, Parabacteroides and Vagococcus showed strong positive correlations with CAT activities and expression of MyD88 and Bcl-2 (p < 0.05 or p < 0.01). Antibiotics are widely used in aquaculture as feed additives and therapeutics to promote growth rate and prevent bacterial infections, respectively. Recently, consumption of antibiotics has raised concerns for its prevalence and high residue rate in foods, such as milk, eggs and aquatic products [27]. Previous studies reported that aquatic animals are more susceptible to low concentration of antibiotics, which induced antibiotic resistance [28]. Excessive antibiotics destroy intestinal structure in teleost, leading to gut dysfunction [10,29]. However, relationships between crustacean intestine and microbiota under environmental and extreme doses remains uncertain. Here, an florfenicol poisoning model was conducted in Chinese mitten crabs to explore the potential mechanism of florfenicol-induced intestinal damage. In the present study, histopathological structure, oxidative stress and apoptosis pathways, as well as gut microbial alteration, were examined in adult E. sinensis exposed to 0, 0.5, 5 and 50 μg/L FF. As a major organ for nutrient digestion and absorption, the intestine of aquatic animals is also intimately involved in host metabolism, immune defense and other crucial activities. All of these procedures require a healthy intestinal structure and complete intestinal barrier [30]. Physical, biochemical and immunological barriers are the three main components of an intact intestinal barrier. According to histopathologic analysis, significant changes were observed in intestinal villus architecture with apical epithelial shedding and villi shortening, implying increased permeability and susceptibility of intestinal barrier after FF exposure. Exposure to 100 μg/L FF also resulted in intestinal morphology and structure damage to zebrafish [9]. Furthermore, expression levels of ZO-1 were detected. Working as a tight junction protein, ZO-1 formed a physical barrier by tightening intestinal epithelial cells to strengthen intestinal function, which caused damage due to increased intestinal permeability [31]. As expected, FF inhibited ZO-1 expression after 14 days of exposure. Previous studies found that adding 2.0 g/kg oxytetracycline to the diet of Nile tilapia (Oreochromis niloticus) reduced the expression of ZO-1 and CLDN3 [29]. Qian et al. reported that exposure to environmentally relevant concentrations of sulfamethoxazole (0.3 μg/L) suppressed ZO-1 expression and induced inflammation and apoptosis in grass carp [9]. However, ZO-1 mRNA levels significantly increased in LC and MC groups at 7 days, indicating an epithelial structure reinforcing phenomenon, which is generally consistent with the histopathologic analysis. Antibiotics were used early as feed additives to promote the growth trait in aquaculture, which might be involved in ZO-1 upregulation on day 7. Additionally, in crustaceans, hemocytes are closely related to innate immune functions. Thus, the immunological status of aquatic animals could be directly reflected through THC. In the present study, FF-treated crabs showed a significant drop in THC. Similar results were reported in mud crabs (Scylla paramamosain) [32]. Decreased HTC might be caused by cell apoptosis [33]. These results suggest that high FF exposure increases intestinal permeability and damages the intestinal barrier, leading to serious destructions in crabs. The extent of intestinal damage is limited in environmental dose exposure. Antibiotics-induced excessive reactive oxygen species (ROS) can elevate oxidative stress in aquatic animals, leading to severe oxidative stress damage [12,34]. As the first part of the antioxidant system, SOD and CAT can eliminate overproduction of ROS under normal conditions, and therefore reflect the overall antioxidant status of organisms. Previous studies on swimming crab (Portunus trituberculatus) reported noticeably decreased SOD activities after intravenous dosing treatment of FF [25]. Chen et al. [35] observed significant decrease of SOD and CAT activities in a dose-dependent manner after 6 days of sulfamethoxazole treatment in marine mussels (Mytilus galloprovincialis). Similarly, in the present study, activities of SOD and CAT were both significantly decreased after 7 days of exposure in three FF exposed groups, suggesting that even the ng/L concentration would cause too much ROS to damage the antioxidant system. Nevertheless, as the FF stress was prolonged, different profiles appeared between SOD and CAT, with a dramatic increase of CAT activity on day 14. CAT has the ability to degrade hydrogen peroxide into molecular water and oxygen. The recovery of CAT activity revealed that oxidative stress of organism is weakened. In general, the reduction of CAT and SOD suggested the degradation of antioxidative status in Chinese mitten crab under FF exposure. Thus, these antioxidative enzymes are widely used as biomarkers in ecological risk assessment [36]. In present results, CAT showed a dose-dependent manner which indicated the feasibility and high efficiency as biomarker for florfenicol stress in crustaceans. In the TLR signaling pathway, MyD88 activates a string of downstream signaling cascades, and finally increases the expression of NF-κB [37]. NF-κB is a key protein in host inflammation and innate immune response in invertebrates [38]. Evidence suggests that high doses of maduramicin contribute to intestinal inflammatory in crayfish (Procambius clarkii) [39]. In this study, MyD88 mRNA transcripts were maintained at a relatively high level during all 14 days of exposure, suggesting the occurrence of intestinal inflammatory response in crabs. In addition, AMPs, the major antibacterial effectors activated by NF-kB pathways, would be secreted into extracellular space to resist environmental pollutions stress and microbial invaders in crustaceans [40]. In the present study, long-term exposure of FF markedly inhibited the transcription of three AMPs, namely, ALF1, Relish and Dorsal. In contrast, at 7 days of exposure, significant changes were found only in the HC group, which is consistent with the results of THC. To reflect the degree of injury in intestine tissue, cytometric TUNEL assay and expression of apoptosis-related genes were conducted. Previous investigation revealed that the apoptosis rate of zebrafish embryonic cells was enhanced with a dose-dependent pattern of 4-Epianhydrotetracycline [41]. The apoptosis index markedly upregulated at both day 7 and 14 compared with the control group, indicating that FF exposure not only damaged the oxidative system but also induced intestinal cell apoptosis. Correspondingly, this point was proved by the expression pattern of apoptotic-related genes. Zhao et al. reported that the Bax/Bcl-2 protein ratio was upregulated in grass carps after exposure to cypermethrin or/and sulfamethoxazole in a 42-day period [42]. Likewise, the ratio was almost threefold higher comparing to the blank control after 7 days of FF exposure. However, as the exposure was extended to 14 days, completely reversed expression profiles of these apoptosis-related genes were observed. Apoptosis functions in programmed cell death by eliminating excess, infected or damaged cells [43]. Combined with SOD activities, we speculate that this phenomenon might be correlated with the regulation of intestinal homeostasis. Accumulating studies illustrated that gut microbiota has a close relationship in maintaining host intestinal barrier integrity, oxidate stress and immunity in aquatic animals when facing with environmental contaminants [9,20]. Here, we used full-length 16S rRNA sequencing technology to investigate dynamic changes on crab gut microbiota diversity at dosage and time dimension. After 14 days of exposure, α diversity index was only remarkedly decreased in the HC group compared with control, indicating that intestinal homeostasis worked in the LC and MC groups. At the phylum level, relative content of Bacteroidota increased obviously while Proteobacteria decreased after 50 μg/L FF exposure, which was in accord with one previous study on E. sinensis subjected to imidacloprid [20]. Together, compared with bacterial quantity, bacterial composition and diversity were more sensitive to high FF exposure. After 75 mg/kg per day oxytetracycline oral treatment, only bacterial composition altered in Atlantic salmon (Salmo salar) intestine [44]. This could be explained by increased abundance of antibiotic-resistant bacteria and drop in antibiotic-sensitive bacteria. OUTs in LC group at 7 and 14 days were relatively few compared to previous studies of antibiotic exposure [9]. This might be attributable to FF residues which eradicated susceptible microorganisms [44]. Furthermore, in contrast with alpha diversity in the four groups at day 14, bacterial quantity indicated by ACE index was changed dramatically. Although no statistical significance changes were seen at the phylum level, the trend of bacterial composition was highly consistent with the results of Hong et al. [20], indicating a conserved biological function of gut microbiota. Bacteroidota participate in immunomodulation and lipid metabolism [45]. Likewise, microbial function also predicted that lipid metabolism is the dominant function of gut bacteria in long-term exposure. Absence of Bacteroidota and induced Bacillota were proved to be associated with a deteriorating of non-alcoholic fatty liver disease leading to hepatic steatosis [46]. The Bacillota/Bacteroidota ratio, correlated with metabolic disorders, was higher in day 14. At the family level, the relative abundance of Tannerellaceae increased with prolonged exposure. The presence of the Tannerellaceae family is reported to have beneficial effects in therapy of collagen-induced mouse arthritis [47]. In addition, some other nondominant families in gut microbiota also changed in FF treated crabs. Pseudomonadaceae, regarded as a degrader of organic pollutants [48], was significantly increased at day 14, implying the removal of antibiotics from crabs. Ratio of Lachnospiraceae/Streptococcaceae is known to be negatively associated with metabolic disorder and proinflammatory processes [49]. A similarly increased ratio was observed in our findings. Results from metabolic function prediction demonstrated that this change is responsible for glycolysis and lipid metabolism in crab intestines, giving us a hint about the function of gut microbiota in FF metabolism. Finally, we explored the bacterial composition at the genus level. Parabacteroides, regarded as beneficial bacteria in amino acid metabolism that also participate in proteolytic functions [50], showed strong augmentation in crabs with long-term exposure in the present study. Notably, the abundance of opportunistic pathogens Vibrio was significantly downregulated. This might be attributed to the broad-spectrum antibacterial function of FF. Our results are in accordance with a report from [51], which said that nanoplastics could invert the proportion of Parabacteroides and Vibrio in the large yellow croaker (Larimichthys crocea). Moreover, Pseudomonas can efficiently decompose organic contaminants [52]. Thus, based on the function of metabolizing xenobiotic compounds, we speculate that gut microbiota contributes to host defense by degrading and detoxifying FF in aquatic crustaceans. Correlation analysis provides the associations between physiological indicators and gut bacteria during FF exposure. In this study, Lactovum, Vibrio and Roseimarinus were positively associated with ZO-1, AMPs and apoptosis-related genes, while Parabacteroides showed strongly negative correlation. However, little is known about Lactovum and Roseimarinus in aquatic animals. These two genera may have similar functions with Vibrio through correlation analysis. Underlying mechanisms of host-gut microbiota interactions need to be further confirmed by in vitro experiments. Florfenicol (FF, CAS No.: 73231-34-2, 98% in purity, Aladdin, China) was purchased from Aladdin Co., Ltd., Shanghai, China. FF is dissolved in dimethyl sulfoxide (DMSO, Solarbio, Beijing, China) with a concentration at 1 mg/mL as a stock solution. At days 7 and 14 after exposure, FF contents in the intestines were detected by HPLC. Healthy crabs (48.5 ± 4.5 g) were collected from a local farm in Gaochun, Jiangsu Province, China. To eliminate the influence of sex differences, only male individuals were used in the experiment. All experimental animal procedures were in accordance with the principles of Institutional Animal and Use Committee of the Nanjing Normal University. In total, 120 crabs were randomly divided into four triplicate groups and distributed in 12 tanks (60 × 40 × 30 cm; length, width, height) that each contained 10 L water. Before exposure, all crabs were cultured for two weeks to acclimatization to the following conditions: 12 h light/dark cycle; pH: 7.9 ± 0.3; temperature: 24.5 ± 2.0 °C; dissolved oxygen: 6.3 ± 0.7 mg/L; ammonia nitrogen < 0.2 mg/L. Crabs were fed with commercial feed (Tongwei, China) twice a day at the ration of 2% of body weight. According to the environmental concentrations in surface water and sediment of aquaculture area [53] and previous studies [9], concentrations of FF conducted in the present immersion exposure experiment were set to 0.5, 5 and 50 μg/L, referred to as low concentration (LC), median concentration (MC) and high concentration (HC), respectively. The group without any FF addition was used as control, identified as C in the following text. During the treatment, one third of the water was replaced daily, and FF was added in the treated tanks to maintain the concentration at the initial levels throughout the experiment. After 7 and 14 days of consecutive exposure, five crabs from each group were captured and the intestine of each individual was quickly sampled for enzyme activity determination and RNA isolation. Simultaneously, the contents of intestine were also squeezed for microbiome analysis. Samples were frozen in liquid nitrogen immediately and then stored at −80 °C for further analysis. Intestines from two crabs were randomly selected from each group at days 7 and 14 postexposure, and then fixed in paraformaldehyde (4%) solution for 24 h. Solid wax blocks were made by dehydrating in ethanol, rinsing in toluene, equilibrating in xylene and then, finally, embedding in paraffin. The intestine sections were then subjected to hematoxylin and eosin (H&E) staining. Villus length and the thickness of the muscle layer were measured using a light microscope (Nikon, Tokyo, Japan). Hemocytes (500 μL per crab) were collected individually from the third pereiopod of crab, and then stirred and fixed with formalin (10%) for 10 min at room temperature. Then, the treated cells were stained using the Giemsa method and uniformly placed in a hemocytometer, forming a single layer without overlapping. Finally, the number of hemocytes was counted under an inverted microscope (Nikon, Tokyo, Japan). TUNEL staining was performed to detect the apoptosis that occurred in the intestine of E. sinensis after FF exposure. Apoptosis cells were identified using the colorimetric TUNEL apoptosis assay kit (Beyotime, Shanghai, China). A Nikon Ti-E-A1R (Japan) fluorescent microscope was used to observe apoptosis. Cell nuclei were stained with DAPI and visualized in dark blue. Apoptotic cells were stained green. Six views of each sample were randomly picked, and total numbers of normal and apoptotic cells were counted respectively in view to calculate the apoptotic index. The intestines from control and from each FF exposure group were homogenized and centrifuged. Then, the concentrations of superoxide dismutase (SOD) and catalase (CAT) were measured with commercial assay kits (Jiancheng Institute, Nanjing, China). All results were normalized to total protein content in the respective sample for comparison. Each assay was run with five replicated samples. Total RNA was extracted from intestinal tissues using TRIzol reagent (Invitrogen, Waltham, MA, USA). The amount of RNA was measured by NanoDrop Spectrophotometer (Thermo Scientific, Waltham, MA, USA). Purified RNA was reversed into cDNA by HiScript® Reverse Transcriptase (Vazyme Biotech, Nanjing, China) and then immediately stored at −20 °C for qRT-PCR. Expression of zonula occludens-1 (ZO-1), apoptosis- (Caspase3, Caspase8, p53, Bcl-2 and Bax) and immune-related (MyD88, ALF1, Relish and Dorsal) genes was measured. Hieff® qPCR SYBR® Green (Yeasen Biotechnology, Shanghai, China) and a Roche LightCycler96 were used for qRT-PCR detection. EF1-α was used as an internal reference gene. Melting curves were generated to investigate the specificity of each amplified products. Relative gene expression was calculated by 2−ΔΔCT method [54]. All samples were run in triplicates. Detailed primer sequences are presented in Table A1. Total bacterial community DNA was extracted by using the Power Soil® DNA Isolation kit (MoBio, Carlsbad, CA, USA) through the manufacturer’s instructions, and was then quantified for further use. Next, the bacterial full-length 16S rRNA gene was amplified with the primers 27-forward (5′-AGRGTTTGATYNTGGCTCAG-3′) and 1492-reverse (5′-TASGGHTACCTTGTTASGACTT-3′). Then, the purified PCR products were mixed in equidensity ratios and used to construct sequencing libraries by the PacBio sequencing platform (Biomarker-Technologies Company, Beijing, China). SMRT-Link (V8.0) software was utilized to demultiplexed raw subreads and aligned it to CCS (Circular Consensus Sequencing) sequence. Sequences with more than 97% identity were clustered to one operational taxonomic unit (OTU) using USEARCH (V10.0). Chimeric and singleton sequences were also filtered. Data processing was conducted and displayed by R software (V3.5). Alpha diversity analysis, including richness abundance (Chao1 and ACE) and diversity (Simpson and Shannon) of communities, were implemented in QIIME (V2.0) software. Nonmetric multidimensional scaling (nMDS) method was performed based on Bray–Curtis similarity to evaluate the beta diversity in samples between groups. In addition, the relative microbial abundance and composition from phylum classification level to genus were also calculated. Significant differences of taxa in each group at phylum, family, and genus level were determined by metastats analysis. Eleven OTUs were chosen for heatmap comparison based on the significant abundance changes under FF-bath treatment on day 14 using the one-way ANOVA method. Linear discriminant analysis (LDA) coupled with LDA effect size (LEfSe) was applied to find the biomarkers with the threshold of 4.0 [55]. PICRUSt was used to predict potential microbial metabolic function [56]. Finally, associations between biochemical indices and generic abundance were calculated by Pearson correlation analysis and shown in heatmap. All data were expressed as the mean ± standard errors of the means (SEM). Statistical analysis was carried out by one-way ANOVA, followed by LSD test, for the comparison between multiple groups using SPSS software (25.0, IBM, Chicago, IL, USA). Different letters (a, b, c, d) and asterisks “*” indicate significant difference among control and three florfenicol exposed groups (p < 0.05). Statistical results are shown in Supplementary Materials. In summary, FF exposure significantly affected E. sinensis gut health by damaging the intestinal barrier function, showing clear histological morphology variation and increased intestinal permeability. Meanwhile, exposure to FF induced and activated the antioxidant and apoptosis system to protecting cells from oxidative damage. Furthermore, the accumulation of FF in crabs resulted in declined microbial diversity and shifted community structure. Our results also suggest that the gut microbial diversity was closely related to crabs’ antioxidant and apoptosis systems, hence demonstrating their important roles in mitigating the effects of toxic stress on crabs’ intestine. Overall, this study highlights the essential role of intestinal homeostasis of crustaceans facing with antibiotic polluting systems.

PMC10002401

Stefan Titu,Vlad Alexandru Gata,Roxana Maria Decea,Teodora Mocan,Constantin Dina,Alexandru Irimie,Cosmin Ioan Lisencu

Exosomes in Colorectal Cancer: From Physiology to Clinical Applications

23-02-2023

exosome,cancer,colorectal cancer,exosomal miRNA,lncRNA

Exosomes are nanosized vesicles that have been found to be involved in many diseases. Exosomes can mediate communication between cells in a variety of ways. Certain types of mediators derived from cancer cells can play a crucial role in the development of this pathology, promoting tumor growth, invasion, metastasis, angiogenesis, and immunomodulation. Exosomes in the bloodstream show promise as a future tool for detecting cancer at an early stage. The sensitivity and specificity of clinical exosome biomarkers need to be enhanced. Knowledge of exosomes is not only important for understanding the significance of cancer progression but also for providing clinicians with useful information for the diagnosis, treatment, and discovery of methods to prevent cancer from recurring. The widespread adoption of diagnostic tools based on exosomes may revolutionize cancer diagnosis and treatment. Tumor metastasis, chemoresistance, and immunity are all aided by exosomes. A potential new approach to cancer therapy involves preventing metastasis by inhibiting miRNA intracellular signaling and blocking the formation of pre-metastatic niches. For colorectal patients, exosomes represent a promising area of investigation for improving the diagnosis, treatment, and management. Reported data demonstrate that the serum expression level of certain exosomal miRNA is significantly higher in primary colorectal cancer patients. The present review discusses mechanisms and clinical implications of exosomes in colorectal cancer.

Exosomes in Colorectal Cancer: From Physiology to Clinical Applications Exosomes are nanosized vesicles that have been found to be involved in many diseases. Exosomes can mediate communication between cells in a variety of ways. Certain types of mediators derived from cancer cells can play a crucial role in the development of this pathology, promoting tumor growth, invasion, metastasis, angiogenesis, and immunomodulation. Exosomes in the bloodstream show promise as a future tool for detecting cancer at an early stage. The sensitivity and specificity of clinical exosome biomarkers need to be enhanced. Knowledge of exosomes is not only important for understanding the significance of cancer progression but also for providing clinicians with useful information for the diagnosis, treatment, and discovery of methods to prevent cancer from recurring. The widespread adoption of diagnostic tools based on exosomes may revolutionize cancer diagnosis and treatment. Tumor metastasis, chemoresistance, and immunity are all aided by exosomes. A potential new approach to cancer therapy involves preventing metastasis by inhibiting miRNA intracellular signaling and blocking the formation of pre-metastatic niches. For colorectal patients, exosomes represent a promising area of investigation for improving the diagnosis, treatment, and management. Reported data demonstrate that the serum expression level of certain exosomal miRNA is significantly higher in primary colorectal cancer patients. The present review discusses mechanisms and clinical implications of exosomes in colorectal cancer. Exosomes are nanosized vesicles that have been found to be involved in many diseases. They are secreted by various cell types upon the fusion of multivesicular bodies and the plasma membrane [1]. Exosomes are typically 40–150 nm in diameter and carry nucleic acids, proteins, lipids, and metabolites [2]. Exosomes eventually generate multivesicular endosomes (MVEs) that are secreted into the extracellular space to travel to other cells [3]. Originally, when released from cells, exosomes were considered cellular garbage collectors following cell degradation or loss of cellular homeostasis without playing an important role in the surrounding body cells. However, more recent findings have showed that they mediate cell–cell communication, being loaded with proteins, lipids and nucleic acids that are delivered to target cells, and they are able to alter the biological behavior of the recipient cells [4]. Various surface molecules are shown to be responsible for the interaction between extracellular vesicles and recipient cells for their uptake. After they bind to the target cell, several processes may occur, receptor–ligand interaction, endocytosis and/or phagocytosis or membrane fusion and further load delivery into the cytosol and the subsequent change in the physiological state of the recipient cell [4]. There have been several studies where all membrane-bound vesicles are largely cited as extracellular vesicles and not particularly referred to as exosomes, microvesicles or other subtypes. Nevertheless, it is necessary to clearly distinguish exosomes from other extracellular vesicles in order to comprehend their action and compare various study results [5]. The biogenesis of exosomes involves their origin in endosomes, and they exhibit membrane protein expression profiles involved in membrane transport and fusion such as Rab GTPases, annexins and flotillin, components of the ESCRT complex, integrins and tetraspanins, including CD9, CD63, and CD81 [6]. One of the basic functions of exosomes is the elimination of excessive proteins or undesirable molecules from the cell, but they are important mediators of intercellular communication and are involved in various pathways being biologically active vesicles released into the extracellular environment [1]. Exosome engineering through genetic and chemical methods for targeted drug delivery may help increase their therapeutic applicability as clinical biomarkers [7]. There are still a lot of aspects to be considered for the design of new cancer treatment strategies, but exosomes exhibit great potential in precision cancer medicine. Figure 1 is broadly depicting all clinical applications that exosomes may have. As exosomes have proved their key role in cancer processes, there are three main research areas with clear participation in cancer progression: exosomes can modulate host immune response and induce immune tolerance; exosome crosstalk with the tumor microenvironment promotes tumor growth and progression; and their significant role in metastasis [2]. More exosomes are produced and released by cancer cells than by healthy ones, and the molecules found in exosomes released by tumor cells are very different from those found in healthy ones. Recent studies have shown that there are substantial differences between colorectal cancer (CRC) patients and healthy controls in the levels of certain microRNAs (miRNAs), long non-coding RNA (lncRNAs), and proteins found in exosomes isolated from blood (NCs). Some research suggests that these exosomal molecules can serve as markers for colorectal cancer. There have been various studies on the role exosomes play in immune regulation, with a more recent one focusing on how exosomes regulate the immune response [8]. It has been demonstrated that human Epstein–Barr virus-infected B cells secrete exosomes carrying Major Histocompatibility Complex (MHC) classes I and II, thus indicating their potential implication in the modulation of immune responses [9]. This finding has triggered numerous other studies that have confirmed that exosomes secreted by antigen-presenting cells, for example, DCs, express class I, class II MHC, adhesion, and co-stimulatory molecules. Such features allow exosomes to directly activate CD8+ and CD4+ T-cells and induce a strong immune response [1]. Peptide-pulsed dendritic cells release immunogenic exosomes and stimulate a strong CD8+ T-cell-dependent anti-tumor immune response [10]. Exosomes derived from cancer cells express tumor antigens able to activate dendritic cells, therefore determining immune priming and triggering a specific cytotoxic response superior to the immunogenicity of tumor cell lysates or soluble antigens in vaccines [11]. It has been shown that one intraperitoneal injection of tumor peptide-loaded dendritic cell-derived exosomes can trigger a very powerful immune response that could lead to tumor growth delay or tumor rejection [12]. While this could be attributed to high antigen density, it is also due to the presence of heat shock proteins as seen in the case of exosomes produced by melanoma cells [13,14]. Exosomes trigger immune response suppression, leading to the low immunogenicity observed in several studies. Exosomes derived from cancer cells can suppress natural killer cells by downregulating NKG2D expression [15]. Dendritic cell maturation is impaired in vivo by tumor cell-derived exosomes, therefore leading to immunosuppression. Breast cancer cell-derived exosomes are internalized by bone marrow myeloid precursors, impairing dendritic cell differentiation by promoting IL6 overexpression and Stat3 phosphorylation [16]. Subsequent research showed that bone marrow precursor cells isolated from an IL6 knockout (KO) model can differentiate into dendritic cells following treatment with exosomes derived from cancer cells. Altogether, these study results indicate the immunosuppressive potential of tumor cell-derived exosomes via NK and DC modulation. Still, not all findings can identify the effector molecules initiating the modulation of the immune response [2]. Cancer progression is determined by the crosstalk between cancer cells and the neighboring cells. This type of cell-to-cell communication is based on dynamic information exchange, inducing a pro-tumor microenvironment where carcinogenesis occurs and the immune response is modulated in order to promote tumor progression and survival [1]. Exosomes are essential components of the intercellular microenvironment, acting as regulators of cell-to-cell communication. It has been widely demonstrated that exosomes can induce phenotypic changes in neighboring cells through the activation of specific cell-signaling pathways leading to cancer progression [17]. Extensive studies have been carried out on intracellular communication, mainly during tumor development. Exosome-associated RNAs, miRNAs, proteins, DNAs, and even metabolites are able to determine changes in the outcome of recipient cells via autocrine and paracrine signaling mechanisms. Exosomal proteins are able to modulate the outcome of exosome-secreting cells through autocrine signaling. More specifically, chronic myeloid leukemia-derived exosomes contain TGFβ1, a cytokine that binds to the TGFβ1 receptor in leukemia cells and further promotes tumor growth by the activation of ERK, AKT and anti-apoptotic pathways in producer cells [5]. Some of their characteristics make exosomes superior to other extracellular vesicles for use as therapeutic agents, such as their stability in vivo and in vitro, bioavailability, good distribution into the surrounding body fluids, their ability to successfully cross the blood–brain barrier, good tolerance and regulation of gene expression by transferring miRNA and siRNA into target cells. All these features indicate their potential role in anti-cancer vaccines as well as natural liposomes for targeted delivery with various options for novel cancer therapies [1]. Mitochondrial DNA components were detected in exosomes, resulting from the culture supernatant of myoblasts and chromosomal DNA (vide infra). Chromosomal DNA was identified in cell culture supernatant in both human and mouse biological fluids, such as blood, seminal fluid, and urine. DNA-loaded exosomes could enhance DNA stability after it leaves the cell [18]. Such findings promote the use of exosomes as novel biomarkers in liquid biopsies, assisting the diagnosis and monitoring of cancer patients [19]. Blood plasma exosomes containing circulating DNA are complex agents in cancer therapies, isolating cancer-specific DNA for circulating cancer cell-derived exosomes [20]. Fibroblast-derived exosomes were shown to stimulate the protrusion of breast cancer cells (BCC) as well as their motility and metastasis dependent on tetraspanins, namely Cd81, which are common EV-associated markers. A study on a mouse model showed that tumor exosomes influence cancer metastasis based on the core PCP pathway in breast cancer cells, indicating that PCP components are almost mutually distributed in the protrusions of single, motile and malignant cells. Exosome activity is associated with the Wnt11 produced in breast cancer cells, and exosomes secreted from fibroblast are internalized by BCCs and further loaded with Wnt11. Therefore, exosomes secreted from fibroblasts play an important role in mediating the mobilization of autocrine Wnt-PCP signaling in BCCs, stimulating invasive behavior and metastasis in murine models [21]. In a recent study, cancer-associated fibroblasts demonstrated enhanced exosome production following gemcitabine injection, which also influenced exosome content by an increase in the presence of SNAIL1 and miR-146a. After treating pancreatic cancer cells with gemcitabine-derived CAF exosomes, cancer cells showed resistance to therapy and increased proliferation. Such results emphasize the ability of stromal cell-derived exosomes to enhance pro-cancer properties, including migration and resistance to therapy [22]. Exosomes are often employed as a novel reservoir for disease biomarker discovery, especially in cancers. There have been reports showing the usefulness of exosomal miRNA-103, tripartite motif-containing 3 protein, glypican-1 proteoglycan protein and hepatocyte growth factor-regulated tyrosine kinase substrate protein in colon cancers. As a result, exosomes proved their potential as tumor markers for various types of cancers, including colorectal cancer [23,24,25]. As cancer cells secrete more exosomes than normal cells, there is a significant difference between molecules found in tumor cell-derived exosomes and those in normal cells. It has been demonstrated that there is a significant difference in certain miRNAs, lncRNAs and proteins in blood-derived exosomes between patients with colorectal patients and healthy subjects. Such exosomal molecules could be used as predictors for colorectal cancer [24]. The serum expression level of exosomal miRNA (let-7a, miR-1229, miR-1246, miR-150, miR-21, miR-223, and miR-23a) was significantly higher in primary colorectal cancer patients, including those with early-stage disease than in healthy subjects, being substantially downregulated following tumor excision. Those seven miRNAs also showed significantly higher secretion by colon cancer cell lines when compared to the healthy colon-derived cell line. Their high sensitivity was validated by receiver operating characteristic (ROC) analysis [26]. It has been stated that 90% of cancer deaths are caused by metastasis. Commonly, colorectal cancer spreads to distant organs (liver, lung, lymph nodes). In the case of patients with distant metastasis, the five-year survival rate is a grim 10%. It is thus very important to detect metastasis early in order to increase the survival of these patients [24]. MiR-203 demonstrated the existence of a link between tumor and host cells, with exosomal miR-203 presented as a novel biomarker to predict metastasis mainly as a promotor of monocyte differentiation to M2-TAMs and the subsequent formation of pre-metastatic niches. There have been significant clinical findings showing the dual functions of miR-203 in the progression of colorectal cancer [27]. Enhanced IRF-2 serum levels in CRC patients with lymph node metastasis present themselves as a novel biomarker for metastasis. Exosomal IRF-2 is able to activate lymph node metastases by remodeling the lymphatic network [28]. Shao et al. demonstrated that serum extracellular vesicles containing miR-21 in colon cancer cells are new macrophage regulators leading to the creation of an inflammatory pre-metastatic niche in colon cancer liver metastasis. While cancer develops, primary CRC cells secrete serum extracellular vesicles containing miR-21 that are transported by the blood flow to the liver where they are engulfed by macrophages. The serum extracellular vesicles deliver the miR-21 load, and by targeting the TLR7 pathway, they polarize macrophages enhancing the synthesis and release of pro-inflammatory cytokines such as IL-6, thus paving the way for a permissive inflammatory pre-metastatic niche in the liver where circulating CRC cells can survive, colonize and subsequently develop macrometastasis [29]. Recent research has shown that miR-375 controls the expression of MMP2 and other genes involved in the epithelial–mesenchymal transition (EMT), such as SNAIL. Colorectal cancer cells proliferate, invade, and migrate when miR-375 is suppressed. Loss of function of the tumor suppressor miR-374 in colorectal cancer (CRC) promotes proliferation, invasion, migration, and intrahepatic metastasis through activation of the PIK3/AKT pathway. To a large extent, miR-374 inhibition upregulates the expression of its targets, which include the transcription factors SNAIL, SLUG, and ZEB1 as well as NCAD and VIM [30]. The regulation of ZEB transcription factors in CRC cells is primarily mediated by two members of the miR-200 family: miR-200c and miR-429. MiR-200c inhibition of ZEB1 expression leads to EMT inactivation and decreased CRC cell invasion and migration. Because of its ability to target ONECUT2, MiR-429 could suppress cell migration and invasion, reversing TGFb’s EMT-inducing effects. MiR-429 is, however, substantially downregulated in colorectal cancer [31]. Because of its ability to target ONECUT2, MiR-429 could suppress cell migration and invasion, reversing TGFb’s EMT-inducing effects. In contrast, miR-429 is considerably downregulated in colorectal cancer [32]. In addition, the loss of ASCL2 function, a target of WNT signaling, can activate the miR-200 cluster, which in turn inhibits the ZEB and SNAIL families of transcription factors and controls the plasticity from EMT to mesenchymal–epithelial transition (MET) [33]. It has been found that the upregulation of the ZEB2 target gene is associated with CRC invasion and metastasis when other tumor suppressors, particularly miR-335, miR-132, and miR-192, are downregulated [34,35,36]. Takano et al. stated that CRC cell-derived exosomal miR-203 promotes the differentiation of monocytes into M2-tumor-associated macrophages (TAMs) involved in colorectal cancer metastasis to the liver [27]. Table 1 summarizes the roles of exosomes in colorectal metastatic disease. One can distinguish the important clinical aspects in which exosomes are involved as well as opposite effects (anticancer/cancer promoter) reported for different exosomes. Efforts have been made to employ miRNAs in serum or plasma as diagnostic biomarkers for more cancers. There are still decisions to be made regarding the type of miRNAs to be selected as markers. The particular properties of exosomes, such as their ability to embed specific miRNAs, their stability in the blood flow, their reproducible detection, and especially their ability to reflect the properties of cancer cells, promote them as important tools in the design of highly sensitive diagnostic strategies for the rapid and non-invasive monitoring of cancer evolution [26]. Exosomal miRNAs could be a biomarker of colorectal cancer. A recent RNA sequencing study on exosomes in colorectal cancer patients indicated high miRNA-139-3p, let-7b-3p and miRNA-145-3p expression in plasma exosomes [37]. Elevated exosomal miRNA-19a levels in the serum of colorectal patients were indicative of cancer recurrence [38]. Moreover, exosomal miRNA-17-92a expression in the blood was associated with cancer recurrence. Certain exosomal miRNAs such as miRNA-1229, miRNA-1246, miRNA-21, miRNA-23a, let-7a, miRNA-223 and miRNA-150 demonstrated great transfer by serum exosomes in colorectal cancer patients, but they were significantly lower following surgical excision [26]. MiRNA-1246, miRNA-21 and miRNA-23a stand out as powerful diagnostic biomarkers of colorectal cancer [39]. Figure 2 illustrates one method to be implemented in the future to analyze cargoes of exosomes in order to highlight different types of miRNA embedded as biomarkers for colorectal cancer. Table 2 shows the types of exosomal miRNAs that are potential cancer diagnostic biomarkers in colorectal cancer. The studies discussed the use of lncRNA-loaded CRC-derived exosomes as diagnostic biomarkers. In another study, Zou et al. observed significantly lower serum exosomal miR-150-5p levels in colorectal cancer patients, therefore being appropriate diagnosis indicators. Diagnostic accuracy was boosted by the combined use of miR-150-5p and the carcinoembryonic antigen. Altogether, these findings emphasize the potential of exosomal miRNAs in diagnosing colorectal cancer [41]. LncRNA is a non-coding RNA that has a size of more than 200 nt in length, and it was found in the blood exosomes of patients diagnosed with colorectal cancer. The results in one study showed an overexpression of lncRNA differentially expressed. This could lead in using lncRNA as a tumor marker due to its non-invasive character, high sensitivity and specificity, as well as stability. It is also highly correlated with aggressive tumor behavior and poor prognosis. Such results provide the grounds for the design of an early diagnostic and prognostic biomarker for colorectal cancer and the corresponding novel therapeutic strategies [42]. In their study, Hu et al. study demonstrated that exosomal lncRNAs, namely LNCV6_98602, LNCV6_98390, LNCV_108266, LNCV6_116109, LNCV6_38772, and LNCV6_84003 plasma expression, was significantly higher in patients with colorectal cancer, promoting them as potential diagnostic biomarkers for this type of cancer [43]. Barbagallo et al. showed in two types of CRC cell lines (HCT-116, Caco-2) that urothelial cancer associated 1 (UCA1), also a lncRNA, can act as a RNA regulator for colorectal cancer progression by modulating the ceRNA network, thus upregulating ANLN, BIRC5, IPO7, KIF2A and KIF23 in two ways: (1) miRNAs sponge effects determining negative expression, and (2) the direct binding of mRNAs to 3′-UTRs to protect them from degradation. Such elaborate RNA-based regulatory signaling for cancer control suggests the design of novel anticancer therapies targeting UCA1 [44]. Granulocytic myeloid-derived suppressor cells were shown to enhance the capability of colorectal cancer cells for self-renewal and differentiation as a result of exosomes and exosomal S100A9 influence in the tumor microenvironment, mainly under hypoxic conditions. Hyperoxia reduces the stemness of colon cancer cells via the inhibition of the production of GM-Exo. Elevated plasma concentration of exosomal S100A9 was linked to the occurrence and recurrence of colorectal cancer. The production of block MDSC exosomes could be used as a new approach for colorectal therapy [45]. The results demonstrate the potential use of exosomal proteins as biomarkers of colorectal cancer. The carcinoembryonic antigen (CEA) was also observed in the serum exosomes of colorectal patients [46]. The value of the area under the curve (AUC) of serum exosomal CEA (0.9354) was greater than that of serum CEA (0.8557). It is thus more significant to detect serum exosomal CEA in order to predict distant metastasis in colorectal cancer. The overexpression of interferon regulatory factor 2 (IRF-2) was observed in the serum exosomes of colorectal cancer patients with lymph node metastasis [28]. From a mechanistic view, exosomal IRF-2 triggers lymph node metastasis by remodeling the lymphatic network. Certain miRNAs were differentially expressed in the plasma exosomes of patients with locally advanced rectal cancer, therefore promoting themselves as potential biomarkers for the poor prognosis of colorectal cancer [47]. Among them, there was a correlation between low miR-181a-5p levels and high miR-30d-5p levels in plasma exosomes and lymph node metastases and liver metastases. There is still no clear definition of the roles these RNAs play in colorectal cancer [24]. In their study, Jun et al. were able to identify several candidate targets with a miRNA–mRNA network (mRNA: CBFB, CDH3, ETV4, FOXQ1, FUT1, GCNT2, GRIN2D, KIAA1549, KRT80, LZTS1, SLC39A10, SPTBN2, ZSWIM4; and exosomal miRNA: hsa-miR-126, hsa-miR-139, hsa-miR-141, hsa-miR-29c, and hsa-miR-423), which could be used as potential biomarkers in the diagnosis of colorectal cancer with the presence of an exosomal miRNA–mRNA network in cancer progression. Their results pave the way for new diagnostic and treatment strategies of colorectal cancer [48]. There have been great attempts to enhance the innate properties of exosomes and to enhance the manufacturing process of exosomes or exosome mimetics. Exosome-based drug delivery tools were divided into three subgroups based on the extent of human manipulation and their natural feel compared with cell-derived exosomes. Frequent protein components that exosomes contain include cytoskeletal (such as actin), cytosolic (for example GAPDH), heat shock (HSP90), antigen presentation (MHC-I, -II), and membrane proteins (CD9, CD63) together with proteins involved in vesicle trafficking (Tsg101) [49]. In the tumor microenvironment (TME), fibroblasts are a major component. MicroRNAs regulate multiple signaling pathways, causing fibroblasts at the primary tumor site to take on a new phenotype and transform into CAFs. Cancer-associated fibroblasts (CAFs) are distinct from normal fibroblasts (NFs) due to their pro-tumorigenic properties and high expression of smooth muscle actin (28). To promote tumor growth, CAFs secrete a variety of pro-inflammatory molecules, such as interleukins, chemokines, and extracellular matrix (ECM) components. Oxaliplatin (Oxa) is a common chemotherapeutic agent for colorectal cancer treatment. The exosome-mediated crosstalk between CRC-associated fibroblasts (CAFs) and CRC cells have demonstrated important roles in chemoresistance to Oxa. It was also confirmed that oncogene miR-21, one of the most oncogenic miRNAs, was enriched in the exosomes from CAFs [50]. After overexpression in exosomes, miR-21 is transported to colorectal cancer cells and enhances AKT phosphorylation strongly related to chemoresistance to Oxa. In another study, lncRNA H19 was expressed to a great extent in the CAFs of colorectal cancer patients, which also increased with cancer progression. LncRNA H19, as a oncofetal transcript, has been shown to promote SIRT1-mediated autophagy in colorectal cancer (CRC) cells, which in turn confers resistance to 5-fluorouracil [51]. One of the most common causes of therapeutic failure is resistance to therapy. The various mechanisms of exomes were shown to determine drug resistance in several recent studies. Exosomes can guide miRNAs, lncRNAs and proteins to the target cells and trigger signal transmission between drug-resistant cells and sensitive cells, stromal cells and tumor cells, which can lead to the drug resistance of tumor cells [52,53]. Other examples of resistance were observed in the microenvironment of ovarian cancer, where exosomes derived from tumor-associated adipocytes and tumor-associated fibroblasts are able to transport miR-21 to ovarian cells, downregulating APAF1 expression and inhibiting tumor apoptosis, thus leading to resistance to paclitaxel [54]. In colorectal cancer, CAF-derived exosomes loaded with miR-92a-3p are aimed at FBXW7 and MOAP1 in the tumor microenvironment and further activate the WNT/β-catenin pathway, inhibit mitochondrial apoptosis, leading to cell stemness, epithelial–mesenchymal transition, tumor metastasis and resistance to 5-FU/L-OHP [55]. Tumors are able to stray from attacks from the immune system by various mechanisms that allow them to avoid being detected. The immunomodulatory potential makes exosomes useful in novel immunological strategies to improve antitumor immunity. Cancer immunotherapy using chimeric antigen receptor (CAR) is a promising therapeutic approach. The clinical use of CAR-modified T cell (CAR-T) therapy in solid tumors was not as successful as in hematological malignancies, such as acute lymphoid leukemia, mainly due to side effects such as cytokine release syndrome (CRS), cytokine storm and on-target/off-tumor responses [56]. Because of their notable accuracy across a wide range of biological datasets, microRNAs have emerged as promising leads in the search for additional CRC cancer biomarkers. The value of serum miRNAs throughout CRC diagnosis, prognosis, and treatment response has been the subject of a plethora of recent papers. Compared to traditional markers such as carcinoembryonic antigen (CEA) and CA19-9, a panel of six miRNAs (miR-21, let-7g, miR-31, miR-92a, miR-181b, and miR-203) has been shown to be a potential marker for CRC diagnosis with over 40% specificity and sensitivity [57]. The absence of trustworthy methods for cancer detection has led to a drawback in the development of colorectal cancer. There is a need for highly efficient detection techniques in order to lower the risk of cancer-associated mortality. More and more findings have demonstrated the strong correlation between the initiation and progression of colorectal cancer and the differentially expressed exosomal RNAs and proteins. These molecules are able to influence the oncogenesis, metastasis, chemoresistance and recurrence of colorectal cancer, thus being potential candidates for this type of cancer. There are several advantages offered by exosomes as novel tumor markers: (i) they could be superior to conventional techniques in terms of sensitivity and specificity; (ii) their bioactive molecular content, without much serum involvement; (iii) they are characterized by high stability and their contents do not degrade in the extracellular environment; and (iv) they are secreted by a variety of body liquids, and thus, they can be extracted in a non-invasive manner. The Food and Drug Administration has already approved the use of certain exosome-based diagnostic kits in clinical trials [24]. Nevertheless, there are certain drawbacks to the use of exosomes as tumor markers. For example, it is essential to rapidly and meticulously isolate exosomes from a sample prior to using them as biomarkers. The present isolation techniques have their own limitations, being bulky, lengthy, including contaminations, and they are expensive. The purity of exosomes is of great importance and the presence of impurities, such as proteins and RNAs in exosomal compounds has been reported, which may have a negative impact on the accuracy of exosome-based diagnosis. It is thus crucial to design highly accurate separation methods to enable the transition of exosome detection to clinical applications. Another aspect is the term “exosomes” itself, which is not recommended nomenclature anymore due to the wide vesicle heterogeneity depending on the purification method, which has dictated the quality and accuracy of the final product. The different results have led to standardization issues, and thus, studies cannot be compared. It is necessary to eliminate all deviations in order to successfully employ exosomes as biomarkers [24]. There are multiple mechanisms by which exosomes act as mediators of intercellular communication. Those derived from colorectal cancer cells are essential mediators in this type of cancer influencing tumor formation by enhancing growth, invasion, metastasis, angiogenesis and immunomodulation. Regardless the stage of the condition, exosomes can transport certain biomolecules into the blood, therefore promoting themselves as promising biomarkers for cancer stage. As exosomes are released into various biofluids, they could be used as a novel diagnostic biomarker in colorectal cancer. There are still a few aspects that need to be thoroughly explained, such as the processes of separation, characterization and validation [40]. Several studies have investigated the potential of exosomes as diagnostic and prognostic biomarkers in cancer in general and colorectal cancer in specific and as targets for novel therapeutic interventions. However, further research is needed to fully understand the complex roles of exosomes in colorectal cancer and to translate this knowledge into clinical practice. Overall, exosomes represent a promising area of investigation for improving the diagnosis, treatment, and management of colorectal cancer. It has become increasingly evident that there are several key aspects regarding the underlying mechanisms of exosome-mediated crosstalk in the tumor microenvironment, distant cell interactions, exosome heterogeneity, and molecular mechanisms that are responsible for resistance and metastasis. Our understanding of exosome-mediated therapy-resistance in different cancers will be directed by the tumor context, which will be directed by the design of different research approaches in this new vast area of study based on the tumor context. The translation of these findings into the clinical realm will provide a novel and effective treatment modality for future cancer patients. Due to their quantity and heterogeneity, exosome biomarkers can produce false positives and negatives in diagnosis and prognosis. Clinical exosome biomarker sensitivity and specificity must be improved. Blocking the formation of pre-metastatic niches and inhibiting miRNAs intracellular signaling to prevent metastasis may be used as a novel cancer therapy strategy.

PMC10002409

Dilara Z. Gatina,Ilnaz M. Gazizov,Margarita N. Zhuravleva,Svetlana S. Arkhipova,Maria A. Golubenko,Marina O. Gomzikova,Ekaterina E. Garanina,Rustem R. Islamov,Albert A. Rizvanov,Ilnur I. Salafutdinov

Induction of Angiogenesis by Genetically Modified Human Umbilical Cord Blood Mononuclear Cells

23-02-2023

human umbilical cord blood mononuclear cells,angiogenesis,recombinant adenoviruses,gene modification,transgene expression,VEGF,FGF2,SDF1α,NUDE mice,Matrigel plugs,cytokine profile

Stimulating the process of angiogenesis in treating ischemia-related diseases is an urgent task for modern medicine, which can be achieved through the use of different cell types. Umbilical cord blood (UCB) continues to be one of the attractive cell sources for transplantation. The goal of this study was to investigate the role and therapeutic potential of gene-engineered umbilical cord blood mononuclear cells (UCB-MC) as a forward-looking strategy for the activation of angiogenesis. Adenovirus constructs Ad-VEGF, Ad-FGF2, Ad-SDF1α, and Ad-EGFP were synthesized and used for cell modification. UCB-MCs were isolated from UCB and transduced with adenoviral vectors. As part of our in vitro experiments, we evaluated the efficiency of transfection, the expression of recombinant genes, and the secretome profile. Later, we applied an in vivo Matrigel plug assay to assess engineered UCB-MC’s angiogenic potential. We conclude that hUCB-MCs can be efficiently modified simultaneously with several adenoviral vectors. Modified UCB-MCs overexpress recombinant genes and proteins. Genetic modification of cells with recombinant adenoviruses does not affect the profile of secreted pro- and anti-inflammatory cytokines, chemokines, and growth factors, except for an increase in the synthesis of recombinant proteins. hUCB-MCs genetically modified with therapeutic genes induced the formation of new vessels. An increase in the expression of endothelial cells marker (CD31) was revealed, which correlated with the data of visual examination and histological analysis. The present study demonstrates that gene-engineered UCB-MC can be used to stimulate angiogenesis and possibly treat cardiovascular disease and diabetic cardiomyopathy.

Induction of Angiogenesis by Genetically Modified Human Umbilical Cord Blood Mononuclear Cells Stimulating the process of angiogenesis in treating ischemia-related diseases is an urgent task for modern medicine, which can be achieved through the use of different cell types. Umbilical cord blood (UCB) continues to be one of the attractive cell sources for transplantation. The goal of this study was to investigate the role and therapeutic potential of gene-engineered umbilical cord blood mononuclear cells (UCB-MC) as a forward-looking strategy for the activation of angiogenesis. Adenovirus constructs Ad-VEGF, Ad-FGF2, Ad-SDF1α, and Ad-EGFP were synthesized and used for cell modification. UCB-MCs were isolated from UCB and transduced with adenoviral vectors. As part of our in vitro experiments, we evaluated the efficiency of transfection, the expression of recombinant genes, and the secretome profile. Later, we applied an in vivo Matrigel plug assay to assess engineered UCB-MC’s angiogenic potential. We conclude that hUCB-MCs can be efficiently modified simultaneously with several adenoviral vectors. Modified UCB-MCs overexpress recombinant genes and proteins. Genetic modification of cells with recombinant adenoviruses does not affect the profile of secreted pro- and anti-inflammatory cytokines, chemokines, and growth factors, except for an increase in the synthesis of recombinant proteins. hUCB-MCs genetically modified with therapeutic genes induced the formation of new vessels. An increase in the expression of endothelial cells marker (CD31) was revealed, which correlated with the data of visual examination and histological analysis. The present study demonstrates that gene-engineered UCB-MC can be used to stimulate angiogenesis and possibly treat cardiovascular disease and diabetic cardiomyopathy. Angiogenesis is the growth of new blood vessels from pre-existing vessels, an essential process for development, wound healing, and the restoration of blood flow and oxygen supply to tissues after injury. One of the main tasks of modern medicine is the stimulation of the processes of angiogenesis in the treatment of vascular diseases. To date, many approaches have been proposed for the induction of therapeutic angiogenesis. Among the proposed methods are surgical methods [1], the use of inducer proteins [2], recombinant DNA molecules [3], inducer genes [4], and the use of various cell types [5,6], including ex vivo genetically modified cells [7]. In this aspect, human umbilical cord blood mononuclear cells (UCB-MC) seem to be a promising “tool” for stimulating angiogenesis through the delivery of genetic engineering systems, expression of recombinant proteins, and possibly direct participation in new vessel formation. The choice of UCB-MC in studies for cell and gene-cell therapy looks promising because of some advantages of this cellular source. Umbilical cord blood contains many stem/progenitor cells and can be obtained easily [8]. The mononuclear fraction of UCB contains populations of different immature cells capable of differentiating into many cell types [9]. Cell populations that have been discovered in UCB are hematopoietic stem cells (HSCs), endothelial progenitor cells, mesenchymal stem cells (MSCs), unrestricted somatic stem cells (USSCs), and side population cells [10,11,12,13]. As cellular material for transplantation or carriers for genetic constructs, UCB-MCs have low immunogenicity because they do not express all the antigens on the cell membrane. This feature enhances the ability to cross donor-recipient HLA disparities. It allows for the usage of UCB-MC for transplantation in non-fully compatible HLA recipients with a much lower incidence of grade II-IV acute GVHD (graft versus host disease) cases after transplantation [14,15,16,17]. Furthermore, UCB-MCs can prevent the oncological transformation of recipient cells after transplantation [15]. Another appealing reason for using UCB cells for cell therapy is their ability to produce various biologically active molecules, such as proteins with antioxidant properties, angiogenic, neurotrophic, and growth factors [18,19,20,21,22], which make them suitable for effective stimulation of regenerative processes in non-compatible recipients for a short time before the immune system eliminates them. Overall, UCB cell transplantation can replace dead cells, prevent further death of surviving cells, and stimulate regeneration by secreting biologically active molecules. A genetic modification of UCB cells can enhance their ability to regenerate tissue [23,24]. This approach unites the advantages of cell- and gene therapy. Genetically modified UCB cells can provide targeted delivery of therapeutic genes and expression of recombinant molecules at the regeneration site. For example, our previous studies showed the positive effect of genetically modified umbilical cord blood mononuclear cells (UCB-MC) simultaneously produces three recombinant molecules (vascular endothelial growth factor (VEGF), glial cell-derived neurotrophic factor (GDNF), and neural cell adhesion molecule (NCAM) in animal models of amyotrophic lateral sclerosis [25], spinal cord injury [26] and stroke [7,27]. Many state-of-the-art methods and models for studying angiogenesis have been proposed, which are well analyzed in the review articles [28,29,30]. Among various models, the in vivo angiogenesis plug assay, which uses basement membrane extracts (BME) or Matrigel, is widely used for evaluating pro- or anti-angiogenic factors during in vivo angiogenesis. This assay is reliable, easy to perform without special equipment, reproducible, quantitative, and quick [31,32,33]. Matrigel predominantly contains laminin III, collagen IV, heparan sulfate proteoglycans, and various growth factors. The assay is performed by injecting the liquid Matrigel into the subcutaneous space of an animal at 4 °C, which solidifies to form a plug at body temperature. Over time, blood vessels sprout into the plug. The number of plug sites per animal can be several, allowing multiple test compounds or concentrations to be tested. Thus, drug screening can also be evaluated for effects on the activity of angiogenic or anti-angiogenic factors [34,35,36,37]. The drug can be placed in the plug with the test factor by mixing with the Matrigel matrix or given to the host animal. Cells or exosomes can also be examined when mixed into the gel to produce angiogenic factors. Furthermore, the assay is highly versatile. For example, the role of certain genes can be evaluated using genetically modified mice (overexpressing or ablating a protein gene) or animal models of diseases. This report aimed to study the effect of genetically modified umbilical cord blood mononuclear cells overexpressing recombinant proangiogenic proteins VEGF165, FGF2, and SDF1α on the induction of angiogenesis. Furthermore, we assessed the influence of all three factors on the tone of the secretory profile of modified UCB-MCs and tubule formation in the in vivo Matrigel plug assay. The present study shows that when combined with UCB, the three factors can enhance angiogenesis and be useful for developing new therapies. Isolated cells demonstrated high viability (>97%) and included CD45+ lymphocytes (58.9%). CD45+CD3+ lymphocytes constituted 59.2%, while CD14+ macrophages constituted 7.3%. This ratio of the central populations of blood cells (lymphocytes, T-lymphocytes, and monocytes) is believed to be typical for human UCB-MCs. We also examined the percentage of CD34+ blood cells among isolated UCB-MCs. According to the obtained data, CD34+ cells constituted 0.4% of CD45+ cells. In addition, 91.8% of CD45+CD34+ cells expressed CD38. Furthermore, 90% of the CD45+CD34+ cells had the phenotype CD90+. The flow cytometry results are shown in Figure 1. Immunophenotyping of a pool of CD34-positive cells showed that genetic modification and expression of recombinant factors by cells did not affect the viability and endothelial cell markers (Figure 2). It has been demonstrated that genetic modification of the UCB-MCs with recombinant adenoviruses (Ad-VEGF, Ad-FGF2, Ad-SDF1α, or Ad-EGFP) did not affect cell viability. Moreover, it has been shown that UCB-MCs transduced with Ad-EGFP exhibited green fluorescence, confirming the efficiency of transduction (Figure 3A). Furthermore, EGFP expression was sustained for 30 days after a genetic modification of UCB-MCs. According to the flow cytometry results, EGFP+ cells constituted 28 ± 2.7% (Figure 3B). Analysis of the mRNA expression of VEGF165, FGF2, and SDF1α in genetically modified human UCB-MCs was carried out using qPCR. It has been established that genetic modification of hUCB-MCs with Ad-VEGF165 results in augmented VEGF expression (190.6 ± 8.9 fold). Simultaneous transduction with Ad-VEGF165, Ad-FGF2, and Ad-SDF1α resulted in the upregulation of VEGF, FGF2, and SDF1α expression (198.6 ± 0.45; 204.2 ± 0.36 and 140.9 ± 0.32 fold respectively) compared to non-transfected cells, and cells modified with Ad-EGFP (Figure 3C). The obtained results are evident for efficient modification of hUCB-MCs with developed genetic constructs which provide a synthesis of target genes in vitro. A complete analysis of all cytokines and chemokines measured in the Luminex assays demonstrated that gene modification and gene expression did not change levels of multiple anti and proinflammatory cytokines as well as chemokines. The results obtained from the eight donors in comparison to the untreated control are shown in Table 1 (Supplementary Table S1). We have not observed any statistically significant differences in cytokine and chemokine secretion between the groups of non-transfected cells and genetically modified ones except for upregulated levels of recombinant proteins in corresponding groups. Multiplex analysis revealed statistically significant (p < 0.05) upregulation of VEGF secretion (1087.12 ± 169.11 pg/mL) in UCB-MCs modified with Ad-VEGF compared to the UCB-MCs treated with Ad-EGFP (52.31 ± 10.36 pg/mL) and non-treated cells (51.75 ± 8.65 pg/mL). Simultaneous transduction with Ad-VEGF, Ad-FGF2, and Ad-SDF1 has resulted in the increased production of VEGF (701.94 ± 96.99 pg/mL), FGF2 (576.27 ± 57.83 pg/mL), and SDF1α (622.39 ± 113.07 pg/mL) (Figure 3D). Obtained results correlate with the data presented above of RT-qPCR and confirm the capacity of recombinant adenoviruses for infection of target cells. It is also worth emphasizing that the UCB-MC-VEGF-FGF2-SDF1 and UCB-MC-VEGF did not differ from UCB-MC-EGFP and UCB-MC-NTC in vitro studies. What can be seen from the data of morphological, and phenotypic studies are the profiles of secreted factors. Therefore, UCB-MC-EGFP is the ideal control in our study in vivo. Matrigel mixtures were implanted into the subcutaneous space of the dorsal region in mice after seven days post-transplantation when implanted Matrigel samples containing genetically modified UCB-MCs were extracted from Balb/c nude mice. Embedded fragments represented discs with d = 10 mm and 2 mm height. The color of the implants correlated with vascularization density. The color of the implants varied from milky-white (Matrigel without cells and Matrigel with UCB-MC + Ad-EGFP) to red-brown (Matrigel with UCB-MC + Ad-VEGF165 + Ad-FGF2 + Ad-SDF1α) which is due to the vascular formation and presence of blood cells, particularly, erythrocytes (Figure 4A). Gross histological hematoxylin and eosin (H&E) staining of extracted plugs showed the absence of inflammatory sites. The skin and subcutaneous tissue in the area of implantation were not visually changed (Figure 4B). We have established that in isolated subcutaneous implants containing hUCB-MC, human-transduced Ad-VEGF165, or a combination of Ad-VEGF165, Ad-SDF1α, and Ad-FGF2, the hemoglobin concentration was significantly higher in comparison with Matrigel fragments without cell administration and implants with UCB-MCs transduced Ad-EGFP. Moreover, the significantly higher concentration of hemoglobin was determined in the samples containing UCB-MCs transduced with Ad-VEGF165, Ad-SDF1α, and Ad-FGF2 compared to the group with UCB-MCs transduced with single Ad-VEGF165 (Figure 4C). Moreover, we observed a two-fold increase of mCD31 mRNA expression in plugs containing hUCB-MC transduced Ad-VEGF165 or a combination of Ad-VEGF165, Ad-SDF1α, and Ad-FGF2 compared to controls. Moreover, we did not discover the difference between Ad-VEGF165 and the group with a mixture of Ad-VEGF165, Ad-SDF1α, and Ad-FGF2. Analysis of the mRNA expression of VEGF165, FGF2, and SDF1α in genetically modified UCB-MCs in Matrigel implants was evaluated by RT-qPCR. Notably, obtained results confirmed the expression of target genes in genetically modified UCB-MCs implanted in Matrigel even at one-week post-transplantation. We discovered that the Matrigel complexes containing UCB-MC Ad-VEGF gave rise to more abundant VEGF mRNA than UCB-MC Ad-EGFP and PBS (Matrigel samples without UCB-MCs). Likewise, UCB-MCs contemporaneously transduced with Ad-VEGF, Ad-FGF2, and Ad-SDF1α exhibited upregulated levels of mRNA expression of VEGF, FGF2, and SDF1α. (Figure 5A). During histological analysis of implants, it has been shown that control—PBS (Matrigel samples without UCB-MCs) contained small amounts of migrated fibroblast-like cells. Visually, the implants were surrounded by a thin connective tissue capsule, which contained rare capillaries in a density of 1.5 ± 0.5 units/mm2. In samples with implanted UCB-MCs transduced with a cocktail of adenoviruses (Ad-VEGF165, Ad-FGF2, and AdSDF1α), Matrigel mass contained a residual amount of VEGF+ cells. These vessels localized close to the capsule and migrated fibroblasts, some of which were positive for SDF1α and FGF2. In Matrigel samples with implanted UCB-MCs genetically modified with Ad-EGFP, we found single and small rounded clusters of EGFP-positive cells and rare migrated fibroblast-like cells. The implants were surrounded by a thin connective tissue capsule, from which strands of connective tissue grew into its depth with capillaries found in a density of 7.5 ± 3 units/mm2. Vessels were located close to the capsule. Fibroblasts that migrated into Matrigel expressed SDF1α and FGF2. Expression of VEGF, FGF2, and SDF1α in the implanted UCB-MCs were not confirmed. In the group with UCB-MCs modified with Ad-VEGF165, implant samples presented Matrigel mass with single small, rounded clusters of VEGF-positive UCB-MCs cells and rare migrated fibroblast-like cells. The implants were surrounded by a thin connective tissue capsule, from which strands of connective tissue grew more profound into the central regions of the implant with capillaries‘ density of 16 ± 5 units/mm2. In the group of UCB-MCs simultaneously transduced with a combination of Ad-VEGF165, Ad-SDF1α and Ad-FGF2, implant samples were represented by the mass of Matrigel with single and small rounded clusters of UCB-MCs, as well as rare migrated fibroblast-like cells. The implant was surrounded by a thin connective tissue capsule, from which the connective tissue and vessels of various calibers grew to the center of the implant with a capillary density of 23 ± 5 units/mm2. Implanted UCB-MCs expressed VEGF, FGF2, and SDF1α (Figure 5B). Adenoviruses mediate gene transfer into dividing and quiescent cells and can be produced with a significant titer. The high immunogenicity of adenoviruses as vehicles for the delivery of therapeutic genes represents one of the main disadvantages resulting in the activation of the immune response in immune-competent organisms and the absence of expression of the target therapeutic genes [38]. However, this negative effect is eliminated when using an ex vivo gene therapy approach. Moreover, adenoviral systems promote transient transgene expression due to their non-integration into the host cell genome [39]. However, this negative point might become beneficial for gene therapy based on growth factors: induction of angiogenesis does not require the prolonged expression of therapeutic proteins but, more importantly, their synergistic effect [40]. The absence of integration of adenoviruses eliminates the risk of insertional mutagenesis, which is a typical problem when using retroviral vectors [39]. Adenoviral vectors demonstrate comparatively low efficiency of genetic modification of hematopoietic cells, which might be increased with a higher concentration of virus [41] or its specific treatment, resulting in augmented tropism [42]. In the present study, we chose the adenovirus delivery vectors containing VEGF, FGF2, and SDF1α to investigate the angiogenic effect of UCB-MC in vitro and the Matrigel plug assay in Nude mice. In our investigation, cellular carriers expressed phenotype typical for UCB-MCs, and about 30% of the cells were efficiently transduced with an MOI of 10. The transduction efficiency correlated with previous results and other research groups’ data [42,43]. After in vitro transduction, the UCB-MCs expressed the recombinant mRNA of proangiogenic factors in the cytoplasm and secreted those factors into the culture medium, which in our study we confirmed by RT-qPCR and immunological studies. The obtained data correlates with our previously published results [7]. Various approaches were proposed for stimulating therapeutic angiogenesis based on the delivery methods of genetically engineered systems expressing a broad range of proangiogenic factors. The therapeutic efficacy of proangiogenic factors has been proven in numerous experiments on animal models [44] and in several clinical studies [45,46]. The key inducers of angiogenesis, VEGF, FGF2, EGF, SDF1α, and PDGF-BB, are most often used as genetic components [2]. In particular, VEGF is perhaps the most characterized and frequently used mitogen in creating gene therapy systems and in the induction of therapeutic angiogenesis. VEGF is a crucial participant in forming new blood vessels and can induce the growth of pre-existing ones [47]. However, Zentilin et al. reported that overexpression of VEGF induced leaky neovessels that missed connecting correctly with existing vessels [48]. The FGF family includes vertebrates’ most versatile growth factors that play critical roles in many biological processes, including angiogenesis [49]. FGF, similar to VEGF, is a pleiotropic molecule capable of acting on various cell types, including endodermal, mesenchymal, and neuroectodermal origin cells. It has been shown that FGF2 induces the expression of VEGF and several other factors by endothelial cells through autocrine and paracrine mechanisms [50,51]. SDF1α is a constitutively expressed and inducible chemokine, associated with various physiological and pathological processes, including embryonic development, homeostasis maintenance, and angiogenesis activation [52]. There is evidence that the administration of SDF1α increases blood flow and perfusion via the recruitment of endothelial progenitor cells (EPCs). SDF1α binds exclusively to CXCR4 and has CXCR4 as its only receptor [53]. Compared with the effects of other angiogenic growth factors, SDF1α has unique properties. The generation of hyperpermeable vessels, a significant characteristic of VEGF-stimulated angiogenesis, may not be observed after injection of SDF1α contributes to the stabilization of neovessel formation by recruiting CXCR4 + PDGFR+ cKit+ smooth muscle progenitor cells during recovery from vascular injury [54]. Extensive evidence suggests that SDF1α up-regulates VEGF synthesis in several cell types, whereas VEGF and basic FGF induce SDF1α and its receptor CXCR4 in endothelial cells [55]. However, it should also be noted that in a wide range of studies using various models, the mutual synergistic role of VEGF, FGF2, SDF1α, and countless other factors responsible for the formation of vessels has been shown [56,57,58,59]. It is generally known that an optional cellular source for allogenic transplantation should meet the following criteria: it must be less immunogenic and contain a sufficient amount of immature cells capable of differentiation in various directions; it should have a prolonged period of storage and potency for expansion. Most gene-cell-mediated therapy protocols intend genetic modification of target cells with different vectors, providing stable expression of target proteins. Human UCB-MCs might be easily isolated and characterized; these cells exhibit low immunogenicity and are composed of unique populations of progenitor cells capable of differentiation into endothelial, muscular, and neural cells, etc. Mononuclear cells from umbilical cord blood are a well-characterized group of cells that are extensively used in pre-clinical and clinical trials of therapy for various human diseases and the induction of therapeutic angiogenesis as well [60]. However, relatively small amounts of UCB-MCs for achieving sufficient therapeutic effect remain the main limitation for its extensive introduction in the clinic [61]. To increase its biological activity, it was proposed to mix different cell pools with further genetic modification [62]. Contemporary cell-mediated approaches to gene therapy suggest UCB-MC as a cell carrier for the delivery of various therapeutic genes. This concept assumes either the differentiation of transplanted cells into different cell types or the realization of therapeutic effects due to the secretion of a broad range of bioactive molecules [63]. Furthermore, our previous study has demonstrated that UCB-MCs are capable of transferring therapeutic genes and promoting evident therapeutic effects using different models, such as ALS [64], SCI [25,26], and stroke [27]. Similar results were obtained in investigations dedicated to therapies for hematologic and non-hematologic disorders [65,66,67,68,69]. At the same time, there is no current data about the influence of the simultaneous transduction of several recombinant adenoviruses on the secretome profile and angiogenic properties of modified hUCB-MCs. A sustained balance of proangiogenic factors and their synergetic effect is essential for functional vascular formation. In the present study, we developed the UCB-MC application to simultaneously deliver many genes (VEGF, FGF2, and SDF1α) to stimulate angiogenesis. Our previous studies also showed the approach of preventive gene therapy with many genes to positively affect stroke. Adenoviral vectors carrying genes encoding vascular VEGF, glial cell-derived neurotrophic factor (GDNF), and NCAM or gene-engineered umbilical cord blood mononuclear cells (UCB-MC) overexpressing recombinant proteins were intrathecally injected before distal occlusion of the middle cerebral artery in the rat. Morphometric and immunofluorescence analysis revealed a reduction in infarction volume and a lower number of apoptotic cells. It decreased the expression of Hsp70 in the peri-infarct region in gene-treated animals [7]. The heterogeneous cell population from the mononuclear fractions UCB-MCs secretes different anti-inflammatory, pro-inflammatory cytokines, chemokines end grow factors [70]. Previously, it was shown that the duration of cultivation, cultivation medium, and the additives used in the culture are the main factors influencing the production of cytokines by UCB-MCs. Our study describes the profile of cytokines and chemokines released by UCB-MC following their in vitro gene modification by adenoviruses. Five groups of secreted factors were investigated: pro-inflammatory cytokines (IL-6, IL-1β, and TNF), an anti-inflammatory cytokine (IL-4 and IL-10), TH1-type cytokines (IL-12 and IFN-γ), chemokines (IL-8, MIP-1α, MIP-1β, and MCP-1) and growth factors (VEGF, FGF2, and SDF1α). Interestingly, the range of cytokine, chemokine, and growth factor concentrations detected in the supernatants of UBC-MC varied between donors, indicating major individual heterogeneity, comparable with previously published data [71]. The highest secretion level by modified and unmodified cells was shown for IL-8 and MCP-1. These factors are known to be produced more intensively than any other chemokines in the human body and are seen as the first line of defense in inflammatory responses [72]. In addition, the cells also secreted high concentrations of GROα, IL-6, MIF, MIP-1α, MIP-1β, and SCGF-β. Unfortunately, adenoviruses are potent activators of the innate and adaptive immune systems. The administration of high doses of Ad-based vectors to animals or patients, primarily through the intravascular pathway, leads to severe immunopathology manifested by cytokine storm syndrome, disseminated intravascular coagulation, thrombocytopenia, and hepatotoxicity, which can lead to morbidity and also death [69]. Research by Teigler et al. on peripheral blood mononuclear cells (PBMCs) showed that their stimulation with the Ad vector increases the secretion of IFN-γ, INF2α, IL-15, G-CSF, MIG, and IP-10. Supporting this perspective, it is worth emphasizing that the study’s authors used 103 vp/cell [73]. Previous studies have shown that treatment of myeloid dendritic cells and plasmacytoid dendritic cells with Ad5 does not lead to an increase in IFN production by them, even at the highest exposed rAd (100 vp/cell) [43]. Our previous examination has shown that genetic modification UCB-MC and expression of transgenes VEGF or EGFP did not influence the global transcriptome landscape [74]. In this study, we demonstrate that a gene-cell system with simultaneous delivery of genes based on UCB-MC can generate the expression of several transgenes both in vitro and in vivo. Furthermore, the UCB-MC-VEGF165 and UCB-MC-VEGF-FGF2-SDF1α Matrigel plugs in mice were filled with red blood cells and showed vessel-like structure formation. We did not find significant differences between the UCB-MC-VEGF and UCB-MC-VEGF-FGF2-SDF1α groups in the present study. Although in line with the RT-qPCR data and immunology tests, levels of expression of VEGF, SDF1α, and FGF varied. Perhaps this is because we used a small amount of cellular material and a short exposure period to Matrigel fragments. Furthermore, the UCB-MC-VEGF165 and UCB-MC-VEGF-FGF2-SDF1α Matrigel plugs in mice were filled with red blood cells and showed vessel-like structure formation. We did not find significant differences between the UCB-MC-VEGF165 and UCB-MC-VEGF-FGF2-SDF1α groups in the present study. Although in line with the RT-qPCR data and immunology tests, levels of expression of VEGF, SDF1α, and FGF varied. Perhaps this is because we used a small amount of cellular material and a short exposure period to Matrigel fragments. The creation of expression constructs based on adenovirus was carried out by using molecular cloning methods of Gateway-cloning technology (Invitrogen), as described previously [75]. Briefly, subcloning of SDF1α from the plasmid vector pBud-VEGF-SDF1α into the intermediate vector pDONR221 was performed [76]. The HEK293A cells were infected with a coarse viral runoff to prepare the necessary amounts of Ad-VEGF, Ad-FGF2, Ad-SDF1α, and Ad-EGFP adenoviruses. To purify viral particles from cell debris, supernatants were filtered through 0.22 µm filters and centrifuged in a gradient of cesium chloride. Virus dialysis was performed using a membrane with a pore throughput of 3.5 kDa in two stages. After purification and concentration, the resulting recombinant adenoviruses were titrated by optical density, as well as by plaque formation. The titer of the recombinant adenoviruses we obtained was from 1 to 3.8 × 109 PFU/mL. The viral titer values were guided by the genetic modification of human UCB-MC. All UCB-MC units were collected from healthy donors with a gestation period of 37–40 weeks in maternity public hospitals in Kazan. Blood collections were carried out into single blood bags of 250 mL, with the blood preservative CPDA-1 (GCMS, Republic of Korea). Exclusion criteria were maternal infections or viral diseases. Isolations of mononuclear cells were conducted within 16–18 h after blood collection. Nucleated blood cells were isolated using SepMate ™-50 tubes according to the manufacturer’s protocol (STEMCELL Technologies Inc., Vancouver, BC, Canada). The viability of the isolated cells was determined in a hemocytometer with a 0.4% trypan blue solution. Cell viability, as measured by trypan blue exclusion, was >97%. The immune phenotype of isolated cells was analyzed by staining with monoclonal antibodies CD45—PerCP (BioLegend, San Diego, CA, USA), CD3-FITC (BioLegend, San Diego, CA, USA) CD14-APC/Cy7 (BioLegend, San Diego, CA, USA), CD38-APC/Cy7 (BioLegend, San Diego, CA, USA) CD34-FITC (BioLegend, San Diego, CA, USA), CD90-PE/Cy5 (BioLegend, San Diego, CA, USA). Expression of CD markers were analyzed by flow cytometry using BD FACS Aria III (BD bioscience, San Jose, CA, USA) Genetic modification of hUCB-MCs with recombinant adenoviruses (MOI 10 for each virus) was performed according to a previously developed protocol [77]. The efficiency of genetic modification was assessed after 72 h by means of fluorescent microscopy on AxioObserverZ1 (Carl Zeiss, Oberkochen, Germany) and flow cytometry using BD FACS Aria III (BD Bioscience, San Jose, CA, USA). Analysis of the mRNA expression of VEGF165, FGF2, and SDF1α in genetically modified cells and isolated Matrigel implants was carried out by qPCR with further statistical analysis. Isolation of total RNA was performed by using the TRIzol (Thermo Fisher Scientific, Waltham, MA, USA) reagent according to the manufacturer’s recommendations with further cDNA synthesis. Real-Time PCR was set up on the Real-Time CFX96 Touch (BioRad Laboratories, CA, USA). The nucleotide sequences of the primers and probes used in RT-qPCR are mentioned in Table 2. All reactions for each sample were performed in triplicate with a further calculation of ΔΔCt values and normalization to the housekeeping gene of β-actin rRNA. Standard curves for quantitative analysis were created using serial dilutions of plasmid DNA with corresponding inserts (VEGF, FGF2, and SDF1α). Expression of target genes in non-transduced UCB-MCs was considered 100%. The level of the murine target gene mCD31 was normalized to the mouse housekeeping gene of mGAPDH. The statistical analysis of the obtained results was carried out in MS Excel 2007 with further calculation using U criteria (Mann-Whitney). Supernatant cytokine profiles were analyzed using Bio-Plex Pro Human Cytokine 27-plex Panel and Bio-Plex Human Cytokine 21-plex Panel (Bio-Rad, Hercules, CA, USA) multiplex magnetic bead-based antibody detection kits, following the manufacturer’s instructions. Supernatant aliquots (50 µL) were used for analysis, with a minimum of 50 beads per analyte acquired. Median fluorescence intensities were measured using a Bioplex 200 (Bio-Rad, Hercules, CA, USA) analyzer. Data collected were analyzed with MasterPlex CT control software and MasterPlex QT analysis software (Hitachi Software, San Bruno, CA, USA). Standard curves for each analyte were generated using standards provided by the manufacturer. In vivo experiments were performed using immune-deficient mice of Balb/c nude lineage of both sexes for 7–8 weeks. The animals were bred using the animal facilities in Puschino’s laboratory. After quarantine, animals were held in an SPF vivarium with HEPA filters according to GLP standards. In the area of the withers, mice were subcutaneously injected with 2 million human transduced or native UCB-MCs mixed with 300 µL of Matrigel matrix. Female and male Balb/c nude mice were randomly assigned to a few groups: 1. Matrigel without UCB-MCs; 2. UCB-MCs transduced with Ad-EGFP in Matrigel; 3. UCB-MCs transduced with Ad-VEGF165 in Matrigel; 4. UCB-MCs transduced with a combination of adenoviruses Ad-VEGF165; Ad-SDF1α and Ad-FGF2 in Matrigel. All experiments were performed in quadruplicates. After seven days post-transplantation mice were taken from the experiment. The status of subcutaneous Matrigel implants was evaluated visually, and concentrations of hemoglobin were evaluated. Levels of the expression of therapeutic genes were analyzed by RT-qPCR. Production of therapeutic proteins was assessed via immunohistochemistry. The analysis of hemoglobin concentration in subcutaneous implants was evaluated colorimetrically. Implants were balanced by weight and homogenized in DPBS using the Mini-Bead Beater-16 (BioSpec, Bartlesville, OK, USA) with zirconium beads (d = 2 mm for 100 mg), during 2 cycles for 20 sec each. The obtained homogenates were centrifuged at 15,000× g for 15 min. Supernatants containing hemoglobin were examined on a microplate reader Tecan Infinite Pro 2000 with an OD of 540 nm (Tecan Austria GmbH, Grödig, Austria). For histological analysis, Matrigel implants were fixed in a 10% buffered formalin solution for 48 h. After fixation, implants were dehydrated in increasing concentrations of ethanol and embedded in paraffin (Histomix, Biovitrum, Saint Petersburg, Russia). Paraffin slides with 5 µm thickness were prepared at the rotary microtome HM 355S (Thermo Fisher Scientific, Waltham, MA, USA). For general morphological characterization, slides were deparaffinized and stained with hematoxylin and eosin according to the standard protocol. For immunological studies, serial sections were deparaffinized and incubated in a citric buffer for 30 min to unmask epitopes. Cell membranes were permeabilized in a 0.1% solution of Tween-20 in PBS. Non-specific binding was blocked by incubation in a 10% solution of donkey serum for 30 min. Sections were stained with the antibodies to VEGF (mab293), FGF2 (sc-1390), and SFD1α (sc-28876), diluted 1:100 for 1 h at room temperature. After washing sections were stained for 1 h with secondary antibodies at room temperature followed by washing and DAPI staining (1:50,000 dilution) for 10 min. Primary and secondary antibodies are shown in Table 3. The microvessel density (MVD) was examined by counting the vessels in each implant, reported as the vessel number per square millimeter (vessel/mm2). Visualization of the results was performed on a scanning laser microscope LSM780 (Carl Zeiss, Oberkochen, Germany). Image analysis was performed using Image J software (https://fiji.sc/ (accessed on 18 December 2021)). GraphPad Prism® 7 software was used to show all data reports (GraphPad, Inc., La Jolla, CA, USA). The data are presented as the mean ± standard error (SE). p-values were analyzed using a one-way analysis of variance (ANOVA) followed by Tukey’s test. Statistical significance is denoted by p < 0.05. All tests with animals, morphometric and statistical analyses were performed in a “blinded” manner with respect to the experimental groups. The study suggests that UCB-MCs continue to be an essential source of stem cells for therapy and various gene-cell strategies for tissue regeneration. It is important to emphasize that the transplantation of genetically modified UCB-MCs is safer and more effective than direct gene therapy. Human UCB-MCs can be efficiently simultaneously modified with adenoviral vectors encoding VEGF, FGF2, and SDF1α. Modified UCB-MCs overexpress recombinant genes. Genetic modification of cells with recombinant adenoviruses (MOI 10) does not affect the profile of secreted pro- and anti-inflammatory cytokines, chemokines, or growth factors, except for an increase in the synthesis of recombinant proteins by cells. Modified cells can induce the formation of new vessels. Although many unresolved problems remain before modified UCB-MCs can be applied in the clinic, our results remain promising in terms of the induction of therapeutic angiogenesis in future clinical trials, including the treatment of decompensated forms.

PMC10002423

Yuer Wang,Huahao Fan,Yigang Tong

Unveil the Secret of the Bacteria and Phage Arms Race

22-02-2023

bacteria,phages,restricting modification systems,CRISPR-Cas systems,aborting infection,quorum sensing

Bacteria have developed different mechanisms to defend against phages, such as preventing phages from being adsorbed on the surface of host bacteria; through the superinfection exclusion (Sie) block of phage’s nucleic acid injection; by restricting modification (R-M) systems, CRISPR-Cas, aborting infection (Abi) and other defense systems to interfere with the replication of phage genes in the host; through the quorum sensing (QS) enhancement of phage’s resistant effect. At the same time, phages have also evolved a variety of counter-defense strategies, such as degrading extracellular polymeric substances (EPS) that mask receptors or recognize new receptors, thereby regaining the ability to adsorb host cells; modifying its own genes to prevent the R-M systems from recognizing phage genes or evolving proteins that can inhibit the R-M complex; through the gene mutation itself, building nucleus-like compartments or evolving anti-CRISPR (Acr) proteins to resist CRISPR-Cas systems; and by producing antirepressors or blocking the combination of autoinducers (AIs) and its receptors to suppress the QS. The arms race between bacteria and phages is conducive to the coevolution between bacteria and phages. This review details bacterial anti-phage strategies and anti-defense strategies of phages and will provide basic theoretical support for phage therapy while deeply understanding the interaction mechanism between bacteria and phages.

Unveil the Secret of the Bacteria and Phage Arms Race Bacteria have developed different mechanisms to defend against phages, such as preventing phages from being adsorbed on the surface of host bacteria; through the superinfection exclusion (Sie) block of phage’s nucleic acid injection; by restricting modification (R-M) systems, CRISPR-Cas, aborting infection (Abi) and other defense systems to interfere with the replication of phage genes in the host; through the quorum sensing (QS) enhancement of phage’s resistant effect. At the same time, phages have also evolved a variety of counter-defense strategies, such as degrading extracellular polymeric substances (EPS) that mask receptors or recognize new receptors, thereby regaining the ability to adsorb host cells; modifying its own genes to prevent the R-M systems from recognizing phage genes or evolving proteins that can inhibit the R-M complex; through the gene mutation itself, building nucleus-like compartments or evolving anti-CRISPR (Acr) proteins to resist CRISPR-Cas systems; and by producing antirepressors or blocking the combination of autoinducers (AIs) and its receptors to suppress the QS. The arms race between bacteria and phages is conducive to the coevolution between bacteria and phages. This review details bacterial anti-phage strategies and anti-defense strategies of phages and will provide basic theoretical support for phage therapy while deeply understanding the interaction mechanism between bacteria and phages. Bacteria and bacteriophages (phages for short) have been engaged in a constant and repeated arms race and have coexisted steadily for billions of years. Phages outnumber bacteria by ten to one and are widely recognized as the most diverse of microbes. Phages can be divided into temperate phages and lytic phages according to their intracellular proliferation modes [1]. After infecting the host, temperate phages do not lysis the host cell but remain in the lysogenic state, integrating the phage genome into the host genome, becoming prophage and replicating with the host replication. Lytic phages can cause lysis and death of the host [2]. The process of lytic phages infecting the host includes five parts: adsorption, injection, biosynthesis, assembly and release. Under certain conditions, temperate phages can enter the lysis cycle from the lysogenic state, causing the host cell to the lysis and die [3]. Bacteria have developed various strategies to resist phage infection, such as preventing phages from adsorbed on the surface of bacteria [4]; through the superinfection exclusion (Sie) block of phage’s nucleic acid injection [5]; by restricting modification (R-M) systems [6], CRISPR-Cas [7], aborting infection (Abi) [8] and other defense systems to interfere with the replication of phage genes in the host; and through quorum sensing (QS) enhancement of phage’s resistant effect [9]. At the same time, phages have also evolved a variety of counter-defense strategies, such as degrading extracellular polymeric substances (EPS) that mask receptors [10] or recognizing new receptors [11], thereby regaining the ability to adsorb host cells; modifying its genes to prevent the R-M systems from recognizing phage genes [12] or evolving proteins that can inhibit the R-M complex [13]; through mutate the target sequence [14], build nucleus-like compartments [15] or evolved anti-CRISPR (Acr) proteins [16] to resist CRISPR-Cas systems; and by producing antirepressors [17] or blocking the combination of autoinducers (AIs) and its receptors [18] to suppress the QS (Table 1). The competition between bacteria and phages has never stopped, and the rapid co-evolution between them has a positive impact on improving the evolution rate of phages and bacteria [19]. Understanding these adversarial strategies is of great significance to the research fields of practical applications. Preventing phages from attaching to host cells is the first step in the bacterial defense battle. Strategies to avoid phage adsorption can be divided into two categories: (i) loss or structural change of receptors and (ii) construction of physical barriers to phage infection. The absence of receptors is a key part of bacterial resistance to phage infection [20] (Figure 1a). Through gene mutation or deletion, the receptor cannot be expressed or the structure of the receptor can be changed to protect bacteria from phage infection. For example, phage OWB uses tail tubular proteins A and B to recognize the transmembrane protein encoded by Vibrio parahaemolyticus vp0980. However, due to the lack of phage-recognized receptors, the mutant hindered the adsorption of phage OWB [4]. Listeria monocytogenes serovar 4b is mutated from serotype 4b to the more virulent serotype 4d by mutating the gene encoding the glycosylation of teichoic acid. Serotype 4d is resistant to infection by phages with wall-teichoic acids as recognition receptors due to the absence of galactose from wall-teichoic acids and lipoteichoic acid molecules [21]. The cell wall is usually thought to protect bacteria from environmental threats [22]. However, because the cell wall contains receptors that the phages can recognize, bacteria with cell wall defects can resist phage adsorption in some cases. Cell wall deficiency is one of the mechanisms of bacteria hindering phage adsorption [23]. By losing or modifying the receptors of jumbo phages, such as lipopolysaccharide and type IV pili, the adsorption of phages to Pseudomonas aeruginosa (P. aeruginosa) PA5oc can be inhibited [24]. At the same time, the virulence and pathogenicity of the bacteria are reduced due to the reduction in the virulence factors such as lipopolysaccharide and type IV pili. In addition, studies have shown that some phage-resistant mutant strains can also affect the growth of bacterial biofilm. The PA1S_08510 gene of P. aeruginosa PA1 strains encodes the O-antigen polymerase Wzy. The phage-resistant mutant strain PA1RG hinders the infection of phage PaP1 which uses O-antigen as a receptor due to the lack of O-antigen on its surface [25]. Meanwhile, the reduction in biofilm in the mutant strain PA1RG can lead to the re-sensitivity of drug-resistant bacteria to some antibiotics. The capsule (K antigen), which acts as a virulence factor, can also act as a receptor for some phages. Phages ΦFG02 and ΦCO01 are able to infect Acinetobacter baumannii (A. baumannii) strains AB900 and A9844, respectively, using the capsule as receptors. The phage-resistant mutants ΦFG02-RAB900 and ΦCO01-RA9844 affected the genes responsible for capsule biosynthesis, gtr29 and gpi, respectively, through a single nucleotide deletion at the K locus. The deletion of the capsule as phage receptors, ΦFG02-RAB900 and ΦCO01-RA9844, results in the interruption of phage adsorption [26]. Phage Phab24 also uses the capsule of A. baumannii as receptors and the outer membrane as secondary receptors [27]. Besides, both studies have found that phage-resistant mutants lacking bacterial capsules can be re-sensitized to antibiotics, which is conducive to the study of phage therapy and phage therapy in combination with antibiotics for the treatment of drug-resistant bacteria. Masking phage receptors by physical barriers such as outer membrane vesicles (OMVs) or EPS can also prevent phages from being adsorbed to bacterial surfaces. Composed of polysaccharides, proteins and nucleic acids, EPS not only enables bacteria to survive in harsh environments but is also significant in fighting against attacks on bacteria by phages and antibiotics [28,29]. The outer membrane protein OmpA of Escherichia coli (E. coli) has been proven to be the receptor of some T-even-like phages. The outer membrane lipoprotein TraT interacts with OmpA to inhibit OmpA-specific phages binding to the receptors and thus inhibit phage adsorption [30]. The cell-binding protein A of Staphylococcus aureus (S. aureus) can also mask teichoic acid which acts as the phage receptor and achieve inhibition of adsorption [31]. There are various types of OMVs secreted by bacteria, which have functions such as transporting virulence factors and bacterial communication [32]. In addition, OMVs can also act as a protective umbrella for bacteria, hindering phage’s adsorption to bacterial surface receptors. The phage infection can be effectively avoided by allowing phages to attach to OMVs [33]. Blocking the injection of phage nucleic acids through Sie is the second line of defense established by bacteria. Sie exists widely in animal [34] and plant viruses [35]. Sie prevents the host cell infected by the temperate phages from being infected again by identical or highly similar phages. Various proteins encoded by phages establish the Sie mechanism by (i) inhibiting phages from binding to receptors, (ii) blocking phage’s DNA injection, and (iii) inhibiting phage tail tube penetration of the plasma membrane [36]. The protein gp05 encoded by temperate phage D3112 is a twitching inhibitory protein that affects the assembly of type IV pilus tail fiber proteins (TFPs) by interacting with the bacterial type IV pilus assembly protein PilB. Thus, preventing phage MP22, which also uses P. aeruginosa TFPs as receptors, from re-infecting the host cell [37] (Figure 2). T-even phages can establish the Sie mechanism after infecting bacteria. For example, when other T-even phages re-infect the host, the T4 phage encodes two proteins, Imm and Spackle, that enable about 50% of the DNA to remain in the head of the phage and the rest to be degraded by endonuclease I in the peripheral space of the host cell, thus preventing the injection of the phage’s DNA into the bacteria [5]. Among them, Imm blocks DNA transfer into the cytoplasm by binding to the receptor on the inner membrane. Spackle blocks the degradation of peptidoglycan. The early phage gene product Sp inhibits the activity of gp5 by forming the Sp-gp5 complex with the phage tail spike protein gp5, hindering the local degradation of host cell peptidoglycan by the lysozyme, and preventing the phage’s tail from entering the cytoplasm. In addition, studies have shown that Sp does not interact with T4 endolysin. On the one hand, the structure of endolysin is different from that of gp5; on the other hand, as a late gene product, endolysin seems more difficult to interact with Sp [38]. Since lytic activity in T4 is not significantly inhibited, host cells containing prophages can still be lysed and release progeny phage particles. Sie has been extensively studied in double-stranded DNA (dsDNA) phages, while single-stranded DNA (ssDNA) phages belonging to the family Microviridae have also been proven to block a phage’s DNA injection through the highly variable region of the DNA pilot protein, thus preventing the repeated infection of other microviruses [39]. The antisense RNA of some phages can regulate the expression of Sie-related genes, thus affecting the mechanism of Sie action. In the lysogenic state, the gene sieB of temperate phage P22 encodes two peptides, namely, the exclusion protein SieB and Esc. The exclusion protein SieB mediated the Sie mechanism, and Esc inhibited SieB. However, since the antisense RNA synthesized by phage P22 has an inhibitory effect on SieB synthesis, and the regulatory gene sieB selectively expresses Esc, P22 still has a chance to re-infect the host cell, thus escaping the Sie mechanism [40]. Blocking the replication of phage nucleic acid in host cells is the third line of defense that bacteria have established. Currently, well-studied defense systems include (i) R-M systems, (ii) CRISPR-Cas adaptive immune systems, and (iii) Abi. In addition, some emerging defense systems are being discovered. Here, several anti-phage defense systems widely distributed in bacteria are introduced in detail. At the same time, the newly discovered defense systems in recent years are briefly listed to enrich the understanding of the mechanism of bacterial anti-phage action. R-M systems are the classical defense system that interferes with the replication of phages in the host, which generally acts in the early stage of phage infection. R-M systems can be classified into four classes (I–IV) [6], among which, type II R-M systems have been the most widely studied. The type I–III R-M system consists of genes encoding restriction endonuclease (REase) and methyltransferase (MTase), while the type IV R-M system contains only REase-related genes. MTase methylates its own DNA recognition site to distinguish unmodified foreign DNA, and REase cleaves the phosphodiester bond of unmethylated foreign phage DNA (Figure 3a). In addition, R-M systems rely on mobile genetic elements to promote bacterial genome evolution through horizontal gene transfer [41,42]. The REase and MTase of the type I R-M system are composed of three subunits, HsdR (DNA-restricted translocation subunit), HsdM (DNA-modified subunit), and HsdS (DNA-specific subunit) encoded by the host-specific determining factor (hsd) gene [43], while the REase of type II R-M system is composed of a single subunit. However, sometimes the R-M systems can lead to the cleaving of its DNA due to false recognition, thus causing autoimmunity [44]. In addition, when the expression of REase and MTase is unbalanced, it can be fatal. When the REase/MTase ratio is too high, the host’s DNA may be cleaved by REase before it is methylated. In contrast, when the REase/MTase ratio is too low, phage DNA is modified by MTase and cannot be cleaved by REase, resulting in the host being infected. After the transformation of E. coli cells with plasmids carrying the Esp1396I type II R-M system, MTase is first synthesized to rapidly modify bacteria genome in order to avoid cleavage of the host DNA by the synthesized REase [45]. On the one hand, the promoter of MTase gene transcription is stronger than that of REase; on the other hand, because type II REase needs to be active in the form of homologous dimer or homologous tetramer, low concentration of gene expression products will limit the formation of polymerization. At present, one of the regulatory mechanisms of type II R-M system expression is dependent on the C protein. Transcription factor C protein in C-dependent R-M systems can regulate the expression levels of REase and MTase and may play different regulatory roles in different R-M systems [46]. The binding site of the C protein, C-box, consists of a pair of reverse repeats that form a negative feedback loop with the C protein homologous dimer. When the C protein concentration is low, the expression of downstream REase is activated. On the contrary, REase expression was inhibited when C protein concentration was high. Although a low concentration of C protein can activate downstream REase expression, the tandem REase promoter still plays a dominant role in the expression of REase gene [47]. C protein is conducive to the expression of MTase before REase, and the absence of the C gene will lead to the premature expression of REase, resulting in REase cleaving of its DNA [48]. The CRISPR-Cas systems defend against phages by recognizing and cleaving phage genes [49]. Moreover, some CRISPR-Cas systems can still exert immunity against methylated phage DNA [50]. CRISPR-Cas are widespread in bacteria, such as the type I-E CRISPR-Cas system found in E. coli K12 [7], the type II CRISPR-Cas system found in Streptococcus agalactiae (S. agalactiae) strain GD201008-001 [51], the type I-C CRISPR-Cas system found in Actinomycetes Eggerthella lenta [52]. CRISPR-Cas systems are of great significance for the evolution of bacteria and for enhancing the adaptability of bacteria to the environment [53]. The process of CRISPR-Cas systems to play an immune role can be divided into three steps: adaptation (acquisition of foreign genes), expression (transcription of CRISPR array, maturation of transcripts and formation of effector complexes), and interference (targeting and cleavage of foreign genes) (Figure 4a). Spacers of the CRISPR-Cas systems are generally obtained at the early stage of phage genes injection into host cells [54]. The CRISPR-Cas systems can insert phage genes (protospacer) into CRISPR sites on the host genome. Different phages protospacer inserts into the bacterial genome constitute rich and diverse spacers in the host genome. The more types of phages that infect the host and the more diverse the spacers, the more beneficial it is for bacteria to defend against infection by different phages [55]. In addition to Cas proteins, the trans-activating CRISPR RNA (tracrRNA), which is partially complementary to crRNA, is also essential in the processing of CRISPR-derived RNA (crRNA) from precursor crRNA (pre-crRNA) [56]. The crRNA and tracrRNA are called guide RNA (gRNA). The gRNA and Cas proteins form the effector complex to target and cleave phage nucleic acids. TracrRNA has only been found in type II and V-B systems, and effector complexes of other CRISPR-Cas systems consist only of crRNA with Cas proteins. The protospacer adjacent motif (PAM) of the CRISPR-Cas systems is located near the protospacer. On the one hand, PAM can assist the Cas protein to recognize foreign genes more accurately and avoid causing host autoimmunity. On the other hand, PAM also provides the possibility for phage point mutations to escape the CRISPR-Cas systems. In addition, crRNA terminal sequences can distinguish host genes by recognizing repeats (about 8 bp) in CRISPR sequences, thereby protecting host genes from cleavage [57]. Nevertheless, the phenomenon of self-targeting spacers still occurs sometimes, and the CRISPR-Cas system mistakenly targets and cuts its own genome, leading to the occurrence of autoimmunity [58]. According to the composition of Cas proteins in effector complexes, CRISPR-Cas systems can be classified into class 1 systems (consistsing of multiple Cas proteins) and class 2 systems (consisting of a single Cas protein) [59]. Each type is further divided into multiple subtypes. Different Cas proteins play different roles, such as the integration of foreign DNA, the maturation of crRNA and the cleavage of foreign genes [60,61]. Cas12a protein in type V CRISPR-Cas acts as the RNase to process pre-crRNA into crRNA, and Cas13 in type VI CRISPR-Cas can play the function of targeting and cleaving single-stranded RNA (ssRNA) [62]. Some Cas proteins play a single role, while some Cas proteins can have multiple functions at the same time. For example, Cas9 plays an indispensable role in all stages of the type II CRISPR-Cas system defense against phages [63]. Obviously, the more phage species, the more abundant and diverse spacers in the host genome, which is more conducive to the evolution of bacterial CRISPR-Cas systems. Additionally, the higher the content of CRISPR spacers, the more sensitive the bacteria to phages [64]. The increase in phage abundance is conducive to the increase in CRISPR-Cas systems abundance. Additionally, when the abundance of phages is constant, the phage species is inversely correlated with the abundance of CRISPR-Cas systems [65]. In addition to phage diversity, bacterial species richness can also have an impact on the evolution of CRISPR-Cas systems. When there are infectious phages in the environment, the more intense the competition between bacterial species is and the more beneficial the anti-phage evolution of bacterial CRISPR-Cas is [66]. Moreover, when there are abundant bacterial species and strong interspecific competition in the environment, bacteria will preferentially adopt CRISPR-Cas rather than surface receptor mutation to defend against phage infection [66]. The acquisition rate of spacers has also been shown to be one of the decisive factors affecting the abundance of spacers. The faster CRISPR-Cas can obtain spacers, the more conducive it is to increaseing the diversity of spacers [67]. In addition, the speed of phage development is also one of the important factors affecting the evolution of bacterial CRISPR-Cas systems. When phages develop too fast in the host cell, it is not conducive for the host to acquire spacers [68]. When the host growth is stagnant, the growth rate of phages in the host is delayed, but the acquisition process of spacers is not affected at this time. Therefore, when the host cell growth is inhibited by the external environment and thus inhibits the speed of phage growth, the evolution of the CRISPR-Cas systems may be beneficial. The Abi immune systems are activated during the middle and late stages of phage maturation. By inhibiting their own metabolism, bacteria lead to their own growth arrest and eventually lead to the death of bacteria, thus avoiding the maturation and release of phages. Although Abi can prevent other bacteria from being infected by mature progeny phages, it is at the cost of host cell suicide, so Abi can also be considered as the last line of defense of the bacterial anti-phage mechanism. Abi can impede phage replication based on (i) CRISPR-Cas systems (Figure 5a), (ii) toxin-antitoxin (TA) systems (Figure 5b), and (iii) cyclic oligonucleotide-based anti-phage signaling system (CBASS) [8] (Figure 5c). This part will introduce the Abi immune systems from three aspects: CRISPR-Cas, TA and CBASS. When phages infect host cells and CRISPR-Cas fails to provide good protection to bacteria, they may choose to mediate Abi to protect other uninfected cells by sacrificing themselves [69]. For example, type I-F CRISPR-Cas of Pectobacterium atrosepticum inhibits the maturation of the lytic phages ΦTE and ΦM1 by mediating Abi [70]. This may be caused by the indiscriminate cleavage of CRISPR-Cas’s own genes, but the specific mechanism of Abi caused by type I-F CRISPR-Cas is still unclear. For example, both Cas14 and Cas12 with a single RuvC domain can nonspecifically target ssDNA [71], and Cas13 can nonspecifically target ssRNA. In fact, the Cas protein with endonuclease activity, which nonspecifically targets phages and bacterial genes, is thought to be one of the causes of CRISPR-Cas mediated Abi. The TA systems can also mediate the Abi immune systems and block phage infection. TA systems also have a variety of functions such as controlling bacterial growth, biofilm formation, maintaining genome stability, and dormancy [72]. This part mainly introduces its function of resisting phage infection. TA systems are widespread in bacteria and consist of toxins and less stable antitoxins. Among them, toxins usually play the role of inhibiting bacterial growth in the form of protein, and antitoxins play the role of inhibiting toxins in the form of protein or RNA [73]. TA systems are divided into six types (Type I–Type VI. The phage activation mechanism of TA systems is not fully clear. At present, a relatively clear molecular mechanism is considered [74]: under normal circumstances, antitoxins can play a role in inhibiting toxins. At this point, the toxin is neutralized, and the TA systems are not activated. However, when phages infect bacteria, the TA transcription function of bacteria is hindered, and unstable antitoxin is degraded before toxin, resulting in toxin accumulation, which leads to bacterial growth arrest or death. Type I TA system toxins (<60 aa) can hinder the synthesis of ATP or act as a nonspecific endonuclease to cleave methylated and unmethylated DNA without distinction, thereby retarding bacterial growth or causing bacterial death [75]. For example, the toxin RalR in the type I TA system of E. coli mediates bacterial death by indiscriminate DNA cleavage, while the antitoxin RalA blocks the translation of toxin proteins by complementary pairing with the mRNA guiding RalR synthesis [76]. The type I TA system Hok/Sok can inhibit phage T4 infection. Phage T4 can block the host transcription process and the antitoxin Sok is not stable. As a result, Sok is degraded first and the toxin Hok is activated, thus inhibiting the growth of host cells [77]. Antitoxins forming complexes with toxins is one of the common ways in the type II TA system that toxins mediate Abi systems by cleaving mRNA as RNase to inhibit bacterial protein synthesis [78,79]. The toxin RnlA in the type II TA system has RNase activity, and the toxin RnlA exists as a homodimer with two conformations [80]. The antitoxin RnlB inhibits RnlA by binding to the HEPN (higher eukaryotes and prokaryotes nucleotide) domain of the toxin RnlA. The antitoxin (the form of RNA) of type III TA interacts with toxins directly to inhibit the toxin [81]. Phage T4 can inhibit the transcription of toxin ToxN and antitoxin ToxI in the type III TA system after E. coli infection. Because antitoxin ToxI is more unstable than toxin ToxN, a large amount of ToxN is accumulated [74]. The toxin ToxN, which acts as an RNase, inhibits phage translation and maturation mainly by recognizing the GAAAU motif and cleaving the mRNA of the phages. When no phages infect bacteria, the ToxI pseudoknot with a sequence length of 36 nt interacts with three ToxNs to form a complex, thereby inhibiting the activity of ToxN [82]. The antitoxin (the form of protein) of type IV TA systems binds to the target of action of the toxin rather than forming a complex with the toxin protein. For example, in the type IV TA system of S. agalactiae, the antitoxin AbiEi inhibits the binding of the toxin AbiEii, which acts as a nucleotidyltransferase (NTase) to GTP [83]. The antitoxin GhoS of the type V TA system can specifically recognize and cleave the U- and A-rich sites in the mRNA of toxin GhoT, thus hindering the dissolution of the cell membrane by GhoT [84]. Different from the TA systems found in the past, in the type VI TA system, the toxin SocB, which can bind to the sliding clamp to hinder the replication process, can be degraded by the protease ClpXP with the participation of the antitoxin SocA [85]. Moreover, the type I TA system in Clostridium difficile (C. difficile) co-localizes with CRISPR arrays [86]. This means that CRISPR-Cas and TA systems may have some potential connection in mediating Abi. In addition, CBASS is also one of the ways to mediate bacterial Abi. The CBASS immune system is widespread in bacteria and can impede phage replication. After a phage infection, cGAS/DncV-like nucleotide transferase (CD-NTases) in bacterial CBASS can synthesize second messengers such as cyclic dinucleotides and cyclic trinucleotides [87]. The CD-NTase-related protein Cap, which acts as an effector protein, is activated after binding to specific second messengers and induces cell death by disrupting cell membranes, cleaving intracellular DNA, or other means. The crystal structure of the Cap protein can determine the type of the second messenger that binds to it [88]. After Yersinia is infected by phages, the receptor domain of oligomeric cyclic dinucleotide formed by the 8-stranded β-barrel scaffold specifically binds to signaling molecules (second messengers) to promote a bacterial inner membrane rupture and mediate bacterial death [89]. Cyclic AMP-AMP-AMP (cAAA) can also induce bacterial death after binding to the homotrimer DNA endonuclease NucC (nuclease, CD-NTase-associated) in CBASS. NucC can induce bacterial death by cleaving intracellular DNA, thereby hindering phage replication in the cell. In addition, the combination of cAAA with the triple symmetric allosteric pocket of NucC can promote the formation of NucC homohexamer, which also has DNA cleavage activity [90]. Cap4 in Enterobacter cloacae (E. cloacae) can form a SAVED domain by the fusion of two CARF (CRISPR-associated Rossman fold). After specifically binding 2′-5′- and 3′-5′ -linked cyclic oligonucleotide signals, the SAVED domain activates the dsDNA endonuclease activity of Cap4, and then cleaves intracellular DNA, thereby hindering the viral replication process [91]. CBASS can be divided into four types according to different cyclase genes, auxiliary genes (cap) and signaling molecules [92]. Type I CBASS (42%) had no cap gene and induced bacterial death by forming small holes in the bacterial membrane; type II CBASS (39%) contained cap2 and cap3 genes; type III CBASS contained three auxiliary genes: cap6 (encoding TRIP13/Pch2 domains), cap7 (encoding a single HORMA domain) and cap8 (encoding two HORMA domains); type IV CBASS contains cap9–11 gene, which is small in quantity and it is not clear whether type IV CBASS can resist phage infection. Among them, the HORMA domain in type III CBASS can form HORMA complexes that can synthesize signaling molecules after binding with CD-NTases [93]. Additionally, TRIP13 ATPase can exert a negative regulatory effect by decomposing the HORMA complex. The aforementioned mechanisms that hinder phage adsorption, injection and replication are all introduced from the level of individual bacteria. In addition, bacteria can also be regulated as a group through QS [94]. QS, which has the function of intercellular communication, is composed of quorum-sensing signal synthase, extracellular signaling molecules called AIs and receptors [95]. AIs synthesized by quorum-sensing signal synthase can play a role in regulating population density [96], regulating virulence factors [97] and resisting phage infection after binding with the corresponding receptor. The ways of QS resisting phage infection can be divided into two: (i) cooperating with CRISPR-Cas systems to stimulate the expression of cas genes [98] or promoting the identification and cleaving of target genes [9]; (ii) inhibiting phage adsorption by directly reducing phage receptors. Under the condition of high population density, the QS system can enhance the recognition and cleavage of phage genes by CRISPR-Cas systems [99]. For example, under the condition of high population density, the QS of P. aeruginosa PA14 can promote the expression of cas gene-encoding nuclease [98]. The smaI/smaR-type QS system can also enhance the immunity of CRISPR-Cas systems to phages in Serratia under the condition of high population density [100]. The quorum sensing signal synthase smaI can synthesize the acyl-homoserine lactones (AHL) signaling molecule N-butanoyl-L-homoserine lactone (C4-HSL). When the host population density is high, C4-HSL increases. C4-HSL can bind to its receptor smaR, thus blocking the inhibition of cas gene by DNA-binding repressor smaR (Figure 6a). In addition, the smaI/smaR-type QS system can also promote the ability of type I-E and I-F CRISPR-Cas systems to capture spacers, thus enhancing the adaptability of CRISPR-Cas systems. Thus, at high host population densities, the facilitation of the CRISPR-Cas system by the QS system may be aimed at reducing the adaptive cost of bacterial resistance to phages. QS can directly reduce phage adsorption receptors to hinder phage adsorption. For example, the binding of AHL in E. coli with its receptor SdiA can reduce phage λ receptor LamB and hinder phage adsorption [101] (Figure 6b). In addition, the QS system can also increase the biofilm of bacteria under the condition of low cell density [102] (Figure 6c). The increase in biofilms may be beneficial for masking phage receptors. However, the use of biofilms to mask receptors or to reduce the number of receptors does not seem to be the main way for the QS system to defend against phages. Under the condition of high population density, biofilm production may be inhibited rather than promoted by the QS system [103]. Bacteria can block phage adsorption by disabling the expression of the receptor, changing the structure of the receptor or masking the receptor with a physical barrier formed by EPS. Given the mechanism of bacteria hindering phage adsorption, phages can (i) recognize new receptors or (ii) degrade EPS by depolymerase [10], and then acquire the ability to recognize and adsorb the host again (Figure 1b). Receptor binding protein (RBP) J is the host-recognizing trimer protein. Phage λ originally recognized the LamB receptor of host E. coli by the J protein. However, when phage λ is co-cultured with E. coli lacking the LamB receptor, phage EvoC, which recognizes the new receptor OmpF, can be isolated [11]. Additionally, both OmpF and LamB are trimeric structures. The RBP mutation of phages is not only an anti-defense measure against bacteria but is also conducive to the expansion of the host selection range of phages [104]. In addition, phages can also degrade EPS through depolymerase and acquire the ability to recognize the receptor. Depolymerases are divided into lyases (water-free molecules after the substrate is cleaved) and hydrolases [105]. Depolymerase is widely found in phages. For example, A. baumannii phage IME200 can express the depolymerase Dpo48 [106], and phage IME205 can express the depolymerases Dpo42 and Dpo43 [107]. Phage Sb-1 can degrade the EPS of methicillin-resistant S. aureus. Furthermore, the lack of EPS protection can increase the sensitivity of bacteria to antibiotics [108]. Therefore, phage depolymerase can not only enhance the recognition ability of the host but also be used in combination with antibiotics as a new type of antibiotics [109]. Aiming at the cleavage of unmethylated phage genes by bacterial R-M systems, phages (i) modify their own genes to block the cleavage of phages nucleic acids by REase (Figure 3b) or (ii) evolve the overcome classical restriction (Ocr) protein and the restriction of DNA A (ArdA) protein that can inhibit the R-M complex (Figure 3c). Gene modification is one of the ways that phages resist the R-M systems. For example, phages T4gt, T4, Xp12, and SP8 modify pyrimidine in their DNA to 5-hydroxymethylcytosine (5hmC) and glucose-5-hydroxymethylcytosine, respectively. 5-methylpyrimidine and 5-hydroxymethyldeoxyuridine resist type II REase [12]. The iron-binding protein Mom produced by phage Mu can methyl carbamoylation its genes after binding to the cofactor acetyl CoA and Fe2+/3+, thus, it can resist various REases [110]. There is a 7-deazaguanine modifier gene cluster with a length of about 5940 bp in phage CAjan, including genes folE, queD, queE, queC, yhhQ and dpdA [111]. GTP can be converted to 7-cyano-7-deazaguanine (preQ0) under the catalysis of four enzymes (FolE, QueD, QueE and QueC) [112]. Phage CAjan inhibits the cleavage of phage DNA by restriction endonucleases with GA and GGC as recognition sites by modifying GTP to preQ0 in the specific sequences GA and GGC in phage DNA. In addition, Ocr and ArdA produced by phages can also inhibit the type I R-M system. Both Ocr dimer and ArdA dimer bind to the gap between HsdRs subunits motor 1 and motor 2 in a manner that mimics DNA, inhibiting the binding of the R-M complex to phage DNA, but not its conformational transition [13]. The Ocr dimer expressed by phage T7 occupies the DNA binding site in the type I R-M system restriction enzyme EcoKI (R2M2S1) by simulating the negative surface charge of DNA (~24 bp) and the approximate 46° bending of the DNA helix axis [113,114]. The Ocr protein has also been proven to inhibit BREX defenses similar to R-M systems. The Ocr dimer expressed by phage T7 binds to the BREX system complex by simulating the shape and surface charge of DNA (~20 bp) and inhibits the BREX methylation of adenine fifth in the specific non-palindromic sequence of host DNA [115]. Furthermore, the Ocr dimer can also compete with sigma factors for nucleic acid binding channels of bacterial RNA polymerase, which inhibits sigma factors from recruiting RNA polymerase to bacterial DNA promoters, thus impeding the host transcription process [116]. CRISPR-Cas adaptive immune systems are one of the most important strategies for bacteria to fight phage infection. In the face of precise attacks by bacteria, phages have also evolved a series of counterattack measures to counter the bacterial defense systems. The counterattack strategies of phages against bacterial CRISPR-Cas mainly include (i) constructing nucleus-like compartments to shield nuclease; (ii) mutating the target sequence to block the recognition of effector complexes; (iii) through Acr to inhibit the recognition or cleaving of foreign nucleic acids by effector complexes (Figure 4b). Recent studies have proven that giant phages can produce a proteinaceous nucleus-like compartment, and this protein shell can act as a physical barrier to protect phage dsDNA from nuclease hydrolysis [15]. For example, the Serratia giant phage PCH45 without the acr gene and DNA-modifying enzyme gene can encode the nucleus-like shell to resist bacterial CRISPR-Cas systems through the gp033 gene, and then successfully infect Serratia with type I-E and I-F CRISPR-Cas [117]. However, because the phage protein translation process is located in the bacterial cytoplasm, mRNA, which is not protected by the nucleus-like compartment, can be targeted for cleavage by type III and VI CRISPR-Cas systems. Therefore, the phage-constructed nucleus-like compartment cannot resist CRISPR-Cas systems of type III and VI (the type III system can target foreign DNA transcripts, and Cas13 in the type VI system can cleave ssRNA [62]). The giant phage ΦKZ without the acr gene can also construct the nucleus-like compartment to resist bacterial types I-C, I-F, II-A and V-A CRISPR-Cas, but not type III CRISPR-Cas systems [118]. Furthermore, the nucleus-like compartment not only antagonizes the CRISPR-Cas adaptive immune systems but also impedes the R-M systems by shielding restriction enzymes [118]. Phages can block the recognition of target genes by CRISPR-Cas systems through the DNA glycosylation or mutation of PAM/protospacer. For example, the glycosylation of 5hmC in the DNA of phage T4 by glycosyltransferase can inhibit the recognition of target genes by type I-E and II-A CRISPR-Cas systems [119]. The mutation of PAM and protospacer of phage M13 enabled the phages to successfully infect E. coli [14]. However, the CRISPR-Cas systems have a certain fault tolerance rate for the mutation of the target gene, and the CRISPR-Cas systems can re-resist the phages by obtaining the new spacer [120]. The Acr evolved by phages is a widely discovered and studied anti-CRISPR-Cas strategy. At present, Acrs have been found to hinder the recognition or cleavage of foreign nucleic acids by effector complexes in CRISPR-Cas systems such as type I-C, I-F, II-A, II-C, III-A, V-A and VI-B by acting alone or in combination [16]. The identified type I-F Acrs include AcrIF1-4, AcrIF7-9, AcrIF11, AcrIF14, etc. For example, AcrIF4 blocks the formation of the active conformation of the Csy1 complex by interacting with the I-F CRISPR-Cas surveillance complex (the Csy complex) [121]; AcrIF7 mimics the bases of the target sequence and occupies the binding site of the Csy complex to DNA [122]; AcrIF9 interacts with Csy3 and induces non-specific binding of the Csy complex to DNA, which makes the Csy complex lose its specific targeting ability [123]; AcrIF11 deprived the Csy complex of dsDNA binding activity [124]. Gene35 of phage JBD30, gene30 of phage D3112 and gene35 of phage JBD5 encode AcrIF1, AcrIF2 and AcrIF3 with anti-I-F CRISPR-Cas function, respectively [125]. Among them, AcrIF1 and AcrIF2 bind to the Csy3 and Csy1-Csy2 heterodimers of the Csy complex, respectively. AcrIF3 affects Cas3 recognition and recruitment by directly binding to Cas3 nuclease rather than binding to the Csy complex, thereby preventing Cas3 from cleaving phage DNA. Currently, the reported type II Acrs include AcrIIA1-2, AcrIIA4, AcrIIA14, AcrIIA22-23, AcrIIC1-4, etc. The mechanism of action of traditional AcrII can be divided into three categories: (i) interacting with the HNH domain of Cas9 (responsible for cleaving complementary chains) or the RuvC domain (responsible for cleaving non-complementary chains), (ii) competing with PAM recognition sites, and (iii) blocking sgRNA recruitment. The C-terminal domain of AcrIIA1 can interact with the HNH domain of Listeria Cas9 to induce the inactivation and degradation of Cas9 through a multi-step mechanism [126]. Both AcrIIA2 and AcrIIA4 can hinder the recognition of target genes by occupying the PAM recognition site of Cas9 [127,128]. In addition, AcrIIA4 can also interact with the RuvC domain of Cas9 to block Cas9 from cleaving target genes [129]. The C-terminal domain of AcrIIA14 binds to the HNH domain of Cas9, which can inhibit the cleavage activity of Cas9 [130]. Similarly, AcrIIC1 can also directly bind to Cas9 and inhibit the cleavage activity of Cas9 [131]. The highly negatively charged AcrIIC2 can bind to the positively charged bridge helix of Cas9. AcrIIC2 prevents gRNA from forming complexes with Cas9 by occupying the gRNA binding site on Cas9 [132]. The two AcrIIC3 links the HNH domain of Cas9 and the other REC2 domain of Cas9, causing Cas9 dimerization and hindering the formation of the active conformation of the HNH domain, thus inhibiting the binding and cleavage of Cas9 to the target gene [133,134]. In addition, AcrIII, AcrV and AcrVI have been gradually discovered. AcrIII-1 with ring nuclease activity degrades cyclic tetra-adenylate (cA4) into A2>P and A2-P in the form of a dimer, thereby inhibiting the activation of RNase by cA4 [135]. Negatively charged AcrVA1 can occupy the PAM binding site of Cas12. Moreover, after AcrVA1 binds to Cas12, crRNA can be cleaved into two parts by Cas12. Additionally, AcrVA4 dimer can block the formation of the active conformation of Cas12a [136]. Similarly, after binding to Cas13a, AcrVIA1 inhibits the formation of active conformations of Cas13a that can bind to target RNA [137]. AcrVIA2 or AcrVIA3 can bind to the Cas13a-crRNA complex instead of Cas13a, hindering the cleavage of the target RNA [138]. In addition, Chevallereau A, et al. demonstrated that when Acr-positive and Acr-negative phages co-infected bacteria, the presence of Acr-positive phages facilitated the maturation of Acr-negative phage replication in the host cell [139]. With the deepening of the research on phages Acr, more and more new Acr are being discovered. This helps people to enrich their understanding of phage’s anti-bacterial mechanisms. Besides, the anti-CRISPR-associated (aca) gene, which is found near the acr gene, encodes the protein Aca, which contains a helix-turn helix domain. The Aca2 homodimer of Pectobacterium carotovorum phage ZF40 binds to the acr promoter and can inhibit the transcription of acr [140]. The aca1 gene located downstream of the acrIF1 gene in P. aeruginosa phage JBD30 can express the Aca1 protein. Aca1 binds to the acrIF1 promoter in the form of the homologous dimer to inhibit the transcription of the acrIF1 gene [141]. For the QS system of bacteria, phages’ counterattack strategies include: (i) expressing anti-repressor and binding cI repressor to make phages enter the lysis cycle (Figure 7a); (ii) synthesizing receptors that can bind to AIs or synthesizing AIs-like proteins to prevent AIs from binding to their corresponding QS receptor. Temperate phages are essential in the evolution and diversity of microbial populations [142]. Temperate phages do not lysis host cells and produce progeny phages during the lysogenic cycle, but make their genes integrate with host bacteria chromosomes and pass along with the division of bacteria. When the phages enter the lysis cycle from the lysogenic state, they can produce mature progeny phages, which mediate the lysis death of bacteria. The switch between the phage’s lysogenic cycle and lysis cycle is the cI repressor in host cells. By binding with the Q promoter, the cI inhibitor inhibits the expression of the phage’s lysing gene, leaving the phages in the lysogenic state, so that no progeny phages can be produced [143]. However, the antirepressor can inhibit cI activity, causing the phages to enter the lysis cycle and promote host cell lysis. The repressor was first identified in phage P22 and was named Ant [17]. A repressor named Qtip was found in phage VP882 [144]. Under the condition of high cell density, Qtip composed of 79 amino acids can recognize and bind the DNA-binding domain of the N-terminal of cI, inhibiting the activity of cI [145,146] (Figure 7b). It is a common strategy for phages to resist the QS system to prevent AIs from binding to their receptors by producing receptors that can bind to AIs or by synthesizing proteins similar to AIs. For example, the gene p37 that encodes the LuxR-type transcription factor was found in phage ΦARM81ld [18]. LuxR encoded by p37 can bind to C4-HSL as AIs, which hinders the binding of C4-HSL and LuxR in the bacterial QS system. Phage VP882 encodes the VqmA QS receptor (VqmAPhage), a homology of Vibrio cholerae VqmA (VqmAVc) [143]. VqmAPhage can bind to 3,5-dimethylpyrazin-2-ol (DPO), which hinders the binding of VqmAVc and DPO (Figure 7c). Furthermore, the binding of VqmAPhage to DPO can activate the expression of the anti-inhibitory factor Gp55. Gp55 can directly act on the cI repressor to inactivate cI, thus causing phages to enter the lysis cycle. Phage DMS3 encodes Aqs1 protein that can bind to LasR, hindering the binding of P. aeruginosa AIs to its receptor LasR [147]. Aqs1, which consists of 69 residues, binds as a dimer to the N-terminal DNA-binding domain of LasR, hindering AHL binding to lasR in the lasR/lasI-type QS system. Aqs1 also inhibits the production of another class of AIs, the quinolone system (PQS), by down-regulating the pqsABCDE and phnAB operons [148] (Figure 7d). The quorum-sensing targeting protein encoded by phage LUZ19 can also inhibit PQS production by interacting with PQS biosynthetic pathway enzymes [149]. The mutual defense strategies between bacteria and phages are gradually being clarified. This review details bacterial anti-phage strategies and phages anti-defense measures to deeply understand the interaction mechanism between bacteria and phages, which is significance for the development and application of modern biotechnology. At present, the defensive and anti-defensive measures between bacteria and phages have been able to solve many practical problems. For example, CRISPR-Cas technology can be used for pathogen nucleic acid detection [150], CRISPR-Cas gene editing technology for designing new strains with enhanced beneficial functions [151], the depolymerase produced by phages can be used as a new anti-biofilm agent [152], the problem of phage contamination during fermentation can be solved based on the interaction between phages and bacteria [153] and phage therapy can be used to treat bacteria, especially to treat infections with drug-resistant bacteria [154]. The emergence of superbugs poses a serious threat to human health, making the problem of drug resistance to bacteria the focus of global attention. Resistant genes can be transferred between bacteria to create new resistant combinations [155]. As a new type of therapy, phage therapy has been used in practical clinical treatment [156]. Phage therapy is expected to be used in combination with antibiotics to address the problem of bacterial resistance. One of the problems facing phage therapy is that bacteria evolve their resistance to phages [157]. The biofilm that forms when P. aeruginosa infects the lungs of people with cystic fibrosis can block antibiotics from entering bacterial cells [158]. At the same time, as a natural barrier, biofilm facilitates bacteria to stay on the surface of living and non-living organisms [159]. This can easily lead to nosocomial infections. Giant phages are often used to treat CF patients infected with P. aeruginosa. Phages can not only lysis bacteria but also reduce biofilms through polysaccharide depolymerase. However, the emergence of phage-resistant mutant strains prevents phage therapy from treating resistant bacteria. Understanding the evolution direction of phage-resistant mutant strains is important for the treatment of drug-resistant bacteria relying on cocktail therapy or the combination of phage therapy and antibiotics. Phage cocktail therapy, which consists of different phages, is more beneficial for the treatment of drug-resistant bacterial infections [160]. In conclusion, some of the defensive or anti-defensive systems found in the current research may be effective tools for solving practical problems in the future. Among the defense and anti-defense measures between bacteria and phages, R-M, CRISPR-Cas, Abi and QS systems have been studied more deeply, while the anti-defense measures taken by phages against bacterial mutant receptors have been less studied. Besides, there are many new defensive and anti-defensive measures whose specific mechanisms of action have not been elucidated and need to be further studied. The arms race between bacteria and phages is conducive to rapid coevolution between them. Meanwhile, temperate phages can also promote the adaptive evolution of the host [161]. Phages and their hosts can exchange genes through horizontal gene transfer, driving coevolution [162]. The coevolution of bacteria and phages can increase the mutation rate of bacteria. The mutant bacterial population can play a greater advantage in the arms race with phages [163]. At the same time, coevolution can also improve the rate of phage’s evolution [164]. However, the role of the community environment in the interaction between bacteria and phages is not fully understood, and many problems still need to be further explored [165].

PMC10002425

Morgan S. Sobol,Anne-Kristin Kaster

Back to Basics: A Simplified Improvement to Multiple Displacement Amplification for Microbial Single-Cell Genomics

21-02-2023

whole genome amplification,miniaturization,cell sorting,microbial dark matter,contact-free liquid dispenser

Microbial single-cell genomics (SCG) provides access to the genomes of rare and uncultured microorganisms and is a complementary method to metagenomics. Due to the femtogram-levels of DNA in a single microbial cell, sequencing the genome requires whole genome amplification (WGA) as a preliminary step. However, the most common WGA method, multiple displacement amplification (MDA), is known to be costly and biased against specific genomic regions, preventing high-throughput applications and resulting in uneven genome coverage. Thus, obtaining high-quality genomes from many taxa, especially minority members of microbial communities, becomes difficult. Here, we present a volume reduction approach that significantly reduces costs while improving genome coverage and uniformity of DNA amplification products in standard 384-well plates. Our results demonstrate that further volume reduction in specialized and complex setups (e.g., microfluidic chips) is likely unnecessary to obtain higher-quality microbial genomes. This volume reduction method makes SCG more feasible for future studies, thus helping to broaden our knowledge on the diversity and function of understudied and uncharacterized microorganisms in the environment.

Back to Basics: A Simplified Improvement to Multiple Displacement Amplification for Microbial Single-Cell Genomics Microbial single-cell genomics (SCG) provides access to the genomes of rare and uncultured microorganisms and is a complementary method to metagenomics. Due to the femtogram-levels of DNA in a single microbial cell, sequencing the genome requires whole genome amplification (WGA) as a preliminary step. However, the most common WGA method, multiple displacement amplification (MDA), is known to be costly and biased against specific genomic regions, preventing high-throughput applications and resulting in uneven genome coverage. Thus, obtaining high-quality genomes from many taxa, especially minority members of microbial communities, becomes difficult. Here, we present a volume reduction approach that significantly reduces costs while improving genome coverage and uniformity of DNA amplification products in standard 384-well plates. Our results demonstrate that further volume reduction in specialized and complex setups (e.g., microfluidic chips) is likely unnecessary to obtain higher-quality microbial genomes. This volume reduction method makes SCG more feasible for future studies, thus helping to broaden our knowledge on the diversity and function of understudied and uncharacterized microorganisms in the environment. The vast majority of bacteria and archaea remain understudied since they have not yet been successfully cultured; thus, their genomes, metabolic potential, and functions in the environment remain unknown [1,2,3,4]. We refer to these microorganisms as microbial dark matter (MDM) [5]. Within MDM hide potentially novel and important solutions for sustainable energy, bioremediation of contaminated environments, and the war against rising antibiotic resistance [6,7,8,9,10]. The use of culture-independent methods to study microorganisms, such as metagenomics, has significantly advanced our understanding of MDM. However, metagenomics still struggles to reliably assemble true, individual genomes due to strain variations and misattribution of sequences to the wrong genomes [11,12]. Furthermore, highly repetitive sequences like those found in CRISPR regions [13,14] are often not accurately assembled and 16S rRNA sequences, as well as mobile genetic elements such as plasmids, are often not attributed to their host organisms [15,16]. As a result, insights into evolutionary mechanisms, like horizontal gene transfer, are lost. Therefore, single-cell genomics (SCG) was developed as a complementary tool to enable the analysis of individual cells, thereby expanding our knowledge of MDM taxa [17,18,19]. In general, a microbial SCG workflow (Figure 1) involves (A) sample collection and preservation, (B) specific or non-specific cell staining, (C) cell sorting, (D) cell lysis, (E) whole genome amplification (WGA), and (F,G) genome sequencing and analysis [17,18]. The WGA step is crucial for generating a sufficient amount of input DNA for library preparation and subsequent sequencing, as a typical microbial cell only contains a few femtograms (fg) of DNA [20,21]. Several different WGA methods have been developed and improved upon over the years. These methods can be categorized as polymerase chain reaction (PCR)-based amplification, isothermal amplification, and hybrid, which combines both methods [22]. Pure PCR-based methods, such as degenerate oligonucleotide primed PCR (DOP-PCR) [23], were not successfully applied to microbial single-cells, likely because of sensitivity issues to the low amount of input DNA. The first method to amplify DNA from a single bacterial cell was the so-called multiple displacement amplification (MDA) [24] (Table 1, Appendix A Figure A1-A). MDA is an isothermal method that uses the high-fidelity phi29 polymerase, which has a lower error rate (1 in 106 bases) compared with standard polymerases used in PCR, 3′ → 5′ exonuclease proofreading activity, and generates fragments larger than 10 kb [25,26,27,28] (Table 1). Currently, MDA remains one of the most widely applied methods for amplifying DNA from microbial single cells for these reasons [17]. Unfortunately, MDA also constitutes one of the major limitations in single-cell sequencing due to its high costs (Table 1), as well as its bias against high GC regions, which leads to uneven genome amplification [29,30,31]. Furthermore, artifacts like chimeras and non-specific products can be produced and are thought to occur randomly since sequences that are over-represented in one MDA reaction can be under-represented in another [29,31]. However, some have found these effects to be reproducible due to the fact that a decreased template copy number increases bias and certain sequences are simply not amplified at all [25,29,30,32]. As a result, treatments such as post-amplification endonuclease degradation and post-amplification normalization by nuclease degradation of dsDNA have been used to reduce chimeric sequences [33] and highly abundant sequences [20], respectively. Other approaches have worked to improve MDA itself, such as WGA-X™, which uses a more thermostable phi29 polymerase for better amplification of high GC organisms [34] (Table 1, Appendix A, Figure A1-A). However, lower genome coverage for organisms with a low GC content compared with standard MDA is reported. More recently, primary template-directed amplification (PTA) was developed, which employs exonuclease-resistant terminators to create smaller amplicons that undergo limited subsequent amplification to limit over-representation of random positions and reduce error propagation [35] (Table 1, Appendix A, Figure A1-B). While this method looks promising to reduce amplification bias, the approach is still in the alpha testing stage for microorganisms (https://www.bioskryb.com/resolvedna-microbiome-alpha/ (accessed on 1 August 2022)) and quite expensive. The hybrid method, multiple annealing and looping based amplification cycles (MALBAC), combines PCR and MDA methods to successfully reduce amplification bias [36,37] (Table 1, Appendix A, Figure A1-C). Yet, MALBAC remains widely unused in microbial SCG, because the Bst and Taq polymerases have higher error rates because they lack proofreading capability [38]. Thus, further work needs to be done to optimize MALBAC, possibly with phi29 or less error-prone enzymes [39]. Even though there is hope to reduce amplification bias in microbial WGA, statistically, inconsistency and bias among the DNA amplification of millions of templates will still persist [41]. In addition, WGA methods are highly sensitive to contamination due to the low amounts of DNA from a single cell. Prior decontamination of reagents with UV [42] can help to remove common reagent contaminants, but this does not prevent other sources of endogenous and/or exogenous contaminants, which become more amplified in larger WGA reaction volumes due to reduced polymerase specificity [43]. Therefore, through bioinformatics, contamination in SAGs needs to be analyzed and removed prior to downstream analysis [12]. Moreover, the large, recommended reaction volumes of these WGA methods also quickly become very costly when applied to high-throughput SCG (Table 1). These high costs limit the depth at which samples can be analyzed, preventing, for example, minority taxa from being captured with SCG. Therefore, a methodically simpler solution is to reduce WGA’s reaction volume. Reduction of total WGA volume has been shown to increase the concentration of the template and lessens the chance of background contamination being amplified [43]. Furthermore, this approach also significantly reduces the high costs of WGA (Table 1). Previous studies have applied this approach at sub-nanoliter (nL) and picoliter (pL) volumes in microfluidic devices [38,40,44,45,46,47,48], nanowells [49,50], planar surfaces [51,52], and hydrogels [53], which are compared in detail in Figure 2. Many of these approaches and their devices remain largely unused outside of their respective publications, likely because most microfluidic chips and other platforms are not commercially available; they require complex fabrication and operation [54,55], and are therefore hard to access and implement in other research groups. Commercially available options, such as 10× genomics®, BD RhapsodyTM, and Fluidigm® C1 are costly, less flexible, and geared towards eukaryotic cells. Additionally, current droplet-based technologies sort based on Poisson distributions of cells, resulting in high unoccupancy and low cell recovery [56], which is not applicable for studies analyzing rare populations [57]. Other approaches such as the use of planar substrates require special care to avoid contamination and evaporation, while hydrogel matrices lack the throughput needed for microbial SCG (Figure 2). Hence, the establishment of a reliable and easy-to-use volume reduction method is needed to widen the accessibility and application of microbial SCG. Surprisingly, there is a lack of information on how bias can be simply reduced between the low- to sub-microliter range within standard 384-well plates. Reduction of standard MDA reaction volumes down to 1.2–2.0 µL have been previously reported [18,58]; however, a systematic assessment of its effect on MDA bias and genome completeness has not yet been done before. Therefore, in this study, we compared the amplification bias in single-amplified genomes (SAGs) of Escherichia coli from 10 µL total MDA reaction volumes down to 0.5 µL using novel acoustic liquid dispensing technology developed by Dispendix GmbH (https://www.dispendix.com/ (accessed on 1 August 2022)). Our results indicated that an MDA reaction volume of 1.25 µL is the “sweet-spot” for significantly reducing amplification bias and increasing assembly coverage up to almost 90%, offering an easily accessible approach for future SCG studies to improve WGA in a cost-effective manner. Previous studies show that volume reduction improves polymerase specificity through “molecular crowding” [59,60]. Molecular crowding reduces competition between amplification of the template and contamination by increasing the probability that polymerase and primers bind to template DNA and reducing spurious binding [59,60]. Moreover, lower reaction volumes reduce the amount of surface area for nonspecific adsorption of nucleic acids to the multi-well plate walls [61,62]. However, too much crowding can also cause adverse effects by causing sterical hinderance and reducing the polymerase from accessing the template [63,64]. Here, we sorted single E. coli cells into 384-well plates to compare SAG amplification bias within total MDA microliter and sub-microliter reactions for the first time. In contrast, previous studies have largely examined volume reduction in the sub-nanoliter to picoliter range [38,44,45,47,49,50,51,65]. MDAs with total reaction volumes of 0.5, 0.8, 1.0, 1.25, 5.0, and 10 µL were conducted in 384-well plates (Appendix A, Table A1). The smallest-sized MDA reaction, 0.5 µL, did not work and the amplification success rate for the 0.8 and 1.0 µL reactions was only 68.75% and 62.50%, respectively. In comparison, the success rate for the 1.25 µL MDA reactions was 87.50%, whereas both the 5.0 and 10 µL MDA reaction volumes had a success rate of 100%. The lower success rates in the lower MDA reaction volumes was likely due to evaporation and/or sterical hinderance of the polymerase in the small volumes [63,64]. On average, the time that it took for the amplification to reach the detection threshold (indicated as Cq; quantification cycle) was earliest for the 1.25 µL MDA reaction volumes (Figure 3A). Previous studies have reported that earlier Cq values indicated higher genome recovery success and quality [34,66]. Additionally, our detected relative fluorescence (RFU) endpoints and DNA yields from the successful reactions decreased as reaction volumes decreased (Figure 3A,B), initially indicating that volume reduction likely limited the exponential nature of MDA [38,45], which should improve genome coverage and uniformity. To further compare the quality of the WGA reactions, a total of five amplified replicates for each different MDA reaction volume were chosen based on their Cq and RFU values, then subjected to Illumina sequencing using equal amounts of DNA (Appendix A, Table A2). There was a significant difference in reads lost during read trimming between the different MDA reaction volumes (Figure 4A, p = 0.0002 Appendix A, Table A3). On average, the 1.25 µL sized MDA reactions lost significantly fewer reads to read quality trimming compared with all other reaction volumes (p ≤ 0.05). After read trimming, all samples were normalized to a 200× sequencing depth before further read processing steps to ensure a fair comparison between the mapping and assembly quality of the different reaction volumes. After depth normalization, the number of duplicated reads was, on average, greater in larger-sized volume reactions (Figure 4B), but the difference between all reactions of the different volumes was not found to be significant (p = 0.0870, Appendix A, Table A3). While some amount of read duplication inevitably results from MDA’s exponential amplification nature, comparisons of the percent duplicates between samples could still provide insight into the specificity of the amplification itself. A higher number of duplicates can be caused by the lower template specificity in large MDA reactions causing more spurious priming and amplification [40,51], especially when template concentrations are very low [67]. Furthermore, the issue of lower template specificity also explains why there was an observed trend that larger reaction volumes had more contaminant reads removed after filtering than the smaller reactions (Figure 4C). Lower specificity, leading to more contamination, is likely due to the increased competition between background contamination and the E. coli single-cell DNA [40,49]. This increase in contamination was also reflected in the higher amplification gain and product yield mentioned previously (Figure 3A,B), which other studies reported as well [45,47,49]. In general, we also observed that 5 and 10 µL MDA reaction volumes gave less consistent results, as evidenced by larger variation between replicates (Figure 4). As a consequence of lower template specificity, the MDA reaction volumes above 1.25 µL also performed worse during read mapping to the reference E. coli MG1655 genome, as indicated by genome coverage breadth and coverage uniformity (Figure 5A,B). MDA in 0.8 and 1.0 µL reaction volumes also resulted in low coverage breadth and uniformity, and a reaction volume of 1.25 µL was therefore determined as the “sweet-spot” for improved MDA in 384-well plates. Likely, the 0.8 and 1.0 µL reaction volumes were simply too low, causing too much molecular crowding, sterically hindering the polymerase from fully accessing the template DNA [63,64], and/or there was too much evaporation. Reduced genome coverage was also recently reported for MDA reaction volumes below 150 nL on a microfluidic system [44], suggesting that platforms independently have a specified “sweet-spot” for efficient MDA. On average, reads from 1.25 µL MDA reaction volumes covered 85 ± 13% of the E. coli genome, which was 19% to 40% more than the other sized reactions (Figure 5A). This increase in coverage was a large improvement when compared to current, well-established methods like WGA-X™, which gives a reported ~36 ± 21% read coverage of E. coli in a standard 10 µL reaction [34]. When compared to 10 µL reactions in this study, we still noted approximately 19% greater coverage breadth than WGA-X™, even though we used ~2 million fewer reads during read mapping. Likely, this difference can be attributed to the lysis modified specifically for E. coli herein. Furthermore, the average genome coverage in our study is ~45% greater than MDA performed in ~60 nL hydrogel reactions [53]. Here, the much lower coverage for E. coli could be due to the fact that the authors performed a second round of MDA, which has been shown to increase bias [38]. Our reported coverages are also well within range of those reported from a different nanoliter microfluidic method [38], as well as from picoliter droplet reactions [47], at the same sequencing depth (Appendix A, Figure A2). It should be mentioned that one other study reports ~15% greater coverage from MDA in nanoliter microwells when compared with our 1.25 µL average genome coverage at the same sequence depth (20×) (Appendix A, Figure A2) [49]; however, the authors only used three single E. coli cells for testing. To assess the uniformity of read coverage across the genome, reads were averaged into 10 kilo-base (kb) bins and their read depths plotted to visualize coverage depth for each reaction volume (Figure 5A). Especially in the larger volumes, more genome regions are not covered by any reads in comparison to MDA performed in 1.25 µL volumes. Furthermore, coverage depths were more uniform across the genome in 1.25 µL, as evidenced by Lorenz curves [68] showing a more equal distribution of reads covering all bases of the genome (Figure 5B). We further verified this by calculating the Gini index of each sample, which is a measure of deviation from uniformity ranging from 0 (perfectly uniform distribution) to 1 (extremely uneven distribution) [69]. The Gini index differs significantly between different reaction volumes (p = 0.0176, Appendix A, Table A4), and is lowest for 1.25 µL reactions (~0.71 ± 0.07, Appendix A, Table A4). These levels of uniformity are similar to those obtained from E. coli in 150 nL microfluidic MDA reaction volumes [38] and hydrogels [53]. Next, we assembled and compared SAGs for all replicates. Prior to assembly, read depths were normalized due to the large differences introduced via MDA, setting a target depth of 100×. However, MDA reaction volumes less than and greater than 1.25 µL resulted in lower final sequence depths due to the fact that more reads were lost during the read pre-processing steps (Figure 6A). Therefore, the resulting assemblies were of lower quality compared with assemblies from 1.25 µL MDA reaction volumes (Figure 6B,C). Specifically, 1.25 µL reactions had the longest average total length and N50 at 3,522,851 bp and 46,179 bp, respectively (Figure 6B,C). N50 constitutes the sequence length of the shortest contig representing 50% of the assembly’s total sequence length and indicates that assemblies from 1.25 µL reaction volumes were more contiguous, resulting in higher quality assemblies than the other MDAs. Next, assembly coverage and completeness were calculated. The difference between these two measurements is that coverage is calculated as the percentage of the assembly (contigs) mapped to the reference genome [70], whereas genome completeness was estimated by MDMcleaner as the presence of marker genes such as small subunit (SSU) rRNA genes, large subunit (LSU) rRNA genes, universal bacterial/archaeal protein coding marker genes, total coding sequences (CDS), and tRNA-genes [12]. In general, the assembly coverage (p = 0.0199) and completeness (p = 0.0128) both significantly differed between the different-sized reactions (Appendix A, Table A5). Not surprisingly, coverage and completeness were highest for assemblies from 1.25 µL MDA reactions and were on average ~75 ± 14% and 94 ± 0.04%, respectively, while contamination was lowest (Figure 6D–F). Three out of five 1.25 µL MDA reaction replicates even achieved over 75% coverage, with the highest being 89.5% (Appendix A, Table A5). Comparatively, WGA-X™ reported E. coli assembly coverages of <60%, even with ~5× more reads [34]. Whereas at 10 µL, our assembly coverages were found to be within the range of those reported from WGA-X™ in 10 µL reactions, highlighting how WGA-X™ could also benefit from further volume reduction. In comparison to other volume reduction approaches, our higher assembly coverages were within range of previously reported E. coli MDA coverages in pL droplets (88–91%) [47] and nL wells (88–94%) [49] at similar sequence depths. Overall, these results demonstrate that MDA performed in 1.25 µL reaction volumes greatly improves this method by producing significantly less-biased, less-contaminated, and more complete SAGs than standard, larger reaction volumes. To assess the benefit of further volume reduction, we also tested the 0.5 µL MDA reaction volume in a droplet microarray (DMA) (Aquarray, Germany) since this reaction size did not work in 384-well plates (Figure 3). The DMA is a platform consisting of a glass slide with super-hydrophobic and hydrophilic patterning to create spots in which nanoliter-sized reactions can take place [71,72]. To prevent evaporation during six hours of MDA, the DMA was placed in a humidity chamber [73] and 5% glycerol was added to the MDA master mix. However, these tests were not successful. Recently, the DMA was used to synthesize cDNA from single HeLa cells [73]; however, the cDNA only spent approximately one hour on the DMA versus six hours needed for MDA, and amplification was performed off-chip. Therefore, we attribute our failed MDAs on the DMA to evaporation and/or sterical hinderance of the polymerase [63,64]. Still, further volume reduction could possibly increase genome coverage by ~12–14% [47,49], but the reproducibility of these picoliter and nanoliter approaches is uncertain since few approaches and their results have been validated outside the original study. This is because microfluidic, droplet, and other volume reduction approaches are not as easily accessible or easy to use in other groups, and many are not high-throughput. Additionally, because DNA yield is limited in smaller volumes, some studies have had to perform two rounds of MDA to generate sufficient amounts of products for library preparation [40,53]. However, library preparation input requirements have decreased from ug to pg in the last few years [74], so lower DNA yield is no longer much of an issue. Escherichia coli K12 MG1655 (DSMZ 18039) was cultured in 1 mL of Luria Bertani (LB) broth at 30 °C and 750 rpm with the Thermomixer Comfort (Eppendorf, Hamburg, Germany) to the exponential growth phase (~4 h; OD600 of ~2.2–2.6). From this point forward, cells were processed in a UV-decontaminated ISO 4 cleanroom. Equipment and gloves were decontaminated with DNA AWAY (Thermo Fisher Scientific, Waltham, MA, USA). Consumables were UV treated for 1 hr in a Crosslinker and 1 × PBS was UV treated for 6 h in a 254 nm shortwave ultraviolet crosslinker at 0.12 Joules/I2 (Analytik Jena GmbH, Jena, Germany). A BD FACSMelody (Becton-Dickson, Franklin Lakes, NJ, USA), fitted with a 100 µM nozzle and equipped with a 488 nm laser for excitation was used to sort single cells. Cells were first diluted to approximately 106 cells mL−1 with sterile 1X PBS to ensure an event rate of <1000 events/s. Gates were defined on side-scatter (cell complexity) and forward-scatter (cell-size). Cells were sorted in single-cell mode into 384-well plates (Bio-Rad, Hercules, CA, USA) containing no sorting buffer (i.e., dry sorting). Plates were sealed with Microseal B (Bio-Rad, Hercules, CA, USA) and stored at −80 °C. Plates containing sorted cells were thawed and centrifuged at 4 °C for 5 min at 3000 rpm (Eppendorf, Germany). Preliminary results found that REPLI-g Single Cell Kit (QIAGEN, Hilden, Germany) lysis buffer was too destructive for E. coli single cells; therefore, a modified lysis buffer from Stepanauskas et al. (2017) was used [34]. Cell lysis buffer (0.2 M KOH, 5 mM EDTA and 50 mM DTT) and neutralization buffer (1 M Tris-HCl, pH 4) were treated with UV for 10 min on an ice-water bath in a 254 nm shortwave ultraviolet crosslinker at 0.12 Joules/cm2 (Analytik Jena GmbH, Jena, Germany) [42]. The lysis solution was then dispensed onto the cells and into wells containing no cells (negative controls) with an I.DOT mini (Dispendix, Stuttgart, Germany) non-contact liquid dispenser. The plate was incubated at 21 °C for 10 min and neutralized by the addition of an equal volume of neutralization buffer (1 M Tris-HCL, pH 4). The amount of lysis and neutralization buffer per MDA reaction can be found in Appendix A, Table A1. Multiple displacement amplification (MDA) was performed with the REPLI-g Single Cell Kit (QIAGEN, Hilden, Germany). REPLI-g sc Reaction Buffer and Polymerase were combined in 0.2 mL DNase, RNase-free PCR tubes (Biozym Scientific GmbH, Hessisch Oldendorf, Germany) and UV treated for 30 min on an ice-water bath in a 254 nm shortwave ultraviolet crosslinker at 0.12 Joules/cm2 (Analytik Jena GmbH, Jena, Germany) [42]. Syto-13 (Invitrogen, Waltham, NJ, USA) was added to the master mix at a final concentration of 1 µM to monitor exponential DNA amplification. The REPLI-g master mix was then dispensed onto the lysed cells and negative controls with an I.DOT mini (Dispendix, Stuttgart, Germany) non-contact liquid dispenser so that the final MDA volumes were 0.5, 0.8, 1.0, 1.25, 5, and 10 µL. The MDA’s were incubated for 6 h at 30 °C in a CFX-384 thermocycler (Bio-Rad, Hercules, CA, USA), then 65 °C for 10 min to stop the amplification and held at 4 °C. Amplified DNA was kept at −20 °C until used for library preparation. The following steps were performed under a UV-decontaminated Laminar Flow PCR workbench (STARLAB International GmbH, Hamburg, Germany), sterilized with DNA AWAY (Thermo Fisher Scientific, Waltham, MA, USA). Prior to library preparation, the amplified DNA was cleaned with DNA Clean & Concentrator—5 (Zymo Research, Irvine, CA, USA). DNA input for library preparation was normalized to 5.98 ng µL−1. Libraries were prepared using the NEBNext® Ultra™ II FS DNA Library Prep Kit for Illumina (New England Biolabs (NEB), Ipswich, MA, USA), following the <100 ng input protocol. Fragmentation was set to 14 min and 7 PCR cycles were used. NEBNext® Multiplex Oligos for Illumina® were used for barcoding. Library concentration and size was quantified with Qubit™ DNA HS assay (Life Technologies, Carlsbad, CA, USA) and a Bioanalyzer High Sensitivity DNA kit (Agilent, Santa Clara, CA, USA). The libraries were sequenced using an Illumina NextSeq 550 with the High Output Kit v2.5 300 Cycles (2 × 150 bp paired-end) (Illumina, San Diego, CA, USA). The sequence reads were quality checked using FastQC v0.11.9 (www.bioinformatics.babraham.ac.uk/projects/fastqc (accessed 1 February 2020)) and quality-trimmed using Trim Galore v0.6.6 [75]. Following trimming, reads were normalized to 3,108,153 read pairs (~200× sequence depth) with BBTools v38.87 reformat.sh [76]. Additionally, reads were also down-sampled to 100×, 80×, 60×, 40×, and 20× with reformat.sh to determine the effects of sequence depth on coverage in Appendix A, Figure A2. Normalized reads (200× depth) were assessed for contamination using FASTQ-Screen v0.15.2 [77] against its standard databases for Homo sapiens, Saccharomyces cerevisiae, Escherichia coli, rRNA, phiX, vectors, adapters, as well as Ralstonia picketti contaminations. E. coli multi-mapping reads were kept. PCR duplicates were counted and removed with dedupe.sh from BBTools [76]. Then, reads were mapped to E. coli MG1655 (ASM584v2) with bbmap.sh. Max indel length was set to 80, as recommended for MDA, then coverage was calculated for 10 kb bins [76]. Prior to de novo assembly, the read coverage was normalized with bbnorm.sh setting target = 100 and min = 5 [76]. SPAdes v.3.15.5 was used as recommended for single cells by using the flag –sc for single-cell mode, kmer lengths of 21 to 101 in 10 -step increments, and setting the flag—careful to reduce the number of mismatches [78]. QUAST v.5.2.0 was used to assess assembly quality [70] and MDMcleaner v0.8.3 [12] was used to estimate SAG contamination and completeness. Statistical differences between sample quality, mapping, and assembly statistics were calculated using Anova: Single Factor with an alpha value of 0.05 in Microsoft Excel®. For data not normally distributed, as determined by Shapiro–Wilk testing, the non-parametric Kruskal–Wallis one-way ANOVA was used with an alpha value of 0.05. Both the Shapiro–Wilk and Kruskal–Wallis tests were calculated using the Real Statistics Resource Pack software (Release 7.6), Copyright (2013–2021) Charles Zaiontz (www.real-statistics.com (accessed on 1 January 2023)). Pairwise comparisons for measurements with statistical significance were determined using the t-Test: two-sample assuming equal variances with an alpha value of 0.05 in Microsoft Excel®. Gini indexes were calculated with the ineq package [68] in R v.3.6.3 [79]. Read depth and Lorenz curve plots were created using ggplot2 [80]. Based on our results, we question whether further volume reduction is really necessary. As reviewed in Figure 2, many of the current nL and pL volume reduction approaches are either too low-throughput, require complex fabrication, and/or are too expensive to make or purchase (>100 USD per device). Therefore, one should gauge for themselves whether the time and cost benefits of volume reduction down to nL and pL reactions make sense in the scope of their study. Meanwhile, volume reduction in standard 384-well plates and with commercially available cell sorters and liquid dispensers makes this approach more easily accessible to other researchers and already drastically reduces the costs by ~97.5% from the standard 50 µL MDA reaction (Table 1). We also found that with our approach, 40× sequence depth is enough for high-quality assemblies (Appendix A, Figure A2), compared to the standard >100× depths generally used in microbial SCG [34,42]. Further cost reduction could also be achieved by applying this approach to the less expensive WGA-X™ method (Table 1), seeing that preliminary work in our group finds WGA-X™ to work in 1.25 µL reaction volumes as well. In the end, we anticipate that the improvements made herein will be of great interest for other single-cell studies and will therefore increase the use of SCG, especially for research focused on elucidating the genomic potential of rare taxa and/or novel microbial dark matter in environmental samples.

PMC10002426

Clara Vianello,Marco Salluzzo,Daniela Anni,Diana Boriero,Mario Buffelli,Lucia Carboni

Increased Expression of Autophagy-Related Genes in Alzheimer’s Disease—Type 2 Diabetes Mellitus Comorbidity Models in Cells

03-03-2023

Alzheimer’s disease,type 2 diabetes mellitus,autophagy,ATG16L1,ATG16L2,GABARAP,GABARAPL1,GABARAPL2,SQSTM1,neuronal cultures

The association between Alzheimer’s disease (AD) and type 2 diabetes mellitus (T2DM) has been extensively demonstrated, but despite this, the pathophysiological mechanisms underlying it are still unknown. In previous work, we discovered a central role for the autophagy pathway in the common alterations observed between AD and T2DM. In this study, we further investigate the role of genes belonging to this pathway, measuring their mRNA expression and protein levels in 3xTg-AD transgenic mice, an animal model of AD. Moreover, primary mouse cortical neurons derived from this model and the human H4Swe cell line were used as cellular models of insulin resistance in AD brains. Hippocampal mRNA expression showed significantly different levels for Atg16L1, Atg16L2, GabarapL1, GabarapL2, and Sqstm1 genes at different ages of 3xTg-AD mice. Significantly elevated expression of Atg16L1, Atg16L2, and GabarapL1 was also observed in H4Swe cell cultures, in the presence of insulin resistance. Gene expression analysis confirmed that Atg16L1 was significantly increased in cultures from transgenic mice when insulin resistance was induced. Taken together, these results emphasise the association of the autophagy pathway in AD-T2DM co-morbidity, providing new evidence about the pathophysiology of both diseases and their mutual interaction.

Increased Expression of Autophagy-Related Genes in Alzheimer’s Disease—Type 2 Diabetes Mellitus Comorbidity Models in Cells The association between Alzheimer’s disease (AD) and type 2 diabetes mellitus (T2DM) has been extensively demonstrated, but despite this, the pathophysiological mechanisms underlying it are still unknown. In previous work, we discovered a central role for the autophagy pathway in the common alterations observed between AD and T2DM. In this study, we further investigate the role of genes belonging to this pathway, measuring their mRNA expression and protein levels in 3xTg-AD transgenic mice, an animal model of AD. Moreover, primary mouse cortical neurons derived from this model and the human H4Swe cell line were used as cellular models of insulin resistance in AD brains. Hippocampal mRNA expression showed significantly different levels for Atg16L1, Atg16L2, GabarapL1, GabarapL2, and Sqstm1 genes at different ages of 3xTg-AD mice. Significantly elevated expression of Atg16L1, Atg16L2, and GabarapL1 was also observed in H4Swe cell cultures, in the presence of insulin resistance. Gene expression analysis confirmed that Atg16L1 was significantly increased in cultures from transgenic mice when insulin resistance was induced. Taken together, these results emphasise the association of the autophagy pathway in AD-T2DM co-morbidity, providing new evidence about the pathophysiology of both diseases and their mutual interaction. Alzheimer’s disease (AD) is the most common cause for dementia, and 55 million people are estimated to live with this condition worldwide. Numbers are expected to increase to 113 million by 2050 [1], causing enormous impacts on global health and imposing a huge economic burden. Therapeutic approaches encompass cholinesterase inhibitors and memantine as symptomatic agents [2,3]. Great hopes have been raised by antibodies which target amyloid-beta aggregates in the brain as potential disease-modifying interventions [4,5]. However, whether meaningful clinical efficacy can be reached as well as cost-effectiveness are still questions, while safety concerns need further analyses and clarification [6,7,8]. Therefore, preventive actions directed at potentially modifiable risk factors are crucial to reduce AD severe disease burden [2]. Both genetic and environmental factors contribute to AD risk. Dominantly inherited mutations in APP, PSEN1, and PSEN2 genes account for rarer early-onset cases, whereas carrying at least one copy of the APOE ε4 allele is the strongest genetic risk factor for the common late-onset form [9]. Although age is the most relevant factor providing the largest impact, additional environmental components present important contributions that are potentially modifiable [2]. Among the latter, consistent evidence supports major roles for education, hypertension, obesity, hearing loss, traumatic brain injury, smoking, depression, physical inactivity, social isolation, type 2 diabetes mellitus (T2DM), and air pollution as potential targets of intervention in different life stages, particularly at midlife [2]. In addition, T2DM has been compellingly associated with significantly greater risk of dementia [10,11,12]. Moreover, metabolic syndrome and obesity, which are often associated with T2DM, represent dementia risk factors per se, thus further complicating the picture [13,14]. It has been suggested that since T2DM is modifiable, its reduction could constitute a possible strategy for reducing future AD incidence. Indeed, it has been estimated that if T2DM was removed as a risk factor, about 1.1% of dementia cases could be prevented. Although this percentage is low, the number of impacted people is nonetheless high when considering global incidence rates [12]. Despite the demonstrated convincing association between AD and T2D, the pathophysiological mechanisms responsible are still unknown. As a result, the best approach to be adopted for prevention still needs to be elucidated [12]. Furthermore, whether antidiabetic treatments represent a useful way forward is uncertain at present, as available data are inconsistent [2,15,16]. Several hypotheses have been proposed to explain the mechanistic link between AD and T2DM [17,18,19]. Among them, insulin signalling is impaired in both AD and T2DM, and the definition of AD as type 3 diabetes is based on the observed insulin resistance [20,21,22,23]. In addition, defects in mitochondrial function are shared by both AD and T2DM, thus a common causative role has been proposed for this defect based on preclinical and clinical findings [24,25]. In a previous study, we adopted a system biology approach to address this important gap in knowledge about the common pathophysiological dysregulations contributing to AD and T2DM comorbidity. We compared molecular mechanistic networks underlying brain T2DM pathophysiology in AD and control subjects by analysing transcriptional datasets with a novel approach. We discovered a central role for the autophagy pathway in the mechanisms shared between AD and T2DM [26]. Autophagy is an intracellular degradation pathway that traffics organelles, dysfunctional proteins, and infectious agents to lysosomes via specific vesicles called autophagosomes [27]. In agreement with our findings, autophagy relevance in AD is supported by a wealth of data, and targeting this mechanism is proposed as a potential avenue for drug discovery [28,29,30]. Moreover, abnormal autophagic responses have been implicated in metabolic disorders [31]. The aim of this study was to further investigate the role of genes identified in our previous studies as relevant for the pathophysiology of Alzheimer’s disease and T2DM comorbidity, namely ATG16L1, ATG16L2, GABARAP, GABARAPL1, GABARAPL2, and SQSTM1. We thus investigated the modulation of these genes in an animal model of AD and in cellular models of insulin resistance in Alzheimer’s disease brains. In immunofluorescence experiments, the following antibodies and dilutions were used: anti-Phospho SQSTM1/p62 (S349) (Abcam, Cambridge, UK, cat # ab211324) 1:100; anti-SQSTM1/p62 (Abcam, cat # ab56416) 1:50; anti-β-Tubulin III (Merck Millipore, Burlington, MA, USA cat # T2200) 1:500; anti-MAP2 (1:500, Merck Millipore, cat # M9942) 1:500; anti-GFAP (Thermo Fisher Scientific, Waltham, MA, USA, cat # 13-0300), 1:800, donkey anti-rabbit-IgG Alexa Fluor 488 (Thermo Fisher Scientific, cat # R37118) 1:1000; donkey anti-mouse-IgG Alexa Fluor 594 (Thermo Fisher Scientific, cat # A-21203, 1:1000); goat anti-mouse IgG1 CF 568 (Merck, cat # SAB4600313 1:1000); goat anti-rat Alexa Fluor 647 (Thermo Fisher Scientific, cat # A21247, 1:1000); and DAPI (4′,6-diamidino-2-phenylindole Merck Millipore, cat #D9542) 1:5000. In Western blotting experiments, the following antibodies and dilutions were used: anti-Phospho SQSTM1/p62 (S349) (Abcam cat # ab211324) 1:2000; anti-SQSTM1/p62 (Abcam cat # ab56416) 1:2000; Anti-GAPDH (Abcam cat # ab8245); 1:5000; anti-phospho-Akt (Ser473 D9E Cell Signaling, Danvers, MA, USA, cat #4060) 1:2000; anti-Akt (Cell Signaling, cat # 9272, 1:1000); goat anti-mouse IgG IRDye 800(Li-Cor, Lincoln, NE, USA, cat # 926-32210) 1:5000; and goat anti-rabbit-IgG Alexa 680 (Thermo Fisher Scientific cat # A21076) 1:5000. A colony of triple-transgenic AD mice (3xTg-AD) expressing three mutant human transgenes—PS1M146V, APPSwe, and tauP301L—was established at the University of Verona by purchasing transgenic mice from The Jackson Laboratory (Sacramento, CA, USA). C57BL/6J mice were purchased from Charles River Italia (Calco, Italia). Mice were housed at 3/cage at a constant room temperature of 21 ± 1 °C and maintained on a 12:12h light/dark cycle with lights on at 7.30 a.m. with freely available food and water. All efforts to minimise animal suffering and number were made. This study is compliant with ARRIVE guidelines [32]. Procedures involving animals were conducted in conformity with the EU guidelines (2010/63/UE) and Italian law (decree 26/14) and were approved by the University of Verona’s ethical committee and the local authority’s veterinary service. The Italian Health Ministry Ethical Committee for Protection of Animals approved the study (approval number: 283/2019-PR). For gene expression studies, 18 (six/group aged 6, 12, and 18 months) female transgenic mice and 18 (six/group) female wild-type mice were used. For immunofluorescence on brain sections, 12-month-old female 3xTg-AD and wild-type mice were used (n = 3/group). Mice for gene or protein expression experiments were anesthetised using Tribromoethanol (Merck Millipore) and sacrificed. Brain dissections were performed in Petri dishes on ice; the hippocampi were collected, flash-frozen in liquid nitrogen, and stored at −80 °C until analysis. The whole procedure did not exceed 5 min to preserve brain integrity. Mice for immunofluorescence experiments were anesthetised using Tribromoethanol, perfused transcardially with 0.1 M phosphate buffered saline solution (PBS), followed by formaldehyde 10% V/V, buffered 4% w/v (Titolchimica, Rovigo, Italy), and brains were extracted and postfixed overnight. Seven dams were used (wild-type: n = 4; 3xTg-AD n = 3) and neuronal cultures were prepared from 5–6 pups/preparation for each genotype. Human glioblastoma H4 cell lines stably expressing the βAPP-Swe mutation (K595N/M596L) were a kind gift from prof. M. Pizzi, University of Brescia, Italy. Cells were cultured in DMEM with 10% foetal bovine serum (FBS), 100 Units/mL penicillin, 2 mM glutamine, and 100 μg/mL streptomycin (Thermo Fisher Scientific) [33]. After reaching 80% confluence and twenty-four hours before starting the experiment, cells were trypsinised and seeded at a density of 4 × 106 cells in T75 cm2 flasks. Treatments were performed in DMEM medium without FBS. Primary mouse cortical cultures were prepared as previously described [34] with modifications [35]. Briefly, newborn C57BL/6 and 3xTg-AD mice (P0-P1) brains were isolated and cortices were dissected in 1X ice-cold DBPS medium (cat # 14200075, Thermo Fisher Scientific). After removal of meninges, cortices were washed twice and enzymatically digested with DPBS solution containing 0.25% (v/v) trypsin (Thermo Fisher Scientific), 1 mM sodium pyruvate, 0.1% (w/v) glucose, and 10 mM HEPES pH 7.3 for 20 min at 37 °C. Following a 5 min incubation with 0.1mg/mL DNAse I (Merck Millipore) at room temperature, the enzymatic reaction was stopped with an MEM solution containing 10% FBS, 0.45% (w/v) glucose, 1 mM sodium pyruvate, 2 mM L-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin (all reagents from Thermo Fisher Scientific). Next, the tissue was triturated through a P1000 pipette, and the cell suspension was passed through a 70 µm MACS SmartStrainer (Miltenyi Biotec, Bergisch Gladbach, Germany). Cells were then counted and diluted to 8 × 105 cells/mL in Neurobasal™-A Medium (NBA, Thermo Fisher Scientific) containing 1X B27 supplement (Thermo Fisher Scientific), 2 mM L-Glutamine (Thermo Fisher Scientific), 100 U/mL penicillin, and 100 μg/mL streptomycin (Thermo Fisher Scientific) and plated on 6-well plates pre-coated with 0.1 mg/mL poly-L lysine (Merck Millipore). Cells were maintained in a standard, humidified 5% CO2 incubator until the day of the experiment (5–7 days in vitro, DIV). To monitor insulin response, cells were challenged with 100 nM insulin (Merck Millipore) for 30 min. To induce insulin resistance, cells were treated for 24 h with 40 mM glucose (Merck) or 20 nM insulin before receiving insulin challenge [36]. Controls were treated with vehicle. At the end of the experiments, both H4Swe cells and primary mouse neurons were washed with PBS and harvested by 5 min centrifugation at 2900× g, and the pellets were re-suspended in RNA later (Thermo Fisher Scientific), stored at 4 °C for 24 h, and transferred at −20 °C until RNA extraction. Treatments were repeated in 3–6 independent experiments. Gene expression was assessed by qPCR as previously reported [37] with slight changes. RNA was extracted with the Aurum total RNA mini kit (Bio-Rad, Hercules, CA, USA) which includes a DNase I digestion step, following manufacturer’s instructions. RNA amount was assessed by UV absorbance in a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific). cDNA was synthesised using the iScript Advanced cDNA synthesis Kit (Bio-Rad). qPCR was performed by Sybr Green technology in a 7900HT Fast Real-Time PCR System (Thermo Fisher Scientific) with SSO Advanced Universal SYBR Green Supermix (Bio-Rad) in 20 µL according to this protocol: stage 1: 95 °C, 20 s; stage 2: 40 × (95 °C, 3 s; 60 °C, 30 s). Primers were selected with the NCBI Primer-BLAST tool and purchased from Eurofins Italia (Torino, Italy). Sequences are reported in Table 1. Data were analysed using the Delta-Delta-Ct method, converting to a relative ratio (2−DDCt) for statistical analysis [38] by normalising to the geometric average of two endogenous reference genes: Gapdh and Ywhaz, as previously reported [39,40]. The specificity of amplification products was evaluated by building a dissociation curve in the 60–95 °C range. Hippocampi were homogenised with a micro-pestle in ice-cold lysis buffer (10% w/v) containing 50 mM Tris-HCl (pH 7.5), 2% Igepal, 10 mM MgCl2, 0.5 M NaCl, 2 mM EDTA, 2 mM EGTA, 5 mM benzamidine, 0.5 mM phenyl-methylsulfonyl fluoride, 8 mg/mL pepstatin A, 20 mg/mL leupeptin, 50 mM β-glycerolphosphate, 100 mM sodium fluoride, 1 mM sodium vanadate, 20 mM sodium pyrophosphate, and 100 nM okadaic acid. Homogenates were clarified by 1 min centrifugation at 10,000× g at 4 °C and protein concentration was assessed by Precision Red Protein Quantification Assay (Cytoskeleton). H4Swe cells were seeded into 6-well plates at a density of 9.5 × 105 cells/well. Following treatments, cells were washed once in Tris-buffered saline (TBS), lysed, and assayed for protein with the Bradford method (Merck). In both instances, lysates were processed for Western blot as previously reported [41], with slight changes. Briefly, lysates were separated using 4–12% Bis-Tris gels (Novex pre-cast gel, Thermo Fisher Scientific) and transferred to 0.45 μm nitrocellulose membranes (Thermo Fisher Scientific). Blots were blocked for 1 h at room temperature in 1X Odyssey blocking buffer (TBS) and incubated with primary antibodies overnight in Odyssey blocking buffer (TBS) plus 0.1% Tween-20 (Tween-20 TBS) at 4 °C. Membranes were washed 3 × 10 min in Tween-20 TBS at room temperature, followed by incubation with secondary antibody conjugated to IRDye diluted in Tween-20 TBS for 1 h at room temperature. Blots were washed 2 × 10 min in TBST, 1 × 10 min in TBS, and visualised with Odyssey Infrared Imaging System (Li-Cor) by quantifying fluorescent signals as Integrated Intensities (I.I. K Counts) using the Odyssey Infrared Imaging System. After background subtraction, protein levels were assessed as total protein to Gapdh loading control ratios or as phosphorylated to total protein ratios. In brain sections, immunofluorescence was carried out as previously reported [42]. Briefly, after post-fixing, brains were embedded in an OCT cryoembedding matrix and sectioned on the coronal plane at 30 mm thickness with a cryostat. Sections were treated with a blocking solution of 2% bovine serum albumin, 2% normal goat serum, and 0.2% Triton X100 in PBS for 20 min at room temperature and incubated overnight at 4 °C in primary antibodies. Secondary antibodies were diluted 1:1000 in the above blocking solution, with the appropriate serum. After immunohistochemical processing, sections were counterstained with the fluorescent nuclear marker DAPI (100 ng/mL) for 10 min at room temperature and mounted on slides with 0.1% paraphenylenediamine in glycerol-based medium (90% glycerol 10% PBS). For H4Swe cell immunostaining, 5 × 105 cells/well were seeded onto 18 mm round coverslips in 24-well plates and left to attach overnight. The next day, cells were washed twice with PBS and fixed with 4% paraformaldehyde for 20 min. Fixed cells were treated for 10 min with blocking and incubated overnight with primary antibodies in blocking solution. After three washes with PBS, samples were incubated with secondary antibodies diluted 1:2000 in blocking solution for 1 h. After final washes, coverslips were treated with DAPI solution. Coverslips were fixed onto glass slides with a drop of anti-fading mounting medium and sealed with nail polish. Primary cortical cells were fixed in 10% (v/v) formalin solution (Titolchimica) for 15 min at room temperature, washed three times in PBS, and blocked in PBS containing 10% (v:v) normal goat serum (Thermo Fisher Scientific) and permeabilised with 0.3% (v:v) TritonX-100 (Merck Millipore) in PBS for 40 min. Next, cells were incubated with mouse anti-Map2 and rat anti-Gfap primary antibodies overnight at 4 °C, and after three PBS washing steps, with anti-secondary antibodies, anti-mouse IgG1 CF 568, and anti-rat Alexa Fluor 647 for 1 h at room temperature. Antibodies were diluted in PBS containing 5% (v:v) normal goat serum. Nuclei were counterstained with DAPI 1:5000 and coverslips were mounted on slides using DAKO fluorescence mounting media (Agilent, Santa Clara, CA, USA). Images at different Z-planes were collected on a Leica tcs-sp5 confocal microscope. Images were processed with the software Imaris (Bitplane AG, Belfast, UK) or ImageJ. The data are presented as the observed mean values ± SEM. The data were analysed using a 1-way ANOVA with treatment (control, insulin, glucose + insulin, insulin + insulin) as the treatment factor or 2-way ANOVA with genotype (wild-type, 3xTg-AD) and age (6, 12, and 18 months) or treatment (control, insulin, glucose + insulin, insulin + insulin). When the samples were analysed in different plates using a complete block design, an additional blocking factor plate was also included in the statistical model in order to account for any plate-to-plate variability [43]. The analyses were followed by planned comparisons of the predicted means. The analysis was performed using the InVivoStat v4.4.0 software [44]. The data were log-transformed, where appropriate, in order to stabilise the variance and satisfy the parametric assumptions. A value of p < 0.05 was considered statistically significant. Since comorbidity is frequently observed between AD and T2DM, in our previous study [26], we applied a systems biology approach to investigate if common pathophysiological alterations could be identified at a molecular level. Similar approaches had previously highlighted the role of shared cellular signalling pathways contributing to both T2DM and AD. Among them, a prominent role was discovered for neurotrophin, PI3K/AKT, MTOR, and MAPK signalling, as well as for microglial-mediated immune responses, which can cross-talk to each other [45]. In addition, our previous data revealed a central role for autophagic mechanisms; in particular, a number of autophagy-related genes were indicated as important players, namely ATG16L1, ATG16L2, GABARAP, GABARAPL1, GABARAPL2, and SQSTM1. Therefore, we first aimed to investigate whether these genes were specifically modulated in association with neurobiological alterations characterising AD. We thus analysed the expression of the respective mouse orthologues (Atg16l1, Atg16l2, Gabarap, GabarapL1, GabarapL2, Sqstm1) in a transgenic mouse model of AD. 3xTg-AD mice harbour three mutant genes for the beta-amyloid precursor protein (βAPPSwe), presenilin-1 (PS1M146V), and tauP301L [46,47]; as a consequence, the mice progressively develop plaques and tangles, as well as cognitive impairments [47,48,49]. We thus compared hippocampal gene expression between 3xTg-AD mice and the respective wild-type controls at different ages. At 6 months, Atg16L1, Atg16L2, and GabarapL1 were expressed at significantly higher levels in 3xTg-AD mice (Figure 1A,B,D). In contrast, at 12 months, GabarapL2 expression was significantly reduced, whereas Sqstm1 levels were elevated (Figure 1E,F). At the protein level, although increased Sqstm1 mRNA expression was observed qualitatively in the hippocampus (Figure 2A), the increase could not be confirmed in semi-quantitative Western blotting experiments, possibly because of the lower sensitivity of the technique (Figure 2B,C). Next, we investigated whether these genes were modulated in the presence of AD-T2DM comorbidity. To model this condition, we first employed the human glioblastoma H4 cell line stably expressing the βAPP-Swe mutation [50,51,52] and applied treatments able to induce insulin resistance [53,54]. In this model, phospho-Akt/Akt levels were significantly increased by the insulin challenge (100 nM), whereas this response was abated after chronic treatment with high-concentration insulin, thus showing that insulin resistance was successfully achieved (Figure 3). Similar to findings obtained in 3xTg-AD mice, in the presence of insulin resistance, Atg16L1, Atg16L2, and GabarapL1 expression levels were significantly increased (Figure 4A,B,D). Subsequently, we examined whether Sqstm1 phosphorylation levels were affected by the onset of insulin resistance. No significant differences were revealed by Western blot or immunofluorescence analyses (Figure 5). Next, we generated a second AD-T2DM cellular model by inducing insulin resistance in neuronal primary cultures obtained from 3xTg-AD mice and wild-type controls. In this model, we confirmed that primary cultures were enriched in neurons (Figure 6). Gene expression analysis confirmed that Atg16L1 was significantly increased in cultures from transgenic mice when insulin resistance was induced (Table 2), whereas no other difference was detected in the other genes analysed. In addition, Gabarap showed a significant reduction by genotype (Table 2). However, the findings showed a very high level of variability within groups. In this study, we examined the modulation of genes recognised as relevant for the common cellular dysregulations sustaining the observed comorbidity between AD and T2 DM in our previous systems biology study [26]. Here, we explored their expression in 3xTg-AD mice, a transgenic mouse model of AD overexpressing mutated human genes associated with early-onset AD (PSEN1 and APP) or with the formation of neurofibrillary tangles (tau) [46]. In this mouse model, the neuropathological features of AD, amyloid plaques and neurofibrillary tangles, as well as neuroinflammation, developed progressively in an age-dependent fashion. In particular, extracellular amyloid beta deposition started at six months of age and progressively increased to reach its full extent at 15 months [47,49]. Tau pathology followed a similar age-related increase, although delayed with respect to amyloid beta pathology [46,47,49]. Likewise, cognitive impairments reproducing the human pathological feature of AD appeared at six months and became progressively more severe at 12 and 20 months [49]. We discovered that at 6 months of age, Atg16L1, Atg16L2, and GabarapL1 were expressed at higher levels in 3xTg-AD mice, whereas at later time points, this increase subsided. The alterations are in agreement with those obtained in the previous study, where pre-frontal cortex samples were analysed in two AD mice models [26]. These findings suggest that the increased expression may occur as an attempt to oppose the neuropathological alterations by activating a neuroprotective response. A limitation of this experimental design is that 3xTg-AD mice were generated in a hybrid C57BL/6:129 genetic background; therefore, the control line we used, although similar, is not identical. However, the use of C57BL/6 as a control strain is well documented in previous studies [55,56,57,58]. To reproduce the molecular dysregulation characterising insulin resistance in AD brains, we used neuronal models of AD based either on a neuronal cell line generating amyloid beta deposits, H4Swe cells, or on the 3xTg-AD mouse primary neuron cultures. H4Swe cells are well established as tools to investigate AD-related cellular dysregulation [50,51,52]. However, a limitation is that they do not share all neuronal characteristics, being a neuroglioma-derived line. Therefore, primary neurons were also investigated. In both in vitro models, we established a condition of insulin resistance by prolonged treatment with high insulin concentrations. As a consequence, the normal response to insulin challenge is hampered by prolonged insulin exposure, and the normal Akt phosphorylation and activation responses characterising the insulin signal transduction pathway are not induced [53]. Similar to findings in 3xTg-AD mice, we found that Atg16L1, Atg16L2, and GabarapL1 were significantly elevated in insulin resistance conditions. The increased expression of these genes in the cell model of AD-T2DM comorbidity corroborates the hypothesis of a neuroprotective role of this response, as hyperglycaemia has been previously associated with the increased beta amyloid plaque production [59]. Atg16L1 was identified as the mammalian orthologue of the corresponding yeast gene, which was known to provide a crucial contribution to autophagic processes [60,61]. Autophagy was discovered as a process occurring in response to cellular stresses such as nutrient deprivation, infection, or hypoxia. Its chief function is providing nutrients for vital cellular activities during fasting by degrading cellular components and releasing them back to the cytoplasm to be used again. However, in addition to this non-selective approach, further studies demonstrated that autophagy can selectively eliminate potentially harmful damaged mitochondria or protein aggregates [61,62]. Consequently, autophagy dysfunction has been implicated in several diseases and its components generated interest as potential pharmacological targets [28,62]. In autophagy, starvation signals promote the recruitment of autophagy proteins to a specific subcellular location, where they assemble a structure called the phagophore. An isolation membrane is gradually formed to isolate a portion of the cytosol and is finally sealed into a vesicle, termed the autophagosome, which contains cytoplasmic material. The autophagosome then fuses with the lysosomal membrane, and the autophagic body together with its cargo are degraded [62,63]. In this process, the role of Atg16L1 is essential for autophagy initiation, as its recruitment in the Atg12-Atg5 complex is required to engage autophagic proteins in the phagophore assembly site and contribute to its scaffolding by Atg8/LC3 protein lipidation [60,64,65,66]. Therefore, the increase observed in the present study suggests an effort to trigger autophagic responses to counteract the increased production of abnormal proteins and rescue insulin response. In addition to its well-demonstrated role in canonical autophagy, Atg16L1 was shown to exert different functions related to a structural component specifically observed in the C-terminal of the mammalian protein compared to the yeast counterpart. This specific component is necessary for the Atg16L1-mediated lipidation of single membranes, a non-canonical autophagy pathway, and specific cargo recruitment [66]. Furthermore, Atg16L1 contributes to modulating the extent of the innate immune response to injuries or infection, with an anti-inflammatory role [66,67]. Recent results showed that aged mice lacking this C-terminal domain of Atg16L1 develop beta amyloid plaques, excessive tau phosphorylation, reactive microgliosis, and memory impairments [68]. The proposed mechanism points to Atg16L1 involvement in a process defined as LANDO (LC3-associated endocytosis), which contributes to TREM2, CD36, and TLR4 recycling [68]. Therefore, the observed increased Atg16L1 levels may contribute to establishing a protective response that goes beyond the activation of autophagic responses, but also involves a rescue from neuronal damages through different mechanisms. Interestingly, we observed increased Atg16L1 expression in all investigated models. This result reinforces the notion of a primary role of this protein in the cellular response to both AD and T2DM pathophysiology, in a fashion independent from the in vivo or in vitro model which is well-conserved through evolution both in mice and in humans. Atg16L2 is a second isoform of Atg16L1, sharing a similar domain structure and a similar ability to bind Atg12-Atg5 and form a complex. However, the Atg16L2 protein is not recruited to phagophores and does not contribute to autophagosome formation; thus, it is not essential to canonical autophagy [69]. However, data suggesting the possibility of a cell-specific involvement in canonical autophagy are also available [70]. In addition, a recent report on the generation of Atg16L2 knock-out mice demonstrated a contribution of this gene to the maturation of immune cells and suggested that distinct functions are associated with respect to Atg16L1 [71]. Data showing its relevance in serious diseases such as Crohn’s disease and various cancers notwithstanding, very incomplete information is available on the role of Atg16L2 [72]. Our findings also support the involvement of this widely expressed gene in the pathophysiology of insulin resistance in AD brains. The GabarapL1 protein belongs to the Atg8/LC3 autophagy proteins, which include six members: LC3A, LC3B, LC3C, Gabarap, GabarapL1, and GabarapL2. The recruitment of Atg8 family proteins to the forming phagophore is mediated by the above-mentioned Atg12-Atg5–Atg16L1 complex and is essential for phagophore elongation and, ultimately, for autophagy [62,63,73]. GabarapL1 has also been implicated in autophagosome fusion with lysosome, and these functions are supposed to contribute to the degradation of oncogenic proteins and exert tumour-suppressive functions [73]. Interestingly, GabarapL1 has been specifically implicated in a newly discovered selective autophagy process termed glycophagy, which is involved in the transport and delivery of glycolytic fuel substrates [74]. Since these pathways regulate cellular energy demand, compelling evidence links glycophagy-mediated glucose availability to energy metabolism, in agreement with our findings. With regard to Sqstm1 levels, contrasting findings have been previously reported. In agreement with the present results, no alterations were detected in the hippocampus or in mitochondria-enriched hippocampal fractions of young 3xTg-AD mice [75,76]. Conversely, a decrease was found in whole brain homogenates and in the mitochondria-enriched hippocampal fractions of old 3xTg-AD mice [76,77,78]. This study investigated the molecular underpinning of the comorbidity between AD and T2DM in cellular models of insulin resistance in the presence of AD-related neuropathological features. Our findings are in agreement with the hypothesis that impaired autophagic mechanisms are important in the pathophysiology of AD through nonstandard mechanisms. In particular, the autophagy-related genes Atg16L1, Atg16L2, and GabarapL1 were highlighted as having a more relevant function in this mechanism, in addition to GabarapL2 and Sqstm1.

PMC10002448

Jessica Marinaccio,Emanuela Micheli,Ion Udroiu,Michela Di Nottia,Rosalba Carrozzo,Nicolò Baranzini,Annalisa Grimaldi,Stefano Leone,Sandra Moreno,Maurizio Muzzi,Antonella Sgura

TERT Extra-Telomeric Roles: Antioxidant Activity and Mitochondrial Protection

23-02-2023

telomerase catalytic subunit,mitochondrion,oxidative response,primary cell lines,electron microscopy

Telomerase reverse transcriptase (TERT) is the catalytic subunit of telomerase holoenzyme, which adds telomeric DNA repeats on chromosome ends to counteract telomere shortening. In addition, there is evidence of TERT non-canonical functions, among which is an antioxidant role. In order to better investigate this role, we tested the response to X-rays and H2O2 treatment in hTERT-overexpressing human fibroblasts (HF-TERT). We observed in HF-TERT a reduced induction of reactive oxygen species and an increased expression of the proteins involved in the antioxidant defense. Therefore, we also tested a possible role of TERT inside mitochondria. We confirmed TERT mitochondrial localization, which increases after oxidative stress (OS) induced by H2O2 treatment. We next evaluated some mitochondrial markers. The basal mitochondria quantity appeared reduced in HF-TERT compared to normal fibroblasts and an additional reduction was observed after OS; nevertheless, the mitochondrial membrane potential and morphology were better conserved in HF-TERT. Our results suggest a protective function of TERT against OS, also preserving mitochondrial functionality.

TERT Extra-Telomeric Roles: Antioxidant Activity and Mitochondrial Protection Telomerase reverse transcriptase (TERT) is the catalytic subunit of telomerase holoenzyme, which adds telomeric DNA repeats on chromosome ends to counteract telomere shortening. In addition, there is evidence of TERT non-canonical functions, among which is an antioxidant role. In order to better investigate this role, we tested the response to X-rays and H2O2 treatment in hTERT-overexpressing human fibroblasts (HF-TERT). We observed in HF-TERT a reduced induction of reactive oxygen species and an increased expression of the proteins involved in the antioxidant defense. Therefore, we also tested a possible role of TERT inside mitochondria. We confirmed TERT mitochondrial localization, which increases after oxidative stress (OS) induced by H2O2 treatment. We next evaluated some mitochondrial markers. The basal mitochondria quantity appeared reduced in HF-TERT compared to normal fibroblasts and an additional reduction was observed after OS; nevertheless, the mitochondrial membrane potential and morphology were better conserved in HF-TERT. Our results suggest a protective function of TERT against OS, also preserving mitochondrial functionality. Telomerase is a specialized reverse transcriptase, with the main function of adding telomeric DNA repeats on chromosome ends in order to counteract telomere shortening and maintain the stability of linear chromosomes [1]. Telomere shortening is caused by the so-called “end replication problem”, which is due to the inability of DNA polymerases to completely replicate the ends of a linear chromosome [2,3]. Furthermore, short telomeres can undergo different events such as degradation of the terminal regions of chromosomes or action of nuclease and other destructive factors [4]. In order to counteract telomere shortening, in humans, the main mechanism of telomere length maintenance involves the enzyme telomerase, a ribonucleoprotein complex consisting of two subunits: a functional catalytic protein called telomerase reverse transcriptase (TERT) and an RNA component (TERC), which works as a template site for DNA elongation [5,6]. The telomerase complex also contains a variety of accessory proteins essential for its biogenesis and function in vivo [7,8,9]. TERT and TERC are essential subunits for the correct functioning of telomerase [10,11]. However, while TERC is constitutively expressed in most somatic cells and in germ cells, TERT is considered a limiting factor because its expression is tightly regulated. Consequently, telomerase expression is finely controlled during embryogenesis and in most differentiated cells, where its activity is low or absent. Indeed, the protein is only expressed in somatic cells with rapid renewal potential, such as hematopoietic and stem cells [12,13]. Telomerase reactivation in normal human cells confers unlimited proliferative ability and may cause genomic instability, with a high probability of developing cancer [14,15]. Telomerase is active in 85% of tumor cells, while the other 15% displays a different mechanism of telomere length maintenance based on recombination, known as Alternative Lengthening of Telomere (ALT) [16,17]. In the last 20 years, several authors have reported that TERT is involved in mechanisms other than telomere maintenance [18], such as gene expression regulation of WNT/β-catenin or NF-kB pathways [19,20]. Its role in crucial physiological processes, including cell cycle, metabolism, differentiation, cell signaling and cell survival, has been recognized [19,21,22,23], even though interest in TERT functions has mainly been referred to anti-apoptotic [24,25] and antioxidant effects [26,27], besides protection against specific DNA-damaging agents [28,29,30]. The enhanced resistance to oxidative stress observed in TERT-overexpressing cells could be related to increased levels of the ROS scavenger glutathione (GSH), as well as of manganese superoxide dismutase (MnSOD or SOD2) and forkhead-box-protein O3 (FoxO3a) proteins [25,26,31]. The nuclear export of TERT is strictly controlled and such regulation includes shuttling between the nucleus and cytoplasm through nuclear export signal (NES), identified at the protein C-terminus. Upon oxidative stress, Scr kinase is the main trigger of the nuclear exclusion of TERT responsible for phosphorylation on tyrosine 707 [32]. In addition, a specific N-terminal sequence, comprising 20 amino acid residues represents the mitochondria transport signal (MTS) [33]. MTS is indispensable for TERT transport via the translocases of outer (TOM 20 and TOM 40) and inner mitochondrial membranes (TIM 23) and for its localization in the mitochondrial matrix, suggesting an active mitochondrial import mechanism [34,35]. Moreover, several authors have shown that hydrogen peroxide treatment increases TERT mitochondrial levels [33,34,36]. The significance of such induction remains unclear, as evidences of either exacerbation of H2O2-induced mtDNA damage or protection against different types of insult have been provided [33,37]. In fact, most data suggest that TERT participates in antioxidant response, to protect mtDNA, reduce mitochondrial ROS production and improve mitochondrial functions [36]. Moreover, the mechanisms underlying such action and the biological role of TERT in mitochondria remain undefined and require further investigation, focusing on this issue. In this research, we aimed to study TERT telomere-independent functions in a normal cellular context, taking advantage of normal human primary fibroblasts (HFFF2), cells without any mutation and that lack TERT expression and, consequently, telomerase activity. After the stable overexpression of the TERT protein in HFFF2 cells, we investigated the response to X-rays and H2O2 treatment, in terms of DNA damage and oxidative stress induction. Subsequently, we focused on the antioxidant response with a particular focus on TERT’s role inside mitochondria. Reactive oxidative species (ROS) level and the expression of factors commonly involved in the antioxidant response were measured. TERT translocation inside mitochondria was demonstrated in this work, and its consequences on the health and morphology of this compartment were investigated. In order to unravel the role of TERT in responding to different genotoxic agents, specifically X-rays and H2O2 treatment, we employed normal human primary fibroblasts (HFFF2), lacking TERT expression and consequent telomerase activity [13] and hTERT-overexpressing HFFF2 cells (HF-TERT). This cell line was obtained by retroviral transduction with a plasmid containing the cDNA for the human TERT and was tested by Western blot and RQ-TRAP (Real-time Quantitative—Telomeric Repeat Amplification Protocol assay) showing TERT expression and telomerase activity (Figure S1). In the first instance, we compared the cell growth curves of normal and transduced fibroblasts subjected alternatively to X-rays and H2O2 treatment, in order to analyze the difference in cell proliferation due to TERT overexpression. As expected, even in untreated conditions HF-TERT displayed a higher growth rate than HFFF2. X-ray irradiation appeared to significantly affect cell proliferation in both cell lines, particularly in HFFF2 that, until 72 h, appeared to have no growth. On the contrary, H2O2 treatment led to a slight decrease in growth but not in a significant manner compared to control samples (Figure S2A,B). Comparing the different cell lines under the same treatment, it is possible to note that HFFF2 cells showed slower growth rates than HF-TERT after X-rays, differentially of hydrogen peroxide treatment in which there is no differences after treatment (Figure S2C,D). To evaluate DNA damage induction after X-rays and H2O2 treatment, we detected by immunofluorescence a marker of DNA damage, the phosphorylated form of histone H2AX (γH2AX) that is recruited not only after a DNA double-strand break (DSB) but also due to replication fork stalling (Figure 1A) [38,39]. In HFFF2 and HF-TERT cell lines, both genotoxic agents induced a significant increase in γH2AX foci (Figure 1B,C). After X-rays irradiation in HFFF2 cells, a higher damage level persisted for many days after treatment while HF-TERT displayed less DNA damage, which was completely repaired in the following days (Figure 1B). After H2O2 treatment, both cell lines showed a similar behavior up to 24 h, while after 48 h HF-TERT completely repaired the genomic damage, differently from normal fibroblasts, in which we observed a further increase in γH2AX foci (Figure 1C). Since telomeric DNA is less efficiently repaired than other parts of the genome [40], and telomeric sequences are characterized by high content of guanine residues, these regions are more susceptible to oxidative damage, especially to the accumulation of oxidized bases such as 8-oxoguanine [41,42,43]; the presence of 8-oxoG is able to induce replication fork arrest at telomeres, resulting in DSB [43]. Thus, we analyzed a marker of telomeric dysfunction, TIFs (Telomere Induced Dysfunctional Foci) by immunofluorescence, detecting the co-localization of TRF1, a specific telomeric protein, with γH2AX (Figure S3A) [44]. In this case, we noted different levels of TIFs induction between the two cell lines. HF-TERT cells had also less telomeric damage both after X-rays and after H2O2 treatment compared to HFFF2 cells (Figure S3B,C). Following TIFs results, we performed Quantitative-Fluorescence “In Situ” Hybridization (Q-FISH), a technique used to evaluate telomere length modulation (Figure S4A,B). X-rays caused modulation in telomere length in HFFF2 (as previously described in our lab, [45]) but not in HF-TERT (Figure S4C–E). This could be observed not only from the mean telomere lengths but above all from the shape of the telomere length distributions analyzed at 72 and 96 h after irradiation (Figure S4D,F). In addition, considering that telomere length stability in HF-TERT is probably due to telomerase activity, we performed RQ-TRAP assay to analyze telomerase activity after X-rays treatment. The data did not show any change in telomerase activity, which seems to be not affected by the treatment and probably not connected with the DNA damage level (Figure S4G). Excluding the role of telomerase, in order to understand why HF-TERT cells had less genomic and telomeric damage than HFFF2, we evaluated oxidative stress (OS), induced by H2O2 or X-ray irradiation. Thus, we quantified reactive oxygen species (ROS) after both treatments. X-ray irradiation-induced oxidative stress in both cell lines, but with some dissimilarities (Figure 2A): in normal cells, OS increased after 48 h, reaching its maximum level at 72 h and subsequently decreasing up to 168 h since X-rays exposure; on the contrary, in HF-TERT cells we can note a statistically significant increase in OS only 72 h after treatment that went down immediately after 96 h. After the treatment with hydrogen peroxide for 1 h, ROS production was measured immediately (t0), after 1 h of cellular growth in culture medium (t1) and after 2 h (t2) of recovery (Figure 2B). Treated HFFF2 cells displayed a high level of ROS production immediately after treatment, which persists until t2 time. The same trend is observed in HF-TERT but with a lower level of ROS that returned at the control level after 1 h of recovery. This lower level of OS than that observed in normal fibroblasts could explain the less genomic and telomeric damage. Since HF-TERT cells showed a lower OS, we hypothesized that TERT could improve cellular antioxidant defense through an increase in specific factors involved in the response, such as Glutamate Cysteine Ligase (GCL) and manganese superoxide dismutase (MnSOD or SOD2) [46]. GCL is a heterodimeric protein composed of a catalytic (GCLC) and a modifier (GCLM) subunit [47], and catalyzes the formation of the cellular antioxidant glutathione (GSH). The gene expression of GCLM, GCLC and SOD2 was measured by RT-qPCR in normal fibroblasts and hTERT-overexpressing cells, after 1 and 3 h of recovery following H2O2 treatment. In HFFF2 we did not observe any change in the expression of GCLC and SOD2 genes while we noted an increase in GCLM expression after 1 h and 3 h after treatment. On the contrary, in HF-TERT cells we observed an increase in gene expression for both GCL subunits and for SOD2 after 1 and 3 h of recovery (Figure 3A–C). It is interesting to note that TERT over-expressing cells are characterized by lower basal levels of GCLC and SOD2 compared to HFFF2; this finding could be ascribable to fainter oxidative stress in HF-TERT cells, which requires lower levels of the proteins involved in the antioxidant response. Considering that we observed SOD2 gene modulation only in HF-TERT cells and that TERT was previously shown to induce an increase in SOD2 protein level [27], we also analyzed the protein levels of SOD2 by Western Blot analysis (Figure 4A) in both cell lines. Differently from the results on gene expression, in HFFF2 cells we noted a slight increase in SOD2 protein level after treatment. On the other hand, TERT-overexpressing cells displayed a significant increase in SOD2 protein level after hydrogen peroxide treatment, in accordance with gene expression (Figure 4B), confirming a different response to the OS of the two cell lines. As observed for gene expression, SOD2 protein levels in HF-TERT cells are lower compared to normal cells (both in untreated and treated samples), but the increase induced by H2O2 treatment is more prominent (after 3 h, 80% in HF-TERT compared to 26% in HFFF2). TERT protein is characterized by a nuclear export signal, responsible for its translocation to the cytoplasm; here, it can be subjected to ubiquitination and consequent degradation or directed into mitochondria, thanks to a specific N-terminal sequence, the mitochondria transport signal (MTS) [33]. TERT protein is principally localized within the nucleus, but it was also demonstrated that upon oxidative stress TERT is transported from the nucleus to the cytoplasm [32]. Thus, we analyzed the TERT level in the cytosolic fraction of untreated HFFF2 cells as well as untreated and H2O2-treated HF-TERT; we observed a decrease in the cytoplasmatic TERT amount after treatment (Figure 5A,B). Considering this result, we have wondered if the observed TERT reduction in cytosol after hydrogen peroxide treatment was caused by its degradation. Lee et al. discovered that the CHIP chaperone physically interacts with hTERT in the cytoplasm, resulting in hTERT cytoplasmatic degradation by the proteasome [48,49]. Western blot showed that CHIP expression in cytoplasm did not change after hydrogen peroxide treatment (Figure S5A,B), suggesting that the reason for TERT modulation could be due to a mechanism different from its degradation. Since the TERT sequence is characterized by the presence of the MTS (mitochondrial targeting sequence) responsible for its mitochondrial localization [33], we supposed that the reduction in cytosolic TERT was not due to degradation but to its translocation in mitochondria. Therefore, we investigated the presence of TERT protein in the mitochondrial fraction in untreated HFFF2 cells and in untreated and H2O2-treated HF-TERT cells (Figure 5C). Western blot analysis showed an increase in TERT protein in mitochondria following treatment with hydrogen peroxide (Figure 5D). These results were also confirmed by double-fluorescent labeling with anti-TERT antibody and the mitochondrial localization probe Mitospy (Figure 5E). Untreated HF-TERT cells showed a prevalence of TERT signals in the nucleus but also a few signals in cytosol (Figure 5E, upper panel). H2O2-treated HF-TERT cells displayed a large amount of colocalization between TERT and mitochondria as shown in Figure 5E, lower panel. Moreover, the RQ-TRAP assay performed after H2O2 treatment showed that telomerase activity is not affected by the observed different TERT localization in cellular compartments (Figure S6). In order to assert with certainty TERT localization within mitochondria, we performed experiments with transmission electron microscopy (TEM) using immunogold particles in untreated and treated HF-TERT cells (Figure 6A). As shown in Figure 6B in HF-TERT cells, we saw most mitochondria without gold particles (mean of 69.4 mitochondria), but there were mitochondria with one or two particles (24.4 and 6.1, respectively). In HF-TERT-treated cells, the distribution of gold particles appears different (Figure 6C). We observed a decrease in mitochondria without gold particles (34.9), instead an increase in mitochondria with more than two particles (22.2 with two particles; 18.2 with three particles; 11.1 with four particles; and 4.7 with five particles). Furthermore, immunogold TEM analysis confirmed the Western blot results, and we can conclude that TERT protein was translocated into mitochondria after H2O2 treatment (Figure 6D), and this translocation does not induce nuclear telomerase activity change. Ultrastructural analysis was performed by Focused Ion Beam/Scanning Electron Microscopy (FIB/SEM), to investigate overall cellular morphology and particularly, mitochondrial features. Both cell lines displayed regular cell shape and electron density, with a well-preserved plasma membrane, nuclear morphology and cytoplasmic organelles (Figure 7A–E). Mitochondria of either HFFF2 or HF-TERT cells showed consistent ultrastructural features, i.e., a typically elongated morphology with cristae arranged in a parallel manner and a moderate electron-dense matrix (Figure 7C–G). Following H2O2 treatment, the general morphological features of either cell lines appeared impaired (Figure 7B–F). Numerous mitochondria showed abnormal outer and inner membrane (IMM and OMM) ultrastructure, with pronounced swelling and consequent expansion of the electron-lucent matrix. Cristae, particularly, were reduced in number and lost their normal topology, being fragmented and irregularly arranged (Figure 7D–H). Occasional rupture of the OMM was also observed (Figure S7). Despite the qualitative similarity of alterations caused by oxidative insult, hTERT-overexpressing cells responded differently, in terms of mitochondrial damage extent and severity. Indeed, quantitative analysis of randomly acquired FIB/SEM microphotographs (see M&M, for details) from longitudinally sectioned cells (Figures S8 and S9), showed a relatively small increase in abnormal mitochondria in HF-TERT cells. A highly significant difference in the number of damaged mitochondria was found in HFFF2 cells after treatment, while in HF-TERT cells this difference fails to reach significance (diagrams in Figure 7). Moreover, numerous lysosomes and autophagic vesicles were observed after treatment, especially in TERT-overexpressing cells. Successively, we analyzed the mitochondrial mass to determine if there were changes in metabolic functions afterwards hydrogen peroxide treatment, by using Mitotracker Green, as shown in Figure 8A,B. In normal cells, we did not observe any change immediately after treatment, but we saw a slight decrease in mitochondrial mass after 3 h of recovery. On the contrary, untreated HF-TERT cells displayed less mitochondrial mass compared to their untreated counterpart. After H2O2 treatment, mitochondrial mass was already significantly reduced immediately and after 3 h of recovery. On the other hand, mtDNA copy number was evaluated by quantification of the MT-ND4 gene and MT-7SDNA in HFFF2 and HF-TERT cells after treatment (Figure 8C,D). qPCR results showed a slight reduction in the mtDNA copy number in HF-TERT after hydrogen peroxide treatment, while in normal cells, the mtDNA copy number was unaffected after H2O2 treatment. To evaluate the mitochondrial function after TERT translocation, we first measured the mitochondrial membrane potential (Δψm) using the fluorescent dye JC-1, a molecule which accumulates in mitochondria in a membrane potential-dependent manner, forming reversible aggregates (red emission) (Figure 8E). Following mitochondria depolarization JC-1 shifts to a monomeric form resulting in green emission [50]. HFFF2 cells exposed to H2O2 for 1 h showed a significant decrease in the red to green fluorescence intensity ratio respect to control cells, indicating a mitochondrial depolarization. Differently, H2O2-treatment of HF-TERT cells did not induce any mitochondrial depolarization as observed by no JC-1 ratio change (Figure 8F). Interestingly, the J-aggregates in treated normal cells were significantly less abundant than treated hTERT-overexpressing cells. In addition, we measured the ATP levels, the source of energy within the cell produced by mitochondria [51]. After H2O2 treatment, HFFF2 cells displayed a significant decrease in ATP level for all the analyzed timepoints (immediately, after 1 h and after 3 h of recovery), while HF-TERT cells showed ATP reduction only after 1 or 3 h of recovery (Figure 8G,H). Mitochondrial transcription factor A (TFAM) is a member of the high-mobility group (HMG) [52], able to directly bind mtDNA, but it is also involved in many functions such as helping mitochondrial DNA in transcription and replication [53,54]. As a possible indication of the modulation of this process, we analyzed the expression of the TFAM gene in HFFF2 and HF-TERT cells at 1 and 3 h after H2O2 treatment. Even though in normal fibroblasts we observed a slight not significant increase in the TFAM expression after treatment (Figure 9A), only in HF-TERT cells, TFAM gene showed a significantly increased expression at both times analyzed (Figure 9B) suggesting likely different regulation of mtDNA functions. TERT, together with TERC, aside from the well-established function in telomere lengthening, has non-canonical functions as a transcriptional regulator of genes in different pathways but also has an anti-apoptotic and antioxidant role [24,26,27]. In recent years, many different reports have focused on the role of TERT in increasing resistance to specific DNA-damaging agents and in reducing cellular ROS levels, with a protective effect on the cellular redox status [26,30]. Taking advantage of normal primary fibroblasts (HFFF2), cells without any mutation and that lack TERT expression and, consequently, telomerase activity, transduction with hTERT gene allowed us to study TERT telomere-independent functions after oxidative stress (OS) damage, induced by physical and chemical agents (X-rays and H2O2). Firstly, we evaluated total and telomeric DNA damage induction using a phosphorylated form of histone H2AX (γH2AX), recruited after DSB [38]. Both genotoxic agents induced a significant increase in DNA damage. However, HFFF2 cells had a greater damage induction that persisted for many days after treatment, while TERT-overexpressing fibroblasts had less DNA damage, which was completely repaired in the following days. The same results have been observed in telomeric sequences. Telomeric DNA is a preferential target of OS due to a high content of guanine residues, which makes it susceptible to oxidative damage. In addition, the telomeric region is also less efficiently repaired than other parts of the genome [40,55]. In this case, we have analyzed the colocalization of γH2AX and TRF1, a specific telomeric protein, in order to analyze Telomere Induced Dysfunctional Foci (TIFs) and we have evidenced different levels of TIFs induction between the two cell lines. Normal cells showed higher telomeric damage after X-rays and H2O2 treatment compared to HF-TERT, confirming results obtained from the analysis of genomic DNA damage. These observations are consistent with the results reported by Sharma and coll. that demonstrated a better response to DNA damage in cells with ectopic expression of TERT, after treatment with ionizing radiation and cisplatin [30]. In addition, in cells with suppression of TERT, an increase in radiosensitivity, diminished capacity for DNA repair and fragmented chromosomes were observed, demonstrating that loss of TERT impairs the DNA-damage response [56]. The lower induction of genomic and telomeric damage observed by us and other authors in normal cells overexpressing TERT, could, however, also result from protective effects of TERT on cellular redox status [30,36]. Following this hypothesis, we decide to assess levels of OS, induced both by H2O2 or by X-ray irradiation, considering that also X-rays are able to induce oxidative stress [45]. Thus, we have quantified reactive oxygen species (ROS) after both treatments. Although the two cell lines displayed a similar trend, HF-TERT cells seemed to have less induction of OS, both after X-ray irradiation and H2O2 treatment. These results are similar to those of Ahmed et al. (2008), in which over-expressing hTERT fibroblasts were shown to display reduced levels of intracellular ROS and therefore reduced oxidative stress under both basal conditions and induced chronic oxidative stress [36]. This evidence was also observed in non-transformed cells, as reported by Yang and colleagues [22]. The authors using embryonic stem cells showed ROS levels lower than that observed after the TERT knockdown, suggesting resistance to oxidative stress. In tumor cells, Indran et al., 2011, have demonstrated that TERT reduces basal cellular ROS level and intracellular ROS in response to different stimuli. Vice versa, TERT downregulation in the same cells induces an increase in ROS. These anti-oxidative effects of TERT are correlated to an increase in Glutathione (GSH) and non-oxidized peroxiredoxin [26]. In agreement with different authors, our results indicate less induction of ROS in HF-TERT than in normal fibroblasts; thus, we have investigated different antioxidant genes such as Glutamate Cysteine Ligase (GCL) and manganese superoxide dismutase (MnSOD or SOD2) [27,57]. GCL is a heterodimeric protein, composed of a catalytic (GCLC) and a modifier (GCLM) subunit [47], which catalyzes the formation of the cellular antioxidant glutathione (GSH). While GCLM increased in both cell lines, GCLC increased only in HF-TERT cells. This increase probably leads to a greater GCL heterodimer formation and consequent increase in antioxidant ability. We also observed increases in SOD2 mRNA and protein levels in HF-TERT cells after treatment, confirming a better antioxidant response compared to normal fibroblasts. Collectively, these observations are in agreement with the ability of hTERT (via the interaction with NF-kB) to induce the expression of antioxidant genes [58,59,60]. Moreover, it is interesting to note that TERT antioxidant activity is present even at basal conditions, as revealed by the observation of a lower level of ROS and of some factors involved in the cellular antioxidant defense (GCL, SOD2). Interestingly, in recent years, it was found that TERT functions are not limited to the nucleus. Haendeler and colleagues have discovered that the Scr kinase family regulates TERT export from the nucleus to the cytoplasm under oxidative stress [32]. ROS provokes rapid activation of the Src kinase, which induces phosphorylation of nuclear TERT on tyrosine 707. This phosphorylated form interacts with the nuclear export receptor CRM1/exportin, and so, it is actively transported through nuclear pores [32,61]. This result clearly identifies oxidative stress as the main trigger for TERT nuclear exclusion. Due to this interesting data and with the aim of studying TERT extra telomeric roles, in our model, we analyzed TERT level in the cytosolic fraction after hydrogen peroxide treatment and we observed a decrease in TERT amount in H2O2-treated HF-TERT cells. Firstly, we considered the hypothesis that this reduction could be due to TERT ubiquitination and consequent degradation, induced by the interaction with the chaperone CHIP [48,49]. We investigated if CHIP protein could be responsible for the observed TERT reduction. However, our results showed that the CHIP level in cytosol did not change after treatment. Thus, the reduction in TERT level seemed to be not ascribable to an increased degradation but probably to its translocation in another compartment. In fact, Santos and co-authors have identified a specific N-terminal sequence of TERT, the mitochondria transport signal (MTS), that allows TERT to be transported into mitochondria through the mitochondrial membrane [33]. Haendeler and co-workers have found that TERT interacts with both TOM20 and TOM40 at the mitochondrial outer membrane and TIM23 at the inner membrane and TERT resides in the mitochondrial matrix [34]. Different authors in recent years have confirmed the presence of TERT in mitochondria, using different cellular models, and have shown that hydrogen peroxide treatment is able to increase its translocation into mitochondria [34,36,37]. Using our cellular models and different techniques we have demonstrated that TERT is located in mitochondria and is further translocated inside the organelle after hydrogen peroxide treatment. Treated HF-TERT cells showed an increased amount of TERT into mitochondria, compared to untreated cells, indicating that oxidative stress could represent a trigger for TERT translocation into mitochondria. We have demonstrated the presence of TERT protein in the mitochondrial fraction by Western blot, which increased after H2O2 treatment. Subsequently, both immunofluorescence and transmission electron microscopy (TEM) have corroborated previously observed results. To date, various methods have been used to reveal the localization of TERT in mitochondria, including immunoblotting, coimmunoprecipitation and immunofluorescence [33,34,36,62]; however, for the first time in this paper, TERT was detected inside mitochondria by TEM immunogold labeling, which is one of the most sensitive methods for localization and quantification of antigens in different cellular compartments or organelles [63,64]. After showing that, in our conditions, TERT moves into mitochondria after oxidative stress, our attention was focused on the study of the possible effects of its translocation, probably influencing mitochondrial status and functionality. Thus, we tested the mitochondrial function after H2O2 treatment evaluating the mitochondrial membrane potential (Δψm) and we showed mitochondrial depolarization only in treated normal cells. Differently, in HF-TERT cells we did not observe any change in Δψm level, suggesting that TERT probably ameliorates mitochondrial health status. Successively, in transduced fibroblasts we have observed a reduction in mitochondrial mass and ATP after H2O2-treatment, consequently accompanied by a reduction in mtDNA. On the other hand, ultrastructural analyses showed in H2O2-treated HF-TERT cells fewer mitochondria abnormalities compared to normal cells. Taking into account all the mitochondrial data, HF-TERT fibroblasts contained a lower number of mitochondria (observed as mass) but are characterized by a better function and morphology compared to normal fibroblasts (see in Δψm and ultrastructural analysis). Furthermore, data obtained from the analysis of TFAM, typically involved in mtDNA transcription and replication, indicated only in HF-TERT an increase in TFAM expression suggesting that under OS, TERT induces TFAM as an early step of a mitochondrial renewal process. Therefore, we can speculate that if on one hand TERT could activate a mechanism to discard damaged mitochondria; on the other hand, this protein may be responsible for the restoration of new healthy mitochondria, as suggested by TFAM induction in TERT-overexpressing cells. This peculiar behavior of TERT, under oxidative stress conditions, could reveal further important roles of this protein, that seem to be apparently contradictory but anyway dedicated to guarantee cellular survival. It will be interesting deeper investigate this hypothesis, in order to clarify which are the specific actors involved in mitochondria biogenesis and in mitochondria disruption. In conclusion, the use of human primary fibroblasts and their transduced counterpart allowed us to study the role of TERT in normal cells, in a normal cellular context, free of any gene mutation. Our results have confirmed previous literature data, highlighting even more the antioxidant role of TERT, both under basal and stress conditions. In fact, TERT untreated cells displayed a better response than normal primary fibroblasts for many parameters. Most importantly, we have demonstrated TERT translocation into mitochondria induced by oxidative stress; this translocation seems to be related to the TERT protective role in preserving mitochondria functionality, as demonstrated by our functional and structural results. Further investigations are needed to understand the reason of TERT translocation into mitochondria induced by oxidative stress and to unravel any targets and the putative mechanism of action inside the mitochondrion. Human Fetal Foreskin Fibroblasts (HFFF2) (ECACC, Salisbury, UK) and Human Epithelial Kidney cells (HEK 293) were grown in D-MEM. Cell medium was supplemented with 10% fetal bovine serum (FBS), 10,000 units/mL penicillin, 10 mg/mL streptomycin and 2 mM L-glutamine (Euroclone, Milan, Italy). All cell lines were maintained in a humidified incubator at 37 °C, with 95% relative humidity and 5% CO2. As described by Counter et al. [65], pBabe-Puro-hTERT (Addgene plasmid #1771) vector plasmid containing the cDNA of hTERT protein was transfected into HEK 293 using Lipofectamine 2000 (Thermo Fisher Scientific, USA) and Ampho Retrovirus Packaging Vector (Novus Biological, Englewood, CO, USA) to obtain retroviral vector particles. Four days later, the supernatant from these cells were used to infect HFFF2 cells to establish the HF-TERT cell line. After the infection, cells were selected with puromycin (2 μg/mL) (Tocris, Ellisville, MO, USA) for 5 days. Human glioblastoma astrocytomas (U251MG) were grown in MEM, supplemented with 10% fetal bovine serum (FBS), 10,000 units/ml penicillin, 10 mg/ml streptomycin, 2 mM L-glutamine, 1% Non-Essential Amino Acids and 1 mM Sodium Pyruvate (Euroclone, Milano, Italy). U251 were used in RQ-TRAP assay as positive control of telomerase activity. Telomerase activity was measured by the SYBR green RT-qPCR assay, which was conducted as described elsewhere [66] with minor modifications. Briefly, the reaction was performed with protein extracts, 0.1 µg of telomerase primer TS, and 0.05 µg of anchored return primer ACX, in SYBR Green PCR Master Mix (Biorad, Hercules, CA, USA). The primer sequences were those reported by Kim and Wu [67]. The reaction was performed using the Agilent AriaMax real-time PCR system (Agilent Technologies, Santa Clara, CA, USA), samples were incubated for 30 min at 30 °C and amplified in 40 PCR cycles with 30 s at 95 °C and 90 s at 62 °C (two step PCR). HF-TERT heat-treated cells samples were obtained by boiling protein extract at 85 °C for 10 min. Telomerase activity was expressed relative to the telomerase positive (U251MG) cells and HF-TERT heat treated cell was used as negative control. Each sample was analyzed in triplicate in at least three independent experiments. Cells were subjected alternatively to two different genotoxic agents, X-ray irradiation or hydrogen peroxide. In the first case, cells were X-rays irradiated at room temperature using a Gilardoni apparatus (200 kV, 6 mA, dose-rate 0.51 Gy/min) with a dose of 4 Gy, then were seeded at the requested density in fresh medium. Unirradiated cells were used as the control in all the experiments and were seeded at the requested density as irradiated cells. In the case of H2O2 treatment, cells were treated with H2O2, 24 h after seeding in a complete medium, for 1 h at 37 °C in incubator at the final concentration of 200 μM. After 1 h of treatment, H2O2 was removed and cells were grown in complete medium for 1 or 3 h of recovery. Cells were examined at different times after treatment and were compared to parallel cultured control HFFF2 cells grown in the medium without H2O2. After treatment, 1 × 105 cells were seeded in a Petri dish and every 24 h were detached and counted up to 120 h. All data points were performed at least in three different experiments. After treatment, the cells were fixed with 4% paraformaldehyde (Sigma Aldrich‚ Burlington, MA, USA), permeabilized in Triton X-100 and blocked with 1% bovine serum albumin (BSA). Slides were incubated with a mouse monoclonal anti-phospho-histone H2AX antibody (Millipore, Burlington, MA, USA) in combination, when needed, with rabbit telomeric protein TRF1 antibody (Santa Cruz Biotechnology, Dallas, TX, USA), overnight at 4 °C. Samples were washed in PBS then exposed to the secondary anti-mouse Alexa 546 antibody (Invitrogen, Waltham, MA, USA) and anti-rabbit Alexa 488 (Invitrogen, Waltham, MA, USA) for 1 h. DNA was counterstained by DAPI in Vectashield (Vector Laboratories Inc, Newark, CA, USA). Slides were analyzed using Axio Imager M1 microscope (Carl Zeiss) equipped with a CCD camera. The frequency of foci per cell were scored in 50 nuclei in at least three independent experiments. After treatment, chromosome spreads were obtained following 30 min incubation in 30 μM calyculin-A (Wako, Germany) [68]. Spreads of these prematurely condensed chromosomes (PCC) were prepared by a standard procedure, consisting of treatment with a hypotonic solution (75 mM KCl) followed by fixation in freshly prepared Carnoy solution. Q-FISH staining was performed as described by Berardinelli et al. [69] with minor modifications. Briefly slides and probes (Cy3 linked telomeric and chromosome 2 centromeric Peptide Nucleic Acid PNA probes; PANAGENE, Republic of Korea) were co-denatured at 80 °C and hybridized for 2 h at room temperature in a humidified chamber. Slides were counterstained with DAPI in Vectashield. Images were captured at a 63× magnification with Axio Imager M1 microscope equipped with a CCD camera. The telomere size was analyzed with ISIS software (MetaSystems, Altlußheim, Germany). In particular, the software calculates telomere lengths as the ratio between the total telomeres fluorescence (T) and the fluorescence of the centromere of the two chromosomes 2 (C), thus data were expressed as a percentage (T/C%). Experiments were repeated at least three times, and at least 10 metaphases were scored for each experiment. The accumulation of intracellular ROS level was detected after irradiation or hydrogen peroxide treatment, with the ROS detection assay kit purchased from BioVision. Cells were irradiated and analyzed at different times. After irradiation 1 × 104 cells were seeded inside 96-multiwell plates. For hydrogen peroxide treatment, 2 × 105 cells were seeded in a Petri dish. The culture medium was discarded, new medium containing the fluorogenic probe H2DCFDA was added (1× ROS label solution, as indicated in the manufacturer’s protocol) and incubated for 30 min at 37 °C. Samples were treated with 200 µM H2O2 for 1 h. Subsequently, hydrogen peroxide was removed and added in a complete medium. In the untreated samples, after probe incubation, were added culture media. At the end of stimulation, cells were analyzed in a FL-1 fluorescence channel using a CytoFlex (Beckman Coulter, Pasadena, CA, USA) flow cytometer. About 20,000 events/samples were analyzed for each condition. Analysis was performed with a CytExpert v2.2 software (Beckman Coulter, Brea, CA, USA). Doublet discrimination was performed by an electronic gate on FSC-Area vs. FSC-Height. Experiments were repeated at least three times. Total RNA was extracted from HFFF2 and HF-TERT cells using the Total RNA Purification Plus Kit (Norgen Biotek Corp, Thorold, ON, Canada) and has been reverse-transcribed using a LunaScript™ RT SuperMix Kit (New England Biolabs, Ipswich. MA, USA). Quantitative reverse transcription PCR (RT-qPCR) analysis has been performed using the TaqMan Universal PCR Master Mix (Applied Biosystems, Paisley, UK), TFAM (Hs00377764_g1), GCLC (Hs00155249_m1), GCLM (Hs00978072_m1) and SOD2 (Hs00167309_m1) have been used and the relative abundance of each target transcript has been normalized to the expression level of GAPDH (Hs99999905_m1). Data have been analyzed using the 2−ΔΔCt method and reported as fold change relative to controls (untreated cells). Experiments were repeated at least three times. Cells were treated as described above and lysed in 20 mM Tris HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton-100X and protease inhibitors, to obtain the whole-cell extract. Protein extracts of mitochondrial and cytosolic fractions were isolated using the Mitochondria/Cytosol Fractionation Kit (Abcam, Waltham, MA, USA) and according to the manufacturer’s protocol. Protein extracts (50 μg) were loaded on an SDS-PAGE and transferred onto a polyvinylidene fluoride (PVDF) membrane (Immobilion-P, Millipore). After blocking in 3% BSA, membranes were incubated with the following primary antibodies: anti-β-Actin (C4) (sc-47778, Santa Cruz Biotechnology, Dallas, TX, USA), anti-TERT (600-401-252S, Rockland), anti-SOD2 (sc-137254, Santa Cruz Biotechnology), anti-Aco2 (NBP1-32781, Novus Biological), anti-CHIP (ab2917, Abcam), anti-COX IV (ab14744, Abcam) and anti-α-tubulin (T6199, Sigma-Aldrich). Finally, membranes were incubated with the appropriate HRP-conjugated secondary antibody (Bio-Rad Laboratories, Hercules, CA, USA). Proteins were visualized using ClarityTM Western ECL substrates (Bio-Rad). Images were acquired on ChemiDoc™ Imaging system (Bio-Rad) and protein levels were quantified using the Image Lab software (Bio-Rad). Experiments were repeated at least three times. Cells were seeded on chambered coverslips (μ-Slide 8 well, Ibidi) and subjected to H2O2 treatment as described above, and after 3 h of recovery, cells were incubated with 500 nM MitoSpyTM (BioLegend) at 37 °C for 30 min. Then, cells were fixed with 4% paraformaldehyde (Sigma Aldrich) for 15 min, permeabilized and blocked in 5% BSA for 1 h. Slides were incubated with a rabbit polyclonal anti-TERT (Rockland) overnight at 4 °C and then exposed to the secondary antibody (anti-rabbit Alexa 488, Invitrogen) for 1 h at 37 °C. Finally, fluorescent images were registered with Leica TCS SP5 confocal microscope and processed with LAS AF software (version 1.6.3, Leica Microsystems CMS GmbH). Cells were centrifuged at 1800 rpm for 10 min and the resultant pellets were fixed in 4% glutaraldehyde for 2 h at 4 °C. After several washes in 0.1 M cacodylate buffer (pH 7.4), samples were post-fixed in 1% osmium tetroxide for 20 min in the dark, dehydrated in ethanol series and embedded in an Epon-Araldite 812 mixture (Sigma-Aldrich, Milan, Italy). Ultrathin sections (70 nm in thickness) were obtained with a Reichert Ultracut S ultratome (Leica, Wien, Austria) and collected on gold grids (300 mesh). After etching with 3% NaOH in methanol, slides were incubated for 30 min in a blocking solution containing PBS, 1% BSA and 0.1% Tween and then, with the rabbit antibody, anti-hTERT (Rockland) was diluted at 1:200 in BSA blocking solution for 1 h. After several washings with PBS, the primary antibody was visualized by immunostaining with the secondary goat anti-rabbit IgG (H + L)-gold conjugate antibody (GE Healthcare, Amersham, UK; particle size, 10 nm) diluted at 1:50 in BSA blocking solution for 45 min. Subsequently, slides were treated for 5 min with PBS containing 0.5% glutaraldehyde, counterstained with uranyl acetate and observed under a Jeol 1010 EX transmission electron microscope (Jeol, Tokyo, Japan). Data were recorded with a MORADA digital camera system (Olympus, Tokyo, Japan) and the frequency of mitochondria scored 100 in cells. The fluorescent dye JC-1 (5,5′,6,6′-Tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine Iodide, AdipoGen) was used to assess the mitochondrial membrane potential. JC-1 is a lipophilic cation dye that exhibits a potential-dependent accumulation in mitochondria indicated by a fluorescence emission shift from green (529 nm) to red (590 nm) [50]. Cells were seeded in a Petri dish and treated with 200 µM of H2O2 for 1 h; subsequently, hydrogen peroxide was removed, cells were washed once with PBS and then incubated with 5 μg/mL JC-1, at 37 °C for 15 min. After two washes with PBS, cells were covered with fresh PBS and their fluorescence was analyzed using a Zeiss Axiophot (Carl Zeiss) fluorescent microscope. Mitochondrial uncoupler FCCP (Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone, Sigma) was used as a positive control. Acquired images were analyzed with ImageJ (NIH, Bethesda, MD, USA) for green and red fluorescence and results were given as red/green fluorescence ratio. At least 100 cells per sample were analyzed. The cellular ATP content was measured using a luminometric assay (ATPLITE 1 STEP, PerkinElmer, Waltham, MA, USA) as follows: 1 × 104 cells per well were seeded in triplicate on 96-well microtiter. The following day cells were treated with 200 µM of H2O2 for 1 h. Subsequently, ATP measurement has been performed using the EnSpire® Multimode Plate Readers. Experiments were repeated at least three times. DNA was extracted from cells using the NucleoSpin Tissue kit for DNA (Macherey-Nagel). A total of 30 ng of DNA were used for qPCR analysis of mitochondrial DNA using a specific Taqman probe for MT-ND4 and MT-7S genes, while nuclear gene RNase P (Copy Number Reference Assay, human, RNaseP) was quantified for normalization. Data were analyzed using the 2−ΔΔCt method and reported as a fold change relative to controls. Experiments were repeated at least three times. Mitochondrial mass per cell was measured using MitoTracker Green FM (Molecular Probes, Eugene, OR, USA). A total of 1 × 105 cells were seeded in a Petri dish and treated with 200 µM H2O2 for 1 h; subsequently, hydrogen peroxide was removed, cells were washed once with PBS and then incubated with 50 nM MitoTrackerTM Green FM (ThermoFisher, Waltham, MA, USA) for 30 min at 37 °C in the dark. After two washes with PBS, fluorescent images were acquired by Axio Imager M1 microscope (Carl Zeiss) equipped with a CCD camera. Acquired images were analyzed with ImageJ (NIH, Bethesda, MD, USA) for green fluorescence. At least 100 cells per sample were analyzed. Cells were grown on glass coverslips and processed for electron microscopy, as previously described [70]. Briefly, samples were fixed with 2% formaldehyde (from paraformaldehyde) and 1% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4, at 4 °C. The subsequent steps, including osmium post-fixation, UranyLess (Electron Microscopy Science, Foster City, CA, USA) contrast and ethanol dehydration, were performed on ice, in the dark. Complete dehydration in 100% ethanol and the subsequent steps were carried out at room temperature. Cells were infiltrated with a 1:1 mixture of anhydrous ethanol and epoxy embedding medium (Sigma-Aldrich™, Cat# 45359-1EA-F, Burlington, MA, USA), then in pure resin for 90 min. The excess resin was gently removed, prior to polymerization at 60 °C for 72 h. This delicate procedure allowed us to readily identify the cell boundaries at FIB/SEM, facilitating the milling process and the cross-sectioning of the sample. Resin-embedded coverslips were secured to stubs using an adhesive carbon disc and made conductive with a thin layer of gold by a K550 sputter coater (Emithech, Kent, UK). The resulting samples were analyzed with a Dual Beam (FIB/SEM) Helios Nanolab 600 (FEI Company, Hillsboro, OR, USA) at the electron microscopy interdepartmental facility (LIME), at Roma Tre University. The cells were longitudinally cut by the ion beam operated at a voltage of 30 KV and a current of 6.5 nA. The resulting cross-sections were examined by the SEM column, scanning the backscattered electrons using a 2 KV voltage and a current of 0.17 nA.

PMC10002457

Cezary Krasnodębski,Agnieszka Sawuła,Urszula Kaźmierczak,Magdalena Żuk

Oligo—Not Only for Silencing: Overlooked Potential for Multidirectional Action in Plants

24-02-2023

oligo technology,plant ASO,DNA methylation,gene modulation,epigenetics,gene silencing,gene up/downregulation,non-GMO method,GMO alternative

Oligo technology is a low-cost and easy-to-implement method for direct manipulation of gene activity. The major advantage of this method is that gene expression can be changed without requiring stable transformation. Oligo technology is mainly used for animal cells. However, the use of oligos in plants seems to be even easier. The oligo effect could be similar to that induced by endogenous miRNAs. In general, the action of exogenously introduced nucleic acids (Oligo) can be divided into a direct interaction with nucleic acids (genomic DNA, hnRNA, transcript) and an indirect interaction via the induction of processes regulating gene expression (at the transcriptional and translational levels) involving regulatory proteins using endogenous cellular mechanisms. Presumed mechanisms of oligonucleotides’ action in plant cells (including differences from animal cells) are described in this review. Basic principles of oligo action in plants that allow bidirectional changes in gene activity and even those that lead to heritable epigenetic changes in gene expression are presented. The effect of oligos is related to the target sequence at which they are directed. This paper also compares different delivery methods and provides a quick guide to using IT tools to help design oligonucleotides.

Oligo—Not Only for Silencing: Overlooked Potential for Multidirectional Action in Plants Oligo technology is a low-cost and easy-to-implement method for direct manipulation of gene activity. The major advantage of this method is that gene expression can be changed without requiring stable transformation. Oligo technology is mainly used for animal cells. However, the use of oligos in plants seems to be even easier. The oligo effect could be similar to that induced by endogenous miRNAs. In general, the action of exogenously introduced nucleic acids (Oligo) can be divided into a direct interaction with nucleic acids (genomic DNA, hnRNA, transcript) and an indirect interaction via the induction of processes regulating gene expression (at the transcriptional and translational levels) involving regulatory proteins using endogenous cellular mechanisms. Presumed mechanisms of oligonucleotides’ action in plant cells (including differences from animal cells) are described in this review. Basic principles of oligo action in plants that allow bidirectional changes in gene activity and even those that lead to heritable epigenetic changes in gene expression are presented. The effect of oligos is related to the target sequence at which they are directed. This paper also compares different delivery methods and provides a quick guide to using IT tools to help design oligonucleotides. Oligo technology, based on short oligonucleotide sequences introduced into cells to modulate gene expression, is a burgeoning alternative for some aspects of genetic engineering in plants, but it is also a new tool that opens up some unique opportunities for gene control. Oligo technology is the result of exploiting elementary biochemical properties of nucleic acids and naturally occurring gene control mechanisms based on short RNA molecules—RNA oligonucleotides. Genes for such molecules are found in viruses and in cells of both prokaryotic and eukaryotic organisms, where their products exert important regulatory functions. The mechanisms of action exploited by oligo technology may therefore be similar to those triggered by endogenous sequences. The most important and crucial role of naturally occurring RNA oligonucleotides in cells (especially miRNA) is the tight regulation of gene expression in the context of RNA interference (RNAi). miRNA genes affect the expression of other genes in a manner that depends on the sequence of the mature oligonucleotide they encode. Apart from controlling expression, RNA oligonucleotides are critical to the basic function of DNA replication by serving as primers for DNA polymers. Genes regulating oligonucleotides are abundant in the vast majority of eukaryotic genomes. They play a key role in homeostasis by establishing the balance between transcripts and proteins in plants and animals. In both kingdoms, miRNA genes are highly conserved and play critical roles in gene regulation [1,2]. The importance of oligonucleotides in cellular functions is underscored by the fact that the last common ancestor of plants and animals already possessed basic components of the miRNA regulatory pathway [3]. The RNA-dependent regulatory machinery of the ancestor of the eukaryotes was composed of proteins from prokaryotes, phages, and archaea. Before these proteins were adapted for the new task of regulating gene expression, their roles appeared to be RNA processing, DNA repair, and protection from foreign or pathogenic genetic material [3,4]. This is also the reason why the RNA-mediated mechanisms of expression control of these two kingdoms show a high degree of similarity. Many researchers believe that the emergence of organisms with highly organized bodies would have been impossible without the sophisticated gene control by miRNA genes, as the explosive evolution of morphological innovations in organisms is highly correlated with the accumulation of new RNA genes that tightly control expression. However, over millions of years of evolution, plants have developed some additional mechanisms of action (mostly related to DNA methylation) or have begun to use homologous proteins for other purposes. DNA oligonucleotides, which occur naturally in cells, are short-lived byproducts of the digestion of longer DNA molecules belonging to a killed pathogen or to apoptotic or necrotic cells of one’s own body. Underestimated by nature, DNA oligonucleotides have become revolutionary to humans as tools of genetic engineering and biochemistry. While it is possible to adapt synthesized RNA molecules in oligo technology, such an approach is far-reaching and problematic because RNA is easily degraded (especially in the cell). The very similar structure of DNA oligonucleotides compared to RNA molecules used naturally by cells makes it possible to use DNA oligos as a low-cost substitute. This takes advantage of the fact that they are readily available, cheap to produce, easy to purify, stable, and already used in numerous applications: as primers in PCR reactions, fluorescent and radioactive probes, or drugs. Nucleic acids (both DNA and RNA) have been used for vector-free modulation of genes (their repair—mutagenesis or modification of their activity) practically since the beginning of the development of molecular biology in both animal and plant organisms. An example of such application is chimeric RNA-DNA oligonucleotides, which are directed to a specific sequence and can cause specific point mutations in this target sequence [5]. Similar techniques have also been used to modify plant genes for example, two teams have independently induced point mutations in the acetolactate synthase gene of tobacco [6,7]. Such technologies for using oligonucleotide sequences 40–200 nt in length to generate small (point) changes/mutations in the sequence of target genes are referred to as oligonucleotide-directed mutagenesis (ODM) and can be successfully used in plants [8,9]. In this work, we focus on a slightly different technology that uses oligonucleotides to induce changes in the activity of plant genes. The main difference between the Oligo technology described here and the ODM techniques is that in the case of the former we do not cause changes in the sequence of the modulated genes, and the expected effect is achieved by inducing epigenetic changes. Moreover, this technology uses much shorter 18–22 nt deoxynucleotides. Exogenous oligonucleotides (20–22 nt), introduced into the cell in the form of single-stranded nucleic acid fragments, are treated by the cellular machinery similarly to those derived from viruses or from the degradation of the transcript by polymerase II. The technology of using oligonucleotide sequences, mostly in antisense orientation (AOS) in mammalian cells, has become the basis of advanced gene therapy techniques for many difficult-to-treat diseases [10]. Its firmly established status has been confirmed by numerous FDA-approved oligo treatments in humans. Basically, the effect of oligos is not only bidirectional but also affects gene expression at each stage via different types of proteins in the internal regulatory system of the cell, which allows modulation of this process at multiple levels [11,12,13,14,15,16]. In plants, the effects of oligos on the steps of transcription, splicing, translation and DNA methylation are the best documented. There is a constant search for new tools for genetic engineering and control of gene expression in organisms, both for basic research and for the development of genetically modified organisms in medicine, agriculture and industry. One of the methods increasingly used for this purpose is oligo treatment. Of particular note is the possibility of DNA methylation modulation by oligos, which leads to heritable changes in gene expression without creating GMO plants [12,17]. Oligo-induced changes are stable and show similar traits to the reference transgenic plants, but without altering the genome sequence [13]. Therefore, oligos provide new tools for plant improvement through noninvasive epigenetic modulation. Due to the mode of action of oligos, similar to small RNAs, sequence-selective inhibition or enhancement of gene expression enables the elucidation of complex gene expression, especially gene functions and regulatory elements [18,19]. Most importantly, oligos enable the study of vital genes, which is virtually impossible using the classical method of gene knockdown. Gene silencing via RNAi and siRNA is also used for this purpose. Treatment with oligos does not require tedious construction preparation and plant transformation, whereas shRNA and artificial miRNA must be inserted into a plasmid before they can be introduced into the cell, which is difficult and time-consuming [20,21,22]. Furthermore, this means that the degree of inhibition of gene expression depends on the level of expression of the plasmid in the cell. A similar problem occurs when generating mutants with overexpression. Oligos, on the other hand, are dose-dependent and can be used to either increase or decrease expression, depending on the level of expression [23]. At the same time, pleiotropic effects were minimized, which is a common problem when generating mutants by genetic transformation [24,25]. The effects observed after transformation may not only be caused by gene silencing but may also depend on changes in genome sequences and structure caused by the insertion. Unfortunately, this may also lead to altered expression of other genes. Apart from that, the external addition of oligos enables the study of genes at different stages of plant development and allows experiments to be conducted over time. Therefore, primary and compensatory effects can be distinguished [17]. Regulatory proteins can also be targets of oligos to alter the expression of specific genes or even entire signaling and metabolic pathways. This provides tremendous flexibility in studying gene function and its global impact on hormone balances. Although small interfering RNAs (siRNAs) can also be delivered directly into the cell, their design and synthesis are more complicated and expensive. Oligos, unlike siRNAs, do not need to be fully complementary to exert an effect [26]. They allow for triggering a change at SNP sites, but because they are inaccurately designed with respect to genome sequences, oligos can lead to expression defects in nontarget genes. In addition, homologous sequences can be targeted with a single oligo, meaning that a single oligo can inhibit more than one gene from the same gene family [27]. Important in the context of the applicability of oligo technology is the ability to target regulatory proteins, which allows the study and modulation of entire signaling pathways and the study of the influence of individual factors on cell function. In general, the action of exogenously introduced nucleic acid sequences (oligo) can be divided into direct interaction with endogenous nucleic acids (genomic DNA, hnRNA or transcript) and indirect interaction via induction of processes regulating gene expression (at transcriptional and translational levels) involving regulatory proteins using endogenous cellular mechanisms [28]. A diagram showing the putative mechanisms of action of oligonucleotides on gene activity can be found in Figure 1. It is usually assumed (mainly on the basis of tests on animal cells) that the introduction of short oligodeoxynucleotides into the cell leads to hybridization of oligos with the homologous region in the transcript sequence (direct interaction). This event activates RNA-dependent RNA polymerase (RdRP) and leads to the formation of double-stranded RNA (dsRNA) that may be cleaved by DICER-like proteins (DCL) and can be integrated into AGO4 or AGO6 proteins, which might be responsible for gene repression as an effect of RNA interference (RNAi) or can also activate particular genes in the process called RNA activation (RNAa). The probable mechanism whereby this occurs is via RNase H1 or RISC pathways that reduce gene expression (this process, mainly using animal cells as an example, was excellently described in [29]), but can also increase gene expression through modulation of splicing or translation or stability by protecting AU-rich element (ARE) of mRNA [30]. The RNase H pathway is unique to DNA oligo as opposed to the natural mechanisms induced by gene-regulating RNA molecules. It was also demonstrated that antisense oligonucleotides can reduce mRNA levels by acting through the no-go decay pathway. It is also possible that the degradation of polyA is induced by the presence of the complementary oligo. This would lead to rapid digestion of an unstable (due to lack of polyA) transcript. Splicing of pre-mRNA, a dynamic process in which introns are removed and exons are joined, is governed by a combinatorial system exerted by overlapping cis-elements unique to each exon and its flanking intronic sequences. Oligonucleotides (mainly antisense oligos (ASOs)) can block splicing cis-elements of splicing and/or affect RNA structure and modulate splicing in vivo [31]. The next step in the expression of the gene is the step of translation. It has been shown that after base pairing with the target RNAs, oligonucleotides can recruit RNase H1 to cleave the RNA substrate within the region complementary to the oligo. RNase H1, which is expressed at low levels in all cells and localized in both the nucleus and cytoplasm, seems to be a limiting factor with respect to ASO-mediated antisense activity [15,32]. Oligonucleotides can efficiently reduce the levels of both nuclear and cytoplasmic RNAs. Results achieved on animal cells show that many oligos can rapidly reduce levels of cytoplasmic mature mRNAs without affecting the levels of nuclear pre-mRNAs [15]. There is no confirmation of this phenomenon occurring in plants so far. RNase H1-dependent oligonucleotides can trigger rapid degradation of mRNAs in the cytoplasm, where most mRNAs are translated under normal conditions. It is therefore possible that oligos can act on translating mRNAs. In such cases, the activity of oligos may be affected by the translating ribosomes. Translating mRNAs are being rapidly scanned by one or more ribosomes per mRNA. Scanning ribosomes can actually remove ASOs from the mRNA before RNase H1 is recruited, resulting in altered ASO activity [14]. The activity of oligonucleotides on mRNAs should depend on several rates: the on/off rate of oligo binding, the rate of RNase H1 recruitment, the rate of RNase H1 cleavage, and the rate of translation. Additionally, formation of oligonucleotide-target hairpin complex involves some type of triple-stranded structure with noncanonical interaction (Hoogsteen hydrogen bonds), therefore leading to more specific recognition and a higher affinity of the bond. The above mechanisms are described for both animal and plant cells [33,34]. Other mechanisms could be responsible for oligo activity, including direct binding to genomic double-stranded DNA and triplex DNA generation. This conjecture is based on our original finding that oligo affects genes and the methylation status of the genome [12,13,17]. Triple helix-forming oligonucleotides can compete with the binding of transcription factors and affect transcription initiation or elongation [35]. Recently, selected CCGG motif has been shown to be differentially methylated in response to plant treatment with oligo [12]. For the time being, there is no clear explanation for this. One hypothesis states that there is a distinction between complexes containing antisense or sense-oriented oligos, with the biggest difference between molecules of both types being the strength and stability of the effect on gene expression. Further, it has been found that there is a difference between complexes containing antisense or sense-oriented oligo. Not only may sense or antisense oligo induce opposite effects on gene expression (up- or downregulation), but can also lead to different strengths of expression changes. As for mechanisms operating on the genome level (epigenetic modulation) in plants, it seems that the mechanism tends to be based on DNA methylation, whereas in mammalian cells it is associated with histone modification [36]. DNA methylation is considered to be one of the most important epigenetic marks in plants [36]. Attachment of a methyl group in the 5‘ position of the DNA cytosine (5-mC) and, as was recently indicated, 6’ position of the DNA adenosine (6mA) are important epigenetic modifications in plants [37]. The process of addition or deprivation of a methyl residue is based on the recruitment of the methyltransferases—in plants by RdRP or polymerase IV and polymerase V complex [38]. Changes in nucleic acid methylation state have been reported to have a robust impact on plant phenotype. It has been shown that transgenic rice (Oryza sativa) and potato (Solanum tuberosum) with altered methylation gave 50% higher yield, biomass and increased resistance to drought stress [39]. Research on flax plants shows that it is possible to obtain both repression and overexpression of the target gene using oligo technology. The mechanism behind overexpression that has been proven is the induction of changes in genomic DNA methylation (most frequently CG sites but also CHG, CHH) leading to the reorganization of the chromatin structure via shift of the nucleosome position [12]. Recently, the possibility of inheriting oligo-induced epigenetic changes and thus changes in gene activity for at least 3 generations has been demonstrated [13,17]. By inducing epigenetic changes (mainly changes in genomic DNA methylation), it was possible to produce a new variety of linseed—Silesia [40]. Tests for distinctness, uniformity and stability (DUS) and value for cultivation and use (VCU) were carried out in the experimental stations of the Research Center for Cultivar Testing (Polish name COBORU). The result was the registration of the first epigenetically modulated variety in the National List and the granting of the breeder’s right for this variety (the status was obtained in March 2020) and would subsequently lead to its registration in the Common Catalog EC. The described experiments were carried out on flax, and it seems very likely that obtaining similar effects for other plant species is possible. The strongest effect of stabilization of changes was obtained via oligos that caused changes in methylation of certain regions of the genome. Thus, it appears that inducing changes in methylation, a key epigenetic mechanism for plants, may be a guarantee of permanent, inherited changes. As mentioned above, the use of oligonucleotides in plant research can be considered to be at an early stage. There is much more knowledge and experience in the field of using oligonucleotides in mammalian research or therapy. This is also true for a technical topic like oligonucleotide design. Homologues of plant epigenetic factors can be identified in animal organisms. In eukaryotes, high concordance of epigenetic mechanisms and their evolutionary conservation can be observed [41]. This implies the possibility of adapting the rules for the evolution of oligonucleotides acting on mammals for the evolution of oligonucleotides for plants. A 15-nucleotide antisense fragment is long enough to make a specific association with the target mRNA sequence. The optimal length of an oligonucleotide is 15–20 [23] or 15–30 [42] nucleotides. A critical feature defining the performance of the oligo is its complementarity with the target fragment. There are conflicting reports on the specificity of the oligo (with respect to the target sequence). Xie et al. [43] state that even two-point mutations can limit the effect of the oligo. This possibility is also confirmed by the results obtained with the gene CAB [11]. However, other results concerning modification of two very homologous isoforms of the CHS protein in flax [12,13] do not confirm this. One of the most important conditions for the efficiency of an oligodeoxynucleotide is that it can hybridize with the target fragment of the nucleic acid [23]. Therefore, the most important task in oligo design is to define the secondary structure of the target nucleic acid as precisely as possible. The appropriate fragment is free of structures that prevent steric hybridization [23,44]. Currently, a major facilitator in this design aspect is the increasing availability of fully sequenced genomes and the ability to use them as input data in IT tools [27]. Certain secondary structures may favor hybridization of oligonucleotides. End-strand fragments, single-stranded loop fragments of harpins, and common sequences longer than 10 nucleotide beads can be good targets [44,45,46]. Base composition should also be considered when designing antisense oligonucleotides. CCAC, TCCC, ACTC, GCCA, and CTCT are motifs whose presence correlates positively with oligonucleotide efficacy. In turn, the motifs GGGG, ACTG, AAA, and TAA are characterized by a negative correlation [47]. Higher efficiency is obtained with ASOs containing at least 11 G or C per 20 nucleotides [48]. IT tools come in handy when it comes to designing antisense oligonucleotides. Mfold, currently accessible from the UNAFold (http://www.unafold.org, accessed on 5 December 2022) domain, developed in 2003 [49], enabled prediction of all optimal or close to optimal structures of a given mRNA using minimum free energy (MFE) prediction algorithm. Mfold’s input is a clear mRNA sequence or in FASTA format as well as parameters such as folding temperature, ionic conditions and constraints of pairing particular bases. It is also possible to determine graphic and file format properties. During query creation, access to documentation is available. The obtained output includes a folding energy dot plot, several predictions of structures, their graphical representation and thermodynamic details. Another program available on the web is Sfold (https://sfold.wadsworth.org, accessed on 1 December 2022), which uses other mathematical algorithms based on stochastic sampling in Boltzmann ensemble and clustering to obtain the best predicted secondary structure of mRNA [50]. Furthermore, hybridization probability of antisense oligonucleotide in specific sites of predicted structure is calculated. Sfold facilitates selecting optimal oligo considering empirical rules like GC content and avoidance of GGGG motifs. Outputs can be displayed in graphical or text format and contain information about predicted binding energy. Therefore, the use of both programs is critical for the optimal design of oligos, giving the chance to achieve the best effect [44]. Let flax (Linum usitatissimum L.) chalcone synthase (Phytozome database, CHS1—Lus10031622, 1194 bp) transcript analyzed by mFold and sFold be an example output for these IT tools. Supplementary Files S1 and S2 shows respectively, energy dot plot and one of proposed folding (deltaG = −371 kcal/mol) of CHS1 mRNA generated using mFold 2.3 with default parameters. SFold output (also default parameters) Supplementary File S3 provides a probability profile of CHS1 displaying predicted accessible sites on the target RNA. It is also possible to receive a probability profile of a specific target fragment. Accessible sites can be targeted by several oligos. It is important to select optimal ones. Text output for antisense oligos is a representation of proposed oligos covering the target sequence iterated by one base and their determined parameters. Figure 2 presents a fragment of filtered output of antisense oligos for CHS1. Selected oligos meet the conditions of binding energy (oligo with a lover binding energy is preferable) and rules like GC content (40–60%) or GGGG motif absence. Cellular factors involved in determining mRNA accessibility are still not well understood. Despite continuous improvement and the possibility of applying comparative methods using several available algorithms, precise prediction of secondary structures remains troubling [45,50]. Therefore, experimental validation of the effectiveness of the designed oligo is essential [12]. In order to avoid aspecific interactions, it is recommended to perform homology analysis on a target organism’s mRNAs using available databases. Great attention should be paid to the use of appropriate controls to prove that the effects (changes in expression) result from the impact of a specific oligo on the target [23]. Effect of the oligo—overexpression or suppression—depends strongly on the target sequence to which it is directed. It is possible to distinguish specific target sites for antisense oligonucleotide activity such as exons, 3′ and 5′ UTR regions, and promoter regions: 100% of oligos targeting the “core promoter,” a region located ∼40 bp upstream of the transcription initiation site, result in upregulation of the gene [12]. Upstream of the core promoter region are the proximal and distal regions of the promoters (up to 1000–1500 bp upstream of the transcription initiation site). Proximal and distal regions of the promoter contain various regulatory sequences such as enhancers, silencers, insulators, and cis-elements that contribute to the fine regulation of gene expression at the transcriptional level. Alignment of the oligonucleotide to the 3′ end can lead to cleavage of the poly-A tail and consequent degradation of the mRNA. Negative regulatory elements may be present in the 5′UTRs of some mRNAs. Work at the 5’ end or at the translation initiation site can result in hindering the assembly of the translation system [46]. It has also been shown (in mammalian cell cultures) that translation can be enhanced by antisense oligonucleotides (ASOs) targeting upstream open reading frames. CHS gene study shows that oligonucleotides targeting the 5’UTR (very close to the start of translation) cause overexpression [12]. Immature mRNA can serve as a target for antisense oligonucleotides. Targeting the splicing site may prevent the correct assembly of the mRNA [51]. An oligo targeting exons can have a differential effect, mostly repression (67% of OLIGO analyzed), but also slight (less than in the case of targeting promoter sequences) gene activation or no effect, as oligo targeting intron sequences repression [12,13]. One of the major challenges associated with the use of antisense oligo is its stabilization, as the natural molecule of oligo–deoxyribonucleic acid is subject to rapid endonucleolytic and exonucleolytic degradation once introduced into the cell. Oligo modifications are much more popular in experiments with animal and mammalian cells, but especially in human therapy. To prolong the operation and availability of oligos, chemical modifications are made to the molecules. Modified oligos are expected to be eliminated from tissues more slowly, to increase the percentage of drug in tissues, and to be lower in dose and/or administered less frequently. The stability of ODNs can be increased by thiophosphorylation (PTO), in which the nonbridging oxygen atom in the phosphorothioate backbone of oligo is replaced with sulfur. The PTO modification is particularly popular because it provides sufficient stabilization against nucleolytic degradation while allowing recognition by RNase H [23]. In the second common modification—MOE—a change occurs at the 2’-position of the ribose units (2’-methoxyethoxy), resulting in greater nuclease resistance and higher binding affinity. Unlike PTO, MOE oligonucleotides are not subject to RNase H action. Therefore, they act by blocking the translational machinery [23,52]. In experiments with plants treated with modified oligo (PTO or methylation of all cytosines in sequence), significantly higher stability of epigenetic changes was observed in successive generations of generative plants. However, a pronounced induction of genes involved in phenylpropanoid metabolism (which was not targeted by the oligo) was observed in plants treated with oligo PTO. This could indicate a defensive response of the organism due to the potential toxicity of the exogenous factors used. No effect was observed with methylated oligo (less stable than PTO) [13]. Moreover, progeny obtained from seeds of plants treated with oligo PTO were significantly less vigorous and produced very few seeds of their own. Oligo technology is less commonly applied to plant cells than to animal cells, but it has already been shown to work in several species: flax, barley, tobacco, tea and recently cucumber and potato [10,11,12,14,15,17,34,53]. This technique was first successfully applied in plant cells to alter the expression of the gene encoding the transcription factor SUSIBA2 [24]. The researchers’ results showed that antisense oligonucleotides were efficiently transported within the leaf and reached the nucleus and chloroplasts [11,24]. In general, plant cells are more susceptible to oligo treatment than animal cells. This is mainly due to the positively charged cell wall, which is not a barrier for the negatively charged oligonucleotide molecules. Moreover, oligonucleotides can enter plant cells through channels specific to sugar molecules [24]. The exact mechanism of import into a plant cell has not been fully described. Oligonucleotides can be introduced into plant cells in several ways: infiltration under reduced pressure, infiltration through the stomata, spraying of cells, forced osmosis, and biolistic particle delivery system (gene gun) [54]. Depending on the plant species, tissue type and age (developmental stage), different delivery methods are most suitable. For details, please see Table 1. Oligos are usually designed for a specific target. Therefore, appropriate validation allows determining which one works “best,” according to our expectations. Based on proper design of oligos, we can expect that they will act in a certain way on the target gene or sequences [12,16]. Nevertheless, not all aspects of oligos have been fully examined or explained yet. Therefore, it is necessary to determine at which stage of the gene expression oligo acts (e.g., transcription or translation) and the direction of those changes (down or up). For this purpose, it is recommended to check both the target transcript (e.g., by qPCR), and if it results in a protein product, also their value (e.g., by Western blot). In some cases, the enzymatic activity or the level of final metabolites are also determined [11,17,55]. Moreover, fluorescence labeling of oligos enables the determination of their subcellular localization. It is especially important when the target sequence is located outside the nucleus, e.g., in chloroplasts [11], which allows determining whether the oligos have reached their destination. In plants, oligos have been shown affect DNA methylation, which can modulate the expression of the target gene. However, oligo usage may also lead to methylation modulation throughout the genome. The resulting epigenetic changes can be stable for up to 3 generations. Therefore, the total degree and profile of methylation would be worth checking with particular regard to the target gene [13]. The advent and development of methods in genetic modification of organisms have revolutionized science, medicine, pharmaceuticals, agriculture, and the chemical industry. To date, seven antisense therapeutics have received regulatory approval and dozens are in clinical development [18,56,57,58,59]. One of the FDA drugs—milases—is a patient-tailored antisense oligonucleotide drug for Batten disease. A successful example of this therapy may indicate that antisense oligonucleotides are useful in developing individualized treatment [60]. The discovery of the principles of epigenetics and the technologies resulting from new knowledge in this field are opening up innovative ways to better adapt living organisms to human needs. GMO technology, which is heavily used in plant biotechnology, has difficulty commercializing GM plants and products due to community disapproval and regulatory issues. Oligo technology faces the challenge of being a GM alternative by modulating the genome rather than altering it. The oligo method undoubtedly has great potential benefits and is also attractive because of its low cost, high efficiency [22], and the flexibility associated with the possibility of adapting the application method/system/mode to the characteristics of the target plant. A great advantage of oligo technology is also the possibility of using it for both dicotyledonous and monocotyledonous plants, which cause difficulties with the classic transformation with Agrobacterium. Most studies using oligonucleotides (mainly antisense) are performed to obtain transient transformants and to study gene function. To date, there has been little attempt to stabilize the changes induced by oligonucleotides. As far as we know, there is only one case in which a crop variety was generated using oligo technology [13,40]. However, there is a chance that by developing the knowledge of stabilizing epigenetic changes in plants, the changes in gene activity induced in this way will be inherited. This will make it possible to use this method for crop improvement as an alternative to genetic modification. The use of this method is further facilitated by the development of accompanying fields such as genome sequencing and in silico techniques (IT tools). The diversity and multistep nature of the mechanisms of action of oligonucleotides and the differentiation of effects in direction and strength offer a wide range of potential applications. Reports of inheritance of epigenetic changes induced by the action of oligos promise efficient practical application of this method in areas such as agriculture, protection of crops from infections, and improvement of plant adaptability to stress and climate change [34,53]. The use of oligonucleotides in research should not be underestimated. Their application in plant tissues can open the way for high-throughput screening for gene function [61]. There is still much to be explored in the field of oligonucleotide use, especially with regard to application in plants. The key challenge appears to be minimizing the occurrence of unanticipated effects and more accurately predicting the changes caused by the oligo used. Oligo technology offers many advantages while not requiring a large investment of time and money.

PMC10002460

Prakash P. Pillai,Muthukumar Kannan,Susmita Sil,Seema Singh,Annadurai Thangaraj,Ernest T. Chivero,Raghubendra Singh Dagur,Ashutosh Tripathi,Guoku Hu,Palsamy Periyasamy,Shilpa Buch

Involvement of lncRNA TUG1 in HIV-1 Tat-Induced Astrocyte Senescence

22-02-2023

aging,astrocyte senescence,HIV-1 Tat,lncRNA TUG1,proinflammatory cytokines

HIV-1 infection in the era of combined antiretroviral therapy has been associated with premature aging. Among the various features of HIV-1 associated neurocognitive disorders, astrocyte senescence has been surmised as a potential cause contributing to HIV-1-induced brain aging and neurocognitive impairments. Recently, lncRNAs have also been implicated to play essential roles in the onset of cellular senescence. Herein, using human primary astrocytes (HPAs), we investigated the role of lncRNA TUG1 in HIV-1 Tat-mediated onset of astrocyte senescence. We found that HPAs exposed to HIV-1 Tat resulted in significant upregulation of lncRNA TUG1 expression that was accompanied by elevated expression of p16 and p21, respectively. Additionally, HIV-1 Tat-exposed HPAs demonstrated increased expression of senescence-associated (SA) markers—SA-β-galactosidase (SA-β-gal) activity and SA-heterochromatin foci—cell-cycle arrest, and increased production of reactive oxygen species and proinflammatory cytokines. Intriguingly, gene silencing of lncRNA TUG1 in HPAs also reversed HIV-1 Tat-induced upregulation of p21, p16, SA-β gal activity, cellular activation, and proinflammatory cytokines. Furthermore, increased expression of astrocytic p16 and p21, lncRNA TUG1, and proinflammatory cytokines were observed in the prefrontal cortices of HIV-1 transgenic rats, thereby suggesting the occurrence of senescence activation in vivo. Overall, our data indicate that HIV-1 Tat-induced astrocyte senescence involves the lncRNA TUG1 and could serve as a potential therapeutic target for dampening accelerated aging associated with HIV-1/HIV-1 proteins.

Involvement of lncRNA TUG1 in HIV-1 Tat-Induced Astrocyte Senescence HIV-1 infection in the era of combined antiretroviral therapy has been associated with premature aging. Among the various features of HIV-1 associated neurocognitive disorders, astrocyte senescence has been surmised as a potential cause contributing to HIV-1-induced brain aging and neurocognitive impairments. Recently, lncRNAs have also been implicated to play essential roles in the onset of cellular senescence. Herein, using human primary astrocytes (HPAs), we investigated the role of lncRNA TUG1 in HIV-1 Tat-mediated onset of astrocyte senescence. We found that HPAs exposed to HIV-1 Tat resulted in significant upregulation of lncRNA TUG1 expression that was accompanied by elevated expression of p16 and p21, respectively. Additionally, HIV-1 Tat-exposed HPAs demonstrated increased expression of senescence-associated (SA) markers—SA-β-galactosidase (SA-β-gal) activity and SA-heterochromatin foci—cell-cycle arrest, and increased production of reactive oxygen species and proinflammatory cytokines. Intriguingly, gene silencing of lncRNA TUG1 in HPAs also reversed HIV-1 Tat-induced upregulation of p21, p16, SA-β gal activity, cellular activation, and proinflammatory cytokines. Furthermore, increased expression of astrocytic p16 and p21, lncRNA TUG1, and proinflammatory cytokines were observed in the prefrontal cortices of HIV-1 transgenic rats, thereby suggesting the occurrence of senescence activation in vivo. Overall, our data indicate that HIV-1 Tat-induced astrocyte senescence involves the lncRNA TUG1 and could serve as a potential therapeutic target for dampening accelerated aging associated with HIV-1/HIV-1 proteins. Although the advent of combination antiretroviral therapy (cART) has normalized the life expectancy of people living with HIV-1 (PLWH) to that of uninfected individuals, there is also an increased prevalence of HIV-1-associated neurocognitive disorders (HAND) and related comorbidities as these individuals live longer lifespans [1,2,3]. Adding further complexity to this is the preponderance of age-related mental health comorbidities of the elderly afflicting PLWH, ultimately manifesting as premature aging [4,5,6,7]. Studies have demonstrated the accumulation of senescent cells in the aging brain [8]. The contribution of HIV-1-associated mediators to the induction of cellular senescence has been demonstrated in mesenchymal stem cells [9], corneal endothelial cells [10], CD8+ T-cells [11], CD4+ T-cells [12], microglia [4,13], and astrocytes [14]. Additionally, life-long dependence on cART as well as dependence on illicit drugs such as methamphetamine have also been implicated in inducing astrocyte senescence [14,15]. Owing to the fact that despite cART, HIV-1 proteins such as transactivator of transcription (Tat) and gp120 continue to persist in tissues such as the brain and lymph nodes [16,17,18], we sought to understand the role of HIV-1 Tat in mediating astrocyte senescence. HIV-1 Tat is an early viral protein released by HIV-1-infected cells such as microglia, macrophages, and astrocytes and crosses the blood–brain barrier (BBB) from the periphery into the central nervous system (CNS), where it exerts is cytotoxicity directly by acting on glial and neuronal cells [19,20,21,22,23,24]. HIV-1 Tat-mediated cytotoxicity further leads to the release of soluble factors underlying neuroinflammation, oxidative stress, and excitotoxicity, ultimately resulting in neuronal damage [16,25]. Astrocytes, one of the most abundant glial cell populations in the CNS, perform key roles involving synapse formation and function, the release and uptake of neurotransmitters, the production of neurotrophic factors, the control of neuronal survival, which regulate BBB integrity and contribute to immunity within the CNS [14,26,27]. An impairment of astrocytes thus plays a significant role in the onset and progression of neurodegenerative diseases [28,29]. Even though HIV-1-infected astrocytes make up a small fraction of the infected cells within the CNS, these glial cells still represent a critical population that exhibits increased sensitivity to inflammatory triggers [30,31], ultimately resulting in neuronal impairment associated with HAND. In addition to active HIV-1 infection, viral proteins such as HIV-1 Tat and gp120 can also modulate astrocyte dysfunction and trigger inflammatory responses [32]. The accumulation of senescent astrocytes has also been demonstrated in Alzheimer’s disease [33] and is associated with exposure of the host to environmental factors leading to the development of Parkinson’s disease [34]. HIV-1 infection has been reported to induce macrophage and microglia senescence involving alterations in senescence-associated β-galactosidase (SA-β-gal) activity, p21 levels, and the production of cytokines such as IL6 and IL8, all of which contribute to HAND [4,13]. There is, however, a paucity of knowledge on the role of HIV-1 infection or HIV-1 proteins in driving astrocyte senescence leading to accelerated aging. Recently, various epigenetic modifications, including DNA methylation, histone modifications, and dysregulated expression of noncoding RNAs that accumulate across the lifespan, have been implicated in predicting several age-related complications such as cellular senescence [35,36,37,38,39]. While there is evidence of accelerated aging based on epigenetic data in PLWH [38,40], the role of noncoding RNAs in the context of HIV-1 protein(s)-mediated cellular senescence remains elusive. Long noncoding RNAs (lncRNAs) that are poorly conserved but abundant heterogeneous regulatory noncoding RNAs regulate gene expression at multiple levels, including transcription, RNA processing, translation, and post-translation. It has also been demonstrated that numerous lncRNAs mediate cellular senescence in various stages of the cell cycle by modulating senescence-associated pathways, such as p53/p21, pRB/p16, and p14 [41,42,43,44,45]. LncRNA Taurine Upregulated Gene 1 (TUG1) is one of the novel lncRNAs that is primarily expressed in retinal and brain tissues [46,47,48]. LncRNA TUG1, along with other lncRNAs, have been shown to regulate the cell cycle, which is specifically increased during aging [49]. LncRNA TUG1 has also been shown to disrupt the expression of the Homeobox (HOX) gene family, such as HOXB7, thereby leading to an aging phenotype [50]. LncRNA TUG1 is highly expressed in the human subependymal zone of the brain and is involved in age-related neurodegenerative diseases such as ischemic stroke and Huntington’s disease [49,51,52,53]. LncRNA TUG1 has an impact on tissue-specific aging such as intervertebral disk and age-related cataract involving the Wnt/β-catenin/caspase pathways [54,55]. The expression of upregulated lncRNA TUG1 has been shown to promote the proliferation of esophageal squamous cells and plays a role in promoting the proliferation of non-small-cell lung carcinoma [56]. Despite these diverse effects, the potential role of lncRNA TUG1 in astrocyte proliferation/senescence has not been examined to date. In this study, we determined the role of the lncRNA TUG1 in the onset of HIV-1 Tat-induced astrocyte senescence. Our findings implicate the role of lncRNA TUG1 as a novel regulator of astrocyte senescence that could be targeted as a therapeutic option for ameliorating HIV-1 Tat-mediated accelerated aging in the context of HAND. To determine whether exposure to HIV-1 Tat induces astrocyte senescence, HPAs were exposed to varying doses of HIV-1 Tat (25, 50, and 100 ng/mL) for 24 h, following which the expression levels of cellular senescence markers such as p21Waf1 and p16Ink4a were determined using western blotting. As shown in Figure 1, exposure of HPAs to HIV-1 Tat significantly (p < 0.05) increased the expression levels of p21 (Figure 1A) and p16 (Figure 1B) in a dose-dependent manner. Chronic exposure of HPAs to HIV-1 Tat (50 ng/mL) for seven days (HIV-1 Tat was added to the cells daily at the same time) also significantly (p < 0.05) increased the expression levels of p16 (Figure 1C) and p21 (Figure 1D). The concentration of 50 ng/mL of HIV-1 Tat was thus chosen for the subsequent experiments. Previous reports have demonstrated the circulating levels of HIV-1 Tat in the cerebrospinal fluid of HIV-1-infected patients to range from 1 to 40 ng/mL [16,25,57], and this is speculated to be even higher, especially in the proximity of HIV-1-positive perivascular cells. We next performed the senescence-associated-β-galactosidase (SA-β-gal) staining in HPAs exposed to HIV-1 Tat (50 ng/mL; 7 days), and as shown in Figure 1E,F, the SA-β-gal staining intensity was notably elevated in HIV-1 Tat-exposed HPAs compared with control cells. The exposure of HPAs to HIV-1 Tat (50 ng/mL) also mediated cell cycle arrest at the G0/G1 phase as determined by flow cytometry. As shown in Figure 1G, exposure of HPAs to HIV-1 Tat increased the accumulation of cells in the G1 phase (from 60.76% to 72.83%) with a concomitant decrease in the S phase of cells (from 16.87% to 9.70%), thereby suggesting cell cycle arrest at the G0/G1 phase during astrocyte senescence induced by HIV-1 Tat. In these studies, exposure of cells to hydrogen peroxide (H2O2) served as a positive control. In addition, we also demonstrated that exposure of HPAs to HIV-1 Tat (50 ng/mL; 7 days) failed to induce cell death in our experimental paradigm (Figure 1H). HIV-1 Tat-mediated astrocyte senescence was further characterized by increased formation of senescence-associated heterochromatin foci (SAHF) as shown by punctated DAPI staining (Figure 1I). Moreover, the percentage of cells containing SAHF-positive foci was significantly increased from day 2 to day 7 in HIV-1 Tat-exposed HPAs compared with the control group (Figure 1J). Next, we sought to examine whether exposure of HPAs to HIV-1 Tat induced the generation of ROS production using a 2′,7′-dichlorodihydrofluorescein diacetate (H2DCF DA) fluorescence assay. As shown in Figure 1K, exposure of HPAs to HIV-1 Tat for 7 days increased ROS production compared with control cells. H2O2 was used as a positive control (Figure 1K). Next, we determined the senescence-associated secretory phenotypes (SASPs) in HPAs exposed to HIV-1 Tat (50 ng/mL; for 7 days) by determining the expression levels of proinflammatory cytokines such as TNF and IL6 mRNAs using both qPCR and ELISA. As shown in Figure 1L,M, HPAs exposed to HIV-1 Tat demonstrated a significant (p < 0.05) increase in the mRNA and protein expression of proinflammatory cytokines, such as IL1β, IL6, and TNF-α compared with control cells. To further confirm whether exposure of HPAs to HIV-1 Tat could impact cellular activation, HPAs were exposed to HIV-1 Tat (50 ng/mL; 7 days) and assessed for the expression of glial fibrillary acidic protein (GFAP) using western blotting. As shown in Figure 1N, HPAs exposed to HIV-1 Tat demonstrated significantly (p < 0.05) increased expression of GFAP compared with control cells. To understand the mechanisms underlying HIV-1 Tat-induced astrocyte senescence in vitro, we next sought to assess the involvement of lncRNAs in cellular senescence and aging. Herein, we monitored the expression levels of known senescence-associated lncRNAs, such as MIAT, MEG3, MALAT1, ANRIL, PLUTO, HOTAIR, H19, lncRNA p21, XIST, GAS5, and TUG1 in HIV-1 Tat-induced astrocyte senescence. As shown in Figure 2A, HPAs exposed to HIV-1 Tat (50 ng/mL) significantly (p < 0.05) downregulated the expression of all the lncRNAs except for lncRNA TUG1, which was significantly upregulated compared with control cells. Since lncRNA TUG1 has been shown to play a role in age-related neurodegenerative diseases and cellular senescence, we further sought to determine the expression levels of lncRNA TUG1 in HIV-1 Tat- (50 ng/mL for 2, 5, and 7 days) exposed HPAs in vitro. As shown in Figure 2B, we found a significant upregulation of lncRNA TUG1 in HPAs exposed to HIV-1 Tat. In addition, the expression of lncRNA TUG1 was significantly (p < 0.05) upregulated in HPAs exposed to H2O2 with no significant changes in HPAs exposed to heat-inactivated HIV-1 Tat (Figure 2C). We also wanted to determine the contribution of upregulated lncRNA TUG1 in HIV-1 Tat-induced astrocyte activation. For this, we used the gene silencing approach by knocking down lncRNA TUG1 in HPAs, followed by exposing cells to HIV-1 Tat (50 ng/mL; 2 days). Figure 2D demonstrates the gene silencing efficiency. We next sought to determine the astrocyte activation in this experimental setup, and as shown in Figure 2E,F, in HPAs silenced for the lncRNA TUG1 expression, there was a significant (p < 0.05) abrogation of HIV-1 Tat-mediated upregulation of GFAP mRNA (Figure 2E) and protein expression (Figure 2F) within 2 days. Next, we sought to determine the contribution of lncRNA TUG1 in HIV-1 Tat-induced astrocyte senescence and increased expression of proinflammatory cytokines in lncRNA TUG1 silenced in HPAs exposed to HIV-1 Tat (50 ng/mL; 2 days). As shown in Figure 3A,B, transfection of HPAs with lncRNA TUG1 siRNA, but not the scrambled siRNA, significantly (p < 0.05) downregulated HIV-1 Tat-induced expression of the senescence markers such as p21 (Figure 3A) and p16 (Figure 3B). As expected, in HPAs transfected with scrambled siRNA, HIV-1 Tat exposure resulted in increased production of ROS and increased numbers of SA-β-gal activity compared with the control cells, whereas, silencing of lncRNA TUG1 in HPAs resulted in a significant (p < 0.05) reduction in both ROS levels (Figure 3C) and the number of SA-β-gal-positive cells (Figure 3D,E). Furthermore, gene silencing of lncRNA TUG1 in HPAs also significantly (p < 0.05) abrogated the expression levels of proinflammatory cytokines, such as IL1β (Figure 3F), IL6 (Figure 3G) and TNFα (Figure 3H) mRNAs compared with control cells. Similarly, and as expected, the protein expression levels of the proinflammatory cytokines (Figure 3I–K) were also normalized in lncRNA TUG1-silenced, HIV-1 Tat-exposed HPAs. Having shown HIV-1 Tat-mediated astrocyte senescence in vitro, we next sought to validate our findings in vivo using HIV-1 Tg rats. For this, 15-month-old HIV-1 Tg rats and age-matched wild-type rats were assessed for the expression of p21 and p16 in the prefrontal cortex. As shown in Figure 4A,B, we performed immunostaining for the expression of p21 and p16 that colocalize with GFAP in these groups of rats and found significant colocalization of p21 (Figure 4A,B) and p16 with GFAP (Figure 4C,D) in the prefrontal cortices of HIV-1 Tg rats compared to that of wild-type rats. Additionally, the expression of lncRNA TUG1 as well as the proinflammatory cytokines, such as IL1β, IL6 and TNFα (both mRNA and protein levels) were monitored in the prefrontal cortices of HIV-1 Tg and wild-type rats using qPCR and ELISA. As expected, the expression of lncRNA TUG1 (Figure 4E) and the mRNA and protein expression levels of proinflammatory cytokines (Figure 4F,G) were significantly (p < 0.05) increased in the prefrontal cortices of HIV-1 Tg rats compared with the wild-type rats. With the availability of effective cART, HIV-1 has transformed from a death sentence to a manageable chronic disease, with many individuals living longer and healthier lives. In fact, with the right treatment regimen and care, the life expectancy of those infected with HIV-1 can be comparable to uninfected healthy individuals. However, there is ample evidence implicating that PLWH have a higher risk of developing certain age-related illnesses [2,58]. As stated above, chronic infection with HIV-1, coupled with early initiation and dependence on cART and drug abuse, results in a slow-smoldering inflammatory milieu in the brain that accumulates over time, culminating into premature aging and neurodegeneration. In the HIV-1-infected, cART-treated, drug-abusing population, these neurodegenerative, cognitive impairments are seen at a much younger age compared with normal healthy (uninfected and drug-naïve) counterparts [59,60]. Some of the likely mediators contributing to HIV-1-associated premature aging include residual viral proteins such as HIV Tat and gp120 and long-term toxicity of the antiretrovirals, among others [61,62,63]. In the current study, we demonstrate that exposure of HPAs to HIV-1 Tat upregulated the expression of lncRNA TUG1, which, in turn, correlated with increased SA-β-gal activity, p16, and p21 levels, the formation of SAHF, and the acquisition of a SASP in astrocytes. Interestingly, our findings also demonstrated that gene silencing of lncRNA TUG1 was able to reverse HIV-1 Tat-induced astrocyte senescence, thereby suggesting a possible involvement of the lncRNA TUG1/p16/p21 axis in HIV-1 Tat-mediated astrocyte senescence. Our previously published reports using HIV-1 Tg rats and HPAs underpinned the role of the HIV-1 Tat protein in inducing a set of microRNAs that played a critical role in astrogliosis with implications in aging [64]. Our findings are comparable to earlier studies implicating the role of HIV-1 infection or the HIV-1 Tat protein and other inducers, such as drugs of abuse (methamphetamine) or antiviral drugs, as contributors of astrocyte senescence [14,15] with the acquisition of classical senescence markers along with the increased expression of proinflammatory cytokines and oxidative stress. It is well recognized that only cells with stable cell cycle arrest are considered senescent and that the pathways underlying this process are driven by cyclin-dependent kinase inhibitors such as p21 and p16 [65]. Our study also demonstrated that increased expression of p21 and p16 proteins in HPAs exposed to HIV-1 Tat plays a critical role in the induction of stable cell cycle arrest in these cells. Additionally, we also demonstrate that in the brains of the HIV-1 Tg rats, there is an increased proportion of astrocytes expressing p21 and p16 in the prefrontal cortices compared with the wild-type rats, thus suggesting increased activation of senescence and cell cycle dysregulation. These observations are consistent with current reports showing that there is an increased expression of p21 and p16 in post-mortem brain tissues of Amyotrophic Lateral Sclerosis/Motor Neurone Disease [66]. Similar observations were also reported in the brains of individuals with Alzheimer’s disease [33]. We also showed an increased number of astrocytes in the G0/G1 phase of the cell cycle following HIV-1 Tat exposure. These observations further confirm impaired cell-cycle regulation in the HIV-1 Tat-exposed HPAs. Senescent cells are morphologically characterized by an enlarged size and flattened shape compared with the non-senescent control cells. The senescent cells also exhibit elevated levels of ROS, increased lysosomal contents, and compromised lysosomal activity, ultimately resulting in increased levels of β-galactosidase [67,68]. Data from the current in vitro astrocyte senescence model confirm these metabolic and morphological changes. A characteristic feature of senescent cells is extensive chromatin reorganization resulting in the formation of SAHF [69,70]; nuclei of senescent cells typically contain 30–50 bright, punctate, DNA-stained, dense foci that can be readily distinguished from chromatin in normal cells. The SAHFs play a role in sequestering proliferation-promoting genes that are required for the progression through the S-phase of the cell cycle [71]. As shown in our study, there is increased DAPI puncta representing SAHFs in senescent astrocytes by day 5 in vitro, which reached more than 50% by day 7 in vitro. Many senescent cells acquire a SASP that comprises a highly complex mixture of secreted cytokines, chemokines, growth factors, and proteases. The chronic inflammation induced by the sustained presence of senescent astrocytes leads to tissue damage and degeneration in the aged brain. Here, we show increased mRNA expression of proinflammatory cytokines following HIV-1 Tat exposure. IL6 is one of the prominent cytokines of the SASP, and our data also indicate a significant increase in the gene expression levels of IL6. This increased proinflammatory cytokine gene expression confirms the involvement of immune activation in astrocyte senescence. Our data are in line with the earlier reports wherein astrocytes were demonstrated to be an important source of the generation of proinflammatory mediators in response to HIV-1/HIV-1 proteins [19,30,72,73]. The astrocytes are not only the producers of neuroinflammation but are also known to be highly susceptible to both the viral proteins and cellular toxins released from HIV-1-infected macrophages/microglia. Upon activation, the astrocytes undergo astrogliosis that is characterized by cellular proliferation and further release of toxic mediators. There are also reports suggesting age-related upregulation of GFAP expression as well as astrogliosis in different brain regions [74,75,76,77]. Here, we show increased gene expression of GFAP mRNA in astrocytes following continuous exposure to HIV-1 Tat, suggesting the involvement of glial activation—one of the key phenotypic features of the senescing astrocytes. Recent advances in high-throughput sequencing and RNA profiling technologies have facilitated the identification of an enormous number of lncRNAs and their critical roles in diverse biological processes including cellular senescence and aging [78,79]. In addition, emerging evidence indicates that aberrant expression of lncRNAs is associated with neuronal aging, cellular senescence, and tumors, suggesting their involvement in regulating cell cycle events [80,81]. lncRNA TUG1 has been linked with promoting cell proliferation [82,83,84,85], suggesting tissue- and cell-type-specific lncRNA functions. In particular, a study conducted by Zhang et al. (2014) showed that lncRNA TUG1 is a direct target of the tumor suppressor p53 and has a role in repressing specific genes involved in cell-cycle regulation [56]. In the present study, we have determined the expression levels of lncRNAs that are known to be associated with cellular senescence, including lncRNA TUG1 in HPAs exposed to HIV-1 Tat. Our results demonstrate that out of several lncRNAs that were dysregulated, lncRNA TUG1 expression was significantly upregulated in HPAs exposed to HIV-1 Tat. Similar to findings in the present study, others have also reported downregulated lncRNA TUG1 expression in glioma and non-small-cell lung cancer [54,56]. In summary, HIV-1 Tat increases SA-β-gal activity, p21 and p16 expression, cell cycle arrest, ROS, the formation of SAHFs, and the expression of proinflammatory cytokines and lncRNA TUG1 in HPAs. Intriguingly, silencing of lncRNA TUG1 retards most of these detrimental effects of HIV-1 Tat, such as the appearance of the proinflammatory phenotype as well as the increase in p21 and p16 expression (Figure 5). Our findings linking lncRNA TUG1 to astrocyte senescence mediated by HIV-1 Tat is unique and previously not reported. Overall, our findings enrich the understanding of how lncRNA TUG1 may potentially drive cellular senescence in HPAs and initiate the accelerated aging process. These observations altogether suggest that lncRNA TUG1 could be considered as a therapeutic target to control accelerated or premature aging in PLWH. Endotoxin-free recombinant HIV-1 Tat-101 (1032–10) was purchased from ImmunoDX, LLC (Woburn, MA, USA). P16-INK4A (Cat No. 10883-1-AP) antibody was purchased from Proteintech Group, Inc. (Rosemont, IL, USA). P21 antibody (Cat No. ab109199) and GFAP (Cat No. ab18258) were purchased from Abcam, Boston, MA, USA. β-Actin (Cat No. sc-47778 HRP) antibody was purchased from Santa Cruz Biotechnology, Inc., Dallas, TX, USA. Senescence β-galactosidase staining kit (Cat No. 9860S) was purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). Hydrogen Peroxide (Cat No. H325-500) was purchased from Thermo Fisher Scientific, Inc. (Pittsburgh, PA, USA). Peroxidase-AffiniPure Goat Anti-Rabbit IgG (H + L) (Cat No. 111–035-003) and Peroxidase-conjugated AffiniPure Goat Anti-Mouse IgG (H + L) (Cat No. 115–035-003) was purchased from Jackson ImmunoResearch Inc. (West Grove, PA, USA). Image-IT™ LIVE Green Reactive Oxygen Species Detection Kit (Cat No. I36007), Lipofectamine™ RNAiMAX transfection reagent (Cat No. 13778150), Opti-MEM® I Reduced Serum Media (Cat No. 31985070), goat anti-rabbit IgG (H + L) cross-adsorbed secondary antibody, Alexa Fluor 488 (Cat No. A-11008), goat anti-mouse IgG (H + L) cross-adsorbed secondary antibody, Alexa Fluor 594 (Cat No. A-11005), and ProLong® Gold Antifade Mountant with DAPI (Cat No. P36935) were purchased from ThermoFisher Scientific, Inc. (Pittsburgh, PA, USA). HIV-1 Tg rats (Harlon, Inc., Indianapolis, IN, USA; 15-month-old male rats) and age- and background-matched controls (F344) were used in this study (N = 3/group). All animal procedures were performed according to the protocols approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Nebraska Medical Center (20-057-07-FC; Approved on 17 August 2020). Animals were euthanized and the prefrontal cortex was dissected and immediately snap-frozen and stored at −80 °C until used for various assays. HPAs were derived from the single-cell isolation process of fetal brain tissues. Briefly, human fetal brain tissues were collected from electively aborted specimens (gestational age 12–16 weeks) after the abortion procedure was completed in collaboration with the Birth Defects Research laboratory at the University of Washington. The protocol complies with all applicable state and federal requirements. All individuals provided informed consent using an Institutional Review Board-approved consent form at the University of Washington. For experiments, HPAs were cultured and appropriately treated in 6-well plates. HPAs with 60–80% confluent (0.8–1.0 × 106 cells/well) in a 6-well plate for about 24 h after seeding, were treated with both lncRNA TUG1 siRNA (Sense: 5′-GGCCAACUUUGACAAAUCUUU-3′; Antisense: 5′-AGAUUUGUCAAAGUUGGCCUU-3′) and scrambled control siRNA (Sense: 5′-UUCUCCGAACGUGUCACGUUU-3′; Antisense: 5′-ACGUGACACGUUCGGAGAAUU-3′), as per standard protocols and reported previously [86]. Briefly, HPAs were transfected with targeted siRNA (20 pM) mixed with 2 μL of Lipofectamine™ RNAiMAX transfection reagent diluted in 100 μL of Opti-MEM® I Reduced Serum Media. The resulting siRNA–lipid complexes were added to the HPAs and incubated for 6 h. Next, the medium was changed to 10% FBS-containing DMEM for 20 h incubation. The transfected HPAs were then ready for further experiments. Cell-cycle status in the HPAs was determined by measuring nuclear DNA content. The cells were collected on days 2, 5, and 7 after treatment with HIV-1 Tat, centrifuged, and washed twice with ice-cold PBS. The cells were then fixed in 70% ethanol at 4 °C for 24 h followed staining by propidium iodide (50 µg/mL) and RNAse A (100 μg/mL). The samples were analyzed using a FAC Calibur flow cytometer (BD Bioscience, San Diego, CA, USA), and the flow cytometry data were analyzed using FlowJo™ v10.6 software (TreeStar Inc., Ashland, OR, USA). SA-β-gal staining was performed in HPAs treated with HIV-1 Tat for 7 days using the Senescence-β-galactosidase staining kit (Cell Signaling Technology, Inc., Danvers, MA, USA, Cat No. 9860S) as per the manufacturer’s instructions. ROS were detected in live cells using the Image-iT™ LIVE Green ROS Detection Kit according to the manufacturer’s instructions. Briefly, cells were seeded in 24-well plates followed by exposure to HIV-1 Tat, H2O2, or left untreated for up to 7 days. The cells were subsequently labeled with carboxy-H2DCFDA working solution (25 µM, 30 min at 37 °C), washed, mounted in the warm buffer, and imaged immediately on a Zeiss Observer using a Z1 inverted microscope (Carl Zeiss Microscopy, LLC. White Plains, NY, USA). HPAs treated with HIV-1 Tat or indicated chemicals for up to 7 days were subjected to SAHF quantification. SAHFs were measured by counting DAPI puncta following DAPI staining [69]. Briefly, control and treated HPAs were washed 3 times with ice-cold PBS and fixed with 4% PFA for 30 min. The fixed cells were then washed thrice with PBS and stained with DAPI (1 μg/mL; 10 min). After washing with PBS, the heterochromatin foci were imaged using a Z1 inverted microscope, and SAHF-positive cells were counted manually by independent blinded assessors. Cell viability was assessed using the MTT assay as per standard protocols and reported previously [87,88]. Briefly, HPAs were cultured in a 96-well plate (1 × 104 cells per well) for 24 h followed by exposure either to HIV-1 Tat, heat-inactivated HIV-1 Tat, or H2O2 for an additional 24 h and incubated with 20 μL MTT tetrazolium salt (5 mg/mL) dissolved in Hank’s balanced salt solution added to each well and incubated in a CO2 incubator for 4 h. Finally, the medium was suctioned from each well, and 200 μL of dimethyl sulfoxide was added to disperse the formazan crystals. The absorbance of each well was obtained using a plate counter at the test and reference wavelengths of 570 and 630 nm, respectively. Formalin-fixed, paraffin-embedded brain tissue sections of wild-type and HIV-1 Tg rats were baked at 60 °C overnight, followed by deparaffinization, rehydration, and antigen retrieval using the standard protocol. Then the slides were incubated with a blocking buffer containing 10% normal goat serum for 1 h at room temperature followed by the addition of primary antibodies, such as p21 (1:400), p16 (1:100), and GFAP (1:500), and incubated overnight at 4 °C. The next day, the sections were washed, followed by incubation with Alexa Fluor 488 goat anti-rabbit (1:500) and Alexa Fluor 555 goat anti-mouse (1:500), respectively, at room temperature for 2 h. The slides were then mounted, and fluorescent images were obtained using a Z1 inverted microscope (Carl Zeiss Microscopy, LLC, White Plains, NY, USA) for the colocalization analysis. Treated cells were lysed using the RIPA buffer (Cell Signaling Technology, Inc., Danvers, MA, USA, Cat No. 9806) and quantified using Pierce™ BCA Protein Assay Kit (ThermoFisher Scientific, Inc., Pittsburgh, PA, USA, Cat No. 23227). Equal amounts of proteins were electrophoresed and transferred on a polyvinylidene fluoride membrane (Millipore, Danvers, MA, USA, Cat No. IPVH00010). The membranes were blocked with 5% non-fat dry milk (in 1X TTBS buffer) and then probed overnight at 4 °C with primary antibodies, including p21 (1:2000), p16 (1:2000), GFAP (1:500), and β-actin (1:4000). After washing, appropriate secondary antibodies were added (1:5000) for 1 h at room temperature. Next, the blots were imaged and the data analyzed using ImageJ software [89]. Real-Time PCR experiments were performed as per the standard protocol as described elsewhere. In brief, total RNA was isolated using Quick-RNA™ MiniPrep Plus (Zymo Research, Orange, CA, USA, Cat No. R1058) and reverse-transcribed using Verso cDNA synthesis kit (ThermoFisher Scientific, Inc., Pittsburgh, PA, USA, Cat No. AB1453B), as per the manufacturer’s protocols. Comparative real-time PCR was performed using RT2 SYBR Green ROX qPCR Mastermix (Qiagen, Germantown, MD, USA, Cat No. 330523) in an Applied Biosystems® QuantStudio™ 3 Real-Time PCR System. The primers used in this study are shown in Table 1. Each PCR reaction was carried out in triplicate, and three independent experiments were run. GAPDH was used as a housekeeping control for the normalization, and the fold change in expression was obtained by the 2−ΔΔCT method. Data are presented as mean ± SEM. For comparison between each group, Student’s t-test or one-way analysis of variance (ANOVA) with Dunn’s post hoc test was used. GraphPad Prism version 6.01 (San Diego, CA, USA) was used for statistical analysis and the preparation of bar graphs. A probability less than 0.05 was considered statistically significant.

PMC10002478

Jingbo Chen,Jiawen Xu,Yan Sun,Yuhuan Xue,Yang Zhao,Dongzi Yang,Shuijie Li,Xiaomiao Zhao

Gut Microbiota Dysbiosis Ameliorates in LNK-Deficient Mouse Models with Obesity-Induced Insulin Resistance Improvement

22-02-2023

gut microbiota,LNK deficiency,obesity,insulin resistance

Purpose: To investigate the potential role of gut microbiota in obesity-induced insulin resistance (IR). Methods: Four-week-old male C57BL/6 wild-type mice (n = 6) and whole-body SH2 domain-containing adaptor protein (LNK)-deficient in C57BL/6 genetic backgrounds mice (n = 7) were fed with a high-fat diet (HFD, 60% calories from fat) for 16 weeks. The gut microbiota of 13 mice feces samples was analyzed by using a 16 s rRNA sequencing analysis. Results: The structure and composition of the gut microbiota community of WT mice were significantly different from those in the LNK-/- group. The abundance of the lipopolysaccharide (LPS)-producing genus Proteobacteria was increased in WT mice, while some short-chain fatty acid (SCFA)-producing genera in WT groups were significantly lower than in LNK-/- groups (p < 0.05). Conclusions: The structure and composition of the intestinal microbiota community of obese WT mice were significantly different from those in the LNK-/- group. The abnormality of the gut microbial structure and composition might interfere with glucolipid metabolism and exacerbate obesity-induced IR by increasing LPS-producing genera while reducing SCFA-producing probiotics.

Gut Microbiota Dysbiosis Ameliorates in LNK-Deficient Mouse Models with Obesity-Induced Insulin Resistance Improvement Purpose: To investigate the potential role of gut microbiota in obesity-induced insulin resistance (IR). Methods: Four-week-old male C57BL/6 wild-type mice (n = 6) and whole-body SH2 domain-containing adaptor protein (LNK)-deficient in C57BL/6 genetic backgrounds mice (n = 7) were fed with a high-fat diet (HFD, 60% calories from fat) for 16 weeks. The gut microbiota of 13 mice feces samples was analyzed by using a 16 s rRNA sequencing analysis. Results: The structure and composition of the gut microbiota community of WT mice were significantly different from those in the LNK-/- group. The abundance of the lipopolysaccharide (LPS)-producing genus Proteobacteria was increased in WT mice, while some short-chain fatty acid (SCFA)-producing genera in WT groups were significantly lower than in LNK-/- groups (p < 0.05). Conclusions: The structure and composition of the intestinal microbiota community of obese WT mice were significantly different from those in the LNK-/- group. The abnormality of the gut microbial structure and composition might interfere with glucolipid metabolism and exacerbate obesity-induced IR by increasing LPS-producing genera while reducing SCFA-producing probiotics. Obesity is becoming a worldwide health risk factor, and obesity-induced morbidity and complications account for huge costs for affected individuals, families, healthcare systems, and society at large. Obesity is a low-grade sustained inflammatory state that alters the whole-body metabolism that frequently leads to insulin resistance (IR) [1], which in turn plays a vital role in the pathogenesis of obesity-associated hyperlipidemia, non-alcoholic fatty liver disease, polycystic ovary syndrome, type 2 diabetes, and atherosclerotic cardiovascular disease [2]. Nutrients and substrates as well as systems involved in host–nutrient interactions, including gut microbiota, have been also identified as modulators of metabolic pathways controlling insulin action and obesity regulation [3]. However, the molecular mechanism of IR has not been exactly clarified. Gut microbiota is the general term for the microbes that inhabit the gastrointestinal tract of the human body. Around 98–99% of the intestinal microbiomes can be classified into four groups: Bacteroidetes, Firmicutes, Proteobacteria, and Actinomycetes. The balance of intestinal microbe species is the key to keeping the intestinal immune function normal and maintaining the homeostasis of the body. Breaking the balance will lead to serious pathophysiological changes, which is called gut microbiota dysbiosis [4]. Increasing studies showed that Bacteroides are associated with high-fat and high-protein diets [5] and the imbalance of intestinal microecology might be involved in the occurrence of many diseases, such as irritable bowel syndrome, obesity, type 2 diabetes, metabolic syndrome (MetS), and cardiovascular diseases [6,7,8]. Metagenomic sequencing and 16S RNA sequencing were used to detect the changes in intestinal microbiota in patients with prediabetes, type 2 diabetes, and MetS. Two studies found that although the races and their diets were different, in type 2 diabetes patients, the proportion of Clostridium butyrate-producing Roche fusobacterium and Clostridium leptum decreased while the proportion of non-Clostridium butyrate increased [9,10]. The levels of Firmicutes and Clostridia in the gut microbiota of type 2 diabetes patients were significantly decreased as compared to normal controls, and the ratio of Bacteroidetes to Firmicutes was increased and positively correlated with blood glucose concentrations [11]. There are changes in the intestinal microbiota in people with abnormal glucose metabolism, and the changes in the intestinal microbiota also seem to be involved in the occurrence and remission of abnormal glucose metabolism. It was reported that feces from mice with abnormal glucose metabolism transplanted into healthy germ-free mice could cause abnormal glucose metabolism [12]. Furthermore, transplanting feces from lean donors into patients with MetS could increase their gut microbiota diversity and insulin sensitivity [13]. The results above suggested that the intestinal microbiota are closely related to the occurrence and development of abnormal glucose metabolism, while IR, as an important link in the occurrence and development of abnormal glucose metabolism, also seems to be related to the intestinal microbiota. Our previous study discovered that ovarian tissues from PCOS patients with IR exhibited higher expression of the SH2 domain-containing adaptor protein (LNK) than ovaries from normal control subjects and PCOS patients without IR [14]. In addition, we found that there were more accumulated intrahepatic triglyceride, higher serum triglyceride (TG), and free fatty acid (FFA) in wild-type (WT) mice as compared to LNK-deficient (LNK-/-) mice fed with a high-fat diet (HFD). LNK deficiency improved glucose metabolism and IR in obese mice, suggesting the LNK might play a pivotal role in controlling glucolipid metabolism and obesity-induced IR by regulating IRS1/PI3K/Akt/AS160 signaling and the AKT/FOXO3 pathway [15,16]. Therefore, we chose LNK-/- mice as the IR-improved model and WT mice as the MetS/IR model. In this study, we compared intestinal microbiota of LNK-/- mice and WT mice that consumed HFD, with the aim to explore the potential influence of gut microbiomes on the glucolipid metabolic disorder and obesity-induced IR. The study protocol was approved by the Research Ethics Board of Sun Yat-sen memorial hospital of Sun Yat-sen University and Guangdong Provincial People’s Hospital. All the experimental procedures were approved by the Committee for Animal Research of Sun Yat-sen University and the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. Four-week-old male C57BL/6 wild-type mice (n = 6) were purchased from the animal research center of Sun Yat-sen University. Whole-body LNK-deficient in C57BL/6 genetic backgrounds mice (n = 7) were created via CRISPR/Cas mediated genome engineering by Cyagen Biosciences Inc. The mouse Sh2b3 gene (GenBank accession number: NM_001306127.1; Ensembl: ENSMUSG00000042594) is located on mouse chromosome 5. Exon 1 to exon 3 were selected as target sites. Cas9 mRNA and gRNA generated using an in vitro transcription were then injected into fertilized eggs for knockout mouse production. All mice were randomly divided into different groups, housed 4 to 5 per cage, with standard laboratory conditions (12 h light:12 h darkness cycle) at a controlled temperature (23 ± 2 °C) and free access to rodent feed and water. All mice (4–5 weeks old) were fed a high-fat diet (HFD, 60% calories from fat, D12492; Research Diets Inc., New Brunswick, NJ, USA) for 16 weeks. When mice were fed with a HFD for up to 16 weeks, fecal samples were collected and immediately kept frozen at −80 °C until processed for analysis. Total DNA was isolated from the fecal samples using the MasterPure Complete DNA&RNA Purification Kit (Epicenter) according to the manufacturer’s instructions with some modifications as described previously [17]. DNA was extracted using a DNA extraction kit for the corresponding sample. The concentration and purity were measured using the NanoDrop One (Thermo Fisher Scientific, Waltham, MA, USA). Next, 16S rRNA/18SrRNA/ITS genes of distinct regions (e.g., Bac 16S: V3-V4/V4/V4-V5; Fug 18S: V4/V5; ITS1/ITS2; Arc 16S: V4-V5 et al.) were amplified used specific primer (e.g., 16S: 338F and 806R/515F and 806R/515F and 907R; 18S: 528F and 706R/817F and 1196R; ITS5-1737F and ITS2-2043R/ITS3-F and ITS4R; Arc: Arch519F and Arch915R et al.) with a 12bp barcode. Primers were synthesized by Invitrogen (Invitrogen, Carlsbad, CA, USA). PCR reactions, containing 25 μL 2× Premix Taq (Takara Biotechnology, Dalian Co. Ltd., Dalian, China), 1 μL each primer (10 μM), and 3 μL DNA (20 ng/μL) template in a volume of 50 µL, were amplified via thermocycling: 5 min at 94 °C for initialization; 30 cycles of 30 s denaturation at 94 °C, 30 s annealing at 52 °C, and 30 s extension at 72 °C; followed by 10 min final elongation at 72 °C. The PCR instrument was BioRad S1000 (Bio-Rad Laboratory, Hercules, CA, USA). The length and concentration of the PCR product were detected via 1% agarose gel electrophoresis. Samples with the bright main strip between (e.g., 16S V4: 290–310 bp/16S V4V5: 400–450 bp et al.) could be used for further experiments. PCR products were mixed in equidensity ratios according to the GeneTools Analysis Software (Version 4.03.05.0, SynGene, Cambridge, UK). Then, the mixture of PCR products was purified with E.Z.N.A. Gel Extraction Kit (Omega, Bellevue, WA, USA). Next, sequencing libraries were generated using NEBNext® Ultra™ II DNA Library Prep Kit for Illumina® (New England Biolabs, Ipswich, MA, USA) following the manufacturer’s recommendations, and index codes were added. The library quality was assessed on the Qubit@ 2.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). At last, the library was sequenced on an Illumina Nova6000 platform and 250 bp paired-end reads were generated. Fastp (version 0.14.1) was used to control the quality of the raw data by sliding the window (-W 4 -M 20). The primers were removed by using cutadapt software according to the primer information at the beginning and end of the sequence to obtain the paired-end clean reads. Paired-end clean reads were merged using usearch -fastq_mergepairs (V10) according to the relationship of the overlap between the paired-end reads; when with at least a 16 bp overlap, the read generated from the opposite end of the same DNA fragment, the maximum mismatch allowed in the overlap region was 5 bp, and the spliced sequences were called Raw Tags. Fastp (version 0.14.1) was used to control the quality of the raw data by sliding the window (-W 4 -M 20) to obtain the paired-end clean tags. R software was used to count the union (pan) and intersection (core) of the target classification level in different samples to evaluate whether the sample size was sufficient. R software was used to analyze the common and endemic species, the composition of the community, and the richness of species. A total of 13 mice (7 LNK-/-mice and 6 WT mice) were included in this study. The average body weights of 0W LNK-/-mice and WT mice were 21 g ± 2.2 g and 21.1 g ± 2.1 g, respectively, with no significance (p > 0.05). During the process of the mice fed with HFD, we observed that LNK-/- mice had a loss of appetite compared with WT mice. The food intakes of LNK-/- mice and WT mice were 18.8 g ± 1.3 g and 19.4 g ± 1.8 g, respectively, with statistical significance (p < 0.05). After 16 weeks, there was a significant difference in body weight between LNK-/- mice (47.5 g ± 4.6 g) and WT mice (52.6 g ± 3.3 g) (p < 0.05). All thirteen feces samples from seven LNK-/-mice and six WT mice were analyzed. A majority of intestinal microbe species of LNK-/-mice and WT mice were similar, however, the diversity of gut microbiomes in the WT mice group was less than that of the LNK-/- mice group (Figure 1A). The α diversity of the gut microbiota calculated using the Shannon index showed that the LNK-/- group species diversity was higher than that of the WT group at the phylum level (p < 0.05, t test) (Figure 1B,C). To compare the composition difference in the intestinal microbiota between LNK-/- and WT mice, we next performed a Bray–Curtis-based principal coordinates analysis (PCoA) (Figure 2A). It was shown that the degree of similarity between the two groups of microbial communities was significantly different (Bray–Curtis PERMANOVA, p = 0.016). In addition, the composition of the microbiota in the samples of the LNK-/- group was more heterogeneous and significantly different from that of the WT group. The heat map showed that gut microbiota compositions between the LNK-/- and WT groups were markedly different (Figure 2B). In the phylum-level taxonomy classification, the WT group was dominated by Proteobacteria, Verrucomicrobia, and Bacteroidetes; the LNK-/- group was dominated by Bacteroidetes, Proteobacteria, and Firmicutes (Figure 2C). Although bacteria are similar at the phylum level between the two groups, Figure 2C showed that their proportion was different. The WT group was dominated by Proteobacteria and had a relative abundance of Verrucomicrobia, with the significance compared with LNK-/- mice (p < 0.05) (Figure 2D), while the LNK-/- group has a relatively large proportion of Firmicutes (p < 0.05) and Bacteroidetes (Figure 2D). According to the results of the linear discriminant analysis effect size (LEfSe) (LDA ≥ 2.0), the abundances of Proteobacteria, Helicobacteraceae, Epsilonproteobacteria, and Campylobacterales were significantly increased in WT mice, while the abundance of Erysipelotrichales, Allobaculum, and Bacteroidales was significantly increased in LNK-/- mice (Figure 2E). To explore the gut microbial differences between LNK-/- and WT mice further, we used STAMP software to analyze the genera with significant differences (p < 0.05). We found that the abundances of some short-chain fatty acid (SCFA)-producing genera in the WT groups were significantly lower than in the LNK-/- groups, such as Prevotella_9, Prevotellaceae_UCG-001, Clostridium_sensu_strict_1, Ruminococcaceae_UCG-010, and Stenotrophomonas (Figure 2F). Our previous study showed that upon the consumption of HFD, LNK-/- mice had a loss of appetite, and WT mice accumulated more intrahepatic triglyceride, TG, and FFA compared with LNK-/- mice. LNK plays a pivotal role in adipose glucose transport by regulating insulin-mediated IRS1/PI3K/Akt/AS160 signaling. In this study, we found that the abundance of Proteobacteria was significantly increased in the WT mice group, which was one of the main LPS-producing bacteria. Some SCFA-producing genera in WT groups were significantly lower than in the LNK-/- groups. LPS is also called endotoxin. The complex of LPS and its receptor CD14 can be recognized by Toll-like receptor 4 (TLR4) on the surface of immune cells to induce an inflammatory response. When the change in diet or the use of antibiotics affects the balance of gut microbiota, the number of harmful bacteria such as G- bacteria increases, and the decomposed product LPS passes into the blood circulation through the intestinal epithelium to cause endotoxemia, which triggers a systemic inflammatory response [18]. This study revealed that inflammation and LPS levels were elevated in patients with type 2 diabetes. Both animal and human experiments have demonstrated that the direct injection of LPS can increase fasting blood glucose and insulin levels, resulting in hyperinsulinemia and insulin resistance. When the number of G- bacteria decreased with the use of antibiotics, the amount of LPS entering the circulation decreased, which could relieve the systemic inflammation and increase insulin sensitivity. LPS receptor CD14 knockout mice fed a high-fat diet or injected with LPS had decreased inflammatory factors in adipose tissue, increased insulin sensitivity in liver and adipose tissue, and had a delayed development of insulin resistance, and their weight gain slowed down [19,20,21,22]. The results suggest that LPS plays an important role in the induction of the inflammatory response and insulin resistance. The intestinal microbiota may affect the content of circulating LPS in the following two ways to induce insulin resistance. For one thing, the structure of intestinal microbiota is unbalanced, the number of G + bacteria is decreased, the proportion of G- bacteria is increased, and the production of LPS is increased. Studies have shown that the number of G + bacteria such as Clostridium decreased and the number of LPS-containing bacteria such as Bacteroides and Proteobacteria increased in diabetic patients. Adding Lactobacillus and Bifidobacterium to the diet of high-fat-induced obese mice could help restore a balance between probiotics and pernicious bacteria in the gut and increase insulin sensitivity. The addition of prebiotic oligosaccharides to a high-fat diet-induced diabetic mouse model also increased the number of bifidobacteria, decreased the level of LPS, and improved insulin secretion and inflammation, which was significantly associated with the number of bifidobacteria [23]. Additionally, intestinal microbiota alter intestinal permeability. Studies have shown that a high-fat diet may interact with intestinal microbiota, alter intestinal permeability, promote the rise of LPS levels, and cause an inflammatory state and insulin resistance [24,25]. The intestinal microbiota selectively regulates the expression of colonic Cannabinoid receptor 1, which affects intestinal permeability by altering the distribution of Claudin-1 [26]. In addition, obesity itself affects intestinal permeability. A study of normal-weight and overweight healthy women showed a positive correlation between gut permeability and waist circumference and visceral fat content [27]. Increased visceral adipose promotes the secretion of the pro-inflammatory factors TNF α, IL-1, and IL-6 by infiltrating macrophages in adipose tissue and reducing the production of the anti-inflammatory factor adiponectin. With the action of multiple pro-inflammatory factors, intestinal mucus production was decreased, and intestinal permeability was increased. TNF-α can also act on tight junction proteins, resulting in the increased permeability of the tight junction of intestinal cells [28,29,30]. These proinflammatory factors can also promote insulin resistance and lipid storage in adipocytes, thereby forming a vicious cycle. Probiotics such as Bifidobacterium, Lactobacillus, and Prevotella_9 can promote the release of SCFAs from the undigested soluble dietary fiber in the colon via fermentation, at the same time reducing the intestinal pH, inhibiting the growth of harmful bacteria, to reduce the production of LPS in the intestinal lumen [31]. SCFAs can also promote the secretion of insulin by pancreatic β cells by regulating the secretion of gut-derived hormones, such as glucagon-like Peptide 1 (Glp-1), Glucagon Peptide 2 (Glp-2), Peptide YY (PYY), and glucose-dependent insulinotropic Peptide (GIP), etc., to increase insulin sensitivity and suppress appetite and food intake, thereby improving insulin resistance. After 8-week oral medication of VSL#3 probiotics containing 8 kinds of viable bacteria, the diet-induced obesity mice had increased GLP-1 production, decreased food intake, reduced body weight, and improved glucose tolerance. Their intestinal microbiota composition also changed the number of probiotics of Firmicutes such as lactobacillus, and Bifidobacterium increased, which was related to the increase in butyrate in SCFAs [32]. Butyrate can improve the function of the intestine, promote the activity of the intestine, and has a better therapeutic effect on patients with a loss of appetite, diarrhea, dyspepsia, and so on. In addition, butyrate can promote the reduction of dietary intake and digestion and is also beneficial to obese or fatty liver patients [32]. In addition, another study showed that healthy volunteers ate inulin-containing foods that promoted probiotic growth and a regular diet, respectively. Moreover, GLP-2 was found to be increased in fasting serum and decreased in intestinal permeability after eating inulin-containing foods [33]. The results demonstrated that probiotics could promote the production of SCFAs and the secretion of GLP-1 and Glp-2 by regulating the balance of intestinal microbiota, further improving intestinal permeability and alleviating IR. Our research explored the changes in the gut microbiota in LNK-/- and ET mice, which provided new ideas for the mechanism and treatment of MetS and IR. Although previous studies had shown that the disorder of intestine microbiota was related to MetS, the underlying mechanism remains unclear. Therefore, this study was a supplement to this research field. Nevertheless, this study still had some shortcomings. Firstly, as is known to all, sex hormones strongly influence body fat distribution and adipocyte differentiation. Estrogen and testosterone differentially affect adipocyte physiology and estrogens play a leading role in the causes and consequences of female obesity. Therefore, in this study, to avoid the influence of estrogen on the occurrence of obesity, we did not put male and female mice together to compare, and only collected fecal samples based on previous obesity-induced IR male mouse models. The sample size was not large enough, and there may be bias in the results for female mice. The results of female mice and the potential effects of sex hormones on gut microbiota need further research. Secondly, in the study, we focused on the difference in gut microbiota between LNK-/- and WT mice. We will continue relevant studies, and the indexes such as LPS, butyrate, gut permeability, and mucosal structural changes will be measured or observed in our next study. The relationship between changes in gut microbiomes and IR needs to be confirmed by further experiments. Finally, the 16S rRNA gene sequencing had some limitations, such as a short reading length, sequencing errors, and difficulty in evaluating and operating taxa. It would be important to combine signaling pathways and metabolomics analysis in the next step. Our research described that the structure and composition of the gut microbiota community between LNK-/- and WT mice were significantly different. The change in the gut microbial structure and composition of obese WT mice might aggravate glucolipid metabolic disorder and IR by increasing the production of LPS while reducing the production of SCFAs.

PMC10002481

Minjung Son,Ga Young Kim,Yejin Yang,Sugyeong Ha,Jeongwon Kim,Doyeon Kim,Hae Young Chung,Hyung Ryong Moon,Ki Wung Chung

PPAR Pan Agonist MHY2013 Alleviates Renal Fibrosis in a Mouse Model by Reducing Fibroblast Activation and Epithelial Inflammation

02-03-2023

kidney fibrosis,PPAR,inflammation,fibroblasts

The peroxisome proliferator-activated receptor (PPAR) nuclear receptor has been an interesting target for the treatment of chronic diseases. Although the efficacy of PPAR pan agonists in several metabolic diseases has been well studied, the effect of PPAR pan agonists on kidney fibrosis development has not been demonstrated. To evaluate the effect of the PPAR pan agonist MHY2013, a folic acid (FA)-induced in vivo kidney fibrosis model was used. MHY2013 treatment significantly controlled decline in kidney function, tubule dilation, and FA-induced kidney damage. The extent of fibrosis determined using biochemical and histological methods showed that MHY2013 effectively blocked the development of fibrosis. Pro-inflammatory responses, including cytokine and chemokine expression, inflammatory cell infiltration, and NF-κB activation, were all reduced with MHY2013 treatment. To demonstrate the anti-fibrotic and anti-inflammatory mechanisms of MHY2013, in vitro studies were conducted using NRK49F kidney fibroblasts and NRK52E kidney epithelial cells. In the NRK49F kidney fibroblasts, MHY2013 treatment significantly reduced TGF-β-induced fibroblast activation. The gene and protein expressions of collagen I and α-smooth muscle actin were significantly reduced with MHY2013 treatment. Using PPAR transfection, we found that PPARγ played a major role in blocking fibroblast activation. In addition, MHY2013 significantly reduced LPS-induced NF-κB activation and chemokine expression mainly through PPARβ activation. Taken together, our results suggest that administration of the PPAR pan agonist effectively prevented renal fibrosis in both in vitro and in vivo models of kidney fibrosis, implicating the therapeutic potential of PPAR agonists against chronic kidney diseases.

PPAR Pan Agonist MHY2013 Alleviates Renal Fibrosis in a Mouse Model by Reducing Fibroblast Activation and Epithelial Inflammation The peroxisome proliferator-activated receptor (PPAR) nuclear receptor has been an interesting target for the treatment of chronic diseases. Although the efficacy of PPAR pan agonists in several metabolic diseases has been well studied, the effect of PPAR pan agonists on kidney fibrosis development has not been demonstrated. To evaluate the effect of the PPAR pan agonist MHY2013, a folic acid (FA)-induced in vivo kidney fibrosis model was used. MHY2013 treatment significantly controlled decline in kidney function, tubule dilation, and FA-induced kidney damage. The extent of fibrosis determined using biochemical and histological methods showed that MHY2013 effectively blocked the development of fibrosis. Pro-inflammatory responses, including cytokine and chemokine expression, inflammatory cell infiltration, and NF-κB activation, were all reduced with MHY2013 treatment. To demonstrate the anti-fibrotic and anti-inflammatory mechanisms of MHY2013, in vitro studies were conducted using NRK49F kidney fibroblasts and NRK52E kidney epithelial cells. In the NRK49F kidney fibroblasts, MHY2013 treatment significantly reduced TGF-β-induced fibroblast activation. The gene and protein expressions of collagen I and α-smooth muscle actin were significantly reduced with MHY2013 treatment. Using PPAR transfection, we found that PPARγ played a major role in blocking fibroblast activation. In addition, MHY2013 significantly reduced LPS-induced NF-κB activation and chemokine expression mainly through PPARβ activation. Taken together, our results suggest that administration of the PPAR pan agonist effectively prevented renal fibrosis in both in vitro and in vivo models of kidney fibrosis, implicating the therapeutic potential of PPAR agonists against chronic kidney diseases. Chronic kidney disease (CKD) affects approximately 10% of the global population, with high mortality due to limited treatment options [1]. CKD often leads to end-stage renal disease, which is fatal without renal replacement therapy, such as dialysis or kidney transplantation. Kidney fibrosis is considered a major underlying pathological process that is commonly detected in CKD development [2]. Understanding the mechanisms of renal fibrosis is essential for developing therapies to prevent or slow CKD progression. Fibrosis is defined by the formation and accumulation of the extracellular matrix (ECM), mainly by tissue-resident fibroblast cells [3]. Under physiological conditions, minimal amounts of ECM support kidney structure and function. In response to tissue injury, wound-healing processes are activated to inhibit the inflammatory response with proper tissue regeneration. However, persistent inflammatory responses result in incomplete regeneration, with the formation of fibrotic scar tissue [4]. Exaggerated deposition of ECM during chronic and pathological fibrosis development disrupts the normal kidney architecture and interferes with kidney function. At a certain stage, unresolved kidney fibrosis becomes irreversible and contributes to renal failure. The mechanisms underlying the development of kidney fibrosis have been studied extensively [5]. Regardless of the trigger, multiple cell types participate in fibrogenesis, including fibroblasts, pericytes, epithelial cells, endothelial cells, and inflammatory cells [6]. The main contributor to fibrosis progression is the accumulation of fibroblasts with a phenotypic appearance of myofibroblasts. During progressive fibrosis, the interstitium is filled with myofibroblasts, which produce large amounts of ECM proteins [6]. Although myofibroblasts are the executing cells of fibrosis, other cells also contribute to the development of fibrosis through both direct and indirect mechanisms. Pericytes, epithelial cells, and endothelial cells have been shown to directly contribute to fibrosis through the transition to mesenchymal-like cell types [7]. Epithelial cells also contribute to fibrosis through the secretion of pro-fibrogenic and pro-inflammatory factors, such as TGF-β, CTGF, and cytokines [8,9]. Considerable evidence suggests that inflammatory cells play a critical role in the initiation and progression of renal fibrosis [10,11]. The chemokine is mainly secreted from tubule epithelial cells during injury and recruits various inflammatory cell types, including monocytes, T cells, dendritic cells, and fibrocytes [9]. The infiltration of inflammatory cells is a major phenotype of kidney fibrosis that promotes fibrosis [12]. Peroxisome proliferator-activated receptors (PPARs), PPARα, PPARβ/δ, and PPARγ, play an essential role in the regulation of various physiological processes, including lipid and energy metabolism [13]. Fibrates (PPARα agonists) are used to treat dyslipidemia, and thiazolidinediones (PPARγ agonists) are used to increase insulin sensitivity in type 2 diabetics. In addition, PPAR dual agonists have been developed to treat type 2 diabetes with secondary cardiovascular complications [14,15]. Many synthetic ligands for PPARs are still under development to expand their therapeutic applications. In addition to their original roles in metabolism, PPAR agonists have been shown to exert various physiological effects. PPAR agonists have been reported to block the development of fibrosis in the liver, heart, kidneys, and lungs [16,17]. Furthermore, several studies have reported the anti-inflammatory action of peroxisome proliferator-activated receptor (PPAR) agonists [18]. Previously, we synthesized and evaluated the role of MHY2013, a potent PPAR pan-agonist, in several metabolic disease models [19,20]. In addition, MHY2013 showed anti-fibrotic effects in an age-related renal fibrosis model by regulating the lipid metabolism in epithelial cells [21]. However, the effects of MHY2013 on general aspects of renal fibrosis have not yet been investigated. In this study, we demonstrated the role and efficacy of MHY2013 in a general renal fibrosis model. Using mouse models of renal fibrosis induced by folic acid, we demonstrated the anti-fibrotic efficacy of PPAR pan agonism in renal fibrosis. MHY2013 treatment significantly reduced fibrosis and inflammation in a mouse model of renal fibrosis. In addition, using in vitro analysis, we found anti-fibrotic and anti-inflammatory effects of MHY2013 in renal fibroblasts and epithelial cells. To evaluate the anti-fibrotic effects of MHY2013, folic-acid-induced renal fibrosis models were used. MHY2013 was intraperitoneally administered at a low (0.5 mg/kg/day) or high dose (3 mg/kg/day) during the experimental period (Figure 1A). The MHY2013-treated group showed lower expression of kidney damage-related genes (Havcr1, Timp2, Igfbp7, and Spp1) than those of the FA-treated group (Figure 1B). Blood urea nitrogen (BUN) levels were increased in the folic acid (FA)-treated group, and high-dose MHY2013 treatment significantly blocked the FA-induced BUN increase (Figure 1C). Structural changes were analyzed with hematoxylin and eosin (H&E) staining. Tubule dilation and damage were detected in the cortex and medulla regions of FA-treated kidneys (Figure 1D). MHY2013-treated kidneys showed a smaller increase in tubule dilation (Figure 1D). These results indicate that MHY2013 has protective effects against folic-acid-induced kidney damage. We further analyzed the effects of MHY2013 on the development of renal fibrosis. The MHY2013-treated group showed lower expression of fibrosis-related genes (Col1a2, Col3a1, Vim) than that of the FA-treated group (Figure 2A). The increased expression of Col1a2 and Vim was confirmed with in situ hybridization (ISH) analysis. FA treatment significantly increased Col1a2 and Vim expression in the interstitial region of the kidney, and MHY2013-treated groups showed lower Col1a2 and Vim expression (Figure 2B,C). The protein levels of fibrosis markers were further checked. FA-induced α-SMA and collagen I levels were significantly decreased with MHY2013 treatment (Figure 3A). An immunohistochemical analysis confirmed that fewer αSMA-positive myofibroblasts were detected in the MHY2013-treated kidneys (Figure 3B). The extent of fibrosis was confirmed using Sirius Red (SR) staining. The FA treatment significantly increased SR-positive regions, whereas the MHY2013 treatment reduced SR-positive regions (Figure 3C,D). Finally, the activation of SMAD proteins was detected. Less SMAD2 and SMAD3 phosphorylation was detected in the MHY2013-treated groups than in the FA groups (Figure 3E). Collectively, these data indicate that MHY2013 effectively blocked FA-induced kidney fibrosis. The development of fibrosis is accompanied by pro-inflammatory responses. FA treatment also increases the inflammatory responses in the kidneys [22]. We further examined the inflammatory responses in animal models. MHY2013 treatment significantly reduced pro-inflammatory gene (Tnfa, Il1b, and Ccl2) expression and the macrophage marker Emr1 in the kidneys (Figure 4A). The activation of NF-κB, induced by FA, was effectively blocked with MHY2013 treatment (Figure 4B). Activated NF-κB was mainly detected in the epithelial cells of dilated tubules, and MHY2013 significantly reduced p-NF-κB expression in tubule cells (Figure 4C). Macrophage infiltration was confirmed using ISH analysis. Increased Emr1 expression was mainly detected in the interstitial region of FA-treated kidneys (Figure 4D). MHY2013-treated groups showed less macrophage infiltration in the kidneys (Figure 4D). We further detected the co-expression of Col1a2 and Emr1. In the FA group, Emr1- and Col1a2-positive cells were colocalized in the kidneys, indicating that inflammation is connected to fibrosis development (Figure 4D). In accordance with the qPCR results, the MHY2013-treated group showed lower Emr1 and Col1a2 expression in the kidney (Figure 4D). These results indicate that MHY2013 exerts anti-inflammatory effects against FA-induced kidney fibrosis. To investigate the anti-fibrotic role of MHY2013 under in vitro conditions, we used kidney-derived fibroblast cells. First, we confirmed the activation of PPAR by MHY2013 in NRK49F kidney fibroblasts. MHY2013 significantly increased PPRE activity under PPARα, PPARβ, and PPARγ expression conditions, confirming MHY2013 as a PPAR pan agonist (Figure 5A–C). TGF-β treatment significantly increased Col1a2, Acta2, and Vim expression in NRK49F fibroblasts, and MHY2013 pre-treatment effectively blocked fibroblast activation (Figure 5D). The protein expression levels of α-SMA and Col1 were analyzed. MHY2013 treatment significantly reduced TGF-β-induced α-SMA and Col1 protein expression (Figure 5E). The increased expression of α-SMA was confirmed using immunofluorescence. TGF-β increased αSMA expression in cells, whereas MHY2013 reduced αSMA expression (Figure 5F). To examine which PPAR subtype influenced fibroblast activation, we overexpressed PPAR before TGF-β treatment. We found that PPARγ overexpression effectively blocked TGF-β-induced fibroblast activation (Figure 5G), whereas other PPAR subtypes did not show a significant reduction (data not shown). These results indicate that MHY2013 effectively blocks TGF-β-induced NRK49F kidney fibroblast activation, mainly through PPARγ activation. To examine the anti-inflammatory effects of MHY2013 under in vitro conditions, kidney tubule epithelial cells were used. Stimulation of NRK52E cells with a lipopolysaccharide (LPS) significantly increased chemokine gene expression, and MHY2013 pretreatment effectively reduced their expression (Figure 6A). We further evaluated NF-κB activity using a luciferase assay. LPS treatment significantly increased NF-κB activity, whereas MHY2013 effectively blocked NF-κB activity (Figure 6B). Finally, to examine which PPAR subtype influences LPS-induced chemokine expression, we overexpressed PPAR before LPS treatment. We found that PPARβ overexpression effectively blocked LPS-induced chemokine expression (Figure 6C), whereas other PPAR subtypes did not show a significant reduction (data not shown). Collectively, these data show that MHY2013 reduces LPS-induced NF-κB activation and chemokine expression in renal epithelial cells, mainly through PPARβ activation. Renal fibrosis, which is generally accompanied by CKD progression, is defined by the loss of renal parenchymal cells and their substitution with ECM proteins. During fibrosis development, both the synthesis and degradation of ECM proteins occur via several intra- and extracellular events. When ECM protein synthesis exceeds degradation, excessive ECM accumulation results in fibrosis [23]. It is well established that various cell types directly and indirectly participate in fibrosis development. Resident fibroblasts are the main responsible cells for the synthesis of ECM proteins [3]. During fibrogenesis, fibroblasts receive signals from other cells and begin to proliferate and become myofibroblasts. Myofibroblasts produce large amounts of ECM proteins that primarily contribute to the pathogenesis of kidney fibrosis. Transforming growth factor-β (TGF-β) is considered a key player of renal fibrosis by stimulating fibroblasts in the kidney, thus making it an interesting target for the treatment of fibrosis [24]. Indeed, anti-TGF-β treatments using neutralizing antibodies, inhibitors against the TGF-β receptor, or antisense oligonucleotides to TGF-β1 halt the progression of renal fibrosis development, suggesting its fibrotic role in CKD [25]. We found that MHY2013 significantly reduced TGF-β-induced fibroblast activation in vitro. MHY2013 effectively inhibits TGF-β-induced α-SMA and collagen I expression in fibroblasts. Several studies have reported that PPARγ activation blocks TGF-β-induced ECM production in fibroblasts. Wang et al. evaluated three PPARγ agonists (15d-PGJ2, troglitazone, and ciglitazone) and found that PPARγ activation directly inhibits TGF-β/SMAD signaling pathways and alleviates renal fibroblast activation, resulting in reduced ECM synthesis [26]. Another PPARγ agonist, pioglitazone, similarly prevents renal fibrosis by repressing the TGF-β signaling pathway [27]. MHY2013 also showed direct anti-fibrotic effects on fibroblasts. Using PPAR transfection, we found that PPARγ overexpression inhibits TGF-β-induced fibroblast activation. Based on these results, we concluded that MHY2013 directly reduces fibroblast activation through PPARγ activation. Renal inflammation is a protective response induced during kidney injury, which eliminates the cause of injury and promotes tissue repair. However, unresolved inflammatory responses can promote abnormal fibrosis in the kidneys, leading to CKD [28]. During prolonged inflammation, bone-marrow-derived leukocytes, including neutrophils and macrophages, are the main players in kidney inflammation. The accumulation of these cells is a major feature of pro-inflammatory kidney disease. In addition to these cells, studies have also revealed the important role of locally activated kidney cells, such as tubular epithelial cells (TECs), mesangial cells, podocytes, and endothelial cells. During the development of interstitial fibrosis, TECs play an important role in initiating the inflammatory response [29]. Under damaged conditions, TECs actively participate in pro-inflammatory responses through chemokine production. Several lines of evidence suggest that chemokines produced from TECs are crucial for the recruitment of monocytes and macrophages [30]. Based on these observations, the regulation of epithelial inflammation has been an interesting target for modulating kidney inflammation and fibrosis. Based on our finding that MHY2013 decreases inflammation in animal models, we further demonstrated its role in epithelial inflammation. MHY2013 significantly reduces NF-κB activation and chemokine production in epithelial cells. Furthermore, using PPAR subtype transfection, we found that PPARβ overexpression decreases chemokine production in epithelial cells. There is evidence that PPARβ exerts anti-inflammatory effects in kidney disease. PPARβ-null mice developed more severe ischemic renal injury with more severe tubule damage than wild-type mice [31]. A macrophage-specific PPARβ-deleted mouse model also showed impaired apoptotic cell clearance and reduced anti-inflammatory cytokine production [32]. These mice were much more likely to develop autoimmune kidney disease, a lupus-like autoimmune disease. In addition, several reports have demonstrated the anti-inflammatory role of PPARβ agonists in kidney disease. GW0742 has been shown to inhibit streptozotocin-induced diabetic nephropathy in mice by reducing inflammatory mediators, including MCP-1 and osteopontin [33]. Another study showed that PPARβ agonists reduced the incidence of hypertension, endothelial dysfunction, inflammation, and organ damage in lupus mice [34]. Collectively, the reduced inflammatory responses observed in our in vitro and in vivo experiments were associated with the PPARβ activation property of MHY2013. All animal experiments were approved by the Institutional Animal Care Committee of the Pusan National University (PNU-IACUC approval No. PNU-2022-3164) and performed according to the guidelines issued by Pusan National University. C57BL/6J mice were obtained from Hyochang Science (Daegu, Republic of Korea). To establish the renal fibrosis mouse model, male mice (7-week-old) were intraperitoneally injected with a single dose of folic acid (250 mg/kg dissolved in 0.3 M NaHCO3) or vehicle. For the MHY treatment groups, MHY2013 was intraperitoneally administered in low (0.5 mg/kg/day) or high doses (3 mg/kg/day) during the experimental period (n = 5~7). All mice were maintained at 23 ± 2 °C with a relative humidity of 60 ± 5% and 12 h light/dark cycles. One week after the folic acid treatment, the mice were sacrificed using CO2 inhalation. Serum was collected for biochemical analyses. Kidneys were collected and then immediately frozen in liquid nitrogen. For long-term storage, kidney samples were moved to a −80 °C deep freezer. Part of kidneys was fixed in neutral-buffered formalin for histochemical experiments. NRK49F rat-kidney fibroblasts were purchased from ATCC (CRL-1570) and grown in Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 10% fetal bovine serum (FBS) and 1% penicillin. All cells were incubated at 5% CO2 and 37 °C in a water-saturated atmosphere. To determine the effect of MHY2013 on TGFβ-induced fibroblast activation and ECM production, a MHY2013 concentration with 10 μM was pre-treated 30 min before the TGFβ (10 ng/mL) treatment. Protein or RNA samples were collected 24 h after the TGF-β treatment to determine the effect of MHY2013. NRK52E rat-kidney epithelial cells were purchased from ATCC (CRL-1571) and grown in DMEM supplemented with 10% FBS and 1% penicillin. To determine the effect of MHY2013 on LPS-induced inflammation, a MHY2013 concentration of 10 μM was pre-treated 30 min before LPS (10 μg/mL) treatment. Protein and RNA samples were collected 1 h after LPS treatment to determine the effect of MHY2013. All cell culture experiments were performed at least 3 times per experiment. Serum samples were obtained using centrifugation at 3000 rpm for 20 min at 4 °C. Blood urea nitrogen (BUN) levels were measured using a commercial assay kit from Shinyang Diagnostics (SICDIA L-BUN, 1120171, Seoul, Republic of Korea) according to the manufacturer’s instructions. Two different solutions were used to extract proteins: ProEXTM CETi protein extract solution (Translab, Daejeon, Republic of Korea) was used to extract protein from tissues, and RIPA buffer (#9806, Cell Signaling Technology, Danvers, MA, USA) was used to obtain the total protein from the cells. Both solutions contained protease inhibitor cocktails to prevent protein degradation and phosphate inhibitor to prevent dephosphorylation. Protein concentration was measured using a BCA reagent (Thermo Scientific, Waltham, MA, USA). Extracted proteins (5–20 μg of protein) were then mixed with 4× sample buffer (Cat#1610747, Bio-Rad, CA, USA) and boiled for 5 min. The proteins were then separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (Millipore, Burlington, MA, USA). The membranes were blocked in 5% nonfat milk and washed with Tris-buffered saline-Tween buffer for 30 min. Specific primary antibodies (1:500 to 1:2000 dilution, Supplementary Table S1) were added to the membranes and incubated overnight at 4 °C. After three washes with the TBS-Tween buffer, the membranes were incubated with a horseradish peroxidase-conjugated anti-mouse, anti-rabbit, or anti-goat antibody (diluted 1:10,000) for 1 h at 25 °C. The resulting immunoblots were visualized using Western Bright Peroxide solution (Advansta, San Jose, CA, USA) and a ChemiDoc imaging system (Bio-Rad) according to the manufacturer’s instructions. All western blot analyses were performed at least 3 times per experiment. Total RNA was prepared using a TRIzol reagent (Invitrogen, Carlsbad, CA, USA). Briefly, kidney tissues (n = 5~7) or cells (n = 3) were homogenized in the TRIzol reagent. To isolate RNA, 0.2 mL chloroform was added to the 1 mL homogenate and shaken vigorously for 15 min. The aqueous phases were transferred to fresh tubes, and an equal volume of isopropanol was added. The samples were then incubated at 4 °C for 15 min and centrifuged at 12,000× g for 15 min at 4 °C. The supernatants were removed, and the resulting RNA pellets were washed once with 75% ethanol and then dried, followed by dissolving in diethyl pyrocarbonate-treated water. Next, 1.0 μg of isolated RNA was reverse-transcribed using a cDNA synthesis kit from GenDEPOT (Katy, TX, USA). qPCR was performed using a SYBR Green Master Mix (BIOLINE, Taunton, MA, USA) and a CFX Connect System (Bio-Rad). Primers were designed using Primer3Plus [35], and the primer sequences used are listed in Supplementary Table S2. For qPCR data analysis, the 2−ΔΔCT method was used as a relative quantification strategy. To visualize histological changes in the kidneys, the kidneys were fixed in 10% neutral formalin, and paraffin-embedded sections were stained with H&E. To assess the degree of renal fibrosis and damage, SR staining was performed using a commercially available kit (VB-3017; Rockville, MD, USA). This staining method is commonly used to visualize collagen fibers, which are a hallmark of fibrosis. Immunohistochemical analysis was performed to visualize the protein expression regions in the kidneys. Briefly, paraffin-embedded sections were deparaffinized and rehydrated. The sections were then incubated with the primary antibodies and visualized using diaminobenzidine substrates. The sections were counterstained with hematoxylin, which allows for the visualization of cell nuclei. Images were obtained using a microscope (LS30; Leam Solution, Seoul, Republic of Korea). ISH was performed using formalin-fixed paraffin-embedded tissue samples. RNAscope 2.5 HD Assay (322300, Biotechne, Minneapolis, MN, USA) or RNAscope 2.5 HD Duplex Detection Kit (322436, bio-techne, Minneapolis, MN, USA) was used to visualize RNA expression in the tissue, in accordance with the manufacturer’s instructions. The following probes were used to perform the RNAscope assay: Mm-Vim cat# 457961, Mm-Emr1 cat# 317969-C2, and Mm-Col1a1 cat# 319379. Images were obtained using a microscope (LS30; LEAM Solution, Seoul, Republic of Korea). Luciferase assays were performed to determine the transcriptional activity of PPAR transcription factors in the NRK49F cells. Briefly, NRK49F cells were transfected with the PPRE-X3-TK-LUC plasmid (0.1 µg) with PPARα, PPARβ/δ, or PPARγ expression vectors (0.01 µg) using Lipofectamine 3000 reagent (Invitrogen, Carlsbad, CA, USA.). The cells were further treated with MHY2013 or WY14643 (a known PPARα agonist), GW501516 (a known PPARβ/δ agonist), and rosiglitazone (a known PPARγ agonist). The luciferase activity was measured using a One-Glo Luciferase Assay System (Promega, Madison, WI, USA). After adding the luciferase substrate, the luminescence was measured using a luminescence plate reader (Berthold Technologies GmbH & Co., Bad Wildbad, Germany). Luciferase assays were performed to determine the transcriptional activity of NF-κB in the NRK52E cells. The cells were transfected with the NF-κB promoter-LUC plasmid, and the luciferase activity was measured using a One-Glo Luciferase Assay System and a luminescence plate reader. Immunofluorescence was performed to visualize protein expression in the cells. The cells were fixed in 4% formaldehyde for 10 min, washed thrice with ice-cold PBS, and exposed to 0.25% Triton-X 100 in PBS for 10 min for permeabilization. To prevent non-specific binding of antibodies, the cells were blocked using a solution containing 1% BSA and 0.1% Tween 20 in PBS at room temperature for 30 min. Next, the cells were incubated overnight with anti-αSMA antibody, which had been diluted in the blocking buffer at 4 °C. After washing off any unbound antibodies with PBS, the cells were incubated with a secondary antibody conjugated with a fluorescent tag for 1 h in the dark. The cells counterstained with Hoechst 33258 in PBS for 1 min to visualize the nuclei. The images were captured using a fluorescence microscope (LS30). Student’s t-test was used to analyze the differences between the two groups, and an analysis of variance was used to analyze intergroup differences. The level of statistical significance was set at p < 0.05. The software used for the analyses was GraphPad Prism version 5 (GraphPad Software Inc., San Diego, CA, USA). Image calculations were performed using the ImageJ software (National Institutes of Health, Bethesda, MD, USA). In conclusion, we investigated the anti-fibrotic and anti-inflammatory roles of the PPAR pan agonist MHY2013 using in vitro and in vivo kidney fibrosis models. When administered to the FA-induced mouse kidney fibrosis model, MHY2013 effectively reduced fibrosis development and inflammatory responses in the kidney. The anti-fibrotic and anti-inflammatory mechanisms of MHY2013 were further demonstrated using NRK49F kidney fibroblasts and NRK52E kidney epithelial cells. MHY2013 directly reduced TGF-β-induced ECM production in fibroblasts mainly through PPARγ activation, whereas MHY2013 suppressed LPS-induced pro-inflammatory responses in TECs mainly through PPARβ activation. Taken together, our results suggest that the administration of the PPAR pan agonist effectively prevented renal fibrosis in both in vitro and in vivo models of kidney fibrosis, implicating the therapeutic potential of PPAR agonists against chronic kidney diseases (Figure 7).

PMC10002482

Franziska Pankratz,Aziza Maksudova,Roman Goesele,Lena Meier,Kora Proelss,Katia Marenne,Ann-Kathrin Thut,Gerhard Sengle,Annkatrin Correns,Jeanina Begelspacher,Deniz Alkis,Patrick M. Siegel,Christian Smolka,Sebastian Grundmann,Martin Moser,Qian Zhou,Jennifer S. Esser

BMPER Improves Vascular Remodeling and the Contractile Vascular SMC Phenotype

03-03-2023

growth factors/cytokines,remodeling,restenosis,vascular disease,BMPER,vSMCs,neointima

Dedifferentiated vascular smooth muscle cells (vSMCs) play an essential role in neointima formation, and we now aim to investigate the role of the bone morphogenetic protein (BMP) modulator BMPER (BMP endothelial cell precursor-derived regulator) in neointima formation. To assess BMPER expression in arterial restenosis, we used a mouse carotid ligation model with perivascular cuff placement. Overall BMPER expression after vessel injury was increased; however, expression in the tunica media was decreased compared to untreated control. Consistently, BMPER expression was decreased in proliferative, dedifferentiated vSMC in vitro. C57BL/6_Bmper+/− mice displayed increased neointima formation 21 days after carotid ligation and enhanced expression of Col3A1, MMP2, and MMP9. Silencing of BMPER increased the proliferation and migration capacity of primary vSMCs, as well as reduced contractibility and expression of contractile markers, whereas stimulation with recombinant BMPER protein had the opposite effect. Mechanistically, we showed that BMPER binds insulin-like growth factor-binding protein 4 (IGFBP4), resulting in the modulation of IGF signaling. Furthermore, perivascular application of recombinant BMPER protein prevented neointima formation and ECM deposition in C57BL/6N mice after carotid ligation. Our data demonstrate that BMPER stimulation causes a contractile vSMC phenotype and suggest that BMPER has the potential for a future therapeutic agent in occlusive cardiovascular diseases.

BMPER Improves Vascular Remodeling and the Contractile Vascular SMC Phenotype Dedifferentiated vascular smooth muscle cells (vSMCs) play an essential role in neointima formation, and we now aim to investigate the role of the bone morphogenetic protein (BMP) modulator BMPER (BMP endothelial cell precursor-derived regulator) in neointima formation. To assess BMPER expression in arterial restenosis, we used a mouse carotid ligation model with perivascular cuff placement. Overall BMPER expression after vessel injury was increased; however, expression in the tunica media was decreased compared to untreated control. Consistently, BMPER expression was decreased in proliferative, dedifferentiated vSMC in vitro. C57BL/6_Bmper+/− mice displayed increased neointima formation 21 days after carotid ligation and enhanced expression of Col3A1, MMP2, and MMP9. Silencing of BMPER increased the proliferation and migration capacity of primary vSMCs, as well as reduced contractibility and expression of contractile markers, whereas stimulation with recombinant BMPER protein had the opposite effect. Mechanistically, we showed that BMPER binds insulin-like growth factor-binding protein 4 (IGFBP4), resulting in the modulation of IGF signaling. Furthermore, perivascular application of recombinant BMPER protein prevented neointima formation and ECM deposition in C57BL/6N mice after carotid ligation. Our data demonstrate that BMPER stimulation causes a contractile vSMC phenotype and suggest that BMPER has the potential for a future therapeutic agent in occlusive cardiovascular diseases. Cardiovascular diseases (CVD) are by far the leading cause of death worldwide, with approximately 17.8 million CVD deaths in 2017 [1]. Myocardial infarction, stroke, or peripheral artery diseases are often caused by atherosclerotic lesions, which are one of the main reasons for vessel occlusion [2]. Percutaneous interventions such as balloon dilatation followed by stent implantation are a common therapeutic strategy. However, a common complication after these interventional procedures is neointimal hyperplasia and restenosis [3]. Under physiological conditions, vascular smooth muscle cells (vSMCs) are highly differentiated cells of the arterial tunica media that regulate the vascular tone. A specific set of genes of the contractile apparatus, such as smooth muscle actin (SMA), calponin (Calp), myosin heavy chain 11 (Myh11), or transgelin (TAGLN), is highly expressed in vSMCs and marks the differentiated, contractile phenotype. In response to vessel injury and mechanical stress, contractile vSMC can dedifferentiate, which allows them to proliferate, migrate, and produce the extracellular matrix (ECM). This synthetic myofibroblastic phenotype of vSMCs is found to accelerate atherosclerosis, hypertension, and neointima formation [4]. Therefore, a profound understanding of the molecular mechanisms that control the vSMC phenotype is necessary to develop preventive and targeted therapeutic strategies for vascular diseases [5]. Several growth factors are well known to modulate vSMC phenotypic plasticity: e.g., platelet-derived growth factor (PDGF) and insulin-like growth factor (IGF) induce the synthetic phenotype, whereas members of the transforming growth factor β (TGF-β) family shift the balance towards a contractile phenotype [6]. In recent years, it has become evident that the TGF-β superfamily plays an important role during vascular development, homeostasis, and remodeling [7]. The bone morphogenetic proteins (BMPs) are the largest subgroup of these extracellular proteins that signal through cell-surface complexes of type I and type II serine/threonine kinase receptors. The components of the BMP signaling pathway, i.e., the ligands, receptors, and intracellular signaling molecules, are all expressed in vascular cells [8]. Upon activation, the receptors mediate intracellular signaling via the classical Smad 1/5 transcription factor phosphorylation and by alternative pathways such as MAP kinases and phosphoinositide 3-kinase (PI3K) pathways [9]. Recently, BMPs have been implicated as important regulators of vSMC plasticity. Several studies have reported BMP signaling to modulate the differentiation of vSMCs by inhibiting proliferation and migration and to promote vSMC differentiation into a contractile phenotype [10,11]. The BMP signaling pathway is highly regulated at several levels. For example, BMP receptor availability is regulated at the cell membrane by co-receptors, pseudo-receptors, and proteases [12,13]. In the extracellular space, the availability of BMPs is regulated by BMP antagonists such as chordin [14], noggin [15], and BMP modulators such as twisted gastrulation (TSG) [16] and the BMP endothelial cell precursor-derived regulator (BMPER) [17]. BMPER, also known as the vertebrate homolog of Drosophila cross-veinless 2, is a secreted glycoprotein that contains five cysteine-rich domains followed by a von Willebrand D domain and a trypsin inhibitor domain. It was originally identified in a screening for differentially expressed proteins in embryonic endothelial precursor cells [17]. We and others have previously shown that BMPER may enhance BMP signaling in a concentration-dependent fashion [18,19,20]. BMPER is able to bind to BMP-2, -4, -6, -7, -9, and BMPRIa/b [17,19,21]. Lately, BMPER has been the subject of intensive research in the area of vascular biology including coronary artery development [22], inflammation [23,24], atherosclerosis [25], and angiogenesis [18,26]. However, the focus of all prior investigations was on endothelial cells, and little is known regarding the role of BMPER in vSMCs biology. In earlier investigations, BMPER expression during embryonic development was detected in the somites and in migrating neural crest cells at E9.5. Later on, BMPER is expressed in the aorto–gonad–mesonephros (AGM) region that gives rise to hemangioblast cells and, therewith, vascular development [17,27]. Of interest, vSMCs have been reported to originate from all these sources [28], suggesting that BMPER might play a role in vSMC biology. Therefore, we aimed to investigate the function of BMPER in a mouse model of neointima formation in the carotid artery and in vSMCs phenotype modulation. Previously, we and others have shown that BMPER is expressed in endothelial cells and in the aortic wall [18,21]. However, to the best of our knowledge, the expression of BMPER in vascular disease has not been reported yet. To investigate BMPER expression in pathological arterial blood vessel remodeling such as neointima formation, we performed a mouse carotid ligation model combined with perivascular cuff placement to induce restenosis of the right common carotid artery (RCCA) (Figure 1A). With this approach, we aimed to focus on the role of vSMCs during arterial remodeling compared to other models such as a wire injury. Complete ligation of the artery distal to the carotid bifurcation induces rapid proliferation of medial SMCs, leading to extensive neointima formation while the endothelial cell layer is not disturbed [29]. In addition, cuff placement further increases mechanical stress and inflammation of the adventitial layer [30]. Moreover, the cuff allows the application of hydrogel for local substance delivery. As a simple approach to test if Bmper expression is regulated during neointima formation, we isolated RNA after 14 days from the RCCA and the left CCA (LCCA) and performed qPCR analysis. Bmper mRNA expression was increased after carotid injury compared to the untreated LCCA (Figure 1F). For histological analysis mice were sacrificed, and the RCCA as well as the untreated LCCA were isolated after 21 days. To investigate neointima formation in our carotid injury model’s hematoxylin and eosin (Figure 1B), Elastic van Gieson (Figure 1C) and Movat’s stain (Figure 1D) were performed. Immunostaining showed that Bmper was expressed in endothelial cells and vSMCs in LCCA. Interestingly, Bmper expression decreased in the intimal and medial layers in the RCCA after carotid injury. In contrast, Bmper expression strongly increased in the adventitia (Figure 1E), which is in line with the finding on the mRNA expression level. Taken together, Bmper expression is differentially regulated in the carotid artery wall after injury, supporting the notion that it may play a role during neointima formation. Given that Bmper expression is increased during neointima formation, we aimed to further investigate the function of Bmper in vivo. Because Bmper homozygous-deficient mice die at birth [20], we used BMPER heterozygous-deficient (Bmper+/−) mice to investigate the effect of Bmper deficiency after vascular injury. Three weeks post carotid injury, mice were sacrificed and carotid arteries were subjected to histopathological examination. Hematoxylin and eosin staining revealed increased stenosis (Figure 2A) in Bmper+/− mice and Elastic van Gieson staining (Figure 2B) demonstrated an increase in neointimal formation compared to wildtype littermates. Quantification of intimal size, media, and lumen revealed a significant increase in intimal/medial ratio in Bmper+/− mice along with reduced vessel lumen compared to littermates. Dedifferentiation of vSMCs is associated with increased ECM deposition, including deposition of collagens, increased MMP expression to facilitate proliferation and migration, and reduced contractile marker gene expression [31]. Therefore, collagen 3 type A1 (Col3a1), Mmp2, Mmp9, and Sma expression were further examined by immunohistological analysis 3 weeks after carotid ligation. In Bmper+/− mice, Col3a1 was increased in the arterial wall, especially in the adventitia, compared to wildtype mice (Figure 2C). Interestingly, Mmp2 and Mmp9 were clearly and especially increased in the medial layer of Bmper+/− mice compared to littermates (Figure 2C). Accordingly, Sma expression was reduced in Bmper+/− mice (Figure 2C). We corroborated and extended our findings by analyzing mRNA expression levels of Col3a1, Col1a1, Mmp2, Mmp9, fibrillin-1 (Fbn1), and Sma two weeks after carotid ligation (Figure 2D). Furthermore, as the transgene lacZ is integrated into the genomic Bmper gene locus as reporter [32], we used lacZ transgene expression to corroborate our findings from wildtype C57BL/6N mice that showed increased Bmper mRNA expression in neointima formation (Figure S1). Taken together, Bmper-deficiency resulted in enhanced expression of ECM components accompanied by increased Mmp2 and Mmp9 and reduced contractile marker Sma expression. These findings are in line with the enhanced neointima formation following injury in Bmper+/− mice. In order to investigate the impact of BMPER deficiency on vSMC phenotype and function, human vSMCs were silenced for BMPER with either of two specific small interfering RNAs (siRNAs) or transfected with scrambled siRNA as control. First of all, we assessed the efficient BMPER knockdown in a time course of 24–72 h on mRNA level by PCR followed by an analysis of contractile marker mRNA expression such as TAGLN, CALP, SMA, and MYH11 (Figure 3A). While BMPER expression was reduced, contractile marker expression was decreased (Figure 3A–C). On the other hand, expression of the synthetic vSMC marker vimentin was increased in BMPER-deficient cells (Figure 3B,C). However, not only was the expression of contractile marker SMA decreased in BMPER-silenced vSMCs but also the assembly of the contractile apparatus was diminished compared to siRNA control (Figure 3D). Changes in cell viability or apoptosis by BMPER silencing were excluded (Figure 3E,F). As a synthetic vSMC phenotype is associated with changes in cell function, we next investigated vSMC migration, contraction, and proliferation (Figure 3G–I). Indeed, cell migration and proliferation were increased 48 h after BMPER siRNA transfection, and collagen gel contraction was reduced. Collectively, we found that silencing of BMPER in vSMCs led to decreased contractile and increased synthetic marker expression, along with increased proliferation and migration, as well as decreased gel contraction emphasizing a synthetic vSMC phenotype. Given that the silencing of BMPER promotes the synthetic vSMC phenotype, we next asked if stimulation with recombinant human BMPER protein had an inverse effect on the vSMC phenotype. Before stimulation with different concentrations of BMPER protein, vSMCs were put on starvation medium for 24 h. Stimulation with BMP4 protein served as positive control to induce contractile markers [10]. As expected, BMPER stimulation of vSMCs increased TAGLN, CALP (Figure 4A), and SMA (Figure S2A) mRNA expression. In addition, MYH11 and SMA protein expression was enhanced and, conversely, VIM protein expression was reduced after BMPER stimulation (Figure 4B,C). After BMP4 and BMPER stimulation, immunocytochemistry staining of SMA showed increased contractile filaments in vSMCs compared to unstimulated control, particularly compared to vSMCs stimulated with PDGF, which is known to evoke a synthetic vSMC phenotype (Figure 4D). Cell viability and apoptosis assays revealed no effect of BMP4 or BMPER on vSMCs (Figure 4E,F). BMP4 and BMPER protein stimulation reduced vSMC migration and proliferation in contrast to PDGF (Figure 4G,H). In summary, we confirmed our notion that recombinant BMPER promotes a contractile phenotype in vSMCs in vitro. As BMPER shifts the vSMC phenotype towards a contractile phenotype, we hypothesized that the expression of BMPER and BMP4 is downregulated in vSMCs that adopt a synthetic phenotype. Therefore, serum-starved vSMCs were stimulated with increasing concentrations of FBS that is commonly used to stimulate SMC proliferation and dedifferentiation in vitro [33]. Diminished expression of the contractile markers SMA and TAGLN were used as positive controls (Figure 5A), indicating that vSMCs had switched towards the synthetic phenotype. Along this line, BMPER and BMP4 expression is downregulated. In addition, expression of BMPER, BMP4, SMA, and TAGLN in vSMCs after FBS stimulation was further confirmed on the protein level (Figure 5B). To investigate if the classical SMAD signaling cascade is affected by decreased BMPER expression, we silenced BMPER in vSMCs and performed Western blot analysis 48 h post-siRNA transfection (Figure S2B). Similar to our previous findings in endothelial cells that reduced BMPER expression decreases SMAD signaling [18], we now confirmed this finding in vSMCs. Taken together, BMPER and BMP4 along with contractile markers are downregulated in vSMCs that switch to a synthetic phenotype. Many extracellular proteins such as the matricellular protein cellular communication network factor (CCN2) mediate pleiotropic effects by binding to different growth factors or other ECM components [34]. We and others have shown that BMPER binds and regulates the function of BMP2, 4, 6, 7, and 9 [35]. However, if we assume that BMPER in high concentrations inhibits BMP4 and both BMPER and BMP4 can induce a contractile phenotype in vSMCs, BMPER seems to act in an additional way to promote the contractile phenotype. In yeast two-hybrid screening performed in the past, we identified that IGFBP4 physically interacts with BMPER. To confirm our findings, we performed a PLA with BMPER and IGFBP4 antibodies to show a physical interaction in vSMCs. The PLA of serum-starved vSMCs with species-specific control antibodies shows only a few unspecific signals compared to PLA with BMPER and IGFBP4-specific antibodies (Figure 6A). Moreover, in order to verify specific fluorescence signals, we stimulated the vSMCs for 20 min with IGF1 to activate the IGF pathway. Indeed, changes in the fluorescent signals in the cellular localization were detectable compared to serum-starved vSMCs. To further confirm a physical BMPER–IGFBP4 interaction, we performed co-immunoprecipitation (IP) assays in HEK293A. Cells were transfected with BMPER-Myc and IGFBP4-V5 tagged or empty control vectors for 24 h before IP, using V5-specific or BMPER-specific antibody, as shown in Figure 6B,C, respectively, was performed. To corroborate our findings and to gain better insight into the interaction of BMPER with IGFBP4 and its potential impact on BMP4 binding, we employed surface plasmon resonance (SPR)-binding studies. When increased concentrations of BMPER were flown over immobilized IGFBP4 on a sensor chip, robust binding signals were detected, as shown by association curves after injection and dissociation curves after injection stop (Figure 6F). By assessing the kinetic parameters of both curves, a high affinity binding constant of KD = 6 nM was determined for this interaction. As shown previously, soluble BMPER flown over immobilized BMP4 also showed a high binding affinity of 1.5 nM (Figure S3A). However, in contrast, when IGFBP4 was flown over immobilized BMP4, no interaction was detected (Figure S3B). To assess whether the presence of IGFBP4 affects the interaction of BMPER with BMP4, a stable concentration of BMPER was flown over immobilized BMP4 in the presence of increased concentrations of IGFBP4. Our results from this competition assay revealed that IGFBP4 does not interfere with the BMPER–BMP-4 interaction (Figure S3C). While BMPER and BMP4 expression was decreased in vSMCs after serum stimulation (Figure 5), IGFPB4 expression was not significantly downregulated in serum-stimulated vSMCs (Figure S4A). The function of IGFBP4 is to bind and inhibit IGF binding its receptor in the extracellular space. IGFBPs are in turn post-translationally modulated by specific proteases that release IGF. In vSMC, activation of the IGF signaling pathway is associated with increased migration and proliferation, as well as neointima formation [36]. Along this line, siRNA-mediated silencing of IGFBP4 in vSMCs decreased the expression of the contractile markers TAGLN, CALP, and SMA (Figure S4B). As we have shown that BMPER binds to IGFBP4, we next asked if BMPER protects IGFBP4 from cleavage by pregnancy-associated plasma protein-A (PAPP-A) and thus inhibits the release of IGF, which could have a protective role in neointima formation. Again, we used HEK293A cells and overexpressed IGFBP4 and BMPER under serum-free conditions. We confirmed uniform overexpression of IGFBP4 in HEK293A lysates 24 h post transfection (Figure 6D). The supernatants containing the extracellular proteins IGFBP4 and BMPER were used in a cell-free assay to determine the relative contribution of PAPP-A to the total proteolysis of IGFBP-4 with and without the presence of BMPER and IGF1, which enhances proteolytic cleavage of IGFBP4 (Figure 6E). As reported, IGF1 allowed proteolytic cleavage of IGFBP4 by PAPP-A [37]. However, if BMPER is co-expressed with IGFBP4, IGF1-induced proteolytic cleavage by PAPP-A is significantly reduced. In situ analysis of Bmper and Igfbp4 mRNA expression in injured and control carotid artery using RNAscope technology showed a similar expression pattern (Figure S4C) and confirmed the results of the immunohistofluorescence Bmper (Figure 1E) and lacZ stain (Figure S1B). Moreover, PLA of injured and control carotid artery showed a physical association in situ that is clearly enhanced in the adventitia of the injured carotid artery (Figure 6G). Taken together, these data demonstrate that BMPER, in addition to the BMPs, also interacts with the IGF signaling pathway regulator IGFBP4 and protects IGFBP4 against proteolytic cleavage. In turn, the release of IGF1 was diminished, suggesting that in this way the contractile vSMC phenotype can be preserved. The therapeutic goal in the treatment of occlusive vessel disease is to prevent restenosis. Previously, we and others have found that BMPER exerts a moderate pro-angiogenic and anti-inflammatory effect on endothelial cells [18,20,23,24,38]. Combined with our novel findings that BMPER supports the contractile vSMC phenotype, BMPER appears to be an attractive candidate for a therapeutic approach in arterial vessel disease. To test this hypothesis, recombinant mouse BMPER (rBmper) protein was mixed in Pluronic gel, which in its sol state acts as a drug-delivery system in a physiological environment [39]. Carotid ligation, in combination with perivascular cuff placement, was used to induce restenosis, and the cuff allowed the local application of Pluronic gel. Three weeks post carotid ligation, together with Pluronic gel application, C56BL/6N mice were sacrificed and carotid arteries were subjected to histopathological examination. As expected, HE staining (Figure 7A) and EvG staining (Figure 7B) of mice treated with solvent control showed clear induction of neointima formation. However, local administration of rBmper reduced the neointimal area, which is indicative of diminished vSMC migration and proliferation. Quantification of intimal size, media, and lumen revealed in total a clear trend towards decreased intimal/medial ratio in rBmper mice along with significantly increased vessel lumen. In order to figure out the vSMC phenotype state, we next investigated changes in the expression of ECM composition and vSMC markers. Immunohistochemistry analysis showed markedly diminished Col3a1, Mmp2, and Mmp9 expression 3 weeks after carotid ligation in rBmper mice compared to solvent control mice (Figure 7C). Additionally, Sma expression was enhanced in the media of rBmper, overall indicating a more contractile phenotype of vSMCs. These results are corroborated and extended by analysis of mRNA expression levels of Col3a1, Col1a1, Mmp2, Mmp9, Fbn1, and Sma two weeks after carotid ligation (Figure 7D). Altogether, these data demonstrate that administration of rBmper in Pluronic gel during the process of restenosis in mice has the capacity to keep vSMCs in the contractile phenotype and, thus, reduce neointima formation. Unlike other muscle cells such as cardiomyocytes, vSMCs are not terminally differentiated. In response to injury or mechanical stress, contractile vSMCs change phenotype, proliferate, and migrate as part of the remodeling process. Dysregulation of this plasticity program is the reason for neointimal hyperplasia, which is the major cause for restenosis after percutaneous interventions [40]. Here, we report that after vessel injury, the expression of ECM protein Bmper is altered during the course of neointima formation. Both loss-of-function and gain-of-function studies suggest that Bmper plays a key role in vSMCs’ phenotypic switch in vitro and in vivo and that, presumably, in this way, Bmper limits neointima formation. Besides activating the BMP pathway, we revealed IGFBP4 as a new physical interaction partner for BMPER. This finding highlights a second way through which BMPER inhibits the IGF pathway and, thus, may promote the contractile vSMC phenotype. Besides vSMC migration and proliferation in the subintimal space, a hallmark of neointima formation is the deposition of high amounts of ECM [41]. Interestingly, Bmper mRNA expression levels of lysates from injured carotid arteries compared to untreated arteries are upregulated after carotid ligation at a time point that is correlated with high ECM deposition. By now, it is well-accepted that there is a tight interplay between the ECM and the BMP signaling pathway [42], which would support the notion that Bmper is available and, thus, its expression is increased. However, examination of the immunohistological Bmper staining in untreated carotid artery displayed a good Bmper signal in the media, which 21 days after carotid ligation was lost, and Bmper expression was detected more in the adventitial tissue. In line with these findings, BMPER and BMP4 expression was reduced in cultured human vSMCs that were stimulated with FBS to promote SMC proliferation and de-differentiation. Further reasons for changes in BMPER expression could be changes in mechanics, e.g., blood flow, and BMP signaling feedback loops. Corriere et al. have reported increased Bmp4 expression after carotid ligation during neointima formation and assumed that endothelial Bmp4 expression is induced by alterations in blood flow [43]. Along the same lines, another study has shown that in response to disturbed flow conditions, BMP4 and BMP antagonists such as noggin are coexpressed in the arterial wall of mouse and human blood vessels [44]. Moreover, the authors hypothesize that the expression of antagonists seemed to play a negative feedback role against the inflammatory response of BMP4. In fact, signaling pathway autoregulation is a common theme for the BMP pathway and has been reported by several groups in different tissue contexts [21,45,46]. Our findings show that BMPER alongside the contractile markers is downregulated in vSMCs that adopt a synthetic phenotype. Moreover, our in vitro data demonstrate that BMPER deficiency triggers vSMC dedifferentiation, proliferation, and migration, supporting the role of BMPER in additionally regulating vSMC phenotypic switching. Accordingly, neointima formation in Bmper heterozygote-deficient mice is aggravated, and this supports the notion that in vivo reduced Bmper expression also increases the synthetic vSMC phenotype as the mouse in vivo model of restenosis used is predominantly a model for investigating the contribution of vSMCs to neointima formation [29,30]. This is in good agreement with several in vitro studies that have reported that BMP signaling influences the differentiation of vSMCs by inhibiting proliferation and migration and promoting vSMC differentiation into a contractile phenotype [11]. For example, Lagna et al. have reported that BMP2, BMP4, and BMP7 displayed a similar preserving effect on contractile vSMC-specific gene expression in different types of vSMCs, which is SMAD-dependent [10]. Moreover, the same group revealed that SMAD transcription factors regulate postranscriptional miRNA biogenesis that is critical for the control of the contractile vSMC phenotype [6]. In accordance with these findings, Corriere et al. reported that, besides increased Bmp4 expression after carotid ligation, constitutively active BMP type IA receptor reduced vSMC proliferation and migration in vitro. They hypothesized that BMP4 expression is increased and the BMP pathway is activated to counterbalance the proliferative and chemoattractant effects of other growth factors such as PDGF that are also upregulated in vivo [43]. Given that stimulation with recombinant BMPER protein also induces a contractile vSMC phenotype in vitro and has the tendency to reduce neointima formation in vivo, we suppose that BMPER at least partially upon activating the BMP pathway promotes the contractile vSMC phenotype. Altogether, these findings indicate that a tightly regulated and functional BMP signaling pathway is necessary to keep vSMCs in a contractile, quiescent, and healthy state. Interestingly, extracellular growth factor modulators are known to regulate different signaling pathways and are rarely pathway-specific [47]. Of note, also some BMP modulators such as the cerberus or connective tissue growth factor (CTGF) have been shown to modulate more than just the BMP pathway. Besides BMPs, CTGF was shown to inhibit BMP signaling on the one hand and to promote TGF-β signaling on the other hand [48]. This circumstance has also been confirmed for BMPER that was shown to interact with low-density lipoprotein receptor-related protein 1 (LRP1) in addition to the BMPs and BMP receptors [49]. Here, we have now add IGFBP4 to the list of BMPER binding partners. IGFBP4 itself is an extracellular modulator that antagonizes IGFs and is cleaved by the protease PAPP-A. The IGF signaling pathway is characterized to promote the synthetic vSMC phenotype and IGF, and PAPP-A expression is increased in restenosis and as a reaction to vascular injury [36,50]. Several studies have described IGFBP4 as being the major IGFBP produced by vSMCs, and its expression is upregulated after carotid ligation [51,52]. Interestingly, IGFBP4 cleavage by PAPP-A is IGF-dependent, and it is hypothesized that PAPP-A can act as an IGF regulator through the proteolysis of IGFBP4, which ultimately causes neointimal hyperplasia [52,53]. This is underlined by a study with a protease-resistant form of IGFBP4 that was reported to inhibit IGF and neointima expansion in a porcine model of neointimal hyperplasia [54]. We have found that BMPER not only binds to IGFBP4, but also inhibits the proteolytic cleavage of IGFBP4 by PAPP-A in the presence of IGF-1. Therefore, it is tempting to speculate that BMPER, in addition to the BMP pathway, promotes the contractile vSMC phenotype by binding to IGFBP4 and in this way prevents the release of IGF, which itself would trigger the synthetic phenotype (please also refer to scheme in Figure 8). By definition, BMPER is a matricellular protein because it is a non-structural protein that is located in the ECM and binds to structural ECM components such as heparan sulfate proteoglycans [55]. Furthermore, BMPER has regulatory roles for the BMP and the IGF pathway, and thus, BMPER belongs to the category of matricellular proteins [56]. By modulation of signaling pathway activities in endothelial cells, BMPER mediates anti-inflammatory [23,24,25], and moderate pro-angiogenic [18,38,49] effects, and increases the expression of endothelial nitric oxide synthase (eNOS) [24]. It is well known that eNOS and its product nitric oxide control vasomotor function ad cardiovascular homeostasis and support the differentiated vSMC phenotype [57]. In our present study, we show that BMPER promotes the contractile vSMC phenotype and diminishes neointima formation in the carotid ligation model. In general, BMPER has beneficial effects on both endothelial and vSMCs and we are tempted to call it a “wellness factor” for blood vessels. Of note, there are very few other examples, including interleukin-19, that have anti-inflammatory and at the same time pro-angiogenic effects [58]. Growth factors or cytokines with these properties are of interest in the context of a therapeutic approach to cardiovascular diseases, such as myocardial infarction or ischemic tissues per se, as they could stimulate neovascularization in ischemic tissue while simultaneously attenuating the existing tissue-damaging inflammation. Collectively, these data emphasize the extracellular matrix protein BMPER with its rare and valuable characteristics as a future therapeutic agent with high potential. Many large population-based cohort studies report differences in atherosclerotic diseases in different sexes [59]. To limit the potential effects of female hormones, only male mice ranging from 10 to 12 weeks of age were studied. Transgenic B6SJL-Bmpertm1Emdr/J mice were originally purchased from the Jackson Laboratory, and a congenic strain on C57/BL6N background was generated (>10 back-crosses). Because BMPER−/− animals die at birth [20], BMPER+/−-deficient mice compared to wildtype littermates were used in experiments. For the application of recombinant mouse BMPER protein wildtype C57/BL6N, mice were purchased from Charles River, Sulzfeld, Germany, or from the local stock of the animal facility at the University Medical Center Freiburg, Germany. Mice were housed under specific pathogen-free conditions and had ad libitum access to water and food. Handling and care of animals were approved and in compliance with the guidelines for the care and use of laboratory animals published by the directive 2010/63/EU of the European Parliament. Experimental animal protocols were approved in advance by the Regierungspraesidium Freiburg, Germany. To induce neointima formation in the mouse carotid artery, we combined the two methods of carotid artery ligation directly under the bifurcation [29] with polyethylene cuff placement below the ligation around the carotid artery [60] (please refer to Figure 1A). Mice were anesthetized with ketamine (Ketaset®, Zoetis Deutschland GmbH, Berlin, Germany)/xylazine (Rompun® BAYER, Leverkusen, Germany) (100/10 mg/kg bw, i.p.) in 0.9 saline, followed by preemptive caprofen (Rimadyl, Zoetis Deutschland GmbH, Berlin, Germany (5 mg/kg bw, s.c.)) for post-operative analgesia. Mice were secured in the supine position on a warm heating pad to avoid cooling. The surgical area was shaved and disinfected with Cutasept (Paul Hartmann AG, Heidenheim, Germany). Dexpanthenol eye and nose cream (Bepanthen, BAYER, Leverkusen, Germany) was applied to prevent corneal desiccation. With the scalpel, an approx. 1 cm long incision was made on the right side of the neck. The right common carotid artery (RCCA) was dissected and ligated with 6-0 black braided silk sutures (LookTM, Surgical Specialties Corporation, Westwood, MA, USA) near the carotid bifurcation. After ligation, a perivascular polyethylene cuff (0.58 mm inner diameter, 0.964 mm outer diameter, length 5 mm, Becton Dickinson, Franklin Lakes, NJ, USA) was placed around the artery, thereby inducing mechanical stimulation of neointima formation. Finally, the wound was closed with single-button sutures braided with 3-0 Mersilene® (Ethicon, Norderstedt, Germany) suture. The animals were placed in a cage with red light to be warmed and observed until full consciousness was restored and then returned to the colony. To administer recombinant BMPER protein (c = 5 µg/mL recombinant mouse BMPER in 0.9% saline; R&D Systems, Darmstadt, Germany) to C57/BL6N mice, the protein was mixed with 30% Pluronic® gel (Sigma-Aldrich, Schnelldorf, Germany) and applied to the perivascular cuff. The Pluronic gel solidifies at body temperature, forming a gel matrix in the cuff to act as a local drug delivery system in a physiological environment. The control group received the same volume 0.9% saline into the Pluronic gel and cuff. For mRNA expression and (immune-) histochemistry analysis, the injured RCCA and the untreated left common carotid artery (LCCA) were harvested at 14 days or 21 days, respectively. Excised carotid arteries were embedded in Tissue-Tek O.C.T. compound (Fisher Scientific GmbH, Schwerte, Germany) and stored at −20 °C for further histological analysis. Eight-micrometer serial cryostat sections were cut starting from the bifurcation towards the aortic arch. Sections were air-dried, fixed in ice-cold acetone, and subjected to standard hematoxylin and eosin (HE), Movat’s, and Elastica van Gieson (EvG) staining (all Morphisto GmbH, Frankfurt, Germany). To assess morphological changes and to identify the neointimal thickness (represented as ratio of intima/media), neointimal area, and media area, slides were analyzed by a blinded investigator with Zeiss Axioplan2/Axiovision Rel. 4.8 software or Zeiss Axio Imager Z2/ZEN 3.1 blue edition software. Immunofluorescent staining was performed for αSMA-FITC, BMPER, Col3A1, MMP2, and MMP9 compared to respective mouse or rabbit IgG controls (Supplementary Table S1). AlexaFluor555–conjugated secondary antibody was used, and nuclei were stained with DAPI (Sigma-Aldrich). Slides were imaged with Zeiss Axioplan2/Axiovision Rel. 4.8. Confocal images of human vSMCs stained with anti-αSMA-Cy3 antibody (C6198, clone 1A4, Sigma-Aldrich) were taken by using a ZEISS LSM5 Live DUO high-speed confocal microscope at the Life Imaging Center, ZBSA, Freiburg, Germany. In situ hybridization with fluorophor–labeled RNA probes was performed by using the RNAscopeTM (ACD Biosciences, Newark, NJ, USA) technology, referring to the manufacturer’s instructions. Briefly, carotid arteries were fixed 14 days after carotid ligation with 10% freshly prepared PFA and embedded in O.C.T. Pretreatment with hydrogen peroxide was followed by target retrieval and specific probe hybridization and signal amplification. Afterwards, nuclei were counterstained with DAPI and slides were mounted to proceed with microscopic evaluation. Human primary pulmonary arterial vascular smooth muscle cells (vSMCs) were purchased and cultured in smooth muscle cell growth medium with 5% fetal bovine serum (FBS) from PELOBiotech (Martinsried, Germany). HEK293A was cultured in DMEM supplemented with 10% FBS (both Gibco®). Recombinant proteins (please refer to supplementary data table) were reconstituted according to the manufacturer’s protocol (R&D Systems, Wiesbaden, Germany). For siRNA transfection, cells were seeded the day before and transfected in endothelial basal medium (EBM) containing 0.4% FBS. For recombinant protein or FBS stimulation, cells were seeded in culture medium, and the next day, the cells were synchronized for 24 h in 0.4% FBS EBM before the stimulation started. Silencing of BMPER by siRNA transfection in vSMC was performed as recently described. BMPER siRNAs (B1 and B2) were purchased from Thermo Fisher Scientific, Karlsruhe, Germany. Allstars negative control Alexa Fluor-488 nm was purchased from Qiagen, Hilden, Germany. For transfection, a final concentration of 100 nmol/L siRNA together with Lipofectamine RNAiMAX was used according to the manufacturer’s protocol (InvitrogenTM). Transfection efficacy was confirmed by quantitative real-time (q) PCR and Western blot analysis. For plasmid transfection in HEK293A cells, FuGENE HD transfection reagent (Promega; Mannheim, Germany) was used. In brief, up to two plasmids were diluted in OptiMEM (Gibco®), and FuGENE® HD transfection reagent was added in the ratio of 1:3 (DNA:Fugene). After 15 min incubation time at room temperature, the transfection mixture was added to 80% confluent cell dishes and incubated for 24 h. For a list of plasmid constructs please refer to supplementary Table S1. Proliferation was assessed using a colorimetric BrdU-incorporation ELISA (Roche, Basel, Switzerland). In brief, after 24 h siRNA transfection or stimulation with recombinant proteins, cells were cultured in fresh BrdU-containing 0.4% FBS/EBM medium for another 24 h. The colorimetric ELISA for BrdU quantification was performed following the manufacturer’s instructions. Cell migration assay was performed as previously described [18]. In brief, vSMCs pre-stimulated with indicated recombinant proteins or transfected with siRNAs for 48 h were labeled with 10 µM CFDA-SE (Life Technologies), harvested by centrifugation, resuspended in migration medium (RPMI with 0.5% FBS, 0.1% BSA), counted, and placed in the upper chamber of a modified Boyden chamber (1 × 105 cells per HTS FluoroBlok 24-well chamber; pore size 8 µm; BD Biosciences, Heidelberg, Germany). The chambers were placed in 24-well culture dishes containing migration medium and indicated recombinant proteins. After incubation for 4 h at 37 °C, 5% CO2, the cells were fixed with 4% PFA and migrated cells were counted manually in 5 random microscopic fields using an Axiovert fluorescence microscope. After 6 h siRNA transfection, 5 × 104 vSMCs were mixed with 250 µL of freshly prepared collagen gel solution (Cultrex Rat Collagen I, R&D Systems, 0.1% acetic acid, 1M sodium hydroxide solution, HEPES buffer (Gibco), 10× Medium 199 (Sigma-Aldrich)), seeded in 48 well plates and incubated for 30 min at 37 °C in a cell incubator to induce gelation. Afterwards, 250 µL 0.4% FBS/EBM was added for 48 h, before the gels were lifted off the bottom of the wells and allowed to float freely. Twenty-four hours later gels were stained with phenol red to increase the contrast, imaged, and gel-size-quantified with a digital imaging system (ChemiDOC XRS (Bio-Rad)). Cell viability was assessed by using the CellTiter-Fluor™ cell viability kit from Promega, Mannheim, Germany. CellTiter-Fluor™ measures a conserved and constitutive protease activity within live cells and therefore serves as a marker of cell viability. Briefly, 7000 vSMCs per 96 wells were seeded in 5% SMC medium before stimulation with recombinant proteins or siRNA transfection in 0.4% EBM was performed. Cell viability assay was performed following the manufacturer’s instructions. The resulting fluorescence signal of each well was measured by using the Gemini fluorescence microplate reader and SoftMaxPro software (Molecular Devices) set to 380 nmEx/505 nmEm. Cell apoptosis was assessed by using the luminescence Caspase-Glo®3/7 assay kit from Promega. Caspase-Glo®3/7 is a proluminescent caspase-3/7 substrate, which is cleaved to release aminoluciferin, a substrate of luciferase used in the production of light. Cells were treated as described for cell viability assy. After 48 h, cell apoptosis assay was performed following the manufacturer’s instructions. The resulting luminescence signal of each well was measured by using the GloMax® 96-microplate luminometer (Promega). To detect direct BMPER–IGFBP4 physical interaction, either vSMCs were grown on 12 well glass coverslips (ibidi, Graefelfing, Germany) or carotid tissue sections were fixed with acetone for 20 min at −20 °C and washed 3 times in PBS. Subsequently, the Duolink® PLA assay was performed following the manufacturer’s instructions (Sigma-Aldrich). In brief, rabbit anti-Igfbp4 antibody and rat anti-Bmper antibody were incubated overnight at 4 °C. Afterwards, anti-rabbit and anti-mouse oligonucleotides labeled secondary antibodies (PLA probes) were incubated, followed by a ligase and polymerase reaction to amplify the signal. DAPI was used to stain the nuclei before photographs were taken by Zeiss Axio Imager Z2 with ApoTome/ZEN 3.1 blue edition software. Total RNA from mouse carotid arteries was extracted using the TriPure isolation method according to the manufacturer’s protocol followed by DNase I treatment (Sigma-Aldrich). Reverse transcriptions were performed with iScript cDNA-Kit, applying 1 µg RNA following the manufacturer’s protocol (Bio-Rad, Munich, Germany). Quantitative real-time PCR was performed using IQ SybrGreen 2 × Supermix (BioRad) or TaqMan® gene expression master mix and assays (Thermo Fisher Scientific) that were analyzed with the iCycler real-time PCR detection system or CFX96 touch real-time PCR detection system (Bio-Rad). Quantification was performed using MyiQ lightcycler or CFX manager version 3.1 software (Bio-Rad). Differences in gene expression were calculated using the ΔΔCT method [61]. The housekeeping gene human RNA polymerase II (hRP) was used for internal normalization. Primers were purchased from Eurofins MWG Operon, Ebersberg, Germany. For primer sequences and TaqMan kits, please refer to Supplementary Table S1. Western blot analysis was performed as previously described. Primary antibodies were incubated overnight at 4 °C and secondary antibodies at room temperature for 1 h in 3% non-fat dried milk/TBST. Visualization was performed by an ECL system (GE Healthcare Europe GmbH, Freiburg, Germany) and a digital imaging system (ChemiDOC XRS (Bio-Rad)). For quantification of protein band intensities, Image Lab (Bio-Rad, Munich, Germany) was used and expression was normalized to Gapdh loading control. Twenty-four hours after plasmid transfection, the supernatants were harvested by centrifugation, and IGF1 (c = 25 ng/mL) was pre-incubated at 37 °C for 15 min at 37 °C in an incubator before PAPP-A (c = 150 ng/mL) was added and incubated for another 2 h at 37 °C. Afterwards, supernatants were transferred for concentration on centrifuge tubes with Amicon® Ultra-4 10 K centrifugal filter units (Millipore) and 50 μL protein concentrate was subjected to Western blot analysis. Co-immunoprecipitation was used to demonstrate the potential protein–protein interaction of BMPER and IGFBP4. Twenty-four hours after plasmid transfection, HEK293A cells were mechanically detached in PBS w/o and centrifuged at 500× g for 5 min at 4 °C and the pellet was resuspended in 500 μL non-denaturing IP lysis buffer. Cell disruption was then performed by repeated freezing in liquid nitrogen. Following lysis, lysates were centrifuged at 4 °C for 10 min at 12,000 g, and aliquots were set aside for direct blot analysis (input). For the IP 5 μg of the specific antibody was added to the cell lysate and incubated overnight at 4 °C rotating. The next day, immunocomplexes were precipitated at 4 °C for 1 h with protein G PLUS-agarose beads (Santa Cruz, Heidelberg, Germany) and subsequently washed three times with 1 mL IP buffer for 10 min and centrifugation at 2000× g for 5 min at 4 °C. Proteins were eluted with 6× Laemmli sample buffer and heated for 5 min at 95 °C before Western blot analysis was performed. SPR experiments were performed as described previously (using a BIAcore 2000 system (BIAcore AB, Uppsala, Sweden)). BMP-4 and IGFBP-4 were immobilized at 430 and 360 RUs, respectively, to a CM5 sensor chip using the amine coupling kit following the manufacturer’s instructions (Cytiva, Uppsala, Sweden). Interaction studies were performed by injecting 0–320 nM recombinant BMPER and IGFBP-4 in HBS-EP buffer (0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA, and 0.005% (v/v) surfactant P20) (Cytiva). For competition assays, BMPER was preincubated for 30 min at a constant concentration of 0.75 nM with 0–6 nM IGFBP-4 prior to injection. The surface was regenerated with a pulse of 10 mM glycine, pH 1.5. Kinetic constants were calculated by nonlinear fitting (1:1 interaction model with mass transfer) to the association and dissociation curves according to the manufacturer’s instructions (BIAevaluation version 3.0 software). Apparent equilibrium dissociation constants (KD values) were then calculated as the ratio of kd/ka. Statistical analysis was performed using GraphPad Prism 5.0, La Jolla, USA. Data are presented as mean ± SEM, and comparisons were calculated by Student’s t-test (2-sided, unpaired). All experiments were repeated at least three times in triplicates. One-way ANOVA was used for multiple comparisons of >2 groups. The Bonferroni post-test for multiple comparisons was used if the p-value for the overall ANOVA comparison was statistically significant. Results were considered statistically significant for p < 0.05.

PMC10002483

Cristina Rodríguez-Díaz,Flores Martín-Reyes,Bernard Taminiau,Ailec Ho-Plágaro,Raquel Camargo,Felix Fernandez-Garcia,José Pinazo-Bandera,Juan Pedro Toro-Ortiz,Montserrat Gonzalo,Carlos López-Gómez,Francisca Rodríguez-Pacheco,Dámaris Rodríguez de los Ríos,Georges Daube,Guillermo Alcain-Martinez,Eduardo García-Fuentes

The Metagenomic Composition and Effects of Fecal-Microbe-Derived Extracellular Vesicles on Intestinal Permeability Depend on the Patient’s Disease

04-03-2023

microbiome,metagenome,fecal-microbe-derived extracellular vesicles,intestinal permeability,diarrhea,morbid obesity,inflammatory bowel disease

The composition and impact of fecal-microbe-derived extracellular vesicles (EVs) present in different diseases has not been analyzed. We determined the metagenomic profiling of feces and fecal-microbe-derived EVs from healthy subjects and patients with different diseases (diarrhea, morbid obesity and Crohn’s disease (CD)) and the effect of these fecal EVs on the cellular permeability of Caco-2 cells. The control group presented higher proportions of Pseudomonas and Rikenellaceae_RC9_gut_group and lower proportions of Phascolarctobacterium, Veillonella and Veillonellaceae_ge in EVs when compared with the feces from which these EVs were isolated. In contrast, there were significant differences in 20 genera between the feces and EV compositions in the disease groups. Bacteroidales and Pseudomonas were increased, and Faecalibacterium, Ruminococcus, Clostridium and Subdoligranum were decreased in EVs from control patients compared with the other three groups of patients. Tyzzerella, Verrucomicrobiaceae, Candidatus_Paracaedibacter and Akkermansia were increased in EVs from the CD group compared with the morbid obesity and diarrhea groups. Fecal EVs from the morbid obesity, CD and, mainly, diarrhea induced a significant increase in the permeability of Caco-2 cells. In conclusion, the metagenomic composition of fecal-microbe-derived EVs changes depending on the disease of the patients. The modification of the permeability of Caco-2 cells produced by fecal EVs depends on the disease of the patients.

The Metagenomic Composition and Effects of Fecal-Microbe-Derived Extracellular Vesicles on Intestinal Permeability Depend on the Patient’s Disease The composition and impact of fecal-microbe-derived extracellular vesicles (EVs) present in different diseases has not been analyzed. We determined the metagenomic profiling of feces and fecal-microbe-derived EVs from healthy subjects and patients with different diseases (diarrhea, morbid obesity and Crohn’s disease (CD)) and the effect of these fecal EVs on the cellular permeability of Caco-2 cells. The control group presented higher proportions of Pseudomonas and Rikenellaceae_RC9_gut_group and lower proportions of Phascolarctobacterium, Veillonella and Veillonellaceae_ge in EVs when compared with the feces from which these EVs were isolated. In contrast, there were significant differences in 20 genera between the feces and EV compositions in the disease groups. Bacteroidales and Pseudomonas were increased, and Faecalibacterium, Ruminococcus, Clostridium and Subdoligranum were decreased in EVs from control patients compared with the other three groups of patients. Tyzzerella, Verrucomicrobiaceae, Candidatus_Paracaedibacter and Akkermansia were increased in EVs from the CD group compared with the morbid obesity and diarrhea groups. Fecal EVs from the morbid obesity, CD and, mainly, diarrhea induced a significant increase in the permeability of Caco-2 cells. In conclusion, the metagenomic composition of fecal-microbe-derived EVs changes depending on the disease of the patients. The modification of the permeability of Caco-2 cells produced by fecal EVs depends on the disease of the patients. Gram-positive and Gram-negative bacteria release membrane vesicles with sizes ranging from 20 to 400 nm in different abundances, structures, and molecular cargo [1,2]. These microbial extracellular vesicles (EVs) represent a secretion and transport mechanism for carbohydrates, lipids and several cell wall components as well as proteins, DNA, RNA and signaling molecules, among others [3]. Therefore, EVs have been related to cell-to-cell communication, virulence, horizontal gene transfer or phage infection [4,5]. Although the outer membrane vesicles (OMVs) of Gram-negative bacteria were the first found and described, recent work has demonstrated the production of other types of EVs by both Gram-positive and Gram-negative bacteria and even mycobacteria and fungi [4]. The types and origins of these EVs were summarized in a previous review, including OMVs, outer-inner membrane vesicles (OIMVs), cytoplasmic membrane vesicles (CMVs) and tube-shaped membranous structures (TSMSs) [1]. EVs play an essential role in bacterial survival and host interactions due to inter-kingdom signaling and their potential properties in the fecal microbiota–eukaryote interaction [6]. For example, it has been shown that EVs of toxigenic Bacteroides fragilis (B. fragilis) contribute to bowel disease and colon cancer [7]. EVs of Akkermansia muciniphila (A. muciniphila) have been shown to play a role in controlling intestinal permeability and regulating intestinal barrier integrity, improving metabolic function and ameliorating obesity in mice [8]. A previous study also demonstrated how Listeria monocytogenes (L. monocytogenes) produces EVs that carry the majority of listerial virulence proteins, and it uses these EVs for toxin release and mammalian toxicity [9]. Recently, several studies have investigated the secretion of EVs by pure bacterial cultures, while less data are available regarding the secretion of EVs by complex microbial communities or environments. Lagos [10] isolated EVs secreted from fresh pig feces in vitro and observed modifications in their composition and abundance in function under the environmental conditions, especially with respect to carbohydrate availability. Tulkens [2] described the presence of bacterial EVs in human plasma and correlated their abundance with immune activation and barrier integrity in patients with Crohn’s disease (CD), human immunodeficiency viruses (HIVs) and cancer. Further research on the fecal microbiota composition and the derived EVs in feces demonstrated the role of bacterial EVs in the regulation of intestinal immunity and homeostasis and highlighted the protective effect of A. muciniphila EVs in the development of dextran-sulfate-sodium-induced colitis [11]. Recently, it was demonstrated how Staphylococcus aureus secretes EVs which can be delivered into macrophage cells, stimulating a potent IFN-β response in recipient cells [12]. In addition to feces, the presence of bacterial EVs has also been studied in human breast milk, suggesting a role in the vertical transfer of the fecal microbiota [13]. Furthermore, milk EVs have been demonstrated to be an important source of mRNA and therefore have important potential as a tool for monitoring the clinical stage of bovine leukemia virus infection [14]. However, less data are available in relation to the EVs present in human feces, their metagenomic profiling, their differences according to different diseases associated with an intestinal dysbiosis and their effects on intestinal permeability. In this study, we first implemented a procedure to isolate fecal-microbe-derived EVs from human feces. To exclude the presence of free DNA in the EV samples, we also investigated the use of a PMA treatment. Second, we compared the fecal microbiota composition with the composition of the fecal-microbe-derived EVs using 16S ribosomal DNA sequencing in healthy subjects and in patients with diarrhea, morbid obesity and CD. Finally, we tested the effect of these EVs on the cellular permeability of Caco-2 cells in vitro. The performance of the purified EVs with qEVoriginal size exclusion columns and PMA treatment was evaluated prior to sequencing and statistical analysis (Figure 1A). Fecal EV purification with qEV IZON columns normally removes free DNA from the samples; however, the objective of this assay was to verify if the final concentration of the EVs (and therefore the DNA concentration) was enough to perform a good-quality sequencing analysis. Therefore, this test allowed us to determine if the purified EVs from feces had too much free DNA, which might have interfered in sequencing results, and if additional PMA treatment was necessary for these samples. The PMA treatment was applied as previously described, and sequencing and statistical analysis were performed. The α-diversity metrics, including the number of observed genera, Chao1, the reciprocal Simpson index and Simpson evenness, were used to assess community richness and diversity. Good’s coverage was >0.99 for all samples, indicating that although the number of generated sequence reads (on average, 7000) was limited, this sampling effort allowed for the production of an accurate caption of the fecal-microbe-derived EV communities. No significant differences in bacterial richness, diversity or evenness were observed at genus level, regardless of whether the PMA treatment was used or not (Supplementary Figure S1). Regarding the microbiota composition, the purified fecal-microbe-derived EVs presented a few significant differences at genus level compared with those treated with PMA. Post-hoc pairwise differences between the two groups (with and without PMA treatment) were detected only in genus Alistipes and Acidibacter, which were found to be increased in samples without PMA treatment (Supplementary Figure S2). These results demonstrated that this treatment is not essential for the characterization of human-fecal-microbe-derived EVs when following the protocol implemented in the present study. After isolation with qEV IZON columns and without PMA treatment, the fecal-microbe-derived EVs showed a typical particle shape and size when analyzed by NTA (Figure 1B) and TEM (Figure 1C). The Western blot analyses revealed the presence a band of bacterial peptidoglycan (Figure 1D). We further compared the microbial profile of the feces from which the EVs were isolated with the microbial profile of their derived EVs. First, a metagenomics analysis of feces was performed to study the microbial profile. The bacterial EVs were purified using qEVoriginal size exclusion columns, and PMA treatment was not performed. Once the EVs were isolated, a metagenomics analysis was performed to sequence and identify the genera from which these EVs originated. The 16S amplicon sequencing yielded 10,000 cleaned reads per sample from which taxonomic identification was obtained. No significant differences were found in bacterial richness (Chao1 richness index), alpha diversity (inverse Simpson index) and evenness (derived from Simpson index) between the feces and fecal-microbe-derived EVs when all samples were compared together (Figure 2). When the same analysis was performed by group of patients (feces vs. fecal-microbe-derived EVs from the control group and feces vs. fecal-microbe-derived EVs from a group of patients with disease (CD, diarrhea or morbid obesity)), no significant differences were observed between the fecal microbiota and the fecal-microbe-derived EVs after multiple comparisons using the Kruskal–Wallis test with Benjamini–Hochberg FDR corrections (Supplementary Figure S3). AMOVA and HOMOVA analyses showed that the genetic diversity in the fecal-microbe-derived EVs was significantly different from that from fecal bacteria (p = 0.037); however, the amount or variation of this genetic diversity in each group (fecal bacteria and EVs) was not significantly different (p > 0.05). Finally, the NMDS and dbRDA analyses are shown in Figure 3A and Figure 3B, respectively. Twenty-one genera presented a relative abundance greater than 1% in both types of samples: fecal bacteria and EVs (Figure 4A(i)). Seven dominant genera with a relative abundance >3% in both groups were identified: namely, Bacteroides, Faecalibacterium, Prevotella_9, Romboutsia, Escherichia-Shigella, Streptococcus and Laschnospiraceae_ge. Significant differences were observed in 18 different genera (Figure 4A(ii)). Among them, only two taxa, identified as Oscillibacter and Saccharimonadaceae_ge, were increased in the EV samples, while the remaining 16 genera were significantly increased in the feces samples The comparison between fecal bacteria and EVs obtained only from the control group revealed the presence of 17 genera with a relative abundance greater than 1% and 6 dominant genera with a relative abundance of >3% in fecal bacteria and the EV samples, which were identified as Bacteroides, Prevotella_9, Prevotellaceae_NK3B31_group, Dialister, Alistipes and Parabacteroides (Figure 4B(i)). Few significant differences were observed between the composition of the fecal bacteria and EVs with only five genera implicated, including Phascolarctobacterium, Rikenellaceae_RC9_gut_group, Pseudomonas, Veillonella and Veillonellaceae_ge. Almost all of these genera were increased in the fecal bacteria and reduced in the EVs with the exception of Rikenellaceae_RC9_gut_group and Pseudomonas, which were found in higher proportions in the EVs (Figure 4B(ii)). The relative abundance of bacterial genera in the fecal microbiota and fecal-microbe-derived EVs for each patient is shown in Supplementary Figure S4. Finally, we compared the composition of the microbiota from fecal bacteria and EVs in the groups of patients with different diseases (morbid obesity, CD and diarrhea groups together). In the fecal microbiota samples, eleven taxa were identified as dominant, with a relative abundance greater than 3% (Figure 4C(i)), while only five genera were observed in these proportions in the EVs (Faecalibacterium, Prevotella_9, Romboutsia, Bacteroides and Parabacteroides). Several significant differences between the fecal composition and the composition of the EVs were detected, with 20 different genera implicated (Figure 4C(ii)). Only Saccharimonadaceae_ge was found to be increased in EVs, while the remaining 19 genera were all increased in the fecal bacterial samples. The relative abundance of bacterial genera in the fecal microbiota and fecal-microbe-derived EVs per type of disease is shown in Figure 4D (Crohn’s disease), 4E (morbid obesity) and 4F (diarrhea group). The relative abundance of bacterial genera in the fecal microbiota and fecal-microbe-derived EVs for each patient is shown in Supplementary Figure S4. Table 1 shows the significant differences found in the composition of EVs between the four groups of patients. A group of genera was increased (Bacteroidales and Pseudomonas) or decreased (Faecalibacterium, Ruminococcus, Clostridium and Subdoligranum) in EVs from control patients with respect to the rest of the groups. Other genera were also found to be decreased in the control patients with respect to most of the groups (Table 1). There were also some genera that were exclusively increased or decreased, depending on the type of disease, when they were compared to the control group (marked with * in Table 1). Moreover, our findings showed that Tyzzerella, Verrucomicrobiaceae, Candidatus_Paracaedibacter and Akkermansia were increased in EVs from the CD group compared to the morbid obesity and diarrhea groups. In addition, Parabacteroides was increased in EVs from the morbid obesity group compared to the CD and diarrhea groups. No other significant differences were found. We first tested whether fecal EVs induced an alteration of the intestinal permeability of Caco-2 cells by measuring TEER and FD4. Fecal EVs from the different groups of patients were used. We found that Caco-2 cells incubated with fecal EVs from the control patients presented an increase of 29.1 ± 4.0% in the TEER value (Figure 5A). However, Caco-2 cells incubated with fecal EVs from patients with morbid obesity, CD and diarrhea presented an increase of 12.1 ± 3.52%, 9.9 ± 1.7% and 4.4 ± 1.6%, respectively, in TEER values. The change produced by EVs from patients with diarrhea was significantly lower than the change produced by fecal EVs from the control patients (p = 0.0 45). Next, we measured paracellular permeability by monitoring the flux of FD4 through the Transwell. As shown in Figure 5B,C, the fecal EVs from control group did not exert a significant effect on the permeability. Fecal EVs from patients with morbid obesity and CD induced a slightly significant increase in the permeability of Caco-2 cells at 30 min. However, the fecal EVs from patients with diarrhea induced the highest increase at each time in the translocation of FD4 to the basolateral compartment when compared to the control group. Moreover, the increase found with the diarrhea fecal EVs was also significantly higher than with the fecal EVs from patients with morbid obesity. Therefore, the permeability of the Caco-2 cells was modified by fecal EVs according to their origin. Sequencing methods do not discriminate between live (dormant cells and non-growing or growing cells, which are metabolically active) or dead bacteria. In our study, the analysis of the PMA-treated EVs did not present enough differences in richness, alpha diversity or Good’s coverage to be statistically different from those that were not treated with PMA. This method is recognized as a valuable tool for the distinction of dead/viable cells since PMA treatment is a DNA-intercalating agent that acts on free DNA and penetrates cells with compromised membranes [14]. Regarding the composition of EVs, significant differences were only found for two genera, indicating that most populations can be found in the same proportions in EVs treated and not treated with PMA. Therefore, this treatment is not essential for the metagenomics analysis of fecal-microbe-derived EVs from human feces when the protocol implemented in the present study is followed. The use of qEV original size exclusion columns for the purification of fecal EVs appears to be sufficiently efficient to remove free DNA from the fecal EVs. Microbe-derived EVs have been directly associated with disease development [15]. However, there are few studies on the composition of fecal-microbe-derived EVs compared with their feces of origin, and a large proportion of these studies focused on colorectal cancer and inflammatory bowel disease (IBD) patients [16,17,18]. In our study, the overall bacterial richness and diversity were not significantly different between the two types of samples studied (fecal bacteria and fecal-microbe-derived EVs) within the control group or within patients with disease. However, differences were detected in the microbial composition of EVs in relation to the fecal microbiota. In general, higher proportions of various genera were found in the fecal microbiota compared with the EVs. Previous studies have also demonstrated that the protein composition of fecal EVs differs from that of fecal samples in IBD patients [16]. These findings may suggest that the different bacterial genera secrete variable proportions of EVs which, in turn, may be influenced by the patient’s intestinal disease. These fecal-microbe-derived EVs may be used as novel biomarkers to detect various intestinal diseases, as proposed by Park for colorectal cancer [15]. We also observed more differences in the microbiota structure between the fecal bacteria and EVs in patients with disease than in the control group. However, the main limitation of this study is the low number of recruited patients in each group, which did not allow us to describe the fecal-microbiota-derived EVs that are candidates for predicting each disease. Nevertheless, our results provide preliminary data to further study how the composition of these fecal-microbe-derived EVs is modified in different diseases and to analyze whether these EVs may be involved in the microbiota–host interaction. Previously, significant compositional differences were demonstrated in obese and diabetic rats compared to normal rats in terms of the composition of microbial EVs [19]. Another study also showed that the composition of intestinal EVs was greatly altered after vertical sleeve gastrectomy in mice [20]. As bacteria proliferate, the secretion of EVs should increase in line with the increase in the relative abundance of taxa [10]. However, it is possible that bacteria, depending on the group to which they belong, are capable of producing a greater or lesser number of EVs. Furthermore, this production may be influenced by the presence of other bacterial communities and by the physiological conditions of the environment. This could be a hypothesis to explain the increase in EVs in certain bacterial groups with respect to their percentage in fecal microbiota. In this study, the control patients presented high proportions of Pseudomonas and Rikenellaceae_RC9_gut_group in fecal-microbe-derived EVs but lower proportions of Phascolarctobacterium, Veillonella and Veillonellaceae_ge when compared with the fecal bacterial samples. The Pseudomonas genus, specifically Pseudomonas fragi, also commonly produced important levels of EVs during growth. These vesicles display considerable proteolytic activity but are not associated with bacteriocinogenicity. They most likely act in the physiological distribution of extracellular proteinases [21]. Regarding Rikenellaceae_RC9_gut_group, only one previous study described an increase in their EVs in patients with colorectal cancer [15]. In addition, a high proportion was found after fecal microbiota transplantation upon Salmonella Enteritidis infection in chicks [22]. Moreover, the supplementation with probiotics in broilers had a promoting effect on the growth performance and increased the colonization of beneficial bacteria in the cecum as Rikenellaceae_RC9_gut_group [23]. It is involved in degrading carbohydrates [24] and metabolizes lipids [25]. In addition, members of the Veillonellaceae family are often found in association with gut inflammation [26] and are more abundant in patients with IBD, fibrosis and other diseases [27,28]. Taken together, these data seem to suggest that the EVs derived from certain bacteria might have an important role in the maintenance of intestinal homeostasis. In our disease patients, we observed differences between the composition of the fecal bacteria and EVs in a total of 20 genera, all of which showed decreased proportions in EVs except Saccharimonadaceae_ge, which was increased in the EVs. In the literature, there is no specific information about the presence of Saccharimonadacea-derived EVs in the feces of patient; therefore, its role in the gut requires further investigation. EVs belonging to other genera that were found to be decreased in our study have been previously studied due to their possible effects on intestinal diseases. An example of this includes Akkermansia (A. muciniphila)-derived EVs, which have been reported to act as a functional moiety for controlling gut permeability and regulating the intestinal barrier integrity in mice [8]. Other important bacterial groups that showed significant differences between the composition of the fecal microbiota and EVs are Lactobacillus and Bifidobacterium. The EVs of Lactobacillus plantarum Q7 have been demonstrated to alleviate induced colitis symptoms and histological damage in mice. They also reduced the levels of proinflammatory bacteria (Proteobacteria) and increased the levels of anti-inflammatory groups (Bifidobacterium and Muribaculaceae) [29]. The EVs of Bifidobacterium longum can export several cytoplasmic proteins that could be involved in bifidobacterial adhesion and survival in the gastrointestinal tract [30]. Our in vitro experiment demonstrated that fecal EVs act as regulators of epithelial barrier integrity with differences depending on the disease of the patients. This is in accordance with the different compositions of fecal-microbe-derived EVs that are dependent on the type of disease. In contrast to most studies, our findings describe the effects of EVs from a mixture of fecal bacteria, not from a specific bacterium. Several studies with different species of fecal microbiota have demonstrated the role of bacterial-derived EVs as modulators of epithelial barrier integrity [31]. In this context, the fecal microbiota-derived EVs, besides the host-derived EVs [31], could be involved in the regulation of gut homeostasis by enhancing the intestinal permeability, a condition that subsequently leads to inflammatory and metabolic diseases [32]. This increased gut permeability would allow for the passage of endotoxins and luminal antigens into the intestinal lamina propria, initiating a mucosal immune response that causes chronic, low-grade inflammation, prompting metabolic disorders such as insulin resistance and obesity [31]. Possible differences in the surface cargo molecules, such as microbe-associated molecular patterns (MAMPs), could be mediating the adhesion of these fecal-microbiota-derived EVs to host epithelial cells and, consequently, the downstream effects [33]. Moreover, in a later study, it would be interesting to analyze the metabolic and transcriptomic changes produced by these fecal EVs in different types of cells. A limitation of this study was that we did not characterize the total composition of these EVs, i.e., we did not analyze their protein, RNA, DNA and lipid contents. These factors could be associated with the effects produced by these EVs. In this study, we only focused on analyzing the genera from which the EVs originated by metagenomic analysis. In addition, although the method used in this study to isolate the EVs has been previously described and used [34,35,36], it is possible that it could be improved by performing EV isolation prior to freezing in order to minimize the presence of intracellular artifacts/contaminants from microorganisms/cells derived from the freezing process. This point will require further study to analyze the differences between these two methodologies. In summary, we conducted a metagenomic study to reveal associations between the fecal microbiota and the microbial composition of EVs in control subjects and in patients with disease. We found that fecal-microbiota-derived EVs from control subjects have a metagenomic profile closely similar to that of the fecal microbiota. However, we have shown that the presence of a dysbiotic fecal microbiota in different diseases is accompanied by an altered composition of fecal-microbe-derived EVs. Therefore, our findings demonstrate that diseases such as diarrhea, CD or morbid obesity alter the microbial composition of EVs in relation to the fecal microbiota. On the other hand, we found an increase in intestinal permeability with fecal EVs from patients with different diseases. We suggest that the fecal-microbiota-derived EVs from certain bacteria might cause increased intestinal permeability as part of their infectious mechanisms, while other bacterial strains attenuate inflammation and reinforce the gut barrier integrity [37]. We postulated the importance of controlling the balance between the different subsets of fecal microbiota and their EVs in the development of diseases associated with altered intestinal permeability. However, the cause-and-effect relationships and the role of these fecal-microbiota-derived EVs, as mediators of interspecies interactions and as novel biomarkers, in the course of a disease require future careful, experimental studies. Our cohort study included 32 patients: 9 healthy volunteers, 10 diarrheic patients, 9 patients with morbid obesity and 4 patients with CD. These diseases were chosen because they demonstrates a clear alteration of fecal microbiota [38,39,40]. In those patients with diarrhea, neither parasites nor Cryptosporidium were isolated in the feces, they had normal flora, no Salmonella, Shigella, Campylobacter, Yersinia and Aeromonas were isolated, and the presence of Clostridium difficille toxin and adenovirus and rotavirus antigens was negative. Fecal samples were collected from all patients (n = 32) and immediately stored at −80 °C in the Virgen de la Victoria University Hospital Biobank (Andalusian Public Health System BioBank) until analysis. All participants were of Caucasian origin. All participants gave their written informed consent, and the study protocol was carried out in accordance with the ethical guidelines of the Declaration of Helsinki. The study was approved by the Malaga Provincial Research Ethics Committee, Malaga, Spain (PI18/01652, PE-0098-2019). A total of 10 g of feces was inoculated into 40 mL of sterile, phosphate-buffered saline (PBS) and homogenized. The EVs were then isolated through centrifugation as previously described with some modifications [10]. Briefly, a first centrifugation of the homogenate was performed (40 min, 4000× g, and 4 °C. The supernatant was recovered and filtered using sterilized vacuum filtration units, Rapid-Flow™ filters MF 75, 1000 mL of Nalgene® and 0.2 μm of cold ice (Thermo Fisher Scientific, Waltham, MA, USA). The filtrate was transferred to 10 mL polycarbonate, open-top, thick-wall tubes and ultracentrifuged at 100,000× g for 3 h at 4 °C with a fixed-angle rotor (Type 70.1 Ti) in a Beckman Optima XL-100K ultracentrifuge (Beckman Coulter Life Sciences, Indianapolis, IN, USA). Pellets were resuspended in 200 µL of PBS and the EVs were purified using qEVoriginal size exclusion columns of 70 nnm (Izon Science Europe Ltd., Oxford, UK), following the manufacturer’s recommendations. Fractions 6–8 (enriched in EVs) were collected, mixed, concentrated with Vivaspin® 6 100K centrifugal concentrators (Sartorius AG, Göttingen, Germany), aliquoted and frozen at −80 °C until use. This protocol was used to separate bacteria and other contaminating soluble molecules, such as toxins and proteins, from the EVs. These aliquots of fecal EVs were used for treatment with propidium monoazide, metagenomic analysis, transmission electron microscopy, nanoparticle tracking analysis, Western blot and for the incubation of Caco-2 cells. The isolated EVs (n = 4; one from a healthy control, one from a CD patient, one from a diarrheic patient and one from a patient with morbid obesity) were fixed in 2% paraformaldehyde—0.1 M PBS for 30 min. A glow discharge technique (60 s, 7.2 V, using a Bal-Tec MED 020 Coating System) was applied over carbon-coated copper grids, and these grids were immediately placed on top of sample drops for 15 min. Then, the grids with adherent EVs were washed in a 0.1 M PBS drop. Additional fixation in 1% glutaraldehyde was performed for 5 min. After washing the grids properly in distilled water, the grids were contrasted with 1% uranyl acetate and embedded in methylcellulose. Excess fluid was removed and allowed to dry before examination with a transmission electron microscope FEI Tecnai G2 Spirit (ThermoFisher Scientific, Waltham, MA, USA). All images were acquired using a Morada digital camera (Olympus Soft Image Solutions GmbH, Münster, Germany). The magnification used for the TEM images was 49,000×. The EV size and concentration were assessed using the NanoSight NS300 system (Malvern Panalytical, Malvern, UK) (n = 3; one from a morbidly obese patient, one from a diarrheic patient and one from a healthy control). Particles were automatically tracked and sized-based on Brownian motion and the diffusion coefficient. The EVs were resuspended and diluted with 0.22 μm filtered PBS at a concentration range 109 particles/mL, and 1 mL was used for NanoSight analysis. Five replicates of 30 s videos were captured to analyze the concentration and size distribution of the EVs at the detection threshold of 5. A data analysis was performed using NanoSight analysis software. Fecal EVs (n = 3; one from a CD patient, one from a morbidly obese patient and one from a healthy control) were lysed with 1× RIPA buffer (Thermo Fisher (Kandel) GmbH, Kandel, Germany) and supplemented with a protease inhibitor cocktail (Merck KGaA, Darmstadt, Germany). The protein lysate was incubated with the same volume of Laemmli Buffer 2× (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and supplemented with 2-mercaptoethanol (5%) at 95 °C for 5 min. The samples were subjected to 4–20% SDS-PAGE (NB12-420) (NuSep, Inc., Germantown, MD, USA) and transferred onto polyvinylidene fluoride membranes (Trans-Blot Turbo Midi 0.2 µm PVDF Transfer Packs) (Bio-Rad Laboratories, Inc., Hercules, CA, USA) at 13 V and 1.1 A for 20 min. The membranes were subsequently blocked in PBS–bovine serum albumin (BSA) 5% for 1 h at room temperature. The membranes were then incubated for 48 h at 4 °C with a mouse monoclonal anti-bacterial peptidoglycan antibody, clone 3F6B3 (Merck KGaA, Darmstadt, Germany). This antibody is specific to the three-dimensional polymer complex structure of bacterial peptidoglycan. The membranes were washed three times with 0.05% Tween-20 washing buffer in PBS and incubated with a horseradish-peroxidase-conjugated secondary antibody (VeriBlot for IP Detection Reagent (HRP), ab131366) (Abcam, Cambridge, UK) for 3 h at room temperature. Finally, after another three washes, the membranes were revealed with Clarity Western ECL substrate (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The proteins were visualized by an ImageQuant LAS 4000 (GE Healthcare, Buckinghamshire, UK). The procedure used to isolate the microbe-derived EVs could result in the presence of a small percentage of free DNA in the sample. Therefore, we tested sample treatment with PMA [41] in order to detect the co-extraction and amplification of the nonprotected DNA of the membrane-compromised EVs. Furthermore, we wanted to evaluate the optimal separation of EVs from free DNA using qEVoriginal size exclusion columns of 70 nnm (Izon Science Europe Ltd., Oxford, UK). For this assay, seven samples from patients were evaluated (two from healthy volunteers, two from diarrheic patients, two from morbidly obese patients and 1 from a patient with CD). In total, 100 µL of each sample was centrifuged at 5000× g in duplicate from which one was left untreated and the other one was treated with PMA (PMAxx™ dye) (Biotium, Fremont, CA, USA) prior to DNA extraction (Figure 1A). The manufacturer’s protocol for PMA treatment was used and involved the use of the PMA-Lite™ LED Photolysis Device (Biotium, Fremont, CA, USA). The statistical analyses were performed with the seven samples tested. The total DNA was extracted from the EVs that were treated and not treated with PMA (Izon Science Europe Ltd., Oxford, UK) and directly from the fecal samples using DNeasy blood and Tissue Kits (QIAGEN Science, Hilden, Germany), following the manufacturer’s recommendations. Briefly, after the isolation and purification of the EVs from feces, DNA extraction was performed. Once this DNA was obtained, seven of these DNA samples from the EVs were aliquoted in duplicate to treat one half with PMA. In parallel, a DNA extraction was also performed in all fecal samples used for the isolation of EVs. The DNA was eluted into DNase/RNase-free water and its concentration and purity were evaluated using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Inc., Wilmington, DE, USA). Extracts were stored at −20 °C until use. Libraries and sequencing were performed as previously described [42]. Briefly, amplification of the V1-V3 regions of the 16S rRNA bacterial gene was performed using the primers 5′-GAGAGTTTGATYMTGGCTCAG-3′ forward and 5′-ACCGCGGCTGCTGGCAC-3′ reverse with overhand adapters. Amplicons were purified using Agencourt AMPure XP bead kit (Beckman Coulter, Pasadena, CA, USA), indexed using Nextera XT index primers 1 and 2 (Illumina, San Diego, CA, USA), quantified by Quant-IT PicoGreen (Thermo Fisher Scientific, Waltham, MA, USA) and diluted to a concentration of 10 ng/μL. DNA samples were quantified by qPCR with a KAPA 170 SYBR®® FAST qPCR Kit (Kapa Biosystems, Wilmington, MA, USA). Samples were normalized, pooled and sequenced using Illumina MiSeq technology with v3 reagents (Illumina, San Diego, CA, USA), using paired end reads by GIGA Genomics platform (Liège, Belgium). A bacterial community composed of known proportions of Carnobacterium maltaromaticum, Lactococcus lactis subsp. cremoris, Leuconostoc carnosum, Pseudomonas aeruginosa and Streptococccus thermophilus was used as a positive control. Negative controls were used in their entirety for DNA extraction, library preparation and sequencing. Sequence reads were processed using Mothur v1.44.3 and VSearch for alignment, clustering and chimera detection, respectively [43,44]. The sequences were clustered into operational taxonomic units (OTUs) at an identity of 97%. The SILVA 138 database of full-length 16S rDNA gene sequences was used for the alignments of unique sequences and taxonomical assignations. For each sample, a subsampling dataset containing 10,000 representative, cleaned reads was retained (mean: 10,000, SD: 0) and used to generate OTUS (cut off: 0.03) as well as to evaluate several ecological indicators. All statistical analyses were performed at the genus level. Regarding alpha diversity (reciprocal Simpson diversity index and Simpson evenness), Goods’s coverage and population richness (Chao1 estimator of richness) were calculated using Mothur v1.44.3 and compared between two groups using a Wilcoxon matched-pairs signed rank test (PRISM 8) (GraphPad Software, Boston, MA, USA) or between three or more groups using Kruskal–Wallis multiple testing with Benjamini–Hochberg FDR corrections (PRISM 8) (GraphPad Software, Boston, MA, USA). Bar plots were built using PRISM 8, including only genera with a relative abundance >1%. The β-diversity was estimated with the Bray–Curtis dissimilarity index using Mothur (v1.44.3) and R for graphical analysis (v1.2.5033). Non-metric multidimensional scaling (NMDS) was performed using Mothur and was considered satisfying when the stress value was <0.20. An AMOVA (analysis of molecular variance) and a HOMOVA (homogeneity of molecular variance) were performed using Mothur in order to reveal eventual significant population structure differences and to determine if the genetic diversity within two or more populations was homogeneous [44]. A distance-based redundancy analysis (dbRDA) was constructed using RStudio. Post-hoc pairwise differences between groups were assessed with Deseq2 package in R, and differences were then identified with Kruskal–Wallis tests using Benjamini–Hochberg FDR correction [45]. Caco-2 (ECACC, Cat. No. 09042001) epithelial cell lines were maintained in complete medium (Dulbecco’s modified Eagle’s Medium (DMEM) of high glucose with L-glutamine (Biowest, Nuaillé, France) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Biowest, Nuaillé, France), 1% penicillin/streptomycin (Biowest, Nuaillé, France) and 1% MEM non-essential amino acids (Sigma-Aldrich, St. Louis, MO, USA) under standard conditions inside a humidified cell culture incubator at 37 °C with 5% CO2. Caco-2 cells were harvested by washing three times in sterile DPBS, followed by treatment with trypsin-EDTA. Harvested cells were counted and seeded in 12-well PET Transwell™ inserts of 0.4 μm pore size (Corning Inc., Corning, MA, USA) at 105 cells/insert by adding 0.5 mL of cell suspension. The apical and basal cell culture media, 0.5 mL and 1.5 mL respectively, were changed every two days. Cells were maintained for approximately 3 weeks in the same medium to allow for full cell differentiation. The culture medium was changed, and 1 μg of protein from the purified EVs suspension was added for 24 h of incubation [8]. The protein concentration of the purified fecal EV suspension was determined using the bicinchoninic acid (BCA) assay (Thermo Fisher Scientific, Waltham, MA, USA). After 24 h of incubation with fecal EVs (n = 4 for each group of patients), the trans-epithelial electrical resistance (TEER) and para-cellular permeability were measured to analyze the EV-induced changes. TEER was measured using a Millicell®® ERS-2 Voltohmmeter (Merck Millipore, Burlington, MA, USA). Once Caco-2 cells reached a TEER > 1000 (Ω·cm2), experiments with the purified fecal EV suspension were performed as described above. TEER values were obtained by subtracting cell-free filter readings and correcting for the surface area (1.1 cm2). All readings of TEER were repeated across triplicate sample Transwells. TEER values were expressed as the percentage of change with respect to the TEER value obtained prior to the incubation with purified fecal EV suspension. Data were presented as means ± SEM (n = 4). The Caco-2 monolayer paracellular permeability was assessed by measuring the unidirectional flux of fluorescein isothiocyanate (FITC)-dextran (FD4; 4000 Da, Sigma-Aldrich, Saint-Louis, MO, USA) from the apical to the basolateral compartments of the Transwell™. The complete DMEM medium was removed from the apical and basolateral compartments, replaced with Krebs Ringer Bicarbonate Buffer Hepes Albumin (KRBHA), and equilibrated for 1 h at pH 7.4. The KRBHA medium was replaced again, and 25 mg/mL stock solution of FD4 was added to the apical compartment at time zero to obtain a final concentration of 1 mg/mL. An aliquot of 100 μL from the basolateral compartment was removed every 30 min over 2 h, followed by replacement with fresh KRBHA. Samples were transferred onto Nunclon®® MicroWell plates (Thermo Scientific, MA, USA) and the fluorescence of FD4 was measured in a microplate fluorescence reader (FLx 800, Bio-tek Instruments Inc., Winooski, VT, USA) with an excitation of 485/20 nm and an emission of 528/20 nm. A negative control was performed with the Caco-2 cells without fecal EV treatment. A positive control was performed with the Caco-2 cells and 5 mM EGTA instead of fecal EVs. EGTA causes a breakdown of the tight junctions by sequestering bivalent ions independently of inflammatory stimuli [46]. Based on the relative fluorescence units, FD4 concentrations were expressed as the percentage of change from Caco-2 cells without fecal EV treatment. All the results were analyzed in triplicate. Data were presented as means ± SEM (n = 4). All data were analyzed with GraphPad Software (Prism 8.1.1) (GraphPad Software, San Diego, CA, USA). Differences between groups were compared using Kruskal–Wallis tests followed by post hoc analyses using Dunn’s test. Values were considered to be statistically significant when p < 0.05.

PMC10002484

Mariaimmacolata Preianò,Serena Correnti,Tahreem Arshad Butt,Giuseppe Viglietto,Rocco Savino,Rosa Terracciano

Mass Spectrometry-Based Untargeted Approaches to Reveal Diagnostic Signatures of Male Infertility in Seminal Plasma: A New Laboratory Perspective for the Clinical Management of Infertility?

23-02-2023

mass spectrometry,proteome,seminal plasma,male infertility,seminal fluid,biomarker,laboratory medicine

Male infertility has been recognized as a global health problem. Semen analysis, although considered the golden standard, may not provide a confident male infertility diagnosis alone. Hence, there is the urgent request for an innovative and reliable platform to detect biomarkers of infertility. The rapid expansion of mass spectrometry (MS) technology in the field of the ‘omics’ disciplines, has incredibly proved the great potential of MS-based diagnostic tests to revolutionize the future of pathology, microbiology and laboratory medicine. Despite the increasing success in the microbiology area, MS-biomarkers of male infertility currently remain a proteomic challenge. In order to address this issue, this review encompasses proteomics investigations by untargeted approaches with a special focus on experimental designs and strategies (bottom-up and top-down) for seminal fluid proteome profiling. The studies reported here witness the efforts of the scientific community to address these investigations aimed at the discovery of MS-biomarkers of male infertility. Proteomics untargeted approaches, depending on the study design, might provide a great plethora of biomarkers not only for a male infertility diagnosis, but also to address a new MS-biomarkers classification of infertility subtypes. From the early detection to the evaluation of infertility grade, new MS-derived biomarkers might also predict long-term outcomes and clinical management of infertility.

Mass Spectrometry-Based Untargeted Approaches to Reveal Diagnostic Signatures of Male Infertility in Seminal Plasma: A New Laboratory Perspective for the Clinical Management of Infertility? Male infertility has been recognized as a global health problem. Semen analysis, although considered the golden standard, may not provide a confident male infertility diagnosis alone. Hence, there is the urgent request for an innovative and reliable platform to detect biomarkers of infertility. The rapid expansion of mass spectrometry (MS) technology in the field of the ‘omics’ disciplines, has incredibly proved the great potential of MS-based diagnostic tests to revolutionize the future of pathology, microbiology and laboratory medicine. Despite the increasing success in the microbiology area, MS-biomarkers of male infertility currently remain a proteomic challenge. In order to address this issue, this review encompasses proteomics investigations by untargeted approaches with a special focus on experimental designs and strategies (bottom-up and top-down) for seminal fluid proteome profiling. The studies reported here witness the efforts of the scientific community to address these investigations aimed at the discovery of MS-biomarkers of male infertility. Proteomics untargeted approaches, depending on the study design, might provide a great plethora of biomarkers not only for a male infertility diagnosis, but also to address a new MS-biomarkers classification of infertility subtypes. From the early detection to the evaluation of infertility grade, new MS-derived biomarkers might also predict long-term outcomes and clinical management of infertility. Infertility is a global health issue defined by the inability to conceive after 12 months or more of regular unprotected sexual intercourse [1]. It affects approximately 15% of reproductive-aged couples worldwide and a contributing male factor may be found in about half of the cases, either alone or in combination with female causes [1,2,3]. To date, semen analysis is the cornerstone for the routine evaluation of male fertility with WHO guidelines providing the basis for the standardization of laboratory procedures and reference values worldwide [4,5]. However, although providing valuable information, semen analysis alone is not sufficient to accurately assess male fertility potential or to distinguish fertile subjects from infertile ones [3,6,7]. Moreover, standard seminal analysis fails to properly identify etiological factors and to unravel the molecular and pathophysiological basis of male reproduction diseases. Male fertility disorders require a more in-depth analysis, especially in the case of unexplained male infertility, a condition with unknown etiology which affects roughly 30% of men with normal semen parameters [3,6,8,9,10]. Other advanced sperm function tests such as sperm DNA damage and oxidative stress assays were recently introduced in routine clinical evaluations to predict reproductive outcomes in a more accurate way [4]. However, these additional screening tools still fail to explain the underlying mechanisms at a subcellular level that are associated with the different infertility phenotypes [11,12]. Hence, in the era of precision medicine, ‘omics’ technologies are in constant development, exploring for new, reliable and disease-specific biomarkers, with the aim of improving the diagnosis and prognosis of male fertility disorders and to select the best fitting therapeutic actions. In this scenario, proteomics has emerged as an important tool for providing insights into the underlying molecular processes associated with male infertility. Proteomics technologies could overcome gaps in information from standard semen analysis and other sperm quality tests with limited diagnostic value [13,14]. The progress in research technology and techniques in the field of proteomics led to the development of innovative and reliable platforms for the non-invasive biomarker-based male infertility diagnosis. The improved resolution, sensibility and accuracy of the mass spectrometry (MS) platforms allowed the high-throughput characterization of proteins associated with male fertility disorders [15,16]. In recent years, due to the presence of tissue-specific molecular mediators, an increased amount of MS-based proteomics investigations was carried out on semen and seminal plasma (SP) to better identify disease-specific biomarkers of male infertility [14,17,18,19,20,21]. In particular, the characterization of proteome profile of SP by MS-based approaches, has revealed its key role in reproductive processes and its potential as a screening, diagnostic and prognostic instrument in male fertility assessment [22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37]. In fact, SP that for a long time was considered only a passive medium for spermatozoa transport and protection is highly enriched in proteins, RNAs, lipids and other metabolites. All these molecules exert important effects over sperm function and male fertility, capturing growing interest for their potential as clinical samples for non-invasive diagnostics [14,17,38]. The huge amount of MS-based proteomic data subjected to bioinformatic analysis has provided extensive examination about distribution, molecular and functional analysis for the SP candidate biomarkers [13,39]. This review provides an overview of the main MS-based untargeted approaches for comparative proteomics between fertile and infertile men categorized according to their quantitative and/or qualitative alteration of seminal parameters. In particular, key MS techniques and strategies adopted (bottom-up or top-down) to analyze the SP proteome with a rapid overview on the major proteomics findings and candidate biomarkers for male infertility diseases are highlighted. Challenges and limitations of these approaches, which could accelerate the development of new MS laboratory diagnostics applications in the area of clinical management of male infertility, will be discussed. Semen is a complex body fluid, which contains a heterogeneous mixture of components produced by different sex accessory glands, including the seminal vesicles, the prostate gland and the bulbourethral glands [40]. Semen is released during ejaculation and is composed of two major fractions: the cellular and the fluid fractions. The cellular fraction is composed of spermatozoa, which are produced in the testes and stored in epididymides, while the fluid fraction contains different liquid secretions, which contribute to generate the SP. Spermatozoa accounts only for approximately the 5% of the whole semen volume, while SP represents the remaining 95% [17]. Immediately after ejaculation, human semen appears as a thick fluid, with a gelatinous structure, referred to as the coagulum, which traps and immobilize the spermatozoa. This coagulum is typically liquefied within 15–30 min, mainly by the activity of the prostate-specific antigen (PSA or KLK3), a chymotrypsin-like serine protease, which hydrolyzes highly abundant proteins forming the coagulum (Semenogelins I and II, fibronectin). These proteolytic cleavages allow the liquefaction of the seminal clot and the increase in spermatozoa motility, which are able to reach the female reproductive tract [41,42]. Other enzymes which participate in the semen liquefaction include KLK2, KLK5 and KLK14, which have been reported to hydrolyze fibronectin and Semenogelins in ex vivo and in vitro studies, and also KLK6, KLK7 and KLK13, which showed catalytic capacity toward fibronectin [43]. The difference between semen and SP lies in the fact that SP is only the fluid portion of the entire ejaculate, while semen comprises both the spermatozoa and SP. SP is the supernatant, easily obtained after the centrifugation of the liquefied semen and the removal of sperm cells and cell debris, which constitute the pellet (Figure 1) [44]. It has a very heterogeneous and complex molecular composition, including lipids, glycans, inorganic ions, metabolites, cell free DNA, RNA, microRNAs, peptides, proteins and oligosaccharides. It contains secretions derived from multiple glands of the reproductive tract, among which seminal vesicles and prostate are the main contributors in terms of volume (~65% and ~25%, respectively). In particular, secretions from seminal vesicles are highly enriched by cytokines, prostaglandins and fructose, sources of energy for spermatozoa [45], while prostate glands secrete a fluid constituted by proteolytic enzymes, citrate, lipids, calcium, magnesium and zinc [46,47]. These secretions also include basic polyamines, which warrant an alkaline environment to the semen, contributing to the survival of sperm cells in the acidic milieu of the vagina. A minor contribution of the seminal fluid volume is represented by the testis and epididymis (~10%), and finally, by bulbourethral (Cowper) and periurethral (Littre’s) glands (~1%). Their secretions also act as lubricants of the semen, allowing a more efficient sperm transfer [48]. SP also contains a large number of extracellular vesicles (EVs) which have heterogeneous dimensions, origin, and “molecular cargo” and are a rich source of proteins [49]. They are especially secreted by the epididymis (epididymosomes) and the prostate (prostasomes) and are implicated in promoting spermatozoa motility, immunomodulation, antibacterial activity and antioxidant protection [50]. Several studies investigated the proteome of the EVs, collecting them by semen centrifugation, the removal of spermatozoa and different ultracentrifugation steps of the supernatant. This allows for exclusively analyzing the proteome associated with these extracellular vesicles [51,52,53,54,55]. Most of the studies on human SP have been focused on the whole SP [25,26,27,28,29,30,31,32,33,34,35,36,37], whose analysis can be affected by the presence of the EVs. Bianchi et al. performed the first functional proteomic study on the vesicle-free (vf) soluble fraction of human SP in normozoospermic healthy donors by combining two-dimensional gel electrophoresis (2-DE), matrix-assisted laser desorption/ionization-time-of-flight (MALDI-TOF) MS and bioinformatic tools for pathway analyses [22]. Proteomic data demonstrated that vf-SP includes a very limited set of highly abundant unique proteins subjected to massive co- and/or post-translational modifications. Functional analysis showed that these proteins were mainly involved in catalytic activity, immune response, apoptotic process, seminal clot regulation, and central nervous system development and morpho-functional maintenance. These results demonstrated that vf-SP represents a catalytic milieu of proteases and their inhibitors which play a key role in semen coagulation and liquefaction. Data analysis also revealed that some vf-SP proteins are probably involved in the modulation of structural and functional properties of spermatozoa during sperm migration in the male and female genital tracts [22]. SP contains numerous proteins, with a concentration that ranges between 35 and 55 mg/mL, making it an accessible source for proteins identification [17,23,44]. Proteomic and functional studies have highlighted the physiological roles of SP proteins. They mediate important functions not only on the sperm activity including metabolism, maturation, motility and capacitation, but also on semen coagulation, liquefaction and fertilization. They are also involved in delivering and providing nutrition to spermatozoa during their travel throughout the male and female reproductive tracts, in the increase of the immune response, interaction with the zona pellucida and modulation of the acrosome reaction [14,56]. All these functions are crucial for natural reproductive success. Because of its functional characteristics, SP reflects the local pathophysiology of the male reproductive system, thus representing an optimal and promising resource for the discovery of biomarkers of male infertility and other related disorders [14,38]. In fact, being located close in proximity to the male reproductive tract, it is highly enriched of specific proteins and peptides, that have a better predictive value as biomarkers for the study of these pathologies than other markers located in serum or urine, where they are much more diluted and less expressed. Therefore, it is easier to identify and quantify them in seminal plasma by analytical techniques [44]. For these reasons, in recent years, SP has gradually captured interest for its promising potential as a clinical sample for non-invasive diagnostics. In fact, as previously outlined in Figure 1, SP is obtained by a non-invasive and safe procedure, based on the simple centrifugation of the semen, collected by masturbation into sterile containers after (3–5) days of sexual abstinence. The discovery of infertility biomarkers in human SP and their use in the clinical setting, might offer a powerful and reliable approach for overcoming and replacing currently invasive surgical techniques and blood or urine-based tests, which have limitations, including a lack of specificity and prognostic significance [44]. SP biomarkers for male reproductive system disorders could also outperform traditional semen analysis and sperm functional tests, which have a poor diagnostic performance and cannot precisely and accurately predict the fertility status of a man alone [6,38]. The human semen liquefaction process is controlled by different factors, including proteases and protease inhibitors, which represent a high percentage of all the proteins identified in human SP [14]. The high number of these components highlights the importance of this system in this body fluid. It is well known that human semen liquefaction is a proteolytic process, occurring after ejaculation and mainly requiring the enzymatic activity of the PSA, also known as kallikrein-related peptidase 3 (KLK3), which changes semen from a gel-like coagulum to a more fluid consistency. This process is fundamental for the spermatozoa to gain their motility and reach the fertilization site in fallopian tubes [43]. There are also other members of the KLKs family which participate in the process, including KLKs 2, 3, 4, 5, 8, 11, 12, and 14 that are secreted by the prostate and act in a protease cascade. Semen liquefaction is also controlled by endogenous inhibitors such as Zn2+, secreted in the prostatic fluid, that maintain KLK3 in an inactive form. After ejaculation, prostatic fluids are combined with seminal vesicle fluids, containing semenogelins, which sequester Zn2+, thus activating KLK3. This performs site-specific cleavages of semenogelins into low molecular weight peptides, determining the dissolution of the coagulum [43]. Residual proteolytic activity if still present in SP might represent one of the major issues in the context of proteome analysis. Robert and colleagues [57] reported the addition of a protease inhibitor cocktail (PIC) containing 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF), a serine protease inhibitor, that blocked PSA activity, in human SP. They demonstrated that, when PSA was treated with the protease inhibitor, the degradation and the hydrolysis of semenogelins were prevented, supporting the notion that PSA is the main semenogelins processing enzyme in the early stage of semen liquefaction. In a very recent study, Correnti et al. [36] investigated the stability of human SP samples, by assessing the variation of the total peak numbers between samples with and without PIC at several time points by MALDI-TOF MS analysis. More precisely, PIC was added to samples after semen liquefaction at 0 and after 60, 90, 120 and 150 min at room temperature and after 1 and 120 days of storage at −80 °C. The samples were then analyzed by MALDI-TOF MS. They observed no statistically significant variation in peaks number neither up to 2.5 h at room temperature nor up to 120 days of storage at −80 °C, and also, no statistically significant difference in peaks number was observed between samples with and without the use of PIC. These data demonstrated that SP samples are stable for at least 2.5 h at room temperature and for at least 4 months when stored at −80 °C and that no residual proteolytic activity is still present in SP after the liquefaction of coagulum [36]. MS has emerged as a key platform for proteomic analyses of human SP, with two main approaches referred to as ‘bottom-up’ (also known as shotgun) and ‘top-down’ proteomics (Figure 2). Most of the studies here reviewed used a bottom-up approach to study the differential expression of SP proteins between fertile and infertile men [25,26,27,28,29,30,31,32,33,34,35], while top-down approach is reported in only two investigations [36,37]. Briefly, in the bottom-up studies, whole proteins extracted are digested into peptides using trypsin (in-solution digestion). Proteolytic cleavage products are then fractionated by mono-dimensional (reverse phase) or bi-dimensional (strong-cation-exchange or strong-anionic-exchange followed by reverse phase) liquid chromatography (LC), and then fractions of peptides are analyzed by MS or MS/MS (Figure 2) [25,26,27,28,30,31,32,33,34,35]. In a more classical approach, extracted proteins are separated by 1D-SDS-PAGE. The bands (1D) of interest are then proteolytically digested (in-gel digestion) and analyzed by LC coupled to MS/MS (LC-MS/MS) (Figure 2) [29]. Unlike the bottom-up strategy, top-down studies [36,37] enable the analysis of intact proteins and peptides naturally occurring in the proteomes, avoiding any enzymatic digestion and therefore sample alteration (Figure 2). MALDI- and SELDI-TOF MS, HPLC-MS and MS/MS analysis of intact proteomes or sub-proteomes can be included in this category [36,58]. The investigation of intact proteins (rather than enzymatically digested peptides) by MS allows for the better characterization of the protein state. In particular, the use of top-down approaches enables the detection of the biologically active forms of the proteins, the location and identity of post-translational modifications (PTMs). Substantially, the top-down strategy provides a deeper understanding of specific protein isoforms (proteoforms), which are not easily detected by standard protein profiling techniques (Figure 2) [58]. The first large-scale proteomic analysis from a single sample of human SP was performed by Pilch and Mann by 1D-PAGE in combination with LTQ-FTICR MS in order to extensively characterize the SP proteome providing better insights into sperm functions [23]. The preliminary protocol adopted in this study before MS analysis was based on the use of 1D-PAGE without any further biochemical purification steps. This choice implied no loss of both hydrophobic and highly charged proteins before MS analysis, achieving significantly better protein coverage. In fact, 923 SP proteins were identified with high confidence by this approach. Protein expression data were further analyzed by the GoMiner program package to assign their cellular localization, molecular functions and biological processes [23]. This study revealed the presence of a large number of extracellular proteins and proteins specific of male accessory glands with a key role in spermatozoa survival. Interestingly, the highly confident proteomic data originated from this investigation, provided an inventory of SP proteins that has served as a reference for SP proteomics studies with a special focus on male infertility and fertilization, and testicular and prostatic cancers. Milardi and colleagues implemented an extensive analysis of human SP in fertile subjects (n = 5) by the use of high-resolution LTQ-FT MS technology [59]. In this study, 1487 unique proteins per single sample were identified by performing the in-solution digestion of the individual samples followed by HPLC using a C18 column before LTQ-FT MS. Additionally, the authors also identified 83 common proteins in all fertile men (such as semenogelin I and II), obtaining the panel of proteins involved in reproduction regardless of interindividual variability. Drabovich and colleagues performed a multi-step strategy to assess biomarker identification for the differential diagnosis of azoospermia [25]. In particular, SP samples from normal fertile men (n = 12), infertile men with proven non-obstructive azoospermia (n = 10) and previously fertile men who had undergone a vasectomy (n = 8) (which simulated obstructive azoospermia) were analyzed by using LC-ESI-triple-quadrupole and ion-trap/Orbitrap MS (Table 1). First, a multiplex label-free selected reaction monitoring (SRM) assay was performed to measure the relative abundance of 31 proteins in the unfractionated digest of both pooled samples (5 normal, 5 non-obstructive azoospermic and 5 post-vasectomy) and 30 individual SP samples. SRM is a quantitative targeted proteomic assay, which is most commonly performed on a triple-quadrupole and allows for the measurement of multiple proteins in a single assay, thanks to the multiplexing capabilities [60]. Of the 31 proteins measured, 18 showed a statistically significant difference in disease samples compared to normal samples. To validate these 18 candidate biomarkers, heavy isotope-labeled internal standards were synthesized to once again assess their concentrations in the same cohort of 30 individual samples. In conclusion, they proposed a panel of biomarkers able to differentiate between normal, non-obstructive azoospermia and post-vasectomy (which simulated obstructive azoospermia). Among these, the most promising candidates were testis expressed 101 (TEX101), lactate dehydrogenase C (LDHC), sperm associated antigen 11B (SPAG11B), prostaglandin D2 synthase (PTGDS), mucin 15 (MUC15), and protein FAM12B, which were down-regulated in non-obstructive azoospermic and post-vasectomy men (Table 2). In 2013, the same group focused on the previous 18 biomarker candidates to select the minimal number of markers necessary for the differential diagnosis of azoospermia [26]. In total, 119 SP samples (42 men with normal spermatogenesis, 25 men with non-obstructive azoospermia, and 52 with obstructive azoospermia/postvasectomy men) were analyzed using the above-mentioned MS-based multiplex SRM assay (Table 1). Two proteins (epididymis-expressed ECM1 and testis-expressed TEX101) were identified for the differential diagnosis of azoospermia. In particular, extracellular matrix protein 1 (ECM1) was increased in fertile and non-obstructive azoospermic men and decreased in obstructive azoospermia, while TEX101 was increased in fertile men and decreased in non-obstructive azoospermic and obstructive azoospermic men. These data were validated by ELISA and immunohistochemistry (Table 2). In summary, this approach provided a two-biomarker decision tree for the non-invasive differential diagnosis of non-obstructive azoospermia and obstructive azoospermia and for the differentiation of non-obstructive azoospermia subtypes. Del Giudice et al. used LC-ESI- quadrupole-Orbitrap MS in order to determine SP biomarkers of testicular function in adolescents with varicocele [27]. In particular, they compared the proteomic profiles among three main groups by a label-free shotgun approach. In the control group, 23 subjects were recruited without varicocoele and with normal semen analysis. The other two groups were formed by patients with varicocoele and normal semen analysis (37 subjects) and by patients with both varicocoele and altered semen analysis (17 subjects) (Table 1) [27]. For each group, the samples were pooled, trypsin digested and analyzed by LC-ESI-quadrupole-Orbitrap MS. A total of 541 proteins were identified and a label-free quantification approach was performed using intensity-based absolute quantification (iBAQ). Briefly, iBAQ provides protein quantification as a fraction between the sum of peak intensities of all peptides matching to a specific protein by the number of theoretically observable peptides [61]. The authors proposed calcium-binding protein (Cab45), left–right determination factor 1 (protein lefty-1), deoxyribonuclease-1 (DNase I) and lipid phosphate phosphohydrolase 1 (PAP2-alpha) as candidate biomarkers of spermatogenesis and homeostasis associated to the control group, while insulin-like growth factor-binding protein 7 (IBP-7), Ig gamma-3 chain C region (HDC) and cysteine rich secretory protein 3 (CRISP-3) as biomarkers associated to varicocoele groups (Table 2). Among the suggested biomarkers selected by a multivariate statistical analysis, confirmatory results were obtained for two proteins using western blot: Cab45, that was underexpressed in varicocoele groups and CRISP-3, that was overexpressed in adolescents with varicocoele and altered semen analysis (Table 2). Sharma et al. used LTQ linear ion trap MS coupled to ESI ion source, in a label-free bottom-up approach, with the aim to compare the protein expression of SP samples derived from 26 normozoospermic men, 22 teratozoospermic, 6 oligozoospermic and 10 oligoteratozoospermic patients (Table 1) [28]. SP samples were pooled into four groups, according to subject category, and precipitated in cold acetone in order to improve the protein recovery in the sample. Then samples were subjected to in-solution digestion and to the LC-MS/MS system. The authors identified 35 proteins and performed a label-free quantification approach using spectral counting (SpC) (Table 1). In this approach, the protein quantification is determined by counting the total number of MS/MS spectra of all the peptides derived from the same protein [62]. Out of the identified proteins, 20 were differentially expressed among the four groups (Table 2). For example, the semenogelin I (SEMG I) isoform b preprotein was upregulated in the oligoteratozoospermic group; clusterin (CLU) isoform 1 was downregulated in the oligozoospermic group; prostatic acid phosphatase (PAP) was downregulated, and protein DJ-1 was overexpressed in the teratozoospermic and oligozoospermic groups (Table 2). Curiously, at the same time, DJ-1 (a protein involved in stress response) was found to be underexpressed in oligoteratozoospermic patients; this contrasting result requires further investigations. The functional bioinformatic analysis demonstrated that biological regulation is the mainly affected process and revealed that the identified proteins are mostly of extracellular origin (Table 2). However, it should be taken into account that the authors did not report any kind of normalization for protein concentration, and this could represent an important pre-analytical bias, potentially affecting the validity of the obtained data, especially in terms of differentially expressed proteins (Table 1). The hybrid ion trap/Orbitrap mass spectrometer was used by Wang et al. to compare the SP proteome between 20 healthy control donors and 38 asthenozoospermic men [29]. SP proteins were first separated by 1D-SDS PAGE, gel bands were then digested, and peptides were analyzed by the LC-MS/MS (Table 1) [29]. In total, 741 proteins were identified and quantified using the SpC method; among these, 45 proteins were upregulated and 56 were downregulated in asthenozoospermic men compared to control subjects (Table 2). The majority of these proteins derive from the epididymis and prostate, suggesting that functional abnormalities of the epididymis and prostate can contribute to asthenozoospermia. The authors proposed intelectin-1 (ITLN1), alcohol dehydrogenase (ADH), delta-aminolevulinic acid dehydratase (ALAD) and DJ-1 as potential biomarkers for oxidative stress in asthenozoospermic patients; in particular they suggested that the downregulation of DJ-1, validated by western blot analysis, lead to oxidative stress, affecting the quality of semen in asthenozoospermia (Table 2). On the contrary, as mentioned previously, Sharma et al. [28] found an over-expression for DJ-1 in teratozoospermic and oligozoospermic patients, while a down-regulation in the case of oligoteratozoospermic ones was observed. These apparently contrasting findings require further investigations to better elucidate the role of this potential biomarker. The same MS technology was used also by Batruch and colleagues to compare the SP proteome of fertile control men (n = 5), non-obstructive azoospermic patients (n = 5) and post-vasectomy/obstructive azoospermic patients (n = 5) (Table 1) [30]. SP samples were pooled, digested, fractionated by SCX chromatography followed by reversed-phase (RP) LC and analyzed by LTQ-Orbitrap MS. In SCX-RP LC, proteins are first separated on the basis of their charge (more specifically on the basis of the interaction between charged groups of the analyte and the negatively charged stationary phase) [63,64], and then on the basis of their hydrophobicity by using reverse phase columns. The most commonly used columns are the C18, silica-based columns derivatized with alkane chain containing 18 carbons to generate a hydrophobic surface [63]. A total of 2048 proteins were identified and quantified by Batruch et al., using two label-free quantitative approach (SpC and extracted ion chromatograms -XIC) [30]. The XIC methods enable to quantify proteins or peptides measuring the signal intensity, m/z values and the retention time of ions from chromatograms obtained in LC-MS measurements for specific peptides [65]. By this approach, the authors obtained candidate biomarkers useful to discriminate non-obstructive and obstructive azoospermia; they found 34 proteins up-regulated and 18 down-regulated in controls compared to non-obstructive azoospermic men, 59 up-regulated and 16 down-regulated in non-obstructive azoospermic men compared to post-vasectomy/obstructive azoospermic patients. Some of these proteins are shown in Table 2. Herwig et al. performed a label-free bottom-up approach in order to compare the proteomic profile of SP from 11 pooled fertile and 11 pooled infertile men with oligoasthenoteratozoospermia, using a hybrid linear ion trap/quadrupole/Orbitrap mass spectrometer (Table 1) [31]. Pooled SP samples were trypsin digested and analyzed by the hybrid linear trap/quadrupole/Orbitrap mass spectrometer. Using SpC and Gene Ontology (GO) functional annotation, 46 proteins were identified, among which 24 proteins were found to be upregulated in the infertile compared to fertile men. These proteins are mainly involved in metabolism and inflammation, defense, and stress responses, suggesting their influence on infertility, particularly due to oxidative stress. In particular, α-1-antichymotrypsin (AACT) and aldose reductase (ALDR) were upregulated in oligoasthenoteratozoospermic patients (Table 2). Wu et al. used a quantitative bottom-up proteomics approach to identify and quantify SP proteins in three normozoospermic and three asthenozoospermic men (Table 1) [32]. SP samples were treated with an acetone solution buffer over night for the precipitation and recovery of the proteins. Then, SP samples were digested, and the resulting peptides were labeled using the tandem mass tag (TMT) strategy. The peptide mixture was analyzed using a LTQ Orbitrap Velos mass spectrometer. TMTs are isobaric tags which label the peptides targeting the N-terminal position. These tags are fragmented by the MS/MS process and generate reporter ions, whose amount is directly proportional to the amount of the labeled peptides in the samples and is used to obtain quantitative information [66] (see also Figure 2). Thanks to this strategy, the authors were able to identify a total of 524 proteins. The biological functions and origins of these proteins were also determined by the integration of different proteomic datasets and bioinformatics databases. Finally, they found 29 differentially expressed proteins between the two analyzed groups. Bioinformatic analysis revealed that most of these proteins are mainly associated to sperm motility and male infertility, suggesting their potential role as biomarkers of asthenozoospermia. Validation analyses, using western blot, were performed for four proteins (KLK2, HSPA2, SORD, ANAX2), which confirmed the up-regulation of heat shock protein family A (Hsp70) member 2 (HSPA2) and the down-regulation of kallikrein related peptidase 2 (KLK2), sorbitol dehydrogenase (SORD), annexin A2 (ANAX2) in asthenozoospermic compared to normozoospermic (Table 2). However, an important limitation was the small pool of participants which compromised the significance of the study and the robustness of the identified biomarkers (Table 1). In another study, Barrachina and colleagues used a bottom-up approach to analyze SP samples from four normozoospermic, four asthenozoospermic, four oligozoospermic and four azoospermic men by LC-ESI-ion trap/ORBITRAP MS (Table 1) [33]. In particular, SP samples were precipitated in cold acetone to improve protein recovery and then trypsin digested. Resulting peptides were labeled with TMT isobaric tags prior to LC-MS analysis. Proteomic data were analyzed by standard statistical analyses of relative protein quantification values (ANOVA and Pearson correlation test) which revealed a set of six differentially expressed proteins, correlated with sperm concentration. These proteins included cysteine rich secretory protein 1 (CRISP1), epididymal secretory protein E1 (NPC2), serine peptidase inhibitor, Kunitz type 3 (SPINT3), and ECM1 that were down-regulated in patients with low or an absence of sperm cells (oligozoospermic and azoospermic), immunoglobulin heavy constant gamma 2 (IGHG2) that was up-regulated in normozoospermic and aminopeptidase N (ANPEP), which was down-regulated in patient with low sperm motility (asthenozoospermic men). Then, western blot analysis for only one differentially expressed protein (ECM1) was performed in an independent set of samples to confirm these results (Table 2). Saraswat et al. in an untargeted approach using shotgun proteomics compared the SP proteome from 7 normozoospermic and 10 asthenozoospermic men (Table 1) [34]. Samples were trypsin digested and analyzed by the ultra-performance liquid chromatography (UPLC)-MS (Table 1). They identified 429 proteins in SP samples. A label free strategy was performed to quantify these proteins, followed by statistical data analysis including principal component analysis and orthogonal projections to latent structures discriminant analysis (OPLS-DA), to identify the proteins significantly different among the two groups. Although some proteins were differentially expressed between the two groups, no statistical significance was found for seminal plasma dataset. Some of the upregulated and downregulated proteins in asthenozoospermic men are reported in Table 2. Finally, using a shotgun approach, Liu et al. performed a quantitative proteomic strategy to identify and quantify the potential biomarkers of oligoasthenozoospermia [35]. SP samples from 10 men with oligoasthenozoospermia and 10 men with normozoospermia were separately pooled and enzymatically digested; the resulting peptides were labelled with isobaric Tags for relative and absolute quantification (iTRAQ) reagents and then analyzed by two-dimensional RP-RP-HPLC and MALDI-TOF/TOF MS (Table 1 and Table 2). The iTRAQ method utilizes isobaric tags to label peptides and proteins at the N-terminus and provides a multiplexing approach to compare up to eight different samples in a single run [67]. The authors performed 2D-HPLC, providing a separation orthogonality in the RP-RP system using pH 10 in the first and pH 3.0 in the second dimension, according to a method described early by Gilar et al. [68]. More than 700 seminal plasma proteins were both identified and quantified. The differential proteomic analysis revealed the downregulation of 20 proteins and the overexpression of 22 proteins in oligoasthenozoospermic patients in comparison to normozoospermic individuals. In particular, they identified CD177 antigen (CD177), prolactin-inducible protein; (PIP), PSA, lactotransferrin (LTF), prostaglandin-H2 D-isomerase (PTGDS), epididymal secretory protein E1 (HE1 or NPC2) and ECM1, as potential biomarkers of oligoasthenozoospermia (Table 2) [35]. Interestingly, for some of these proteins, i.e., NPC2 and ECM1, the results confirm the observations of previous studies [26,33], which described an overexpression of these species in fertile subjects compared to infertile ones, validating their utility as biomarkers of male fertility status. The characterization of the human SP proteome by a top-down approach has been first explored by Fung and colleagues, describing the direct analysis of a pooled (n = 5) unfractionated SP sample by MALDI-TOF MS [24]. The strategy described by Fung et al., showing intact molecular features in a m/z range from 500 to 10.000, enabled the detection of endogenous peptides of SP in addition to multiple protein isoforms [24]. They also performed a comprehensive analysis of the peptide and protein constituents of the SP sample by combining classical 1D/2D gel electrophoresis with both MALDI-TOF and ESI-LC MS/MS, allowing the identification of over 100 different proteins [24]. Until now, only a few top-down investigations have been reported for the differential expression analysis of SP proteins between fertile and infertile patients (see Table 1 and Table 2) [36,37]. Very recently, our group has optimized a practical and efficient method for SP peptide enrichment before MALDI-TOF MS analysis, with the aim to reveal a diagnostic signature of male infertility [36]. Using a top-down strategy, we applied a dispersive-solid-phase extraction (d-SPE) coupled to MALDI-TOF MS to reveal SP peptides in their native and biologically active forms. In particular, commercially available octadecyl (C18)- and octyl (C8)-bonded silica sorbents and hexagonal mesoporous silica (HMS) were used to finely modulate the low molecular weight profiling of SP samples and best performances were obtained for C18-bonded silica. Finally, to assess the diagnostic potential of the platform, C18-bonded silica d-SPE and MALDI-TOF-MS were used to generate enriched endogenous peptide profiles from 15 fertile and 15 non-fertile donors and a key peptide-pattern within spectra was found to discriminate the two groups (Table 1 and Table 2). Seven differentially expressed peptides were identified, which were downregulated in the infertile patients compared to the fertile men. These peptides were fragments of SEMG I and SEMG II, which are abundant proteins in SP with key roles in coagulation and liquefaction processes. Interestingly, these findings are in contradiction with those of Sharma and others, who reported an augmented expression of SEMG I in oligoteratozoospermic patients as mentioned above [28]. On the other hand, a recent shotgun proteomics investigation by Martins and colleagues [39] reported both SEMG I and SEMG II to be under-expressed both in primary (inability to achieve pregnancy) and secondary infertile (inability to achieve pregnancy after at least one previous successful pregnancy) subjects compared to the fertile controls. It should be noticed that validation experiments performed by western blot only confirmed SEMG II decrease in primary infertility, while no change in the expression of both SEMG I and SEMG II was observed, again by western blot, in the secondary infertility group [39]. These apparently contradictory findings on semenogelins expression may derive from the intrinsic limitations of bottom-up approaches and by the presence of both intact semenogelins and peptide-derived semenogelins in SP. In fact, all the proteins are digested before LC-MS/MS analysis in a bottom-up approach. As a consequence, it is not possible to determine whether the peptides contributing to the identification of both SEMG I and II originated from intact precursors or from a fragmented protein. It is important to emphasize that a top-down strategy does not require the use of trypsin or more in general proteolytic digestion in comparison to bottom-up strategy. Our findings strongly consolidate the importance of semenogelins in male infertility (Table 2) [36]. Another top-down investigation which analyzed human SP proteins in association with the male fertility status was performed by Cadavid and colleagues using SELDI (surface-enhanced laser desorption/ionization)-TOF MS technology [37]. In SELDI-MS, analytes are applied to a protein chip array, which may be composed by different surfaces commonly used in chromatographic techniques (cationic, anionic, hydrophobic surfaces, etc.) or by biochemical bait molecules (immobilized antibodies, receptors) or also by DNA oligonucleotides. These surfaces are designed to retain proteins according to their chemical and physical characteristics (i.e., hydrophobic, hydrophilic, acidic, basic, metal affinity). It has been extensively used in proteomics studies, thanks to its high throughput, but suffers some drawbacks, including low resolution, poor reproducibility, both within and also between laboratories, and the inability to directly identify proteins because of the lack in MS/MS capabilities [69,70]. Cadavid and colleagues analyzed SP samples obtained from seven healthy fertile men and nine men with fertility alterations, including altered sperm progressive motility and sperm count (Table 1) [37]. Protein profiles of the SP samples were obtained by SELDI-TOF MS over a strong anion exchanger ProteinChip® Q10 array. By performing ROC curves, they found 10 SP proteins statistically upregulated in the infertile group compared to the fertile one, though they did not finalize the identification of the proteins. Using the previously published database for seminal proteins [23,29] the authors hypothesized the potential identity of the differentially expressed proteins by using the m/z value obtained for each peak with differential expression between fertile and infertile subjects (see Table 2 for biomarkers). It is interesting to note that, in the case of statistically significant peaks in the low mass range (<20,000 Da), they were not able to propose the potential identity of some biomarkers or alternatively, in some cases, the attribution of putative identity appears forced considering the elevated mass error of such species [37]. One of the major limitations in fact, in SELDI mass spectrometer, is the lack of the MS/MS identification step, which precludes the possibility to assign the identity of potentially new endogenous peptides easily detectable by a top-down approach or to identify post-translational modifications and proteolytic products. Additionally, the only other two investigations based on SELDI-TOF MS were used to assess protein changes in the SP of oligozoospermic [71] and non-obstructive azoospermic patients [72], although these data are not publicly available in an online database. In the first study, Yang and colleagues performed the differential analysis between the SP of fertile men and oligozoospermic patients by SELDI-TOF MS with H4 ProteinChip array surface (hydrophobic/reverse-phase array) and SAX-2 ProteinChip array surface (strong anionic exchanger array) [71]. The authors observed three differentially expressed protein peaks, that could be useful for screening potential oligozoospermic individuals [71]. In another study, Bai et al. analyzed and compared the SP proteome among non-obstructive azoospermic, severe oligozoospermic and fertile group by SELDI-TOF MS with the CM10 protein chip [72]. They stated that SP proteins compositions of severe oligozoospermic and healthy fertile men were similar, but both differed from non-obstructive azoospermic men [72]. The studies here reviewed reported the SP proteome profiling of infertile patients with quantitative and/or qualitative alteration of semen parameters. Although routine semen analysis is still the cornerstone for the clinical evaluation of male fertility/infertility, the exploding research on SP proteome could revolutionize the field of male infertility diagnosis and its clinical management. In fact, MS-based proteomics investigations on SP identified a plethora of potential biomarkers that could be useful for the non-invasive assessment of male reproductive conditions and for the differentiation of the various infertility etiologies. However, candidate markers identified in the here reviewed differential proteomic investigations were not always confirmed among studies. Such discrepancies among different studies yet sharing patient cohorts with same infertility conditions, shed light on the challenges for restricting to a single or at most a few putative biomarkers for male infertility issues (see Table 2). There are several pre-analytical and analytical factors which can influence data reproducibility, limiting the comparison of the here reviewed study results (Table 1). Such data variability could arise from many sources, among them differences in sample collection and processing, patient selection, analysis of individual or pooled samples, intra and inter-individual biological variations, MS techniques used for proteomic investigation and differences in proteomics strategies, the use or not of quantification methods, bioinformatics and the interpretation of collected data, etc. [73]. To date, the effects of different sample processing and the influence of both pre-analytical and analytical variables on SP proteomic profiling have not been extensively investigated in the proteomics studies which analyzed this specific biological fluid. In the following sections, the effects of different sample processing and the influence of both pre-analytical and analytical variables on SP proteomic profiling will be discussed in relation to the here reported investigations. The inherent features of SP, namely complexity and heterogeneity, pose many hurdles especially in comparative studies, which require standardized procedures for the normalization of MS data. It is well established that sample collection and processing methods significantly influence the mass spectra profile [74,75,76]; therefore, one of the major pre-analytical challenges in the proteomic analysis of SP is the lack of a standardized processing protocol. In fact, as pointed out in Table 1, SP specimens were obtained from semen samples with striking differences in the force, the number and the duration of centrifugation steps, in the timing and temperature of semen liquefaction. Specifically, the above reviewed investigations, indicated one [25,26,28,30,32,36] or more steps [27,29,31,33,34,35,37] for sample centrifugation at different force conditions (ranging from 500 to 100,000× g); some of them did not provide precious protocol details about timing of semen liquefaction [27,33,34,37]. Obviously, all these variables could contribute to the heterogeneity of SP analyzed by MS. Furthermore, SP is characterized by a high dynamic range of protein abundance, with the top 10 most abundant proteins that account for about 80% of the total protein content [16,20,23]. The wide range of protein concentrations could mask the presence of low-abundance components, which may still play a key role in reproductive processes. To address these issues, the depletion and pre-fractionation of high abundance proteins as well as enrichment procedures of lower abundant ones should be applied [20]. However, also the use or not of sample separation or fractionation before analysis may result in high variation for protein detection and identification among different experiments and studies (Table 1). In fact, it is quite conceivable that various subproteomes in seminal plasma were extracted and converted into MS-profiles in those studies which used different chemical groups and chromatographic features for sample fractionation (Table 1). Protease activity as well as proteins PTMs contribute to strongly increased SP protein complexity providing different variants detected in independent studies and data heterogeneity [73,77]. MS-based proteomics of SP may accelerate biomarker discovery of infertility by comparative differential proteomics in untargeted approaches. However, heterogeneity or small-sample size of fertile vs. infertile populations necessarily lead to differences in study results (Table 1 and Table 2). In view of the intrinsic intra- and inter-individual variability of SP samples, larger cohorts would be necessary in order to compare the proteomes in the presence or absence of male infertility disorders. As a matter of fact, one of the main concerns in clinical proteomics remains to obtain a number of samples high enough to reach statistical significance. Hence, proteomic studies on sufficiently large cohorts of patients with standardized protocols are warranted to validate the preliminary markers identified in the clinical practice. In the context of intra and inter-individual variability, pooling or not samples (see Table 1) might also have implications that should be taken into consideration during proteomic experimental design, in order to avoid some unforeseen methodological and statistical bias. In fact, although the biological variance among pools is reduced compared to that among individuals, proteins visible in individual samples are not always detectable when the pooling of samples are performed, with a potential loss of information due for example to dilution effects [78,79]. With the exception of the study by Drabovich and colleagues [25], which provided a complementary proteome analysis on pooled and individual samples, about half of the here reviewed studies performed only pooling of samples [27,28,30,31,33,35]. Sample pooling is preferably suitable for proteomic analyses when it is representative of the individual samples used to constitute the pool. Additionally, pooling samples could decrease the study power and modify the mean value or standard deviation of such analyte, and this could affect the value of statistical tests [78,79]. In summary, the choice to analyze individual or pooled samples is a crucial step due to its potential impact on the study design and consequently on the identification of reliable biomarkers of diseases. Concerning proteomics and peptidomics studies, the quest for quantitative strategies is stringently desirable in order to make data comparable among each other. MS-based quantitative proteomics can be divided into two main strategies: label-based and label-free. Although label-based methods provide an accurate and precise quantification, they have limitations, including the increased cost and sample preparation time, which limit the number of samples to be compared [80]. On the contrary, label-free strategies show a large dynamic range of quantification and allow for the quantitative comparison of different numbers of samples, reducing costs and the complexity of sample preparation [81]. It should be taken into account that low-abundant peptides and proteins, peptides derived from proteolytic cleavages and PTMs variants could be of elusive detection using shotgun proteomic approaches. The intrinsic limitations of such strategies, which require the use of trypsin or more in general proteolytic digestion before MS analysis, do not allow the detection of SP peptides and proteins in their native and biologically active forms. This may cause loss of information related to potentially important markers of male infertility. Otherwise, top-down studies, which are well suited for the detection of naturally occurring peptides and small proteins, appear in a limited number (see Table 1 and Table 2). MS profiling strategies aiming at harvesting intact peptide signatures from SP samples may reveal one or more specific molecular patterns for the different infertility phenotypes. It is important to notice that discordant results between reports could also derive from the different performances and detection capabilities of mass spectrometers which may affect spectral readouts. In fact, the use of ultra-high-resolution mass spectrometers or hybrid instruments with combined mass analyzers can achieve enhanced resolution, sensitivity and mass accuracy, increased dynamic range and fast acquisition rates with better qualitative and quantitative MS and MS/MS performances. In light of the above considerations, in the next years the research focus should be directed towards the assessment of high-throughput MS-tools able to capture snapshots from SP containing diagnostic clinical information of infertility issues on an individual and population scale. Hence, there is an urgent need to identify a MS-based simple decision tree for the non-invasive differential diagnosis of infertility related diseases. Interestingly, seminal protein-based assays of TEX101 and ECM1 are under final validation for clinical use, providing the ability of MS to develop reliable clinical-grade assays [82,83,84]. In this scenario, it is worth noting that MALDI-MS has already become a routine laboratory diagnostic tool for the rapid, accurate and cost-effective identification of cultured bacteria and fungi in clinical microbiology, also proving to be capable of supporting screening and clinical decision-making [85,86,87]. Similarly, this high-throughput MS-technology may emerge, together with the clinically useful biomarkers discovered by proteomics, as a powerful diagnostic platform for translating the validated biomarkers of male infertility from bench to bedside. This could facilitate the development of drugs, devices, treatment options and especially new clinical-grade assays providing better management and care of infertile patients. Therefore, ‘top-down’ approaches should become necessary for valuable improvements in translational research exploiting their potential to revolutionize the field of diagnostics and therapeutics of male infertility associated scenarios. Careful attention should be paid in the appropriate interpretation of the huge amount of collected data, which could contribute to the discrepancies between reports from different authors. The huge amount of scientific data generated during proteomic studies requires bioinformatic analysis using advanced software tools. The advancement in computational tools and user compatible analysis software provide a reliable interpretation of MS-based data. Concerning issues related to bottom-up or top-down related MS data, most of data in spectral libraries is built from the in-silico enzymatic digestion of proteins, limiting the analysis only to peptides obtained from expected enzyme cleavage sites and also the number of PTMs considered; therefore, the datasets are difficult to leverage for peptidomics [88]. Moreover, the limited length of the aminoacidic sequences makes MS-peptidomics data analysis more challenging compared to bottom-up spectral data. The MS/MS patterns obtained from top-down approaches provide ‘non-tryptic’ peptides less informative than ‘tryptic digested ones’. In the light of the above considerations, the same bioinformatics strategies adopted for bottom-up approaches could not adequately perform in the case of less predictive and informative MS/MS spectra obtained from endogenous peptides [89]. Noteworthy, unlike bottom-up proteomics, software tools for peptidome characterization are not fully developed and MS data frequently requires skilled manual interpretation and rigorous validation [88,89]. De novo sequence algorithms together with classical database search provide sensitive and accurate peptide identification [90,91]. The application of software able to predict fragmentation patterns is one of most cost-effective ways to validate the identification [92]. More in general, the integration of proteomic and bioinformatic data provides insight into the function of proteins and peptides in cellular pathways. Hence, the advancement of infertility diagnostics and therapeutics depends on the integration of high-throughput “omics” data to identify accurate and specific biomarkers for infertility-related conditions. Network and pathway analysis using bioinformatic tools have been successfully used to obtain a wider picture on the putative pathways associated with the statistically significant biomarkers and their involvement in various infertility-related scenarios [16,93]. Most of the proteomics studies on SP used GO analysis to provide information about the localization, distribution and biological functions of the identified proteins. The integration of the datasets in male infertility still requires improved functionality, as currently few mathematical algorithms are available for cross omics data integration [94]. The integration of various proteomic datasets and bioinformatics databases should help to comprehensively annotate the biological functions and disease associations of the putative diagnostic markers of male infertility. Due to the presence of specific proteins secreted from organs of the reproductive tract, SP represents an excellent clinical diagnostic fluid for the discovery of biomarkers for the diagnosis of male infertility and their clinical translation. It would be highly desirable that multi-analyte panels extrapolated by a high-throughput MS tool in an untargeted discovery phase might in future be able to reveal a distinctive pattern of molecular features correlated by specific relative expression to male fertility or infertility. Currently, proteomics untargeted approaches mainly relying on the MS technological platform, provide not only the discovery of multiple biomarkers between fertile and unfertile populations (Figure 3), but also the accurate quantification of these signature molecules by MS [26,32,33,35]. However, the correct diagnosis of male infertility requires an accurate validation of putative signature molecules from the discovery phase (Figure 3). From the highlighted literature here reviewed, only a few studies have addressed this validation step [25,26,27,29,32,33,35]. Future developments for laboratory medicine diagnosis also require cheap MS instrumentation with ultra-fast high-throughput features as well as highly specialized and experienced personnel for the correct interpretation of the quality of the MS outputs. It is important to notice that for what assisted reproduction technology (ART) is concerned, the major focus of proteomic investigations has been on the ejaculated spermatozoa. In fact, up to now, some differential proteomics investigations performed on semen have already identified putative biomarkers indicative of pregnancy outcome after ART [95,96,97,98,99]. In light of main findings from the here reviewed studies, which revealed how SP contains diagnostic clinical information of infertility issues, ART success rates may increase by placing the right focus on the critical role of SP proteome in fertilization. Therefore, further studies could be useful to clarify the role of SP in a successful pregnancy using ART. Last, but not least, smart and fully automated software for time costs optimization and to further provide increasing accuracy might accelerate the application of MS in clinical laboratories.

PMC10002487

Shubhasmita Mohapatra,Jared Cafiero,Khosrow Kashfi,Parag Mehta,Probal Banerjee

Why Don’t the Mutant Cells That Evade DNA Repair Cause Cancer More Frequently? Importance of the Innate Immune System in the Tumor Microenvironment

06-03-2023

glioblastoma,tumor microenvironment,macrophages,repolarization,chemokines,immunotherapy

The standard of care for most malignant solid tumors still involves tumor resection followed by chemo- and radiation therapy, hoping to eliminate the residual tumor cells. This strategy has been successful in extending the life of many cancer patients. Still, for primary glioblastoma (GBM), it has not controlled recurrence or increased the life expectancies of patients. Amid such disappointment, attempts to design therapies using the cells in the tumor microenvironment (TME) have gained ground. Such “immunotherapies” have so far overwhelmingly used genetic modifications of Tc cells (Car-T cell therapy) or blocking of proteins (PD-1 or PD-L1) that inhibit Tc-cell-mediated cancer cell elimination. Despite such advances, GBM has remained a “Kiss of Death” for most patients. Although the use of innate immune cells, such as the microglia, macrophages, and natural killer (NK) cells, has been considered in designing therapies for cancers, such attempts have not reached the clinic yet. We have reported a series of preclinical studies highlighting strategies to “re-educate” GBM-associated microglia and macrophages (TAMs) so that they assume a tumoricidal status. Such cells then secrete chemokines to recruit activated, GBM-eliminating NK cells and cause the rescue of 50–60% GBM mice in a syngeneic model of GBM. This review discusses a more fundamental question that most biochemists harbor: “since we are generating mutant cells in our body all the time, why don’t we get cancer more often?” The review visits publications addressing this question and discusses some published strategies for re-educating the TAMs to take on the “sentry” role they initially maintained in the absence of cancer.

Why Don’t the Mutant Cells That Evade DNA Repair Cause Cancer More Frequently? Importance of the Innate Immune System in the Tumor Microenvironment The standard of care for most malignant solid tumors still involves tumor resection followed by chemo- and radiation therapy, hoping to eliminate the residual tumor cells. This strategy has been successful in extending the life of many cancer patients. Still, for primary glioblastoma (GBM), it has not controlled recurrence or increased the life expectancies of patients. Amid such disappointment, attempts to design therapies using the cells in the tumor microenvironment (TME) have gained ground. Such “immunotherapies” have so far overwhelmingly used genetic modifications of Tc cells (Car-T cell therapy) or blocking of proteins (PD-1 or PD-L1) that inhibit Tc-cell-mediated cancer cell elimination. Despite such advances, GBM has remained a “Kiss of Death” for most patients. Although the use of innate immune cells, such as the microglia, macrophages, and natural killer (NK) cells, has been considered in designing therapies for cancers, such attempts have not reached the clinic yet. We have reported a series of preclinical studies highlighting strategies to “re-educate” GBM-associated microglia and macrophages (TAMs) so that they assume a tumoricidal status. Such cells then secrete chemokines to recruit activated, GBM-eliminating NK cells and cause the rescue of 50–60% GBM mice in a syngeneic model of GBM. This review discusses a more fundamental question that most biochemists harbor: “since we are generating mutant cells in our body all the time, why don’t we get cancer more often?” The review visits publications addressing this question and discusses some published strategies for re-educating the TAMs to take on the “sentry” role they initially maintained in the absence of cancer. Glioblastoma (GBM) is one of the deadliest forms of cancer, with an average life expectancy of about 14–18 months from detection [1]. The standard of care (SOC) for GBM is surgical resection of the tumor followed by chemo and or radiation therapy to eliminate the residual cancer cells [2,3]. However, in most cases, GBM returns soon after the SOC and eventually overcomes the patient. In chemotherapy, the general strategy has been the use of antimetabolites that inhibit DNA replication and other compounds that target specific signaling proteins that are often overactivated by mutations [4]. Unfortunately, such chemotherapeutic agents also inhibit the normal signaling proteins needed by healthy cells, such as the immune cells, thus causing severe side effects linked to lymphopenia. On the other hand, though targeted to cancer cells, radiation therapy kills the juxtaposed normal cells, such as the microglia, macrophages, and the infiltrating immune cells, thereby weakening the overall system. Overall, such therapeutic strategies have not significantly extended the life of GBM patients. Despite the ongoing effort to develop effective therapeutic strategies, an important question remains: “biochemical knowledge reveals that, despite effective DNA repair, we are constantly generating some cells with mutant DNA molecules, so why don’t we, relatively speaking, develop cancer more frequently?” A general belief is that our vigilant immune cells promptly eliminate such mutant cells. However, when our immune system is weakened, the mutant cells can proliferate to precipitate cancer, especially in the later years. We found hard evidence corroborating this belief in a study conducted by Afshar-Sterle and coworkers, which showed that the loss of the tumor-suppressor function of the gene BLIMP1 or deregulated expression of the BCL6 oncogene occurs in a large portion of B-cell lymphomas in human patients, however, the deliberate introduction of Blimp1 deficiency or Bcl6 overexpression in the B cells of mice does not precipitate lymphoma unless the T-cell receptor CD28- and Fas-ligand activities are simultaneously impaired in the CD8+ Tc cells [5]. Additionally, consistent with the hypothesis that cells with mutated DNA are eliminated by microglia and macrophages, Shi and coworkers observed that consequent to Knl1 deletion, neural progenitor cells accumulate DNA damage on mis-segregated chromosomes in the mitotic spindle, which triggers apoptosis and phagocytosis by the microglia [6]. However, Kasapi and Triantafyllopoulou note that the “role of genotoxic stress as an instructor of macrophage-mediated immune defense and tissue remodeling is only beginning to be understood” [7]. Therefore, more research is required to elevate the currently-held belief to a widely-accepted phenomenon. During the last decade, a significant focus has been placed on the adaptive immune system as a tool to eliminate cancer cells. This has resulted in the development of cytotoxic T-cell (Tc)-based immunotherapy [8], which has shown considerable success in several cases of melanoma [9,10], but, unfortunately, not for GBM and some peripheral cancers, such as endometrial/ovarian, pancreatic, liver, and colon cancers, to name a few. Furthermore, adverse events, mainly due to autoimmune reactions, have been reported following Tc-based immunotherapy [11]. As for the immediately-acting innate immune cells such as microglia and macrophages, they are recruited into the GBM tumor and changed from the tumoricidal, “classically-activated” “M1”-type to a tumor-promoting, “alternatively-activated” “M2” phenotype by cytokines secreted by the GBM cells [12]. Thus, the GBM microenvironment harbors mostly M2-type tumor-associated microglia/macrophages (TAMs), very few M1-type microglia, and some nonactivated M0-type microglia [12]. Since the direct killing of GBM cells has proven to be ineffective in eliminating all tumor cells and reliably preventing cancer relapse, an attractive strategy could be to re-educate the M2-type microglia or macrophages in the tumor microenvironment (TME) to the M1-type, thereby launching a Trojan horse-like attack from inside the tumor. A few such strategies have been discussed here. The M2 state of the microglia and macrophages is centrally controlled by the transcription factor STAT-3, which is known to stimulate the expression of immune-suppressive cytokines like IL-10, IL-4, and IL-13 [13]. In addition, it upregulates the expression of the key enzyme Arginase-1 (Arg-1) that marks the M2-type microglia and macrophages [12,13,14]. The cytokine IL-10 causes inhibition of STAT-1 by suppressing the phosphorylation of this transcription factor [12,14]. Therefore, the inhibition of STAT-3 would cause an activation of STAT-1 and subsequent STAT-1-mediated events such as the induction of inducible nitric oxide synthetase (iNOS; also known as NOS2), MCP-1, and IL-12, which is typical of M1-type microglia and macrophages [12,14,15,16]. Furthermore, we know from earlier studies that upon release from the microglial cells in the brain, the chemokine MCP-1 (a.k.a. CCL2) crosses the blood-brain barrier into the peripheral system to bind to its receptor, CCR2, expressed by activated macrophages and natural killer (NK) cells, and thereby cause recruitment of these cells into the GBM tumor in the brain [15,16,17]. Thus, the inhibition of STAT-3 in the microglia is central to a process that links to the recruitment of an army of M1 macrophages and activated NK cells into the GBM to eliminate GBM cells and GBM stem cells [15,16,18,19,20]. Due to the mechanisms discussed in the previous section, finding or designing agents to inhibit STAT-3 has been a popular strategy among researchers keen on developing therapies against cancer [21]. Yet no FDA-approved STAT-3 inhibitor is available currently. Several natural products (mostly polyphenols and antioxidants) that inhibit STAT-3 have been used in preclinical studies against various types of cancer [22]. We have shown in a series of publications that curcumin (CC) and synergistic formulations of CC and other polyphenols, such as resveratrol (Res) and epicatechin gallate (ECG) (TriCurin), can inhibit STAT-3 in the microglia and macrophages in GBM as well as HPV+ cervical cancer, thereby repolarizing these cells in the TME to the M1 phenotype [15,19,20,23]. Almost all chemotherapeutic agents (CAs) are designed to block DNA replication in fast-dividing cells such as cancer cells. Intriguingly, at least one, paclitaxel (Taxol), functions by blocking the microtubule-dependent cell division machinery. Additionally, this same chemotherapeutic agent (Taxol) is known to inhibit cytokine-mediated STAT-3 activation and its interactions with microtubules [24]. Therefore, the efficacy of Taxol in eliminating tumor cells through the repolarizing of the TAMs from the M2 to M1 type deserves further investigation. As mentioned in the previous section, Arg-1 is highly expressed by M2-type microglia and macrophages. This urea cycle enzyme is believed to deplete the amino acid arginine, which is also a substrate for the enzyme iNOS that is highly expressed by the M1-type microglia and macrophages [12,15,19,23]. iNOS uses arginine as its substrate to generate nitric oxide (NO), a crucial signaling molecule that combines with reactive oxygen species to generate cytotoxic, reactive nitrogen species inside the tumor [25], thereby eliminating cancer cells and cancer stem cells. A high Arg-1 expression in the M2 microglia and macrophages is expected to disrupt the supply of arginine to iNOS, thus inhibiting the generation of NO and reactive nitrogen species. Therefore, Arg-1 expression by TAMs is a critical event determining their polarization states. Yet, currently, there is considerable debate over using Arg-1 to mark the activation state of human microglia and macrophages [26,27]. It seems that although mouse monocytes show IL-4-evoked induction of Arg-1, this is not observed in human monocytes [26,28]. The literature on the induction of Arg-1 by interleukins is replete with many mechanisms, and Makita and coworkers report that human IL-10 augments IL-4-mediated induction of Arg-1 in monocytes [29]. Furthermore, Kupani and coworkers observed IL-10- and TGFβ-induced expression of Arg-1 in human monocytes [30]. To address this dichotomy, Thomas and Mattila state that cultured monocytes from various sources can elicit responses that are different from macrophage responses in vivo [27]. Their second but legitimate argument is that these debating groups had attempted to identify the Arg-1 protein instead of measuring its activity. However, most biochemists will agree that the apparent absence of a protein is not a full-proof sign of non-expression of the enzyme mainly due to the differing sensitivities of the antibodies and the high Vmax value of Arg-1, which, therefore, can produce ornithine at very low concentrations. Cognizant of this controversy, all of our studies of M2 → M1 repolarization of TAMs have used in vivo analysis using either immunohistochemistry (IHC) or flow cytometry analysis of dissociated tumor cells after fixing and antibody staining [15,16,19,20,23]. Thus, it is likely that data from mice and humans would be similar if the experiments were conducted on intact tumor tissue rather than cultured monocytes. Among the innate immune cells, interferon gamma (IFNγ)-activated NK cells are known to play a crucial role in eliminating cancer cells and cancer stem cells [16,20]. The mechanisms of NK cell-mediated elimination of microglia and macrophages have been studied earlier. Thus, Lunemann and coworkers used human microglia and human NK cells to show that IL-2-activated NK cells formed immune-synapses with resting (M0) microglia to kill them but sparing the lipopolysaccharide (LPS)-activated microglia (M1) [31]. This microglia recognition occurred mainly through the NK-cell-harbored receptor proteins NKG2D and NKp46 since the antibodies to these proteins blocked the killing completely. Furthermore, MHC class I molecules modestly expressed by the microglia appeared to abrogate NK-cell-mediated killing due to toll-like receptor 4 (TLR4) stimulation by LPS, thus protecting these microglial cells. Intriguingly, in vitro cultured peripheral blood monocyte-derived macrophages were not protected from the NK cells following LPS activation. Based on the ability of NK cells to eliminate tumor cells, they have been considered for use in clinical trials involving immunotherapy [32,33,34,35,36]. Although it is accepted that NK cells are recruited into the GBM tumor, how they are drawn into the brain has been an important question, with multiple chemokines proposed to be involved by various research teams. A study showing NK cell chemotaxis into the liver during infection noted the involvement of the chemokine MIP-1a (a.k.a. CCL3) [37]. Morrison and coworkers observed that CCL2 was involved in NK-cell recruitment into the lungs during aspergillosis [38], and Hokerness and coworkers showed that this NK-cell recruitment required CCL2 plus its receptor, CCR2 [39]. Additionally, Trifilo and coworkers report that CXCL10 promotes innate defense against coronavirus infection by recruiting and stimulating NK cells [40]. In our studies in the GBM mouse model, we have observed that repolarization of the TAMs from M2 to the M1 state is associated with a dramatic increase in the expression of CCL2 (a.k.a MCP-1) in the microglia/macrophages, which is concomitant with the recruitment of activated NK cells into the TME [15]. Earlier research has demonstrated that CCL2 is expressed as a marker by M1 microglia and macrophages [41,42]. Furthermore, CCL2 reportedly can compromise the blood–brain barrier (BBB) and translocate from the brain to the peripheral system, thereby affecting recruitment of immune cells such as M1-type macrophages and NK cells, which express the CCL2 receptor CCR2 [43,44,45,46]. Based on these findings, we have proposed that after the initial repolarization of the microglia in the GBM TME from M2 → M1 in a syngeneic mouse model after curcumin treatment, CCL2 released by the TAMs causes intratumor recruitment of activated M1-type macrophages and IL-12-activated NK cells from the periphery [15]. Once inside a tumor, the role of activated NK cells in eliminating tumor cells has been more generally accepted. As mentioned earlier, NK cell-based immunotherapy has been considered for clinical trials [32,33,34,35,36]. In the clinical application of NK cell therapy, deliberate intratumor infusion of NK cells is followed by IL-2 administration to activate the introduced NK cells. However, several factors render in vivo IL-2-mediated activation of NK cells a risky strategy. In addition to toxicity due to IL-2 administration, this causes the proliferation of immunosuppressive regulator T (Treg) cells [47]. Therefore, NK cell therapy has relied on in vitro IL-2 activation of NK cells followed by infusion of the activated cells. In our syngeneic mouse models of GBM and human papillomavirus (HPV)-mediated cancer, we have consistently observed the recruitment of activated NKp46+ NK cells and Tc cells into a tumor in mice treated with curcumin or a synergistic formulation containing curcumin, resveratrol, and epicatechin gallate, Tricurin (Figure 1) [15,19,20]. During our studies in the GBM mice, we also discovered an additional property of NK cells. NK cell recruitment was responsible partly for the curcumin-triggered repolarization of the TAMs from M2- to M1-type [15]. The intriguing offshoot of our studies is that both curcumin and Tricurin appear to be safe tools that can replace IL-2 in causing the activation and intratumor recruitment of NK cells [15,19]. Taken together, safe strategies appear to be available to turn both the TAMs and the NK cells against tumor cells that may have acquired diverse mutations in the process of becoming malignant. Currently, immunotherapy, popularly known as “immune checkpoint inhibitor therapy,” mainly refers to a strategy of empowering CD8+, cytotoxic Tc cells of the adaptive immune system to eliminate cancer cells [48]. In order to prevent autoimmune attacks, the antigen-presenting cells of an organism express program cell death ligands (PD-L1), PD-L2, as well as the major histocompatibility complex (MHC), which bind to the protein program cell death one (PD-1) and the TCR/CD3 complex, respectively, thereby dampening the cytotoxicity of the Tc cells [48]. Most cancer cells also express high levels of PD-L1 to evade attack by Tc cells. Currently, two FDA-approved PD-1 antibody-based immunotherapy drugs are marketed under the names, Keytruda and Opdivo [49,50]. Additionally, the Tc cells express a protein, receptor protein cytotoxic T lymphocyte antigen four (CTLA4), which binds to ligands CD80 and CD86 expressed by antigen-presenting cells and cancer cells [8]. This CTLA4–CD80/CD86 interaction antagonizes the interaction of the Tc antigen CD28 with CD80/CD86, which activates the Tc cells. Therefore, CTLA4 inhibition would cause the activation of Tc cells against cancer cells. To achieve this, the FDA-approved, antibody-based drug Yervoy has been used for various types of cancer, including melanoma [51]. Another potential candidate protein to be included in immunotherapy is COP9 signalosome 5 (CSN5). Lim and coworkers have shown that CSN5, which is induced by nuclear factor kappa B (NF-κB) p50:p65 heterodimer (NF-κB p65), is required for tumor necrosis factor alpha (TNFα)-mediated stabilization of PD-L1 in cancer cells [52]. In their study, curcumin-evoked inhibition of CSN5 caused a decrease in PD-L1 expression in cancer cells, sensitizing them to anti-CTLA4 therapy. Intriguingly, curcumin also inhibits CTLA-4 [53]. Therefore, the inclusion of CSN5 as a target could increase the efficacy of immunotherapy. Finally, a newly invented strategy involving “Base Editing” appears to have given some leukemia patients a new lease on life [54]. Developed six years ago by David Liu, this technique of base editing uses a mutated version of the CRISPR Cas9 protein to target DNA sequences containing a “C” or an “A” to convert them through deamination to U and inosine (I), respectively. Coupling this step with inhibitors of the base excision repair enzymes enabled Liu and coworkers to produce mutations that can correct or introduce pathogenic changes [55]. Using this strategy and allogenic Tc cells, mutant Tc cells were created that could eliminate malignant and normal Tc cells of a leukemia patient, thus making the patient cancer-free [54]. Typically, this step is followed by a transfer of healthy bone marrow-derived T cells to the patient. As for immunotherapy for glioblastoma in particular, most attempts have yielded only limited success [56]. Nevertheless, experimental evidence has suggested that manipulating the innate immune system might be beneficial. The discussion above shows that immunotherapy has so far involved the adaptive immune system. Two questions remain: (one) does the innate immune system influence the adaptive immune system, and does manipulating the adaptive immune system enhance the immunotherapy in use so far? and (two) has the innate immune system been considered as a primary mode of attack on cancer cells? Answers to the first question come from studies on dendritic cells (DCs), which, as innate immune cells, are known to be involved in the recruitment and activation of the Tc cells. Unlike the macrophages, which are recruited into the brain, likely by chemokines, the DCs, though not present in the brain parenchyma, are concentrated in the blood vessel-rich regions around the ventricles, such as choroid plexus and meninges [57,58,59]. From these niches, the DCs migrate to the brain and spinal cord under pathological conditions via lymphatic ducts or blood capillaries [60]. Among the different types of DC cells, the plasmacytoid DCs (pDCs) recognize pathogens such as viruses through toll-like receptor TLR7- and TLR9-signaling and secrete type 1 interferons (IFN1), which strongly activate CD8+ Tc cells [61,62,63]. Intriguingly, similar to the M1-type microglia and macrophages, the DCs respond to inflammation and infection by secreting inflammatory cytokines like IL-6 and IL-12 and chemokines CCL3, CCL4, CXCL8, and CXCL10 to recruit immune cells [64]. Similar to the DCs, the microglia are also known to cause recruitment of T cells from the periphery, although through a mechanism that involves the noncanonical nuclear factor κB (NF-κB)-inducing kinase (NIK) [65]. Additionally, other researchers have reported reciprocal signaling between the CNS microglia and the effector Tc cells in the context of neurodegenerative diseases and glioblastoma [66,67], and general surveillance of the CNS [68]. Thus, antigen-presenting cells such as DCs and microglia can regulate the recruitment and activation of Tc cells. The innate immune cells, like engineered DCs and activated NK cells, have been used in glioblastoma therapy [33,34,36,69]. It appears from the observations of repolarization of the microglia and the macrophages, intratumor recruitment of NK cells, and inhibition of CSN5 [15,16,19,20,23,52], that involvement of the innate immune system to assist the adaptive immune system may yield an effective and safe strategy of eliminating GBM as well as other solid tumors. Finally, question (two) has been answered in therapeutic applications of dendritic cells and NK cells, as discussed before [33,34,36,69]. Due to our particular interest in GBM therapy using microglia and macrophages, we next focused our attention on the use of these antigen-presenting innate immune cells. Reports of using macrophages in cancer therapy appear to be undergoing explosive growth. These studies can be classified roughly into two groups: (one) elimination of tumor-promoting TAMs (M2-type) and the inhibition of further recruitment into the tumor, and (two) reprogramming of the M2 TAMs into the tumoricidal M1 TAMs. However, the more effective antitumor treatments appear to involve a combination of traditional chemotherapy with the targeting of TAM, followed by the emerging immunotherapy involving immune checkpoint inhibition by targeting PD-1, PD-L1, or CTLA4, as discussed earlier. Inhibition of macrophage recruitment into tumors to inhibit M2-type macrophage-evoked tumor progression and metastasis was attempted by blocking some chemokine signaling pathways that cause the intratumor recruitment of macrophages. Such chemokine signaling involved the receptors for CCL2 (CCR2), CCL5 (CCR5), and CXCL12 (CXCR4) [70], which also enhanced STAT-3 activity and M2 polarization of macrophages and retention inside the tumor [71,72]. Fourteen clinical trials targeting the CCL2/CCR2 axis using five investigational drugs have been conducted with disappointing results. One of these drugs (BMS813160) is currently in Phase II trials for colorectal and pancreatic cancer [73,74,75]. The disappointing outcome was attributed to improper patient selection. It was felt that patients selected for high CCR2 expression in their tumors would have shown a more robust response in the clinical trials. Selective inhibition of a single, specific target with drugs has been attempted in most clinical studies, however, they have yielded inconsistent results and precipitated many side effects. For example, the CCL2 pathway is also crucial for the normal functioning of the lungs and the digestive system; therefore, such attempts to shut down specific signaling pathways could be detrimental to the patient [73,76]. Similar inhibition of the CCL5/CCR5 pathway has been studied as a possible target for eliminating TAMs, and the FDA-approved HIV drug Maraviroc is currently being considered for cancer [77,78]. Among the other attempts to deplete TAM, targeting the colony-stimulating factor (CSF-1)/CSF-1R pathway, which is known to trigger TAM recruitment into tumors and polarization of these cells into the M2-like phenotype, has been considered. Some preclinical models showed that CSF-1R inhibition causes reduced TAMs and tumor growth [79]. However, other reports indicated that inhibition of the CSF-1/CSF-1R axis did not obliterate all macrophages but pushed the TAMs toward the M1-like phenotype triggering CD8+ T cell activation and inhibiting tumor progression [80,81]. Furthermore, CSF-1R blockade only caused a modest delay in tumor growth, thus yielding only limited therapeutic success [80,81]. An FDA-approved, small-molecule CSF-1R kinase inhibitor, BLZ945, did not eliminate macrophages in lung cancer but reprogrammed them into the M1 phenotype and triggered the recruitment of IFNγ-wielding NK and Tc cells and also IL-12-secreting dendritic cells with antitumor activity [82]. The macrophage repolarization strategy has also been tested by using agonists for the toll-like receptors TLR3, TLR4, and TLR7/8, which are known to repolarize M2-like TAMs into M1-like phenotypes to levels comparable to that achieved with lipopolysaccharides and IFNγ [83]. Another preclinical study was conducted using poly-ICLC (polyinosinic-polycytidylic acid), a TLR3 agonist, with promising results [84]. A relatively new strategy of repolarizing TAMs involves the cyclic guanosine monophosphate-adenosine monophosphate synthase (cGAS)-stimulated interferon gene (STING) pathway, which appears to be sensitive to cytosolic DNA, typically observed in tumor cells. This cGAS-STING signaling pathway launches innate immune responses, producing type I interferons, which trigger M1 polarization of TAMs and subsequent adaptive immune response [85,86,87]. A STING agonist, 5,6-dimethyl xanthenone-4-acetic acid (DMXAA) (a.k.a. Vadimezan), was used in clinical trials [88,89], however, the phase III clinical trial failed to show any positive outcome. Possible reasons were proposed for this failure, pointing to the species-specificity of the DMXAA, which may not activate human STING, and that DMXAA targets only highly vascularized cancers. In contrast, the ones included in the Phase III trial had normal vasculature. Furthermore, DMXAA may cause hypoxia after vasculature reduction, which would induce the production of the vascular endothelial growth factor (VEGF) and angiogenesis. Thus, DMXAA treatment and an angiogenesis inhibitor could prove more effective against cancers. It should be noted that almost all of the studies involving the strategies of repolarizing the TAMs were combined with other treatments, such as checkpoint-inhibition immunotherapy, chemotherapy, or radiotherapy. Since the M1-polarized microglia and macrophages are known to cause recruitment and activation of NK cells and Tc cells, which have antitumor effects, most future studies will test “multi-therapy” rather than monotherapy. One more valuable message can be derived from the failed clinical trials: most therapeutics against diseases follow the general concept of targeting one signaling molecule since in vitro studies in test tubes and cultured cells are used to confirm the specificity of the targeting agent against that signaling molecule. In this process, a novel targeting molecule is synthesized and patented. However, after the FDA eventually approves this molecule to target a specific protein and treat a specific symptom, the same molecule is often found to also function on another target, allowing it to be repurposed for an unrelated condition. As a good example, the diabetes medication Metformin has been repurposed to treat Fragile X-syndrome-linked symptoms [90]. Most beneficial compounds found in nature and in our diet are similarly pleiotropic, functioning on multiple targets (Figure 1). The major difference between the dietary compounds and the synthetic compounds is centered around the fact that none of the beneficial dietary compounds shut down one specific biochemical pathway completely. In sharp contrast, many synthetic compounds do so, which often causes injury to normal, noncancerous cells, thereby precipitating significant adverse effects. Although many such beneficial natural compounds are currently being studied in preclinical studies, they are rarely considered for clinical trials. In a double-blinded placebo-controlled Phase I clinical trial of 25 subjects, including an arm of biopsy-proven head and neck cancer patients, the subjects received a synergistic drug combination (APG-157) derived from the dietary spice turmeric [91]. The drug was delivered in a pastille form that enabled topical absorption into the tumors in the oral cavity and into the oropharyngeal tumor through salivary transport and systemic absorption through sublingual and buccal absorption. Thus, the drug was rapidly absorbed directly into the tumor and showed rapid systemic absorption [92]. This study used circulating plasma cell-free RNA (cf-RNA) as an effective indicator of drug response on tumor breakdown [90]. The promising observation made by this group included the upregulation of RNA transcripts bearing signatures of an inflammatory response, leukocyte activation, and upregulation of inflammatory cytokines in APG-157-treated patients but not in the healthy or placebo-treated patients. These changes indicate an immune response and a mobilization of immune cells triggered by the treatment. An especially striking observation was the increase in TNF-α response which points to an increase in tumor apoptosis. Since inflammatory cytokines secreted by immune cells in the TME play a vital role in TAMs’ repolarization into the M1 phenotype, the increase in TNF-α transcripts in cf-RNA observed in this case reflected M2 → M1 repolarization of TAMs in the TME (Figure 2). M1 macrophages release proinflammatory cytokines, such as TNF-α, along with IL-1β, and IL-6, to activate innate immunity and kill tumor cells [93]. The pleiotropic action of the drug was further confirmed by (i) the ability of the drug to reverse the cancer-driven dysbiosis of the oral microbiome, as measured by 16S RNA sequencing, and (ii) immunofluorescence of the tumor tissues before and after the drug administration showing immune system activation by recruitment of CD8+ Tc cells to the tumor as expected when TME experiences M2 to M1 reprogramming of TAM. The cells and signaling activities of the innate immune system have been discussed in the preceding paragraphs. Still, it is equally important to understand that innate immunity also arises within aberrant cells, which are different from the innate immune cells, causing their self-elimination through apoptosis [94]. It is quite likely that such innate immunity within aberrant cells is one of the reasons why a defect in the nucleic acid sequence or structure rarely leads to cancer. Viewed from a different angle, continuous inflammatory signals released by such aberrant cells may also create a condition conducive to carcinogenesis [95]. Named as “R-loops”, cells acquire nucleic acid structures comprising an RNA–DNA hybrid and a non-template, single-stranded DNA. The R-loops have been implicated in human diseases, including repeat-expansion disorders, neurological syndromes, and cancer [96,97]. In cancer cells with mutations in, for example, the breast cancer predisposition gene BRCA1, which is known to code for a protein involved in DNA repair [94,98], a significant portion of the RNA–DNA hybrids exit the nucleus and accumulate in the cytoplasm. This gives rise to “innate immunity”, which can also occur in the presence of cytoplasmic DNA from pathogens. The signaling that results from such cytoplasmic DNA or RNA–DNA hybrids involves two major types of proteins, cGAS and the toll-like receptors TLR-3 and TLR-9, which selectively bind to cytoplasmic DNA hybrids and trigger downstream signals [99,100,101]. Although both cGAS and TLRs are expressed mainly by the innate immune cells, they are also expressed by the tumor cells. Using the classic cervical cancer cell line HeLa in culture, Crossley and coworkers achieved induction of cytoplasmic RNA–DNA accumulation by knocking down the RNA–DNA helicase (SETX) or the breast cancer gene BRCA1. Thus, they demonstrated that induction of cytoplasmic RNA–DNA hybrids sets off an innate immune response even in cancer cells, thereby triggering Ser386 phosphorylation of the interferon regulatory transcription factor 3 (IRF3), which in turn induces apoptosis [94]. The induction of cytoplasmic RNA–DNA hybrid levels also caused a dramatic increase in the signaling proteins interferon β (IFNβ), interferon-stimulated gene 15 (ISG15), ISG20, chemokine ligand 5 (CCL5), and tumor necrosis factor (TNF). In the presence of the cGAS inhibitor RU.521 or after depletion of TLR3, a sharp decrease in phosphorylated IRF3 and these downstream effectors was observed in the HeLa cells, thus establishing the involvement of cGAS and TLR3 in the RNA–DNA hybrid-triggered innate immune response. To further study the effect of the RNA–DNA hybrids in innate immune cells, Rigby and coworkers synthesized a 60-basepair RNA–DNA hybrid and transfected it into isolated and cultured dendritic cells [101]. Their experiments demonstrated that TLR9 selectively binds to the nucleic acid hybrid, thereby causing IRF3-mediated activation of type I interferons and boosting the secretion of cytokines such as IL-6 and IFN-α3. Finally, Boros-Oláh and coworkers considered the R-loop-forming genes as drug targets for cancer therapy [102]. In silico analysis by this group used The Cancer Genome Atlas (TCGA) to study 33 primary cancer types. To investigate the correlation between R-loop gene expression and survival rate among cancer patients, the authors used data from TCGA to generate Kaplan–Meier survival curves. In 70% of cases, low expression of R-loop genes, such as RNASEH2A, THOC6, PRMT1, and P1F1, was observed to be associated with prolonged survival of cancer patients with mesothelioma and a low expression of FANCM was linked to prolonged survival among breast cancer patients. However, in 30% of cases, high expression of R-loop genes, such as TREX1 and BUB3, was associated with prolonged survival of patients with cervical squamous cell carcinoma and endocervical adenocarcinoma. For ten R-loop genes (ATXN2, BRCA2, CARM1, DDX19A, RNASEH1, THOC2, THOC3, TOP1, U2AF1, and ZNF207), long-term survival was observed only in the low-expressing group of patients. This study also reported an 80% association between the expression levels of R-loop genes in cancer cell lines and their sensitivity to chemotherapeutics approved by the US Food and Drug Administration (FDA). However, they also observed significant variability in drug interactions; for example, lung small cell carcinoma and ovarian cancer cells were sensitive to most of the drugs, however, B-cell leukemia, Hodgkin’s lymphoma, head, and neck cancer, and Ewing sarcoma cells were less susceptible to the FDA-approved chemotherapeutics. The current “standard of care” involves mainly strategies of direct attack and killing of cancer cells in a tumor. In this strategy, the mutating cancer cells often develop chemoresistance, however, the chemotherapy-mediated killing of fast-dividing immune cells precipitates unwanted infections. Additionally, immunotherapy, currently used for many types of cancer, sometimes causes autoimmune attacks. A comprehensive analysis of immune checkpoint inhibitor therapy of 4489 patients with primary melanoma and a median age of 74.9 was recently reported. This study also had a follow-up survey, in which 1575 patients displayed immune-related adverse events (AE) [11]. Other AEs result from inhibiting an array of diverse signaling pathways, summarized elegantly in a few reviews [103,104]. As for successes, a report published by Merck for the PD-1 antibody drug Keytruda (pembrolizumab) in non-small cell lung carcinoma (NSCLC) showed an overall five-year survival (OS) rate of 23% in treatment-naïve patients (n = 101) and 15.5% OS in patients receiving prior treatment (n = 449). Among patients with PD-L1-expressing tumors, the OS was higher at 29% (n = 27) and 25% (n = 138), respectively [105]. As mentioned earlier, among the immune checkpoint inhibitors, several PD-1 antibodies, some CTLA-4 antibodies, and some PD-L1 antibodies have been used in clinical trials. Among these agents, atezolizumab, a PD-L1 inhibitor, appeared to have the best safety profile [106]. However, some patients treated with atezolizumab experienced chills, pyrexia, and flushing, possibly due to the activation of innate immunity by the intact human Fc region in this antibody. These relatively mild AEs were managed with paracetamol, antihistamine, and steroids only when required. In this review, we have attempted to give an overview of cancer therapy strategies at the preclinical and clinical levels, which mainly involve the innate and adaptive immune systems. First, we cite the work of Afshar-Sterle and coworkers showing that although deregulated expression of the BCL6 oncogene is observed in many B-cell lymphoma patients, deliberate overexpression of this gene in mice does not cause lymphoma unless CD28- and Fas-ligand activities are simultaneously impaired in CD8+ Tc cells [5]. We also cited the work of Shi and coworkers; knl1-deletion-mediated DNA damage concomitantly triggers apoptosis and phagocytosis of neural progenitor cells by microglia [6]. Therefore, synchronous involvement of both innate and adaptive immune cells protects an organism from DNA mutation-evoked cancer. We have also discussed some promising strategies involving immunotherapy involving the empowerment of Tc cells and the difficulties experienced in the clinic with immunotherapy. Cognizant of the promise of using innate immune cells such as activated dendritic cells and NK cells in cancer therapy, several preclinical studies have been conducted, revealing that in the presence of pathogens, the dendritic cells secrete IFN1, which causes the activation of CD8+ Tc cells [61,62,63]. We have also discussed strategies to eliminate the tumor-promoting M2 macrophages and repolarizing them into the tumoricidal M1 phenotype. The first group of studies revealed that the inhibition of the (CSF-1)/CSF-1R pathway, which triggers TAM recruitment into tumor and M2-polarization of the recruits, only pushes the TAMs to the M1-like phenotype, which also causes CD8+ Tc cell activation [80,81,82]. A few relatively new methods of TAM repolarization were also discussed, using the cGAS-STING axis and the TLR3 agonist poly-ICLC [84,85,86,87,88,89]. In our preclinical studies of both GBM and peripheral cancers, we have noticed a profound role of TAMs in initiating a cascade of events involving activated NKp46+ NK cells and CD68+ Tc cells [15,19,20,23]. Thus, it can be concluded that the innate and adaptive immune systems work in close coordination. This notion must be front and center in designing safer and more effective cancer therapy strategies. We have argued that, in contrast to many synthetic CAs that completely shut off a specific signaling axis, thus causing adverse side effects, the most beneficial dietary anticancer compounds are pleiotropic and do not completely shut off any particular pathway. However, recently, they have rarely been considered for clinical studies. Two such studies, conducted recently by Basak and coworkers and Tosevska and coworkers, used a turmeric-based drug, APG-157, in head and neck cancer patients and measured cf-RNA to note leukocyte activation and the upregulation of transcripts bearing an inflammatory response and also a reversal of cancer-driven dysbiosis of the oral microbiome [91,92]. It is perhaps understood from a large number of attempts to develop an effective strategy for difficult-to-treat cancers that we may need to divert our attention from designing molecules to directly kill the cancer cells to empowering the immune system as a whole so that patients regain the ability to eliminate the mutated cells quickly before they cause cancer. For years, epidemiological studies have shown a link between cancer and diet. Still, we have continued to synthesize new antimetabolites and drugs to selectively activate some specific immune cells without making a concerted effort to take lessons from our diet and lifestyle and apply them to empower the human body to eliminate such aberrant cells. It is time that we change our approach to conquer many deadly cancers, such as pancreatic cancer and GBM.

PMC10002489

Chiu-Jung Huang,Kong Bung Choo

Circular RNA- and microRNA-Mediated Post-Transcriptional Regulation of Preadipocyte Differentiation in Adipogenesis: From Expression Profiling to Signaling Pathway

25-02-2023

circular RNA,microRNA,post-transcriptional regulation,signaling pathways,preadipocyte differentiation,adipogenesis,species conservation

Adipogenesis is an indispensable cellular process that involves preadipocyte differentiation into mature adipocyte. Dysregulated adipogenesis contributes to obesity, diabetes, vascular conditions and cancer-associated cachexia. This review aims to elucidate the mechanistic details on how circular RNA (circRNA) and microRNA (miRNA) modulate post-transcriptional expression of targeted mRNA and the impacted downstream signaling and biochemical pathways in adipogenesis. Twelve adipocyte circRNA profiling and comparative datasets from seven species are analyzed using bioinformatics tools and interrogations of public circRNA databases. Twenty-three circRNAs are identified in the literature that are common to two or more of the adipose tissue datasets in different species; these are novel circRNAs that have not been reported in the literature in relation to adipogenesis. Four complete circRNA–miRNA-mediated modulatory pathways are constructed via integration of experimentally validated circRNA–miRNA–mRNA interactions and the downstream signaling and biochemical pathways involved in preadipocyte differentiation via the PPARγ/C/EBPα gateway. Despite the diverse mode of modulation, bioinformatics analysis shows that the circRNA–miRNA–mRNA interacting seed sequences are conserved across species, supporting mandatory regulatory functions in adipogenesis. Understanding the diverse modes of post-transcriptional regulation of adipogenesis may contribute to the development of novel diagnostic and therapeutic strategies for adipogenesis-associated diseases and in improving meat quality in the livestock industries.

Circular RNA- and microRNA-Mediated Post-Transcriptional Regulation of Preadipocyte Differentiation in Adipogenesis: From Expression Profiling to Signaling Pathway Adipogenesis is an indispensable cellular process that involves preadipocyte differentiation into mature adipocyte. Dysregulated adipogenesis contributes to obesity, diabetes, vascular conditions and cancer-associated cachexia. This review aims to elucidate the mechanistic details on how circular RNA (circRNA) and microRNA (miRNA) modulate post-transcriptional expression of targeted mRNA and the impacted downstream signaling and biochemical pathways in adipogenesis. Twelve adipocyte circRNA profiling and comparative datasets from seven species are analyzed using bioinformatics tools and interrogations of public circRNA databases. Twenty-three circRNAs are identified in the literature that are common to two or more of the adipose tissue datasets in different species; these are novel circRNAs that have not been reported in the literature in relation to adipogenesis. Four complete circRNA–miRNA-mediated modulatory pathways are constructed via integration of experimentally validated circRNA–miRNA–mRNA interactions and the downstream signaling and biochemical pathways involved in preadipocyte differentiation via the PPARγ/C/EBPα gateway. Despite the diverse mode of modulation, bioinformatics analysis shows that the circRNA–miRNA–mRNA interacting seed sequences are conserved across species, supporting mandatory regulatory functions in adipogenesis. Understanding the diverse modes of post-transcriptional regulation of adipogenesis may contribute to the development of novel diagnostic and therapeutic strategies for adipogenesis-associated diseases and in improving meat quality in the livestock industries. Adipose tissue mass may expand via increasing the size of the constituent adipocyte cells. On the other hand, adipogenesis is an adipocyte biogenesis process in which new adipocytes are generated from multipotent progenitor stem cells. Adipogenesis begins with the progenitor cells being committed to become preadipocytes, which undergo growth arrest, followed by preadipocyte differentiation into mature adipocytes [1,2]. Adipose tissues are morphologically divided into the white (WAT) and brown adipose tissue (BAT), and the beige adipose tissue, with each type playing a different physiological role. Morphologically, a mature WAT adipocyte carries a large lipid droplet but few mitochondria within the cell. Hence, WAT serves mainly as an energy reservoir; excessive lipids are stored as triglyceride, which is converted, on energy demand, to free fatty acids for circulation [3,4,5]. Dysfunctional WAT is associated with obesity, insulin-resistance in diabetes, cardiovascular disorders and cancers, among other human conditions [6,7,8,9]. On the other hand, small lipid droplets and high numbers of mitochondria are found in BAT adipocytes. Functionally, BAT serves a thermogenic function in producing heat from the lipid droplets, regulating body temperature, in addition to other secretory functions [10,11,12]. While excessive WAT in obesity is unhealthy, BAT is favorable in health in its effects on reducing the accumulation of adipose tissues and in lowering insulin resistance in diabetes patients (reviewed in [13]). Beige adipocytes may be converted from WAT adipocytes, a process dubbed WAT browning. Beige adipocytes possess smaller oil droplets and enriched mitochondria contents; hence, beige adipose tissues share many features and physiological functions with BAT [14]. WAT browning is typically induced by exposure to cold and on demand for heat [15]. In term of regulation, a key gateway to adipogenic gene expression that drives preadipocyte differentiation is the peroxisome proliferator-activated receptor (PPAR) family nuclear proteins, particularly the PPARγ isoform [16,17,18,19,20]. In several human clinical conditions, including inflammation, insulin sensitivity, obesity and cancer, PPARγ has been shown to play important causative roles (reviewed in [18]). In thermogenesis, there is also crosstalk between PPARs and thyroid hormone receptors in the adipogenesis process via competing for binding to the heterodimeric partner, retinoid X receptor or other targets [21,22,23]. Hence, adipogenesis, acting through PPARs or otherwise, is a crucial cellular process for human wellbeing and survival. In animal husbandry, studies on adipogenesis are focused on improving the quality and nutritional values of meat of livestock, which are a major source of proteins and lipids for humans. A major factor that affects meat quality is intramuscular fat content, which controls not only meat texture and taste, but also supplements of essential fatty acids [24]. Hence, extrapolations from animals to the human, and vice versa, may rapidly help expand understanding in the regulation of adipogenesis and the adipogenesis-associated proteins across species. The assumption is that if a transcript or protein sequence is highly conserved through evolution, critical biological functions are implied. Interspecies sequence conservation has, indeed, been useful in identifying causes of congenital diseases in humans [25,26]. Based on this supposition, we have previously reviewed possible extrapolation of species-conserved microRNAs (miRNAs) and the miRNA-targeted mRNAs of chicken in adipogenic gene expression in the adipogenesis process [16]. Studies have shown that adipogenesis is regulated by a wide array of genes [17]. On reaching the cytoplasm, mature mRNAs may further be modulated post-transcriptionally by regulatory RNA species, including circular RNA (circRNA) and microRNA (miRNA), in deciding “go or no go” in translation into a functional protein (Figure 1). Translation of an mRNA may be blocked by a targeting miRNA, or the mRNA may suffer degradation induced by the targeting miRNA (Figure 1B). The miRNA is, in turn, under the whip of another single-stranded but larger-sized circular RNA (circRNA) via base-pairing resulting in the “sponging” off, i.e., in depleting, of the miRNA pool to free the targeted mRNA for translation (Figure 1C). Subsequently, the translated protein goes into signal transduction or other biochemical pathways to regulate adipogenic gene expression (Figure 1A–C) [16,27]. Clinical studies have indicated that circRNAs and miRNAs may be encapsulated in exosomes, or simply released as free forms into the blood stream of patients to be transported to destination cells, which may or may not express the circRNA or miRNA, to exert gene regulatory functions in the destination cells (Figure 1D) [28]. It is noted here in passing that the long noncoding RNAs (lncRNAs) constitute another unique group of noncoding RNAs that interact with other biomolecules, including circular RNAs and microRNAs, to affect biological functions at multiple regulatory levels [29,30,31]. However, lncRNAs are not the focus of this work, and are discussed only in relation to their interactions with the circRNA and miRNA networks being analyzed. The biogenesis of and functions of miRNA have been extensively reviewed [32,33,34,35]. Only a synopsis of key features relevant to this review is given here. Each miRNA is encoded by a single gene, which may occur in clusters; an extensively studied example is the chromosome 19 miRNA cluster, or C19MC, which includes 46 individual miRNA genes [36,37]. Evolutionary transposition has also generated miRNA families, each with multiple family members with identical or highly homologous sequences; a notable example is the let-7 miRNA family [38,39]. It is noteworthy that many miRNA genes are located within the intron sequences of many protein-coding genes, a genomic framework also shaped by evolution [40,41]. In the process of miRNA biogenesis, a hairpin precursor is first formed, which is further processed to generate either or both the 3- or 5-prime (3p or 5p) mature miRNA species with different sequences targeting different mRNAs in action (Figure 1B) [42,43]. Hence, the suffix -3p or -5p is important in designing a specific miRNA to avoid confusion. The soul of the 17–21-nucleotide miRNA is the “seed sequence” of 5–8 nucleotides located at the 5′-end of the short RNA sequence. Members of the same miRNA family share identical seed sequences. In targeting an mRNA, the miRNA seed sequence interacts with a complementary sequence in the 3′-untranslated region (3′-UTR) of the mRNA; the sequences on both sides of the seed regions often show low sequence homology without affecting the mRNA-targeting action of the miRNA. The biogenesis of circRNAs is more elaborate [44]. Unlike miRNA, which each have a specific-gene status, a circRNA is the backsplicing offspring of a pre-mRNA of a specific host gene by stitching together one or more selected exons and/or intron sequences of the pre-mRNA into a circular RNA molecule of assorted sizes (Figure 1C). Backsplicing requires canonical pre-mRNA spliceosomal machinery, recognizing the same consensus intron–exon GU-AG junctions. The process is also modulated by the pairing of intronic repetitive sequences, such as Alu, and contributions by cis complementary sequences and trans-acting protein factors [45,46]. Multiple circRNA species, or isoforms, may be generated from a specific host transcript. Different isoforms carry different sequences and, therefore, sponging off of different miRNA targets. In the literature, the circRNA nomenclature has not been standardized, and is, therefore, rather confusing. A circRNA designation may be provided as: (i) a circBase (http://www.circbase.org, accessed on 20 December 2022) ID, which is specific for each circRNA, but does not provide any hint of the isoform on first sight; (ii) the host gene name but often without indication of the isoform being studied; or (iii) the chromosomal location of the circRNA sequence, which is not reader friendly. In this review, efforts are put into identifying both the circBase ID and the host gene name of the reported circRNAs being discussed. We have previously reviewed the miRNA–mRNA signaling axis that impacts the PPARγ gateway in the adipogenesis process [16]. In this review, we are extending the literature appraisal to include circRNAs in the post-transcriptional regulatory network, focusing on circRNA–miRNA and miRNA–mRNA crosstalk in regulating preadipocyte differentiation. The approach used is integration and analysis of information and datasets harvested from literature scrutiny, interrogation of circRNA and miRNA databases and analysis of the acquired information using bioinformatics and algorithmic tools. To uncover novel circRNAs associated with adipogenesis, whole-transcriptome profiling has been applied in humans and assorted experimental and domesticated farm animals. Reports of such studies are obtained through searches in the PubMed (https://pubmed.ncbi.nlm.nih.gov, accessed on 20 December 2022) database. Many studies are focused on the critical step of differentiation of preadipocytes into mature adipocytes by examining differential expression (DE) data before and after differentiation induced in vitro (Table 1A). Besides preadipocyte differentiation, comparative studies on subjects relevant to adipogenesis are also presented, e.g., circRNA profiling in adipocytes between obese and lean individuals, and between developmental stages in calf and adult cattle (Table 1B) [47,48]. We also include here a paper that does not have DE data but provides useful datasets on visceral (VAT) and subcutaneous adipose tissues (SAT) in humans and mice (Table 1B, rows B2 and B3) [49]. In total, 11 studies with 12 datasets that encompass the human and six animal and bird species are included in this review (Table 1). In early circRNA profiling studies, a human circRNA microarray platform was used, which generated non-discriminating and not-user-friendly datasets [47,50]. Subsequently, most circRNA profiling works used the high-throughput whole-transcriptome RNA sequencing (RNA-seq) platform. The quality of the profiling datasets obtained are affected by the RNA preparations used in the RNA-seq work. In a study in which total RNA preparations were used in a microarray platform, a high number (4080) of differentially expressed (DE) circRNAs are reported, while the total number of circRNAs was not revealed (Table 1A, row A1) [50]. In most studies, ribosomal RNA-free RNA preparations were used. In a few studies, circRNA was further enriched by RNase R treatment to remove other linear RNA species [47,51,52]. The RNA preparations were then used in constructing unidirectional strand-specific RNA libraries for RNA-seq, followed by data analysis using appropriate algorithms and bioinformatics tools. In most cases, the circRNA expression datasets, after various degrees of annotation and organization, were submitted either as Supplementary Materials for online access on publication of the papers, or were deposited in some public databases, including miRbase (https://www.mirbase.org, accessed on 20 December 2022) and miRDB (https://mirdb.org, accessed on 20 December 2022). In this review, the availability of useful circRNA profiling datasets is indicated (Table 1). For whole-transcriptome profiling, the availability of useful and user-friendly datasets is important for comparative analysis. Many authors provide circRNA names either in the host name nomenclature, e.g., circSAMD4A, or as circBase identification (ID) numbers, e.g., hsa_circ_0004846. However, authors, except Arcinas et al. (2019) [49], often neglect to be more specific when different circRNA isoforms are detected [49]. Other useful circRNA identification information includes the GenBank (https://www.ncbi.nlm.nih.gov/genbank/, accessed on 20 December 2022) accession number of the host gene transcript from which the circRNA is derived, the exons retained in the circRNA and the chromosomal positions of the retained exons [49]. In many circRNA profiling papers, only circBase ID numbers are provided, which make it difficult to prompt the identity of the circRNA concerned. Hence, we have put in efforts in this review on better identifying the specific host genes and translating the circBase ID into the host gene nomenclature whenever is possible. In some works, the supplementary circRNA datasets provided are in chromosomal positions of the predicted retained exons without the putative circRNA IDs or host gene names, which makes cross referencing difficult [53,54]. In some cases, accessible and useful datasets are unavailable (Table 1). Discounting the microarray-based work on human by Sun W et al. (2020) [50], the total number of circRNAs expressed in the adipocytes in different species ranges from 2172 in pig to 7203 in yak (Table 1A) [53,54]. On differentiation, the fraction and the number of differentially expressed (DE) circRNAs in the mature adipocytes relative to the preadipocytes before differentiation ranges from 1.09% (41 circRNAs) to 3.02% (117) in the stromal cells of mouse WAT and BAT, respectively (Table 1A, rows A2 and A3). A much higher, but seemingly unrealistic, DE number of 4080 was observed when total RNA was used in the microarray platform [50]. In a study in pig in which lncRNA and mRNA were also included in the analysis [54], a high fraction (13.67%, 297 circRNAs) of DE circRNA was also reported, probably due to the use of different algorithms in analyzing the sequencing datasets. Hence, these two studies are excluded from further analysis. Taken together, the small number of DE circRNAs may indicate that only a small number of circRNAs participate in the preadipocyte differentiation process (Table 1A). Furthermore, DE circRNAs may be up- or downregulated, suggesting that circRNAs may act as positive or negative modulators in modulating downstream adipogenic gene expression. Besides differentiation, four comparative studies, generating six datasets, on adipogenesis-related subjects have been included (Table 1B). In the work comparing human obese vs. lean subjects, 244 DE circRNAs are revealed (Table 1B, row B1) [47]. In comparing two breeds of pig with different fat contents, 275 circRNAs, mostly downregulated, are reported (Table 1B, row B4) [55]. Further examination of the DE circRNAs presented in these two works may contribute to producing slimmer persons in the clinic and leaner farm animals in the meat industry. In development-related adipogenesis, 67 (1.38%) DE circRNAs are reported between young and old rats and 307 (7.08%) in calf and adult cattle (Table 1B, rows B5 and B6) [48,56]. A comprehensive and informative profiling work on preadipocyte differentiation in mouse WAT-1 is presented by Zhang P.P. et al. (2021) [51] (Table 1A, row A2). In the study, 28 circRNAs are identified as upregulated and 13 as downregulated on differentiation, including four circRNAs, each with two isoforms (Table 2, row 1). Based on the log2fold changes of these 41 DE circRNAs provided in this work, comparative assessments are made with other WAT and BAT circRNA datasets in mouse, human and yak, reported by other authors (Table 2, rows 2–5; Supplementary Table S1). When compared with the mouse WAT-2 dataset of Arcinas et al. (2019) [49], only 25 (61.0%) of the 41 circRNAs in the Zhang P.P. et al. (2021) [51] dataset are found in both mouse datasets, including 20 up- and 5 downregulated in expression on differentiation (Table 2, rows 1 and 2; Supplementary Table S1). The 25 circRNAs may be considered as validated circRNAs that are involved in preadipocyte differentiation in the mouse WAT. Interestingly, PubMed interrogations indicate that none of these circRNAs have been reported before in relation to adipogenesis, indicating that they are novel circRNAs awaiting further investigation into their regulatory role in adipogenesis. These circRNAs are not further discussed here. In the human and yak WAT, 23 (56.1%) and 19 (46.3%) circRNAs are in common in the mouse WAT-1 dataset, respectively (Table 2, rows 3 and 4; Supplementary Table S1). The common circRNAs that appear in the WAT datasets of mouse, human and yak may be interpreted as that these circRNAs that are conserved in mandatory functions in modulating preadipocyte differentiation in different species. Thirty-two (78.0%) circRNAs, including three isoforms, are commonly expressed in WAT and BAT differentiation in the different species analyzed (Table 2, row 5; Supplementary Table S1), indicating common circRNA regulatory pathways in WAT and BAT differentiation. It is also noted that 9 of the 41 mouse WAT-1 circRNAs are WAT-specific and are not detected in mouse BAT (Supplementary Table S1), suggesting that some regulatory pathways are exclusive to WAT differentiation. The nine circRNAs are chr17:34877211-34956589, Cacna1d and Fancl in the upregulated group and Rad18, Megf8, Trpc6, Zfp532, Dcbld2, Zfx in the downregulated group. The comparative analysis presented here is based on a limited number of datasets and species analyzed. The predictions made should be taken with caution pending further confirmation of their involvement in adipogenesis differentiation. In two RNAseq profiling studies (Table 1B, rows B1 and B5), the circRNA–miRNA–mRNA trio association, validated in luciferase assays, are reported: circSAMD4A-miR-138-5p-EZH2 mRNA in the human and the circFUT10-let-7c-5p-PGC1β/PPARGC1B in the cattle; circRNA-mediated preadipocyte differentiation was also demonstrated in both cases (Table 3) [47,48]. Further PubMed searches were conducted using the stringent criteria of validation of circRNA–miRNA and miRNA–mRNA interactions by luciferase and mutational or pulldown analysis. Two other circRNAs are identified: the bovine bta_circ_Pparγ and bta_circ_Flt1 (Table 3). Taken together, four circRNAs with established circRNA–miRNA–mRNA connections are found in one or more of the general WAT and BAT circRNA datasets of human, mouse and yak presented in Table 1 and Table 2 above (Table 3). However, they are not among the list of 41 differentially expressed circRNAs in the mouse dataset, most likely because of the imposed criteria constraints in the identification of these circRNAs, and possibly because of species differences. The basic molecular features of the four selected circRNAs, including the exons of the host transcript, retained, size, transcript ID of the host transcript and chromosomal positions are shown in Supplementary Table S2. It is noted that circPPARγ- and circFLT1-modulated miR-92a-3p and miR-93-5p affect more than one mRNA species and, therefore, different cellular processes, and that two long noncoding RNAs also participate in the circFLT1-miR-93-5p regulation (Table 3; see below). Sequence conservation in modulatory RNAs across species is a good indicator of the importance of the regulated cellular functions. If found highly conserved, similar action and function may be predicted across species [16]. Since miRNA plays a central role in connecting circRNA to mRNA, the miRNAs in the validated interactions described above are first examined (Table 3). It is first noted that each of the four miRNAs belongs to a specific miRNA family, and that the genes of miR-92a-3p and miR-138-5p are found in two different chromosomal clusters (Table 3; Supplementary Figure S1). Since members of the same family share identical seed sequences, family members may target the same transcript and share regulatory functions. On the other hand, appearance in chromosomal clusters is an indication of active sequence evolutionary histories [63,64]. The miRNA sequences of the human, mouse, rat, pig, bovine and chicken obtained through miRBase and TargetScan (https://www.targetscan.org, accessed on 20 December 2022) interrogations are aligned (Figure 2A and Table 4). Except for the unavailability of the pig and chicken miR-93-5p sequences, the 6–7-nucleotide seed sequences of all four adipogenesis-associated miRNAs are identical in the six species analyzed, including the avian chicken. Moreover, the non-seed sequences of the miRNAs are also highly conserved (Figure 2A). The observed miRNA sequence conservation supports the proposition that these miRNAs play crucial functional roles in adipogenesis across species. CircRNA sponging of miRNA kickstarts the regulatory role of circRNA in modulating mRNA translation and the subsequent cellular functions. For circRNA–miRNA alignments, circRNA sequences in human, mouse and bovine are obtained from the circBase and circBank (http://www.circbank.cn, accessed on 20 December 2022) databases. The analysis shows that the miRNA seed sequences align perfectly with the targeted sequences of the circRNAs (Table 4), allowing for the rare wobble G-U base-paring in the RNA species (Figure 2B(i,ii), underlined nucleotides). However, a single mismatch in the circFLT1-miR-93-5p pair and two mismatches in circFUT10-let-7c-5p in the seed sequences in the mouse are noted (Figure 2B(ii,iii), in black letters). Furthermore, the circRNA sequences outside the seed regions also show a high degree of sequence homology in the three species, further supporting species conservation of the circRNA and miRNA interactions. Literature scrutiny has revealed that miR-92a-3p targets the C/EBPα (CCAAT Enhancer-Binding Protein alpha) and p130/Rb2 (Retinoblastoma 2/p130) transcripts and miR-93-5p targets SIRT7 (Sirtuin-7) and TBX3 (T-Box Transcription factor 3) (Table 3). On alignment of the miRNA and mRNA sequences of interaction, all seven to eight nucleotides of the miRNA seed sequences are found to align perfectly with the complementary targeted sites in the 3′-UTR (3′-untranslated region) of the mRNAs in most cases, with only rare single-nucleotide disparity in some cases, particularly in chicken (Figure 2C; Table 4, last column). Furthermore, two miRNA target sites are found in the 3′-UTR of PGC1β/PPARGC1B (PPARγ Coactivator 1-β) mRNAs of all species analyzed, both of which are also fully conserved in the let-7c-5p seed sequence (Figure 2C(iii)). The interaction of miR-138-5p-EZH2 (Enhancer of Zeste Homolog 2) is also perfectly aligned (Figure 2C(iv)). Taken together, the high degree of miRNA seed sequence conservation in the miRNA–mRNA interactions between species predicts conservation of the modulation mechanism of adipogenesis across species. In this section, the four selected circRNAs and the associated miRNAs and mRNAs that modulate preadipocyte differentiation are individually discussed. The salient molecular features of the four circRNAs are shown in Supplementary Table S2. Emphasis in our discussion is on the molecular events controlled by the proteins, the translation of which is regulated by the circRNAs and miRNAs. We have shown above that the circRNAs, miRNAs and mRNAs concerned are highly conserved in the interacting sequences (Table 4; Figure 3). Hence, the events are discussed in general without species specification. However, the human or animal species of the RNAs investigated in the cited studies is specified. On induced differentiation of bovine adipocytes, bta_circ_Pparγ (bta_circ_0010660) inhibits adipocyte apoptosis and proliferation while promoting adipocyte differentiation via the sponging of miR-92a-3p [57]. However, the authors did not identify the mRNA targeted by miR-92a-3p. In an early study, the whole of the miR-17/92 cluster, amongst which is miR-92a (Supplementary Figure S1), was used in preadipocyte differentiation studies in mouse 3T3L1 preadipocyte cells [58]. Upregulated expression of members of the miR-17-92 cluster is shown to promote the clonal expansion stage of adipocyte differentiation via targeting p130/RB2 (Retinoblastoma 2), echoing a previous report [65], and supported by our seed sequence analysis that miR-92a-3p targets a 3′-UTR sequence of the p130 mRNA, and the seed sequence is conserved in different species (Figure 2C(i); Table 4). Before terminal differentiation, differentiating preadipocytes are arrested in growth when re-entry of the cell cycle is blocked [66]. circPPARγ-induced downregulation of miR-92a-3p results in increased p130 levels to enhance p130/E2F dimerization [67] and association with the transcription factor DP-1 [66,68]. The consequence is the exit of the cell cycle, growth arrest and terminal differentiation to form mature adipocytes. In uncommitted human bone marrow adipose tissue-derived stromal cells, absence of p130 has, indeed, been shown to hamper terminal adipocyte differentiation [69]. Hence, targeting p130 is the first route by which circPPARγ exerts its influence on adipogenesis via miR-92a-3p by influencing cell cycle, growth arrest leading to terminal differentiation (Figure 3A, left panel, route I). It has also been reported that C/EBPα, when induced in mouse 3T3-L1 preadipocytes in the early stage of differentiation, prompts p130/E2F association via p21 upregulation [59]. In this way, C/EBPα may also regulate E2F availability in the activation of the cell cycle in the clonal expansion stage, leading to terminal differentiation and formation of mature adipocytes (Figure 3A, left panel, route II). In a different study, chronic myeloid leukemia (CML)-derived exosomes that harbor human miR-92a-3p are shown to promote adipogenesis of adipose-derived mesenchymal stem cells [60]. Our analysis supports that the seed sequence of miR-92a-3p perfectly complements a specific sequence of the 3′-UTR of the C/EBPα transcript, and that the seed sequence is conserved (Figure 2C(i); Table 4). In the same CML-exosome study, in vitro studies show that miR-92a-3p suppresses PPARγ and C/EBPα expression and, consequently, the expression of the adipogenic genes, FABP4 (Fatty Acid Binding Protein 4) and AdipoQ (Adipocyte, C1q And Collagen Domain-Containing Protein) (Figure 3A, right panel, route III) [60]. The results are consistent with a previous report that high C/EBPα levels trigger preadipocyte differentiation and WAT development [70]. The C/EBPα-C/EBPβ heterodimer often acts in concert with PPARγ to form a gateway to adipogenic expression and the maintenance of the differentiated state of adipocytes (Figure 3A, right panel, route III) [16,71,72,73]. Importantly, CML-derived exosomal miR-92a-3p is linked to induction of loss of bodyweight via WAT browning and increased energy expenses in cancer-associated cachexia [8]. In induced differentiation of bovine preadipocytes, miR-93, which should be miR-93-5p, is identified as the top expressing miRNA [61]. Through TargetScan analysis and luciferase assays, circFLT1 (bta_circ_002673) and the long noncoding RNAs, lncCCPG1 and lncSLC30A9, are shown to bind competitively to miR-93-5p. On the other hand, circFLT1 and lncCCPG1 also compete to deplete miRNA-93-5p from binding to lncSLC30A9 to offset the lncSLC30A9 action in inducing upregulated expression of PPARγ, C/EBPα and FABP4, leading to preadipocyte differentiation (Figure 3B, route I). The mechanism proposed by the authors is that lncSLC30A9 binds to and transports c-Fos into the nucleus to activate the PPARγ promoter, leading to differentiation (Figure 3B, route I) [61,74]. MiR-93 is a member of the miR-106b/25 cluster (Supplementary Figure S1). In another study using miR-106b/25-knockout mice, miR-93-5p is shown to target SIRT-7 (Sirtuin-7) [62]. SIRT-7 is a NAD-dependent deacetylase of histones that induces transcriptional repression [75]. Since SIRT-7-knockout mice have less visceral fat, the gene is linked to adipogenesis [76]. In knockout mice, SIRT-7 is shown to deacetylate and, hence, activate another SIRT protein, SIRT-1, in the preadipocyte differentiation process (Figure 3B, route II) [77]. On the other hand, FOXO1 (Forkhead Box O1), previously inactivated upstream by being acetylated and also phosphorylated by AKT signaling, is now being re-activated by deacetylation by SIRT-1 and dephosphorylation by PP2A (protein phosphatase 2). Subsequently, the activated FOXO1 protein binds to the PPARγ promoter to block PPARγ expression in cis, or interacts with PPARγ in trans, to deplete PPARγ for utilization in adipogenesis (Figure 3B, route II) [16,78,79,80,81]. When miR-93-5p is sponged by circFlt-1, expression of SIRT-7 is upregulated, suppressing SIRT-1 expression to prevent deacetylation and re-activation of FOXO1, thus, allowing PPARγ to participate in adipogenesis. In the same study, miR-93-5p targeting of TBX3 (T-Box Transcription Factor 3) is demonstrated, which results in suppression of self-renewal in adipocyte precursors before commitment to differentiation in the very early stage of adipogenesis (Figure 3B, route III) [62]. TBX3 has previously been shown to contribute to osteogenic differentiation of human adipose stroma cells and in maintaining pluripotency via targeting the promoter of Oct-4, one of crucial pluripotency inducers in precursor stem cells [82,83]. In RNAseq analysis, one of top differentially expressing circRNAs in adipose tissues of both young and adult cattle is circFUT10 (Table 1, row B6) [48]. In cattle, circFUT10 (circRNA ID not available) is further shown to promote adipocyte proliferation by increasing the number of adipocytes in the S and G2 phases of the cell cycle, while suppressing adipocyte differentiation in in vitro assays in bovine adipocyte cells. CircFUT10 sponges let-7c, which we show is let-7c-5p, and that let-7c-5p targets PGC1β/PPARGC1B (PPARγ coactivator 1-β) (Figure 3C). PPARγ acts in collaboration with retinoid X receptor (RXR) to bind to PPARγ response elements in the promoters’ PPARγ-modulated genes in the adipogenesis process [84,85]. PGC1β also associates with PPARγ to induce further PPARγ interactions with other transcription factors in various processes, including tumorigenesis [86,87]. In adipogenesis, however, PGC1β acts as a PPARγ repressor in adipocyte differentiation [88,89]. It is also noteworthy that PGC1β has been shown to be activated, at least in part, by PRDM16 (PR/SET Domain 16 Protein) in the earlier fate determination of brown fat adipogenesis [90]. Whether PRDM16 also activates PGC1β in preadipocyte differentiation remains to be shown. In a circRNA profiling work on adipose tissues in obese and lean human individuals, one of the top-expressing circRNAs in preadipocytes is circSAMD4A (hsa_circ_0004846). CircSAMD4A expression levels are correlated with bodyweight (Table 1, row B1) [47]. Knockdown of circSAMD4A downregulates expression of PPARγ and C/EBPα and inhibits preadipocyte differentiation via sponging of miR-138-5p, which releases EZH2 (Enhancer of Zeste Homolog 2) mRNA from miR-138-5p translational suppression (Figure 3D, top portion). miR-138-5p has, indeed, been shown earlier to be a negative modulator of adipogenic differentiation of human adipose-derived mesenchymal stem cells [91]. EZH2 is a histone methyltransferase that epigenetically regulates gene expression through methylation of the histone H3K27, resulting in chromatin changes [92,93,94]. Under normal circumstances, methylated H3K27 binds to the WNT promoter, thereby suppressing WNT expression [95]. Involvement of the EZH2-WNT signaling in adipogenesis has been independently reported by several laboratories [95,96,97]. The WNT-1 protein, on entering the nucleus of preadipocyte cells, downregulates expression of PPARγ and C/EBPα through β-catenin association with other transcription factors to consequently suppress preadipocyte differentiation [16,98]. In short, circSAMD4A upregulates EZH2 expression via sponging miR-138-5p to boost H3K27 histone methylation, thereby suppressing canonical WNT/β-catenin signaling and activating the PPARγ-C/EBPα gateway to advance adipogenesis (Figure 3D). It is noteworthy that miR-138-5p and EZH2 are hot subjects in cancer research. circSAMD4A also sponges another miRNA, viz. miR-1244, which targets the transcript of ubiquitin protein ligase MDM2 in promoting proliferation and enhancement of stem cell characteristics of osteosarcoma cells [99]. Besides circSAMD4A, miR-138-5p is also targeted by lncHCP5, lncSNHG7 and lncDSCAM-AS1 in promoting tumor growth in various cancers, all of which also act through EZH2 [100,101,102]. Furthermore, besides the WNT/β-catenin signaling in adipogenesis, the miR-138-5p and EZH2 act in concert to affect other signaling pathways in the tumorigenesis process [103,104,105]. All such observations further support that there exist cross talks between adipogenesis and tumorigenesis. In this review, we have identified and elucidated the regulatory role of sets of circRNAs and miRNAs that modulate post-transcriptional expression of proteins directing chemical signals to the PPARγ-C/EBPα gateway and other entry points to activate adipogenic gene expression in preadipocyte differentiation. A summary of the pathway of analysis and the major findings of the four complete circRNA- and miRNA-mediated regulatory pathways leading to adipogenesis is shown in Figure 4. Part (I) involves dissection of circRNA profiling datasets, which leads to the identification of 32 novel WAT and BAT differentiation-associated circRNAs that await elucidation in their regulatory roles in adipogenesis. In part (II) of our analysis, four circRNAs and the respective interacting miRNAs and mRNAs are identified and the downstream signaling and biochemical pathways are analyzed. A salient finding is conservation in the seed sequence of interactions in the circRNA–miRNA and miRNA–mRNA pairs, supporting that these circRNAs and miRNAs play crucial roles in post-transcriptional regulation in preadipocyte differentiation in the adipogenesis process. Sequence conservation may also justify extrapolations and projections of data between the human and animal species, pending more direct demonstration cross species, but speeding up clinical studies in the human in adipogenesis-associated diseases. No less important, elucidation of regulatory circRNAs and miRNAs in adipogenesis may also have impacts on improving meat quality in the livestock industry.

PMC10002496

Ameneh Ghaffarinia,Ferhan Ayaydin,Szilárd Póliska,Máté Manczinger,Beáta Szilvia Bolla,Lili Borbála Flink,Fanni Balogh,Zoltán Veréb,Renáta Bozó,Kornélia Szabó,Zsuzsanna Bata-Csörgő,Lajos Kemény

Psoriatic Resolved Skin Epidermal Keratinocytes Retain Disease-Residual Transcriptomic and Epigenomic Profiles

25-02-2023

psoriasis,keratinocyte,relapse,transcriptomics,epigenomics,5-mC,5-hmC

The disease-residual transcriptomic profile (DRTP) within psoriatic healed/resolved skin and epidermal tissue-resident memory T (TRM) cells have been proposed to be crucial for the recurrence of old lesions. However, it is unclear whether epidermal keratinocytes are involved in disease recurrence. There is increasing evidence regarding the importance of epigenetic mechanisms in the pathogenesis of psoriasis. Nonetheless, the epigenetic changes that contribute to the recurrence of psoriasis remain unknown. The aim of this study was to elucidate the role of keratinocytes in psoriasis relapse. The epigenetic marks 5-methylcytosine (5-mC) and 5-hydroxymethylcytosine (5-hmC) were visualized using immunofluorescence staining, and RNA sequencing was performed on paired never-lesional and resolved epidermal and dermal compartments of skin from psoriasis patients. We observed diminished 5-mC and 5-hmC amounts and decreased mRNA expression of the ten-eleven translocation (TET) 3 enzyme in the resolved epidermis. SAMHD1, C10orf99, and AKR1B10: the highly dysregulated genes in resolved epidermis are known to be associated with pathogenesis of psoriasis, and the DRTP was enriched in WNT, TNF, and mTOR signaling pathways. Our results suggest that epigenetic changes detected in epidermal keratinocytes of resolved skin may be responsible for the DRTP in the same regions. Thus, the DRTP of keratinocytes may contribute to site-specific local relapse.

Psoriatic Resolved Skin Epidermal Keratinocytes Retain Disease-Residual Transcriptomic and Epigenomic Profiles The disease-residual transcriptomic profile (DRTP) within psoriatic healed/resolved skin and epidermal tissue-resident memory T (TRM) cells have been proposed to be crucial for the recurrence of old lesions. However, it is unclear whether epidermal keratinocytes are involved in disease recurrence. There is increasing evidence regarding the importance of epigenetic mechanisms in the pathogenesis of psoriasis. Nonetheless, the epigenetic changes that contribute to the recurrence of psoriasis remain unknown. The aim of this study was to elucidate the role of keratinocytes in psoriasis relapse. The epigenetic marks 5-methylcytosine (5-mC) and 5-hydroxymethylcytosine (5-hmC) were visualized using immunofluorescence staining, and RNA sequencing was performed on paired never-lesional and resolved epidermal and dermal compartments of skin from psoriasis patients. We observed diminished 5-mC and 5-hmC amounts and decreased mRNA expression of the ten-eleven translocation (TET) 3 enzyme in the resolved epidermis. SAMHD1, C10orf99, and AKR1B10: the highly dysregulated genes in resolved epidermis are known to be associated with pathogenesis of psoriasis, and the DRTP was enriched in WNT, TNF, and mTOR signaling pathways. Our results suggest that epigenetic changes detected in epidermal keratinocytes of resolved skin may be responsible for the DRTP in the same regions. Thus, the DRTP of keratinocytes may contribute to site-specific local relapse. Psoriasis is a relapsing–remitting immune-mediated skin disorder characterized by epidermal hyperplasia, and massive inflammatory infiltrates as hallmarks of scaly erythematous lesions [1,2]. The global prevalence of psoriasis is around 125 million people worldwide [3]. Successful treatment of psoriasis leads to the normalization of epidermal thickness, reduced leukocyte infiltrate, and the return to a clinically and histologically normal condition [4,5,6], referred to as resolved skin. However, many patients do not attain sustained remission and often experience the reappearance of cutaneous symptoms within weeks or months of cessation of the therapy [7,8]. The psoriatic lesions frequently occur on the so-called predilection sites, such as knees and elbows, and they usually reappear on the same body parts. On the other hand, patients can also identify body regions never affected by psoriasis, which we can refer to as never-lesional skin. Previously, we showed that the psoriatic uninvolved skin in patients with severe plaque-type psoriasis maintains molecular, cellular, and extracellular alterations and displays a “pre-psoriatic” phenotype; however, in these studies, we did not discriminate between psoriatic never-lesional and resolved uninvolved skin [9,10,11,12,13]. Since disease flare-up most commonly occurs in the resolved skin, we decided to distinguish between resolved and never-lesional skin to find clues for the recurrence of the disease. Tissue-resident memory (TRM) T cells retained in resolved skin have been suggested as the main drivers of psoriasis relapse [9,14,15,16]. However, whether structural skin cells such as epidermal keratinocytes play a role in the recurrence of the clinical symptoms or not is unclear. At the molecular level, previous bulk microarray and RNA sequencing studies have shown abnormal expression of immune and skin structural genes within resolved skin [14,15,16,17,18]. Nonetheless, it is difficult to dissect the cell-specific transcripts in a whole-skin transcriptome due to its diverse cell populations. Furthermore, it is now apparent that epigenetic regulatory mechanisms such as DNA methylation and DNA hydroxymethylation play a pivotal role in the pathogenesis of several skin diseases, such as psoriasis [19,20,21,22,23,24,25,26,27,28,29,30]. However, the epigenetic changes contributing to psoriasis relapse remain unknown. In order to clarify the role of keratinocytes in the recurrence of lesions in resolved skin, we looked at epidermal keratinocyte transcriptomic and epigenomic differences in resolved vs. never-lesional skin. Since the specific transcripts of epidermal keratinocytes often remain masked in whole-skin transcriptomic analyses, we performed high-throughput RNA sequencing separately on the epidermal and dermal compartments. The visualization of the general pattern of DNA methylation and DNA hydroxymethylation epigenetic marks, 5-mC (5-methylcytosine) and 5-hmC (5-hydroxymethylcytosine) was performed on the paired resolved and never-lesional skin sections using immunofluorescence staining. Finally, we examined the overlap between resolved vs. never-lesional skin in terms of differentially expressed genes (DEGs) from our study, and the lesional vs. healthy skin DEGs from the available datasets. We observed decreased contents of 5-mC and an apparent loss of 5-hmC in resolved epidermis vs. never-lesional epidermis. Additionally, the 5-hmC decreased levels were accompanied by decreased ten-eleven translocation (TET) 3 mRNA expression in resolved epidermis as compared to never-lesional epidermis. There were 102 genes that overlapped among the DEGs of resolved vs. never-lesional epidermis and lesional vs. healthy epidermis. These data suggest that epidermal keratinocytes of resolved skin may contribute to a local relapse in psoriasis, possibly because they are not epigenetically and transcriptionally fully recovered to the baseline level (never-lesional skin). The experimental workflow is summarized in Figure 1. Using immunofluorescence staining and confocal microscopy, we observed that 5-mC and 5-hmC amounts were overall lower in resolved epidermis as compared to never-lesional epidermis. We detected a uniform 5-mC distribution pattern in both never-lesional and resolved epidermis and enhanced 5-hmC intensity in the suprabasal layer of never-lesional vs. resolved epidermis (Figure 2a). On higher resolution images (oil immersion objective), 5-mC and 5-hmC nuclei staining were apparently weaker in the resolved epidermis compared to the never-lesional samples (Figure 2a, closeups). False-colored low-magnification images reported these findings in each patient, although a few 5-hmC hot patches were observed in the resolved epidermis (Figure 3). The 5-hmC epigenetic mark is an intermediate product of active DNA demethylation. This refers to an enzymatic process in which ten-eleven translocation (TET)1, TET2, and TET3 enzymes oxidize the methyl group of 5-mC and convert it to 5-hmC [31,32]. Therefore, we analyzed mRNA expression levels of these enzymes in paired never-lesional and resolved epidermal samples to verify whether there was a link between 5-hmC loss in resolved epidermis and gene expression alterations of these enzymes. The real-time RT-PCR results showed a significant decrease in TET3 (p = 0.0002) and no changes in TET1 and TET2 mRNA expression level in resolved vs. never-lesional epidermis (Figure 2b). Overall, we observed a correlation between a 5-hmC decreased level and TET3 mRNA expression in resolved epidermis. These results are consistent with previous findings confirming that the loss of 5-hmC was associated with lower mRNA expression of TET enzymes in the epidermis of psoriatic lesions [28]. A pairwise comparison of RNA sequencing profiles from paired resolved and never-lesional epidermal and dermal compartments was performed to determine differences in gene expression. Transcriptome analysis of four patients samples yielded 476 DEGs in resolved epidermis vs. never-lesional epidermis (p < 0.05). Of these, 275 (≈57.77%) were down-regulated, and 201 (≈42.23%) were up-regulated. The same analysis of three patients yielded 2966 DEGs in resolved dermis vs. never-lesional dermis (p < 0.05). Of 2966 DEGs, 1360 (≈45.85%) were identified as down-regulated, and 1606 (≈54.15%) as up-regulated genes in the resolved dermis vs. never-lesional dermis. Principle component analysis (PCA) was performed using the resolved vs. never-lesional DEGs as input. The analyses showed a clear separation between all resolved and never-lesional paired samples (Figure S1a,b, see Supplementary Materials). To narrow the search for candidate genes that may play a role in psoriatic lesion recurrence, we examined the 25 most down-regulated and up-regulated DEGs in resolved vs. never-lesional epidermal and dermal compartments (Table 1). Overall, three genes with |FC| ≥ 2.5 were decreased in the resolved epidermis compared with the never-lesional epidermis, including nuclear enriched abundant transcript 1 conserved region 3 (NEAT1_3), sterile alpha motif and HD domain-containing protein 1 (SAMHD1), and Homeobox protein Hox-B2 (HOXB2) (Table 1). Among these, NEAT1_3 and SAMHD1 can act as inflammation regulatory genes. The NEAT1_3 gene produces a long non-coding RNA (lncRNA) with critical roles in innate immune responses [33,34,35]. Screening of the expression of some common lncRNAs in psoriasis using real-time RT-PCR has shown decreased expression of the NEAT1 gene in lesional skin compared to healthy skin [36]. The gene SAMHD1 is the only deoxynucleoside triphosphohydrolase (dNTPase) in eukaryotes, and is required for genome integrity by tightly controlling the intracellular deoxynucleoside triphosphate (dNTP) pools to prevent toxic dNTP accumulation [37]. Using RNA sequencing analyses, SAMHD1 was shown to be significantly up-regulated in psoriatic lesional skin compared to healthy skin (fold of change, 1.5) [38]. A total of six genes, including SH3 and cysteine-rich domain 2 (STAC2), aquaporin 5 (AQP5), family with sequence similarity 25 member C (FAM25C), ELOVL fatty acid Elongase 3 (ELOVL3), chromosome 10 open reading frame 99 (C10orf99), and aldo-keto reductase family 1 member B10 (AKR1B10), were increased with a change greater than 2.5-fold in resolved epidermis vs. never-lesional epidermis (Table 1). Two of these transcripts, C10orf99 and AKR1B10, were highly expressed in the lesional skin of psoriasis patients compared to healthy controls [39,40,41]. AKR1B10 is a human NADPH-dependent oxidoreductase [42]. The AKR1B10 enzyme plays a role in lipid metabolism and functions as a positive regulator of inflammation. Inhibition of AKR1B10 can suppress the inflammatory response triggered by various stressors in cellular and animal models [43,44,45,46,47]. AKR1B10 plays a role in psoriasis lesion formation by dysregulating the retinoic acid signaling pathway, thereby inducing the excessive proliferation of keratinocytes [41]. The gene C10orf99 encodes the protein GPR15L, a novel antimicrobial peptide with broad activity against various bacteria and fungi [48]. GPR15L is mainly expressed in epithelial tissues and functions as a homeostatic chemokine for recruitment of mouse dendritic epidermal T-cell into skin [49,50,51]. Because all 25 most down-regulated and up-regulated DEGs in resolved dermis vs. never-lesional dermis had a change greater than 2.5-fold, we decided to increase the limit to ≥10 to restrict the search to the most down-regulated and up-regulated DEGs (Table 1). The most down-regulated genes in resolved dermis vs. never-lesional dermis included small proline-rich protein 4 (SPRR4), ATPase H+/K+ transporting non-gastric alpha2 subunit (ATP12A), cystatin-E/M (CST6), Small proline-rich protein 1A (SPRR1A), cysteine-rich C-Terminal 1 (CRCT1), microseminoprotein β (MSMB), keratin 34 (KRT34), and gap junction protein beta 4 (GJB4). To the best of our knowledge, only MSMB gene down-regulation has been shown to be associated with pathogenesis of psoriasis. MSMB has been previously shown to be down-regulated in the lesional skin of psoriasis patients compared to non-lesional and healthy skin [40,52,53]. The most up-regulated genes in resolved dermis vs. never-lesional dermis were immunoglobulin heavy variable 3-7 (IGHV3-7), immunoglobulin heavy variable 3-33 (IGHV3-33), and immunoglobulin heavy joining 4 (IGHJ4) (Table 1). Interestingly, of the 25 most up-regulated DEGs in resolved dermis, 14 transcripts, accounting for 56%, were immunoglobulin coding gene segments. The high prevalence of B cell receptor gene segments in resolved vs. never-lesional dermis raises the possibility that B cells may play a role in psoriasis local relapse. However, its importance is unknown and further studies are needed to investigate the contribution of B cells to the local recurrence of psoriatic lesions. We performed gene ontology (GO) enrichment analysis for resolved vs. never-lesional DEGs. The results showed that 75 BPs, 32 MFs, and 7 CCs were significantly overrepresented in resolved vs. never-lesional epidermis (p < 0.05). In the GO analysis of the five most significant BPs, genes were mainly associated with the following terms: regulation of protein phosphorylation, regulation of phosphorylation, regulation of stress-activated MAPK cascade, positive regulation of stress-activated MAPK cascade and secondary heart field specification. In the five most significant MFs, genes were associated with the following terms: interleukin-6 receptor binding, death receptor activity, cysteine-type endopeptidase activity involved in the apoptotic process, positive regulation of telomerase activity, and regulation of cysteine-type endopeptidase activity. Most of these responses occurred in the early endosome lumen (Figure 4). The results in the dermal samples revealed that 37 BPs, 6 MFs, and 29 CCs were significantly overrepresented in resolved vs. never-lesional (p < 0.05). In the five most significant BPs, genes were associated with the following terms: detection of chemical stimulus, detection of stimulus involved in sensory perception, sensory perception of the chemical stimulus, G protein-coupled receptor activity, and detection of stimulus. In the most significant MF, genes were associated with transmembrane signaling receptor activity. These responses occurred mainly in the membrane-bound cell organelles (Figure S2). To determine the disease-residual transcriptomic profile (DRTP), we examined the overlap between the DEGs of resolved vs. never-lesional skin from our study and the DEGs of the lesional vs. healthy skin from the available datasets at GEO for both epidermal and dermal compartments. Since we did not find any meaningful overlap between DEGs of the resolved vs. never-lesional dermis and DEGs of the lesional vs. healthy dermis, we focus on our results for the epidermal samples below. Gene expression data from lesional and healthy epidermal skin samples were downloaded from the Gene Expression Omnibus (series matrix files of GSE68937 and GSE68923 datasets). A total of 4104 DEGs were identified in lesional epidermis vs. healthy epidermis. The psoriatic lesional-specific DEGs showed a 7-fold overrepresentation in resolved vs. never-lesional DEGs compared to non-DEGs of the same comparison (Fisher’s exact test p = 5 × 10−53). Overall, 102 genes overlapped between 476 DEGs of resolved vs. never-lesional and the 4104 DEGs of lesional vs. healthy in epidermal compartments. We found that 95% of these 102 overlapping genes had the same fold change direction in the two comparisons. The amount of change strongly correlated (Spearman’s rho: 0.75, p < 2.2 × 10−16); the data are shown in Figure 5. Of these 102 overlapping genes, 67 were down-regulated, and 35 were up-regulated in the resolved epidermis (Table S1). The five most down-regulated genes were CACNA2D1, ODF3L1, WNT2, LIF, and TPPP, whereas the five most up-regulated genes were AKR1B10, FABP5, PYDC1, WNT5A, and TMEM52 (gene abbreviations are explained in Table S1). Of note, the AKR1B10 gene was the top up-regulated overlapped gene between resolved and lesional epidermis. Combined transcriptomic analysis identified the AKR1B10 gene as the most differentially expressed gene in psoriatic lesions compared to healthy skin [41]. Microarray studies showed significantly higher expression of AKR1B10 (24-fold) in psoriatic lesions compared to healthy skin [39]. In addition, we determined the function of the overlapping genes through GO enrichment analysis. The results showed 54 significant BPs (p < 0.001 and FDR < 0.2). In the five most significant BPs, the genes were associated with the following terms: cellular response to stimulus, cellular response to transforming growth factor beta stimulus, response to transforming growth factor beta, positive regulation of the biological process, and Wnt signaling pathway involved in midbrain dopaminergic neuron differentiation (Table S2). Then, the protein–protein interaction (PPI) networks of the 102 overlapping genes were generated using the STRING database. Based on the results, the largest network (NGFRAP1-PAMR1-TRAF6-ZC3H12A-RC3H1-PAN3-TREX2) was most likely associated with TNF signaling, whereas a cluster of WNT-signaling was also evident (Figure 6). The KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis was also performed on the 102 overlapping genes, which revealed the enrichment of the WNT and mTOR signaling pathways. Surprisingly, the overlapping genes were most significantly enriched in the basal cell carcinoma pathway (Table 2). Interestingly, the signaling pathways associated with breast cancer, gastric cancer, and hepatocellular carcinoma were also significantly up-regulated in the resolved epidermis. Over the years, many studies have provided fundamental insights into the pathogenesis of psoriasis, and the knowledge has been translated into highly effective therapies. However, pathologic changes associated with lesion recurrence are only partially understood, making it difficult to develop practical strategies to prevent frustrating psoriasis flare-ups. A psoriasis relapse is defined by the appearance of new lesions and mainly by the recurrence of old lesions. The DRTP throughout psoriatic healed skin [14,15,16,17,18] and epidermal tissue-resident memory T cells (TRMs) [9,14,15,16] have been considered critical for the recurrence of old lesions. However, the role of resolved skin epidermal keratinocytes in disease recurrence is an unresolved issue. This is the first study to highlight the potential significance of epidermal keratinocytes in psoriasis local relapse by elucidating the DRTP and methylation/hydroxymethylation status in resolved epidermis. Our transcriptional results confirmed the distinction between transcriptomic profiles of never-lesional and resolved uninvolved skin in psoriasis. Among the DEGs in resolved epidermis compared with the never-lesional epidermis, the most down-regulated genes were functionally related to inflammation regulation (NEAT1-3) and genome integrity (SAMHD1). In addition, we found that genes involved in lipid metabolism (ELOVL3 and AKR1B10) and inflammation flare-up (C10orf99) were among the most up-regulated DEGs in resolved epidermis vs. never-lesional epidermis. Three of the most dysregulated genes, namely SAMHD1, C10orf99, and AKR1B10, were identified as likely to be important for lesion development in the resolved epidermis. We suggest these genes could confer a hypersensitivity of resolved epidermal keratinocytes to a secondary assault (Figure 7). Since it was previously shown that SAMHD1-deficient fibroblasts from patients with Aicardi Goutières syndrome (AGS), a rare inflammatory disease, had elevated dNTP pools associated with chronic DNA damage and up-regulation of interferon (IFN)-stimulated genes [54], it is possible that SAMHD1-deficient resolved epidermal keratinocytes also have elevated dNTP levels compared to never-lesional cells, which may lead to latent and persistent DNA damage and inflammation. Thus, resolved epidermal keratinocytes may be more susceptible to genotoxic stress than never-lesional cells. C10orf99 or GPR15L may act as a chemotactic ligand for GPR15 (G protein-coupled receptor 15). Transcriptomic studies and genomic analyses have shown that GPR15L is strongly up-regulated in the lesional skin of patients with psoriasis, atopic dermatitis, and in related animal models [26,40,55]. Single-cell RNA sequencing data showed that mainly epidermal keratinocytes express GPR15L in psoriatic lesional skin [56]. GPR15 is specifically expressed by effector/memory B cells, T cells, and regulatory T cells [50]. Whether the interaction between GPR15L and GPR15 plays a role in the initiation of psoriasis is unknown. However, higher expression of the chemotactic factor GPR15L in resolved epidermal keratinocytes may lead to retaining GPR15+ cells in resolved epidermis. The AKR1B10 gene encodes a retinaldehyde reductase that plays a critical role in the retinoic acid (RA) or vitamin A metabolism and its activity causes decreased RA synthesis. Overexpression of AKR1B10 has been reported in patients with psoriasis [39], atopic dermatitis [57] and keloids [58]. This suggests that an imbalance in retinoic acid (RA) metabolism is a common feature of these relapsing–remitting inflammatory skin diseases, regardless of their pathological background. The lower intracellular level of RA is considered an inducing signal for cell proliferation and prohibiting for cell differentiation, both of which promote tumorigenesis in affected tissues [59]. In resolved epidermal keratinocytes, significant overexpression of AKR1B10 may decrease the level of RA by suppressing the RA synthesis pathway and increase cell survival and proliferation by activating the retinal–retinol pathway. On the other hand, the lower expression of AKR1B10 gene in never-lesional epidermal keratinocytes may increase the level of RA and induce keratinocytes differentiation while decreasing cell survival. However, the abovementioned hypotheses need to be investigated and are open to future research on psoriasis local relapse. In addition, we defined the DRTP as a set of 102 expressed genes that overlapped between the DEGs of resolved vs. never-lesional and the DEGs of lesional vs. healthy skin. Remarkably, we found that the AKR1B10 transcript was not only among the most up-regulated DEGs in the resolved epidermis compared to the never-lesional epidermis, but was also the most up-regulated transcript overlapping between resolved and lesional epidermis. This finding suggests that the retinoic acid signaling pathway plays an essential role in the local recurrence of psoriatic lesions. Indeed, acitretin, a widely used drug for psoriasis, acts through the retinoic acid signaling pathway and affects keratinocytes proliferation and differentiation [60]. Since AKR1B10 is also a druggable target [61], the repurposing of the already known AKR1B10 inhibitors could be of translational importance. We also found that Wnt5a, a negative regulator of epidermal keratinocyte differentiation, remained strongly up-regulated in resolved epidermal keratinocytes. This is consistent with previous findings that recognized the Wnt5a gene as DRTP in psoriatic healed skin after successful etanercept therapy [14]. Our STRING and KEGG pathway analyses also revealed the up-regulation of genes involved in the Wnt pathway. Overall, these data support the hypothesis that the WNT signaling pathway may contribute to the recurrence of psoriatic lesions. Furthermore, TNF and mTOR signaling pathways, which play a role in the pathogenesis of psoriasis, were among the major disease-residual pathways identified in our STRING and KEGG pathway analyses. We also demonstrated a clear difference in the 5-mC and 5-hmc general pattern between psoriatic never-lesional and resolved, uninvolved skin. In resolved epidermis, in addition to 5-mC and TET3 mRNA expression levels compared with never lesional epidermis, we also found greatly decreased 5-hmC contents. Loss of 5-hmC has been reported in different solid tumors [62] and immune-mediated skin disorders, such as psoriasis [28]. Since 5-mC is the only substrate to produce 5-hmC in vivo [63], the loss of 5-hmC in resolved epidermal keratinocytes could result from a global decrease in 5-mC content. On the other hand, TET enzyme dysfunction or decreased levels of it may also result in reduced 5-hmC generation in cells. Hence, low 5-mC content and TET3 enzyme deficiency within the resolved epidermis possibly account for the loss of 5-hmC in resolved epidermal keratinocytes (Figure 8). Interestingly, perturbation of TET-5-hmC pathway in the epidermis of psoriatic lesions has been already reported. It has been suggested that this is related to the loss of the self-renewal capacity of basal keratinocytes that become FABP5-expressing transient amplifying cells (TACs) [28]. In our study, FABP5 remained up-regulated (2.09-fold) in the hypo-hydroxymethylated resolved epidermis (Table 1). These results and the disrupted Wnt5a signaling pathway in the resolved epidermis suggest that epidermal keratinocytes in healed/resolved skin harbor cell differentiation defects at the transcriptomic and epigenetic levels. As mentioned previously, we also observed that DNA was hypo-methylated in resolved epidermal keratinocytes compared to never-lesional cells. The hypo-methylated DNA profile, defined as a decreased 5-mC content of the genome, is a hallmark of several cancers [64] and autoimmune diseases [65,66,67,68,69]. It has been reported that auto-reactive T cells in systemic lupus erythematosus (SLE) [65,66], synovial fibroblasts in rheumatoid arthritis (RA) [67], and cells in the white matter of multiple sclerosis (MS) scars are globally hypo-methylated [68,69]. DNA methylation profiling in psoriasis patients showed intermediate methylation differences in psoriatic uninvolved skin compared to healthy and psoriatic lesional skin [26]. In summary, our findings suggest that disease-residual epigenomic and transcriptomic profiles are still present in resolved epidermal keratinocytes after successful therapy. Seven volunteer patients with moderate-to-severe plaque-type psoriasis, aged >25 years, were enrolled in our study. The disease severity was evaluated using the Psoriasis Area and Severity Index (PASI) scoring system. The PASI is a widely used and gold standard measurement tool that grades the severity of psoriatic lesions and assesses the treatment response of psoriasis patients [70]. This study recruited patients who matched the criteria irrespective of sex and therapeutic regimen. Our inclusion criteria were to have patients with moderate-to-severe (PASI above 15) plaque-type psoriasis before initiating systemic therapy and being on systemic therapy for at least 1 year before taking their skin samples. Psoriatic patients’ detailed information and experimental techniques applied to their samples are available in Table S3. Psoriatic tissue collection was obtained after written informed consent according to the rules of the Helsinki Declaration. Protocols for obtaining patient biopsies were approved by the Regional and Institutional Research Ethics Committee (PSO-CELL-01, 90/2021, 4969, 26 April 2021, Szeged, Hungary; HCEMM-001, 10/2020, 4702, 20 January 2020, Szeged, Hungary) for the protection of human subjects. Full-thickness 6 mm paired never-lesional and resolved skin punch biopsies (PBs) were taken under aseptic conditions with local anesthesia from psoriasis patients. At the time of sampling, their resolved skin had been resolved for at least 6 months. Both the clinician and the patient determined the resolved skin, and the lesional skin was obvious to detect. First, we set up the protocol for equilibrium binding of 5-mC and 5-hmC antibodies to the target antigens in frozen paraformaldehyde-fixed healthy skin sections (Appendix A). Shortly, never-lesional and resolved skin PBs were immediately fixed in a freshly made 4% paraformaldehyde (PFA 4%) solution in PBS (containing 0.01% TritonX-100) for 6 h on a rocker at room temperature (RT). Then the samples were washed once with PBS for 5 min. To minimize freezing-induced damages, the samples were infused gradually with sucrose by placing them in 10% and 20% sucrose in PBS (1 h each by rocking at RT) and 30% sucrose (overnight at 4 °C) [71]. The tissues were subsequently frozen in the cryogenic matrix (Thermo Fisher Scientific, Waltham, MA, USA) using liquid nitrogen, and 12 μm sections were cut and stored at −20 °C until further processing. The sections were washed 2 × 15 min in PBS to remove the infused sucrose and were incubated for 10 min in 0.1% Triton X-100 in PBS at RT. For antigen retrieval, sections were incubated for 1 h with freshly made 2 N hydrochloric acid (HCl) in PBS at RT. After DNA denaturation, sections were neutralized using 0.1 M Tris-HCl (pH 8.3) for 10 min. For blocking, the sections were incubated in blocking solution ((1% normal goat serum, 1% bovine serum albumin (Sigma-Aldrich, St. Louis, MO, USA) in PBS)) for 1 h, at RT in a humidified chamber. Samples were incubated with the following primary antibodies: polyclonal rabbit anti-human 5-hmC (1:1000, Active motif, Carlsbad, CA, USA, Cat No. 39769), mouse monoclonal anti-human 5-mC (1:500, Epigentek, East Farmingdale, NY, USA, Cat No. A-1014) for over-night at 4 °C. Purified mouse IgG1 κ isotype (Biolegend, San Diego, CA, USA, Cat No. 400102) was used for 5-mC isotypic control. As secondary antibodies, AlexaFluor 546-conjugated anti-rabbit IgG and AlexaFluor 647-conjugated anti-mouse IgG (Life Technologies, Carlsbad, CA) were used (both 1:500). We visualized the nuclei with DAPI (Sigma-Aldrich, St. Louis, MO, USA) staining. Olympus Fluoview FV1000 (Olympus Life Science Europa GmbH, Hamburg, Germany) confocal laser scanning microscope was used for fluorescence imaging following immunolocalization. Dry (10×, 20×) and oil immersion (PLANAPO 40× and 60×) objectives were used during imaging. The default DAPI and AlexaFluor dye combination setup of the microscope was used to capture images. Transmitted light detector of the microscopes were used for bright field images and merged onto DAPI images using Olympus Fluoview software. As regards Figure 3, identical microscope configuration is used for all intensity comparison images where ICA look up table of FIJI open software was employed after identical enhancing of brightness values of never-lesional and resolved pairs [72]. The skin PBs were washed in cold saline containing antibiotic/antimycotic (AB/AM) to remove the blood. Subsequently, the skin PBs were incubated in Dispase II solution (neutral protease, grade II, 2 U/mL, Roche Diagnostics, Basel, Switzerland) for 3 h at 37 °C and the epidermis was then carefully peeled from the dermis [73]. Following that, the epidermal and dermal compartments were immediately submerged in RNAlater (Invitrogen, Waltham, MA, USA, Cat No. AM7020) for storage until further processing for RNA extraction. Only the Ustekinumab-treated patients were included for full-length transcriptome sequencing (Table S3). Epidermal and dermal samples were mechanically homogenized using the Ultra Turrax T8 homogenizer (IKA-WERKE, Staufen, Germany). Total RNA was extracted using TRI-Reagent (Molecular Research Center; Cincinnati, OH, USA). The purity and concentration of RNA samples were determined via Nanodrop (Colibri Mikrovolüm Spektrometre, Berthold, Germany). The qualified RNA samples were used for cDNA synthesis using UltraScript cDNA Synthesis Kit (Thermofisher, Waltham, MA, USA). Changes in mRNA expression were detected with real-time RT-PCR using the FAM dye-labeled TaqManTM probes (Thermofisher, Waltham, MA, USA). TaqManTM probes used in our study are listed in Table S4. Real-time RT-PCR experiments were carried out using the qPCRBIO Probe Mix Lo-ROX (PCR Biosystem Ltd., London, UK) on a C1000 Touch Thermal Cycler (Bio-Rad Laboratories, Hercules, CA, USA). All reactions were duplicated, and the data were normalized to the 18S ribosomal RNA gene. Relative mRNA expression was calculated using the ΔΔCt method. Data from never-lesional and resolved epidermis were compared using an unpaired two-tailed t-test. Differences were considered significant when p < 0.05. A high-throughput mRNA sequencing analysis was performed on the Illumina sequencing platform to obtain global transcriptome data. Total RNA sample quality was checked on Agilent BioAnalyzer using the Eukaryotic Total RNA Nano Kit according to the manufacturer’s protocol. Samples with RNA integrity number (RIN) value > 7 were chosen for the library preparation process. RNA sequencing libraries were prepared from total RNA using an Ultra II RNA Sample Prep kit (New England BioLabs, Ipswich, MA, USA) according to the manufacturer’s protocol. Briefly, poly-A RNAs were captured by oligo-dT conjugated magnetic beads and the mRNAs were then eluted and fragmented at 94 °C. First-strand cDNA was generated through random priming reverse transcription, and double-stranded cDNA was developed after the second-strand synthesis step. After repairing ends, A-tailing, and adapter ligation steps, adapter-ligated fragments were amplified in enrichment PCR; finally, sequencing libraries were generated. Sequencing runs were executed on Illumina NextSeq 500 instrument using single-end 75-cycle sequencing. Raw sequencing data (fastq) were aligned to human reference genome version GRCh38 using the HISAT2 algorithm, and BAM files were generated. Downstream analysis was performed using StrandNGS software (www.strand-ngs.com, The access date was 8 May 2022). BAM files were imported into the software, and the DESeq algorithm was used for normalization. Moderated t-test was used to determine DEGs in resolved versus never-lesional skin. Genes with p < 0.05 were considered DEGs. Raw sequencing data is available in the NCBI under the BioProject ID: PRJNA938026 (https://www.ncbi.nlm.nih.gov/sra/PRJNA938026, accessed on 20 December 2022). Cytoscape v3.4 software with ClueGo v2.3.5. application was used for identifying over-represented gene ontology (GO) terms. A two-sided hypergeometric test with Benjamini–Hochberg FDR correction was performed using the list of DEGs and the GO biological process database. Gene expression data of psoriatic and healthy epidermis samples were downloaded from Gene Expression Omnibus (series matrix files of GSE68937 and GSE68923 datasets). GSE68937 and GSE68923 contained the microarray results from two models, including three lesional and two healthy epidermis samples in sets one and three lesional and three healthy epidermis samples in set 2 [74]. Differentially expressed genes between lesional and healthy epidermal samples were identified using Wilcoxon’s rank-sum test. p values were adjusted according to Benjamini and Hochberg [75], and genes with a false discovery rate lower than 0.1 (FDR < 0.1) were classified as DEGs. IDs were collated using data from the Human Gene Nomenclature Organization (HUGO) website. Fold change values were calculated by dividing the median gene expression values in psoriatic with the ones in healthy samples. The overlapping gene function was identified with a GO enrichment analysis using the GOrilla software [76,77]. We classified overlapping DEGs as target genes and all DEGs identified in resolved as background genes. Biological processes with p < 0.001 and FDR < 0.2 were considered significant. KEGG pathway enrichment analysis was carried out with WebGestalt online tool. The classification of genes was the same as for the study with GOrilla. Pathways with an FDR < 0.2 were considered significantly enriched. PPI networks were generated with STRING v11.5 using default parameters and at least medium confidence interactions. R program (version R 3.6.3) was used for statistical analyses. Statistical analyses and the graphs for the real-time RT-PCR results were performed using GraphPad Prism 8.0.2 (GraphPad Software, San Diego, CA, USA).

PMC10002498

Mei-Ling Cheng,Cheng-Hung Yang,Pei-Ting Wu,Yi-Chin Li,Hao-Wei Sun,Gigin Lin,Hung-Yao Ho

Malonyl-CoA Accumulation as a Compensatory Cytoprotective Mechanism in Cardiac Cells in Response to 7-Ketocholesterol-Induced Growth Retardation

23-02-2023

7-ketocholesterol,malonyl-CoA,mevalonic acid

The major oxidized product of cholesterol, 7-Ketocholesterol (7KCh), causes cellular oxidative damage. In the present study, we investigated the physiological responses of cardiomyocytes to 7KCh. A 7KCh treatment inhibited the growth of cardiac cells and their mitochondrial oxygen consumption. It was accompanied by a compensatory increase in mitochondrial mass and adaptive metabolic remodeling. The application of [U-13C] glucose labeling revealed an increased production of malonyl-CoA but a decreased formation of hydroxymethylglutaryl-coenzyme A (HMG-CoA) in the 7KCh-treated cells. The flux of the tricarboxylic acid (TCA) cycle decreased, while that of anaplerotic reaction increased, suggesting a net conversion of pyruvate to malonyl-CoA. The accumulation of malonyl-CoA inhibited the carnitine palmitoyltransferase-1 (CPT-1) activity, probably accounting for the 7-KCh-induced suppression of β-oxidation. We further examined the physiological roles of malonyl-CoA accumulation. Treatment with the inhibitor of malonyl-CoA decarboxylase, which increased the intracellular malonyl-CoA level, mitigated the growth inhibitory effect of 7KCh, whereas the treatment with the inhibitor of acetyl-CoA carboxylase, which reduced malonyl-CoA content, aggravated such a growth inhibitory effect. Knockout of malonyl-CoA decarboxylase gene (Mlycd−/−) alleviated the growth inhibitory effect of 7KCh. It was accompanied by improvement of the mitochondrial functions. These findings suggest that the formation of malonyl-CoA may represent a compensatory cytoprotective mechanism to sustain the growth of 7KCh-treated cells.

Malonyl-CoA Accumulation as a Compensatory Cytoprotective Mechanism in Cardiac Cells in Response to 7-Ketocholesterol-Induced Growth Retardation The major oxidized product of cholesterol, 7-Ketocholesterol (7KCh), causes cellular oxidative damage. In the present study, we investigated the physiological responses of cardiomyocytes to 7KCh. A 7KCh treatment inhibited the growth of cardiac cells and their mitochondrial oxygen consumption. It was accompanied by a compensatory increase in mitochondrial mass and adaptive metabolic remodeling. The application of [U-13C] glucose labeling revealed an increased production of malonyl-CoA but a decreased formation of hydroxymethylglutaryl-coenzyme A (HMG-CoA) in the 7KCh-treated cells. The flux of the tricarboxylic acid (TCA) cycle decreased, while that of anaplerotic reaction increased, suggesting a net conversion of pyruvate to malonyl-CoA. The accumulation of malonyl-CoA inhibited the carnitine palmitoyltransferase-1 (CPT-1) activity, probably accounting for the 7-KCh-induced suppression of β-oxidation. We further examined the physiological roles of malonyl-CoA accumulation. Treatment with the inhibitor of malonyl-CoA decarboxylase, which increased the intracellular malonyl-CoA level, mitigated the growth inhibitory effect of 7KCh, whereas the treatment with the inhibitor of acetyl-CoA carboxylase, which reduced malonyl-CoA content, aggravated such a growth inhibitory effect. Knockout of malonyl-CoA decarboxylase gene (Mlycd−/−) alleviated the growth inhibitory effect of 7KCh. It was accompanied by improvement of the mitochondrial functions. These findings suggest that the formation of malonyl-CoA may represent a compensatory cytoprotective mechanism to sustain the growth of 7KCh-treated cells. High levels of 7-ketocholesterol (7KCh), an oxysterol derived from the oxidation of cholesterol (Chol), is detected in the vascular plaques of atherosclerosis patients and in the plasma of those at high risk of cardiovascular diseases [1]. It can be catabolized via the intra- and extrahepatic pathways [2,3,4]. The latter pathway involves sterol O-acyltransferase-mediated acylation and the reverse transport of the esterified form to high-density lipoprotein (HDL) [4]. A reduction of the sterol O-acyltransferase expression in heart leads to 7KCh accumulation and tissue damage. Highly elevated 7KCh levels were found in the red blood cells (RBCs) of heart failure patients [5]. It is envisaged that 7KCh can be an important risk factor for cardiovascular diseases and heart failure. Heart tissue utilizes a number of energy sources for maintaining its continual contraction. The preference for fuel sources varies with developmental stages, shifting from glucose and lactate for fetal hearts to fatty acids for adult hearts. The remodeling of energy metabolism occurs in response to stressful conditions. Different physiological and pathophysiological conditions, including physical exercise, hypoxia, hypertrophy, and heart failure, causes the cardiac cells to switch to glucose for ATP production [6]. Altered energy metabolism in cardiac tissues is postulated to contribute to the pathogenesis of cardiovascular diseases such as cardiomyopathy and heart failure [7,8]. Exposure to 7KCh has an impact on redox homeostasis and cellular metabolism. It induces reactive oxygen species (ROS) generation in the endothelial cells and cardiomyocytes [5,9,10] and inflicts cellular damage. The other oxidized products of cholesterol cause ROS production, which is implicated in cognitive impairment and the development of cataracts [11,12]. Mitochondria represent an important endogenous source of ROS [13], which can damage the mitochondria in a reciprocal manner. Oxysterols and Chol accumulate in the mitochondria in the cardiac tissues of the animal models subject to ischemia reperfusion [14,15]. It is associated with anomalous changes in the mitochondria, such as the peroxidation of membrane lipids and the loss of the mitochondrial membrane potential [15]. Given the central roles of mitochondria in the metabolism, mitochondrial dysfunction is accompanied by the reprogramming of the metabolism in cardiomyocytes. We have recently reported that a 7KCh treatment induced changes in the cholesterol and lipid metabolism of HL-1 cells [16]. The genes involved in mevalonic acid biosynthesis and the metabolism of fatty acids, triacylglycerides and ketone bodies were down-regulated, while those involved in cholesterol transport and esterification were up-regulated [16]. Intriguingly, 7KCh enhanced the transcription of the genes regulated by ATF4, which has been recently identified as a key regulator of mitochondrial stress [17]. These findings have prompted us to study whether 7KCh induces mitochondrial stress and reprogramming of the cellular metabolism, in particular, energy metabolism. In the present study, we demonstrate that 7KCh exposure induces the reprogramming of energy metabolism and mitochondrial dysfunction in cardiomyocytes and inhibits their growth. It is associated with a compensatory increase in the mitochondrial mass and malonyl-CoA accumulation. Malonyl-CoA suppresses carnitine palmitoyltransferase-1 (CPT-1) activity and β-oxidation. Studies involving the use of pharmaceutical inhibitors and Mlycd−/− knockout cells indicated that an increase in malonyl-CoA reverses the 7KCh-induced mitochondrial defects and growth inhibition. Decreases in the plasma HDL levels and increases in 7KCh in RBCs correlated with the prevalence of cardiovascular diseases and heart failure [5], implying that 7KCh may affect the physiology of cardiomyocytes. The treatment of cardiomyocytes resulted in the cellular uptake of 7KCh. The immunofluorescence assay with an anti-7KCh antibody was used to detect the localization of 7KCh in the HL-1 cells. The fluorescence of the antibody-stained 7KCh was significantly elevated in the 7KCh-treated HL-1 cells (Figure S1; lower left panel) compared with that of untreated cells (Figure S1; upper left panel), suggesting the uptake and intracellular accumulation of 7KCh. We examined the effect of 7KCh on the growth of cardiac cell lines HL-1 and AC16. The HL-1 and AC16 cells were treated without or with 10 or 20 μM 7KCh. The concentrations used were in the physiological range of 7KCh. The erythrocytic 7KCh levels in the healthy volunteers were between 1 and 2 μM [5,18], and the blood 7KCh levels in the heart failure patients were at least 10-to 20-fold higher than those of the normal controls [5]. 7KCh inhibited growth of HL-1 and AC16 cells. The numbers of HL-1 and AC16 cells that were treated with 20 μM 7KCh were about 15% lower those of the untreated cells (Figure 1A,B). Our previous study showed that 7KCh induces oxidative stress in cardiomyocytes [5]. Consistent with this, the HL-1 or AC16 cells were treated with 7KCh, stained with MitoSOX Red, and analyzed cytometrically. The fluorescence of MitoSOX Red was significantly elevated in the treated cells (Figure 2A,J), indicating that 7KCh may cause mitochondrial ROS generation. It is possible that 7KCh may induce the functional alteration of the mitochondria. The effect of 7KCh on the mitochondrial membrane potential ΔΨm was studied. The 7KCh-treated cells were stained with JC-1 dye and analyzed by flow cytometry. The mean fluorescence intensities (MFI) of channels FL2 and FL1 were measured, and their ratio (i.e., FL2 MFI/FL1 MFI), an indicator of ΔΨm, was calculated. The treatment of the HL-1 and AC16 cells with 20 μM 7KCh resulted in 40% decreases in ΔΨm (Figure 2B,K). To further delineate how 7KCH affects the mitochondrial respiration of cardiomyocytes, we resorted to mitochondrial respirometry to study such changes. The oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) of the 7KCh-treated cells were assessed in real-time using a Seahorse XF24 Extracellular Flux Analyzer. The HL-1 cells that were treated without or with 10 or 20 μM 7KCh for 24 h had a reduced basal OCR, but an increased ECAR (Figure 2C–E). Such a shift in energy metabolism toward glycolysis was also observed in the AC16 cells treated with 20 μM 7KCh (Figure 2L–N). The use of respiration inhibitors (oligomycin, FCCP, rotenone, and antimycin A) allowed us to define the OCRs due to maximum respiration, the spare respiratory capacity, and proton leak. Both the maximum respiration and spare respiratory capacity of the HL-1 and AC16 cells diminished in response to 7KCh (Figure 2F,G,O,P). The proton leak was substantially reduced in the 7KCh-treated HL-1 and AC16 cells (Figure 2H,Q). Consistent with the reduction of respiratory function, the ATP levels in the HL-1 and AC16 cells treated with 20 μM 7KCh were about 30% and 35% lower than that of the control cells, respectively (Figure 2I,R). These findings suggest that 7KCh inhibits electron transport and oxidative phosphorylation and induces ROS generation. The deteriorative changes in respiratory function are accompanied by an increase in the mitochondrial mass. As shown in Figure 3A,B, the porin level increased with the 7KCh concentration. In addition, the mitochondrial mass was evaluated by Mitotracker Green staining and cytometric analysis. The mean fluorescence intensity (MFI) of the stained HL-1 cells that had been treated with 20 μM 7KCh was 65% higher than that of the control cells (Figure 3C). These findings suggest that 7KCh causes mitochondrial biogenesis. Consistent with this, the levels of various mitochondrial respiratory chain proteins increased in the 7KCh-treated cells (Figure 3D). The nuclear gene-encoded complex I protein NADH:ubiquinone oxidoreductase subunit B8 (NDUFB8), complex II protein succinate dehydrogenase complex iron sulfur subunit B (SDHB), complex III protein ubiquinol-cytochrome C reductase core protein 2 (UQCRC2), and ATP Synthase F1 Subunit α (ATP5F1A), as well as mitochondrial gene-encoded complex IV protein cytochrome C oxidase I (MTCO1), increased in their expression, albeit to different extents, in the 7KCh-treated cells (Figure 3E). These findings suggest that 7KCh induces compensatory mitochondrial biogenesis in cardiac cells. We applied a metabolomic approach for studying the metabolic reprogramming associated with the functional changes in the mitochondria. The metabolites involved in pathways such as glycolysis, pentose phosphate, and the citric acid cycle were analyzed by the liquid-chromatography coupled with tandem mass spectrometry (LC-MS/MS). A twenty-four hour treatment of HL-1 cells with 20 μM 7KCH caused a concentration-dependent change in the metabolome (Figure 4A). Increased levels of glucose-6-phosphate, fructose-1,6-bisphosphate (FBP), and lactate and decreased levels of 2-phosphoglycerate and 2,3-bisphosphoglycerate indicate an increase in the glycolytic rate (Figure 4B,D). Notably, malonyl-CoA accumulated at a high level in these cells. In contrast, the 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) level decreased in the 7KCh-treated cells compared with that of the untreated cells. The accumulation of malonyl-CoA and the reduction of HMG-CoA formation were also observed in the 7KCh-treated AC16 cells (Figure 4E). We further examined the metabolism of malonyl-CoA using [U-13C] glucose labeling and tracking. The HL-1 cells were treated with 20 μM 7KCh and labeled with [U-13C] glucose. The relative abundances of the isotopologues of various metabolites were analyzed. As shown in Figure 4D,E, the M and M+2 isotopologues of malonyl-CoA increased, while the M, M+2, M+4, and M+6 isotopologues of HMG-CoA decreased. Interestingly, the total levels of the TCA cycle intermediates, citrate, succinyl-CoA, and oxaloacetate (OAA), remained unchanged or slightly increased in the cells treated with 10 μM 7KCh. However, these metabolites declined substantially in abundance upon exposure to the treatment with 20 μM 7KCh. An analysis of the isotopologues of the metabolites revealed that in addition to the M+2 and M+4 isotopologues formed from TCA cycle, the M+3 isotopologue of oxaloacetate generated in the anaplerotic reactions (Figure 4D). The proportions of different isotopologues of citrate were nearly unchanged in the cells treated with 20 μM 7KCh. In contrast, the proportions of isotopologues other than that of the M+2 succinyl-CoA decreased substantially. These findings suggest that the flux of the conversion of citrate to the downstream TCA intermediates is inhibited by 7KCh. The formation of malonyl-CoA has its functional implication. Malonyl-CoA is an inhibitor of CPT-1 involved in the fatty acid uptake into the mitochondria [19,20]. We isolated the mitochondria from the HL-1 cells treated without 7KCh and assayed the CPT-1 activity. The expression levels of CPT-1 were similar in the mitochondrial preparations to those of the control and treated cells (Figure 5A). The CPT-1 activity level was significantly lower in the mitochondria from the 7KCh-treated cells than it was in those from the control cells (Figure 5B). As control, malonyl-CoA directly inhibited the CPT-1 activity of the mitochondria (Figure 5C). These findings suggest that 7KCh inhibits the CPT-1 activity, and probably, β-oxidation in the cardiomyocytes. To study the possibility that 7KCh inhibits the cardiomyocytic β-oxidation, we studied the fatty acid oxidation (FAO) in the HL-1 cells. The HL-1 cells were treated with or without 7KCh, and the palmitate (PA)-stimulated (or vehicle-stimulated) oxygen consumption was measured using the Seahorse XF24 Extracellular Flux Analyzer. The PA-stimulated oxygen consumption represents the reliance of the cells on β-oxidation for energy. 7KCh inhibited the basal respiration, maximum respiration, and spare respiratory capacity in the cells supplied with palmitate as an energy source (Figure 5D). In contrast, 7KCh did not significantly alter these respiration parameters in the cells the treated with the vehicle. The effect of 7KCh on PA-stimulated oxygen consumption is not attributable to expression of different CPT-1 isoforms. The transcripts of Cpt1a, Cpt1b, and Cpt1c genes, as quantified by RT-qPCR, did not differ between the 7KCh-treated and control cells (Figure 5E). These findings suggest that 7KCh may suppress the cardiomyocytic β-oxidation through the malonyl-CoA-mediated inhibition of CPT-1 activity (Figure 5A). The reduction of the HMG-CoA level in the 7KCh-treated HL-1 cells suggests that the inhibition of the MVA pathways may retard the growth of cardiac cells. To study this possibility, we treated the HL-1 cells with 0.25, 1, and 10 μM lovastatin, in addition to 20 μM 7KCh, and determined the cell number. Lovastatin is an inhibitor of enzyme 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) (Figure 6A) [21]. The treatment with 1 μM lovastatin alone caused a three-fold increase in the HMG-CoA level, and a 40% decrease in the cell number (Figure 6B). The co-treatment with 7KCh and lovastatin reduced the HMG-CoA level by 94%. It is consistent with our previous finding that 7KCh represses the expression of Acat2 and Hmgcs1 genes [16], the products of which act upstream of HMGCR in the MVA pathway. Such a co-treatment caused a 54% decrease in the cell number (Figure 6B). These findings suggest that the products of the MVA pathway, such as coenzyme Q (CoQ), may play essential roles in growth of cardiomyocytes. It is in agreement with a decline in the CoQ10 level in the 7KCh-treated cells (Figure S2). To study the physiological roles of malonyl-CoA accumulation in the 7KCh-treated cells, we studied the effect of the inhibitors of malonyl-CoA decarboxylase and acetyl-CoA carboxylase on cell growth (Figure 6A). Malonyl-CoA decarboxylase (MLYCD) catalyzes the degradation of malonyl-CoA to acetyl-CoA; acetyl-CoA carboxylase catalyzes the carboxylation of acetyl-CoA to form malonyl-CoA. The HL-1 cells were co-treated with 20 μM 7KCh and 0.5, 2.5, or 10 μM malonyl-CoA decarboxylase inhibitor CBM 301940, and the cell number was determined. The CBM 301940 treatment, which significantly increased the intracellular ratio of malonyl-CoA level to acetyl-CoA, alleviated the growth retardation effect of 7KCh (Figure 6C,F). For the study with acetyl-CoA carboxylase inhibitor, the HL-1 cells were co-treated with 25,100,250 nM ND 646, and the cell number was determined. The ND 646 treatment, which suppressed the intracellular formation of malonyl-CoA, acted synergistically with 7KCh to inhibit cell growth (Figure 6D,G). These findings suggest that the formation of malonyl-CoA is beneficial to the growth of cardiomyocytes and may represent a compensatory cytoprotective response to 7KCh. To study if malonyl-Co-A accumulation in the cardiomyocytes offers a cytoprotective effect, we derived Mlycd−/− cells from AC16 cells using the CRISPR/Cas9 technology. The expression of the MLYCD protein was nullified in the Mlycd−/− cells (Figure 7A). Mlycd knockout mitigated the growth inhibitory effect of 7KCh (Figure 7B). It was associated with the restoration of ΔΨm and decreased mitochondrial ROS production (Figure 7C,D). The knockout of the Mlycd gene maintained the steady state level of malonyl-CoA, while it reduced that of acetyl-CoA (Figure 7F,G). This increased the malonyl-CoA/acetyl-CoA ratio by about 70%. The malonyl-CoA/acetyl-CoA ratio was elevated four-fold in the 7KCh-treated Mlycd−/− cells as compared to that of the treated AC16 cells (Figure 7F,G). These findings validate that an increase in malonyl-CoA production can relieve the cardiomyocytes from the 7KCh-induced growth inhibition and mitochondrial dysfunction. We studied the effect of 7KCh on the expression of the genes encoding acetyl-CoA carboxylase (ACAC) and acyl-CoA synthetase family member 3 (ACSF3). The level of the acetyl-CoA carboxylase gene transcript, as measured by RT-qPCR using ACAC universal primers, was elevated in the 7KCh-treated HL-1 cells (Figure 8A). There are two ACAC isoforms, namely acetyl-CoA carboxylase α (ACACA) and β (ACACB), in mammals [22]. The Acacb transcript level increased, while that of the Acaca transcript declined (Figure 8B). In addition, there was a significant increase in the Acsf3 transcript level (Figure 8D). In contrast, the Mlycd transcript level remained nearly the same after the 7KCh treatment. These findings suggest that 7KCh enhances the expression of Acacb and Acsf3 genes. It is possible that the 7KCh-induced disruption of normal energy metabolism and metabolic reprogramming in the cardiomyocytes may play important pathogenic roles in cardiovascular diseases. The way in which 7KCh induces metabolic changes in cardiomyocytes remains elusive. In the present work, we demonstrate that 7KCh causes mitochondrial dysfunction, ROS production, and growth retardation. Malonyl-CoA accrues in the cardiomyocytes in response to 7KCh and inhibits CPT-1 activity and β-oxidation. The studies involving pharmacological inhibitors and Mlycd-knockout cardiomyocytes indicate that an experimental rise in the malonyl-CoA level restores the mitochondrial ΔΨm and redox homeostasis and reverses the growth inhibitory effect of 7KCh. It is speculative whether 7KCh directly affects the cardiomyocytic mitochondrial functions. The treatment of the translocator protein ligand 4′-chlorodiazepam suppressed the formation of oxysterols and partially restored the mitochondrial respiration [15], implying an adverse effect of 7KCh on the mitochondria. 7KCh damages the mitochondrial DNA and causes a loss of ΔΨm in retinal pigment cells [23]. We found that the 7KCh treatment resulted in reduction of ΔΨm in the HL-1 and AC16 cells, which were treated with 20 μM KCh (Figure 2B,K). It was associated with the decreases in basal respiration, maximum respiration, as well as spare respiratory capacity in these cells (Figure 2E,F,G,N,O,P). Both the flux of mitochondrial metabolism and the efficiency of respiratory processes, but not the expression of mitochondrial proteins, are adversely affected by 7KCh. The expression of the typical respiratory complex proteins actually increased after the 7KCh treatment (Figure 3D,E). The flux of the TCA cycle decreases in the treated cells (Figure 4), contributing to the decline in oxygen consumption. The efficiency of electron transport is lowered by 7KCh, which is indicated by an increase in ROS generation in the treated cells (Figure 2A,J). The proper assembly of the respiratory supercomplex is essential to efficient electron transport. It has been found that the basal and maximum respiration and the spare respiratory capacity are reduced in the PHB-deficient cells, with impaired organization of the respiratory supercomplexes [24,25]. Consistent with such a notion, our previous study has shown that the Phb1, Phb2, Higd1a, and Higd2a transcript levels declined in the 7KCh-treated cells [16]. Prohibin 1 (PHB1) and 2 (PHB2) participate in the assembly of respiratory supercomplexes and the maintenance of their stability [26]. HIGD1A and HIGD2A are involved in the dynamic assembly of complexes III and IV and in supercomplex formation [27]. Moreover, changes in the anabolic pathways may hamper the electron transport. The reduction of the flux of the MVA pathway dwindled the supply of HMG-CoA (Figure 4B,D,E), which is the precursor of CoQ biosynthesis. As CoQs serve as an important electron carrier, their decrease is likely to impair the normal functioning of the electron transport chain (Figure S2). The 7KCh-induced changes in the mitochondrial functions correlate with the reprogramming of cardiomyocytic energy metabolism. Increases in the levels of glucose-6-phosphate, fructose-1,6-bisphosphate, and lactate (Figure 4A), together with an increase in ECAR (Figure 2), indicate an increase in the glycolytic rate in the 7KCh-treated cardiac cells. Substantial decreases in TCA intermediates, such as citrate, succinyl-CoA, and oxaloacetate, in the cells treated with 20 μM 7KCh are suggestive of a reduction of the TCA cycle flux in these cells (Figure 4D). The isotopologue analysis reveals that the proportions of the M+2 and M+4 isotopologues of succinyl-CoA were similar to those of the corresponding isotopologues of citrate in the control cells. However, the levels of the M+2 and M+4 isotopologues of succinyl-CoA were substantially reduced by the 7KCh treatment. It is likely that the flux of the onward reaction of citrate in TCA cycle decreased upon 7KCh treatment. It also implies that citrate is probably involved in to reactions such as that catalyzed by citrate lyase. Interestingly, most of succinyl-CoA molecules remained unlabeled. These molecules formed from the unlabeled α-ketoglutarate, which may be derived from glutamate in a glutamate dehydrogenase-catalyzed reaction. Additionally, anaplerotic reactions made significant contribution to the oxaloacetate pool in the control and 7KCh-treated cells. Overall, pyruvate derived from glycolysis is converted either to lactate or to oxaloacetate (through anaplerotic reactions). Citrate synthase converts oxaloacetate to citrate, which can be metabolized in the TCA cycle or converted to acetyl-CoA for malonyl-CoA synthesis. In the 7KCh-treated cells, glycolysis is enhanced, while the TCA cycle is inhibited. More citrate molecules contribute to malonyl-CoA synthesis. Malonyl-CoA itself is an inhibitor of CPT-1, which is involved in the fatty acid uptake into the mitochondria and β-oxidation (Figure 5C) [19]. The accumulation of malonyl-CoA in the 7KCh-treated cells reduced the use of fatty acids as fuel molecules for β-oxidation. Indeed, 7KCh inhibited the CPT-1 activity (Figure 5B) and palmitate-stimulated respiration (Figure 5D). It is evident that malonyl-CoA is cardioprotective. The pharmacological inhibition of MLYCD activity and the knockdown of the Mlycd gene improve the biomechanical functions of the post-ischemic heart [28,29]. The 7KCh-induced malonyl-CoA accumulation probably represents a compensatory mechanism to reduce the toxic effect of 7KCh. The pharmacological treatment with CBM 301940, which increased the intracellular malonyl-CoA level, lessened the growth inhibitory effect of 7KCh (Figure 6C,F). The growth of Mlycd−/− cardiac cells were inhibited to a much lesser extent than the control cells were after 7KCh exposure (Figure 7B). Such a cytoprotective effect was associated with the restoration of ΔΨm and a decrease in mitochondrial ROS generation, suggesting an improvement of the mitochondrial functions. These findings advocate that metabolic reprogramming occurs in the 7KCh-treated cardiomyocytes to compensate for mitochondrial dysfunction and growth defects. An additional point about the malonyl CoA-inhibited β-oxidation is noteworthy. Our previous study has revealed that fatty acids can be released from phosphatidylcholines by phospholipase A2, and are used for esterification [16]. The fatty acid molecules are spared from β-oxidation through the action of malonyl-CoA. Malonyl-CoA appears not to serve as a precursor of fatty acid synthesis. The expression of the genes involved in fatty acid synthesis, such as fatty acid synthase and desaturases, are down-regulated [16]. The shift in energy metabolism from oxidative phosphorylation to glycolysis is physiologically relevant. Fatty acids are the major fuel for healthy adult hearts, responsible for 60–80% of ATP generation [30]. The remaining ATP molecules are derived from the metabolism of glucose and lactate. A change in fuel preference and energy metabolism may occur under different physiological or pathophysiological conditions. For instance, an increased glycolysis level was observed in the cardiac tissues isolated from the animal models of left ventricular hypertrophy [31,32,33]. The uptake of 2-deoxy-2-[18F] fluoro-D-glucose increased in the right ventricles of pulmonary arterial hypertension patients, and the uptake value correlated with the disease severity scores [34]. Additionally, it has been recently shown that myocardial infarction or pressure loading induces the cycling of specialized cardiomyocytes that may be involved in cardiac repair [35]. The activation of glycolysis is apparently essential to cardiomyocytic proliferation after an injury [36,37]. It is possible that the remodeling of energy metabolism in cardiomyocytes may affect cellular proliferation and tissue regeneration under certain circumstances, for example, the presence of high plasma 7KCh levels [38]. It is not unprecedented that the compensatory mitochondrial biogenesis occurs in response to mitochondrial dysfunction. Knockout of the leucine-rich pentatricopeptide repeat containing protein (LRPPRC)-encoding gene, whose mutations have been identified in Leigh syndrome, causes the defective assembly of the electron transport chain and triggers compensatory mitochondrial biogenesis [39]. Compensatory mitochondrial biogenesis occurs in cells with mitochondrial DNA mutations [40]. The carriers of LHON mutations display increases in the mitochondrial mass [41]. The declines in mitochondrial functions and ATP supply probably induce the expression of respiratory complex proteins and mitochondrial biogenesis in the 7KCh-treated cells (Figure 3). This is consistent with increases in the transcripts of the Tfam, Tfb1m, Tfb2m, and Pprc1 genes [16], which are involved in mitochondrial transcription, ribosome assembly, and biogenesis [42,43]. The transcription of the genes encoding the malonyl-CoA-metabolizing enzymes is differentially modulated by 7KCh. The Acacb transcription is up-regulated at the expense of the transcription of the Acaca gene (Figure 8B,C), resulting in an overall increase in the ACAC-encoding transcript (Figure 8A). The ACACB protein, located at the outer mitochondrial membrane, promotes the formation of malonyl-CoA, which allosterically inhibits CPT-1 activity [22,44]. Such a change in the expression of ACAC isoforms in the 7KCh-treated cells is congruous with the inhibition of β-oxidation. ACSF3 is involved in the detoxification of malonate, which is a competitive inhibitor of succinate dehydrogenase [45]. The ACSF3-derived malonyl-CoA is used in the malonylation of proteins, which may be involved in the post-translational regulation of mitochondrial proteins and metabolism [45]. The knockout of the Acsf3 gene in HEK 293T cells was found to alter their metabolism [46]. The enhanced transcription of the Acsf3 gene in the 7KCh-treated cells may imply the role of protein malonylation in the observed metabolic changes. The present study has limitations, such as cardiac cell lines were used and more work with animal models is needed to fully understand the cytoprotective mechanism. The results might be subject to technical variation caused by the methods used (e.g., cell counting). Unless otherwise stated, all the chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA). We dissolved 7-Ketocholesterol (7KCh; Sigma-Aldrich) in dimethyl sulfoxide (DMSO). In most of the experiments, 7KCh was used at a concentration of 10 or 20 μM. Lovastatin (Caymen Chemical, Ann Arbor, MI, USA), CBM 301940 (Torcis Bioscience, Bio-Techne, Minneapolis, MN, USA), and ND 646 (Caymen Chemical) were dissolved in DMSO. Lovastatin was used at the concentration range 0.25–10 μM; CBM 301940 was used at the concentration range 0.5–10 μM; ND 646 was used at the concentration range 25–250 nM. HL-1 atrial myocytes (Research Resource Identifier (RRID): CVCL_0303) were cultured in the fibronectin-gelatin-coated flasks containing the Claycomb medium (51800C, Sigma-Aldrich), which was supplemented with 10% HL-1 qualified fetal bovine serum (FBS; TMS-016, Sigma-Aldrich), 100 U/mL penicillin, 100 μg/mL streptomycin, 2 mM L-glutamine, and 0.1 mM norepinephrine in a humidified atmosphere of 5% CO2 at 37 °C, as previously described [47]. The AC16 human cardiomyocyte cell line (RRID: CVCL_4U18; EMD Millipore Corp., Temecula, CA, USA) was cultured in Dulbecco’s modified Eagle medium/nutrient mixture F-12 medium (DMEM/F-12; Thermo Fisher Scientific, Waltham, MA, USA) containing 12.5% FBS according to manufacturer’s instructions. The malonyl-CoA decarboxylase (Mlycd) gene knockout AC16 cells (i.e., Mlycd−/− cells) were generated using the CRISPR/Cas9 knockout service provided by the RNA Technology Platform and Gene Manipulation Core (National Core Facility for Biopharmaceuticals, Taipei, Taiwan). For determination of the growth curves of the 7KCh-treated cells, 5 × 104 cells were seeded in 12-well culture plate and treated with the indicated 7KCh concentrations (i.e., the concentrations indicated in the legends of the respective figures) for various periods. Cardiac cells were fixed in 3.7% formaldehyde for 10 min, and then stained with 5 μg/mL of Hoechst 33342 for 15 min. The cell number was determined using IN Cell Analyzer 1000 (GE Healthcare Life Sciences, Chicago, IL, USA) [48]. As a control for the 7KCh treatment, the cells were treated with the dimethyl sulfoxide (DMSO) vehicle. DMSO also served as the vehicle for lovastatin, CBM 301940, and ND 646. The measurement of the OCR and ECAR was performed using Seahorse XF24-3 Analyzer (Agilent Technologies, Santa Clara, CA, USA) as previously described [49]. In brief, 8 × 103 cells were seeded per well in an XF24 cell culture microplate and maintained in DMEM (Thermo Fisher Scientific). They were treated with the indicated concentrations of 7KCh. After 24 h, the medium was replaced with DMEM without sodium bicarbonate. The XF24 assay cartridge was prepared, loaded with 10 μM oligomycin, 3 μM FCCP, 10 μM antimycin A, and rotenone in the injection ports, and calibrated according to the manufacturer’s instruction. The oxygen consumption of the cells was measured using Seahorse XF-24 analyzer (Agilent Technologies) under basal condition and after the sequential injection of oligomycin (32 min), carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP) (56 min), and antimycin A/rotenone (88 min). The proton flux in the medium was also monitored. The OCR and ECAR were automatically calculated. Superoxide anion production was determined by MitoSOX Red staining and flow cytometry as previously described [50]. The mitochondrial mass and mitochondrial membrane potential were determined, respectively, by staining with MitoTracker Green and JC-1 and performing cytometric analyses as previously described [51,52]. BD LSR II flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA) was used for the analyses. At 24 h before the experiment, 8 × 103 cells were cultured per well in an XF-24 plate. The cells were treated with the indicated concentrations of 7KCh for 24 h. The medium was replaced with 375 μL of FAO Assay Medium (111 mM NaCl, 4.7 mM KCl, 1.25 mM CaCl2, 2 mM MgSO4, 1.2 mM NaH2PO4 supplemented with 2.5 mM glucose, 0.5 mM carnitine, and 5 mM HEPES, pH 7.4) and incubated at 37 °C for 45 min. Before the analysis, 87.5 μL of 1 mM palmitate–1% BSA conjugate or 1% BSA were added to each well. During the experiment, respiration was measured under basal condition and after the sequential injection of oligomycin, FCCP, and antimycin A/rotenone as described in Section 4.3.1. The metabolites were extracted as previously described [53,54]. Briefly, the cells were treated with the indicated concentrations of 7KCh for 24 h, and then washed twice with cold phosphate-buffered saline (PBS). The metabolites were extracted with 80% (v/v) MeOH/H2O pre-equilibrated at 80 °C. The extract was collected into 1.5 mL tubes, vortexed for 5 min, and centrifuged at 12,000× g for 30 min. The resulting supernatant was dried using a centrifugal evaporator under reduced pressure. The samples were re-suspended in 200 μL of 1% acetic acid for mass spectrometric analysis, and then were analyzed using the Xevo TQ-XS Triple Quadrupole Mass Spectrometry System (Waters Corp., Milford, MA, USA) [16]. The chromatographic separation was achieved on a BEH C18 (100 × 2.1 mm, particle size of 1.7 µm; Waters Corp.) at 45 °C. The mobile phase consisted of eluent A (10 mM tributylamine (TBA)/15 mM acetic acid) and eluent B (10 mM TBA/15 mM acetic acid/50% ACN). The flow rate was set at 0.3 mL/min. The elution profile was as follows: 4% B, 6 min; linear gradient 4–50% B, 0.1 min; 50–60% B, 2.9 min; 60–100% B, 0.8 min, and 100% B for an additional 2.2 min. The mass spectrometer was operated in negative ion mode at an ESI voltage of 3 kV. The metabolites were analyzed using MassLynx software (v4.1; Waters Corp., Milford, MA, USA). Stable isotope-labeling was performed as previously described with slight modifications [55]. The cells were treated with or without 7KCh for 24 h, and then incubated in the medium containing 0.5 mM glucose (i.e., low-glucose medium) for 0.5 h. [U-13C] Glucose was added to a final concentration of 20 mM and the incubation continued for 1 h. Metabolites were extracted in the ice-cold 80% methanol and analyzed using the Vion IMS QTOF Acquity UPLC system (Waters Corp.). Chromatographic separation was achieved on a BEH C18 (100 × 2.1 mm, particle size of 1.7 um; Waters Corp.) at 45 °C. The mobile phase consisted of eluent A (10 mM TBA/15 mM acetic acid) and eluent B (10 mM TBA/15 mM acetic acid/50% ACN). The flow rate was set at 0.3 mL/min. The elution profile was the same as that which is described in the preceding section. The mass spectrometer was operated in negative ion mode at an ESI voltage of 2.5 kV. Metabolites were analyzed using UNIFI software (v1.0.6171; Waters Corp.). Isolation of the mitochondria was performed by a modification of previously described method [56]. Around 1.5 × 106 HL-1 cells were harvested and resuspended in ice-cold mitochondria isolation buffer (20 mM HEPES, 5 mM KH2PO4, 50 μM MgCl2, 250 mM sucrose, and 0.2% BSA, pH 7.5). The sample was homogenized with 20 passages using a 27 gauge needle. The homogenate was centrifuged at 500× g for 10 min, and the supernatant was retained. The pellet was washed with the mitochondria isolation buffer, and then subjected to centrifugation. The supernatant fractions were combined and centrifuged at 10,000× g for 30 min at 4 °C to obtain a crude mitochondrial pellet. The CPT-1 activity was measured as described elsewhere with modifications [57]. The enriched mitochondria were suspended in 80 μL of mitochondria isolation buffer and incubated at 37 °C for 2 min. The enzymatic reaction was initiated by addition of 10 μL 1 mM palmitoyl-CoA and 10 μL 10 mM carnitine and incubated at 37 °C for 5 min. The product palmitoylcarnitine was analyzed using LC-MS under conditions that had been described previously [57]. The cells were rinsed with cold PBS, scraped, and collected for centrifugation. They were immediately lysed in a lysis buffer (20 mM Tris·HCl (pH 8), 1% Triton X-100, 137 mM NaCl, 1.5 mM MgCl2, 10% glycerol, 1 mM EGTA, 50 mM NaF, 1 mM Na3VO4, 10 mM β-glycerophosphate, 1 mM PMSF, 1 μg/mL leupeptin, and1 μg/mL aprotinin). The protein concentration of the lysate was determined using the Bradford method. The sample was analyzed by SDS-PAGE and immunoblotting with primary antibodies (including anti-VDAC1/porin antibody (ab14734; Abcam, Cambridge, UK), anti-actin (clone AC-40; Sigma-Aldrich), total OXPHOS rodent WB antibody cocktail (ab110413 (MS-604), Abcam), anti-CPT1A antibody (15184-1-AP; Proteintech group Inc., Rosemont, IL, USA), and anti-GAPDH antibody (GTX100118; GeneTex Inc., Irvine, CA, USA)), and appropriate secondary antibodies (including the horseradish-peroxidase (HRP)-conjugated goat anti-mouse antibody (sc-2005; Santa Cruz, Dallas, TX, USA), and the HRP-conjugated mouse anti-rabbit antibody (SC-2357; Santa Cruz Biotechnology, Dallas, TX, USA)). For immunofluorescence staining, the cells were cultured in a 35 mm glass-bottomed culture dish (Mattek Life Sciences, Ashland, MA, USA), and after the 7KCh treatment, the cells were fixed in 4% paraformaldehyde/PBS for 1 h. The fixed cells were rinsed with PBS and stained with 1 μg/mL anti-7-ketochosterol antibody (MKC-100n; Japan Institute for the Control of Aging (JaICA) Nikken Seil Co., Ltd., Shizuoka, Japan) in PBS/0.1% Triton X-100/1% bovine serum albumin (BSA) at 4 °C overnight. After rinsing it with PBS, the sample was stained with 2 μg/mL anti-mouse DyLight 488-conjugated secondary antibody (Thermo Fisher Scientific) and counterstained with Hoechst 33342 at room temperature for 2 h. The sample was examined under a Zeiss LSM780 fluorescence microscope (Carl Zeiss Microscopy, Oberkochen, Germany). The HL-1 cells were treated with 0, 10, 20, or 50 μM 7KCh for 24 h. They were washed with PBS, and subsequently, lysed with TRIzol reagent (Life Technologies, Carlsbad, CA, USA) for total RNA isolation. The total RNA was quantified using NanoDrop (Implen, Munich, Germany). The cDNA was synthesized from the total RNA using a RevertAid first-strand cDNA synthesis kit (catalog no. K1622; Thermo Fisher Scientific) according to the manufacturer’s protocol. The cDNA sample was mixed with the SsoFast EvaGreen supermix (catalog no. 172-5201; Bio-Rad, Hercules, CA, USA) in a reaction volume of 10 μL. PCR was performed with the CFX96 Touch real-time PCR detection system (Bio-Rad, Hercules, CA, USA). The relative expression of mRNAs was calculated using the comparative CT method and normalized to GAPDH expression. The primer sequences are listed in Table S1. The number of independent experiments (each with a triplicate set of samples) is stated or given as the N value in the respective figure legend. Data are mean ± SD. All the statistical analyses were performed with IBM SPSS 20.0 (IBM, Armonk, NY, USA). Two-way analysis of variance (ANOVA) with Sidak’s multiple comparison test, Mann–Whitney test, Kruskal–Wallis test with Dunn’s multiple comparison test, and Student’s t test were used where appropriate. A p value of < 0.05 is considered to be significant.

PMC10002500

Meng Tang,Chaohan Li,Cheng Zhang,Youming Cai,Yongchun Zhang,Liuyan Yang,Moxian Chen,Fuyuan Zhu,Qingzhu Li,Kehu Li

SWATH-MS-Based Proteomics Reveals the Regulatory Metabolism of Amaryllidaceae Alkaloids in Three Lycoris Species

24-02-2023

alkaloids,galanthamine,Lycoris,metabolism,SWATH-MS

Alkaloids are a class of nitrogen-containing alkaline organic compounds found in nature, with significant biological activity, and are also important active ingredients in Chinese herbal medicine. Amaryllidaceae plants are rich in alkaloids, among which galanthamine, lycorine, and lycoramine are representative. Since the difficulty and high cost of synthesizing alkaloids have been the major obstacles in industrial production, particularly the molecular mechanism underlying alkaloid biosynthesis is largely unknown. Here, we determined the alkaloid content in Lycoris longituba, Lycoris incarnata, and Lycoris sprengeri, and performed a SWATH-MS (sequential window acquisition of all theoretical mass spectra)-based quantitative approach to detect proteome changes in the three Lycoris. A total of 2193 proteins were quantified, of which 720 proteins showed a difference in abundance between Ll and Ls, and 463 proteins showed a difference in abundance between Li and Ls. KEGG enrichment analysis revealed that differentially expressed proteins are distributed in specific biological processes including amino acid metabolism, starch, and sucrose metabolism, implicating a supportive role for Amaryllidaceae alkaloids metabolism in Lycoris. Furthermore, several key genes collectively known as OMT and NMT were identified, which are probably responsible for galanthamine biosynthesis. Interestingly, RNA processing-related proteins were also abundantly detected in alkaloid-rich Ll, suggesting that posttranscriptional regulation such as alternative splicing may contribute to the biosynthesis of Amaryllidaceae alkaloids. Taken together, our SWATH-MS-based proteomic investigation may reveal the differences in alkaloid contents at the protein levels, providing a comprehensive proteome reference for the regulatory metabolism of Amaryllidaceae alkaloids.

SWATH-MS-Based Proteomics Reveals the Regulatory Metabolism of Amaryllidaceae Alkaloids in Three Lycoris Species Alkaloids are a class of nitrogen-containing alkaline organic compounds found in nature, with significant biological activity, and are also important active ingredients in Chinese herbal medicine. Amaryllidaceae plants are rich in alkaloids, among which galanthamine, lycorine, and lycoramine are representative. Since the difficulty and high cost of synthesizing alkaloids have been the major obstacles in industrial production, particularly the molecular mechanism underlying alkaloid biosynthesis is largely unknown. Here, we determined the alkaloid content in Lycoris longituba, Lycoris incarnata, and Lycoris sprengeri, and performed a SWATH-MS (sequential window acquisition of all theoretical mass spectra)-based quantitative approach to detect proteome changes in the three Lycoris. A total of 2193 proteins were quantified, of which 720 proteins showed a difference in abundance between Ll and Ls, and 463 proteins showed a difference in abundance between Li and Ls. KEGG enrichment analysis revealed that differentially expressed proteins are distributed in specific biological processes including amino acid metabolism, starch, and sucrose metabolism, implicating a supportive role for Amaryllidaceae alkaloids metabolism in Lycoris. Furthermore, several key genes collectively known as OMT and NMT were identified, which are probably responsible for galanthamine biosynthesis. Interestingly, RNA processing-related proteins were also abundantly detected in alkaloid-rich Ll, suggesting that posttranscriptional regulation such as alternative splicing may contribute to the biosynthesis of Amaryllidaceae alkaloids. Taken together, our SWATH-MS-based proteomic investigation may reveal the differences in alkaloid contents at the protein levels, providing a comprehensive proteome reference for the regulatory metabolism of Amaryllidaceae alkaloids. Lycoris spp. Amaryllidaceae is found in tropical and temperate regions, especially in China and Japan. It has long been widely used as folk medicine to treat various diseases [1,2]. Lycoris bulbs are used as traditional Chinese herbal medicine to treat sore throat, abscess, cancer, suppurative wound, poliomyelitis, mastitis, otitis media, ulcer, and neurodegenerative diseases [3]. According to the “Compendium of Chinese Materia Medica”, Lycoris also works in detoxifying, diminishing inflammation, pain relief, and diuresis [4]. Alkaloids are the main medicinal chemicals in the bulbs of Lycoris plants. Alkaloids are heterocyclic nitrogen compounds and have significant biological activities in human health applications. For example, the application in medicine is morphine, which has a powerful analgesic and sedative effect in clinical practice. Diterpenoid alkaloids isolated from Ranunculaceae were found to have antibacterial properties [5,6]. Carbohydrate alkaloids extracted from Solanum nigrum berries are medically used in the prevention of HIV infection and AIDS-related intestinal infections [7,8]. More than 600 alkaloids with various structures have been isolated from Lycoris plants including galanthamine, lycoramine, and lycorine thus far and have a variety of biological activities including anticancer, anti-inflammatory, anti-plasmodium, and antibacterial activities [3,9]. Galanthamine is a representative alkaloid of Lycoris, which is a selective, long-acting, reversible, and competitive acetylcholinesterase inhibitor. Symptomatic treatment for AD slows the progression of the disease and helps relieve memory loss [10]. Recently, transcriptomic and metabolomic analyses of L. radiata were performed to investigate the biosynthesis of galanthamine in different organs. It was found that LrNNR, LrN4OMT, and LrCYP96T were highly expressed in bulbs, which is consistent with the observation that more galanthamine is present in bulbs than in roots and leaves [11]. Functional characterization of the lycoris phenylalanine ammonia-lyase gene (LrPAL) revealed that LrPAL may be the limiting factor for the biosynthesis of galanthamine [12]. On the other hand, NpNBS condenses tyramine and 3,4-DHBA into norbelladine which is the first step in benzylisoquinoline alkaloid biosynthesis [13]. In addition, recombinant LlOMT catalyzes norbelladine to generate 4′-O-methylnorbelladine and overexpression of LlOMT in Lycoris longituba could increase the galanthamine content [14,15]. To date, previous research on the biosynthesis of galanthamine only focused on a few key genes, and the whole biosynthetic pathway of galanthamine in plants has not been comprehensively investigated particularly for the metabolism regulatory network from the perspective of proteome level. This study was conducted by SWATH-MS technology to explore the proteome of Lycoris. SWATH-MS is a specific data-independent-acquisition (DIA)-based method and is emerging as a technology that combines deep proteome coverage capabilities with quantitative consistency and accuracy [16]. SWATH overcomes the problem of stochasticity in data-dependent acquisition (DDA), making the detection results more reproducible and consistent. The wider coverage and higher detection sensitivity will further provide more ways to verify the function of the target protein in plants, which makes SWATH-MS a huge prospect in plant proteomics research [17]. In this study, the relative quantitative comparison of proteins among three groups of Lycoris samples with different alkaloid contents was carried out by quantitative proteomics technology based on SWATH-MS, and a total of 2193 proteins were detected. The differentially expressed proteins among the three groups were enriched in the amino acid metabolism pathway, galanthamine synthesis pathway, and sucrose metabolism pathway. Analysis of bioinformatics data will offer considerable information on the proteome among the three Lycoris with different alkaloid contents, and also provide deeper insights into the molecular mechanisms of Amaryllidaceae alkaloid’s regulatory metabolism. In summary, our SWATH-MS-based proteomics study can provide new insights into the mechanism of alkaloid-regulated metabolism in Amaryllidaceae at the protein level, and provide new ideas for the yield enhancement of alkaloid biosynthesis in the future. To further investigate the biosynthesis of Amaryllidaceae alkaloids in Lycoris species, we performed a SWATH-MS-based proteomic analysis of Lycoris longituba (Ll), Lycoris incarnata (Li), and Lycoris sprengeri (Ls) with different alkaloid contents. The three Lycoris materials were collected from the Institute of Botany, Chinese Academy of Sciences, Jiangsu Province, China. Among them, the total content of alkaloids in Ll was the highest, whereas in Ls was the lowest (Figure 1B). We selected three groups of samples as biological replicates for proteomic analysis. The correlation coefficient of the samples within the group and the principal component clustering results strongly reflected the high reliability of the appropriate sampling and suitability for subsequent analysis (Figure S1A). A total of 2193 proteins were quantified, and the group with the lowest alkaloid content Ls served as the control group. Proteins with a fold change above 2 or below 0.5 (p < 0.05) were considered differentially expressed proteins (DEPs) in this study. As shown in the volcano diagram, 485 upregulated proteins and 235 downregulated proteins were detected in Ll compared with Ls, whereas 288 upregulated proteins and 175 downregulated proteins were detected in Li (Figure 2A). The cluster heatmap showed significant changes in protein abundance. When we compared Ll with Ls, red and green colors, respectively, indicated upregulation and downregulation; when we compared Li with Ls, red and blue colors, respectively, indicated upregulation and downregulation (Figure S1C,D). This suggests that some crucial regulators or pathways have been induced, potentially affecting the biosynthesis of Amaryllidaceae alkaloids in Lycoris. KEGG pathway classification was performed on the two groups’ differential proteins and enriched in the top 10 pathways in order to quickly view the pathways that affect the biosynthesis of Amaryllidaceae alkaloids in Lycoris. For example, amino acid biosynthesis and starch and sucrose metabolic pathways were significantly activated and may be involved in contributing to the biosynthesis of Amaryllidaceae alkaloids in Lycoris (Figure 2C). Subsequently, we used the MapMAN BIN system to functionally classify the differentially expressed proteins into two groups. For example, compared with Ls, all differential proteins were assigned to 32 functional categories in Ll, including “protein metabolism”, “RNA processing and transport”, “signal transduction”, and other categories which constitute the main part of differential proteins, indicating these proteins have a great impact on the biosynthesis of Amaryllidaceae alkaloids in Lycoris. Among these, 21 of the 26 proteins involved in amino acid metabolism were downregulated. In the following sections, these upregulated or downregulated proteins will be mapped to different physiological and biochemical pathways, and the correlation between these biochemical pathways and the biosynthesis of alkaloids will be discussed (Figure 2D). Amino acids are closely related to the synthesis of plant alkaloids, and the nitrogen in complex alkaloids comes from amines derived from amino acid metabolism [18]. The amines contributing to alkaloid biosynthesis are derived from various amino acids, and are divided into two categories including polyamines and aromatic amines. Tyrosine is the precursor of multiple alkaloid families, including benzylisoquinolines (BIAs), Amaryllidaceae alkaloids, and betalains. TyrDC from Amaryllidaceae alkaloid biosynthesis has recently been discovered, it enables the incorporation of tyramine into the structures [15]. Through previous analyses of KEGG and MAPMAN BIN, we integrated the identified proteins embedded into different amino acid biosynthesis pathways by KEGG and MAPMAN BIN analysis (Table S1). Proteins with significant changes in abundance were enriched in the biosynthesis of alanine, aspartate, glutamate, cysteine, methionine, and arginine (Figure 3), of which alanine and arginine are the raw materials for the biosynthesis of plant alkaloids. Conversely, various alkaloids can negatively regulate the biosynthesis of glutamate in cells [19]. We detected a total of 38 differential proteins in the amino acid metabolism pathway, mainly concentrated in the biosynthesis and metabolism pathways of amino acids such as alanine, arginine, and cysteine. Among them, 18 genes were upregulated in Ll, 12 genes were upregulated in Li, and 8 genes were upregulated in Ls. In general, the highly expressed proteins were mainly concentrated in Ll, which was consistent with the highest alkaloid content of Ll, implicating that amino acid metabolism is closely related to the biosynthesis of Amaryllidaceae alkaloids in Lycoris. Galanthamine is a unique isoquinoline alkaloid and competitive acetylcholinesterase inhibitor that has broad prospects for future medical applications. The difficulty and high cost of synthesizing galanthamine have been the major obstacles to industrial production. The biosynthetic pathway of Gal has recently been elucidated including phenylalanine ammonia lyase (PAL), cinnamic acid-4-hydroxylase (C4H), coumaric acid 3-hydroxylase, tyrosine decarboxylase (TYDC), desmethylbelladine synthase (NBS), desmethylbelladine 40-O-methyltransferase (OMT), desmethoxymalidine synthase (CYP96T1), and N-methyltransferase (NMT) [15,20]. OMT is a methyltransferase that is involved in the biosynthesis of many alkaloids. For example, OMT has been shown to be responsible for multiple substrate and region-specific methylation in the roots of G. flavum and the biosynthesis and metabolism of plant alkaloids [21,22]. NMT is also a methyltransferase that has been shown to catalyze phenylisoquinoline alkaloids, one biosynthetic precursor of morphine [23]. In this study, we detected ten differentially expressed proteins among the three Lycoris species in the biosynthetic pathway of galanthamine. Among them, three proteins collectively known as OMT probably catalyzed norbelladine into 4′-O-methylnorbelladine and subsequently oxidized to N-demethylnawedine, further reduced to N-demethylgalanthamine. Seven other identified proteins known as NMT are probably responsible for the methylation of N-demethylgalanthamine to galanthamine (Figure 4). According to our MapMAN BIN system results, some proteins were detected to be enriched in starch and sucrose metabolism pathways suggesting a reciprocal regulation between alkaloids and sugar metabolism on the seasonal variation of alkaloids and total polyphenol contents, showing a contrasting tendency in sugar content and tissue development [24] (Figure 5A). Likewise, some proteins showed differences in abundance in secondary metabolic pathways, which are likely to contribute to the biosynthesis of alkaloids. For example, the upregulated 187456_c3_g1.p1 and 189251_c7_g1.p1 in Ll participate in the biosynthesis of sinapyl alcohol, resulting in an increase in the alkaloid raw material sinapyl alcohol, which would eventually lead to a rise in alkaloid content [25] (Figure 5B). Multiple-omics technologies have been extensively applied to various aspects of Lycoris research. Transcriptome analysis facilities the identification of key regulatory genes involved in anthocyanin metabolism during flower development in Lycoris radiata [26]. Proteomic analysis helps us better understand the molecular mechanism of L. radiata development and provides valuable information about the proteins involved in the development and stress response of other Lycoris genera [27]. Multi-omics such as transcriptomic and metabolomic analyses reveal that exogenous methyl jasmonate regulates galanthamine biosynthesis in Lycoris longituba seedlings; furthermore, anthocyanin biosynthesis, steroid biosynthesis, and R2R3 MYB TFs may play vital regulatory roles in petal color development in L. sprengeri [15,28]. Therefore, we used innovative SWATH-MS quantitative proteomics to investigate the molecular regulatory network of Amaryllidaceae alkaloids metabolism in Lycoris, which provides a new strategy for the future exploration of the biosynthesis and function of secondary metabolites. Furthermore, single-cell-based omics reveals that alkaloids are localized in Catharanthus roseus stem and leaf tissues [29], which would be an effective strategy to determine intercellular localization of alkaloids from different tissues in Lycoris, thus understanding the mechanism of alkaloids biosynthesis at the cellular level. A flow cytometry study of the nuclear DNA contents of Lycoris species (Amaryllidaceae) with different chromosome numbers revealed that the Lycoris genome contains approximately 30 G [30]. Since the Lycoris genome is so large and undiscovered, Pacbio’s new revolutionary long-read length sequencing system, revio, will significantly reduce sequencing costs, and is expected to be applied to the Lycoris genome, thus providing a high-quality map of the Lycoris genome and providing a scientific basis for the metabolic regulatory mechanisms in Lycoris. Seeking some target proteins among the differentially expressed proteins is necessary to illuminate the biosynthesis process of Amaryllidaceae alkaloids in Lycoris. Here, we collected five candidate proteins with dramatically increasing abundances (fold changes > 100), as shown in Table 1. Representatively, non-specific lipid-transfer protein 3 (LTP3: TRIBITY_DN163977_c0_g2.p1) exhibited 118.2-fold upregulation in Ll and 173.7-fold upregulation in Li. Plant lipid transfer proteins (LTPs) exhibit the ability to transfer lipids between membranes in vitro and have been shown to promote the production of abscisic acid (ABA) production, thereby stimulating the accumulation of alkaloids in cells [31]. Interestingly, crosstalk was detected between the JA and abscisic acid (ABA) signaling pathways in the regulation of tobacco (Nicotiana tabacum) alkaloid biosynthesis [32]. Therefore, further research is needed to explore secondary metabolites which can regulate the biosynthesis of Amaryllidaceae alkaloids through the ABA signaling pathway. In short, further genetic and molecular strategies with bioinformatics analysis will confirm the roles of these candidate proteins related to the biosynthesis of Amaryllidaceae alkaloids in Lycoris. RNA processing includes mRNA capping, splicing, cleavage, and polyadenylation, which participates in various key cellular processes and probably plays an important role in the biosynthesis of alkaloids. It was recently discovered that two potent anticancer alkaloids SANG and CHEL could directly bind to single-stranded RNAs, which reveals the fundamental structural and calorimetric aspects of the interaction of the natural benzophenanthridine alkaloids with single-stranded RNAs, facilitating the development of next-generation alkaloid therapeutics targeting single-stranded RNA [33]. Coincidentally, the polyadenylate [poly(A)] tail of mRNA was also found to have recently been identified as a potential drug target due to its important role in translational initiation, maturation, and stabilization of mRNA, and production of alternative proteins in eukaryotic cells. Some small molecular alkaloids with isoquinoline groups can bind to poly A with high affinity to form self-structures. It provides a reference for the development of novel bio-base molecules targeting poly(A) structures [34]. Eukaryotic pre-mRNAs are spliced to form mature mRNA. Alternative splicing greatly expands the transcriptomic and proteomic diversities related to physiological and developmental processes in higher eukaryotes. Alternative splicing is a posttranscriptional regulatory mechanism that generates multiple protein isoforms from a single gene through the use of alternative splice sites during splicing. However, the biosynthesis of alkaloids is a special metabolic pathway in plants, in which many genes have alternative splicing at different developmental stages and under stress conditions [35]. The finding of PR3b splicing regulation by JA/ET and NIC loci in Burley 21 is valuable to the genetic studies of low-alkaloid mutants and could provide clues to unravel the mechanisms by which JA/ET-signaling pathways regulate PR protein gene splicing [36]. Homoharringtonine (HHT) is a natural alkaloid with potent antitumor activity, which regulates the alternative splicing of Bel-x and Caspase 9 through a PP1-dependent mechanism, revealing a novel mechanism underlying the antitumor activities of HHT [37]. However, its role in the biosynthesis of galanthamine remains unclear. Interestingly, we found nine differentially expressed proteins related to RNA processing in the tested data in Table 2. This indicates that RNA processing is an important step in the biosynthesis of Amaryllidaceae alkaloids in Lycoris, and further experimental investigations are needed for functional validation. Plant materials of the three Lycoris varieties Lycoris longituba, Lycoris incarnata, and Lycoris sprengeri were collected from the Institute of Botany, Chinese Academy of Sciences, Jiangsu Province, China (32.05° N, 118.83° E) in late March when the leaves were growing vigorously (Figure 1A). Voucher specimens of Lycoris longituba (0653969), Lycoris incarnata (0653959), and Lycoris sprengeri (0653967) were deposited at the herbarium in the Institute of Botany, Chinese Academy of Sciences. The bulb samples with similar diameters and specifications to the three Lycoris varieties were selected, chopped, mixed well, and frozen at −70 °C. Tissue samples were first ground into a powder with liquid nitrogen and added to an appropriate volume of lysis buffer (2.5% SDS/100 mM Tris-HCl, pH = 8.5), sonicated in an ice-water bath for 15 min and centrifuged at 16,000× g for 20 min until clarified. The protein in the supernatant was precipitated by the acetone method. After washing with acetone and drying, 8 M urea/100 mM Tris-HCl solution (pH 8.0) was added to the protein precipitate to fully dissolve the protein. The samples were centrifuged at 12,000× g for 15 min, the supernatant was collected, dithiothreitol (DTT) was added to a final concentration of 10 mM, and the samples were incubated at 37 °C for 1 h to perform a reduction reaction to open disulfide bonds. Then, iodoacetamide (IAA) was added to a final concentration of 40 mM, and an alkylation reaction was performed at room temperature in the dark to block sulfhydryl groups. An appropriate volume of 100 mM Tris-HCl solution (pH 8.0) was added, the Bradford method was used to quantify the protein concentration, the urea concentration was diluted to below 2 M, and trypsin was added to each of the samples according to the ratio of protein amount trypsin amount =50:1, and incubated at 37 °C overnight with shaking. The next day, TFA was added to terminate the digestion and the pH value of the solution was adjusted to approximately 6.0. The solution was centrifuged at 12,000× g for 15 min and a homemade C18 cartridge was used for desalting. The desalted peptide solution was dried by a centrifugal concentrator and then stored frozen at −20 °C for on-board detection. The SWATH methods were used for subsequent MS-analysis using Triple TOF 5600 (Sciex) LC/MS system. The prepared peptide samples were first bound to the trap column and then separated by the analytical column (45 min gradient, 60 min total duration). Two mobile phases used to establish the analytical gradient were buffer A-0.1% (v/v) formic acid, 5% DMSO in H2O, buffer B-0.1% (v/v) formic acid, and 5% DMSO in acetonitrile. During SWATH scanning, each scan cycle consisted of one MS1 scan (ion accumulation time 250 ms, scan range 350–1500 m/z) and 100 variable window MS2 scans (ion accumulation time 33 ms, scan range 100–1800 m/z). Galanthamine, lycoramine and lycorine were purchased from Shanghai TCI Development Co. Their characteristics have been described previously [38]. A total of 0.2 g of bulb samples of Lycoris longituba, Lycoris incarnata, and Lycoris sprengeri were taken to be tested, freeze-dried, and ground into powder. Extraction was performed by sonication with 2 mL ethanol (70% high-performance liquid chromatography (HPLC)–grade) for 30 min. After centrifugation at 12,000 rpm for 10 min, the supernatant was taken and dried under vacuum. The samples were redissolved in 1 mL 0.1% formic acid-acetonitrile (v/v = 95/5) for liquid phase analysis [15]. Waters ACQUITY UPLC BEH C18 column (150 mm × 2.1 mm, 1.7 μm) was used as the liquid chromatography column and the separation was conducted using 0.1% formic acid (v/v) (A) and acetonitrile (B) with a 6-min linear gradient of 5–60% B at a flow rate of 0.2 mL/min. Quantification of galanthamine, lycoramine, and lycorine used 288 → 231, 290 → 189, 288 → 146 (m/z) transition reactions respectively. Experiments were conducted with three independent biological replicates. Least-significant difference test (LSD, p < 0.05) was used to compare the means. Different letters represent significant differences between groups (n = 3, p < 0.05). The mass spectrum files obtained by SWATH scanning are processed by DIA-Umpire to obtain secondary mass spectrum files that are used for database search. TPP software, Comet, and X!tandem search engines were used to search the database, and the search results were used as the spectral library for subsequent target extraction. The algorithm used for SWATH targeted extraction quantification is OpenSWATH. The test results were screened with 1% FDR. The protein quantitative intensity information obtained by SWATH analysis was used for difference comparison and T-test analysis after log2 transformation, data filling (imputation algorithm in Perseus software), and data normalization. Differential proteins were screened by fold difference (Ratio) and BH-corrected p-value (P.adjust). Proteins with a fold change above 2 or below 0.5 (p < 0.05) were considered differentially expressed proteins (DEPs) in this study. In the bioinformatics analysis, the DEPs identified were used for the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis by the clusterProfiler R package, and pathways of KEGG enrichment analysis results were drawn with reference to the KEGG mapper [39]. After the DEPs were compared with the homologous proteins in Arabidopsis thaliana, the MapMAN BIN system was used for functional classification. Amaryllidaceae plants are rich in alkaloids such as galanthamine (Gal), lycoramine (Lycm), and lycorine (Lyc) with a variety of biological activities, including anticancer, anti-inflammatory, anti-plasmodium, and antibacterial activities. In this study, three alkaloids profiling of Lycoris longituba (Ll), Lycoris incarnata (Li), and Lycoris sprengeri (Ls) were determined, and a comprehensive proteomic analysis was carried out with the aim of investigating the regulatory metabolism of Amaryllidaceae alkaloids in Lycoris species. The significant proteome changes in amino acid metabolism, starch, and sucrose metabolism, and galanthamine biosynthesis strengthen our understanding of the regulatory metabolism of Amaryllidaceae alkaloids in Lycoris. Additionally, we have also identified important candidate genes involved in the galanthamine biosynthesis pathway of alkaloid production in Lycoris. Taken together, we have offered new thoughts on the use of SWATH-MS to explore the regulatory metabolism of Amaryllidaceae alkaloids, providing a new strategy for the future exploitation of alkaloids.

PMC10002503

Phaedon D. Zavras,Ilias Sinanidis,Panagiotis Tsakiroglou,Theodoros Karantanos

Understanding the Continuum between High-Risk Myelodysplastic Syndrome and Acute Myeloid Leukemia

06-03-2023

MDS,AML with MDS-related changes,molecular alterations,classification,therapeutic approaches

Myelodysplastic syndrome (MDS) is a clonal hematopoietic neoplasm characterized by bone marrow dysplasia, failure of hematopoiesis and variable risk of progression to acute myeloid leukemia (AML). Recent large-scale studies have demonstrated that distinct molecular abnormalities detected at earlier stages of MDS alter disease biology and predict progression to AML. Consistently, various studies analyzing these diseases at the single-cell level have identified specific patterns of progression strongly associated with genomic alterations. These pre-clinical results have solidified the conclusion that high-risk MDS and AML arising from MDS or AML with MDS-related changes (AML-MRC) represent a continuum of the same disease. AML-MRC is distinguished from de novo AML by the presence of certain chromosomal abnormalities, such as deletion of 5q, 7/7q, 20q and complex karyotype and somatic mutations, which are also present in MDS and carry crucial prognostic implications. Recent changes in the classification and prognostication of MDS and AML by the International Consensus Classification (ICC) and the World Health Organization (WHO) reflect these advances. Finally, a better understanding of the biology of high-risk MDS and the mechanisms of disease progression have led to the introduction of novel therapeutic approaches, such as the addition of venetoclax to hypomethylating agents and, more recently, triplet therapies and agents targeting specific mutations, including FLT3 and IDH1/2. In this review, we analyze the pre-clinical data supporting that high-risk MDS and AML-MRC share the same genetic abnormalities and represent a continuum, describe the recent changes in the classification of these neoplasms and summarize the advances in the management of patients with these neoplasms.

Understanding the Continuum between High-Risk Myelodysplastic Syndrome and Acute Myeloid Leukemia Myelodysplastic syndrome (MDS) is a clonal hematopoietic neoplasm characterized by bone marrow dysplasia, failure of hematopoiesis and variable risk of progression to acute myeloid leukemia (AML). Recent large-scale studies have demonstrated that distinct molecular abnormalities detected at earlier stages of MDS alter disease biology and predict progression to AML. Consistently, various studies analyzing these diseases at the single-cell level have identified specific patterns of progression strongly associated with genomic alterations. These pre-clinical results have solidified the conclusion that high-risk MDS and AML arising from MDS or AML with MDS-related changes (AML-MRC) represent a continuum of the same disease. AML-MRC is distinguished from de novo AML by the presence of certain chromosomal abnormalities, such as deletion of 5q, 7/7q, 20q and complex karyotype and somatic mutations, which are also present in MDS and carry crucial prognostic implications. Recent changes in the classification and prognostication of MDS and AML by the International Consensus Classification (ICC) and the World Health Organization (WHO) reflect these advances. Finally, a better understanding of the biology of high-risk MDS and the mechanisms of disease progression have led to the introduction of novel therapeutic approaches, such as the addition of venetoclax to hypomethylating agents and, more recently, triplet therapies and agents targeting specific mutations, including FLT3 and IDH1/2. In this review, we analyze the pre-clinical data supporting that high-risk MDS and AML-MRC share the same genetic abnormalities and represent a continuum, describe the recent changes in the classification of these neoplasms and summarize the advances in the management of patients with these neoplasms. Myelodysplastic syndrome (MDS) is a heterogeneous group of clonal hematopoietic stem cell disorders characterized by bone marrow dysplasia, hematopoiesis failure and high risk of progression to acute myeloid leukemia (AML) [1,2,3]. One out of three patients diagnosed with MDS will progress to AML, characterized by an increased percentage of blasts over 20% [4]. Based on the 2016 world classification of hematologic malignancies, AML arising from MDS was classified as a distinct clinicopathologic entity entitled “AML with MDS-related changes” (AML-MRC) [5]. AML-MRC is associated with an overall poor response to both induction chemotherapy and low-intensity therapy, a high incidence of relapse and worse overall survival compared to patients with de novo AML [6]. Due to the poor outcomes of individuals with this disease, it has been the focus of intense research at both the pre-clinical and clinical levels. Recent large-scale studies have demonstrated that various cytogenetic abnormalities and gene mutations, such as deletion of 5q or 7q/7, mutations in spliceosome genes and genes encoding epigenetic modifiers, are common in MDS and AML-MRC but appear to be less frequent in de novo AML [7,8,9]. On the contrary, other genomic alterations, such as mutations in transcriptional factors or genes encoding proteins involved in signal transduction, are identified more frequently in AML-MRC compared to MDS [8,10,11], suggesting that these are probably later biological events during disease progression. Single-cell multi-omic studies have recently revealed specific patterns of clonal evolution driving the progression of MDS to AML-MRC [12,13], further highlighting that high-risk MDS and AML-MRC probably represent different stages of the same myeloid disease. These research advances have significantly improved our understanding of the biology of high-risk MDS and its transformation to AML-MRC, leading to significant improvements in the classification of these presentations and a more comprehensive therapeutic approach incorporating novel agents and broadening opportunities for clinical trials. These changes are reflected in the recently updated classification of myeloid neoplasms by the International Consensus Classification (ICC) and the World Health Organization (WHO) [14,15]. Notably, the recognition of a continuum between high-risk MDS and AML-MRC has yielded the enrollment of high-risk MDS patients into AML clinical trials and the enrollment of AML-MRC patients into MDS clinical trials, which is expected to tremendously improve our understanding of this disease’s biology and the survival of these patients. In this review, we summarize the existing pre-clinical studies that support the notion that high-risk MDS and AML-MRC represent a continuum and provide a novel understanding of the mechanisms of disease progression. Moreover, we describe the most recent changes in the classification of MDS and AML-MRC, as well as the implications of these advances in the management of patients with these myeloid neoplasms. Chromosomal abnormalities and copy number alterations are common in MDS and AML-MRC and have important prognostic implications. MDS and AML-MRC share various chromosomal abnormalities, such as deletion of 5q, 7/7q, 20q and complex karyotype, that typically lead to copy number alterations as opposed to de novo AML, which is frequently characterized by balanced re-arrangements, such as translocations 15;17 and 8;21 and inversion 16 [7]. Most of the chromosomal abnormalities identified in patients with MDS have critical prognostic implications and are associated with variable risk of disease progression. Deletion of 5q is the most common chromosomal abnormality in MDS, particularly among patients with lower-risk disease, and has been associated with an overall good prognosis [16,17,18]. Hematopoietic cells with 5q deletion have defective ribosomal biogenesis, resulting in elevated levels of free ribosomal proteins in their cytoplasm, which bind and promote the degradation of MDM2, a key P53 regulator [19]. This leads to P53 activation, which promotes cell-cycle arrest and induces apoptosis of erythroid progenitors and ineffective erythropoiesis [20]. These molecular events render P53 activity critical for the prevention of this disease’s progression to AML. Indeed, analysis of 55 individuals with low or intermediate-1 MDS with 5q deletion showed that TP53 mutation is associated with an increased risk of AML transformation [21]. Interestingly, TP53 mutations were identified in the majority of patients at an early disease stage and were always detectable before AML transformation [21]. Thus, the presence of a TP53 mutation should be an alarm for a high risk of progression in MDS with 5q deletion, even if identified at an earlier disease stage. It should be also noted that MDS patients with a complex karyotype, including 5q deletion, have a poor response to chemotherapy or lenalidomide and a dismal prognosis, which are linked to genomic instability due to P53 dysfunction [22]. Thus, MDS with 5q deletion in the context of a complex karyotype should not be managed as a low-risk disease; it needs to be approached as a high-risk disease. Consistently, 5q deletion is associated with poor survival, high genomic complexity and biallelic TP53 variants among patients with high-risk MDS and AML-MRC [23]. Loss of chromosome 7 or 7q is the second most common chromosomal abnormality among MDS patients. It is associated with overall poor prognosis and a higher risk of progression to AML-MRC with worse outcomes for patients with loss of the entire chromosome 7 [24]. Similarly to 5q, the loss of 7q is associated with high genomic complexity and particularly short survival among patients with AML-MRC [23]. A number of genes with a critical role in the pathogenesis of myeloid malignancies, such as SAMD9, SAMD9L, EZH2, CUX1 and MLL3, are located on chromosome 7 [25]. Most of these genes encode proteins that act as tumor suppressors, and deletion or missense mutations provide survival and growth benefits to hematopoietic stem and progenitor cells [26,27]. Recent data also support the notion that loss of chromosome 7 or 7q is associated with significant alterations in the expression of interferon-gamma pathway genes, promoting an immunosuppressive microenvironment enriched in T regulatory immune cells, potentially conferring with the inferior outcomes of patients with myeloid neoplasms with these abnormalities [28]. Overall, these results support that loss of chromosome 7 or 7q contributes to the pathogenesis and, more importantly, to the progression of MDS to AML-MRC. The presence of three or more chromosomal abnormalities is defined as a complex karyotype (CK) and is always considered a predictor of adverse outcomes, independent of actual specific abnormalities [29]. CK, particularly monosomal karyotype (MK), is strongly associated with one or more mutations in TP53 and frequently with loss of heterozygosity (LOH) of this gene [30,31,32]. A number of studies have highlighted that CK is associated with particularly poor survival and a high risk of transformation to AML among MDS patients, independently of blast percentage and other clinical features such as cytopenias and transfusion dependence [33,34,35]. These findings further support that the notion that the biology of MDS and its progression is defined by genetic alterations, probably at the earlier stages of this disease. Next-generation sequencing studies have demonstrated that MDS and AML-MRC share mutations in genes implicated in cellular processes, such as RNA splicing, epigenetic and transcriptional regulation, and DNA damage repair [8]. Based on an analysis of 299 AML patients, mutations in spliceosome genes such as SRSF2, SF3B1, ZRSF2 and U2AF1, EZH2, BCOR and STAG2 are >95% specific for AML-MRC and can distinguish AML-MRC from de novo AML, even in the absence of an antecedent MDS diagnosis [8]. The identification of mutations in genes implicated in different cellular processes highlights the progression of clonal myeloid neoplasms from earlier stages, such as clonal hematopoiesis of indeterminate potential (CHIP) to MDS and AML-MRC. Particularly, mutations in epigenetic modifiers, such as TET2 and DNMT3A and TP53, are common across a spectrum of clonal hematopoietic disorders, such as CHIP, clonal cytopenias, MDS and AML-MRC [36]. On the contrary, mutations in spliceosome genes are more common in MDS and AML-MRC [37,38], while mutations in transcriptional factors, such as RUNX1 and CEBPA, and activating signaling genes, such as NRAS and FLT3, are more common in AML-MRC [8,9]. These findings suggest that clonal evolution through the acquisition of additional somatic mutations probably drives the progression of these clonal myeloid neoplasms. SF3B1 is commonly mutated in MDS [39], with the majority of mutations being missense substitutions affecting the spliceosome machinery, resulting in altered proteome [40]. SF3B1 mutations have been associated with MDS with ring sideroblasts and a relatively low risk of AML transformation [41]. However, specific substitutions, such as K666N hotspot mutation, result in distinct patterns of RNA splicing and increased risk of progression to AML-MRC [42]. SRSF2 mutations are found in 10–15% of patients with MDS and are associated with older age, higher levels of hemoglobin and a normal karyotype [43]. SRSF2 mutations frequently co-occur with RUNX1, IDH2 and ASXL1 mutations and are associated with a higher risk of transformation to AML-MRC [43,44]. The adverse biology of MDS with SRSF2 mutations could be associated with the altered splicing of genes such as EZH2 [45] and their implications in DNA damage accumulation through alterations of the P53 function [46]. U2AF1 mutations occur in about 5–10% of MDS patients and are associated with a high risk of disease progression to AML-MRC [47]. Aberrant splicing due to U2AF1 mutations results in suppression of ATG7 levels and reduced autophagy, causing increased oxidative stress and chromosomal instability [48]. A recent study also revealed that U2AF1 mutations cause an increase in the expression of the long isoform of interleukin-1 receptor-associated kinase 4 (IRAK4), which promotes the activation of nuclear factor kappa-light-chain-enhancer of B cells (NF-kB) that have been implicated in leukemic growth [49]. ZRSR2 is mutated in about 5% of patients with MDS, affect predominantly men and cause aberrant splicing by retaining U12-dependent introns [31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50]. While most studies indicate a negative impact of ZRSR2 mutations on the survival of MDS patients, the impact is not as strong compared to SRSF2 and U2AF1 mutations [51]. These results highlight that mutations in spliceosome genes can mediate leukemic growth and promote the progression of MDS to AML-MRC. STAG2 encodes a subunit of the cohesin complex, which regulates the separation of sister chromatids during mitosis [52] and is mutated in close to 10% of patients with MDS [53] with male predominance [54]. STAG2 mutations have been associated with poor overall survival among MDS patients [53] but have not been directly linked to progression to AML-MRC. It has been hypothesized that the haploinsufficiency of genes encoding subunits of the cohesin complex affects the expression of genes that are essential for lineage priming and differentiation [55]. Thus, mutations in these genes likely cause abnormal hematopoietic stem cell maturation contributing to dysplasia, but cooperating oncogenic mutations are required for progression to AML-MRC [55]. Indeed, RUNX1, CEBPA, NPM1, and RAS mutations co-occur with STAG2 mutations in AML [9,56,57]. EZH2 is a core component of polycomb group complex 2 (PRC2), which is responsible for trimethylation of lysine 27 of histone 3 and plays a critical role in epigenetic gene silencing [58]. EZH2 is mutated in 4–5% of MDS patients [59] and is frequently co-mutated with ASXL1 and RUNX1 [60,61]. Concurrent TET2 and EZH2 deletion is sufficient to induce an MDS/myeloproliferative phenotype in mice [62], and in a RUNX1 mutant model, EZH2 loss promotes myelodysplasia development but inhibits transformation to AML [63]. In fact, EZH2 is upregulated in AML with CK [64] and EZH2 deletion suppresses the proliferation of MLL-AF9-transformed cells and delays AML transformation in transgenic mice [65]. Consistently, EZH2 mutations have been associated with poor overall survival in patients with chronic myeloid neoplasms [30,54,66] and poor response to hypomethylating agents [67] but not with a higher risk of AML-MRC transformation. Thus, similarly to STAG2 mutations, it can be hypothesized that EZH2 mutations are early events in MDS [68], and additional oncogenic mutations are required for transformation to AML-MRC. The BCOR gene is located on the chromosome X and encodes transcription factor that is essential for normal embryonic development [69]. BCOR mutations are detected in less than 5% of patients with MDS and are typically frameshift insertions or deletions, or stopgain or non-sense mutations, causing a suppression of function of the BCOR protein in hematopoietic cells [70,71]. Recent studies have demonstrated that loss of BCOR function is associated with altered activity of polycomb group repressive complex 1 (PRC1), resulting in induced myeloid cells’ proliferation [71]. Most of the studies on large cohorts have demonstrated that BCOR mutations have a neutral impact on overall survival of MDS patients [72,73]. RUNX1 encodes the alpha subunit of the core binding transcription factor and is implicated in the differentiation of hematopoietic stem cells [74]. Germline point mutations of RUNX1 have been identified in an autosomal dominant platelet disorder with a high risk of transformation to AML [75]. RUNX1 mutations occur in 10–15% of MDS and have been associated with thrombocytopenia, poor survival with a high risk of progression to AML-MRC, co-occurrence of RAS mutations and loss of chromosome 7/7q [76,77,78]. Recent data have also associated RUNX1 mutations with poor response to hypomethylating agents [67,68,69,70,71,72,73,74,75,76,77,78,79]. Recently, it was reported that RUNX1 mutation is associated with rapid progression of low-risk MDS and that mutations in this gene in CD34+ cells from low-risk MDS patients result in dysregulated DNA damage repair and cellular senescence [78]. The authors also found that the transcriptional profiles of CD34+ cells from low-risk MDS patients with RUNX1 mutation resemble those of high-risk MDS at diagnosis [78]. These findings further suggest that specific genomic alterations identified even at the earlier stage of MDS can alter disease biology and induce disease rapid progression. NRAS and KRAS mutations occur in about 4–5% of patients with MDS and are associated with higher white blood cell counts and a higher risk of transformation to AML-MRC [80,81,82,83]. Mutations in RAS genes result in RAS/Raf/MEK and PI3K/AKT signaling activation and are sufficient to induce a myeloproliferative phenotype or even AML in mice [84,85]. Similarly, FLT3-ITD mutations are relatively uncommon among MDS patients with a frequency of 0.6–6% [86] and have been associated with higher blasts percentage, increased risk of leukemic transformation and worse overall survival [87]. FLT3-ITD mutations in hematopoietic stem and progenitor cells cause constitutive autophosphorylation of the FLT3 receptor, leading to the activation of downstream signaling, such as STAT3/5, MAPK and PI3K/AKT [88]. Analysis of 278 MDS patients with low- or intermediate-1-risk MDS revealed that the detection of a RAS or FLT3-ITD mutation is associated with a particularly high risk of progression to AML-MRC with very poor response to treatment with hypomethylating agents [11]. Similarly, the detection of RAS or FLT3-ITD mutations at the time of transformation of MDS to AML-MRC is correlated with significantly worse survival [89]. Thus, independently of the disease stage at diagnosis, the acquisition of these signaling genes mutations significantly alters disease biology and is connected with poor outcomes. TP53 is a tumor-suppressor gene located on the short arm of chromosome 17 and encodes transcription factor P53, which is critical for cell-cycle arrest, DNA repair and apoptosis in the setting of DNA damage [90]. As a result, cancers with TP53 mutation accumulate DNA damage and demonstrate a poor response to cytotoxic therapy [91]. TP53 alterations are discovered in about 5–10% of MDS patients [34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92], with the majority of cases having “multi-hit” involvement with more than one genomic and/or chromosome 17 abnormalities and fewer alterations in other genes [92]. A recent analysis of 3324 MDS patients revealed that high-risk presentation with CK, high incidence of transformation to AML-MRC and poor overall survival are associated with multi-hit TP53 and not monoallelic TP53 alterations [92]. It should be noted that these associations were independent of the Revised International Prognostic Scoring System (IPSS-R) [92], once more indicating that, independent of the disease stage at diagnosis, the molecular alterations that affect malignant cell biology define disease course and outcomes. The incidence, biological mechanisms and clinical significance of genetic abnormalities in MDS and AML-MRC are summarized in Table 1. Single-cell DNA and RNA sequencing has significantly improved our understanding of heterogeneity and clonal evolution in myeloid neoplasms [94,95,96]. Chen et al. utilized longitudinal, paired samples from seven MDS patients who progressed to AML-MRC and performed ensemble deep sequencing and single-cell sequencing of sorted pre-malignant and malignant stem cells based on the expression of previously identified stem cell markers and blasts [95]. The authors demonstrated a significantly higher complexity of sub-clonal mutations in stem cells at the MDS stage compared to blast cells [95]. Moreover, the authors presented data that support a model of parallel clonal evolution at the stem cell level during MDS progression [95]. Particularly, all the patients studied demonstrated a highly diverse pool of pre-malignant MDS stem cells with most of the patients showing relatively early branching of these pre-malignant MDS stem cells towards progression to AML stem cells [95]. These findings suggest that the drivers of progression are probably present in MDS stem cells even at the earliest stages and dictate the natural history of this disease. Finally, the identification of actionable mutations in MDS stem cells could provide a particularly promising opportunity for the early prevention of disease progression to AML-MRC. Guess et al. performed single-cell DNA sequencing of paired MDS and AML-MRC to better characterize clonal shifts upon MDS progression [12]. The authors found two different patterns of clonal evolution during MDS progression to AML-MRC [12]. The first pattern, called the static group, was characterized by relatively stable clonal architecture during progression [12]. In this group, patients had founder mutations in DNA methylation genes, such as DNMT3A, TET2 and IDH1/2, and potentially epigenetic alterations drive progression to AML-MRC through blast growth [12]. On the contrary, the second pattern, called the dynamic group, was characterized by significant clonal architectural changes defined by new chromosomal abnormalities and TP53 mutations or genomic alterations with more prominent mutations in signaling genes [12]. Single-cell transcriptomic analysis demonstrated that pathways such as cell-cycle regulation via E2F activation and inflammatory signaling mediated by cytokine receptors are amongst the top upregulated cellular pathways associated with transformation to AML-MRC [12]. These results are consistent with mechanistic studies highlighting that inflammatory signaling is linked with cell-cycle progression promoting the progression of MDS to AML-MRC and affecting the disease biology and sensitivity to current therapies [93,97,98,99,100,101]. The International Consensus Classification (ICC) and the World Health Organization (WHO) have recently updated the classification of myeloid neoplasms that come with a better understanding of the biology of the disease and therapeutic advancements. Compared to the WHO 2016 classification, the ICC introduced changes in the predefining blast cut-offs, and both committees incorporated genetic alterations of the diseases. The new guidelines recognize clonal hematopoiesis (CH) as the precursor of cytopenias, ranging from clonal cytopenia of undetermined significance (CCUS) to MDS. Cases with cytopenia but without dysplasia lack three cytogenetic abnormalities, namely del(5q), –7/del(7q) and complex karyotype, and are now classified under CCUS. The definitions of the pre-malignant clonal hematopoiesis disease spectrum are summarized in Table 2. The ICC and the WHO recognize MDS as myeloid neoplasms characterized by clonal hematopoiesis, dysplastic changes in the BM of PB and cytopenia in at least one cell line [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102]. Clonality is established via next-generation sequencing (NGS) or karyotype. In cases where clonality cannot be proven, the diagnosis of MDS can still be made with qualifying dysplasia and cytopenia(s). On the other hand, irrespective of the presence of dysplasia, MDS-defining genetic abnormalities are sufficient for diagnosis, especially in the setting of persistent cytopenia. Emphasis has been placed on the molecular categorization of MDS and AML in the new ICC guidelines because the genetic footprint seems to be a stronger predictor of prognosis than the degree of dysplasia [103]. The previously defined categories of MDS with excess blasts (EBs) have been refined. Recent data show that MDS with a higher blast percentage is more likely to transform to AML and carries a worse prognosis, similar to overt AML [104,105]. To more accurately reflect the continuum between MDS and AML, a new entity “MDS/AML” has been suggested to define MDS with blasts between 10 and 19%, lacking AML-defining genetics. Patients with MDS/AML should be eligible for either MDS or AML treatment modalities or clinical trials. MDS/AML is further subclassified into cases with mutated TP53, cases with myelodysplasia-related gene mutations, myelodysplasia-related cytogenetic abnormalities and MDS/AML-NOS. Blasts of greater than 20% of define AML in these cases. AML positive for the Philadelphia chromosome (Ph+) or BCR:ABL1/t(9;22)(q34.1;q11.2) is an exception and is still defined as blasts ≥20%. The MDS/AML category is not applicable in this case so as to distinguish it from the progression of Ph+ CML. For cases with 5–9% and 2–9% blasts in BM and PB, respectively, the term MDS with EB is used. The previously defined category of AML-MRC has been eliminated by the ICC. Genetic characteristics, rather than clinical history (i.e., de novo AML, or AML arising from MDS, or therapy-related AML), are shown to be better primary determinants of disease classification [8]. Two new categories have been suggested instead, irrespective of any history of MDS, namely AML with mutated TP53 and AML with myelodysplasia-related gene mutations. The latter encompasses the prior entity of AML with mutated RUNX1, expanded to include the other eight genes shown to be related to AML arising from MDS or MDS/MPN as follows: ASXL1; BCOR; EZH2; SF3B1; SRSF2; STAG2; U2AF1; and ZRSR2. These mutations are strongly associated with a history of MDS or MDS/MPN and carry a poor prognosis [8]. In the absence of class-defining genetic abnormalities, a separate category of AML with myelodysplasia-related cytogenetic abnormalities incorporates cases with a complex karyotype (defined as ≥3 unrelated clonal chromosomal abnormalities) and/or other unbalanced chromosomal changes that previously fell into the prior AML-MRC category but lack a TP53 mutation or myelodysplasia-related gene mutations, i.e., del(5q)/t(5q)/add(5q), –7/del(7q), +8, del(12p)/t(12p)/add(12p), i(17q), –17/add(17p)/del(17p), del(20q), or idic(X)(q13) [106]. The WHO has retained the 20% blast cut-off between MDS and AML for disease without AML-defining mutations because altering the threshold was felt to be arbitrary and carries the risk of overtreatment. Hence, the family of MDS with increased blasts (IBs) includes disease with <20% blasts, and this is further subclassified into IB1 (5–9% BM and/or 2–4% PB blasts), IB2 (10–19% BM and/or 5–19% BM blasts), and MDS with fibrosis (in which case 5–19% BM and/or 2–19% PB blasts are allowed) [15]. The AML-MRC category is maintained in the WHO guidelines, and it encompasses de novo AML or AML transformed from MDS or MDS/MPN. A key change is that diagnosis can no longer be made solely based on morphology, and it requires the identification of genetic abnormalities and/or history of MDS or MDS/MPN. The group of single-gene somatic mutations again includes the same group of myelodysplasia-associated genes mentioned above, present in the vast majority of AML arising from MDS or MDS/MPN with the exception of RUNX1, claiming insufficient unifying characteristics [15]. As described above, MDS and AML with TP53 mutation are usually associated with a complex karyotype, carry a much worse prognosis and are recognized as separate disease entities by the ICC [9,102,108,109]. The TP53 mutation is an independent predicting factor of poor response to cytarabine or hypomethylating agent (HMA)/Venetoclax-based therapy [110]. MDS with mutated TP53 (i.e., blasts < 10%) requires the presence of a multi-hit TP53 mutation or TP53 mutations with variant allele frequency (VAF) > 10% and a complex karyotype, often with a loss of 17p. Cases with a complex karyotype but not a TP53 mutation do not qualify for this category as they have different, more favorable outcomes [111]. Multi-hit TP53 mutation is defined as two distinct TP53 mutations (each with VAF > 10%), or a single TP53 mutation with either 17p deletion (VAF > 50%) or loss of heterozygosity (LOH) in the 17p TP53 locus [92]. In the absence of LOH information, a single TP53 mutation in the context of a complex karyotype also qualifies for diagnosis. Monoallelic TP53-mutated MDS has different disease biology to multi-hit disease; thus, it is not included in this category and is instead categorized under MDS-NOS (if no EBs) or MDS with EB. On the contrary, monoallelic TP53-mutated MDS/AML or AML have a worse prognosis and are allowed in this category. TP53-mutated AML is not defined separately by the WHO [15]. The therapy-related subgroup of MDS and AML is eliminated by the ICC, and therapy-related cases are now subclassified following primary diagnosis [102]. For example, MDS-NOS with single lineage dysplasia is therapy-related. This definition applies to therapy-related MDS, AML or MDS/AML, and AML progressed from MDS or MDS/MPN. Although it is recognized that secondary disease carries a worse prognosis, the priority is to first define the disease based on its morphologic and genetic features. The secondary myeloid neoplasm category has been preserved by the WHO and it includes therapy-related diseases (renamed to “myeloid neoplasm, post cytotoxic therapy or MN-pCT”), and diseases associated with germline predisposition (Down-syndrome-related AML falls into this category) [15]. Exposure to PARP inhibitors has been incorporated as a qualifying criterion for MN-pCT, whereas methotrexate exposure has been excluded, based on a lack of a significant association that has been reported [112,113]. Of note, AML transformed from another myeloid neoplasm is no longer classified as secondary by the WHO. AML arising from MDS or MDS/MPN is now classified under AML-MR (as per above) and AML arising from MPN is retained under the MPN category. In the fourth edition of the WHO classification of myeloid neoplasms released in 2016, there was a significant overlap between the various AML subcategories [5]. A 20% blast cut-off was required for a diagnosis of AML, apart from cases of APL or AML with mutated CBF. Another major change introduced by both the ICC and the WHO groups is the expansion of AML-defining recurring genetic abnormalities to include t(9;11)(p21.3;q23.3)/MLLT3::KMT2A (or other KMT2A rearrangements), t(6;9)(p22.3;q34.1)/DEK::NUP214, inv(3)(q21.3q26.2)/MECOM(EVI1) (or other MECOM rearrangements) and NPM1, in-frame basic leucine zipper-region (bZIP) CEBPA mutations [114,115,116,117]. Biallelic CEBPA mutations are no longer required, with several studies showing favorable prognosis with monoallelic bZIP mutations [118,119]. In the case of NMP1-mutated AML, the WHO allows diagnosis irrespective of the blast percentage, highlighting the fact that NMP1-mutated neoplasms with blasts <20%, namely MDS and MDS/MPN cases, are associated with aggressive disease and rapid progression to overt AML within 12 months of diagnosis [120]. As newer, targeted therapies have emerged, early identification of the molecular subtype of the disease can lead to earlier initiation of treatment. The classification algorithm of MDS and AML based on the new changes can be seen in Figure 1 and Figure 2. The IPSS-R has been used for years as the gold standard method for MDS risk stratification, as well as the numerous clinical implications this entails, such as clinical trial design and treatment modalities [121]. The IPSS-R is solely based on the hematologic and cytogenetic abnormalities of the disease and does not take into account individual mutations leading to clonality. As new knowledge around the significance of gene mutations in the prognostication of myeloid neoplasms has emerged, a new risk stratification scoring system has been recently suggested [30,31,73]. The IPSS-Molecular (IPSS-M) prognostic model was recently developed [122]. This new evidence is based on data from 2957 patients, from whom mutations of 152 different genes that were implicated in the development of myeloid neoplasms were assessed. The variables of the model were selected based on a stability selection algorithm that was applied to three different clinical outcomes as follows: overall survival (OS), leukemia-free survival (LFS) and AML transformation. As opposed to the IPSS-R, the absolute neutrophil count (ANC) is not considered part of the IPSS-M. At least one gene abnormality was found in 94% of the patients and the increasing number of mutations was negatively correlated with leukemia-free survival and thus disease severity [122]. This model uses hematologic (i.e., medullary blast percentage, hemoglobin, and platelet counts) and cytogenetic data (the five clusters that were included in the IPSS-R scoring system) and, in addition, includes molecular data, namely mutations identified in 16 “main effect” genes and 15 additional “residual” genes. Mutations in three of the main effect genes, TP53 multihit (but not mono-allelic TP53 mutations), MLLPTD (partial tandem duplication) and FLT3 (tyrosine kinase domain and internal tandem duplication) strongly correlated with adverse outcomes, highlighting the importance of screening for those genes at the time of diagnosis [122]. FLT3 and MLLPTD mutations exhibited the strongest correlation with AML transformation [122]. Other main effect genes included ASXL1, CBL, DNMT3A, ETV6, EZH2, IDH2, KRAS, NPM1, NRAS, RUNX1 and SF3B1 (computational pattern SF3B1a) or SF3B15q, SRSF2, and U2AF [122]. In the case of SF3B1 mutations, there are three different clusters with different prognostic outcomes based on the co-mutation pattern identified. SF3B15q (7% of SF3B1 mutant cases) refers to the concomitant presence of isolated 5q deletion; SF3B1β (15% of the cases) refers to co-mutation of SF3B1 along with any of the BCOR, BCORL1, NRAS, RUNX1, SRSF2, or STAG2 genes; and SF3B1α (78% of the cases) is defined as any other mutant SF3B1, in which case 37% of the patients were found to have simple co-mutation patterns, including one of the DNMT3A, TET2, and/or ASXL1 genes [122]. Favorable outcomes were observed for the SF3B1α group, whereas they were not for the other two groups [122]. Fifteen residual genes were also determined (BCOR, BCORL1, CEBPA, ETNK1, GATA2, GNB1, IDH1, NF1, PHF6, PPM1D, PRPF8, PTPN11, SETBP1, STAG2 and WT1) based on adverse effects identified by univariate analysis and greater than 1% recurrence among all patients with MDS [122]. Six disease risk strata were identified based on this new model (instead of five per IPSS-R), which correlated with leukemia-free survival, overall survival and transformation to AML, as follows: very low (14%); low (33%); moderate low (11%); moderate-high (11%); high (14%); and very high (17%). Almost half (46%) of the patients of the cohort were re-classified based on IPSS-M. Of these, 74% were upstaged and 26% were downstaged [122]. The IPSS-M was applied to a separate Japanese cohort of 718 patients, and its discriminative power in terms of LFS was found to be superior to the IPSS-R across all key endpoints. The IPSS-M is applicable to both de novo disease and treatment-related MDS. It is a dynamic metric, i.e., it can be applied at any point during treatment. Its use in predicting treatment outcomes is yet to be determined. Both risk stratification and treatment management of AML are based primarily on the age, the molecular and cytogenetic footprint, and early measurable residual disease (MRD) status [123]. MRD positivity following intensive or non-intensive induction therapy is a major, independent, post-diagnosis poor prognostic factor [124,125,126]. The European Leukemia Net (ELN) has introduced some changes in regard to the genetic risk stratification of AML [127]. FLT3-ITD-mutated AML irrespective of the allelic ratio (formerly FLT3-ITDhigh for biallelic mutation and FLT3-ITDlow for mono-allelic mutation) or the concurrent NMP1 mutation status is now considered intermediate risk disease. This is based on evidence of disease modifying impact of midostaurin-based therapy [128]. The previously recognized RUNX1 and ASXL1-mutated AML categories are expanded to include the whole AML category with gene-related mutations classified under high-risk disease. This new category encompasses the nine above-mentioned gene mutations (ASXL1, RUNX1, BCOR, EZH2, SF3B1, SRSF2, STAG2, U2AF1, ZRSR2) associated with AML arising from MDS. NMP1-mutated AML with concurrent adverse-risk-associated cytogenetic abnormalities has been reclassified as a high-risk disease, based on new data that shows an association with worse outcomes [129]. AML with in-frame monoallelic mutations in the basic leucine zipper region (bZIP) of CEBPA is now also classified under favorable-risk disease, along with biallelic mutated disease [130]. The following cytogenetic abnormalities are added to the adverse risk group: t(3q26.2;v), involving the MECOM gene, and t(8;16)(p11.2;p13.3), associated with KAT6A::CREBBP gene fusion (39, 40). Last, hyperploidy with multiple trisomies or polysomies is no longer considered a complex karyotype and is therefore not classified as an adverse risk [14]. Understanding the biology of MDS progression to AML-MRC is critical for the discovery of effective treatments and the development of a personalized therapeutic approach for patients with high-risk MDS and AML-MRC. Hypomethylating agents (HMAs) followed by allogeneic bone marrow transplantation (AlloBMT) remains the most frequently used approach [2]. Despite the requirement for further interpretation and exploitation of the new biological findings, significant progress has been made in the therapeutic approach of patients with MDS and AML-MRC. Venetoclax has now been approved in combination with hypomethylating agents by the United States Food and Drug Administration (FDA) [131], and targeted therapies, such as IDH1/2 inhibitors, have been effectively used in high-risk MDS and AML-MRC patients with IDH1/2 mutations [131]. Clinical trials testing novel agents, such as magrolimab and epretanoport, and anti-CD123 approaches in high-risk MDS and AML-MRC are currently ongoing [131]. Importantly, the new classifications of WHO and ICC have allowed clinicians to enroll high-risk MDS patients into AML clinical trials and AML-MRC patients into MDS clinical trials [15,132], which will undoubtedly expedite the results of the numerous ongoing clinical trials. The current therapeutic approach for patients with MDS relies on (1) symptoms and clinical manifestation; (2) risk assessment using IPSS or Revised-IPSS or the most recently added IPSS-M calculator [122]; and (3) overall performance status and patient’s preference. AlloBMT constitutes the only potentially curative approach for patients in the clinical spectrum of high-risk MDS and AML-MRC. The decision of AlloBMT candidacy is based on factors such as age, physical and mental status, and comorbidities and should be accessed individually [133]. Specifically, age is not an absolute exclusive criterion but a relative one. Although younger patients (<65 years old) tend to have longer overall survival, there is no significant difference in survival in older patients (>65 years old) [134]. The timing of the AlloBMT remains a significant challenge in clinical practice. Cutler et al. demonstrated that the optimal point of alloBMT is the time of progression to intermediate risk as per the IPSS-R. Particularly, low- and very-low-risk patients should be monitored closely and they should undergo delayed transplantation, but intermediate to very-high-risk patients benefit from early transplantation [135]. It is not yet clear which regimen is more effective at inducing remission in patients as a bridge to alloBMT. HMAs or induction chemotherapy have been used for this purpose [136] with an ongoing clinical trial comparing these two approaches before AlloBMT in patients with MDS (NCT01812252). The RICMAC and MDS200 trial failed to show any superiority of intensive chemotherapy over low-dose chemotherapy in patients with high-risk MDS or in AML-MRC patients [137]. Interestingly, another important consideration for MDS and AML-MRC patients undergoing alloBMT is the conditioning regimen. Scott et al. showed that even though a MAC regimen leads to increased transplant-related mortality, 4-year overall and relapse-free survival were significantly better compared to RIC, with AML-MRC patients deriving increased benefits compared to MDS patients [138]. Despite initial enthusiasm, maintenance strategies following AlloBMT for patients with high-risk MDS and AML-MRC have not led to significant survival benefits [139,140,141], except for specific subsets, such as TP53-mutated diseases [142]. The use of HMAs as first-line therapy is well established for patients with high-risk MDS and AML-MRC, and it is overall considered superior to intensive chemotherapy, partially due to a more tolerant and modest toxicity profile [143]. HMAs reduce the risk of death by 50% in high-risk MDS when used as first-line therapy in comparison with conventional care regimens, while most of the patients respond within 6 months of treatment [143]. Unfortunately, the HMA effect is transient, with responses usually maintained for less than 12 months, particularly in patients with adverse disease biology [144]. The prognosis of patients who progress to HMA is particularly dismal [144]. High-risk MDS and AML-MRC patients with poor response to HMAs showed a response rate of 41% and 32% when treated with chemotherapy and consequently underwent AlloBMT at a rate of 40% and 42%, respectively [145]. Despite the mechanisms of HMA resistance being the subject of intense pre-clinical research [101,146,147,148,149,150], novel combinational therapies based on these mechanisms have not yielded clinically significant results. As opposed to post-AlloBMT maintenance, maintenance therapy with HMAs and particularly the oral formulation of azacitidine, CC-486 improve the outcomes of AML patients including a number of AML-MRC patients in first remission who did not receive AlloBMT [151]. Venetoclax (VEN, ABT-199) is a BH3 mimetic that inhibits the anti-apoptotic effects of BCL-2 proteins, inducing cell death in AML blasts, and was approved by the FDA in 2018 for the treatment of older patients with AML who are not eligible for intensive chemotherapy [152]. Azacitidine/venetoclax combination outperformed azacitidine alone in both de novo AML and AML-MRC based on overall survival (14.7 months vs. 9.6 months) and complete-remission rates (66.4% vs. 28.3%) [153]. AML with IDH1/2 mutations demonstrated the highest sensitivity to the combination of azacitidine/venetoclax compared to azacitidine alone. Despite a more prolonged neutropenia and higher incidence of neutropenic fevers with combination therapy, the rate of discontinuation was similar between the two groups [153]. Recent evidence also supports that azacitidine/venetoclax is an effective combination in treatment-naïve high-risk MDS [154,155]. Liu et al. analyzed 1057 patients with high-risk MDS and AML-MRC treated with HMAs and venetoclax [156]. Fifty-six (56%) patients had complete remission and the overall response rate was 68% [156]. Subgroup analysis revealed that the complete-remission rate was 61% among MDS patients, 68% among patients with newly diagnosed AML-MRC and 39% for relapsed/refractory AML-MRC [156]. The combination of azacitidine/venetoclax is being evaluated in a phase 3 clinical trial for treatment-naïve high-risk MDS patients (NCT04401748). The combination of azacitidine/venetoclax is also effective in relapsed/refractory MDS, as it outperformed azacitidine monotherapy and led to marrow response and complete-remission rates of 38.6% and 6.8%, respectively, within 1.5 months of treatment for approximately 9 months [157]. Of note, the response rate was stratified based on molecular profile rather than the conventional predicting tools (IPSS-R and blast percentage), with IDH2 and DNMT3A mutations showing the best response rate and TP53 mutations being associated with worse response rates. Venetoclax-based regimens appear to be promising approaches as a bridge to alloBMT. It has been reported that high-risk MDS and AML-MRC patients responding to venetoclax-based therapy and then receiving AlloBMT have a 1-year overall survival rate of 75–79% with a relapse rate of 16.7% within the first 12–14 months [158,159]. It remains unclear if azacitidine/venetoclax is a better approach compared to induction chemotherapy for fit patients with high-risk MDS and AML-MRC. Ongoing clinical trials focus on comparing induction chemotherapy to the azacitidine/venetoclax combination, with emphasis on the outcomes of patients with AML-MRC (NCT04801797). Low-dose cytarabine and induction chemotherapy have been evaluated for the management of patients with AML-MRC with responses of approximately 50–60% but a median survival of 6.5 months [160]. Prior treatment with HMAs or lenalidomide and longer time to transform to AML-MRC were associated with worse response [160]. A recent retrospective analysis showed high response rates of induction chemotherapy, followed by AlloBMT in MDS and MDS/MPN patients with NPM1 mutations [116], suggesting that molecular profile may be critical for the identification of MDS/AML-MRC patients that may benefit from induction therapy. Lancet et al. compared the activity of CPX-351, a dual-drug liposomal encapsulation of cytarabine and daunorubicin that delivers a synergistic 5:1 drug ratio into leukemia cells to a greater extent than normal bone marrow cells, to the activity of 7 + 3 in older individuals with AML-MRC [161]. CPX-351 demonstrated significantly improved remission rates and overall survival, despite a prolonged time to neutrophil and platelet engraftment [161]. This survival benefit of CPX-351 over 7 + 3 was maintained after 5 years of follow-up based on a subsequent study [162]. Overall, these results indicate that chemotherapy may be an option for fit individuals with high-risk MDS with elevated blasts or AML-MRC. As described above, the molecular profile of MDS and AML-MRC define the disease biology and affect its natural history independent of the disease stage at diagnosis. Consistently, treatment approaches for specific molecular subtypes are currently the focus of intense research. Biallelic TP53 alteration is one of the most established adverse genomic subtypes of high-risk MDS and AML-MRC and, as described above is associated with particularly poor outcomes. HMAs and HMAs/venetoclax combinations show some efficacy in TP53-mutated high-risk MDS and AML-MCR [153,163], but the incidence of relapse and long-term overall survival remains disappointing [164]. Similarly, it remains unclear if chemotherapy provides any significant benefit for TP53-mutated high-risk MDS and AML-MRC patients [91]. AlloBMT is the only approach that can prolong the survival of TP53-mutated MDS and AML-MRC patients [110]. However, TP53-mutated patients, compared with TP53 wild-type patients, showed a higher relapse rate after alloBMT [34]. The achievement of a lower burden of TP53 mutation before alloBMT has been associated with improved survival [34]. These results clearly support that TP53-mutated MDS and AML-MRC patients represent an unmet need. Recent studies have led to the introduction of two novel agents with promising efficacy for these patients. Eprenetapopt (ARP-246) is a small molecule that stabilizes when converted to methylene quinuclidinone, the misfolded P53 core domain, and restores the function of the mutated protein [165,166,167]. Pre-clinical studies supported that APR-246 has synergistic effects when combined with azacitidine, ara-c and daunorubicin in TP53-mutated AML cells [168,169]. Despite the encouraging findings of two phase I/II trials showing that the APR-246/azacitidine combination resulted in high response and complete-remission rates, as well as high rates of TP53-mutation clearance [170,171], the survival benefit of this combination over azacitidine alone for TP53-mutated high-risk MDS/AML-MRC has not been confirmed. However, recent data support that the APR-246/azacitidine combination has promising efficacy as a maintenance therapy following AlloBMT for TP53-mutated MDS and AML patients [142]. Finally, evaluation of the triplet combination azacitidine/venetoclax/APR-246 is ongoing in high-risk MDS and AML-MRC with adverse cytogenetic and molecular features [172]. Magrolimab is a monoclonal antibody that targets the CD47 surface marker of tumor cells, also known as the “don’t eat me” signal, promoting their phagocytosis by macrophages [173]. Among TP53-mutated MDS and AML-MRC patients, the combination of azacitidine/magrolimab showed a response rate of about 40%, which is similar to TP53-unmutated patients with anemia due to hemolysis being a well-described side effect [174]. Of note, the combination of magrolimab/azacitidine/venetoclax demonstrated complete-remission rates higher than 80% in newly diagnosed AML patients with ELN adverse risk, suggesting that this approach may become the new standard of care for high-risk MDS/AML-MRC patients who are ineligible for induction chemotherapy but can tolerate this triple therapy [175]. Magrolimab-based therapies are currently being compared to standard of care, such as azacitidine and azacitidine/venetoclax, in various ongoing phase three clinical trials (NCT05079230, NCT04313881, NCT04435691, NCT04778397, NCT04435691). IDH1/2 mutations are relatively uncommon in the chronic phase of MDS, but their incidence increases with the transformation to AML-MRC and are associated with overall poor outcomes [176,177]. The IDH1 inhibitor ivosidenib results in high complete-remission rates in patients with IDH1-mutated relapsed/refractory AML [178] and newly diagnosed AML [179]. Phase III clinical trials have shown that the addition of ivosidenib to azacitidine significantly increases remission rates and prolongs the survival of patients with IDH1-mutated AML, including AML-MRC [180]. This led to the FDA approval of this combination for patients with IDH1-mutated AML who are ineligible for induction chemotherapy. Enasidenib is an IDH2 inhibitor that leads to response rates of 30–50% in relapsed/refractory IDH2-mutated AML, including AML-MRC [181,182]. Enasidenib/azacitidine was evaluated for high-risk IDH2-mutated MDS, demonstrating a response rate of 74% and a complete remission of 26% in 27% of patients proceeding to alloBMT and survival after 2 years [182]. Finally, enasidenib has been evaluated as maintenance therapy for high-risk MDS and AML-MRC patients with IDH2 mutation with overall, progression-free and relapse rates being 74%, 69% and 16%, respectively, and relatively rare severe adverse effects [183]. Ongoing clinical trials evaluate the efficacy of IDH1/2 inhibitors in MDS and AML-MRC patients (NCT03503409, NCT04603001, NCT04493164, NCT03471260) and their combination with venetoclax, azacitidine and intensive chemotherapy (NCT03471260, NCT03839771). High-risk MDS is a clonal myeloid neoplasm characterized by genetic abnormalities in stem and progenitor hematopoietic cells with a high incidence of disease progression to a disease stage with a high percentage of blasts defined as AML-MRC or AML with myelodysplasia-related gene mutations or MDS/AML. Recent pre-clinical data support that genetic alterations, such as chromosomal abnormalities and somatic mutations, are detectable in hematopoietic cells even at the earlier stages of the disease and predict disease progression defining its natural history and biology. These results suggest that MDS and AML-MRC represent a continuum of the same disease and suggest that early-risk stratification and intervention may be a particularly promising approach for patients with MDS and a high risk of progression. These advances are reflected in the new changes in the classification of MDS and AML and are utilized in the evaluation of novel therapies.

PMC10002507

Eleonora Giagnorio,Claudia Malacarne,Paola Cavalcante,Letizia Scandiffio,Marco Cattaneo,Viviana Pensato,Cinzia Gellera,Nilo Riva,Angelo Quattrini,Eleonora Dalla Bella,Giuseppe Lauria,Renato Mantegazza,Silvia Bonanno,Stefania Marcuzzo

MiR-146a in ALS: Contribution to Early Peripheral Nerve Degeneration and Relevance as Disease Biomarker

27-02-2023

amyotrophic lateral sclerosis,microRNA-146a,axon degeneration,biomarker

Amyotrophic lateral sclerosis (ALS) is characterized by the progressive, irreversible loss of upper and lower motor neurons (UMNs, LMNs). MN axonal dysfunctions are emerging as relevant pathogenic events since the early ALS stages. However, the exact molecular mechanisms leading to MN axon degeneration in ALS still need to be clarified. MicroRNA (miRNA) dysregulation plays a critical role in the pathogenesis of neuromuscular diseases. These molecules represent promising biomarkers for these conditions since their expression in body fluids consistently reflects distinct pathophysiological states. Mir-146a has been reported to modulate the expression of the NFL gene, encoding the light chain of the neurofilament (NFL) protein, a recognized biomarker for ALS. Here, we analyzed miR-146a and Nfl expression in the sciatic nerve of G93A-SOD1 ALS mice during disease progression. The miRNA was also analyzed in the serum of affected mice and human patients, the last stratified relying on the predominant UMN or LMN clinical signs. We revealed a significant miR-146a increase and Nfl expression decrease in G93A-SOD1 peripheral nerve. In the serum of both ALS mice and human patients, the miRNA levels were reduced, discriminating UMN-predominant patients from the LMN ones. Our findings suggest a miR-146a contribution to peripheral axon impairment and its potential role as a diagnostic and prognostic biomarker for ALS.

MiR-146a in ALS: Contribution to Early Peripheral Nerve Degeneration and Relevance as Disease Biomarker Amyotrophic lateral sclerosis (ALS) is characterized by the progressive, irreversible loss of upper and lower motor neurons (UMNs, LMNs). MN axonal dysfunctions are emerging as relevant pathogenic events since the early ALS stages. However, the exact molecular mechanisms leading to MN axon degeneration in ALS still need to be clarified. MicroRNA (miRNA) dysregulation plays a critical role in the pathogenesis of neuromuscular diseases. These molecules represent promising biomarkers for these conditions since their expression in body fluids consistently reflects distinct pathophysiological states. Mir-146a has been reported to modulate the expression of the NFL gene, encoding the light chain of the neurofilament (NFL) protein, a recognized biomarker for ALS. Here, we analyzed miR-146a and Nfl expression in the sciatic nerve of G93A-SOD1 ALS mice during disease progression. The miRNA was also analyzed in the serum of affected mice and human patients, the last stratified relying on the predominant UMN or LMN clinical signs. We revealed a significant miR-146a increase and Nfl expression decrease in G93A-SOD1 peripheral nerve. In the serum of both ALS mice and human patients, the miRNA levels were reduced, discriminating UMN-predominant patients from the LMN ones. Our findings suggest a miR-146a contribution to peripheral axon impairment and its potential role as a diagnostic and prognostic biomarker for ALS. Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterized by the loss of motor neurons (MNs) in the motor cortex, brainstem, and spinal cord [1]. The ALS clinical spectrum includes extremely heterogeneous and complex phenotypes distinguished by a varying involvement of upper (UMN) and lower (LMN) MNs, site of onset, and rate of progression [2]. Of note, growing evidence demonstrated that terminal MN axonal degeneration, along with neuromuscular junction denervation, are early events in ALS pathogenic cascade [3,4]. Functional and morphological analysis showed early damage of the blood–nerve barrier followed by acute axonal degeneration associated with macrophage response in motor nerve compartments [3]. Progressive axonal degeneration and motor nerve fiber loss were found to correlate with magnetic resonance (MR) imaging and neurophysiologic changes in the ALS rat model [4]. By diffusion-tensor MR imaging, our previous study in the G93A-SOD1 mouse ALS model showed a gradient of degeneration in spinal cord white and gray matter, starting early in the ventral white matter, likely due to a cascade of early pathological events, including axonal dysfunction [5]. In the same model, straightforward evidence showed an early impairment of the peripheral nervous system compared to the central nervous system, with the presence of adaptive and innate immune cell infiltration [6]. In human patients, the relationship between motor axonal dysfunction at the early disease stage and disease progression was explored by compound muscle action potential amplitudes (CMAP), providing evidence that early peripheral motor axon dysfunction can influence ALS progression [7]. ALS can occur in two different forms: sporadic (sALS) in ∼90% of individuals and familial (fALS) [8]. Different genes have been associated with fALS and/or sALS: C9orf72–SMCR8 complex subunit (C9orf72) is the gene most commonly linked to inherited ALS, followed by TAR DNA-binding protein 43 (TARDBP), superoxide dismutase 1 (SOD1), and FUS RNA-binding protein (FUS) [9,10]. These genes affect several cellular functions, including RNA metabolism [11]. However, the exact molecular mechanisms implicated in ALS are not fully elucidated, and biomarkers of disease progression are not available yet. The identification of molecular biomarkers associated with disease course represents an important medical need since they could significantly improve the diagnostic process with relevant prognostic implications, aiding in patients’ stratification into distinct phenotypes and, where appropriate, in enrolling patients with a specific molecular signature in clinical trials [12,13]. To our knowledge, only the neurofilament light chain (NFL) protein, encoded by the neurofilament light chain (NFL) gene, has been identified as a promising disease biomarker. Indeed, CSF and serum NFL levels were found to be elevated in ALS patients [14,15,16] due to NFL release by the axonal plasma membrane as a consequence of axonal damage or degeneration [16]. Notably, NFL in CSF can differentiate ALS subgroups, in particular, bulbar compared to spinal onset ALS patients, as well as patients with SOD compared to C9orf72 mutations, with C9orf72-ALS patients being characterized by higher plasma levels of NFL than SOD-ALS patients [17]. In a more recent study, the NFL levels were higher in patients with UMN predominance [14]; however, the exact molecular events underlying this increase in relationship with axonal degeneration are not entirely clear. Growing evidence suggests that RNA metabolism alterations are critical for ALS pathogenesis [18,19]. MicroRNAs (miRNAs) are small non-coding RNAs that are key determinants of mRNA stability [20]. Several studies indicated a pivotal role for miRNAs in all aspects of neuronal development, function, and plasticity [21,22]. Moreover, the contribution of miRNA dysregulation to MN diseases, including ALS, is increasingly emerging [23,24,25,26,27]. Among miRNAs, miR-146a is known to negatively regulate the expression of the NFL gene [28], thus suggesting its potential involvement in ALS via NFL modulation. In particular, the role of miR-146a in the selective decrease in NFL mRNA and the formation of neurofilamentous aggregates in ALS has been described [28]. Furthermore, the elevated expression of miR-146a has been suggested to play a role in the inhibition of the production of inflammatory cytokines by antagonization of NF-κB activation and in the reduction of muscle mass, thus representing a molecular link between muscle atrophy and inflammation [29,30]. On the other hand, miR-146a deficiency has been associated with astrocyte and microglial cell transformation into the neurotoxic and proinflammatory phenotypes, which might contribute to MN degeneration, thus supporting the idea that miR-146a overexpression may exert a protective effect in ALS [31]. However, since miR-146a up-regulation in ALS patients’ spinal cords was associated with decreased NFL proteins that contribute to the maintenance of neuronal morphology [28], a side effect of miR-146a overexpression can be postulated. Nevertheless, whether miR-146a participates in early peripheral nervous system impairment, and specifically, in NFL dysfunction and axon degeneration at the peripheral nervous system in ALS, still needs to be elucidated. Herein, we analyzed the expression of miR-146a in sciatic nerve and serum of G93A-SOD1 ALS mice at different disease stages by molecular analyses and in situ hybridization method. Moreover, to corroborate the potential role of this molecule as a non-invasive biomarker, we analyzed its expression in serum samples of ALS patients with signs of predominant axonal damage in UMNs and LMNs. We found an up-regulation of miR-146a in the sciatic nerve of affected mice that negatively correlated with transcriptional levels of the Nfl target gene, suggesting an association between the miRNA dysregulation and selective decrease in NFL in the sciatic nerve that could contribute to early axon degeneration in the peripheral nervous system. Contrariwise, the miRNA expression was down-regulated in the serum of affected mice, as well as in serum samples of human patients compared to healthy controls. These data suggested the involvement of miR-146a in human ALS, and its possible role, to be further explored as a therapeutic target to modulate the disease progression. Of note, serum miR-146a down-regulation was more prominent in patients with UMN than LMN predominant impairment, and the miRNA expression was significantly different between the two groups of patients, thus suggesting miR-146a potential value as a non-invasive biomarker to stratify ALS patients according to the disease phenotype. To verify whether miR-146a expression was altered in the sciatic nerve tissue of G93A-SOD1 mice, we performed a longitudinal analysis of its expression levels by real-time PCR in control B6.SJL and affected G93A-SOD1 animals at the following disease stages: pre-symptomatic (week 8), onset (week 12), and symptomatic phase (week 18). Interestingly, our data revealed a significant up-regulation of miR-146a in G93A-SOD1 mice compared to controls (p < 0.05) already at week 8 (Figure 1a). MiRNA-146a levels were higher also at week 12, and particularly at week 18 (p < 0.05), in affected than control mice (Figure 1a). These results suggested that miR-146a overexpression in the sciatic nerve may be an early phenomenon in ALS, appearing before disease symptoms, then becoming evident and persisting during the disease course. To investigate possible alterations in Nfl expression in the sciatic nerve tissue of ALS animals, along with those of miR-146a, we performed real-time PCR analysis. We found that Nfl mRNA levels were significantly down-regulated in G93A-SOD1 compared to control animals at weeks 8 and 18 (Figure 1b, p < 0.05). These levels showed a trend to be also decreased at week 12 in affected versus control mice (Figure 1b), although differences did not reach statistical significance. Using Spearman’s correlation analysis, we assessed the relationship between miR-146a and Nfl expression in the sciatic nerve tissue of G93A-SOD1 mice. We found a significant negative correlation between the miR-146a levels and those of Nfl mRNA at the different time points (Figure 1c), suggesting a link between miR-146a increase in ALS sciatic nerve and Nfl reduction, which could contribute to axon degeneration. To investigate the relationship between miR-146a-5p and the Nfl gene, we performed functional studies in NSC-34 motor neuron-like cells. By transfection with an miR-146a-5p inhibitor in NSC-34 motor neuron-like cells, we observed a significant up-regulation of the expression levels of the Nfl gene in miR-146a-5p inhibitor-transfected NSC-34 cells compared to scrambled-transfected cells (Figure 2, p < 0.05). We performed in situ hybridization to detect and localize miR-146a in the sciatic nerve tissue of G93A-SOD1 and control mice at weeks 8, 12, and 18. According to real-time PCR data, we showed a marked expression of miR-146a in the sciatic nerve of G93A-SOD1 compared to control mice already at week 8 and until week 18 (Figure 3a). The immunofluorescence intensity of U6 snRNA, examined as endogenous control, did not show a difference between the two groups of animals (Figure 3b). The scramble probe, a negative technical control, did not display any signals of fluorescence (Figure 3b). Increased expression of miR-146a in affected versus control animals was confirmed by quantification analysis of the miRNA immunofluorescence intensity normalized to that of U6 from weeks 8 to 18 (Figure 3c). NFL protein levels showed a trend to be decreased in the sciatic nerve of G93A-SOD1 compared to control mice at weak 18, by Western blot analysis, as shown by quantification analysis of band intensity (Figure 4a,b, Supplemental Figure S1). At weeks 8 and 12, we did not observe a decrease in the expression of NFL protein; these results are not in line with those of mRNA analysis, showing a decrease already at these weeks, which could be explained by the exceptional stability of NFL protein (approximately 3 weeks in mice optical axons) [32]. To verify whether dysregulated expression of miR-146a in the sciatic nerve was accompanied by alterations of the circulating miRNA, we assessed its levels in the serum of G93A-SOD1 mice at the symptomatic disease stage, i.e., week 18, compared to control mice. We found that miR-146a levels were significantly down-regulated in sera of ALS compared to healthy mice (Figure 5), suggesting that the miRNA overexpression was a specific feature of the sciatic nerve and that it was associated with a reduction at the serum level. Based on these findings, we wondered whether miR-146a levels may also be altered in ALS patients and whether the miRNA may represent a useful, non-invasive molecular biomarker in the disease. We thus assessed miR-146a expression in the serum of ALS patients and age- and sex-matched healthy donors (Table 1). As observed in affected animals, we found a significant down-regulation of miR-146a levels in ALS compared to control sera (Figure 6). This down-regulation was more evident in ALS patients with UMN compared to patients with LMN predominance, where the two groups did not differ in terms of disease duration at sampling (p = 0.797) (Figure 6a). By receiver operating characteristic (ROC) curve analyses, we obtained sensitivity and specificity diagnostic performance results, which supported a possible role for serum miR-146a as a disease biomarker for ALS, particularly in discriminating ALS patients with UMN from those with LMN predominance (Figure 6b). These findings suggest the potential role of miR-146a, to be validated in larger patients’ cohorts, as a potential biomarker for stratifying ALS phenotypes. The ALS clinical spectrum includes extremely heterogeneous phenotypes marked by a varying involvement of UMN and LMN, site of onset, and rate of progression [1]. The specific molecular determinants and potential biomarkers of disease progression remain to be fully elucidated. Thus far, only NFLs have been identified as promising disease biomarkers in ALS, reflecting axonal damage and degeneration [14,33]. Nevertheless, the exact molecular mechanisms linking NFL alterations and MN axon damage during the disease course are still unknown. Several data suggest that axonal degeneration already occurs at the early stages of ALS [34], supporting the need for a deeper understanding of the molecular changes implicated in early pathogenic events to develop efficient therapeutic approaches to counteract disease progression. Most aspects of neuronal function and plasticity are regulated by miRNAs [22]. Among them, miR-146a is a crucial molecule for axon architecture and function [35]. Of note, this miRNA is known to modulate the expression of the NFL gene, which is a key gene for maintaining neuronal morphology in the spinal cord [28,36]. Based on this relevant function of miR-146a, in the present study, we analyzed its expression in the sciatic nerve of ALS mice during disease progression compared to control animals. Interestingly, we obtained real-time PCR data indicative of a significant up-regulation of miR-146a in the sciatic nerve of G93A-SOD1 ALS compared to control mice. This increase was already significant at the pre-symptomatic disease phase (week 8) and was maintained until the symptomatic phase (week 18). In situ hybridization performed in sciatic nerve sections of affected and control mice, confirmed a marked overexpression of the miRNA in the G93A-SOD1 mouse compared to the control group at early week 8 of life and throughout the disease course (week 12 and 18). We, therefore, suggested that early miR-146a dysregulation at the peripheral axonal level might be of relevance in the ALS pathogenic events, possibly contributing to the initial stages and perpetuation of axon degeneration. To deepen this hypothesis, we verified whether the miR-146a increase in ALS sciatic nerve was associated with dysregulated expression of genes implicated in axon function and structure. We focused on the NFL gene because it encodes for the light chain NFL protein, which is a crucial factor for the neuronal structure. Indeed, along with medium and heavy chains, the light chain is a component of NFLs, type IV intermediate filament heteropolymers that encompass the axoskeleton and functionally maintain the neuronal caliber, also playing a role in intracellular transport to axons and dendrites [37]. The ability to target the Nfl gene has been specifically demonstrated for miR-146-3p, known as miR-146a* [28]. Here, we showed a significant increase in miR-146a-5p in ALS mice, which could well be associated with decreased expression of Nfl since 5p/3p types of the same miRNA have been found to be co-expressed and co-target the same transcripts [38]. Moreover, the interaction between miRNA-5p and -3p with mRNAs have been shown to be fully complementary, as binding characteristics of different miRNA-5p/3p pairs in complementary binding sites of the same genes have been established [39]. Of note, Nfl mRNA levels showed an opposite trend than those of miR-146a, being significantly down-regulated in the sciatic nerve of ALS compared to control mice, already at week 8. In affected animals, reduced Nfl expression was maintained at week 12 and was marked at week 18. A significant negative correlation was found between miR-146a levels and the transcriptional levels of the Nfl gene, thus suggesting that decreased Nfl gene expression in the sciatic nerve of G93A-SOD1 mice might be related to abnormal miR-146a up-regulation at the early disease stage and during disease progression. Our functional studies in NSC-34 motor neuron-like cells by transfection with miR-146a inhibitor demonstrated a relationship between miR-146a-5p and Nfl gene expression, as observed for miR-146a-3p, suggesting that Nfl changes may be an effect of miR-146a-5p dysregulation in ALS mice. However, further functional studies are needed to clarify the exact relationship between miR-146a-5p and Nfl gene expression and whether this gene is a direct target of the -5p miRNA type, as observed for miR-146a-3p. By biochemical analysis, we confirmed the lower expression of the NFL protein in the sciatic nerve of ALS mice at the symptomatic phase of the disease. These findings are in line with literature data showing that high levels of NFL protein in serum and/or CSF of patients with neurodegenerative diseases are not due to NFL overexpression (or over-production of its encoded protein) but to flaking and breaking of axons and subsequent release of NFL protein in biological fluids [40]. Our molecular data reveal miR-146a overexpression as an early molecular alteration occurring in ALS peripheral nerve, potentially implicated in axon impairment in association with reduction of Nfl gene expression and light chain NFL protein synthesis. Our findings open the hypothesis, to be deeply explored, that inhibiting the expression of miR-146a could directly or indirectly affect Nfl expression, potentially favoring motor axon re-organization and recovering neuronal integrity. Since miRNAs are stable in body fluids and may reflect specific pathophysiological states or altered molecular mechanisms, they represent promising biomarkers [41,42]. These molecules can be released into the circulation by pathologically affected tissues and display remarkable stability in body fluids [43]. In the ALS field, the identification of miRNAs as disease biomarkers would be relevant to improve the diagnostic process and prognostic potential. Indeed, miRNAs could allow a more accurate diagnosis, giving the opportunity to start an earlier treatment with higher chances to modify the disease course. In addition, they could help in monitoring the disease progression and in the stratification of ALS patients in distinct pathological phenotypes to drive proper patients’ enrollment in clinical trials and, prospectively, for tailored therapies. MiRNAs may represent a link between the results obtained in animal models and human patients [44]. To explore the hypothesis of miR-146a as biomarkers for ALS and to translate our results to human patients, we first verified whether miR-146a up-regulation in ALS mouse sciatic nerve may be associated with changes in the serum miRNA levels. Of interest, we observed a significant decrease in the miRNA in affected mice compared to controls at the symptomatic disease stage, thus implying an enrichment of miR-146a in the sciatic nerve but not in the circulation. We, thus, investigated miR-146a levels in ALS patients’ serum samples and obtained molecular results indicative of a significant down-regulation of the miRNA compared to healthy controls, in line with the animal model data. Down-regulation of miR-146a was more evident in the serum of ALS patients with UMN predominance compared to LMN patients. By ROC curve analysis, we revealed a potential role of miR-146a as a biomarker for ALS. Indeed, we obtained sensitivity and specificity results indicative of the ability of serum miR-146a levels to discriminate between ALS patients and controls and between patients with UMN or LMN predominance. These findings strongly suggest that miR-146a might be implicated in human ALS pathogenic events and could participate in axon degeneration, as observed in the G93A-SOD1 model. Serum NFL levels were found to be elevated in ALS patients and to mirror the severity of axonal degeneration, corroborating their value as disease biomarkers [14,15,16]. Similarly, miR-146a might represent a circulating biomarker for the disease, whose levels in serum follow an opposite trend to that of NFL, likely in view of a negative miR-146a/NFL correlation, as observed in our study. Reduced levels of the miRNA in ALS serum could be explained by its enrichment in the peripheral nerve with a lack of release in the blood. However, this issue needs to be deeply explored in a further study, taking into account the complexity of the regulatory mechanisms involving miRNAs and their release in the serum and additional sites in which miR-146a expression could be altered to influence its levels in circulation. Nevertheless, our serum data, and their consistency across the animal model and human patients, strongly support a role for miR-146a as a non-invasive molecular biomarker in ALS, particularly for stratifying ALS patients according to the pattern of MN involvement underlying the clinical phenotype. The discovery of a possible miR-146a contribution to peripheral axon damage could have relevant therapeutic implications. Considering that MN axon degeneration is an early phenomenon in ALS [5,7] and that miR-146a up-regulation is also an early event, an advanced molecular approach specifically targeting this miRNA could represent an innovative future method to be tested and developed for counteracting the peripheral degenerative cascade of events occurring in the disease. Greater understanding in this research area could have relevant implications for ALS treatment. All animal experiments were carried out in accordance with the EU Directive 2010/63 and with Italian law (D.L. 26/2014). Transgenic G93A-SOD1 (B6SJL-Tg (SOD1*G93A)1Gur/J) and control B6.SJL mice were purchased from Charles River Laboratories, Inc. (Wilmington, MA, USA), maintained, and bred at the animal house of the Fondazione IRCCS Istituto Neurologico Carlo Besta. The project was approved by the Ethics Committee of the Institute and the Italian Ministry of Health (ref. 03/2018, 78/2022-PR). Transgenic G93A-SOD1 progenies were identified by quantitative real-time PCR amplification of the mutant human SOD1 gene as previously described [45]. They were sacrificed for tissue collection by exposure to CO2 at week 8 (pre-symptomatic stages of disease), week 12 (onset of disease), and week 18 (late stage of disease) [43]. NSC-34 motor neuron-like cells are hybrid cells produced by the fusion of motor neurons of the spinal cords of mouse embryos with mouse neuroblastoma cells N18TG2 [46]. NSC-34 cells were maintained in DMEM high glucose with 10% fetal bovine serum (FBS) (Thermo Fisher Scientific, Waltham, MA, USA), 1% glutamine, 1% sodium pyruvate, and 1% penicillin-streptomycin (P/S) (Euroclone, Milan, Italy), in a humidified atmosphere and 5% CO2 at 37 °C (hereafter referred to as standard culture conditions). NSC-34 was grown for a maximum of 20 passages. To induce NSC-34 differentiation, cells were plated onto Matrigel-coated plates (Corning, Glendale, AZ, USA) in differentiating medium composed of DMEM/F12, 1% FBS (Thermo Fisher Scientific), 1% P/S, 1% sodium pyruvate, 1% MEM non-essential amino acids (Euroclone), supplemented with 1 μM all-trans RA (Sigma-Aldrich, St. Louis, MO, USA) for 12 days, as previously described [47]. The medium was replaced every 2 days. A total of 4 × 104 NSC-34 were seeded into Matrigel-coated 6-well plates and differentiated for 12 days. miR-146a-5p mirVana miRNA inhibitor (IDMH10722, Thermo Fisher Scientific) and Silencer Select Negative Control (Thermo Fisher Scientific), as scrambled miRNA, were allowed to form transfection complexes with Lipofectamine RNAiMax transfection reagent (Thermo Fisher Scientific) in Opti-MEM I Reduced Serum Medium (Thermo Fisher Scientific) at a final concentration of 30 nM for 15 min at room temperature. After 24 h, the medium was replaced with NSC-34 differentiation medium; after 48 h, samples from the 6-well plates were collected for RNA extraction. Three replicates per condition were performed. Total RNA was extracted with Trizol reagent from sciatic nerve tissues (3–4 mg). Sciatic nerve tissues were maintained at −80 °C until use. Total RNA was reverse-transcribed to cDNA using the TaqMan microRNA Reverse Transcription Kit (Thermo Fisher Scientific) with specific primers for miR-146a-5p (termed miR-146a throughout the article). cDNA aliquots corresponding to 15 ng total RNA were amplified by quantitative real-time PCR in duplicate, with Universal PCR Master Mix and specific pre-designed TaqMan microRNA assays (Thermo Fisher Scientific). As endogenous control for data normalization, we used U6 snRNA, which was stably expressed in the G93A-SOD1, and control sciatic nerve tissues. MiR-146a levels were expressed as relative values normalized toward U6, according to the following formula 2−ΔCt × 100. Sciatic nerves were collected from G93A-SOD1 and B6.SJL mice at weeks 8, 12, and 18. Sciatic nerves were immediately frozen in isopentane, pre-cooled in liquid nitrogen, and stored at −80 °C. Frozen tissues were then cut into 10 μm thick sections and stored at −80 °C until usage. Sciatic nerve sections were then fixed with paraformaldehyde 4% and then permeabilized with cold methanol (Merck, Darmstadt, Germany). Slides were then washed with Saline-sodium citrate (SSC) buffer, composed of 3 M NaCl and 0.3 M sodium citrate (Carlo Erba Reagents, Milan, Italy) in distilled water and with PBS. Sciatic nerve slices were then covered with miRCURY LNA miRNA Detection Probes 5′-fluorescein and 3′-fluorescein labeled (Qiagen, Hilden, Germany), specific for miR-146a-5p (80 nM, sequence UGAGAACUGAAUUCCAUGGGUU), U6 (20 nM) and scramble (80 nM), resuspended in situ hybridization (ISH) buffer (Qiagen) overnight at 37 °C. U6 was used as endogenous control, and scramble as negative technical control. The slides were then washed with ISH buffer and PBS and then blocked with a BSA solution (5%) (Thermo Fisher Scientific) at room temperature for 1 h. Next, 4,6-diamidino-2-phenylin-dole (DAPI) (1:1000) (Thermo Fisher Scientific) was applied for 5 min to stain cell nuclei. The slices were then mounted with FluorSave (Merck) and air dry overnight. Images were acquired using the C1 laser scanning confocal microscope system (Nikon, Minato City, Tokyo, Japan) and analyzed using Image J software (version 1.8.0_172). Quantification of the immunofluorescence intensity of miR-146a in sciatic nerve tissues was calculated by correcting for background and normalizing to U6 immunofluorescence intensity as endogenous control, using Image J (version 1.8.0_172). Total RNA was extracted with Trizol reagent from NSC-34-treated cells (4 × 104 cells per sample). For neurofilament light chain (Nfl) gene expression analysis, total RNA from NSC-34-treated cells, and the total RNA previously analyzed for miR-146a from mouse sciatic nerve tissue was retro-transcribed using SuperScript VILO cDNA Synthesis kit (Thermo Fisher Scientific). cDNA (10 ng) was amplified by real-time PCR in duplicate, with TaqMan Fast Advanced Master Mix and TaqMan gene expression assays, specific for Nfl on the ViiA7 Real-time PCR system (Thermo Fisher Scientific). The 18S was stably expressed in the G93A-SOD1 and control sciatic nerve tissue and in the NSC-34 cell line and used as endogenous control. Transcriptional levels of the Nfl gene were expressed as relative values normalized toward 18S levels, according to the following formula 2−ΔCt × 100. To obtain total proteins, sciatic nerves were homogenized in a lysis buffer composed of distilled water supplemented with sodium chloride (NaCl) (150 mM) (Merck), Nonidet P-40 (1%) (Merck), sodium deoxycholate (0.5%) (Merk), sodium dodecyl sulfate (SDS) (0.1%) (Merck), Tris hydrochloride (50 mM, pH 8.0) (Merck), and Halt Protease and Phosphatase Inhibitor Cocktail (100×) (Thermo Fisher Scientific), using Tissue Lyser LT and stainless steel beads (Qiagen) for 3 min at 50 Hz. Extracts were incubated on ice for 30 min and then centrifuged for 20 min at 14,000 rpm at 4 °C to remove particulate matter. Supernatant protein concentration was determined by the Bradford method (Coomassie Plus Assay Kit, Thermo Fisher Scientific ). Western blot analysis was performed on NuPAGE4–12%, Bis-Tris, 1.5 mm, Mini Protein Gels (Thermo Fisher Scientific), loading 30 µg of total proteins, previously denatured at 70 °C for 10 min. Samples were then electrotransferred to iBlot™ 2 Transfer Stacks, PVDF, mini (Thermo Fisher Scientific) using the iBlot 2 Dry Blotting System (Thermo Fisher Scientific). Membranes were treated with a blocking solution containing 5% skim milk powder in Tris-buffered saline with 0.1% Tween 20 (TBS-T) for 1 h at room temperature. Membranes were then incubated with the primary antibodies: anti-β-actin, rabbit polyclonal (dilution 1:3000, ab8227, Abcam, Cambridge, U.K.); anti-NFL, mouse monoclonal dilution (1:1000, MA1-2010, Thermo Fisher Scientific) overnight at 4 °C. Immunoreactivity was detected using secondary Western blot fluorescent antibodies: goat anti-rabbit red (LI-COR Biosciences, NE, USA, dilution 1:10,000) was used to identify β-actin; goat anti-mouse green (LI-COR Biosciences, dilution 1:10,000) was used to identify NFL. Immunoreactive NFL bands were visualized by Odyssey Infrared Imaging System (LI-COR Biosciences) and quantified using the Image Lab software (Bio-Rad Laboratories, Hercules, CA, USA). Total RNA was extracted with the miRNeasy Serum/Plasma Kit (Qiagen) from mouse serum. Mouse blood was collected in serum separation tubes, and sera were isolated by incubating the whole blood for 30 min at room temperature and then centrifuging it for 10 min at 3000 rpm at 4 °C. The supernatant was collected as serum. Sera were maintained at –80 °C until use. Total RNA was reverse-transcribed to cDNA using TaqMan microRNA Reverse Transcription Kit (Thermo Fisher Scientific) with specific primers for miR-146a-5p (termed miR-146a throughout the article), and cDNA aliquots were amplified by real-time PCR in duplicate, as described above. As endogenous control for data normalization, we used miR-24, which was stably expressed in G93A-SOD1, and control sera. Levels of miR-146a were expressed as relative values normalized toward miR-24, according to the following formula 2−ΔCt. A cohort of 24 clinically defined ALS patients was enrolled in the study, including 9 ALS with UMN predominance; 8 ALS with LMN predominance; 7 ALS with UMN + LMN predominance (OMIM: #105400 Amyotrophic Lateral Sclerosis, ALS1) followed-up at Neurology III Unit, and genetically assessed at Unit of Medical Genetics and Neurogenetics at Fondazione IRCCS Istituto Neurologico Carlo Besta (Milan, Italy). Twenty-three sex- and age-matched healthy controls were included in the analyses. Patient clinical features are reported in Table 1. The study was performed in accordance with the standards of The Code of Ethics of the World Medical Association (Declaration of Helsinki). The investigation was approved by the Ethics Committee of Fondazione IRCCS Istituto Neurologico Carlo Besta (project identification code 92/2019: date January 2019–January 2022). Written informed consent was obtained from each subject. Biological samples were stored at −80 °C in the Biobanks of Fondazione IRCCS Istituto Neurologico Carlo Besta until use. Total RNA was extracted with miRNeasy Serum/Plasma Kit (Qiagen) from human serum. Human blood was collected in serum separation tubes, and sera were isolated by incubating the whole blood for 30 min at room temperature and then centrifuging it for 10 min at 300 rmp at 4 °C. The supernatant was collected as serum. Serums were maintained at −80 °C until use. Expression analysis of miR-146a, and that of miR-24 endogenous control, were performed using the same protocol described for miR-146a analysis in mouse serum, but with primers and TaqMan probe (Thermo Fisher Scientific) specific to humans. The nonparametric distributed data, verified via Shapiro–Wilk test, were analyzed by Mann–Whitney test for comparison of two groups as indicated in the figure legends. p-values < 0.05 were considered statistically significant. A nonparametric Spearman correlation test was applied to evaluate the correlation between expression levels of the miRNA and those of its Nfl target gene in the sciatic nerve of G93A-SOD1 mice during disease progression. ROC curves were used to assess the sensitivity and specificity of miR-146a in human serum samples as a biomarker able to discriminate between ALS patients and healthy controls and among UMN, LMN, and UMN + LMN patients. GraphPad Prism version 4.0 (GraphPad Software, San Diego, CA, USA) was used for data elaboration and statistical analysis. Our findings suggest that miR-146a may be implicated in early peripheral nerve degeneration in ALS by lowering the levels of NFL. Thus, it could represent a novel target for future molecular strategies for treating the disease or delaying its progression. In addition, we reveal that, along with NFLs; in addition, miR-146a may be a suitable non-invasive biomarker of MN degeneration in ALS. In particular, after confirmation of our data in larger patient cohorts, it could be applied to the clinical practice as a biomarker for patients’ stratification according to UMN or LMN predominance and clinical phenotypes.

PMC10002509

Byeong Hee Kang,Woon Ji Kim,Sreeparna Chowdhury,Chang Yeok Moon,Sehee Kang,Seong-Hoon Kim,Sung-Hwan Jo,Tae-Hwan Jun,Kyung Do Kim,Bo-Keun Ha

Transcriptome Analysis of Differentially Expressed Genes Associated with Salt Stress in Cowpea (Vigna unguiculata L.) during the Early Vegetative Stage

01-03-2023

cowpea,salt-stress,NGS,RNA sequencing,reference sequencing

Cowpea (Vigna unguiculata (L.), 2n = 22) is a tropical crop grown in arid and semiarid regions that is tolerant to abiotic stresses such as heat and drought. However, in these regions, salt in the soil is generally not eluted by rainwater, leading to salt stress for a variety of plant species. This study was conducted to identify genes related to salt stress using the comparative transcriptome analysis of cowpea germplasms with contrasting salt tolerance. Using the Illumina Novaseq 6000 platform, 1.1 billion high-quality short reads, with a total length of over 98.6 billion bp, were obtained from four cowpea germplasms. Of the differentially expressed genes identified for each salt tolerance type following RNA sequencing, 27 were shown to exhibit significant expression levels. These candidate genes were subsequently narrowed down using reference-sequencing analysis, and two salt stress-related genes (Vigun_02G076100 and Vigun_08G125100) with single-nucleotide polymorphism (SNP) variation were selected. Of the five SNPs identified in Vigun_02G076100, one that caused significant amino acid variation was identified, while all nucleotide variations in Vigun_08G125100 was classified as missing in the salt-resistant germplasms. The candidate genes and their variation, identified in this study provide, useful information for the development of molecular markers for cowpea breeding programs.

Transcriptome Analysis of Differentially Expressed Genes Associated with Salt Stress in Cowpea (Vigna unguiculata L.) during the Early Vegetative Stage Cowpea (Vigna unguiculata (L.), 2n = 22) is a tropical crop grown in arid and semiarid regions that is tolerant to abiotic stresses such as heat and drought. However, in these regions, salt in the soil is generally not eluted by rainwater, leading to salt stress for a variety of plant species. This study was conducted to identify genes related to salt stress using the comparative transcriptome analysis of cowpea germplasms with contrasting salt tolerance. Using the Illumina Novaseq 6000 platform, 1.1 billion high-quality short reads, with a total length of over 98.6 billion bp, were obtained from four cowpea germplasms. Of the differentially expressed genes identified for each salt tolerance type following RNA sequencing, 27 were shown to exhibit significant expression levels. These candidate genes were subsequently narrowed down using reference-sequencing analysis, and two salt stress-related genes (Vigun_02G076100 and Vigun_08G125100) with single-nucleotide polymorphism (SNP) variation were selected. Of the five SNPs identified in Vigun_02G076100, one that caused significant amino acid variation was identified, while all nucleotide variations in Vigun_08G125100 was classified as missing in the salt-resistant germplasms. The candidate genes and their variation, identified in this study provide, useful information for the development of molecular markers for cowpea breeding programs. Cowpea (Vigna unguiculata (L.) Walp.; 2n = 2x = 22) is a tropical herbaceous crop that has adapted to various abiotic stresses, including drought and heat stress [1,2]. Globally, the estimated area of cowpea cultivation is about 15 million hectares, with more than 8.8 million tons being produced annually. The whole of Africa occupies more than 95% of this cultivated area, especially the arid and semiarid regions of West Africa, which can be identified as the main cultivation areas for cowpea [3]. However, salt compounds in the soil of arid and semi-arid regions are generally not eluted due to the low frequency of rainfall [4], and the resulting accumulation of salt in the soil can increase the salt stress for cowpea and other important crops. This problem has been exacerbated by climate change, which has increased the rate of desertification and created larger areas of arid and semiarid land in West Africa [5,6], potentially reducing the yield and quality of crops, including cowpea in the region. Salt stress causes various types of damage at all stages of a plant’s life cycle, from germination to seed production [7,8,9], with the proportion of cropland subject to salt damage reported to be increasing worldwide [10,11]. Therefore, understanding the effects of salt stress and the mechanisms associated with it is important from the perspective of meeting food demand in the future. Plants exposed to high soil salinity generally experience high osmotic and ionic stress, which affects a range of complex physiochemical processes [12,13,14,15]. The salinity reduces the leaf water potential and turgor pressure, leading to osmotic stress that induces abscisic acid biosynthesis, which in turn causes stomatal closure [16]. As a result, photosynthesis is reduced and oxidative stress increases [17]. In addition, excessive salinity around the roots can lead to ion toxicity, which increases the reactive oxygen species (ROS) levels, resulting in nutritional imbalances and damage to the cell structure [18]. This form of ion toxicity is commonly observed with sodium and chloride ions, which accumulate in highly saline soils [19]. One way to address high soil salinity is to create more salt-tolerant crop germplasms. However, salt stress is a complex process, and there are varying degrees of tolerance, both between and within species [20,21]. It has even been found that the response to salt stress can differ depending on the time of exposure and the stage of plant growth, with more rapid exposure resulting in more stress [22]. One study has reported that the difference in germination rates within a particular species ranged from 5.8% to 94.2% [23]. Within a particular species, individual plants with salt-sensitive genotypes tend to exhibit greater ion accumulation than salt-resistant genotypes do, leading to toxic effects [24]. These results suggest that salt tolerance is an independently evolved trait that can arise from completely unrelated mechanisms. This means that the genes associated with salt tolerance found in genetically distant species may not be effective if transplanted into germplasms of cowpea. However, genetic diversity within crops can be used to create germplasms with ideal traits, including salt tolerance [25]. Thus, further research is needed to identify genes related to salt stress in cowpea and to understand the mechanisms underlying their variation. In particular, it is important to understand plant ion homeostasis, osmotic responses, and oxidative stress in relation to increases in soil salinity. Recently, the development of high-throughput sequencing technologies, such as next-generation sequencing (NGS), has made it possible to better understand plant genomes, which is essential for understanding complex traits, including those associated with salt stress [26]. For example, NGS-based RNA sequencing (RNA-seq) makes it possible to identify differentially expressed genes (DEGs) across the genome and analyze stress-related metabolic pathways via the functional annotation of the identified DEGs [27,28]. This approach provides an opportunity to search for candidate genes involved in the stress response of crops under salt stress, including the detection of rare transcripts, thus revealing the function and pathway of genes related to salt tolerance [29,30,31,32]. Though it is challenging to identify target genes from among the numerous DEGs generated via RNA-seq, reference genome information can be used to narrow down the range of candidate genes. As an example of this, kompetitive allele-specific PCR (KASP) genotyping assays have been widely used for SNP allele scoring and indel discrimination with various crops in a way that makes use of allele-specific primers [33]. For example, DEGs have been identified using RNA-seq in rices that differed in their salt tolerance, and SNP variation in the identified DEGs was then successfully used for KASP marker development [34]. These KASP markers can subsequently be employed for plant breeding through marker-assisted selection (MAS). In the present study, we analyzed the expression patterns of genes associated with salt stress using cowpea germplasms with different levels of salt tolerance. We also conducted transcriptome profiling based on the fact that a plant absorbs salt from the roots, and that the damage is most severe at the seedling stage. This study thus aims to analyze the genetic network and related metabolic pathways for cowpea DEGs associated with salt tolerance. In this study, four cowpea germplasms with different levels of salt tolerance were exposed to a 250 mM NaCl solution for three weeks to simulate salt stress (Figure 1 and Figure 2). Ion accumulation was generally higher in the salt-sensitive germplasms compared with the salt-resistant plants. In particular, the sodium and chloride ion levels for the salt-resistant germplasms (Vu_191 and Vu_328) were 27.65 mg/g and 82.12 mg/g, respectively, compared with 51.86 mg/g and 139.35 mg/g, respectively, for the salt-sensitive germplasms (Vu_393 and Vu_396). A total of 24 library samples were obtained for sequence processing, consisting of control (0 h) and NaCl treatments (24 h) for each of the four germplasms, with three replicates each. These library samples were sequenced using the Illumina Novaseq 6000 platform (Table S1). RNA-seq analysis showed that the total number of clean reads generated for each sample was 1,144,868,572 (average length 101 bp). To obtain high-quality transcriptome short reads, bases with a Phred score (Q) of less than 20 were trimmed, and those trimmed reads with a length of less than 25 bp were eliminated. The total number of filtered reads was 1,107,552,070, with an overall average of 85.31% passing through the preprocessing stage, of which 1,038,301,592 (93.81%) were uniquely mapped to the cowpea reference genome sequence (Vunguiculata_540_v1.2). Of the 31,948 standard genes used for analysis, 27,559 had expression values and 25,476 had functional descriptions. DEGs were screened using DESeq2 software based on a false discovery rate (FDR) of ≤0.01 and absolute values for the log2fold change (FC) of >1, with up-regulation defined as log2FC > 1 and down-regulation as log2FC < −1. The gene expression profiles of the cowpea germplasms with different salt tolerance levels were compared between the salt treatment and control samples (Figure 3, Tables S2–S5). Overall, 5997 and 5532 DEGs were detected in the salt-resistant germplasms Vu_191 and Vu_328, respectively. The DEGs identified for Vu_191 typically included senescence-associated genes and genes encoding LEA proteins, while those identified for Vu_328 included genes encoding stress-induced proteins and pectin lyase. In addition, 5031 and 7444 DEGs were detected in the salt-sensitive germplasms Vu_393 and Vu_396, respectively. The DEGs identified for Vu_393 included genes encoding phosphatase family proteins and cytochrome P450 family proteins, while those identified for Vu_396 included nitrate transporters and auxin efflux carrier family proteins. In addition, individual DEGs induced by salt treatment in four germplasms were compared (Tables S6 and S7). Overall, a higher number of up-regulated genes were identified in the salt-resistant germplasms, while the majority of the down-regulated genes were detected in the salt-sensitive plants. In the salt-resistant germplasms, 65 common DEGs, including LEA 4–5, were up-regulated, compared with 60 in the salt-sensitive germplasms, a group which included wall-associated kinase 3 (Figure 4a). In addition, 59 common DEGs, including metallothionein 2A, were down-regulated in the salt-resistant germplasms, compared with 99 for the salt-sensitive germplasms, including the cytochrome P450 family (CyP-89-A-5) (Figure 4b). Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were conducted in order to understand and classify the functions of the common DEGs identified for the different germplasms (Tables S8 and S9). In the salt-resistant germplasms, the up-regulated genes had nine enriched GO terms in the molecular function (MF) category, with many of the DEGs associated with catalytic and transferase activity. In addition, the down-regulated genes had only one enriched GO term in the MF category, with one DEG identified associated with ADP binding, unlike the up-regulated genes. On the other hand, there were no GO terms identified as being at a significant level (p < 0.05) from among the common DEGs for the salt-sensitive germplasms. KEGG analysis classified the common DEGs into five major groups for the salt tolerance germplasms. Most of the DEGs, for both salt tolerance levels, were associated with metabolism at the major classification and with the global/overview maps related to pathways or metabolism at the sub-classification. Hierarchical clustering analysis was conducted to confirm the gene expression patterns using information from the 9784 DEGs that were significantly expressed for each salt tolerance germplasm (Table S10). The identified DEGs were classified into six clusters containing 3710, 1762, 2961, 384, 276, and 691 genes, respectively (Figure 5). For the C1 and C6 clusters, most of the DEGs were generally down-regulated while, for the C2 and C3 clusters, the DEGs were generally up-regulated across the four germplasms. Most of the relative expression levels were found to be similar in the four germplasms, but the C4 and C5 clusters exhibited distinct differences. The C4 cluster contained DEGs that were down-regulated in salt-resistant germplasm Vu_191 and those that were up-regulated in salt-sensitive germplasm Vu_396. This cluster contained DEG-encoding nodulin MtN21/EamA-like transporter family protein and NAD(P)-binding Rossmann-fold superfamily protein. On the other hand, the C5 cluster contained DEGs that were up-regulated in some salt-resistant germplasms and down-regulated in some salt-sensitive germplasms. This cluster contained DEG-encoding nitrate transporter 1.5 and the heavy-metal transport/detoxification superfamily protein. GO and KEGG analysis was conducted to understand the functions of the DEGs included in each cluster (Tables S11 and S12). Overall, 384 DEGs in the C4 cluster were enriched for 29 GO terms, with 28 being independently classified into the MF category and 1 as a biological process (BP). In addition, many of the DEGs were associated with catalytic activity in the MF category, as is the case with the common DEGs, with a difference in that DEGs were detected for the response to oxidative stress in the BP category. However, in the C5 cluster, no GO terms were identified at a significant level (p < 0.05). As a result of the KEGG analysis of these genes, the C4 cluster was grouped into five major classifications and the C5 cluster into three (excluding genetic information and cell processing). In particular, metabolic terms were most common in both clusters, with sub-classifications dominated by global/outline maps related to pathways or metabolism, carbohydrate metabolism, and the biosynthesis of other secondary metabolites. However, this sub-classification had differences for six items, including membrane transport, transport, and catabolism. qRT-PCR was employed to validate the expression of these DEGs (Figure S1). Of the DEGs with significant expression patterns, six were selected and analyzed further. The relative expression levels obtained from qRT-PCR exhibited trends similar to those of the Log2FC from RNA-seq. A total of 27 candidate genes were obtained based on the RNA-seq results, and two of these were not annotated (Table 1). The 25 annotated genes included various genes related to salt stress, such as LATE EMBRYOGENESIS ABUNDANT PROTEIN 4–5, and POTASSIUM TRANSPORTER 6. Reference sequencing (re-seq) was conducted on the four germplasms to narrow down the candidate genes, and the results were integrated with those of RNA-seq. The whole genome re-seq results for each germplasm are presented in Table S13. Of the 27 candidate genes, two containing variations that may be associated with salt resistance and sensitivity (Vigun_02G076100 and Vigun08G125100) were selected (Figure 6). Vigun_02G076100 and Vigun08G125100 were identified as encoding POTASSIUM TRANSPORTER 6 and EXOCYST COMPLEX PROTEIN EXO70, respectively. A total of eight coding SNPs (cSNPs) were found in the exons of these two genes. The five cSNPs found in Vigun_02G076100 included three synonymous SNPs (sSNP) without amino acid substitutions and one synonymous variation without an amino acid substitution. However, the other SNP caused the substitution of lysine (Lys, K) in the positive amino acid with glutamic acid (Glu, E) in the negative amino acid when compared to the reference. This SNP was found in the salt-resistant germplasm Vu_191. The three cSNPs found in the other candidate gene Vigun_08G125100 included two non-synonymous SNPs (nsSNPs). When compared with the reference, one SNP led to the substitution of aspartic acid (Asp, D) in the negative amino acid with histidine (His, H) in the positive amino acid, while the other SNP led to the substitution of glycine (Gly, G) in the special case amino acid with Asp. These two SNPs were found in both salt-sensitive germplasms Vu_393 and Vu_396. Interestingly, the salt-resistant germplasms Vu_191 and Vu_328, for which no SNPs were found, had a missing allele rather than the reference allele. The variations in the two candidate genes were confirmed using DNA-seq and PCR analysis of the four cowpea germplasms. Of the cSNPs identified in Vigun_02G076100, one SNP, causing the substitution of another type of amino acid, was confirmed through DNA-seq. This was the same variation as observed in the four germplasms used for RNA-seq analysis, and the confirmed SNP was used to develop the KASP marker. In order to confirm that Vigun_08G125100 was a missing allele, a primer producing a 1465 bp PCR product was designed (Table S14). Interestingly, PCR products were only generated for this gene with the salt-sensitive Vu_393 and Vu_396. The variations in the two candidate genes were validated using a total of 20 cowpea germplasms that included the 4 germplasms used for RNA-seq (Table 2). Vigun_02G076100 was validated through the developed KASP marker. As a result, Vu_191 exhibited the same variation as seen in the re-seq analysis, and the SNP variation was observed in the salt-resistant germplasm Vu_111 (Figure S2). On the other hand, for Vigun_08G125100, PCR products were only generated for five salt-sensitive germplasms, including Vu_393 and Vu_396 (Figure S3), although this was not found in all salt-resistant cowpea germplasms. Cowpea is a legume crop that is widely grown in arid and semiarid regions because it is both heat- and drought-tolerant [1]. However, salt stress is becoming an increasingly serious issue for crops in these regions due to climate change [4]. In this study, we evaluated the ion accumulation response to salt stress using four cowpea germplasms with different levels of salt tolerance and conducted RNA-seq analysis (Figure 2). It was found that ion accumulation was significantly higher in the salt-sensitive germplasms rather than the salt-resistant germplasms. These results are in agreement with those reported by previous studies [19]. Of the four germplasms investigated in the present study, Vu_191 was classified as strongly salt-resistant, Vu_328 as weakly salt-resistant, Vu_393 as having a medium tolerance, and Vu_396 as a salt-sensitive germplasm. Based on these results, we screened candidate genes by focusing on DEGs with significant expression patterns in the comparison between Vu_191 and Vu_396. The expression profiles of these DEGs were compared between control and treatment groups to identify genes associated with salt stress. Consequently, numerous DEGs related to salt stress were identified in the present study. For example, Vigun_11G140800 encodes senescence-associated gene 12 (SAG 12), which is related to cysteine protease [35], and plays a role in plant aging and programmed cell death in response to biotic and abiotic stresses [36]. This was up-regulated in all four germplasms, suggesting that aging was accelerated by salt stress. In addition, Vigun_09G159100 is a gene-encoding wall-associated kinase 3 (Wak3), which is associated with the pectin molecule in the cell wall and is essential for cell expansion [37]. It has been reported that a decrease in protein levels affects cell expansion and cell shape. In the present study, there was no significant expression value in the salt-resistant germplasms, but overexpression was observed in the salt-sensitive germplasms. This may be a form of plant defense to maintain homeostasis in response to stress such as excessive salt accumulation in salt-sensitive plants. Another gene of note was Vigun_03G195700, which encodes CyP-89-A-5. The Cyp family is a large collection of proteins found in higher plants, and it has been assumed that they provide protection from various biotic and abiotic stresses. In particular, it has been reported that the suppression of CaCyP1 in pepper, which has a high homology with CyP-89-A-5 in Arabidopsis, increased the susceptibility to bacterial pathogens [38]. This gene is down-regulated in salt-sensitive germplasms, which is consistent with our results. Therefore, the Cyp gene found in cowpea is also assumed to affect salt tolerance via a similar mechanism. We also conducted GO and KEGG analysis of the common DEGs and each cluster. The DEGs identified for the salt-resistant germplasms were generally related to catalytic activity (GO:0003824) and transferase activity (GO:0016740), while the DEGs corresponding to Cluster 4 were associated with small-molecule binding (GO:0036094), anion binding (GO:0043168), and ribonucleotide binding (GO:0032553). These GO terms play important roles in several salt tolerance mechanisms, including osmotic regulation. In particular, catalytic activity (GO:0003824) exhibited functions related to osmotic regulation and ionic change when exposed to salt stress [39]. These results also suggest that protein-coding genes, related to molecular structure and function, can be regulated in response to salt stress, and further indicate that anion reactions are associated with salt stress. As a result of our KEGG analysis, most of the DEGs were associated with metabolism with four sub-classifications (global and overview maps, amino acid metabolism, carbohydrate metabolism, and biosynthesis of other secondary metabolites). This suggests that abiotic stress not only regulates metabolic processes via enzyme activity but also causes indirect or direct changes in proteins by affecting amino acids. Our results also suggest the involvement of the metabolism of various amino acids and the biosynthetic pathways of secondary metabolites such as phenylpropanoids. It has been reported that phenylpropanoids are activated under various abiotic stress conditions, including salt stress, to remove ROS [40]. These results thus help us to understand the molecular biological response to salt stress. We subsequently selected 27 target genes, related to salt tolerance, based on the expression patterns and annotations for the DEGs in the RNA-seq analysis. One of these was Vigun_01G124200, which encodes LATE EMBRYOGENESIS ABUNDANT PROTEIN 4–5 (LEA 4–5). The LEA protein is a polypeptide that accumulates in later embryonic stages and is associated with the acquisition of desiccation tolerance [41]. It also increases resistance to osmotic and cold stress in various crops and is associated with water-deficient conditions [42]. This protein generally accumulates during periods of stress-induced growth arrest and is involved in stress recovery [43]. For example, AtLEA4–5 in Arabidopsis is known to be a member of the genes encoding the LEA protein involved in water deprivation tolerance [44]. This gene is usually suppressed by the repressor AtMYB44, but it has been reported that, when exposed to osmotic stress, the repressor is removed and normal expression occurs [45]. In the RNA-seq results, Vigun_01G124200, which encodes LEA4–5, was significantly up-regulated in the salt-resistant germplasms. However, the re-seq results did not detect significant variations. Interestingly, Vigun_03G281700 encoding MYB44 was up-regulated in the salt-sensitive germplasms. This suggests that the expression of LEA4–5 in cowpea can be regulated by the same mechanism used in Arabidopsis thaliana, but that it is also regulated by an additional pathway. Because interpreting the large volumes of data from the 27 target DEGs was difficult, we conducted re-sequencing to narrow down the range of the candidate genes. Most of the target genes had many SNPs in each salt-tolerant germplasm, but these SNPs were in the UTR or intron regions, which may not be involved in regulating gene expression. However, some of the SNPs in the two candidate genes, Vigun_02G076100 and Vigun_08G125100, exhibited significant associations with salt tolerance. Vigun_02G076100, a gene-encoding POTASSIUM TRANSPORTER 6, was up-regulated in salt-resistant Vu_191. Potassium (K+) is an essential cation for plant growth and development and the regulation of enzyme activity, membrane potential, and turgor pressure [12]. High salinity is the result of the accumulation of excessive sodium (Na+) ions, which leads to ion stress. Plants are consequently unable to maintain K+ homeostasis, which ultimately adversely affects plant growth. Accordingly, one of the primary mechanisms associated with salt tolerance in plants is the maintenance of a balanced cation ratio in the cytoplasm [46]. In addition, the Arabidopsis KUP6 subfamily transporter is related to cell growth and potassium homeostasis and has been reported to be a major factor associated with osmotic control [47]. This suggests that the strong salt resistance of Vu_191 occurs as a result of the overexpression of potassium transporter 6. Vigun_08G125100 encodes EXOCYST COMPLEX PROTEIN EXO70 and was up-regulated in both Vu_393 and Vu_396. The exocyst subunit EXO70 protein has been reported to be involved in anchoring and regulating membrane fusion and actin polarity in the plasma membrane of exocysts [48]. Some genes included in the exocyst gene family have been reported to be up-regulated with exposure to salt stress, but their exact functions have not been identified [49]. The variations in the two candidate genes were validated using KASP genotyping and PCR products. The cSNPs found in Vigun_02G076100 were found in Vu_111 and Vu_191, both of which were salt resistant. This has been identified as a specific variation in some salt-resistant germplasms. Based on this, it can be assumed that Vu_111 had the same salt tolerance mechanism as Vu_191. Conversely, Vigun_08G125100 was identified as a missing allele in NGS analysis. To validate these results, 20 cowpea germplasms were tested, with 75% classified as having the same salt tolerance type as before. Thus, it can be assumed that the loss of this gene has occurred as a result of the development of various salt resistance mechanisms, but the functional part has not been confirmed. In summary, we identified two candidate genes related to salt tolerance that differed between cowpea germplasms with different levels of salt tolerance. These variations were developed as KASP and indel markers, respectively. The two developed markers thus have the potential to be useful molecular markers for the screening of germplasms in salt tolerance breeding programs. In this study, 20 cowpea germplasms with different levels of salt tolerance (10 salt-resistant and 10 salt-sensitive) were used, and among them four showed distinct differences in salt tolerance under controlled conditions and were used for RNA-seq analysis (Table S15). The 20 cowpea germplasms were then used to verify the SNP variation. The germplasm seeds were obtained from the Rural Development Administration (RDA) Genebank at the National Agrobiodiversity Center, Republic of Korea. The four cowpea germplasm were treated with 250 mM NaCl for seedlings in the V2 stage with the same growth after germination. After three weeks of NaCl treatment, the entire plant was sampled to evaluate the accumulation of sodium and chloride ions. The ion content was extracted from dried and pulverized leaf samples (150 mg) using 30 mL of distilled water for 1 h and then filtered through Whatman filter paper. The sodium ion levels were determined using a Na+ measuring instrument (Horiba, Kyoto, Japan), while the chloride ion levels were determined using an ion-selective electrode (Mettler Toledo, Columbus, OH, USA). One hundred seeds from each germplasm were sterilized with 70% ethanol for 1 min and then washed with sterile water. The sterilized seeds were germinated in a plant growth chamber under long-day conditions (16 h light and 8 h dark), and similar seedlings were selected and transplanted into 1/2 Hoagland Nutrient Solution for hydroponic use. After two weeks of salt treatment, seedlings with the same growth were treated with 250 mM NaCl, while the control seedlings were placed in a solution without NaCl. After 24 h of salt treatment, the roots of the NaCl-treated and control seedlings were sampled. Each treatment and control group had three biological replicates, which were randomly sampled at 10 points and mixed into a single sample. The samples were frozen using liquid nitrogen and stored at −80 °C for use in subsequent experiments. Overall, a total of 24 RNA library samples were analyzed. Total RNA was extracted using an RNeasy Plant Mini Kit (Qiagen, Hilden, Germany). The quality and integrity of the extracted RNA were determined using a 2100 Bioanalyzer RNA instrument (Agilent, Santa Clara, CA, USA). Poly-A+ libraries were prepared using an Illumina Truseq Stranded mRNA Library Prep Kit (Illumina, San Diego, CA, USA), and the generated libraries were sequenced using an Illumina NovaSeq6000 platform. Both RNA extraction and cDNA library construction were conducted according to the manufacturer’s instructions. In the sequenced transcriptome short reads, the adapter sequence was removed with cutadapt [50] and pre-processing was conducted using DynamicTrim and LengthSort in the SolexaQA package [51]. DynamicTrim removes low-quality bases at both ends of short reads to purify them, while LengthSort excludes trimmed reads of 25 bp or fewer from the analysis process. The clean trimmed reads were mapped onto the Vigna unguiculata (v1.2) reference genome from the Phytozome database (http: //phytozome.jgi.doe.gov/ (accessed on 1 December 2022)) using HISAT2 software [52]. HTSeq (v.0.11.0) [53] was used to measure expression as the total number of reads mapped to each gene. In order to avoid bias due to the germplasm in the sequencing numbers, normalization was conducted using the DEseq library [54]. DEGs were selected based on a twofold change in the number of mapped reads and an FDR of ≤0.01, with the adjusted p value calculated using Benjamini–Hochberg correction. Hierarchical clustering analysis was conducted using the amap [55] and gplot libraries [56] in R to determine gene expression patterns, which were calculated using Pearson’s correlation, and grouping was conducted through the complete method. GO enrichment was analyzed using reference GO information [57]. The significance level was set at 0.05 and the GO terms were classified into biological process (BP), cellular component (CC), and molecular function (MF) categories. Functional annotation was conducted for an e-value of ≤ 1 × 10−100 and best hits using amino acid sequences from the KEGG database [58] and BLASTP. First-strand cDNA was synthesized using SuperScript™ III First-Strand Synthesis SuperMix (Invitrogen, Waltham, MA, USA) following the manufacturer’s instructions. qRT-PCR was conducted on a StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) using a Bio-Rad iQ™ SYBR Green Supermix Kit (Invitrogen, CA). The reaction mixture, containing 20 ng of cDNA, was analyzed according to the manufacturer’s instructions. The PCR conditions were as follows: holding, 1 cycle at 95 °C for 10 min; cycling, 40 cycles at 95 °C for 15 s and at 60 °C for 60 s. Then, the melting curve analysis was conducted to confirm the absence of a product and the dimer formation of the primers. The primers were designed using Primer3 software (v2.3.5) [59]. The CT values were normalized using the ubiquitin-conjugating enzyme E2 variant 1D (UE21D) gene stable under salt stress as a housekeeping gene [60] and gene expression was analyzed using the 2−ΔΔCT method [61]. Three biological replicates were analyzed using the average of two technical replicates. Genomic DNA was extracted using a DNeasy Plant Mini Kit (Qiagen) following the manufacturer’s instructions, and the integrity and purity of the extracted DNA samples were determined using 2.0% agarose gel and a Nanodrop ND 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The cDNA library was constructed and sequenced using the same NGS protocol as for RNA-seq. Paired-end reads were mapped onto the cowpea genomic reference genome and then entered into the nf-core/sarek’s analysis pipeline [62]. The DNA was sequenced using the PCR products of the candidate gene on an ABI 3730XL analyzer (Applied Biosystems). The primers used to generate the PCR products were prepared in the same way as the primers used for qRT-PCR. More detailed information on this process is provided in Table S14. KASP primers were designed to detect SNP variation in the candidate genes according to the standard KASP protocol. Allele-specific primers included FAM (5′-GCTATAACCAGAACAGGCCATCTCAATTT-3′) and HEX (5′-TAACCAGAACAGGCCATCTCAA-TTC). The KASP primers were used to genotype the 20 cowpea germplasms using StepOnePlus software (Applied Biosystems). Genotyping was conducted using a mixture consisting of 50 ng/5 uL of DNA, 0.14 uL of KASP assay mix, and 5 uL of KASP master mix. The KASP cycling conditions were as follows: pre-PCR reading, 1 cycle at 30 °C for 1 min; holding, 1 cycle at 94 °C for 15 min; cycling, 10 cycles at 94 °C for 20 s and 61–55 °C for 1 min (reduction of 0.6 °C per cycle), and 26 cycles at 94 °C for 20 s and 55 °C for 1 min; and post-PCR reading at 30 °C for 30 s. Statistical analysis was conducted using analysis of variance (ANOVA) and least significant difference (LSD) tests in SPSS 27 (IBM, Armonk, NY, USA), with p < 0.05 employed to determine statistically significant differences between groups. Four cowpea germplasms with different levels of salt tolerance were used to investigate transcriptome variations in roots under salt stress. RNA-seq analysis of the salt treatment and control groups, with three biological replicates assessed for each germplasm, led to the selection of 27 candidate genes related to salt stress. Of these, two candidate genes with significant variation were investigated further in this study. The two candidate genes contained cSNPs in the exon region and represented a missing allele, respectively. The information provided on the two candidate cowpea genes in relations to salt stress and presented in the present study has the potential to be used for genetic improvements in cowpea breeding programs.

PMC10002510

Francesca Veronesi,Silvia Brogini,Angela De Luca,Davide Bellini,Veronica Casagranda,Milena Fini,Gianluca Giavaresi

Cell Adhesion and Initial Bone Matrix Deposition on Titanium-Based Implants with Chitosan–Collagen Coatings: An In Vitro Study

02-03-2023

bone implants,biomaterials,biocompatibility,Ti-alloy coatings,collagen,chitosan

In orthopedics, titanium (Ti)-alloy implants, are often considered as the first-choice candidates for bone tissue engineering. An appropriate implant coating enhances bone matrix ingrowth and biocompatibility, improving osseointegration. Collagen I (COLL) and chitosan (CS) are largely employed in several different medical applications, for their antibacterial and osteogenic properties. This is the first in vitro study that provides a preliminary comparison between two combinations of COLL/CS coverings for Ti-alloy implants, in terms of cell adhesion, viability, and bone matrix production for probable future use as a bone implant. Through an innovative spraying technique, COLL–CS–COLL and CS–COLL–CS coverings were applied over Ti-alloy (Ti-POR) cylinders. After cytotoxicity evaluations, human bone marrow mesenchymal stem cells (hBMSCs) were seeded onto specimens for 28 days. Cell viability, gene expression, histology, and scanning electron microscopy evaluations were performed. No cytotoxic effects were observed. All cylinders were biocompatible, thus permitting hBMSCs’ proliferation. Furthermore, an initial bone matrix deposition was observed, especially in the presence of the two coatings. Neither of the coatings used interferes with the osteogenic differentiation process of hBMSCs, or with an initial deposition of new bone matrix. This study sets the stage for future, more complex, ex vivo or in vivo studies.

Cell Adhesion and Initial Bone Matrix Deposition on Titanium-Based Implants with Chitosan–Collagen Coatings: An In Vitro Study In orthopedics, titanium (Ti)-alloy implants, are often considered as the first-choice candidates for bone tissue engineering. An appropriate implant coating enhances bone matrix ingrowth and biocompatibility, improving osseointegration. Collagen I (COLL) and chitosan (CS) are largely employed in several different medical applications, for their antibacterial and osteogenic properties. This is the first in vitro study that provides a preliminary comparison between two combinations of COLL/CS coverings for Ti-alloy implants, in terms of cell adhesion, viability, and bone matrix production for probable future use as a bone implant. Through an innovative spraying technique, COLL–CS–COLL and CS–COLL–CS coverings were applied over Ti-alloy (Ti-POR) cylinders. After cytotoxicity evaluations, human bone marrow mesenchymal stem cells (hBMSCs) were seeded onto specimens for 28 days. Cell viability, gene expression, histology, and scanning electron microscopy evaluations were performed. No cytotoxic effects were observed. All cylinders were biocompatible, thus permitting hBMSCs’ proliferation. Furthermore, an initial bone matrix deposition was observed, especially in the presence of the two coatings. Neither of the coatings used interferes with the osteogenic differentiation process of hBMSCs, or with an initial deposition of new bone matrix. This study sets the stage for future, more complex, ex vivo or in vivo studies. Titanium (Ti) is a quasi-bioinert biomaterial and, in the form of alloys, Ti is the most preferred metal for bone implants, due to its high specific strength, low density, corrosion resistance, and biocompatibility [1,2]. Ti-alloys also show a lower Young’s modulus as compared to others, such as stainless steel or cobalt chrome [3]. Nevertheless, the failure of Ti-based implants is not uncommon, due to toxic outcomes of Ti, as well as to the “stress shielding” effect, a result of the tissue–implant mechanical mismatch [4]. An effective strategy for reducing or eliminating stress shielding, and simultaneously enhancing stable long-term fixation, by means of full bone ingrowth and remodeling, has been the development of open pored metallic structures, through additive manufacturing (AM) techniques [5]. The success or failure of an implant is largely dependent on the extent to which it induces bone matrix ingrowth and integrates into the surrounding bone. The greater the osseointegration, the higher the initial mechanical stability, and the lower the probability of implant loosening, with the formation of fibrous tissue at the interface. In addition to mechanical and topographic characteristics, an appropriate surface modification, such as a proper biocoating, also plays a pivotal role for the success of an implant [6]. The implant surface covering enhances bone matrix ingrowth and biocompatibility, improving the long-term stability between new bone formation and the implant, and reducing the healing time [7,8]. Active biomolecules, such as peptides, collagens, and proteins, or growth factors (GFs), such as bone morphogentic protein-2 (BMP2), are used to increase Ti-based implant biocompatibility and physicochemical performance, leading to an increase in cell responses, bone matrix ingrowth and, thus, osseointegration [9]. Among them, type I collagen (COLL I), due to its hemostatic properties, superior biocompatibility, bioactivity, bioresorbability, and low immunogenicity and antigenicity, has been widely used as a tissue replacement material in several medical applications [10]. COLL I, the most abundant protein in mammals, accounts for nearly 95% of the organic matrix in bone, and induces osteoblast (OB) activities, such as proliferation, differentiation, migration, and secretion of extracellular matrix (ECM), providing a suitable environment for bone formation [11]. In preclinical and clinical studies, a significant increase in bone growth and bone-to-implant contact (BIC) around Ti implants has been observed after surface treatment with COLL I [10,12,13,14,15,16,17]. Chitosan (CS), a deacetylated derivative of the natural polysaccharide chitin, is widely used in several biomedical applications [18,19,20,21,22,23,24,25]. It is a bioactive natural alkaline polysaccharide, positively charged, biocompatible, biodegradable, inexpensive, nontoxic, with wound-healing activity, and antibacterial and antimicrobial properties. CS is comparable to bone and cartilage ECM in its composition and chemical structure, and its products are easily metabolized inside the human body by some enzymes [22]. Some studies have shown that scaffolds with layers of CS display improved implant biocompatibility, osteoconductivity, and bone regeneration, and induce OBs and MSC proliferation and neovascularization. Finally, CS also increases cell adhesion and protein adsorption on Ti-alloy implants, consequently improving the osseointegration [18,19,25]. In the literature, there are still very few studies that fabricated composite CS/COLL coverings for bone implants. CS/COLL/hydroxyapatite nanofibers, coated with platelet-rich plasma (PRP), affected the osteogenic differentiation of OBs [26], and a bilayered COLL/CS membrane induced significant expression of osteogenic genes in mesenchymal stem cells (MSCs), and bone formation with no inflammation, in calvarial defects, in an animal model [27]. No study in the literature has evaluated the osteogenic potential and matrix deposition of a Ti-alloy, coated with three COLL/CS multilayers. Recently, we have published a study, in which Ti-alloy (Ti-6Al-4V) (Ti-POR) cylinders, coated or not with COLL I, were evaluated for their osseointegration ability, in an ex vivo study, developed by culturing rabbit cortical bone segments with these cylinders. It was observed that COLL I improved osseointegration and bone growth of Ti-POR [28]. The present study is an upgrade of the previous one [28], where the same Ti-POR samples were covered with two types of multilayered coatings: one made of collagen-chitosan-collagen (COLL–CS–COLL) and the other of chitosan–collagen–chitosan (CS–COLL–CS). Since the antibacterial and antimicrobial properties of CS-coated implants have already been tested, the aim of this study was to observe the ability of these two types of coating to induce cell adhesion and proliferation and form bone matrix. To reach this aim, after cytotoxicity evaluations, human bone marrow MSCs (hBMSCs) were seeded onto Ti-POR specimens, and coated or not with COLL–CS–COLL or CS–COLL–CS coatings for 48 h, and 14 and 28 days. Cell adhesion, proliferation, and osteogenic gene expression, associated with histological and scanning electron microscope (SEM) analyses, were performed. Figure 1 shows the overlays of the IR spectra of Ti-POR, COLL–CS–COLL, and CS–COLL–CS specimens. In the Ti-POR spectra (Figure 1A), the peaks relating to the salts present in PBS were observed between 1200 and 850 cm−1, as expected. On the other hand, as regards the coated specimens, in addition to the salt peaks, the characteristic peaks of COLL and CS, between 1650 and 1550 cm−1, were present (Figure 1B,C). The solution containing the COLL–CS–COLL specimen showed these peaks in a more marked way, suggesting a slightly higher release of material than the CS–COLL–CS samples (Figure 1B,C). Figure 2 shows representative microscope images of Saos-2 cells in the presence of Ti-POR, COLL–CS–COLL, and CS–COLL–CS specimens. It was observed that Saos-2 cells in direct contact with Ti-POR specimens maintained their typical morphology over 72 h of culture, and had an increasing proliferation. The Saos-2 cells in contact with COLL–CS–COLL and CS–COLL–CS samples had a slower proliferation than those cultured in the presence of Ti-POR, and mainly maintained their typical morphology, and some cells acquired an elongated shape. No significant differences among the types of specimens were observed, in the viability of hBMSCs after 48 h of culture. After 14 days of culture, CS–COLL–CS showed significantly higher hBMSC viability than COLL–CS–COLL and Ti-POR (p < 0.05). After 28 days of culture, both CS–COLL–CS and COLL–CS–COLL induced significantly higher cell viability than Ti-POR (p < 0.05). Over time, the cell viability on COLL–CS–COLL significantly increased (p < 0.05), while for CS–COLL–CS and Ti-POR, cell viability showed a significant increase from 48 h to 14 days (p < 0.05) (Figure 3). In Figure 4, the results of RUNX2, SP7, COLA1, and ALPL gene expression are shown. As regards RUNX2, for all the samples, its expression was very low (under 1 value), and no significant differences were observed among samples at 48 h and 28 days of culture. At 14 days, COLL–CS–COLL significantly increased RUNX2 expression, in comparison to 48 h and 21 days (p < 0.05), and CS–COLL–CS and Ti-POR, in comparison to 21 days (p < 0.05) (Figure 4A). SP7 was not expressed after 21 days of culture, and did not show statistically significant differences among samples at all experimental times. For all samples, its expression significantly decreased from 14 to 28 days of culture (p < 0.05) (Figure 4B). On the contrary, COL1A1 showed a very low expression (below 1 value) at 48 h and 14 days, up to 28 days of culture, when its expression was significantly higher than at the other experimental times, for all the samples (p < 0.05). No significant differences were observed among the samples (Figure 4C). As regards ALPL expression, for all specimens, its expression was significantly higher at 48 h and 28 days of culture in comparison to 14 days (p < 0.05), when its expression was almost absent. No significant differences were observed among specimens for any of the experimental times (Figure 4D). For Ti-POR, no matrix deposition was observed at 48 h. It gradually increased, starting from 14 days, and it was observable both on the surface and within the cylinder. At higher magnification, an initial mineralization of the matrix, colored in green after Toluidine blue/Fast green staining, can already be seen at 14 days, which becomes more evident at the longest experimental time of 28 days (Figure 5A). As for the COLL–CS–COLL and CS–COLL–CS samples, the presence of a filamentous reticular structure, reactive to the Toluidine blue dye, can be seen, both on the surface and inside the material, regardless of the experimental time. After 28 days, for both types of coatings, the presence of areas reactive to the Fast green dye was observed, both on the surface of the materials and along the filaments described above (Figure 5B,C). Compared to Ti-POR, these areas are clearer, larger, as well as more easily identifiable, and uniformly distributed along the entire perimeter of the samples. The staining with Stevenel’s blue and Picrofuchsin confirmed the presence of areas with different degrees of mineralization in the samples, at 28 days (Figure 6). The analysis of the BS-SEM images showed the degree of colonization of the specimens and the de novo matrix production by the hBMSCs, confirming the histological data (Figure 7). On Ti-POR, already at 48 h, the hBMSCs were clearly visible, and the colonization increased over time, both in terms of the number of cells visible under high magnification and in the newly deposited matrix, which progressively covered the surface of the material. The images of the COLL–CS–COLL samples showed the amorphous matrix covering the material, and at higher magnifications the cells were visible, which together with the matrix produced, formed a further layer enveloping the COLL–CS–COLL multilayer coating. At 28 days, cells were visible at intermediate magnifications, indicating increased cell growth and matrix deposition by cells, which made it more difficult to discriminate individual cells, except at high magnifications (Figure 7). Similarly, for CS–COLL–CS, cells colonizing the material were clearly visible at low and intermediate magnifications. At 28 days the hBMSC colonization was easily visible, even at intermediate magnifications, forming a layer of cells and a new matrix, deposited on the CS–COLL–CS multilayer (Figure 7). The present in vitro study evaluated the cell adhesion and bone matrix formation ability of two new multilayered coatings of a Ti-alloy implant (Ti-POR), for further employment in the orthopedic field. The study showed that both Ti-alloy coatings, made of composite COLL/CS layers, were cytocompatible, improving hBMSC adhesion and subsequent colonization from the shortest experimental time, of 48 h. Furthermore, a progressive increase in matrix deposition was observed, and an initial mineralization started from 28 days of culture. For this purpose, hBMSCs were chosen as cells to be seeded onto samples, to evaluate their osteogenic commitment and bone matrix production ability up to 28 days of culture. Cell viability and gene expression analyses were performed, in association with histological and SEM evaluations. Ti-POR samples, manufactured using the additive manufacturing technique of EBM, on which the coverings were applied, were the same used in two previous ex vivo and in vivo studies [11,28]. The samples showed a 3D superficial open porous structure of Ti6-Al-4V, interconnected with a solid central metallic core. The superficial porous layer (1 mm thick) had an average pore diameter of 700 μm. This pore dimension falls in the range 300–1000 μm, the upper and lower bounds of which are considered to be the minimum and maximum pore dimensions that can promote bone ingrowth [29]. After having prepared the 3D CAD model of the porous cylinders, a dedicated software sliced the 3D model into 50 μm layer thicknesses [28]. Surface composition plays an essential role in determining the initial cellular and molecular response of cells that meet the implant. In the literature, the biocompatible and antifungal properties of CS are well recognized, and together with its high antimicrobial and biodegradability characteristics, it and its derivatives have already been employed in several medical applications [30]. It is often used as a dressing material, in the manufacture of drugs as a controlled-release active substance carrier, or in tissue engineering involving soft tissues, nerves, cartilage, bones, or arteries [31,32]. Furthermore, CS is also reported to be an excellent material for growing OBs, such as glycosaminoglycans and hyaluronic acid, thanks to its structural characteristics [33]. Given the promising results of CS for implant applications, its performance and capabilities need to be further assessed and developed for bone. On the other hand, several studies have already tested the osteogenic potential of COLL, the most abundant protein of bone ECM, in numerous orthopedic and dental applications [11,12,14,16,17,28]. For the above-mentioned reasons, in the present study, CS and COLL were evaluated in combination, as an innovative covering of Ti-alloy samples. Through an innovative spraying technique, the two types of multilayered coverings were fabricated, made by COLL–CS–COLL or CS–COLL–CS trilayers. The PBS release test evidenced the presence of the characteristic peaks of COLL and CS between 1650 and 1550 cm−1, in association with the peaks of PBS salt. A first evaluation of cytotoxicity, showed that Saos-2 cells reduced their proliferation and acquired an elongated shape in the presence of both COLL–CS coatings, indicating that cells started to differentiate towards more mature cells and started to produce bone matrix. As regards hBMSCs’ viability, the results showed that, up to 28 days of culture cells remained vital, with a significant increase from 48 h to 14 days of culture, and then maintaining stable growth from 14 to 28 days. In addition, both coated specimens induced a higher cell proliferation than the non-coated one, at 14 and 28 days. For gene expression analysis we chose the most representative osteogenic differentiation genes, that encode for both early and late bone formation markers. Among them, RUNX2 and SP7 genes encode for transcription factors that are considered early osteogenic markers, showing an essential role in the development of OBs, driving the differentiation of BMSCs into OBs and eventually osteocytes with their expression, that decreases when BMSCs differentiate towards OBs [34,35]. COLA1 is a gene that encodes for the pro-alpha1 chains of COLL I, and is considered a late marker of osteogenesis. It is a fibril-forming collagen found in most connective tissues, abundant especially in bone [36]. Finally, ALPL encodes for ALP, an early marker of OB differentiation, which is elevated during osteodifferentiation of MSCs. In the present study, all coated or not coated specimens showed the same gene expression trend at all experimental times, without observing statistically significant differences among the specimens. This result underlined that all samples, analyzed in this study, induced a similar osteogenic hBMSC differentiation. However, differences in gene expression were highlighted over time. A peak of RUNX2 expression was detected starting from 14 days of culture and it significantly decreased at 28 days of culture. On the contrary, major levels in the expression of SP7 were detected at 48 h and the declined at 28 days of culture, at which time no gene activity was observed. Hence, this shift of expression probably determines a later up-regulation of the genes directly regulated by RUNX2, such as ALPL that showed an opposite trend respect RUNX2 expression. In literature it was observed that ALPL level expression increased after two days of culture, decreased at the intermediate time of 14 days and subsequently upregulated again at 28 days, when matrix mineralization largely covered the culture vessels [37]. On the other hand, COL1A1 expression was significantly higher at 28 days of culture than at 48 h and 14 days, underlining its late marker nature. To appreciate the ability of the specimens to induce cell adhesion and bone matrix formation, even at high magnification, we decided to use histology and SEM. After 48 h of culture, in Ti-POR samples no matrix deposition was observed, while, the other two coated specimens, both COLL–CS–COLL and CS-COLL-CS, induced filamentous reticular structure already at 48 h of culture. Ti-POR gradually increased matrix deposition starting from 14 days of culture, both at the surface and inside the implant. Also, COLL–CS–COLL and CS–COLL–CS specimens increased the production of mineralized matrix on surface and inside the implants, but these areas are larger than those present on Ti-POR material. To better confirmed the presence of these matrix areas, Stevenel’s blue/Picrofuchsin histological staining was performed. This staining confirmed the presence of areas with different degrees of mineralization in the samples at 28 days. The Stevenel’s blue stain shows cells and extracellular structures in various shades of blue (except for mineralized tissues). Counterstaining with Picrofuchsin makes collagen fibers a bluish-green color, bone orange or purple, and osteoid yellow-green [38]. The use of BS-SEM helped to appreciate and confirmed at higher magnification what histology already did, and that is the degree of cell colonization and the de novo bone matrix production. After 48 h of culture, Ti-POR was colonized by hBMSCs and both cell colonization and new matrix deposition increased over time. Also, COLL–CS–COLL specimen and similarly CS–COLL–CSone were covered by an amorphous matrix with cells that created a layer of matrix over the coating, with an increase in cell growth and matrix deposition from 48 h to 28 days of culture. The results of the present in vitro study are certainly preliminary, but are a first step to observing whether new innovative coatings can induce the formation of bone matrix in humans. Obviously, these data will need to be confirmed in subsequent in vivo or ex vivo studies, where it is also possible to evaluate the parameter of osseointegration, in a microenvironment as similar as possible to the physiological condition. Concerning that, the ethical adhesion to the 3R principles requires the progressive decrease and replacement of animal use for preclinical studies [39]. So, to find alternative methods to preclinically evaluate implants, bone matrix formation, and osseointegration, with more advanced methods, able to simulate the clinical condition as much as possible, in a previous study we set up an alternative ex vivo model. This model has been confirmed to be useful in evaluating the osseointegration of implant materials, by culturing cortical bone segments adhered to Ti-POR cylinders, and evaluating bone-to-implant contact (BIC) and new bone formation (nBAr/TAr) after 30, 60, and 90 days of culture. This ex vivo model is seen to reproduce the complex in vivo microenvironment, with a view to reducing the number of animals used in vivo [40]. The Ti-alloy (Ti-6Al-4V) cylinders (4 mm in diameter and 8 mm in length) were manufactured by Adler Ortho® SPA (Cormano, Milan, Italy), as reported in the previous study [28]. The specific structure and shape of the implants were designed based on a previous virtual Computer-Aided Design (CAD) model. Each implant has an interconnected porous structure of Ti-6Al-4V, with a solid central core, and an external porous (average porosity ∅ = 700 μm) layer, with dimensions like trabecular bone. Metal powders were added layer by layer by electron beam melting (EBM) (ARCAM EBM-GE Additive, Gothenburg, Sweden) on Ti-6Al-4V (Ti-POR). Subsequently, all Ti-POR implants were washed in an ethanol/water (80%/20%) mixture for 48 h, and then left in 100% ethanol for 72 h, before air drying. The bio-layers used in the present work consisted of equine COLL I (Euroresearch S.r.l., Milan, Italy) and CS (HMC 90/200-Heppe Medical Chitosan GmbH, Saale, Germany). Both COLL I and CS were used in fluid form, to cover the Ti surface of Ti-POR, with an innovative spraying technique developed in Novagenit S.r.l. laboratories. It consists in spraying fluidified solution of both COLL I and CS directly onto metal samples through a nebulizer. About 100 g of equine COLL I gel was weighed and fluidified at about 45–50 °C, with a magnetic stirrer (IKA RCT Basic, Staufen Germany), monitoring the temperature with a temperature probe. About 500 mg of CS powder was solubilized in 50 mL of 1.5% acetic acid solution in water. The formation of each single layer involved: The spraying of fluidized COLL or CS on the surface of the Ti-POR implants; Rapid freezing at −40 °C; Freeze-drying. In this way, it was possible to create Ti specimens with a triple coating layer: some implants with a coating made of collagen-chitosan-collagen (COLL–CS–COLL) and others with a chitosan–collagen–chitosan coating (CS–COLL–CS). At the end of the lyophilization of the last layer, all specimens were dried in a vacuum oven at 37 °C for 96 h. Finally, they were individually packaged and sterilized at 25 kGy. Uncoated cylinders (Ti-POR) of the same dimension were used as control substrates. Figure 8 shows the three types of specimens tested in the present study. The porous and interconnected structure of the specimens does not allow for the measurement of the layers of COLL and CS accurately and homogeneously on the surface of the Ti-POR specimens. The quantification of both materials anchored on the Ti-POR samples was made by calculating the difference in weight of the specimens before and after the coating treatment. Each of the sterile coated and non-coated specimens were dipped into a well of a 24-well plate, with 2 mL of Dulbecco’s phosphate-buffered saline (PBS) (Sigma-Aldrich, Milan, Italy), to perform the phosphate buffer release test. The evaluations were performed in triplicate for each type of specimen. The 24-well plate was placed in a CO2 incubator at 37 °C for 28 days. At this experimental time the specimens were separated from the PBS. The specimens and PBS were freeze-dried separately, and infrared (IR) spectra acquisition (FT-IR Agilent Cary 630 Spectrometer, Agilent Technologies, Santa Clara, CA, USA) was performed on the lyophilized solution, to verify the presence of COLL I and/or CS in the solution. IR spectra of PBS were obtained, to evaluate the release of the multilayers, and therefore the possible presence of COLL I and/or CS. Saos-2 cells (commercial source Saos-2 human primary osteogenic sarcoma. Cod. 89050205. Sigma-Aldrich, Milan, Italy) were expanded in T75 flasks (Corning Cell Culture Flask, Sigma-Aldrich Corp., St. Louis, MO, USA) in DMEM/F12, with 10% FBS (Sigma-Aldrich Corp., St. Louis, MO, USA), L-glutamine 2 mM, and penicillin-streptomycin 0.1 mg/mL, in standard culture conditions (at 37 °C, 5% CO2, and humidified atmosphere), and the medium was replaced every two–three days. After reaching 85–90% of confluence, 1 × 105 cells/mL were seeded in 6-well plates, with 3 mL of medium per well, and incubated in standard culture conditions for 24 h, to allow the cells to adhere to the wells. Then, the sterile COLL–CS–COLL (n = 2), CS–COLL–CS (n = 2), and Ti-POR (n = 2) specimens were positioned in the wells for 6, 24, 48, and 72 h. At the end of the experimental times, cell proliferation and morphology were qualitatively documented by means of an optical inverted microscope (Eclipse TS100, Inverted Routine Microscope, Nikon Instrument S.p.A., Campi Bisenzio, FI, Italy), and a digital sight camera (Digital Sight DS-L2, Nikon Instrument S.p.A., FI, Italy) at 4X magnification. hBMSCs (commercial source, Cod. 492-05a, Cells Applications Inc., San Diego, CA, USA) were expanded in mesenchymal basal medium (MesenCultTM-MSC Basal Medium; STEMCELL Technologies Inc., Vancouver, BC, Canada), completed with the appropriate supplements (MesenCult™ MSC Stimulatory Supplement; STEMCELL Technologies Inc.), 100 U/mL penicillin, and 100 μg/mL streptomycin, (SIGMA, St. Louis, MO, USA) for about 14 days, in T75 flasks (Corning Cell Culture Flask, Sigma-Aldrich Corp., St. Louis, MO, USA), in standard conditions. The medium was replaced every two–three days. An amount of 2.15 × 105 hBMSCs was seeded on each sterile specimen, taking care to uniformly sow the samples: Ti-POR (n = 15), COLL–CS–COLL (n= 15), and CS–COLL–CS (n = 15) cylinders. Test samples were placed in a 12-well plate and incubated for 1 h in standard conditions, to enhance cell adhesion; then 2 mL of fresh basal medium was added. After 24 h of culture, growth medium was completely replaced with osteogenic differentiation medium, composed of completed basal medium supplemented with β-glycerolphosphate (10−4 M), ascorbic acid (50 µg/mL), and dexamethasone 10−7 M. The medium was replaced every two–three days. After 48 h, and 14 and 28 days, the specimens of each type were analyzed for hBMSC viability. The viability of hBMSCs was quantified by Alamar blue assay (Invitrogen, Waltham, MA, USA) at each experimental time. Alamar blue dye, mixed with culture medium (1:10 v/v), was added to hBMSCs, seeded onto the samples, and incubated for 4 h in standard conditions. The amount of fluorescence is proportional to the number of living cells and corresponds to the cells’ metabolic activity. The fluorescent product was quantified at 530ex–590em nm, using a microplate reader (VICTOR X2030, Perkin Elmer, Milano, Italy), and expressed as relative fluorescence units (RFU). After 48 h, and 14 and 28 days, total RNA was extracted from the cells seeded on the samples, using the commercial RNeasy Mini Kit (Purelink™ RNA miniKit, Ambion by Life Technologies, Carlsbad, CA, USA), quantified by a NANODROP spectrophotometer (NANODROP 2720, Thermal Cycler, Applied Biosystem), and reverse transcribed using the Superscript Vilo cDNA synthesis kit (Life Technologies). Gene expression was evaluated by semiquantitative PCR analysis, using the SYBR green PCR kit (QIAGEN GmbH, Hilden, Germany), in a Light Cycler 2.0 Instrument (Roche Diagnostics, GmbH, Manheim, Germany). Ten nanograms of cDNA were tested in duplicate for each sample. The protocol included a denaturation cycle at 95 °C for 15 min, 25 to 40 cycles of amplification (95 °C for 15″, appropriate annealing temperature for each target, as detailed in Table 1, for 20″, and 72 °C for 20″), and a melting curve analysis to check for amplicon specificity. The mean threshold cycle was determined for each sample and used for the calculation of relative expression using the 2−ΔΔCt method, with GAPDH as the reference gene and Ti-POR as the calibrator [41]. For histological and SEM analyses, the same processing method was performed on each specimen until the dehydration step in ethanol 70%. Briefly, the samples were fixed in 2.5% glutaraldehyde, in pH 7.4 phosphate buffer 0.1 M, for 1 h, and subsequently dehydrated in a graded ethanol series (30–50–70%) for 15 min each. As regards histological analyses, after the preliminary phase, the samples dedicated to histological processing were further dehydrated in successive passages in 95% and 100% alcoholic solutions. Subsequently the samples were infiltrated in two different acrylic-based solutions (Methacrylate; Merck, KGaA, Darmstadt, Germany) until polymerization. Then, the samples were cut perpendicularly to the major axis of the cylindrical samples, with the Leica SP1600 diamond blade microtome (Leica Microsystems S.r.l., Buccinasco, Italy), obtaining about 15 cross sections for each sample. Consecutive sections were thinned and smoothed using abrasive papers with different granulation, using the Saphir sanding system (Saphir 550, ATM GmbH, Mammelzen, Germany), to obtain a final thickness of the sections of about 50 ± 10 µm. Three of these sections were then subjected to histological staining with Toluidine blue and Fast green, observed under the Olympus BX51 optical microscope (BX51, Olympus Optical Co. Europe GmbH, Hamburg, Germany), and acquired digitally using the Aperio Scanscope digital scanner (CS System, Aperio Technologies, Vista, CA, USA). A further staining with Stevenel’s blue and Picrofucsin, according to Van Gieson, was carried out on another three histological sections of the Ti-POR, COLL–CS–COLL, and CS–COLL–CS samples, only at 28 days. After the preliminary phase, samples were further dehydrated for 15 min in 95% ethanol and 100% ethanol, for 1 h. Then they were placed in Hexamethyldisilazane (Sigma Aldrich, Co., St. Louis, MO, USA) for two successive passages of 5 min each, and left to air dry. At the end of these steps the samples were coated in gold for 60 s, using an Agar Sputter Coater (Agar Scientific, Stansted, UK), and mounted on sample holders for SEM, using carbon conductive adhesive discs. The backscattered electronic images (backscattered-BS) were acquired using an SEM, model EVO LS HD (Carl Zeiss S.p.A, Milan, Italy), using the SmartSEM software (version 5.07, Carl Zeiss AG, Oberkochen, Germany), at different magnifications, and in backscattered, to evaluate the adhesion and colonization of the surface by hBMSCs. Data were analyzed by using the R software, version 4.2.1 [42]. After having checked the non-normal distribution (Shapiro–Wilk test) and non-homogeneity of variance (Levene test) of data, Kruskal–Wallis χ2 test, followed by non-parametric Mann–Whitney U test were used for comparisons among groups. Data are reported as mean ± SD, at a significant level of p < 0.05. In conclusion, the new two types of coatings made by three layers, COLL–CS–COLL or CS–COLL–CS, are not cytotoxic and induce hBMSC adhesion, sustain their viability, and induce bone matrix deposition on the surface of, and inside, a Ti-alloy implant, without differences between the two coatings. In comparison to the non-coated specimen (Ti-POR), the cell viability results were higher, and bone matrix deposition was observed earlier on the coated specimens. It can therefore be stated that both coatings used do not interfere with the osteogenic differentiation process of hBMSCs and with the deposition of new bone matrix.

PMC10002511

Chokri Zaghdoud,Irene Ollio,Cristóbal J. Solano,Jesús Ochoa,Juan Suardiaz,Juan A. Fernández,María del Carmen Martínez Ballesta

Red LED Light Improves Pepper (Capsicum annuum L.) Seed Radicle Emergence and Growth through the Modulation of Aquaporins, Hormone Homeostasis, and Metabolite Remobilization

01-03-2023

aquaporins,Capsicum annuum,LED lighting,radicle emergence,seed germination,water uptake

Red LED light (R LED) is an efficient tool to improve seed germination and plant growth under controlled environments since it is more readily absorbed by photoreceptors’ phytochromes compared to other wavelengths of the spectrum. In this work, the effect of R LED on the radicle emergence and growth (Phase III of germination) of pepper seeds was evaluated. Thus, the impact of R LED on water transport through different intrinsic membrane proteins, via aquaporin (AQP) isoforms, was determined. In addition, the remobilization of distinct metabolites such as amino acids, sugars, organic acids, and hormones was analysed. R LED induced a higher germination speed index, regulated by an increased water uptake. PIP2;3 and PIP2;5 aquaporin isoforms were highly expressed and could contribute to a faster and more effective hydration of embryo tissues, leading to a reduction of the germination time. By contrast, TIP1;7, TIP1;8, TIP3;1 and TIP3;2 gene expressions were reduced in R LED-treated seeds, pointing to a lower need for protein remobilization. NIP4;5 and XIP1;1 were also involved in radicle growth but their role needs to be elucidated. In addition, R LED induced changes in amino acids and organic acids as well as sugars. Therefore, an advanced metabolome oriented to a higher energetic metabolism was observed, conditioning better seed germination performance together with a rapid water flux.

Red LED Light Improves Pepper (Capsicum annuum L.) Seed Radicle Emergence and Growth through the Modulation of Aquaporins, Hormone Homeostasis, and Metabolite Remobilization Red LED light (R LED) is an efficient tool to improve seed germination and plant growth under controlled environments since it is more readily absorbed by photoreceptors’ phytochromes compared to other wavelengths of the spectrum. In this work, the effect of R LED on the radicle emergence and growth (Phase III of germination) of pepper seeds was evaluated. Thus, the impact of R LED on water transport through different intrinsic membrane proteins, via aquaporin (AQP) isoforms, was determined. In addition, the remobilization of distinct metabolites such as amino acids, sugars, organic acids, and hormones was analysed. R LED induced a higher germination speed index, regulated by an increased water uptake. PIP2;3 and PIP2;5 aquaporin isoforms were highly expressed and could contribute to a faster and more effective hydration of embryo tissues, leading to a reduction of the germination time. By contrast, TIP1;7, TIP1;8, TIP3;1 and TIP3;2 gene expressions were reduced in R LED-treated seeds, pointing to a lower need for protein remobilization. NIP4;5 and XIP1;1 were also involved in radicle growth but their role needs to be elucidated. In addition, R LED induced changes in amino acids and organic acids as well as sugars. Therefore, an advanced metabolome oriented to a higher energetic metabolism was observed, conditioning better seed germination performance together with a rapid water flux. Seed germination is usually considered the most critical stage in seedling establishment, which determines successful crop yield and quality [1]. The process of germination begins with the seed imbibition of water, triggering arrested physical and metabolic activities, including cell expansion, cell division, and reserve mobilization [2], and ends with radicle protrusion [3]. Generally, seeds uptake water during germination in a typical triphasic pattern, initiating with a rapid water absorption (Phase I, imbibition) followed by a plateau phase (Phase II, little net water uptake with metabolic preparation for radicle protrusion), and a second burst of water uptake (Phase III, coupled with radicle emergence) [4]. The micropyle and hilum of seeds are the main regions involved in the absorption of water [5], which is moved into the seed tissue and organs via three different pathways: apoplastic, symplastic, and transcellular paths. The movement of water and other small neutral molecules via the transcellular path—traversing through cell membranes—involves aquaporins (AQPs), which are water-selective channels [6,7]. In higher plants, AQPs are divided into five major subfamilies: plasma membrane intrinsic proteins (PIPs), tonoplast intrinsic proteins (TIPs), NOD26-like intrinsic proteins (NIPs), small basic intrinsic proteins (SIPs), and the uncategorized X intrinsic proteins (XIPs) [8]. In turn, these subfamilies can be further divided into multiple isoforms, depending on the localization and functional properties [9]. Numerous studies have showed the important function of orthodox seed AQPs in water imbibition and subsequent germination, with the expression patterns being spatiotemporal specific at the transcription level [10,11,12]. Indeed, the dynamics of gene expression during germination reveals the implication of two time-separated sets of AQPs [13]: (i) the first set of AQPs is potentially involved in the initial germination process, being the transcript level abundant in dry seed just before initial rehydration, and dramatically reduced during imbibition and germination; and (ii) the second set of AQPs drives the rapid cellular expansion and embryo growth, where the transcripts are increased at, or immediately after, germination. The low levels of AQP gene expression during Phase I of germination [12,14,15] are attributed to the dry seed AQP transcripts stored, ready for immediate translation upon initial rehydration [13,16,17], rather than AQPs non-functionality. PIPs and TIPs are the predominant AQPs in plant seeds. However, whereas PIPs function as water uptake and transport across cell-plasma membranes during seed imbibition and the early growth of the embryo [18], seed-specific TIPs act mainly from the beginning of Phase II as markers of protein remobilization to maintain turgor pressure [12,19]. During Phase II, vacuolation occurs through transition from cells small protein storage vacuoles (PSVs) to large central lytic vacuoles (LVs), leading to significant increase in cellular turgor pressure and the activation of several hydrolytic enzymes [11,20]. This transition was accompanied with a shift from TIP3s to TIP1/TIP2 gene expression [17,19]. TIP3 are enriched in PSVs, whereas TIP1 and TIP2 are preferentially localized in LVs and vegetative storage protein vacuoles, respectively [21], which revealed implication of these isoforms in protein mobilization, a process that is crucial for embryo-cell elongation and radicle emergence [22]. Nevertheless, the dual localization of TIP3 in tonoplast and plasma membrane highlights their role in favouring optimal water uptake during seed imbibition [20]. In addition, transcriptionally, TIP4;1 expression begins post germination [23]. Amino acids play vital roles in the metabolism of seeds during germination. They are used for the synthesis of seed-storage proteins and are precursors of secondary metabolites that are a source of energy for seeds [24]. Also, organic acids may represent the store pools of carbon in metabolic pathways, taking part in the mobilization of stored lipids and carbohydrates during seed germination [25]. Finally, seed germination is also controlled by a crosstalk between plant hormones, including gibberellins (GAs), abscisic acid (ABA), ethylene (ET), auxins (AUX), cytokinins (CKs), and brassinosteroids (BRs) [26], and metabolizable sugars derived from storage that appear to have both nutritional and signalling stimulant functions [27]. ABA specifically inhibits the endosperm rupture and Phase III water uptake by the emerging embryo, but does not alter the spatial and temporal pattern of Phase I and II water uptake [28]. In contrast, GAs, ethylene, and BRs enhance germination by antagonistically suppressing ABA, weakening the tissue surrounding the embryo, which increases the embryonic growth potential, thereby helping radicle protrusion [29,30]; the signal-transduction pathways and mechanisms are hormone specific [31]. GAs and ABA antagonistically modulated α-amylase activity and the transformation of small PSVs into LVs in the aleurone cells of seed endosperm, with GA and ABA acting as an inducer and inhibitor of this transformation, respectively [32]. AUXs (particularly indole-3-acetic acid [IAA]) are involved in the cell-wall loosening and polysaccharide mobilisation mediated by hydroxyl radicals (•OH), processes needed for cell elongation and division [33]. Interestingly, Finkelstein and Lynch [27] reported that glucose suppressed the inhibitory effects of ABA on radicle emergence in a light-dependent manner, although the genetic components involved were not identified. During Phase III of germination, sucrose and hexose transporters, as well as H+-ATPase, become the fundamental proteins involved in the active cell elongation to support radicle extension [34]. Hormone-driven regulatory responses also control water uptake during seed germination through the regulation of AQP gene expression [13]. In small-seeded plants such as Arabidopsis, lettuce, tomato [35,36], and pepper [37], germination was under phytochrome-mediated photocontrol, with light being a critical environmental determinant during this process. Recently, the use of light-emitting diodes (LEDs) has spread to plant production purposes; they are more efficient than fluorescent lamps [38], since they allow a tight control of waveband emission and light intensity with low energy consumption [39]. Red (R, 600–700 nm) and blue (B, 400–500 nm) lights are the most-used source of radiation for plant production in a controlled environment, as these wavelengths are most readily absorbed by photoreceptor phytochromes [40]. Thus, R and B LED lights were more efficient for seed germination and plant growth than other wavelengths, such as green (500–600 nm) and far red (700–800 nm) [41]. The effects of LED light on seed germination are species specific and spectra dependent [42]. Indeed, when lettuce (Lactuca sativa L.) seeds were exposed to different LED light wavelengths, the highest induction of germination was observed in cv. Banchu Red Fire under R light (at 50 Hz and duty ratio of 20%) [43], and in cv. Levistro under lights with a higher B component [44]. In oilseed rape (Brassica napus ‘Modena’), R (638 nm) and B (450 nm) LED lights significantly increased and decreased the seed germination percentage, respectively [45]; whereas, at the same intensity and photoperiod, both LED lights inhibited germination in soybean (Glycine max) seeds [46]. Cho et al. [47] reported that light enhanced the germination rate through the activation of photoreceptor phytochrome B, which mediates GA signalling and metabolism. Most reports on the effect of LED lights on germination have studied seed parameters (germination percentage, GP; germination speed index, GSI; radicle length, RL; hypocotyl length, HL). However, to the best of our knowledge, no works were done illustrating their impact on water uptake and radicle emergence processes with the involvement of aquaporins. Since water plays a crucial role in germination, the aim of this study was to investigate the relationship between better germination performances, through the application of Red LED irradiation, and seed water status. In addition, AQP isoforms involved in water transport will be determined, as well as the changes in the hormone and metabolite profiles under these conditions. The kinetics of water uptake by pepper seeds during Phases I, II, and III of germination (from 0 to 96 h) were monitored in the presence or absence of azide, with or without a previous pulse of R LED irradiation (Figure 1). Upon exposure to sufficient moisture, the water content for the different treated seeds showed a rapid increase during the first 6 h of imbibition, followed by a gradual progression from 6 h to 10 h, reaching a plateau phase (Phase II), then the curve of variation stabilized from 10 h to 48 h. After radicle emergence (at 48 h), a second gradual increase (Phase III) was reached until 96 h, when radicles were 1 cm long. In all phases, hydration was faster in R LED-irradiated seeds in comparison with the rest of the treatments. Seed hydration was the lowest under azide treatment (7.35% lower than Controls after the first 6 h). However, a pulse of R LED irradiation before azide addition suppressed the inhibitory effect of azide on seed imbibition, with the values of water uptake under combined R LED and azide treatments similar to those of Controls. A time course (from 48 h to 96 h) of the percentage of radicle protrusion was obtained for Control and R LED-irradiated seeds (Figure 2). At all measured time points, the highest rates of radicle emergence were recorded for seeds previously irradiated by R LED light. For both treatments, the rates of radicle emergence followed a sigmoid pattern of evolution, with an exponential behaviour starting from 60 h for R LED-treated seeds and from 67 h for Control. At 67 h of imbibition, the percentage of emerged radicles was 2.42-fold higher in R LED-lighted seeds, compared to Control. The tetrazolium test showed no statistical differences regarding the percentage of embryo viability between Control and R LED-irradiated seeds (Table 1), reflecting the innocuous effect of R LED light on seed tissues. R LED-irradiated seeds showed at 96 h a significantly increased germination percentage (GP) and higher germination speed index (GSI) than Controls, by about 5.7%, that together with a significantly higher percentage (~23.86%) of humidity after imbibition during 96 h, and subsequent lyophilisation, reflect the higher ability of R LED-treated seeds to adsorb water faster. The expression of different AQP isoforms was determined in seeds with protruding radicles after 96 h of imbibition in water. NIP1;2, NIP3;1, NIP4;1, NIP4;6, and TIP5;1 isoforms were not detected in our sweet pepper seeds. A heat-map graph characterizes the relative expression profile of detected isoforms: 12 PIPs, 15 TIPs, 5 NIPs, 2 XIPs, and 1 SIP in Control and R LED-irradiated pepper seeds (Figure 3). This map reflects the isoforms’ expression relative to NIP1;1 expression of each individual treatment. Transcript analysis demonstrated that, for both treatments, most PIP and TIP subfamily genes were much more highly expressed than NIP, SIP, and XIP subfamily genes, being the expression treatment specific. In Control seeds, TIP1;7, TIP1;8, TIP3;1, and TIP3;2 were the TIP isoforms that showed the highest expression levels; among PIPs, PIP1;4, PIP1;5, PIP2;3, PIP2;5, and PIP2;8 were the highest expressed. In R LED-lighted seeds, in addition to those mentioned for Controls, other AQPs isoforms showed high expression compare to NIP1;1 including TIP1;4, TIP1;6, TIP4;1, and PIP2;3. The expression profiling of AQP isoforms, comparing both Control and R LED light treatments, were also determined at 96 h of pepper seed imbibition (Figure 4). Comparison between isoforms was realized considering NIP1;1 expression of CON rootlets as reference for both treatments. Regarding PIP gene expression (Figure 4A), significant differences among the treatments were found for PIP2;3 and PIP2;5, which showed a significant increase in expression (by 43.99% and 44.02%, respectively) in irradiated R LED seeds with respect to the Control, and PIP1;5 and PIP2;8, which had a significant decrease in expression by 49.71% and 41.73%, respectively. Among the TIP isoforms, significant differences between treatments were observed for TIP1;7, TIP1;8, TIP3;1, and TIP3;2 gene expressions, where reductions by about 41.08%, 34.43%, 34.99% and 37.24%, respectively, were recorded in R LED-lighted seeds, compared to Controls (Figure 4B). The expression profiling of NIPs showed that NIP1;1 and NIP4;2 isoforms were also significantly lower expressed (by 35.12% and 24.75%, respectively) in seeds previously exposed to R LED irradiance, relative to Controls, while NIP4;5 was significantly higher expressed by 70.91% (Figure 4C); NIP6;1 and NIP7;1 expression did not differ significantly among the treatments. Significant differences in XIP expression were only observed for XIP1;1, which showed an increase in expression by about 46.72% in irradiated seeds with respect to the Controls (Figure 4D). SIP1;1 was detected but no significant differences were found between both treatments. Significant differences (p < 0.05) in the hormone levels in pepper seeds with protruding radicles—after 96 h of imbibition—among Control and R LED-lighting treatments were observed with changes in the amount of salicylic acid (SA), indole-3-butyric acid (IBA), ABA, and jasmonic acid (JA) (Table 2). Endogenous levels of SA were significantly increased (by about 32.03%) in R LED-irradiated seeds compared to Controls, while those of JA were significantly decreased (by 11.65%). Interestingly, IBA appeared only in R LED-irradiated seeds and ABA was found exclusively in Control seeds. IAA, GA3, and 6-BA were not detected (n.d). The sugars quantified with 1H-NMR in the seeds of sweet pepper were fructose, glucose, and sucrose (Figure 5A,B). Myo-inositol was not detected for any treatment in the pepper seeds, although its determination was also carried out. Of those quantified, the highest amount quantified was for sucrose (Figure 5B), followed by fructose and glucose (Figure 5A). For all of them, except glucose, the R LED irradiation affected their concentration. Regarding to Control, previously R LED irradiated pepper seeds had a significant percentage of 8.53% lower sucrose content after radicle emergence during germination (at 96 h of imbibition) (Figure 5B), and 17.55% significantly higher fructose concentration (Figure 5A). However, no statistical differences were observed between treatments with regard to glucose levels. The amino acids detected with 1H-NMR in pepper seeds were GABA, Alanine (Ala), Asparagine (Asn), Aspartate (Asp), Glutamate (Glu), Glutamine (Gln), Isoleucine (Ile), Leucine (Leu), Phenylalanine (Phe), Proline (Pro), Tryptophan (Trp), Tyrosine (Tyr), and Valine (Val). The most abundant transported amino acids with a high N:C ratio during germination in seeds were Glu followed by Asn and Pro (Figure 5C). In general, no changes in the amino acids content were found in R LED-irradiated seeds compared to Control seeds. However, significant differences between treatments were recorded for Ala and Gln, which showed significantly lower concentrations (reductions of 10.42% and 38.93%, respectively) in R LED-lighted seeds, compared to Controls, and Ile, of which the concentration was significantly lower by about 9.52% (Figure 5C). In addition, the other major forms of amino acids, Asn and Pro, showed similar seed levels under both treatments (Figure 5C). The organic acids detected in pepper seeds were acetate, citrate, lactate, and malate, whereas formate, fumarate, and succinate were not detected. Changes in the concentrations of organic acids were observed for citrate and malate, which were significantly accumulated (by 20.04% and 17.88%, respectively) in R LED-irradiated seeds compared to the Controls (Figure 5D). Interestingly, other metabolites such as chlorogenate and trionelline were not detected although they were analysed, but choline concentration was reduced by 13.19% in R LED-lighted seeds relative to Controls (Figure 5E). For a better and simpler visual interpretation of all the data (metabolites for the different experimental repetitions of Control and R LED-irradiated seeds), a principal component analysis (PCA) was conducted (Figure 6). The PC1 component explained 92.75% of the variability observed, showing that the variability was fundamentally due to acetate, Tyr, Asp, Trp, Phe, Val, Glu, Leu, glucose, malate, citrate, SA, and IBA in R LED-irradiated seeds. PC2 (8.60%) was positively associated with Control seeds and metabolites as choline, ABA, ALA, ILE, GABA, sucrose, JA, and lactate were represented. The rapid and uniform seedling emergence and growth is a basic requirement for an adequate crop yield and quality. In different reports it has been shown that the time course of water uptake by seeds during germination includes three phases [2]. The results of this study clearly showed these three phases for pepper seed germination and Red LED irradiation greatly modified the kinetics of water uptake by pepper seeds, increasing not only the amount of absorbed water, but also the rate of water uptake during Phase I and Phase II of germination. Sodium azide (NaN3) was reported to induce intracellular acidosis by blocking respiration via the cytochrome pathway. This acidosis results in H+-dependent closure of PIPs [48,49], thereby inhibiting water permeability across cell membranes [50]. In pepper seeds, NaN3 reduced water uptake in Control and R LED-irradiated seeds, especially during Phase III. Different studies have indicated the importance of AQPs in seed imbibition and subsequent germination [10,51,52,53]. The function of seed AQPs may be related to the water imbibition and activation of the metabolic system in the seeds, which results in higher germination [52]. However, Vander Willigen et al. [15] applied mercury as an AQP inhibitor and found that AQPs are not involved in the early imbibition phase (Phase I). Similarly, sodium azide had no significant effect on the water uptake of pepper seeds during the Phase I with regard to Control, but it reduced the water adsorption when the seeds were previously irradiated with R LED light. It was shown that R LED light increased the activity of cytochrome c oxidase (CCO), the terminal enzyme of the electron transport chains of mitochondria [54]. However, the exact mechanism by which LED irradiation enhanced photorespiration in plants is still unclear. In human cells, it has been proposed that R LED light induced the absorption of a photon by the copper subunit of the CCO [55], which enhances the ability of the mitochondria to increase the rate of oxidative phosphorylation [56]. In addition, Quiroga et al. [57] found that some insensitive AQPs (ZmPIP2;4, ZmPIP2;5 and ZmTIP1;1) increased their protein levels in maize plants when they were treated with NaN3 under normal irrigation. Therefore, the observed suppressed effect of R LED light on NaN3 inhibition of seed hydration from the first 10 h could be due in part to the recovery in mitochondria function, modulating pH-dependent gating of sodium-azide sensitive AQPs, together with an enhancement of sodium azide-insensitive AQP activity. The kinetic of water uptake was also faster under R LED light treatment, from 18 h to 48 h (plateau phase or phase II), compared with the rest of treatments. These results could be explained by previous findings where far lower concentrations of reactive oxygen species (ROS) and higher intracellular calcium ([Ca2+]i) were recorded in mitochondria actively making ATP [58]—such as under R LED irradiance, leading to higher AQP activity [59,60], and thereby faster water movement into seeds. Finally, during Phase III, the phase most influenced by AQP activity, faster and more effective hydration of the embryonic tissues may lead to a reduction of the germination time [61], which is the case of our R LED-irradiated seeds where GSI was significantly higher than Controls. In this work, radicle emergence occurred from 48 h after imbibition in both irradiated and Control seeds, being the percentage higher in the former at all measured times and the exponential radicle protrusion earliest (from 60 h, compared to 67 h for Controls), reflecting a faster cell-wall loosening and greater water uptake for cell elongation. Cellular expansion during Phase III of germination is driven by important water uptake enabling radicle emergence through both the loosened endosperm and testa [62,63]. At 96 h of imbibition, the significantly higher humidity of R LED irradiated seeds also confirmed their higher capacity for water absorption during radicle emergence compared to Controls, without affection of seed viability. Several works have reported the role of different AQP isoforms in seed water uptake during radicle emergence [12,15,23]; however, this study is the first to describe effects of R LED light on AQP gene expression during that process. Interestingly, here differences in the expression pattern of all pepper AQP isoforms in response to R LED irradiance were observed inside and between treatments. Obroucheva [64] reported that PIP and TIP genes were the most strongly activated AQPs during radicle emergence, due to their implication in providing sufficient water entry into embryo-axis cells that are beginning to elongate. These results were in accordance with our results for both Control and R LED-irradiated seeds. In Control seeds, higher levels of transcripts were observed for TIP1;7, TIP1;8, TIP3;1, and TIP3;2 isoforms with regard to the NIP1;1. TIP genes were functionally mainly related to the water transportation required for the enzymatic metabolism of storage nutrients in protein storage vacuoles (PSVs), process that directly preceded cell elongation [17,65], maintaining pressure potential in the vacuolar lumen (VL) at the first hours of radicle emergence [66]. However, whereas TIP3;1 was shown to facilitate water transport [67], TIP3;2 was found to facilitate transport of the osmolyte glycerol rather than water [68]. Control seeds showed also high expression levels of PIP1;4, PIP1;5, PIP2;5, and PIP2;8 isoforms after 96 h of imbibition. Alterations in PIP1s and PIP2s gene expression may have a crucial role in governing effective water transport from cell to cell in the expanding tissue after radicle protrusion [61]. PIP1;4 and PIP1;5 isoforms were reported to pre-exist in dry seeds, participating in their development, with the expression being maintained at high levels in mature seeds, as reported by Shiota et al. [69]. Here, it seems that these two isoforms were also implicated in all the process of seed germination. In contrast, the expression of PIP2;5 and PIP2;8 isoforms were strongly increased in the embryo of rice seeds only a few hours before radicle emergence, suggesting their involvement in that process [52]. In R LED-irradiated seeds, in addition to those of Controls, three TIP (TIP1;4, TIP1;6, and TIP4;1) and PIP2;3 isoforms also showed higher expression regarding NIP1;1. It has been shown that TIP4;1 had dual functions in both glycerol and water transport [68], in which the transcript abundance became more important after radicle protrusion, indicating its involvement in the rapid expansion of radicle cells [13,23]. Similarly, PIP2;3 was not expressed until radicle emergence in germinating rice seeds, followed by increased expression with seedling growth, suggesting it functions in seedling establishment rather than seed germination [52]. These findings suggest that R LED-irradiated seeds expressed some of the AQPs involved in the post-germination radicle growth. Comparing both non-irradiated and R LED-irradiated seeds, PIP2;3 and PIP2;5 isoforms were significantly more expressed in the latter, whereas PIP1;5 and PIP2;8 expression was significantly lower. These results showed a transition in the expression of AQP genes from those implicated in early seed germination phases to those involved in radicle emergence and growth. This expression pattern was in consonance with a faster germination rate in R LED-irradiated seeds. The fact that PIP2;8 expression—which was showed to be expressed before radicle emergence—was lower in R LED-irradiated seeds, indicating the involvement of PIP2;8 in this process. Regarding TIP isoforms, lower transcript abundance was observed for TIP1;7, TIP1;8, TIP3;1, and TIP3;2 in R LED-irradiated seeds, relative to Controls. It has been reported that expression of TIP genes in PSVs ceased when radicles emerged from the seed coat and these proteins disappeared in parallel with the transformation of PSVs to LVs [17]; they were most likely replaced by other TIP isoforms [70]. From our results, it appeared that R LED light accelerated LV biogenesis in the embryonic axes of pepper seeds for the accumulation of endogenous osmotic solutes in elongating cells. Differential gene expression for NIPs was also observed in this work, where NIP1;1 and NIP4;2 were significantly lower in R LED-irradiated seeds, while NIP4;5 was significantly higher. NIP1;1 is localised in the plasma membrane [71], with a developmental pattern that closely resembles TIP3;1 and TIP3;2 [72,73]. Therefore, the involvement of NIP1;1 with TIP3 in a developmentally controlled compensation/complementation of PIPs during early seed germination cannot be discarded. It also showed to be important for direct membrane traffic from the endoplasmic reticulum to the tonoplast [74,75]. Therefore, the lower expression of this isoform in R LED-irradiated seeds seems to be the result of reducing the delivery of proteins to the transformed PSVs after radicle emergence. Finally, our results showed a significantly higher expression of XIP1;1 isoform in R LED-irradiated seeds, even though its function in seed germination is not yet determined. We could state that R LED-irradiated seeds are in a more advanced statement of radicle emergence with regard to Controls, as higher AQPs isoforms involved in water uptake for cell elongation rather than for stored reserve mobilisation were observed compared to Control, manifested by greater PIP and lower TIP gene expression. In accordance with the transcript profiles of AQP isoforms, R LED light also induced metabolic changes in pepper seeds at 96 h of imbibition, in order to fulfil the growing energy demand of the developing embryo and endosperm, and to provide sufficient osmotic pressure in elongating cells. Indeed, R LED-irradiated seeds accumulated more fructose than Controls that showed higher levels of sucrose. It has been reported in germinating Glycine max [76], Moringa oleifera [77], and Quercus ilex [78] seeds that sucrose, the most abundant reserve carbohydrate in quiescent seeds, was conversed to fructose and glucose throughout germination and early seedling development. Therefore, the substantial rate of import of the products from reserve mobilization into glycolysis in R LED-irradiated seeds could increase the flux of pyruvate into the tricarboxylic acid (TCA) cycle [79], thereby providing more energy, as a carbon source, for faster radicle emergence [80,81]. Also, the content of total sugars was higher for Controls than for R LED-irradiated seeds, which explains their higher expression in TIP genes involved in the remobilization of carbohydrates. An advanced metabolome in R LED-lighted seeds relative to the Controls was also observed in our study by significantly higher levels in key intermediates for TCA and glyoxylate cycles, citrate, and malate, which also implies a higher energetic metabolism that would be required for better seed germination performance [82]. In contrast, the significantly lower levels of alanine and isoleucine in R LED-irradiated seeds suggests their incorporation into the gluconeogenesis pathway to provide energy supply for protruded seeds. Indeed, several analytical works based on 14C labelling showed that amino acids in germinating seeds contribute a large amount of carbon substrate to the respiratory system and sugar synthesis [83,84]. Glucogenic alanine and isoleucine are catabolized into TCA cycle pyruvate and succinyl CoA, respectively, which converts to oxaloacetate, the substrate for the gluconeogenic enzyme PEP carboxykinase [85]. However, the oxidative deamination of amino acids produces ammonia [86], which if not re-assimilated, induces defects in the TCA cycle [87]. To ensure ammonium re-assimilation, glutamate dehydrogenase specifically in the developing embryo axis contributes ammonium delivery to glutamine synthetase for glutamine synthesis, in the absence of primary NO3- assimilation [86]. In turn, after being transported into the cytosol, citrate can be metabolized by aconitase to support nitrogen assimilation by producing glutamate or glutamine as an end-product [88,89]. Thus, compared to Controls, the significantly higher glutamine content in R LED seeds at 96 h of imbibition suggested greater nitrogen assimilation [90]. Citrate was also shown to mitigate drought, salinity, temperature, and heavy metal stresses in a variety of plant species [91]. Thus, when growing directly in soil, previous R LED-lighted seeds pepper will probably show better tolerance to these abiotic factors than Controls. However, this fact cannot be deduced from the present data and new studies are needed. The results of metabolic profiling showed that R LED light significantly stimulates gluconeogenesis and mitochondrial respiration in germinating pepper seeds, which provides sufficient energy needed for faster cell elongation and radicle protrusion. The tetrazolium (2,3,5-triphenyl-tetrazolium chloride) test is used to assess seeds’ viability based on the ability of mitochondrial respiration through the electron transport chain [92]. Living cells in seed tissue use the hydrogen released by dehydrogenase enzymes during the chemical reduction of tetrazolium to form triphenyl-formazan, a red, stable, and not diffusible compound [93]. Here, no significant difference between the treatments was observed regarding seed viability, confirming that a pre-germination pulse of R LED light for pepper seeds did not affect their ability to germinate and establish their seedlings. In this study, pepper seeds showed a preferential photoblastic response, as they germinated under both irradiated and not irradiated conditions, as observed for Murdannia nudiflora seeds [94], being the GP significantly higher in R LED-lighted seeds at 96 h of imbibition. Similar results were obtained by Wang et al. [95], where R LED light promoted the seed germination of Momordica charantia (bitter gourd), and kept the germination potential at a high level. The light induction of germination is exclusively mediated by phytochrome B and other phytochromes that are maximally induced by a saturating pulse of monochromatic R light [96], in which the photoreceptor pigment is activated as a switch between 640 nm and 670 nm. The absorption of R light converts the inactive cytosol-localized Pr form of phytochrome (which inhibits germination) to the active nucleus-accumulated Pfr form (which promotes germination) [97]. Previous studies have reported that phytochrome light perception regulates the germination of dicot seeds via the modulation of expression of genes involved in phytohormone metabolism [98,99]. Here, at 96 h of imbibition, ABA was detected only in Control seeds, but not in R LED-irradiated ones. It has been reported that R light induces the expression of the gene CYP707A2 via phytochrome B, which provided ABA degradation and inactivation in both embryo and endosperm during germination [100,101]. In addition, a loss of ABA extrusion from the hypocotyl-radicle transition zone in Medicago truncatula seeds was observed to delay radicle emergence by preventing cell-wall loosening and cell elongation [102], which could be the case in our Control seeds. JA levels were significantly lower in R LED -lighted seeds than in Controls. JA and its derivates were found to inhibit seed germination by disrupting the peroxisomal ATP binding cassette transporter or the core β-oxidation process [103]. However, in accordance with our results, a promoting effect of R light on adventitious root initiation was also observed in Picea abies seedlings, likely by reducing JA biosynthesis [104]. It is important to note that JA generally works in synergy with ABA during germination through a crosstalk signalling, where ABA acts as suppressor and JA enhances ABA function [105,106]. In contrast to JA, the SA concentration was higher in R LED-lighted seeds with regard to the Controls. An increase in SA levels was also reported in Pachyrhizus erosus grown under R LED-light irradiation [107] and in soybean sprouts germinated under R light [108], as a result of PAL stimulation and IC synthase gene up-regulation, which are enzymatic pathways for SA biosynthesis. The accumulation of SA under R light was found to induce SA signalling, mediating the production of ROS [109]. Among ROS, hydrogen peroxide (H2O2) had a direct or indirect negative effect on ABA transport from the cotyledon to the embryonic axis, resulting in a decrease in ABA, which induced a mitogen-activated protein kinase (MAPK)-mediated reduction in the ethylene precursor 1-aminocyclopropane carboxylic acid, favouring epicotyl and radicle emergence [110]. Many reports have also described an antagonistic interaction between the SA and JA pathways [111,112]. SA was also reported to stimulate the accumulation of glycine betaine (GB), providing a nitrogen source for better germination [113], protecting the membrane functions, and increasing the activity of the antioxidant system [114]. However, GB biosynthesis occurs mainly through the degradation of choline [115], which could explain the significantly lower levels of choline in our R LED-irradiated seeds. Here, in opposition to ABA, IBA was only detected in R LED seeds. IBA is an IAA precursor which is converted into active IAA through a β-oxidation process in the peroxisome [116]. However, IBA β-oxidation releases not only free IAA, but also acetyl-CoA [117]. Thus, because IBA is important for early seedling growth, a stage in which peroxisomal activity is high in metabolizing storage oils [118], and because no amounts of IAA were detected, it may be possible that IBA is used in R LED seeds to provide energy to drive embryo growth. As unexpected, no GA3 or 6-BA was detected in both Control and R LED-lighted seeds at 96 h of imbibition, suggesting that GA3 and 6-BA are either not produced until the radicle emerges or are produced at extremely low levels, below the detection limit of our equipment. The lack of IAA could also induce a negative feedback on the biosynthesis of GA3 [119]. Taken together, R LED light adjusts phytohormone metabolism to be in consonance with the advanced germination stage of irradiated seeds. Taking all metabolites into consideration, differences in the influence of those on Control and R LED-irradiated seeds are clear, with metabolites involved in the advanced phases of germination and early seedling development conditioning R LED-irradiated seeds. The seeds of sweet pepper (Capsicum annum L. var. Medrano F1) from Ramiro Arnedo S.L were used for all experiments. For different measurements, seeds were lined in transparent Petri dishes with moist Whatman filter paper (Merck, Darmstadt, Germany). Then, R LED light was applied for 15 min. After that, 30 mL of distilled water was added to cover the seeds for assays, except for viability determination, where the seeds were placed in a recipient with 200 mL of distilled water. For kinetic imbibition measurement, the initial imbibition time was counted after 15 min of water addition. For the rest of measurements and after water addition, the seeds were place in a germination chamber (in darkness at 25 °C, 60% RH) until 96 h after starting imbibition (seeds with radicle or rootlets were analysed). Control seeds (CON) were established in similar conditions without the R LED light-pulse application. The number of repetitions of each treatment (CON and R LED) as well as the amount of the seeds in each determination was variable and depended on the analysed parameter (see distinct methods sections). In this work, a prototype of modulable spectrum plant experimental chamber (MSPEC) was developed for testing the effects of specific R LED light (Osram, Madrid, Spain) on the first 96 h of pepper seed germination (imbibition and radicle emergence steps). To visualize the light distribution pattern along the chamber, the photosynthetic active radiation (PAR) was measured at different coordinates on the work surface of the MSPEC for 10 R LEDs (Figure 7A). The acquired data were processed with a Matlab script (Mathworks, Natick, MA, USA), from which a graphical representation of PAR distribution values was derived (Figure 7B). Ten LEDs were used for the germination tests (Figure 7C). They provided PAR values of around 100 µmol m−2 s−1 in most of the surface which allowed the arrangement of up to four Petri dishes on it. The sensors for temperature and humidity provided data ranging from 23 °C to 25 °C and from 55% to 60% RH, respectively, on all the work area. The LED chosen was manufactured by the company OSRAM. Specifically, it was selected the GH CSSRM5.24 OSLON ® Square model, a compact high-power hyper red LED with proven robustness, high reliability, long lifetime, and low thermal resistance. It presents a peak wavelength at 660 nm (Figure 7D), a typical radiant flux of 1064 mW @ 700 mA and a photosynthetic photon flux of 5.82 μol/s @ 700 mA. A total of 50 seeds were placed in Petri dishes with moist Whatman paper. Before measuring the kinetic of water adsorption, half of the seeds from each treatment (CON and R LED-irradiated seeds) were immersed in sodium azide (NaN3) at a final concentration of 7 mM [48]. The seeds were maintained under NaN3 immersion during the first hour of imbibition to allow the azide to penetrate through the seed coat. After that and for the rest of time-course of measurement of water imbibition, the seeds were placed in distilled water in order to avoid azide toxicity. The standard germination test (SGT) was carried out according to the International Seed Testing Association (ISTA) rules [120] with modifications. For that, 50 seeds of each treatment (CON and R LED) were distributed in five rows in transparent Petri dishes with Whatman paper previously moistened with distilled water. The seeds were then placed inside a germination chamber at a temperature of 25 °C and 60% relative humidity in the dark. A total of five replications (each 50 seeds) for each treatment (CON and R LED) were considered. Counting started from the first day. Germination was calculated using Equation (1) and expressed as a percentage, where GP is the germination percentage, ni the total number of germinated seeds, and N the total count seeds sampled. The final germination was considered after 96 h when the radicle was 0.50 cm. The germination speed index (GSI) was obtained through the methodology proposed by Martínez-Solís et al. [121]. Germinated seeds were counted daily, seeds with sprouted radicles were considered, and Equation 2 was applied, where GSI is the germination speed index, Ti is the time in hours passed between the test start and the end of the interval, and Ni is the number of germinated seeds within consecutive time intervals. The viability analysis was determined using tetrazolium chloride, as it was described by ISTA [120]. One hundred seeds were placed in a recipient with 200 mL of distilled water, and this was put in a water bath at a temperature of 35 °C for 14 h. Later, a 1% tetrazolium chloride solution (2 mL) was added, and the recipient with the seeds was put in a water bath at a temperature of 35 °C for 4 h. Finally, the seeds were rinsed with distilled water and examined under a microscope (LEICA, EZR®, L’Hospitalet de LLobregat, Spain). The embryos were classified according to colour intensity: (1) alive with high vigour, when they were completely dyed with an intense red colour, (2) alive with low vigour, when their coloration was a pale red, and (3) not viable, when they remained colourless. The classified seeds were expressed as the percentage of viable and unviable embryos. A total of five replications for each treatment were realized. Using an analytical balance (Model: 161 RADWAG PS 4500/C2), 25 g of seeds were weighed for each treatment and grouped on transparent Petri dishes with filter paper in 5 pools per treatment (CON, LED, azide, and LED+azide). Next, 6 mL of distilled water was added in order to moisten the filter paper. Then, R LED light was applied for 15 min. After that, 30 mL of distilled water was added to the seeds to cover them for 96 h at room temperature (23 °C). The increase in weight was registered during this time and the amount of water adsorbed (g g−1 DW) was expressed through Equation (3), where Wa is the water adsorbed, Wi is the initial weight, Wf is the final weight, and Hf is the percentage humidity content (that it was similar for Control and R LED-treated seeds before the start of imbibition) [122]. A total of five replications for each treatment was realized. After 96 h of imbibition, 0.5 g of seeds with radicle or rootlets, per treatment and experimental replicate (n = 3), were frozen and ground to a fine powder in liquid nitrogen. Total RNA was extracted using the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany), according to the manufacturer’s instructions. Contaminating DNA in the samples was removed with DNase I, using the DNA-free Kit (Ambion, Applied Biosystems, Austin, TX, USA), and the RNA concentration was quantified with a Nanodrop 1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). After that, the extracted RNA was stored at −80 °C until use. The cDNA was synthesised from 2 μg of total RNA, using the M-MLV reverse transcriptase from the RETROscript Kit (Ambion, Applied Biosystems, Austin, TX, USA). Reverse transcription was carried out with heat denaturation of the RNA, according to the manufacturer’s instructions. To compare the expression of distinct AQP isoforms under the different treatments, qRT-PCR was performed as described by Muries et al. [123], in an Applied Biosystems 7500 Real-Time PCR system. The primers and the lengths of the amplicons for PIP, TIP, NIP, SIP, and XIP isoforms were those described by Uppuluri et al. [124] (Supplemental Table S1). After denaturation at 95 °C for 10 min, amplification occurred in a two-step procedure: 15 s of denaturation at 95 °C and 1 min of annealing and extension at 60 °C, followed by a dissociation stage. Data collection was carried out at the end of each round in step 2. These conditions were used for both target and reference genes and the absence of primer dimers was checked in controls lacking templates. The amplifications were performed on three independent samples for each treatment and seeds (biological replicates) and triplicate reactions were carried out for each sample (technical replicates) in 96-well plates. The transcript levels were calculated using the 2−ΔΔCt method [125], for both target and reference genes. Standard curves (log of the cDNA dilution vs. Ct) using serial 10-fold dilutions of the cDNA were built for each pair of selected primers, obtaining 95% PCR efficiency, corresponding to a slope of −3.561. Actin and ubiquitin were used as house-keeping genes for the standardisation of each sample [126]. ΔCt (target gene) calculates as Ct (target gene) − Ct (house-keeping gene). ΔΔCt was then calculates as ΔCt (target gene)—ΔCt (reference sample gene). The final result of this method is presented as the fold change of target gene expression in a target sample, relative to a reference sample (CON NIP1;1), normalized to each reference gene. With both standard genes, similar relative expressions were obtained. A “non-targeted” metabolic analysis was conducted in the seeds with radicle or rootlets after 96 h of water hydration. For that, 0.5 g of rootlets were ground with liquid nitrogen with a mortar and pestle and lyophilized. Afterwards, the samples were prepared for analysis according to the protocol by Van de Weijer and Schrauwen-Hinderling [127]. For this analysis, a Nuclear Magnetic Resonance (NMR) system coupled to a 500 MHz Bruker spectrometer (Bruker Biospin, Rheinstetten, Germany) equipped with a broadband 5 mm N2 CryoProbe Prodigy BBO. The seed extracts were measured at 300.1 ± 0.1 K without rotation and with 4 test scans before the 32 scans performed for the experiment. The acquisition parameters were set in the following manner: the size of the FID = 64 K, spectral band = 12.4345 ppm, receiver gain = 28.5, acquisition time = 2.18 s, relaxation delay = 2 s, and line broadening = 0.50 Hz. The acquisition of data was performed through the NOESY pulse sequence of pre-saturation (Bruker 1D, noesypr1d) with water suppression through the irradiation of the water frequency during the recycling and mixing times. In the processing of the samples and for each spectrum separately, a reduction of noise was produced, based on the deconvolution of the multi-level signal. Afterwards, a correction was performed of the baseline, and to complete the process, an interpolation technique of the areas of the signal was utilized. All of this provides us with a “fingerprint” of the sample, a general view of the metabolites that are most represented produced by the cells at time of harvest, expressing the chemical shifts (d) in parts per million (ppm). The NMR equipment detects the signals and records them as frequency versus intensity graphic, known as the “acquisition spectrum”. The resulting 1H-NMR spectra were processed with the Chenomx NMR Suite program version 8.3 (Chenomx, Edmonton, AB, Canada), in order to identify and quantify the metabolites of interest. All the samples were calibrated with the signal from the internal standard (IS), the deuterated Trimethylsilylpropionic acid sodium salt (TSP-d4) and the pH was set to a value of around 6. The software utilized includes a broad range of spectrum data that can be utilized to detect the metabolites that are over 5–10 mM: among the metabolites that were found and/or quantified, the following are highlighted: Alanine, Aspartate, Asparagine, Glutamate, Glutamine, Isoleucine, Leucine, Phenylalanine, Proline, Tryptophan, Tyrosine, Valine, Acetate, Citrate, Lactate, Malate, Fructose, Glucose, Sucrose, GABA, and Choline. A total of five replications for each treatment were realized. Sample preparation was carried out according to Müller and Munné-Bosch [128]. For that, frozen material (about 100 mg fresh weight of seeds with radicles or rootlets) was ground in liquid nitrogen with the mixer mill MM400 (Retsch GmbH, Haan, Germany) in a 2 mL Eppendorf tube, and then extracted with 1 mL of extraction solvent (methanol:isopropanol, 20:80 (v/v) with 1% of glacial acetic acid) using ultra sonication (4–7 °C). The labelled forms of the compounds d4-SA, d6-ABA, d5-JA, d5-IAA, d2-GA1, d2-GA4, d2-GA9, d2-GA19, d2-GA20, d2-GA24, d4-ACC, d6-2iP, d6-IPA, d5-Z, and d5-ZR were added as internal standards. D5-Z and d5-ZR were used as internal standards for DHZ and DHZR, respectively. After centrifugation (10,000 rpm for 15 min at 4 °C), the supernatant was collected, and the pellet was re-extracted with 0.5 mL of extraction solvent, and the extraction repeated three times again. Then, supernatants were combined and dried completely under a nitrogen stream and re-dissolved in 300 μL of methanol, centrifuged (10,000 rpm for 5 min) and filtered through a 0.22 μm PTFE filter (Waters, Milford, MA, USA). Samples (5 μL) were then analysed by UPLC/ESI-MS/MS. Hormones were determined in five independent samples for each treatment. Quantification was done by the creation of calibration curves including each of the 17 unlabelled analyte compounds (SA, ABA, JA, IAA, GA1, GA4, GA9, GA19, GA20, GA24, ACC, 2iP, IPA, Z, ZR, DHZ, and DHZR). Ten standard solutions were prepared ranging from 0.05 ng mL−1 to 1000 ng mL−1 and for each solution a constant amount of internal standard (as described above) was added. Calibration curves for each analyte were generated using Analyst™ software (Applied Biosystems, Inc., California, USA). The limit of detection (LOD, S/N = 3) and the limit of quantification (LOQ, S/N = 10) were also calculated with the aid of this software. UPLC-QToF-MS/MS was performed using a Waters ACQUITY UPLC I-Class System (Waters Corporation, Milford, MA, USA) coupled to a Bruker Daltonics QToF-MS mass spectrometer [maXis impact Series with a resolution ≥ 55,000 FWHM Bruker Daltonics, Bremen, Germany], using ESI for both positive- [ESI (+)] and negative- [ESI (−)] ionization modes. UPLC separation was performed using an HSS T3 C18 100 × 2.1 mm column with 1.7 µm of size particle (Waters Corporation, Milford, MA, USA) at a flow rate of 0.4 mL min−1. The separation was performed using H2O with acetic acid (AcOH) at 0.05% (pH ~3.20) (PanReac AppliChem, Barcelona, Spain) as the weak mobile phase (A) and ACN with AcOH at 0.05% (J. T. Baker, Morristown, NJ, USA) as the strong mobile phase (B). The gradient started with 1% of B at 0 min, which was progressively increased to 99% at minute 5; after that, it was held until minute 7.50 and then decreased to 1% at 10 s, where it remained until minute 10.50. Nitrogen was used as the desolvation gas with a flux of 9 L min−1 and as the nebulizing gas with a pressure of 2.0 bars. The drying temperature was 200 °C and the column temperature was 25 °C. The voltage source was 4.0 kV for ESI (−) and 4.5 kV for ESI (+). The MS experiment was carried out using HR-QTOF-MS, applying 24 eV for ESI (+) and 20 eV for ESI (−) and using broadband collision-induced dissociation (bbCID). MS data were acquired over an m/z range of 50–1200 Da. The external calibrant solution was delivered by a KNAUER Smartline Pump 100 with a pressure sensor (KNAUER, Berlin, Germany). The instrument was calibrated externally before each sequence with a 10 mM sodium formate solution. The mixture was prepared by adding 0.5 mL of formic acid and 1.0 mL of 1.0 M sodium hydroxide to an isopropanol/Milli Q water solution (1:1, v/v). A total of five replications for each treatment were realized, considering each replication a different sample extraction. The statistical analysis included a one-way ANOVA with the SPSS program v24. When the ANOVA was significant (p < 0.05), Tukey’s HSD test for the separation of means was applied for p < 0.05. A principal component analysis (PCA) was carried out to identify and classify the principal metabolites to be used in distinguishing Control and R LED-irradiated seeds. This PCA was conducted on the mean of distinct values recorded for different metabolites, from five tests for each treatment (Control and R LED) using SPSS (PASW Statistics18.0) computer package (SPSS 2006). R LED light applied to pepper seeds in a short pulse enhanced the kinetics of water uptake by seeds at all germination phases, increasing their germination performance. The symplastic water movement for embryo growth during germination was modulated by the R LED irradiance. Indeed, a faster vacuolization manifested by reduction in the expression of TIP genes (TIP1;7, TIP1;8, TIP3;1, and TIP3;2), together with increased water absorption for cell elongation through the higher expression of PIP genes involved in water transport (PIP2;3 and PIP2;5) led to faster radicle emergence and greater GP in R LED irradiated seeds, compared to Controls. Reduced NIP1;1 and NIP4;2 expression could be related with the reduction of delivery of proteins from the protein storage vacuoles after radicle emergence. In any case, the role of NIP and XIP in seed germination needs further studies. The faster water uptake in R LED-irradiated seeds was in consonance with a more advanced energetic metabolism involving monosaccharides, organic acids, and amino acids, in order to satisfy the energy required for faster germination. The R LED light also affected the accumulation of the hormones of germination activating those involved in radicle emergence and growth. Therefore, after R LED irradiation, a direct link between the rates of water transport, seed carbohydrate metabolism, hormone regulation, and radicle growth can be established. If advanced germination induced by R LED light is an advantage for the plant to cope with stress in the environment, it must be studied as well as the effect of different spectral lights on seed germination stages.

PMC10002514

Xianshu Luo,Zhuoyu Han,Qing Kong,Yuming Wang,Haijin Mou,Xuefeng Duan

Clostridium butyricum Prevents Dysbiosis and the Rise in Blood Pressure in Spontaneously Hypertensive Rats

04-03-2023

Clostridium butyricum,butyrate,hypertension,short chain fatty acid,dysbiosis

Hypertension is accompanied by dysbiosis and a decrease in the relative abundance of short-chain fatty acid (SCFA)-producing bacteria. However, there is no report to examine the role of C. butyricum in blood pressure regulation. We hypothesized that a decrease in the relative abundance of SCFA-producing bacteria in the gut was the cause of spontaneously hypertensive rats (SHR)-induced hypertension. C. butyricum and captopril were used to treat adult SHR for six weeks. C. butyricum modulated SHR-induced dysbiosis and significantly reduced systolic blood pressure (SBP) in SHR (p < 0.01). A 16S rRNA analysis determined changes in the relative abundance of the mainly SCFA-producing bacteria Akkermansia muciniphila, Lactobacillus amylovorus, and Agthobacter rectalis, which increased significantly. Total SCFAs, and particularly butyrate concentrations, in the SHR cecum and plasma were reduced (p < 0.05), while C. butyricum prevented this effect. Likewise, we supplemented SHR with butyrate for six weeks. We analyzed the flora composition, cecum SCFA concentration, and inflammatory response. The results showed that butyrate prevented SHR-induced hypertension and inflammation, and the decline of cecum SCFA concentrations (p < 0.05). This research revealed that increasing cecum butyrate concentrations by probiotics, or direct butyrate supplementation, prevented the adverse effects of SHR on intestinal flora, vascular, and blood pressure.

Clostridium butyricum Prevents Dysbiosis and the Rise in Blood Pressure in Spontaneously Hypertensive Rats Hypertension is accompanied by dysbiosis and a decrease in the relative abundance of short-chain fatty acid (SCFA)-producing bacteria. However, there is no report to examine the role of C. butyricum in blood pressure regulation. We hypothesized that a decrease in the relative abundance of SCFA-producing bacteria in the gut was the cause of spontaneously hypertensive rats (SHR)-induced hypertension. C. butyricum and captopril were used to treat adult SHR for six weeks. C. butyricum modulated SHR-induced dysbiosis and significantly reduced systolic blood pressure (SBP) in SHR (p < 0.01). A 16S rRNA analysis determined changes in the relative abundance of the mainly SCFA-producing bacteria Akkermansia muciniphila, Lactobacillus amylovorus, and Agthobacter rectalis, which increased significantly. Total SCFAs, and particularly butyrate concentrations, in the SHR cecum and plasma were reduced (p < 0.05), while C. butyricum prevented this effect. Likewise, we supplemented SHR with butyrate for six weeks. We analyzed the flora composition, cecum SCFA concentration, and inflammatory response. The results showed that butyrate prevented SHR-induced hypertension and inflammation, and the decline of cecum SCFA concentrations (p < 0.05). This research revealed that increasing cecum butyrate concentrations by probiotics, or direct butyrate supplementation, prevented the adverse effects of SHR on intestinal flora, vascular, and blood pressure. The intestinal flora’s potential role in influencing host health has attracted considerable attention in recent decades. Many metabolic diseases, inflammatory bowel diseases, and cardiovascular diseases have been reported to be linked to flora disorders [1,2,3]. In recent years, much seminal evidence has shown that abnormal intestinal flora is closely associated with changes in blood pressure in the host. Iñaki Robles-Vera and co-workers reported that fecal microbiota transplantation from adult spontaneously hypertensive rats (SHR), to adult Kyoto rats (WKY), resulted in a chronic rise in blood pressure (BP), vascular oxidative stress, and impaired endothelial function. Conversely, fecal microbiota transplantation from WKY to adult SHR induced a blood pressure reduction and improvement of endothelial dysfunction [4]. Mechanisms linking the gut microbiota to hypertension include dysbiosis, inflammation, gut permeability, and decreased production of short-chain fatty acids (SCFAs), particularly butyrate [5,6,7]. SCFAs are mainly produced through fermentation of indigestible carbohydrates by bacteria in the intestine [8], which can regulate body weight, energy metabolic balance, lipid metabolism, and other pathophysiological processes [9,10], including hypertension. SCFAs can be transported to the surface of the cell membrane, through mono-carboxylate transporter (MCT), to bind with G protein-coupled receptors (GPCRs), and modulate vascular tone and inflammation to regulate blood pressure through the interaction with GPCRs or inhibition of histone deacetylases [11,12]. Onyszkiewicz and co-workers demonstrated that butyrate could enter the circulation through the intestinal-vascular barrier, and act on GPR41/GPR43 to relax mesenteric arteries and decrease blood pressure [13]. In addition, treatment of Olfr78 knockout and non-knockout mice with propionate revealed that Olfr78 knockout mice showed lower blood pressure levels [14]. It has been reported that the probiotics Bifidobacterium breve CECT7263 and Lactobacillus fermentum CECT5716, could regulate the blood pressure of SHR through modulating short-chain fatty acid-producing genera in the intestine [15]. SHR is an established model of genetic hypertension characterized by elevated blood pressure, arterial remodeling, endothelial dysfunction, hyperlipidemia, vascular inflammation, and immune system dysregulation [16,17,18]. In addition, dysbiosis and a decrease in the relative abundance of SCFA-producing bacteria have been reported in SHR [19,20]. Clostridium butyricum (C. butyricum) is an anaerobic, gram-positive bacillus, known as a butyrate producer and a regulator of gut health [2]. It resides in the gastrointestinal tract and has a protective role against pathogenic bacteria and intestinal injury, via the modulation of gut microbial metabolites [21,22,23]. C. butyricum are capable of utilizing a range of carbohydrates, and produce several fermentation products, with butyrate as the major product. Butyrate plays a crucial role in promoting gut health and has been used in clinical practice to treat various chronic enteric diseases [24]. While probiotics and SCFAs have the potential to regulate blood pressure, we did not know if C. butyrcium, the main butyrate producer, could regulate blood pressure in SHR. Based on the potential of probiotic and bacterial-derived SCFAs for blood pressure regulation, we hypothesized that C. butyricum could modulate intestinal flora, improve vascular inflammation, and lower blood pressure. We investigated the hypotensive effect of C. butyricum on SBP in SHR, using SHR animals as a model, and investigated the effects of SHR, WKY, and C. butyricum on intestinal immunity and vascular inflammation, a key component of SHR-induced hypertension. In addition, we analyzed the potential mechanistic link between dysbiosis and hypertension by 16S rRNA V1–V9 sequencing. In fed rats, the SBP of WKY remained around 100 mmHg, which was highly significantly different from that of SHR (p < 0.01). The SBP of SHR gradually increased between week 9 and week 13, and stabilized at 189 ± 7 mmHg by week 15. Given that multiple animal models of hypertension exhibit reduced SCFA-producing bacteria, we treated SHR with C. butyricum or butyrate (the main metabolite of C. butyricum). The SBP of SHR remained steadily lower throughout the treatment period of C. butyricum or butyrate. By week 15, the SBP of rats in the SHR-Cb group was 50 mmHg lower than that of the rats in the SHR group (Figure 1b,c). In contrast, there was no significant difference in SBP between rats in the WKY-C. butyricum group compared to those in the WKY group (Figure 1a). This suggests that C. butyricum was able to prevent elevated SBP in SHR but did not affect normotensive rats. Captopril lowers blood pressure in hypertension. The results of the repeated measures ANOVA showed that C. butyricum treatment significantly reduced the SBP in SHR after two weeks (Tables S1–S3). In addition, to clarify the effect of C. butyricum in preventing elevated blood pressure, we treated SHR with captopril for 6 weeks. The results showed that the SBP of SHR was significantly reduced after 2 weeks of treatment (Tables S1–S3), but C. butyricum or butyrate had a significant effect in preventing elevated SBP after 4 to 6 weeks of treatment (Figure 1d). To investigate the effect of C. butyricum and butyrate intervention on gut microbiology, we analyzed the full-length 16S rRNA sequencing of the colon contents. The richness and diversity of the intestinal flora were lower in SHR than in WKY (p < 0.05). C. butyricum increased the richness and diversity of the colonic flora by increasing the values of Chao and Shannon at the OUT level (Figure 2a), while butyrate did not alter the abundance of colonic flora. Similarly, C. butyricum and butyrate showed a clear separation from SHR alone in both unweighted and weighted UniFrac principal coordinate analyses (Figure 3). Firmicutes, Bacteroidetes, Actinobacteria, Proteobacteria, Fusobacteria, and Verrucomicrobia are the main phyla in the gut, with Firmicutes and Bacteroidetes accounting for 90% of the intestinal flora [25,26]. As shown in Figure 2c, Firmicutes were significantly increased in the SHR colon compared to WKY (WKY: 67%, SHR: 80%), while the relative abundance of Proteobacteria in the SHR colon was 6% higher than that of WKY. C. butyricum and butyrate significantly increased the relative abundance of Verrucomicrobia and Bacteroidetes in the SHR colon (p < 0.01). This indicates that C. butyricum and butyric acid altered the intestinal microbial community of SHR. On the genus level, the relative abundance of Akkermansia in the colon of SHR was significantly increased by C. butyricum and butyrate (SHR: 4%, SHR-C. butyricum: 33%, SHR-butyrate: 18%) (Figure 2c). The Firmicutes/Bacteroides ratio is a marker of intestinal health. Some diseases such as inflammatory bowel disease and excessive obesity, including hypertension, have been reported to have significantly higher Firmicutes/Bacteroides ratios, and regulation of the ratio is effective in treating the disease [27,28]. We analyzed Firmicutes/Bacteroides in all groups of rats, and we demonstrated that Firmicutes/Bacteroides were significantly higher in SHR than WKY (p < 0.01). Treatment with C. butyricum and butyrate prevented Firmicutes/Bacteroides from being elevated (p < 0.01) (Figure 2b). In addition, we analyzed changes in colonic microorganisms on the species level of SHR after 6 weeks of C. butyricum and butyrate treatment. Consistent with previous studies, short-chain fatty acid-producing bacteria were decreased in the SHR gut. Akkermansia muciniphila, Lactobacillus amylovorus, Muribaculum sp002492595, Agthobacter rectalis, Romboutsia ilealis, and lleibacterium valens had their relative abundance altered by C. butyricum and butyrate (Figure 2c–h). We demonstrated that C. butyricum and butyrate prevented the decrease in the relative abundance of Akkermansia muciniphila, Muribaculum sp002492595, and Agthobacter rectalis in the colonic contents caused by SHR (Figure 2c–e). C. butyricum and butyrate significantly increased Akkermansia muciniphila (SHR-C. butyricum: 34%, SHR-butyrate: 23%) (Figure 2c), interestingly, Roshanravan N. and co-workers reported that supplementation with butyrate promoted Akkermansia muciniphila levels in the intestine, improved vascular inflammation and oxidative stress, and lowered blood pressure [29,30], which is consistent with the results of our study. Both Lactobacillus amylovorus and lleibacterium valens have been reported to reduce obesity [31,32]. The results showed that colonic Lactobacillus amylovorus increased after 6 weeks of C. butyricum treatment, but that butyrate did not alter their relative abundance in the colon (Figure 2d), while treatment with C. butyricum and butyrate did not alter the colonic levels of lleibacterium valens (Figure 2h). In addition, treatment with both C. butyricum and butyrate reduced the levels of the SHR colonic pathogen Romboutsia ilealis (Figure 2g). We examined the effects of SHR on the colon and vascular system with and without C. butyricum and butyrate. Figure 4b shows that SHR caused severe intestinal mucosal detachment in the colon, goblet cells in the colon were reduced, and there was a proliferation of inflammatory fibrous tissue; all of these were prevented by treatment with C. butyricum and butyrate. In addition, we observed that SHR caused endothelial cell damage, proliferation, and shedding of outer membrane cells in the aorta. Treatment with C. butyricum and butyrate improved the vascular damage caused by SHR (Figure 4b). Since SHR caused inflammation in the colon and vascular system, we next examined serum levels of inflammatory factors. ELISA demonstrated that C. butyricum and butyrate decreased tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), interleukin-17A (IL-17A), and lipopolysaccharide (LPS), and increased the inhibitory inflammatory factor interleukin-10 (IL-10) in the SHR circulation (Figure 4a,c–f). This suggests that the regulation of the SBP in SHR by C. butyricum and butyrate possibly correlates with reduced levels of inflammation. To clarify the effect of dysbiosis on SHR, we next tested whether the SHR-induced changes to the microbiota led to changes in microbial metabolites that may contribute to the adverse effects of SHR on the gut and blood pressure. Microbiota analysis showed that some short-chain fatty acid producers were modified by SHR, which was prevented by C. butyricum and butyrate (Figure 2d–h). Therefore, we measured the SCFAs concentrations in colonic contents and plasma. SHR caused a significant decrease in colonic and plasma total SCFAs, particularly butyrate concentrations (Figure 5). C. butyricum and butyrate significantly prevented the SHR-induced decrease in colonic butyrate, acetate, and propionate concentrations, but the effects on acetate and propionate were not significant (Figure 5c and Figure 6). SHR caused a decrease in total SCFAs in plasma, but we found that SHR did not change the concentration of propionate in plasma (Figure 5b and Figure 6d). We tested the relative expression levels of three monocarboxylate transporter proteins involved in transporting butyrate and other SCFAs into and through the intestinal epithelium [11,33]. Figure 7a,b shows that the relative expression levels of MCT1 and MCT4 were significantly reduced in the colon of SHR compared to WKY (p < 0.05), which was enhanced by treatment with C. butyricum or butyrate. Similarly, the relative expression level of Slc5a8 was also enhanced by C. butyricum or butyrate, but it was not significant (n = 4 per group, p = 0.229 for SHR vs. SHR-C. butyricum, p = 0.324 for SHR vs. SHR-butyrate). This result suggested that the decrease in circulating total SCFAs and butyrate concentration may be due to a decrease in SCFAs transport proteins caused by SHR. To evaluate the direct effects of SCFAs on the vasculature, we examined the relative expression levels of SCFA-sensing receptors in the aorta. Most notable among the SCFA targets is the mammalian G protein-coupled receptor pair of GPR41 and GPR43, that can be expressed in blood vessels [34]. We observed reduced relative expression levels of GPR41 and GPR43 in the SHR aorta (GPR41, p = 0.051; GPR43, p = 0.004 for WKY vs. SHR). After 6 weeks of treatment, the mRNA expression levels of GPR41 and GPR43 in the SHR- C. butyricum group and SHR-butyrate group were upregulated compared with those in the SHR group, meanwhile, compared with the C. butyricum group, the results showed that butyrate treatment group had a more significant regulatory effect (p < 0.01). Therefore, the mechanism for sensing SCFAs appears to be partially compromised in the SHR aorta, but was ameliorated by treatment with C. butyricum or butyrate. The host inflammation is associated with a balance between the pro-inflammatory and anti-inflammatory actions of regulatory T cells. The Th17/Treg balance has been shown to normalize endotoxemia, prevent endothelium-dependent diastolic damage to acetylcholine, and reduce blood pressure in spontaneously hypertensive rats [30]. Therefore, we examined the distribution of Th17 and Treg in the aorta and spleen. Figure 8 shows that SHR leads to Th17 increases and Treg decreases in the spleen and aorta. These were ameliorated by treatment with C. butyricum or butyrate. In recent years, much seminal evidence has demonstrated for the first time that abnormal gut flora is closely associated with changes in blood pressure in the host [35]. Both animal models of hypertension and human hypertension are accompanied by dysbiosis [36]. Through the study of the effect of C. butyricum on the intestinal flora, we have demonstrated that C. butyricum could regulate the intestinal flora and increase the relative abundance of short-chain fatty acid-producing bacteria [37]. We hypothesized that a reduction in SCFAs was responsible for SHR hypertension. Probiotics such as Bifidobacterium breve CECT7263 and Lactobacillus fermentum CECT5716, have been shown to restore SHR-induced dysbiosis and reduce SBP [15]. These probiotics were found to ferment dietary fiber in the gut to produce SCFAs, which protect against the vascular oxidative stress and endothelial dysfunction caused by hypertension. Several studies have shown that the relative abundance of many SCFA-producing bacteria is reduced in animal models of hypertension [19,20,38]. Bhanu P. Ganesh and co-workers have demonstrated that C. butyricum reduced the effect of hypertension in a model of obstructive sleep apnea on altered microbiota, and increased the relative abundance of many SCFA-producing genera such as Parabacteroides, Roseburia, Clostridium, Bifidobacterium, Ruminococcus, and Blauti [39]. We suggested that C. butyricum and butyrate reduced the effect of SHR on altering the microbiota. Unweighted and weighted UniFrac principal coordinate analyses were used to show that the flora of SHR and WKY were significantly separated, altered by C. butyricum and butyrate. Chao and Shannon showed that C. butyricum and butyrate restored the reduction in flora diversity and abundance caused by SHR. In addition, treatment with C. butyricum and butyrate significantly reduced the Firmicutes/Bacteroides ratio (p < 0.01). We concluded that SHR induced dysbiosis and elevation of Firmicutes/Bacteroides, which was prevented by C. butyricum or butyrate. On a species-level analysis of all rat colon flora, we found that C. butyricum and butyrate prevented SHR-induced decreases in colonic Akkermansia muciniphila, Muribaculum sp002492595, and Agthobacter rectalis levels, which is consistent with previous findings. Akkermansia muciniphila is a native bacterium in the gut that ferments carbohydrates in the intestine to produce acetate and propionate, reducing metabolic disorders and improving low levels of inflammation [40]. Levels of intestinal Akkermansia muciniphila are negatively correlated with diabetes, obesity, and other metabolic syndromes [41]. Interestingly, previous studies have reported that supplementation with butyrate promotes intestinal levels of Akkermansia muciniphila, reduces vascular inflammation and oxidative stress, and lowers blood pressure [30]. Thus, Akkermansia muciniphila might play a significant role in reducing the SBP in SHR. In conclusion, C. butyricum was demonstrated to modulate the intestinal flora of SHR, reduce the Firmicutes/Bacteroides ratio, and prevent the SHR-induced reduction of short-chain fatty acid-producing species, thereby potentially maintaining colonic butyrate levels. SCFAs, a major source of energy for epithelial cells, have many beneficial effects, including maintaining the integrity of the intestinal barrier [42], reducing mucosal inflammation, and improving intestinal health [43,44]. We found that SHR caused a significant reduction in colon and plasma SCFAs, particularly butyrate concentrations. C. butyricum and butyrate prevented the SHR-induced decrease in colon butyrate concentrations, but had no significant effect on acetate and propionate. SHR caused a decrease in the total SCFAs of plasma, but it did not modify plasma propionate concentrations. SCFAs could affect the host by activating the G protein-coupled receptor (GPCR) or by inhibiting histone deacetylases, to stabilize the intestinal epithelial barrier, regulate cytokine secretion, alter the T lymphocyte population, increase the protective mucus layer, and regulate antibody secretion [45,46,47,48]. If SCFAs entered the circulatory system, it would also affect tissues and organs outside the intestinal tract [49]. We observed that SHR induced a reduction in plasma concentrations of SCFAs, which was not significantly improved by C. butyricum and butyrate; so we examined the relative expression of SCFAs transporters in the proximal colon and SCFA-sensing receptors. We demonstrated that SHR caused a reduction in colonic MCT1 and MCT4 expression, which was prevented by C. butyricum or butyrate. Similarly, the reduction in the SCFA-sensing receptors GPR-41 and GPR-43 mRNA expression in the aorta of SHR, was ameliorated by C. butyricum or butyrate. Treatment with C. butyricum and butyrate increased the concentration of intestinal SCFAs, while having no significant effect on plasma SCFAs. Therefore, we examined the possibility of colonic and aortic effects. Our findings in this study showed that the number of mucus-producing goblet cells of the colon was reduced, and there was intestinal mucosal detachment in SHR, and this could be prevented by C. butyricum and butyrate treatment. Similarly, Santisteban and co-workers reported a reduction in goblet cells in the SHR colon [50]. Similarly, C. butyricum and butyrate improved SHR-induced vascular injury. LPS could increase the intestinal permeability, and an increased intestinal permeability would allow more LPS to enter the circulation and exacerbate the inflammatory state [51]. Iñaki Robles-Vera and co-workers reported that LPS levels in SHR serum were significantly higher than WKY, and after treatment with the probiotic Bifidobacterium bifidum and SCFAs, blood pressure and LPS were normalized in the treated group of rats. They hypothesized that a reduction in LPS in rat serum was a key factor in the modulatory effect of probiotics on hypertension [15]. Our results have shown that treatment with C. butyricum and butyrate reduced LPS levels in serum of SHR. In addition, TNF-α is a pro-inflammatory factor produced by macrophages, and plays an important pathogenic role in inflammatory bowel disease. The downregulation of TNF-α can significantly downregulate the inflammation of Crohn’s disease, and the content is positively correlated with inflammation [52]. IL-10 is an anti-inflammatory cytokine produced by immature T cells. Previous studies have shown that 17 strains of clostridium from healthy human microbiota can induce the production of IL-10, thus inhibiting colitis. The probiotic C. Butyricum can promote the production of IL-10 by T cells and thus prevent the occurrence of colitis through an IL-10-dependent mechanism [53]. IL-6 and IL-17A are pro-inflammatory factors produced by Th17 cells. IL-6 is an important cytokine in the process of immune inflammatory reaction, its abnormal content will damage the vascular endothelium and lead to an increase in blood pressure in hypertensive patients [54]. We observed that serum levels of the inflammatory factors TNF-α, IL-6, and IL-17A were increased by SHR and decreased by C. butyricum or butyrate. Conversely, IL-10 was increased by C. butyricum or butyrate. These findings suggest that C. butyricum and butyrate ameliorate vascular inflammation and modulate the SBP of SHR, which may be linked to the modulation of inflammatory factor levels. The balance between Th17 and Treg is critical for maintaining immune homeostasis, with the over-activation of Th17 cells exacerbating intestinal inflammation, while the lack of Treg in intestine-associated lymphoid tissue, or its inability to circulate naturally to sites of inflammation, has been shown to cause an immune response in commensal flora, and to induce colitis [55]. In this study, to analyze the effect of SHR on Th17 and Treg, we performed immunofluorescence staining of the spleen and aorta. We concluded that SHR induced an increase in splenic and aortic Th17 and a decrease in Treg, which improved through treatment with C. butyricum or butyrate. We have shown that (1) SHR caused a decrease in colonic SCFAs, especially butyrate concentrations, and increased intestinal and vascular inflammation, and hypertension. (2) in SHR rats, the probiotic C. butyricum, and SCFA butyrate, increased total colonic SCFAs and butyrate concentrations, reduced dysbiosis and colonic injury, and prevented vascular inflammation and hypertension. (3) C. butyricum and butyrate regulated the expression level of the major SCFAs transporters MCT1 and MCT4, and the receptors GPR41 and GPR43, in SHR rats. They may effectively regulate the colonic inflammation of SHR by regulating the concentration of SCFAs. (4) C. butyricum and butyrate modulated the signal pathway, to elevate the anti-inflammatory level by reducing the expression level of pro-inflammatory factor TNF-α, IL-6 and IL-17A, and improving the expression level of anti-inflammatory factor Il-10. (5) C. butyricum and butyrate decreased the level of intestinal LPS, and then ameliorated the intestinal barrier dysfunction caused by SHR. These findings demonstrate the critical role of impaired butyrate production in the development of SHR-induced hypertension, and suggest that treatment targeting increased C. butyricum and microbial butyrate production may prove effective in treating hypertension (Figure 9). C. butyricum CGMCC 1.5205 (C. butyricum) was preserved at the China General Microbiological Culture Collection Center (CGMCC). A thermo-static incubator (DRP-9052, Senxin, Shanghai, China) was used to cultivate C. butyricum in reinforced Clostridium medium for 24 h at 37 °C. 16S rDNA sequencing (27F: AGAGTTTGATCCTGGCTCAG, 1492R: TACGGCTACCTTGTTACGACTT) was performed on the broth of C. butyricum, and the sequencing results were compared on the NCBI website (https://www.ncbi.nlm.nih.gov (accessed on 25 January 2022)) to confirm that the strain cultivated was C. butyricum. The C. butyricum fermentation broth was centrifuged at 10,000× g rpm for 10 min and 20% trehalose was added to the bacterial pellets as a protective agent before freeze-drying. The freeze-dried bacteria were counted using the blood counting chamber, with a result of 1010 cfu/mL. The freeze-dried powder was stored until use. All studies in animals should be conducted in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals, or the equivalent. All experimental procedures were approved by the Animal Ethics Committee of Ocean University of China (Approved protocol no: SPXY2017050402). Thirty-two male, 5-week-old spontaneously hypertensive rats (SHR) and 12 Kyoto rats (WKY) were purchased from Beijing Viton Lever Laboratory Animal Technology Co. (Beijing, China). After 4 weeks of acclimatization feeding, 32 SHR rats were randomly divided into 4 groups (n = 8) and 12 WKY rats were randomly divided into 2 groups (n = 6), according to body weight and blood pressure. The WKY group (gavaged with 3 mL of sterilized saline NaCl 0.8%), WKY C. butyricum group (WKY-Cb) (3 mL of C. butyricum freeze-dried powder dissolved in sterilized normal saline (108 CFU/mL) was administered to the rats), SHR group (gavaged with sterilized saline NaCl 0.8%), SHR Clostridium butyricum (SHR-C. butyricum) (3 mL of C. butyricum freeze-dried powder dissolved in sterilized normal saline (108 CFU/mL) was administered to the rats), SHR butyrate group (0.5 mg/kg), and SHR captopril group (SHR-CAP) (captopril 10 mg/kg, Changzhou Pharmaceutical factory). All groups were gavaged once a day. Rats were housed in individual ventilated cages, in a pathogen-free animal facility under temperature (20–22 °C) and humidity (50–60%) control, and with a pre-set light-dark cycle (12 h:12 h). During the experimental period, the rats had free access to tap water and feed. All rats were executed after 6 weeks, and the intestinal contents, spleen, aorta, and serum were collected and stored at −80 °C until use. In a consistent environment, to reduce disturbances for blood pressure measurement of the rats, 44 rats were acclimatized and fed for 4 weeks before being prepared for grouping. The SBP was measured in unanesthetized rats using the CODA volume-pressure relationship tail-cuff system (Kent Scientific Corporation, Muscatine, IA, USA). Blood pressure was measured at regular intervals, every 2 weeks, during the treatment period, consecutive measurements were taken for each one, and the blood pressure data were averaged for each measurement. Gene expression in colonic tissue was measured by quantitative real-time PCR with SYBR Green. Quantitative real-time PCR was performed on a Thermo Lifetech ABI QuantStudio 3 from Applied Biosystems (Applied Biosystems, Thermo Fisher, Waltham, MA, USA). All amplification reactions were carried out in 96-well optical-grade PCR plates in triplicate (Applied Biosystems, Thermo Fisher, USA), each with 20 μL, sealed with optical sealing tape (Ruibiotech, Qingdao, China). The primers were designed with the GenBank database or DNAMAN for Windows, and synthesized commercially by Ruibiotech. All primers used are shown in Table S4. The TIANGEN Bacterial Genomic DNA Extraction Kit (DP302-02, TIANGEN, Beijing, China) was used to extract colonic fecal genomic DNA. Colonic feces were collected under aseptic conditions and stored at −80 °C until subsequent analysis. Genomic DNA from the colonic fecal samples was extracted using the TIANGEN Fecal Genomic DNA Extraction Kit (DP328-02, TIANGEN, Beijing, China) according to the manufacturer’s instructions. A NanoDrop spectrophotometer ND 3.0 1000 (NanoDrop Technologies, Wilmington, DE, USA) was used to quantify the concentration of extracted DNA. The purity of the extracted DNA was checked by 1% agarose gel electrophoresis. PacBio 16S rRNA V1–V9, sequenced and analyzed on the Pacbio Sequel II sequencing platform, was performed by Biozeron (Lingen, Shanghai, China). The recovered purified PCR products were detected and quantified by a QuantiFluor™-ST Blue Fluorescence Quantification System (Promega, Madison, WI, USA), then mixed in the appropriate proportions according to the sequencing volume required for each sample, and analyzed in PacBio libraries. The raw data from PacBio were processed using the SMRT analysis software, version 9.0, to obtain demultiplexed circular consensus sequence (CCS) reads. OUT clustering was performed using UPARSE (version 7.1), based on 98.65% similarity (http://drive5.com/uparse/ (accessed on 12 February 2022)), and chimeric sequences were identified and removed by UCHIME. The phylogeny of each 16S rRNA gene sequence was analyzed by the RDP classifier (http://rdp.cme.msu.edu/ (accessed on 18 February 2022)) against the Silva (SSU132) 16S rRNA database, multiple diversity index analysis based on OUT data, and statistical analysis of community structure. The levels of IL-6, TNF-α, IL-10, IL-17A, and LPS in serum were measured using ELISA kits (Lianke, Hangzhou, China). All procedures were performed according to the steps of the kits’ instructions. Plasma was centrifuged at 3500 rpm for 15 min and three duplicate samples were prepared for each group. An enzyme marker was used to collect the fluorescence intensity. After removing background and normalization, the concentration of each cytokine in the sample was calculated from the standard curve. A suitably sized colonic, splenic, and aortic ring was fixed in 10% (v/v%) neutral buffered formalin, then washed in running water, dehydrated in alcohol, cleaned in xylene, and treated with paraffin. Thin tissue sections (3–5 μm) were then rehydrated and stained with hematoxylin and eosin (H&E) [56]. After staining, the dried slides were photographed and preserved using a microscope (BX41, Olympus Corporation, Tokyo, Japan) to observe the state of the transverse colon, spleen, and aorta. Immunofluorescence assays were performed by Servicebio (Wuhan, China). Aortic slides were dewaxed, hydrated, and then subjected to microwave antigen repair in ethylenediaminetetraacetic acid buffer (pH 9.0). After serum closure (4% goat serum for 40 min), sections were incubated with the following primary antibody combinations: anti-CD4 and anti-IL-10, anti-CD4, and anti-IL-17, diluted overnight at 4 °C at 1:100. After incubation with the primary antibodies, the slides were incubated with the secondary antibodies for 50 min, protected from light. In addition, the slides were stained with a 1:200 dilution of 4′-6-diamidino-2-phenylindole (DAPI) solution in the dark for 10 min. Finally, the stained cells were observed by fluorescence microscopy (BX41, Olympus Corporation, Tokyo, Japan) and images were collected. The Graphpad prism 8.0 software was used for graphing and the SPSS 22.0 software was used for data analysis. Student’s t-tests or one-way ANOVA was used to analyze the data between groups. Two-way repeated-measures ANOVA was used when analyzing blood pressure at multiple time points. p < 0.05 indicates a significant difference. Line and bar plot data are expressed as mean ± SEM.

PMC10002515

Kotaro Sano,Hiroaki Kobayashi,Hirotaka Chuta,Nozomi Matsuyoshi,Yuki Kato,Hiroshi Ogasawara

CsgI (YccT) Is a Novel Inhibitor of Curli Fimbriae Formation in Escherichia coli Preventing CsgA Polymerization and Curli Gene Expression

22-02-2023

CsgA,curli,Escherichia coli,OmpR,biofilm,periplasmic protein

Curli fimbriae are amyloids—found in bacteria (Escherichia coli)—that are involved in solid-surface adhesion and bacterial aggregation during biofilm formation. The curli protein CsgA is coded by a csgBAC operon gene, and the transcription factor CsgD is essential to induce its curli protein expression. However, the complete mechanism underlying curli fimbriae formation requires elucidation. Herein, we noted that curli fimbriae formation was inhibited by yccT—i.e., a gene that encodes a periplasmic protein of unknown function regulated by CsgD. Furthermore, curli fimbriae formation was strongly repressed by CsgD overexpression caused by a multicopy plasmid in BW25113—the non-cellulose-producing strain. YccT deficiency prevented these CsgD effects. YccT overexpression led to intracellular YccT accumulation and reduced CsgA expression. These effects were addressed by deleting the N-terminal signal peptide of YccT. Localization, gene expression, and phenotypic analyses revealed that YccT-dependent inhibition of curli fimbriae formation and curli protein expression was mediated by the two-component regulatory system EnvZ/OmpR. Purified YccT inhibited CsgA polymerization; however, no intracytoplasmic interaction between YccT and CsgA was detected. Thus, YccT—renamed CsgI (curli synthesis inhibitor)—is a novel inhibitor of curli fimbriae formation and has a dual role as an OmpR phosphorylation modulator and CsgA polymerization inhibitor.

CsgI (YccT) Is a Novel Inhibitor of Curli Fimbriae Formation in Escherichia coli Preventing CsgA Polymerization and Curli Gene Expression Curli fimbriae are amyloids—found in bacteria (Escherichia coli)—that are involved in solid-surface adhesion and bacterial aggregation during biofilm formation. The curli protein CsgA is coded by a csgBAC operon gene, and the transcription factor CsgD is essential to induce its curli protein expression. However, the complete mechanism underlying curli fimbriae formation requires elucidation. Herein, we noted that curli fimbriae formation was inhibited by yccT—i.e., a gene that encodes a periplasmic protein of unknown function regulated by CsgD. Furthermore, curli fimbriae formation was strongly repressed by CsgD overexpression caused by a multicopy plasmid in BW25113—the non-cellulose-producing strain. YccT deficiency prevented these CsgD effects. YccT overexpression led to intracellular YccT accumulation and reduced CsgA expression. These effects were addressed by deleting the N-terminal signal peptide of YccT. Localization, gene expression, and phenotypic analyses revealed that YccT-dependent inhibition of curli fimbriae formation and curli protein expression was mediated by the two-component regulatory system EnvZ/OmpR. Purified YccT inhibited CsgA polymerization; however, no intracytoplasmic interaction between YccT and CsgA was detected. Thus, YccT—renamed CsgI (curli synthesis inhibitor)—is a novel inhibitor of curli fimbriae formation and has a dual role as an OmpR phosphorylation modulator and CsgA polymerization inhibitor. In their natural environment, many bacterial species form a biofilm after attaching to a solid surface [1]. These biofilms are predominantly composed of major components of the bacteria extracellular matrix, including extracellular DNA, lipids, polysaccharides, and proteins [2,3]. Biofilms allow bacteria to grow and survive in conditions of stress [4]. The Gram-negative bacterium Escherichia coli and the related species Salmonella spp. produce amyloid fibers, known as “curli fimbriae,” which are the major protein component of the extracellular matrix [5,6]. Curli fimbriae promote bacterial adhesion to solid surfaces and aggregation at the initial stage of biofilm formation. They facilitate the formation of floating biofilms called “pellicles” that are produced at the air–liquid interface and biofilms on solid surfaces [7,8,9,10]. Therefore, curli fimbriae are considered functional amyloids that facilitate bacterial survival [11]. In E. coli and Salmonella, factors required for curli fimbriae formation are encoded by the following two adjacent, divergently oriented operons: csgBAC and csgDEFG [5,12]. The expression of the curli fimbriae constituents CsgA and CsgB is encoded by the csgBAC operon and induced by CsgD—a biofilm master regulator. Both CsgE and CsgF are chaperone-like proteins involved in the extracellular transport of CsgA and CsgB from the periplasmic space [13,14]. Specifically, after crossing the inner membrane through the Sec transport system, CsgA and CsgB are transported via CsgE and CsgF to the outer membrane transporter CsgG; through this transporter, CsgA and CsgB are exported to the extracellular space [15]. On the cell surface, CsgB serves as an aggregation nucleus for CsgA and is important to anchor CsgA fibrils [12,15]. Accumulation of amyloid fibrils such as curli fimbriae in cells is cytotoxic and even lethal [16,17,18]. Factors that inhibit CsgA aggregation in E. coli have been identified, including chaperone proteins such as DnaK, Hsp33, and Spy [19]. CsgD induces the expression of CsgC; additionally, CsgC has been reported to efficiently transport CsgA to the extracellular space as a monomeric precursor protein of curli fimbriae by preventing CsgA aggregation in the periplasmic space [16,20]. Therefore, multiple mechanisms have been developed to prevent CsgA aggregation before its release in the extracellular environment. Furthermore, to appropriately induce or suppress CsgD expression, a minimum of 15 transcriptional regulators control the expression of the csg genes and are regulated by changes in factors including temperature, osmotic pressure, and pH [21,22,23]. These transcriptional regulators include the following: nucleoid proteins, such as Fis, H-NS, and IHF; two-component systems (TCSs), such as BasS/BasR, BtsS/BtsR, CpxA/CpxR, EnvZ/OmpR, and RstB/RstA and other transcriptional regulatory factors, such as MlrA, Cra, CRP, Crl, and RcdA [12,23,24,25,26,27,28,29,30,31,32,33,34]. Recently, we identified multiple new transcription factors, including YhjC (PlaR) and YjgJ (RcdB), that are potentially involved in the regulation of csgD expression. Thus, E.coli cells appear to respond to various environmental changes and finely regulate the formation of curli fimbriae [35]. Despite considerable progress in terms of the recent identification of the function of csgC—a constituent gene of the csg operon—the complete mechanism of curli fimbriae formation requires elucidation. Aside from controlling curli protein expression, CsgD regulates the expression of at least 20 other genes [36]; these include genes of unknown function that were predicted to be associated with curli fimbriae formation. Among these proteins, YccT is induced by CsgD, which directly binds to the 35–58 base pairs located upstream of the yccT transcription start site [36]. When taking its amino acid sequence into consideration, it is plausible that YccT may have a signal peptide at the N-terminus and be transported to the periplasmic region. STY1099, a Salmonella YccT ortholog, is reportedly involved in nitrate reductase activity, oxidoreductase activity, cellular uptake, and stress response and it plays a role in redox homeostasis under peroxide stress conditions [37]. However, the role of YccT remains largely unknown. The purpose of the present study was to elucidate YccT function. Herein, we demonstrated that YccT inhibited curli formation. Additionally, we provided new insights into the mechanisms underlying biofilm formation in bacterial species containing CsgD and YccT and causing infectious diseases in humans such as Shigella spp. and Salmonella. These data might provide the basis for developing inhibitors against biofilm formation by pathological bacteria. Strains without CsgD—a protein that activates the transcription of the csgBAC operon—lose their ability to form curli fimbriae and are not stained by Congo red [12], Figure 1A). In contrast, overexpression of csgDEFG induces csgBAC expression [38]. Indeed, we confirmed that the wild-type E. coli strain overexpressing csgDEFG derived from plasmid DNA induced curli formation to stain red, whereas the appearance of the CsgD-deletion strain was confirmed as a white colony on the Congo red plate (Figure 1B). Even when only CsgD is induced, csgBAC expression is significantly induced; however, the csgDEFG expression is induced only about 1.5 times higher than that of the control strain transformed with the empty vector [36]. For that reason, in CsgD overexpression from a multicopy plasmid, the expression of the CsgA monomer is induced; however, the transport proteins (CsgE/CsgF/CsgG) for exporting the CsgA monomer from the E. coli cell are hardly induced from the genome DNA [39]. Indeed, we confirmed that the colony of wild-type E. coli strain (BW25113) overexpressing CsgD is hardly stained by Congo red, thereby indicating the inhibition of the curli formation (Figure 1A,B). Additionally, a previous study reported that CsgD overexpression promoted cellulose synthesis and prevented colony staining by Congo red [40]. The BW25113 strain is useful for elucidating gene function as it can use an all-single-gene knockout mutant collection. However, the E. coli BW25113 strain used in this study did not produce cellulose fibers owing to the presence of a nonsense mutation at the sixth leucine residue of bcsQ, a gene of the cellulose synthesis operon, leading to a premature stop codon [38]. Therefore, the decreased Congo red staining induced by CsgD overexpression does not depend on the induction of cellulose synthesis. Previously, we were successful in identifying multiple novel target genes of CsgD [36]. Since these newly identified genes may comprise genes inhibiting curli fimbriae formation, the same experiment was performed using strains lacking one CsgD target gene (deaD, fliF, fliE, nlpA, wrbA, yccT, yccU, yhbT, yhbU, and ymdF). Curli fimbriae formation inhibition was prevented only in the yccT deletion strain (Figure 1B), suggesting that YccT was involved in the inhibition of curli fimbriae formation via CsgD overproduction. CsgC is the downstream gene of the csgBAC operon. Owing to the fact that CsgC inhibits CsgA polymerization [16], we determined whether CsgD-induced inhibition of curli fimbriae formation was mediated by CsgC. Curli fimbriae formation in the csgC deletion strain was remarkably inhibited through CsgD overexpression, which was similar to what was observed in the wild-type strain. This indicated that CsgD-induced inhibition was CsgC-independent (Figure 1D). To confirm the effects of YccT on curli fimbriae formation, we constructed a YccT overexpression plasmid, pBADyccT. The effects of YccT overexpression on curli fimbria formation were assessed using Congo red plates in WT E. coli BW25113, csgA or yccT deletion mutants, and in E. coli with a deletion of dgcC, which encodes the c-di-GMP synthase involved in the regulation of cellulose synthesis by Bcs proteins under the control of CsgD. Owing to the fact that the E. coli BW25113 strain is unable to produce cellulose, the red colony color on the Congo red plates was considered to depend purely on curli fimbriae formation. The colonies of the csgA deletion strain were white, whereas the WT and dgcC deletion strains formed red colonies (Figure 2). All colonies, except those formed by the yccT deletion mutant, were white after CsgD overexpression (Figure 2, strains transformed with pBADcsgD). All strains overexpressing CsgA formed red colonies (Figure 2, strains transformed with pBADcsgA). Additionally, all strains overexpressing YccT formed white colonies (Figure 2, strains transformed with pBADyccT). The expression of all three genes was positively regulated by CsgD, which suggested that YccT was associated with the inhibition of curli fimbriae formation induced by CsgD overexpression. csgBAC expression is controlled by one promoter (csgBp) and its transcription is activated in a CsgD-dependent manner [12,36,41]. Similarly, the activation of yccT gene expression depends on CsgD [36]. Therefore, we investigated the mRNA and protein expression levels of CsgA, a major curli subunit, and YccT in cells overexpressing CsgD. To detect the YccT protein, a nucleotide sequence encoding a FLAG tag was inserted in frame before the stop codon of the yccT open reading frame (ORF) in the E. coli strain leading to the expression of a FLAG tag at the C-terminal ending of YccT. The resulting BWyccT-FLAG strain was prepared and grown for 12 h in preculture and 16 h in main culture (yeast extract casamino acids [YESCA] medium, final concentration of L-arabinose 0.02%). Western blot analyses were subsequently performed using an anti-FLAG antibody. YccT and CsgA protein expression was induced by CsgD overexpression (Figure 3A). Furthermore, the analysis of csgD, csgB, and yccT mRNA levels via Northern blot indicated that the expression of csgD and csgB were remarkably induced and yccT was moderately increased through CsgD overexpression, indicating a CsgD-dependent activation of their transcription (Figure 3B). Notably, although CsgD overexpression induces the expression of the curli component proteins CsgA and CsgB, it inhibited curli fimbriae formation (Figure 1 and Figure 2). These results suggested that YccT affected CsgA after CsgA transport in the extracellular space. The YccT sequence contains a signal peptide comprising 20 amino acids, suggesting that it is transported to and functions in the periplasmic space. Therefore, we investigated YccT expression in the periplasmic space after CsgD overexpression by Western blot analyses of BWyccT-FLAG strains containing the control (pBAD18) or csgD overexpressing (pBADcsgD) vector. YccT-FLAG was detected in the periplasmic fraction, but not in the spheroplast fraction, in both strains (Figure 3C). These data indicated that YccT, like CsgA, secretes into the periplasmic space. To determine whether YccT affected CsgA in the periplasmic region and to elucidate the underlying mechanism, the effects of purified YccT on CsgA fiber formation were examined in vitro using the thioflavin-T (ThT) assay. CsgA amyloid fiber formation involved a nucleation (3–4 h of gradual increase in the fluorescence intensity) and an elongation (4–10 h of a remarkable increase in fluorescence intensity) reaction. The maximal fluorescence intensity was reached after 10 h (Figure 4A,B). The addition of YccT—YccT/CsgA ratio of approximately 1/10 (0.094:1)—prolonged the nucleation by 6–7 h, and the inhibition of CsgA aggregate formation in the elongation phase was dependent on YccT concentration (Figure 4A). These results suggested that YccT interacted directly with CsgA and contributed to the inhibition of CsgA polymerization, thereby promoting its degradation, or inhibiting its transport from the periplasmic space to the extracellular space. To clarify the effects of YccT overexpression on CsgA expression, we introduced the pBADyccT-FLAG vector into the yccT deletion strain to induce YccT-FLAG fusion protein expression. YccT*-FLAG (mature YccT-FLAG with the signal peptide cleaved by transport to the periplasmic region) was expressed for all arabinose concentrations of >2.0 × 10−8% (Figure 5A, upper panel). Moreover, YccT-FLAG is not transported to the periplasmic region, and it accumulated intracellularly when arabinose was added in amounts greater than 2.0 × 10−4% (Figure 5A, upper panel). In contrast, treatment with 2.0 × 10−4% L-arabinose induced a decrease in CsgA expression and almost no CsgA expression was detected when the arabinose concentration was 2.0 × 10−2% (Figure 5A). The effects on csgD and csgBA transcription of YccT overexpression were confirmed via Northern blot. Indeed, pBADyccT was introduced in the yccT deletion strain, and the expression of csgBA and csgD in this strain was significantly inhibited compared with that in the strain containing the control vector (Figure 5B). These results indicated that the repression of csgBA and csgD expression owing to YccT overexpression occurred at the transcriptional level. To confirm the effects of the deletion of the YccT signal peptide sequence on the repression of curli fimbriae formation induced by YccT overexpression, the yccT deletion strain was transformed with the plasmid containing YccT*. The color of colonies formed by bacteria transformed with the yccT or yccT* expression plasmid (pBADyccT or pBADyccT*, respectively) was investigated using the YESCA agar medium containing Congo red. The strain containing pBADyccT formed white colonies, whereas the strain transformed with pBADyccT* formed red colonies as did bacteria expressing the control vector (pBAD18) (Figure 6A). These results suggested that the transport of YccT to the periplasmic region plays an important role in repressing curli fimbria formation (Figure 6A). The YccT-sfGFP and YccT*-sfGFP vectors in which sfGFP was subsequently fused with YccT and YccT*, respectively, were expressed in the BW25113 strain to investigate the localization of YccT in E. coli cells. Unipolar or bipolar distribution of the YccT-sfGFP signal was observed (Figure 6B). Since YccT was transported to the periplasm (Figure 3C), this distribution indicated that a part of YccT-sfGFP that was intended for transport to the periplasm remained in the cells. In contrast, most of the YccT*-sfGFP fluorescence signal followed a unipolar pattern and formed spots larger than those observed for YccT-sfGFP. The transport of YccT*-sfGFP to the periplasm space was expected to be blocked; as a result, YccT*-sfGFP remained unipolar owing to aggregation or interaction with some cell factors. Therefore, YccT might localize to the cell’s pole by interacting with unknown factors to transport it to the periplasmic region. Additionally, YccT without a signal peptide formed remarkable aggregates in cells and its localization was different, which may have been the main etiology for the loss of the inhibitory effect on curli fimbriae formation. Next, the transcription of csgBA and csgD after YccT* overexpression was investigated to clarify the effects of the deletion of the YccT signal peptide on the expression of csgBA and csgD. The examination of each promoter activity using lacZ reporter strains revealed that deleting the YccT signal peptide restored csgD expression to the same level as that in the WT strain expressing the control vector. Similarly, csgBA expression was recovered to about 50% of that of the WT strain containing the control vector (Figure 6C,D). However, because csgB promoter activity was not recovered by approximately 50%, YccT*-sfGFP—which remains in cells—may be directly or indirectly involved in CsgD-dependent csgB induction. These results suggested that the signal peptide contributed to the bipolar or unipolar localization of YccT cells and its transport to the periplasm was directly or indirectly, through its interaction with unknown factors, involved in csgD and csgBA transcriptional inhibition. Numerous transcription factors have been involved in the regulation of csgD expression [35,42]. In particular, the signal transduction systems CpxA/R, RcsBAD, and EnvZ/OmpR, which are involved in the stress response, play a major role in regulating csgD expression [23,43,44]. To determine the involvement of these stress response systems in the repression of csgD expression induced by YccT overexpression, csgD expression in cpxAR, rcsB, and envZ deletion strains was investigated. YccT overexpression in the cpxRA and rcsB deletion strains induced a significant reduction in csgD expression compared with control strains (Figure 7A,B). However, this decrease was similar to that observed in the WT strain, suggesting that the Cpx and Rcs systems were not involved in the mechanism of csgD expression repression induced by YccT overexpression. In contrast, in the envZ deletion strain, the decrease in CsgD expression by YccT overexpression (~25%) was less compared to that of the WT strain (Figure 6C and Figure 7C) or the strain expressing the control vector (Figure 7C). The colonies of the WT strain containing the control vector were red on Congo red plates, whereas those of the envZ-deficient strain were pale red (Figure 7D, first and third pictures, respectively). In contrast, the WT strain overexpressing YccT formed white colonies, whereas the pale red color of the envZ deletion strain was maintained after YccT overexpression (Figure 7D, second and fourth pictures, respectively). Next, we performed a Phos-tag sodium dodecyl sulfate–polyacrylamide electrophoresis (SDS–PAGE) assay to assess the effects of YccT overexpression on the phosphorylation state of intracellular OmpR. The amount of phosphorylated OmpR was greater in the YccT overexpressing WT and envZ deletion strains (Figure 7E). Additionally, an increase in non-phosphorylated OmpR was observed in the envZ deletion strain, which was attributable to envZ deletion. OmpR phosphorylation in the stationary phase is thought partially independent of EnvZ, and OmpR in envZ deletion strains is phosphorylated using other phosphate group donors such as acetyl phosphate [21,45]. Transcriptome analysis of YccT overexpressing cells revealed significant repression of ompF expression, suggesting that the stabilization of intracellular phosphorylated OmpR by YccT overproduction caused the repression of csgD expression. Owing to the fact that the amount of OmpR in cells increased after YccT overexpression, YccT might be involved in activating OmpR expression or stabilizing OmpR. Therefore, we investigated YccT-sfGFP localization in E. coli strains lacking envZ or ompR. In the WT strain, YccT-sfGFP was localized at both poles of the cells, whereas in the envZ and ompR deficient strains, multiple fluorescent spots were observed in most cells (Figure 8). Additionally, the analysis of the fluorescent spots in the ompR deletion strain by deconvolution revealed that bright spots were present along the cell membrane (Figure S1). These results suggested that the deletion of ompR and envZ facilitated the formation of YccT-sfGFP aggregates and OmpR/EnvZ was involved in YccT localization to the cell poles. In contrast, the localization of YccT-sfGFP was not affected in the csgA-deficient strain, suggesting that CsgA was not involved in the localization of YccT to the cellular poles and YccT and CsgA did not interact in the cytoplasm. In the present study, we aimed to clarify the mechanisms through which the induction of CsgD expression inhibits curli fimbria formation and identified a CsgD-regulated periplasmic protein YccT as a potential inhibitor. Therefore, we propose to rename “YccT” to “CsgI” as the inhibitor of curli synthesis. We confirmed that the expression of curli subunits and YccT was induced by CsgD overexpression, suggesting that YccT inhibited the aggregation or extracellular transport of CsgA in the periplasmic region. Previously, a model in which CsgC specifically inhibits CsgA polymerization in the periplasmic space and facilitates CsgA extracellular secretion has been proposed [20]. Although the functions of YccT and CsgC are equivalent, their amino acid sequences are significantly different. The structure and active sites of YccT are currently under investigation. The present data suggest that YccT interacted with CsgA in the periplasmic space and was involved in the inhibition of CsgA extracellular transport and/or degradation. Additionally, CsgC and YccT were simultaneously induced via CsgD overexpression. However, since curli fimbriae formation was inhibited in the csgC deletion strain as in the WT strain, it was likely that CsgC mediated the inhibition of curli fimbriae formation in CsgD overexpression conditions, suggesting that YccT and CsgC function independently of each other in the periplasmic space. In the early stage of biofilm formation in E. coli, CsgD is induced by multiple transcription factors such as IHF, which is a nucleoid protein, and OmpR, a transcription factor, and activates curli fimbriae (constituent subunits CsgA and CsgB) formation [21,23,29]. Our data suggested that YccT was simultaneously induced with curli fimbriae and interacted with the curli subunit CsgA in the periplasmic space to participate in CsgA degradation and reduce the excessive accumulation of curli constituting subunits in the periplasmic region. In the future, the interaction between CsgA and YccT in the cell and periplasmic space will be further investigated. Furthermore, we demonstrated that csgD expression was repressed when YccT was overexpressed. In YccT* (YccT without signal peptide), csgBA and csgD expression was restored and curli fimbriae formation occurred. When periplasmic proteins are overexpressed and aggregate in the periplasmic space, csgD expression is predicted to repress and envelope stress response systems, such as the Cpx and Rcs systems, thereby inducing the expression of genes encoding proteases and molecular chaperones are activated [25,43,44,46,47]. However, in this study, the involvement of Cpx and Rcs in the repression of csgD expression induced by YccT overexpression was not confirmed; however, the repression of csgD expression was lost in the envZ deletion strain transfected with the YccT overexpression plasmid. Therefore, the repression of csgD expression by YccT might be induced through the two-component regulatory system EnvZ/OmpR. EnvZ regulates the expression of many target genes including genes encoding outer membrane proteins such as OmpF and OmpC by controlling the phosphorylation state of the paired cytoplasmic response regulator OmpR in response to changes in external osmolality. OmpC expression is preferentially induced via high osmotic conditions, whereas OmpF expression is induced by low osmotic conditions [48,49,50]. The EnvZ/OmpR system is activated not only by osmotic pressure changes but also by low pH and oligotrophic conditions [51,52,53]. Previous transcriptome analysis has shown an altered expression of more than 100 genes in ompR/envZ deletion mutants [54]. Additionally, the inner membrane protein MzrA modulates EnvZ and affects OmpR phosphorylation levels [55]. The detailed molecular mechanisms underlying YccT-associated inhibition of curli fimbriae formation remain unclear. In particular, it is necessary to investigate the interaction of YccT with EnvZ and OmpR and its relationship with the products of genes controlled by the EnvZ/OmpR system. In the present study, YccT overexpression induced an increase in OmpR levels and OmpR was more phosphorylated in the WT and envZ deletion strains overexpressing YccT, whereas the levels of OmpR dephosphorylation were higher in the envZ deletion strain. Therefore, YccT is likely stimulating the expression or stabilization of OmpR. OmpR phosphorylation depends on its cognate sensor kinase EnvZ and phosphate donors such as acetyl phosphate and carbamoyl phosphate [56,57,58,59]. The mechanism of YccT-induced OmpR phosphorylation is unknown; however, YccT might contribute to OmpR phosphorylation by activating EnvZ in the periplasmic space or by stabilizing OmpR. The promoters of ompC and ompF, which are regulated by EnvZ/OmpR, contain binding sites (F1, F2, F3, and F4 for ompF and C1, C2, and C3 for ompC) for phosphorylated OmpR. The hierarchical binding of phosphorylated OmpR to the F1 and F2 sites or F1, F2, and F3 sites in the ompF promoter activates ompF transcription, whereas higher levels of intracellular phosphorylated OmpR bind to the F4 site and repress the transcription of ompF [60,61,62]. In Salmonella, phosphorylated OmpR binds the csgD promoter at the D1, D2, and D3–D6 sites. OmpR binding to D1 is important for activating csgD expression, whereas its further binding to D2 represses csgD expression in vivo [21]. In contrast, in E.coli, only the D1 site was observed with the homologous region of the csgD promoter region [22,23], whereas the recent Genomic SELEX study using phosphorylated OmpR has shown the OmpR binding site is located downstream of csgDP1. [63]. Thus, the elevated phosphorylated OmpR levels induced by YccT overexpression might promote phosphorylated OmpR binding downstream of csgDP1 of the csgD promoter and repress csgD expression. In the envZ deletion strain, YccT overexpression also induced substantial levels of dephosphorylated OmpR, suggesting that the inhibitory effect on csgD expression was prevented by the competition of dephosphorylated OmpR with phosphorylated OmpR. The mechanisms by which YccT promotes OmpR expression or stabilization are currently under investigation. In addition, the genomic SELEX study has shown that the yccT promoter is a target of OmpR; however, the mechanisms of yccT expression regulation by OmpR remain elusive [63]. Microscopic analyses of YccT localization in E. coli cells revealed that YccT-sfGFP foci were found at one of both cellular poles in the WT strain. In contrast, the signal peptide deletion mutant (YccT*-sfGFP) formed aggregates in one cellular pole. Recently, the bipolar localization of Salmonella enterica YccT (homology with E. coli YccT: 71%) has been shown [37]. Our results in E.coli cells are partially in agreement with these data, suggesting that similar mechanisms are involved in YccT localization in E.coli and Salmonella. Additionally, coimmunoprecipitation assays showed that Salmonella enterica YccT may interact with MreB, DnaK, or OmpR [37]. MreB and the molecular chaperone DnaK are involved in localizing several response regulators to the poles of cells [64]. Moreover, OmpR-GFP foci have a bipolar distribution [64,65]. Analyses using OmpR-YFP have shown that EnvZ-dependent phosphorylated OmpR-YFP foci are found near the poles in the presence of the OmpR target gene promoter on plasmid DNA [66]. Thus, the distribution pattern of response regulators might be closely related to their function as transcription factors [64]. In ompR or envZ deletion strains, YccT-sfGFP formed many foci and presented a spiral localization pattern along the membrane. Additionally, the localization of YccT-sfGFP did not change in the csgA deletion strain suggesting that CsgA does not interact with intracellular YccT and does not affect YccT localization. However, in the csgD deletion strain, YccT-sfGFP formed many foci in the cells comparable to those observed in the ompR or envZ deletion strain. These data suggested that the factors involved in YccT localization to the poles and transport to the periplasm are regulated by CsgD. Since OmpR/EnvZ is essential for csgD expression, the changes in YccT-sfGFP localization in ompR or envZ deletion strains might be due to the effects on CsgD expression (Figure 8). The strains without major CsgD-controlled genes csgA and dgcC showed YccT-sfGFP localization similar to that of the wild-type strain, suggesting the involvement of other CsgD-controlled genes in the localization. The molecular chaperone DnaK is involved in the transport of stable CsgA into the periplasmic space and the stabilization of CsgD and plays an important role in curli fimbria formation [67,68]. Because the expression of YccT was induced in a CsgD-dependent manner, DnaK might be important for YccT localization. It is also possible that the inhibition of curli fimbria formation induced by YccT overexpression results from the competition of YccT with CsgA for DnaK-mediated transport into the periplasmic space. Based on these results, we propose a model of the mechanisms triggered by YccT to inhibit curli fimbriae formation (Figure 9). Interestingly, in Salmonella, FitT, which prevents flagellar protein aggregation and contributes to extracellular transport, plays a role in fine-tuning the expression of the flagellar gene [69]. Therefore, YccT might act similarly to FliT. In conclusion, our data demonstrate that YccT is a novel inhibitor of curli formation by inhibiting both CsgA polymerization and curli gene expression. Thus, we propose to rename YccT CsgI as an inhibitor of curli synthesis. E. coli BW25113 and the gene deletion strains used in this study are listed in Table 1. These strains were grown in YESCA medium at 28 °C or in lysogeny broth (LB) medium at 37 °C with constant shaking at 140 rpm. E. coli BL21(DE3) strain was used for the expression and purification of YccT without signal peptide and CsgA. The one-step recombination system and PCR products were used to add the FLAG tag sequence to the YccT gene in the E.coli genome [70]. For the reporter assay to detect csgD promoter activity, a single copy promoter-lacZ fusion was generated in the genome of E.coli according to a method described previously [71]. All constructed strains are shown in Table 1. For the construction of His-tagged YccT or CsgA expression plasmids, DNA fragments containing the yccT or csgA ORFs were amplified by PCR using E. coli BW25113 genome DNA as template and a pair of specific primers (primer sequences are in Table 1). After digestion with NotI and NdeI, each PCR fragment was inserted into pET21(a) at the corresponding site. For the construction of an arabinose-inducible YccT or YccT-FLAG expression plasmid, a DNA fragment containing each gene’s ORF was amplified by PCR using E. coli BW25113 genome DNA as a template and a pair of gene-specific primers (see Table 1). After digestion with EcoRI and XbaI, the PCR-amplified fragment was inserted at the corresponding site of pBAD18 [75] to generate the plasmids pBADyccT and pBADyccT-FLAG. For the construction of the expression plasmids containing YccT-sfGFP or the mutant YccT*-sfGFP with a deletion of the signal peptide, a linearized DNA fragment of pBAD24 was prepared by PCR using Prime STAR Max (Takara) and fused with YccT-sfGFP or YccT*-sfGFP fragment using Infusion cloning kit (Takara) (for primer sequences see Table 1). The Big Dye Terminator Cycle Sequencing V 3.1 Kit (Applied Biosystems) was used for the sequence analysis of each plasmid. Overnight cultures were incubated in 30 mL YESCA medium at 28 °C for 16 h to prepare total RNA for Northern blot analysis. RNA purification and Northern blot analysis were performed as described previously [23]. DIG-labeled DNA fragments were amplified by PCR using BW25113 genomic DNA (50 ng) as a template, DIG-11-dUTP (Roche Basel, Switzerland) and dNTPs as substrates, gene-specific forward and reverse primers (csgD-N-F and csgD-N-R for the csgD probe; csgB-s-N and csgB-t-N for the csgB probe; yccT-s-N and yccT-t-N for the yccT probe; Table 1), and Ex-Taq DNA polymerase (Takara bio, Shiga, Japan). Briefly, 4 μg of total RNA was denatured in formaldehyde-MOPS gel loading buffer at 65 °C for 10 min, separated by electrophoresis on a 2% agarose gel containing formaldehyde, and transferred onto a nylon membrane (Roche). Hybridization with the DIG-labeled probe was performed overnight at 50 °C using the DIG easy Hyb system (Roche). To detect DIG-labeled probes, the membranes were treated with anti-DIG-AP Fab fragments and CDP-Star (Roche) and images were scanned with Typhoon Trio (cytiva, Marlborough, MA, USA). The expression and purification of His-tagged YccT were conducted following the standard procedure [76]. His-tagged transcription factors were expressed in E. coli BL21(DE3) transformed with YccT expression plasmid (pETyccT-His). First, pETyccT and pETcsgA (Table 1) were introduced into BL21(DE3). The transformed bacteria were cultured in 500 mL LB medium containing ampicillin (100 µg/mL) to an OD600 of 0.9. Then, IPTG at a final concentration of 1 mM was added and the culture was incubated for 3 h shaking. Afterward, the cells were collected by centrifugation (5000 rpm for 20 min). For YccT purification, 5 mL of Bugbuster (Takara bio) per 1 g of pellet, 10 µL of 0.1 M PMSF, 2 µL of DNase (5 U/µL), and 2 µL of RNase (10 mg/mL) were added to the pellet. The samples were rotated for 20 min and then centrifuged at 10,000 rpm for 20 min. The supernatants were mixed with 1 mL of Ni-NTA agarose (QIAGEN, Hilden, Germany) and shaken on a rotator for 1 h. The solutions were then transferred to a column (Muromachi Chemicals, Inc., Tokyo, Japan), washed with 10 mL lysis buffer (50 mM Tris/HCl pH 8.0 at 4 °C and 100 mM NaCl), and eluted with 2 mL lysis/imidazole buffer. A 2-mL sample was transferred to a PD-10 column (GE Healthcare, Chicago, IL, USA) equilibrated with Kpi buffer, and the sample drained with 3.5 mL Kpi buffer was used for the ThT assay. For CsgA purification, the collected pellet was dissolved in 25 mL of 8 M guanidine hydrochloride (GdnHCl) and 50 mM K2HPO4 and stirred with a stirrer at 4 °C for 2 days. The mixture was centrifuged at 10,000 rpm for 20 min at 4 °C, the supernatant was collected, ultrasonically crushed (Astrason XL2020, MISONIX, Farmingdale, NY, USA), mixed with 1 mL of Ni-NTA agarose, and shaken on a rotator for 1 h. The solution was transferred to a column (Muromachi Chemicals, Inc.), washed with 10 mL of 50 mM Kpi buffer, and CsgA protein was eluted with 2 mL of 125 mM imidazole/Kpi buffer. The eluted sample was passed through an ultrafiltration filter (Millipore; Amicon Ultra-4 Centrifuged Filter Devices 30 KDa) by centrifugation at 7500× g. Then, the filtered solution was transferred to a PD-10 column equilibrated with Kpi buffer and the sample drained with 3.5 mL Kpi buffer was passed through an ultrafiltration filter (Millipore; Amicon Ultra-4 Centrifuged Filter Devices 10 KDa) by centrifugation at 7500× g. The solution in the concentrated column was measured and used for the ThT assay. The purity of YccT and CsgA used in this study was verified by SDS–PAGE. The BWyccT-FLAG introduced with pBADcsgD or pBAD18 were cultured with shaking at 28 °C for 16 h in 10 mL of YESCA medium. These culture solutions were transferred into a 50 mL Falcon tube and collected by centrifugation (4 °C, 5000× g, 10 min). After removing the supernatant from each sample, 5 mL of Spheroplast buffer (30% sucrose, 5 mM EDTA-2Na, 10 mM Trisbase) was added to suspend, and 5 mg of lysozyme was added and mixed. After cooling on ice for 45 min, ultracentrifugation (4 °C, 26,000× g, 10 min, himac CS100GX) was performed, and the supernatant was used as a periplasmic fraction. The pellet was suspended by adding 5 mL of Spheroplast buffer, which was used as a spheroplast fraction. Each fraction was transferred to a size exclusion spin column (Vivaspin 6; GE Healthcare), centrifuged (room temperature, 10,000× g, 30 min) and concentrated. The periplasmic fraction and the spheroplast fraction were electrophoresed by SDS-PAGE, and YccT-FLAG was detected by the Western blot assay using a FLAG antibody. The ThT assay was performed as described previously [77]. Purified CsgA and YccT samples and 20 µM ThT were added to a Nunc F96 MicroWell Black Polystyrene Plate (Thermo Fisher Scientific, Waltham, MA, USA). A negative control was prepared with BSA. The background was measured using a sample containing only Kpi buffer and ThT. The fluorescence intensity was measured using ARVO (Perkin Elmer, Waltham, MA, USA) (emission 495 nm, excitation 438 nm). To generate a graph, the fluorescence intensity was standardized using the following formula: Fi = Fluorescence intensity of each sample Fmax = the highest fluorescence intensity in samples with CsgA F0 = background fluorescence intensity obtained for samples containing only Kpi buffer and ThT E. coli BW25113 strain transformed with an expression plasmid containing the sequence of YccT or YccT-sfGFP with or without signal peptide were shaken in LB medium supplemented with L-arabinose (final concentration: 0.02%) at 37 °C for 2 h. Then, 200 μL of 1% glutaraldehyde was added to 100 μL of culture solution, and the mixture was incubated at room temperature for 2.5 h. For the staining of the E. coli membrane, FM4-64 (final 1 μM) was added and then centrifuged at 9,100 g for 1 min, and pellets were re-suspended in 20 μL of PBS (-). Each sample was observed using a green or red filter of an optical fluorescence microscope (Axioimager M1; Carl Zeiss Meditec AG, Jena, Germany) with 630 times magnification. The automatic adjustment mode was used for the exposure time when taking a photomicrograph. The photomicrographs obtained with both filters were merged using ImageJ software (NIH). The csgD-lacZ and csgB-lacZ reporter strains used in this study are listed in Table 1. The measurement of ß-galactosidase activity was performed as described previously [78]. Overnight cultures of E. coli strains were inoculated into 10 mL YESCA medium and cultured for 16 h shaking at 28 °C. Then, the cultures were transferred to a 15-mL Falcon tube and centrifuged at 10,000 rpm for 10 min at 4 °C. The supernatant was discarded and 5 mL Bugbuster Protein Extraction Reagent (Novagen, Darmstadt, Germany), 1 µL rLysozyme™ solution (Merck, Darmstadt, Germany), and 1 µL of 100 mM PMSF were added to 1 g of pellet. The mixture was gently inverted and stirred on a rotator for 20 min. It was then centrifuged at 15,000 rpm for 20 min at 4 °C and the supernatant was transferred to a new 1.5-mL tube. To precipitate CsgA protein, 300 µL of the supernatant was mixed with 60 µL 1 M NaCl in a new 1.5-mL tube and left on ice for 10 min to precipitate CsgA. The mixture was then centrifuged at 20,000 rpm for 10 min at 4 °C and the supernatant was discarded. The pellet was resuspended in 30 µL of 90% formic acid and dried in a centrifugal concentrator (CC-105: Tomy Seiko, Tokyo, Japan) before being suspended in 15 µL of 8 M urea. After SDS–PAGE, proteins were transferred to a PVDF membrane using Trans-blot SD semi-dry transfer cell (Bio-Rad, Hercules, CA, USA). YccT-FLAG and CsgA were detected with primary antibodies recognizing the FLAG tag (Sigma-Aldrich, St. Louis, MO, USA) or CsgA peptide (Eurofins; gifted by Dr. Shinya Sugimoto) and anti-rabbit IgGs (H + L) (MP Biomedicals, LLC-Cappel Products, Santa Ana, CA, USA) as a secondary antibody. Each plasmid (pBAD18, pBADyccT, or pBADdspyccT) was introduced into BW25113 or JW3367 cells, which were cultured in 10 mL LB medium containing ampicillin (100 µg/mL) to an OD600 of 0.9. Then, arabinose at a final concentration of 1 mM was added and the culture was incubated for 4 or 12 h shaking. Afterward, the cells were collected by centrifugation (5000 rpm for 20 min). Bug Buster (Takara) was added to the pellets, which were then stored at −80 °C for 1 day. The pellets were thawed on ice, and lysozyme solution (1 μL/1 mL Bugbuster) and 1 μL of 100 mM PMSF were added and mixed well. The mixture was incubated at room temperature for 20 min. After centrifugation at 15,000 rpm for 20 min at 4 °C, the cells were dissolved in 4 × SDS loading dye, and proteins were separated by electrophoresis on a Super Sep Phos-tag SDS–PAGE (Fujifilm-Wako, Osaka, Japan) containing 25 μM Phos-tag™ acrylamide and 50 μM MnCl2. The gel was shaken in 5 mM EDTA/Western buffer (Nacalai Tesque, Kyoto, Japan) three times 10 min. The gel was shaken for an additional 10 min in Western buffer and then transferred to a nitrocellulose membrane using a Bio-Rad semi-dry transfer device. Proteins were then detected following a standard Western blot protocol and using rabbit anti-OmpR primary antiserum (gifted by Akira Ishihama) and anti-rabbit IgGs (H + L) (MP Biomedicals, LLC-Cappel Products) as a secondary antibody. Bands were detected on LAS-1000 (Fuji Film, Tokyo, Japan) using an Immobilon western chemiluminescent HRP substrate (Millipore, Darmstadt, Germany). The position of the phosphorylated OmpR was determined using a molecular weight marker and purified His-tag OmpR treated with acetyl phosphate. The Congo red plate assay was performed as described previously [79]. E. coli strains were grown at 28 °C for 48 h on a YESCA plate containing 50 μg/mL Congo red and 10 μg/mL Coomassie blue.

PMC10002520

Marek Drozdzik,Joanna Lapczuk-Romanska,Christoph Wenzel,Lukasz Skalski,Sylwia Szeląg-Pieniek,Mariola Post,Arkadiusz Parus,Marta Syczewska,Mateusz Kurzawski,Stefan Oswald

Protein Abundance of Drug Metabolizing Enzymes in Human Hepatitis C Livers

25-02-2023

hepatitis C,liver,drug metabolizing enzymes

Hepatic drug metabolizing enzymes (DMEs), whose activity may be affected by liver diseases, are major determinants of drug pharmacokinetics. Hepatitis C liver samples in different functional states, i.e., the Child–Pugh class A (n = 30), B (n = 21) and C (n = 7) were analyzed for protein abundances (LC-MS/MS) and mRNA levels (qRT-PCR) of 9 CYPs and 4 UGTs enzymes. The protein levels of CYP1A1, CYP2B6, CYP2C8, CYP2C9, and CYP2D6 were not affected by the disease. In the Child–Pugh class A livers, a significant up-regulation of UGT1A1 (to 163% of the controls) was observed. The Child–Pugh class B was associated with down-regulation of the protein abundance of CYP2C19 (to 38% of the controls), CYP2E1 (to 54%), CYP3A4 (to 33%), UGT1A3 (to 69%), and UGT2B7 (to 56%). In the Child–Pugh class C livers, CYP1A2 was found to be reduced (to 52%). A significant trend in down-regulation of the protein abundance was documented for CYP1A2, CYP2C9, CYP3A4, CYP2E1, UGT2B7, and UGT2B15. The results of the study demonstrate that DMEs protein abundances in the liver are affected by hepatitis C virus infection and depend on the severity of the disease.

Protein Abundance of Drug Metabolizing Enzymes in Human Hepatitis C Livers Hepatic drug metabolizing enzymes (DMEs), whose activity may be affected by liver diseases, are major determinants of drug pharmacokinetics. Hepatitis C liver samples in different functional states, i.e., the Child–Pugh class A (n = 30), B (n = 21) and C (n = 7) were analyzed for protein abundances (LC-MS/MS) and mRNA levels (qRT-PCR) of 9 CYPs and 4 UGTs enzymes. The protein levels of CYP1A1, CYP2B6, CYP2C8, CYP2C9, and CYP2D6 were not affected by the disease. In the Child–Pugh class A livers, a significant up-regulation of UGT1A1 (to 163% of the controls) was observed. The Child–Pugh class B was associated with down-regulation of the protein abundance of CYP2C19 (to 38% of the controls), CYP2E1 (to 54%), CYP3A4 (to 33%), UGT1A3 (to 69%), and UGT2B7 (to 56%). In the Child–Pugh class C livers, CYP1A2 was found to be reduced (to 52%). A significant trend in down-regulation of the protein abundance was documented for CYP1A2, CYP2C9, CYP3A4, CYP2E1, UGT2B7, and UGT2B15. The results of the study demonstrate that DMEs protein abundances in the liver are affected by hepatitis C virus infection and depend on the severity of the disease. Hepatitis C virus (HCV) is a small, enveloped RNA virus targeting hepatocytes, and its local replication and immune responses lead to liver damage, ultimately resulting in cirrhosis and/or hepatocellular carcinoma. The implementation of new treatment modalities, i.e., direct acting antiviral drugs (DAAs) have markedly improved therapeutic outcomes. However, the treatment success rate is determined by the HCV genotype, and its high rate and the erroneous nature of its viral replication result in a high prevalence of resistance-associated substitutions (with and without drug pressure) that may impact the efficacy of pharmacotherapy [1]. The currently available DAAs are subjected to metabolism via enzymes located mainly in the intestine and liver [2]. DAAs are substrates of cytochrome P450 (CYP)1A2—pibrentasvir, CYP2B6—velpatasvir, CYP2C8—dasabuvir, velpatasvir, CYP2C19—simeprevir, CYP3A4/5—glecaprevir, grazoprevir, voxilaprevir, daclatasvir, elbasvir, pibrentasvir, simeprevir, velpatasvir, and uridine 5′-diphospho-glucuronosyltransferase (UGT)1A1—pibrentasvir. Apart from being enzymatic substrates, some of the drugs produce inhibitory activity against enzymes, which complicates the enzyme/substrate/inhibitor interplay. The following are the enzymatic inhibitors: CYP1A2—glecaprevir, pibrentasvir; CYP3A4/5—asunaprevir, daclatasvir, dasabuvir, elbasvir, glecaprevir, grazoprevir, paritaprevir, pibrentasvir, simeprevir, velpatasvir; and UGT1A1—glecaprevir, and pibrentasvir [3,4]. The summaries of product information of some DAAs define that safety and efficacy have not been studied in HCV-infected patients with moderate or severe hepatic impairment (Child–Pugh score B or C) due to ethical concerns and methodological challenges. However, the European Medicines Agency (EMA) and the United States of America Food and Drug Administration (FDA) regulations recommend pharmacokinetic studies in patients with impaired hepatic function when it is likely that liver dysfunction may significantly affect pharmacokinetics (especially metabolism and biliary excretion) and dose adjustments might be needed [5,6]. Therefore, quantitative information about drug metabolizing enzymes (DMEs) levels in HCV livers is of clinical relevance. In this regard, physiologically based pharmacokinetic (PBPK) modeling and simulation may enable stratification of potential risks derived from the altered pharmacokinetics of administered drugs, prediction of oral drug bioavailability, and drug–drug interactions (DDIs). The findings of this study can be applied not only to agents targeting HCV, but also to other drugs (DMEs substrates) that are administered to patients with hepatitis C. This study provides information about DMEs proteomic data from only one pathological state of the liver stratified according to the Child–Pugh classification, and includes the largest number of cases published so far. Our preliminary study revealed the impact of liver diseases on DMEs protein levels, and it included 21 cases of HCV livers (also used in the current analysis). The findings revealed that the disease entails significant decrease in CYP2E1, CYP3A5, and UGT2B7 protein abundance [7]. However, due to the limited number of cases, we could not analyze the data according to liver dysfunction stage. This study suggested that the protein abundance of DMEs was affected by both the type of liver pathology (hepatitis C, alcoholic liver disease, autoimmune hepatitis, primary biliary cholangitis, and primary sclerosing cholangitis) as well as the organ functional status (according to the Child–Pugh classification); however, in the latter analysis, samples with different liver pathologies were merged. The combined analysis of all liver pathologies suggested that the worsening of liver functions was associated with a significant decrease in protein abundance of CYP2E1 in class A, downregulation of CYP3A4 and UGT2B7 starting from class B, and CYP1A2, CYP2C8, and CYP2C9 appeared to be reduced in the Child–Pugh score C livers [7]. In the present study, we were able to substantially expand the number of HCV samples, which not only allowed us to provide quantitative proteomic (LC–MS/MS) data on DMEs status in HCV livers, but also to stratify DMEs expression levels according to the Child–Pugh classification, i.e., functional state of the liver. The information presented in this study is complementary to the findings on drug transporters in the liver in patients suffering from HCV infection, as the same subjects were included in the analysis [8]. In the control samples, a strong correlation (rs > 0.6) between DMEs mRNA expression and protein abundance was observed. Only CYP2E1/CYP2E1 and CYP2C19/CYP2C19 did not demonstrate significant correlations (Table 1). However, HCV infection affected correlations between DMEs gene expression and protein levels, i.e., loss of strong correlation was seen in most enzymes, but in general the same trend in mRNA and protein abundance changes was observed (Table 1, Figure 1 and Figure 2, Supplementary Tables S1 and S2). Protein abundance levels of several enzymes were not significantly affected by HCV infection, i.e., CYP1A1, CYP2B6, CYP2C8, CYP2C9, CYP2D6, and CYP3A5. The Child–Pugh class A livers were characterized by significant up-regulation of UGT1A1 (to 163% of the controls). The Child–Pugh class B was associated with down-regulation of CYP2C19 (to 38% of the controls), CYP2E1 (to 54% of the controls), CYP3A4 (to 33% of the controls), UGT1A3 (to 69% of the controls), and UGT2B7 (to 56% of the controls). Significant reductions in CYP1A2 (to 52% of the controls) and UGT2B7 (to 20% of the controls) in the Child–Pugh stage C were also noted (Figure 1 and Figure 2, Supplementary Table S2). The significant trend in the downregulation of CYP1A2, CYP2C9, CYP3A4, CYP2E1, UGT2B7, and UGT2B15 was documented. The merged results of all HCV samples revealed that the disease did not significantly affect the protein levels of CYP1A1, CYP2B6, CYP2C8, CYP2C9, CYP2D6, CYP2E1, and UGT1A3 as compared to the control samples. Significant decrease in CYP1A2 (to 64% of the controls), CYP2C19 (to 45% of the controls), CYP3A4 (to 47% of the controls), and CYP3A5 (to 55% of the controls) as well as UGT2B7 (to 70% of the controls), UGT2B15 (to 79% of the controls), and marked increase in UGT1A1 (to 177% of the controls) abundances, were noted (Figure 1 and Figure 2, Supplementary Table S2). The percentage contributions of all investigated CYPs proteins stratified according to the Child–Pugh score are given in Figure 3. The protein amounts decreased parallel to the disease progression, and the rank order of the enzymes was not markedly affected by the functional state of the liver. CYP2C9, CYP2E1, CYP1A2, and CYP3A4 showed the highest abundances, while CYP2B6, CYP1A1, and CYP2C19 were only found in traces (~1–2%) (Figure 3, Table S2). The Jonckheere–Terpstra test evidenced a significant trend in the downregulation of CYP1A2 (p = 0.012), CYP2C9 (p = 0.029), CYP3A4 (p = 0.019), CYP2E1 (p = 0.004), UGT2B7 (p = 0.0001), and UGT2B15 (p = 0.013) protein abundances from class A to C classified livers. Genotyping studies of CYP2C19, CYP2D6, and CYP3A5 resulted in exclusion of samples genetically determined with the enzyme deficiency, i.e., CYP2C19 – 2 controls, 1 HCV subject, and CYP2D6 – 1 control, 3 HCV subjects. As for CYP3A5 expression, 7 out of 58 HCV patients and 3 out of 20 control subjects were defined as expressers (*1/*3) with a protein abundance of 297.7 fmol/mg (±259.0). The CYP3A5 non-expressers (*3/*3) protein levels were defined at 39.6 fmol/mg (±34.07). Information about the proteomic status of DMEs in the liver in its healthy and disease states provides insights into potential effects of liver pathologies on drug pharmacokinetics, and thus therapeutic responses as well as potential side effects. The correlation between DMEs mRNA and protein levels is not always satisfactory (in the present study it was especially affected by HCV infection). Therefore, reliable protein quantification information allows for better predictions than mRNA expression data. Our previous findings from studies including various liver pathologies (hepatic virus-induced liver damage, alcoholic liver disease, autoimmune hepatitis, primary biliary cholangitis, and primary sclerosing cholangitis) demonstrate a disconnection between the gene expression and protein abundance of DMEs [7]. So far, only limited information about the proteomic data of DMEs in relation to liver diseases has been published. The above-mentioned study [7], demonstrated that different liver pathologies affected various CYPs and UGTs. The HCV samples of the aforementioned study (21 samples are also included in the present analysis) demonstrated that the disease was associated with significant down-regulation of CYP2E1, CYP3A5, and UGT2B7 protein abundances (and unchanged levels of CYP1A1, CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP3A4, UGT1A1, UGT1A3, and UGT2B15). Those results mostly corroborate the findings of the present study, but analysis of a larger number of cases also revealed significant protein abundance reduction in CYP1A2, CYP2C19, CYP3A4, and UGT2B15, as well as a marked up-regulation of UGT1A1. These findings are supported by the proteomic (LC–MS/MS) analysis of Prasad et al. [9] for CYP2D6 (unchanged levels) as well as CYP1A2, CYP2E1, UGT2B7, and UGT2B15 (down-regulation) in 30 HCV liver specimens. The observed differences could be related to methodological specificities; however, they could also be related to analysis of liver tissues from patients with various stages of HCV liver disease (not defined in the study of Prasad et al. [9]). The present study also provides as-of-yet unavailable information about the changes of DMEs in HCV livers in dependence on the organ functional stage (based on the Child–Pugh classification). Our previous study suggested that the stage of liver dysfunction (analyzed in the combined samples from five different liver pathologies) affected DMEs protein levels. In detail, the study revealed protein abundance down-regulation of CYP2E1 in class A livers, decreased levels of CYP2E1 and CYP2C8, CYP3A4, CYP3A5, and UGT2B7 in class B, as well as reduction in CYP2E1, CYP2C8, CYP3A4, CYP3A5, and CYP1A2, CYP2C9 in the class C tissues [7]. However, these findings, as stated above, are based on the merged results of five liver pathologies, with various contributions to each Child-–Pugh class. Thus, the advantage of the current study relies on the analysis of only one liver pathology, i.e., HCV. The results of the present study indicate that the Child–Pugh class A livers are characterized by significant up-regulation of UGT1A1, the class B livers by down-regulation of CYP2C19, CYP2E1, CYP3A4, UGT1A3, and UGT2B7, as well as the class C livers by significant reduction in CYP1A2 and UGT2B7. The present analysis provides evidence that progression of liver dysfunction produced by HCV infection is associated with constant down-regulation of DMEs. The range of protein abundance changes were in most cases over two-fold, which suggested potential clinical relevance. The results of the present study can be used to explain pharmacokinetic changes observed in HCV infected patients [10]. This study demonstrated that CYP1A2, CYP2C19, CYP2D6, and CYP2E1 enzyme activity was differentially affected by the presence of liver disease. The activity of all enzymes decreased significantly, but the reduction depended on the organ functional state. However, Frye at al. did not specify underlying liver pathology, and stratified subjects according to the Child–Pugh classification into compensated liver disease (Child–Pugh score of 5) or decompensated liver disease (Child–Pugh score ≥ 6) [10]. Our quantitative proteomic data are in keeping with altered enzymatic activity in the liver disease patients, e.g., lower (40%) chlorzoxazone (CYP2E1 substrate) metabolic ratio in patients with moderate–severe liver disease, significantly (69%) lower caffeine (CYP1A2 substrate) metabolic ratio in decompensated liver disease (Pugh score ≥6), and no effects on the drug pharmacokinetics of the compensated disease (Pugh score =5) [10]. From this perspective, the current study findings are in keeping with clinical data on metabolic ratios of probe substrates for the studied enzymes (i.e., CYP1A2, CYP2C19, and CYP2E1) [10]. The CYP3A4 results of the present study are in line with clinical observations on midazolam (CYP3A4 substrate) pharmacokinetics reported by Pentikäinen, who demonstrated significantly higher oral bioavailability (38%) and 41% lower total clearance of the drug (and unchanged plasma protein binding and distribution) in patients with chronic liver disease (with undefined etiology) [11]. The pharmacokinetic study, which revealed unaltered pharmacokinetics of debrisoquine (substrate of CYP2D6) in mild or moderate (Child–Pugh classification) liver disease patients also corroborate our results. However, reduced urinary excretion of 4-hydroxydebrisoquine in patients with the liver disease was also observed [12]. The present study also demonstrated significantly elevated protein levels of UGT1A1 in Child–Pugh class A livers, and this observation is not in keeping with the study of Prasad et al. [9], who reported UGT1A1 to be below the lower limit of quantification in livers infected with HCV. However, our finding fits to the changes in the gene expression reported by Congiu et al. [13]. The latter study revealed higher levels of UGT1A1 in early stages of liver fibrosis; however, they noted down-regulation in more advanced stages of the disease. In the present study, mRNA UGT1A1 levels in the liver were similar in all Child–Pugh classes. Clinical pharmacokinetic study suggests that glucuronidation pathways are not affected by liver diseases (revised in [14]), which is also supported by the stable protein levels of UGT1A3 and UGT1A1 in the Child–Pugh class B and C HCV livers revealed in this study. However, in most studies reporting preservation of drug glucuronidation only patients with mild to moderate liver disease were recruited. More recent reports have revealed that some glucuronidation pathways could be down-regulated in more advanced stages of liver failure. Our study demonstrated lower levels of UGT2B7 and UGT2B15 in HCV affected livers, especially in the Child–Pugh class C livers. These findings may explain the reduced glucuronidation of morphine (UGT2B7) [15], lamotrigine (UGT1A4, UGT2B7) [16], zidovudine (UGT2B7) [17], mycophenolic acid (UGT1A9, UGT2B7) [18], or oxazepam (UGT2B15) [19] in patients with liver dysfunction. However, those studies might not identify the effects of the disease on individual UGT isoforms levels, since the substrate overlap of UGTs activity is frequently observed. The protein abundance of DMEs in the control group from the present study are mostly in keeping with the results of other reports. It should be stated that the control liver source can (to some extent) affect results. The preliminary study compared DMEs protein abundances in organ donor livers and metastatic livers (used in studies as reference/control values) and demonstrated some differences [20]. The type of specimen analyzed, i.e., whole liver tissue or microsomal fraction applied for proteomic analysis [21], along with tissue preservation process [22] as well as methodological issues [23,24], can also produce some discrepancies. In the present study, the control group results of the most abundant CYPs (CYP2C9, CYP2E1, and CYP3A4) in metastatic livers free from pathological changes are in line with Vasilogianni et al.’s targeted proteomics results, but measured in microsomal fraction [25], and Couto et al., who measured CYPs in microsomal fraction using a global proteomic approach in the same type of control (metastatic liver) tissues [26]. In the current study, only Caucasian liver samples were included so that we could exclude the ethnic bias. In the literature the impact of ethnicity on drug pharmacokinetics was postulated, suggesting indirect differences in DMEs levels/activities [27]. Some of the inter-ethnic differences can be ascribed to genetic polymorphisms, which affect protein levels or enzymatic activity, e.g., higher frequencies of CYP3A5 allele expression (CYP3A5*3, CYP3A5*6, and CYP3A5*7) in African Americans compared to individuals of European, Native American, and Asian ancestry (affecting pharmacokinetics of tacrolimus) [28] or higher number of the slow metabolizer of CYP2B6 516 G > T allele in Africans Americans (46.7%) and Sub-Saharan Africans (45%) compared to Asians (17.4%), Hispanics (17.4%), Japanese (18%), and Caucasians (21.4%) (affecting efavirenz or atazanavir pharmacokinetics) [29,30]. In the present study 7 out of 58 HCV patients and 3 out of 20 control subjects were defined as expressers of CYP3A5 (*1/*3 genotype) with protein abundance levels of 297.7 fmol/mg (±259.0), in comparison to 39.6 fmol/mg (±34.07) in non-expressers (*3/*3 genotype). However, there is a paucity of expression and protein abundance information about DMEs in different ethnic groups. The available data suggest that ethnic origin does not have a substantial impact on DMEs levels [31]. It seems that nongenetic factors such as diet, weight, and environmental factors should be also highlighted as potential sources of inter-individual variation in drug pharmacokinetics. There is also evidence of age-dependent changes in the expression levels, protein abundances, and activities of DMEs [32,33]. However, our study groups are age-similar, and the impact of age on the results can be disregarded. The changes in the DMEs expression/abundance can be, in part, ascribed to altered cytokine status produced by HCV infection in the liver. Hepatitis C is an inflammatory disease associated with elevated expression levels of IL(interleukin)-6 and TNF(tumor necrosis factor)-α [34]. It is also documented that liver-infiltrating T cells from chronic hepatitis C patients produced IFN(interferon)-γ [35], apoptotic hepatocytes released IL-1α [36], and macrophages exposed to HCV secreted IL-1β and IL-18 [37]. These cytokines could be involved in the transcriptional regulation of some DMEs and may explain the expression/abundance changes observed in the present study. Human hepatocyte culture experiments demonstrated that IL-6 exposure resulted in the down-regulation of genes coding for CYP3A4 and CYP2B6, as well as up-regulation of CYP1A2. This study did not evidence any impact of IL-1β, and no synergism between IL-6 and IL-1β on the CYPs genes expression [38]. The exposure of HepaRG cells to IL-6 produced the suppression of CYP1A2, CYP2B6, and CYP3A4 mRNA levels. Similar findings on CYP1A2, CYP2B6, and CYP3A4 mRNA expressions were observed in primary hepatocytes [39]. However, no suppression of CYP1A2, CYP2B6, and CYP3A4 mRNAs after exposure to IL-18 and IL-1β was observed [38,39]. The TNF-α suppressed expression of cyp1a1 gene in the hepatocyte cell line Hepa1c1c7 [40], and IFN-γ induced down-regulation of CYP1A2 and CYP3A4 expression in human hepatocytes [41]. The down-regulation of transcription factors such as hepatocyte nuclear factors (HNFs), NF(nuclear factor)-κB, along with several nuclear receptors such as pregnane X receptor (PXR) and constitutive androgen receptor (CAR), have been proposed to be responsible for suppression of the CYPs expression by inflammatory stimuli [40,42,43]. It was shown that PXR was involved in the IL-6-mediated down-regulation of CYP3A4 in HepG2 cells [44], and IL-1 mediated regulation of CYP3A4, with possible contribution of HNF(hepatocyte nuclear factor)4 [45]. An involvement of NF-κB in the down-regulation of CYP1A1/1A2 expression in hepatocytes was also reported [40]. Furthermore, PXR and CAR could also regulate phase II enzymes in hepatocytes [46]. However, not all results of the present study are in keeping with in vitro and ex vivo experimental findings; contrary results have also emerged from those reports, most likely due to different cell models, culture conditions, or experimental protocols. The present study results can be also used to better scale PBPK models of DAAs, as there is missing information about pharmacokinetics in HCV patients with advanced liver disease (especially in Child–Pugh class C subjects). The study findings are in keeping with the available pharmacokinetic data and recommendations specified in the summaries of product information of DAAs. The significantly down-regulated levels of CYP1A2 (and CYP3A4 and UGT1A1) could explain the altered pharmacokinetics of pibrentasvir, whose area under the concentration–time curve (AUC) differed by 26% or less in patients with Child–Pugh class A or B cirrhosis and increased to 2.1-fold for those with class C [47]. The protein levels of CYP2B6 were not affected by the stage of liver failure. Velpatasvir is metabolized via this enzymatic pathway (also via CYP2C8 and CYP3A4), and its pharmacokinetics was not affected by liver dysfunction, which supports the proteomic findings of the present study. The drug AUC was comparable in non-HCV Child–Pugh classes B and C patients with normal hepatic function subjects [48]. The CYP2C8 protein abundance, similar to CYP2B6, was not altered in the samples from HCV-infected patients. The enzyme contributes to dasabuvir (and as stated above of velpatasvir) metabolism. The AUC values of dasabuvir were similar in healthy subjects and the Child–Pugh class A patients. However, in the Child–Pugh class B patients a 16% AUC reduction of the drug was observed, which was paralleled by a 57% decrease in M1 (major metabolite) AUC values. The class C subjects were characterized by elevated AUCs for dasabuvir (325%) and M1 (77%) [49]. According to the summary of product characteristics, dasabuvir should not be administered to the Child–Pugh class B and C patients. The down-regulation in CYP2C8 protein levels in the liver could contribute to the observed changes in the drug pharmacokinetics. The pharmacokinetic information about simeprevir, a CYP2C19 substrate (also a substrate of CYP3A4 and CYP2C8) in liver dysfunction patients is not equivocal. Sekar et al. [50] observed equal AUC in non-HCV Child–Pugh class A and B subjects. Other trials reported two-fold higher AUC in non-HCV Child–Pugh class B and C patients [51] or 2.4- and 5.2-fold AUC increases in the Child–Pugh class B and C patients, respectively, compared to healthy individuals. However, contribution of CYP3A4 down-regulation to the observed changes cannot be excluded. Therefore, simeprevir should not be used in Child–Pugh class C patients and caution should be taken with Child–Pugh class B subjects, as stated in the manufacturer recommendations [52]. The reported alterations in simeprevir pharmacokinetic characteristics can be considered to be in keeping with the protein abundance changes observed in the present study, since the down-regulation of CYP2C19 protein abundance was found, significant from the Child–Pugh class B. The documentation in the present study of significant down-regulation of CYP3A4 levels are in line with pharmacokinetic studies and recommendations for the clinical use of elbasvir/grazoprevir, glecaprevir/pibrentasvir, and sofosbuvir/velpatasvir/voxilaprevir (all agents, except for sofosbuvir, are substrates of CYP3A4). Clinical guidelines mark those combined medications as not recommended or contraindicated in the Child–Pugh class B and C patients (also due to unavailability of the relevant data). Results of pharmacokinetic studies are in part inconsistent (due to complexity of factors affecting drug kinetics in liver failure patients) with the present study findings; however, some could indicate reduced CYP3A4 metabolic capacity of the liver, i.e., increased steady-state exposure and Cmax of grazoprevir changing with the Child–Pugh class [53] or pibrentasvir (also a substrate for CYP1A2 down-regulated in the present study) AUC increase by 51%, 31%, and 5.2-fold in patients with the Child–Pugh A, B, and C, respectively [47]. Other substrates of CYP3A4, i.e., daclatasvir or elbasvir are highly protein bound (>99%) and characterized by a low hepatic extraction ratio. Therefore, liver function deterioration could not influence the unbound fraction of daclatasvir in the Child–Pugh class B and C patients, in comparison with HCV-infected controls [3]. Likewise, elbasvir exposure was comparable in HCV patients with Child–Pugh class B liver cirrhosis and healthy controls [54]. In addition to disease-related changes in the protein abundance of distinct CYP enzymes as investigated in our study, the applied pharmacotherapy in HCV patients may be affected by the inhibitory potential of several DAAs, i.e., CYP1A2—glecaprevir, pibrentasvir; CYP3A4/5—asunaprevir, daclatasvir, dasabuvir, elbasvir, glecaprevir, grazoprevir, paritaprevir, pibrentasvir, simeprevir, velpatasvir; and UGT1A1—glecaprevir, pibrentasvir [3,4]. This information is of importance as some of these agents are available as fixed-dose combinations, e.g., elbasvir/grazoprevir or glecaprevir/pibrentasvir. This information, apart from CYPs protein abundance levels, should be implemented in the construction of PBPK models. The control samples were harvested from metastatic livers, from a site at least 5 cm distance of the tumor site. The tissues were collected from Caucasian patients, aged 63 ± 10 years, 11 males and 9 females, diagnosed with metastatic colon cancer. The collected tissues did not show any pathological signs as confirmed by histological examination (the samples were used as the controls in the previously published study [7]). HCV (diagnosed according to the standard clinical criteria) liver parenchymal tissue samples were dissected from the patients requiring liver transplantation. The liver tissue specimens were harvested during elective liver transplantation from the organ immediately after excision. The stage of liver dysfunction was classified according to the Child–Pugh score. Characteristics of the subjects are presented in Table 2. The whole medication information is available for the control samples, i.e., one patient was treated with bisoprolol, furosemide, and tamsulosin (hypertension and prostate hypertrophy), one was treated with bepridil (hypertension), and another one was medicated with amlodipine (hypertension). None of these drugs are known to be a potent regulator of CYP or UGT enzymes. The HCV liver samples were collected in the years 2007–2019, and treatment standards for HCV were modified several times in this period, which is a limitation of the samples. We were only able to select samples without co-existing co-morbidities. Tissue biopsies were taken from livers (control and pathological) under standard general anesthesia not later than 15 min after blood flow arrest. The liver samples were immediately snap-frozen in liquid nitrogen for protein analysis or immersed in RNAlater (Applied Biosystems, Darmstadt, Germany) for RNA analysis, and then stored at −80 °C. The study protocol was approved by the Bioethics Committee of the Pomeranian Medical University. Total RNA was isolated from 25 mg of each tissue sample using a Direct-zol RNA MiniPrep kit (Zymo Research, Irvine, CA, USA). RNA concentration and purity was assessed using a DS-11 FX spectrophotometer (Denovix, Wilmington, DE, USA). cDNA was prepared using a SuperScript® VILO™ cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA), with 500 ng of total RNA for 20 µL of reaction volume, according to the manufacturer’s procedure. The gene expression levels were examined in duplicate using TaqMan Fast Advanced Master Mix and pre-validated TaqMan assays: CYP1A1 (Hs00153120_m1), CYP1A2 (Hs00167927_m1), CYP2B6 (Hs03044631_m1), CYP2C8 (Hs02383390_s1), CYP2C9 (Hs02383631_s1), CYP2C19 (Hs00426380_m1), CYP2D6 (Hs00164385_m1), CYP2E1 (Hs00559367_m1), CYP3A4 (Hs00604506_m1), CYP3A5 (Hs01070905_m1), UGT1A1 (Hs02511055_s1), UGT1A3 (Hs04194492_g1), UGT2B7 (Hs00426592_m1), UGT2B15 (Hs00870076_s1) in ViiA 7 Real-Time PCR System (Life Technologies, Waltham, MA, USA). Threshold values for each gene were set manually and mean CT (cycles of threshold) values were recorded. Relative mRNA expression was calculated by the 2−ΔCt method, which was normalized to the mean expression value obtained for the housekeeping genes: GAPDH (Hs99999905_m1), HMBS (Hs00609297_m1), PPIA (Hs04194521_s1), RPLP0 (Hs99999902_m1), RPS9 (Hs02339424_g), and by 2−ΔΔCt method, which was additionally normalized to the mean value for the control group. Genomic DNA was extracted from tissue samples using a Tissue DNA Purification Kit (EURx, Gdansk, Poland) and subsequently standardized to a uniform concentration (20 ng/μL) before being stored at −20 °C. All samples were genotyped for common lack-of-function variants affecting protein concentration (i.e., stop-codons, frameshifts, and splicing defects) using ViiA7 Fast Real-Time PCR System and pre-validated TaqMan assays (Life Technologies, Carlsbad, CA, USA). The following variants were evaluated: CYP2C19*2 (rs4244285, Assay ID: C__25986767_70), CYP2D6*3 (rs35742686, C__32407232_50), CYP2D6*4 (rs3892097, C__27102431_D0), and CYP3A5*3 (rs776746, C__26201809_30). Additionally, CYP2D6 gene deletion (CYP2D6*5) was evaluated using the qPCR method with TaqMan probes for CYP2D6 (Hs00010001_cn) and reference RPPH1 gene. Tissues placed in liquid nitrogen were mechanically disrupted in a stainless-steel mortar system. Approximately 40 mg of tissue powder of each sample was lysed with 1 mL of 0.2% SDS and 5 mM of EDTA containing 5 µL/mL of Protease Inhibitor Cocktail Set III (Merck, Darmstadt, Germany) for 30 min at 4 °C on a platform shaker with 40 rpm (Polymax 1040, Heidolph, Schwabach, Germany). Total protein content of the whole tissue lysates was determined by bicinchinonic acid assay (BCA, Thermo Fisher) and 100 µg of each sample was processed using a filter aided sample preparation (FASP) [55]. Protein quantification of nine CYP (CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, CYP3A4, and CYP3A5) and four UGT (UGT1A1, UGT1A3, UGT2B7, and UGT2B15) enzymes were measured by mass spectrometry-based targeted proteomics using the validated LC−MS/MS method [56]. With the exception of UGT1A3, an additional proteospecific peptide was analyzed for each protein in the same manner as the 13 validated peptides (i.e., 2 proteospecific peptides have been used for each enzyme). One peptide was used for quantification whereas the other served as a qualifier for the presence of the specific protein. For all peptides and their isotope-labeled internal standard peptides, three mass transitions were used, respectively. The calculated protein values represent the mean of at least 2–3 mass transitions/peptide. The final protein abundance for each enzyme was normalized to the individual mass of tissue used in the tryptic digest (fmol/mg). The mRNA and protein expression data were means ± standard deviation and coefficient of variation %. The median as well as minimum and maximum values are given in Tables S1 and S2 of Supplementary Information. Differences between the study groups were evaluated using the nonparametric Kruskal–Wallis test for multiple comparisons with the post hoc Dunn’s test, and correlations with the Spearman rank test. The Jonckheere–Terpstra test for ordered differences was used to determine the significance of a trend in protein abundance along the liver functional state (using the Child–Pugh classification). The p values of <0.05 were considered significant. The statistical calculations were performed using Statistica 13.3 Software Package (TIBCO Software Inc., Palo Alto, CA, USA). In conclusion, it can be stated that the study provides information about the proteomic data of clinically relevant DMEs in hepatitis C-infected livers, also in relationship to the disease stage classified according to the Child–Pugh score. The disease significantly down-regulated the protein abundance of CYP1A2, CYP2C19, CYP2E1, CYP3A4, UGT2B7, and UGT2B15. The levels of CYP1A1, CYP2B6, CYP2C8, CYP2C9, CYP2D6, as well as UGT1A3, remained stable, whereas the protein amount of UGT1A1 was up-regulated. DMEs down-regulation mostly developed in the Child–Pugh class B (only CYP1A1 started to be decreased in the class C). The rank order of the enzymes was not markedly affected by the liver functional states, i.e., CYP2C9, CYP2E1, CYP1A1, and CYP3A4 showed the highest abundances, while CYP2B6, CYP1A1, and CYP2C19 were found in trace amounts (~1–2%). These findings indicate that the HCV liver has preserved capacity of drug metabolism in the Child–Pugh class A stage. The results from the present study can be incorporated into PBPK models in order to get more precise predictions of drug pharmacokinetics or drug–drug interactions and thus appropriate drug dose-adjustments in patients with HCV liver dysfunction. Refinement of the existing models can be of special importance for drugs where their clinical application is currently limited to mild stages of HCV liver disease (e.g., prescription of ombitasvir/paritaprevir/ritonavir or dasabuvir is restricted to the Child–Pugh score A, since the efficacy and safety of these agents were not studied in the Child–Pugh score B and C patients). This approach can open, based on more adequate estimations, clinical studies on drugs not registered for application in patients with advanced liver diseases since, for ethical reasons and risks, such trials without sufficient entry data are not implemented. The results of the present study can be combined with the findings on drug transporters protein abundances in the same set of HCV patients [8]. The combined picture of DMEs and transporters in the pathological livers can contribute to building PBPK models in HCV patients.

PMC10002521

Jose D. Puentes-Pardo,Sara Moreno-SanJuan,Jorge Casado,Julia Escudero-Feliu,David López-Pérez,Paula Sánchez-Uceta,Paula González-Novoa,Julio Gálvez,Ángel Carazo,Josefa León

PARP-1 Expression Influences Cancer Stem Cell Phenotype in Colorectal Cancer Depending on p53

01-03-2023

colorectal cancer,PARP-1,cancer stem cells,p53

Poly(ADP-ribose) polymerase-1 (PARP-1) is a protein involved in multiple physiological processes. Elevated PARP-1 expression has been found in several tumours, being associated with stemness and tumorigenesis. In colorectal cancer (CRC), some controversy among studies has been described. In this study, we analysed the expression of PARP-1 and cancer stem cell (CSC) markers in CRC patients with different p53 status. In addition, we used an in vitro model to evaluate the influence of PARP-1 in CSC phenotype regarding p53. In CRC patients, PARP-1 expression correlated with the differentiation grade, but this association was only maintained for tumours harbouring wild-type p53. Additionally, in those tumours, PARP-1 and CSC markers were positively correlated. In mutated p53 tumours, no associations were found, but PARP-1 was an independent factor for survival. According to our in vitro model, PARP-1 regulates CSC phenotype depending on p53 status. PARP-1 overexpression in a wild type p53 context increases CSC markers and sphere forming ability. By contrast, those features were reduced in mutated p53 cells. These results could implicate that patients with elevated PARP-1 expression and wild type p53 could benefit from PARP-1 inhibition therapies, meanwhile it could have adverse effects for those carrying mutated p53 tumours.

PARP-1 Expression Influences Cancer Stem Cell Phenotype in Colorectal Cancer Depending on p53 Poly(ADP-ribose) polymerase-1 (PARP-1) is a protein involved in multiple physiological processes. Elevated PARP-1 expression has been found in several tumours, being associated with stemness and tumorigenesis. In colorectal cancer (CRC), some controversy among studies has been described. In this study, we analysed the expression of PARP-1 and cancer stem cell (CSC) markers in CRC patients with different p53 status. In addition, we used an in vitro model to evaluate the influence of PARP-1 in CSC phenotype regarding p53. In CRC patients, PARP-1 expression correlated with the differentiation grade, but this association was only maintained for tumours harbouring wild-type p53. Additionally, in those tumours, PARP-1 and CSC markers were positively correlated. In mutated p53 tumours, no associations were found, but PARP-1 was an independent factor for survival. According to our in vitro model, PARP-1 regulates CSC phenotype depending on p53 status. PARP-1 overexpression in a wild type p53 context increases CSC markers and sphere forming ability. By contrast, those features were reduced in mutated p53 cells. These results could implicate that patients with elevated PARP-1 expression and wild type p53 could benefit from PARP-1 inhibition therapies, meanwhile it could have adverse effects for those carrying mutated p53 tumours. Worldwide, colorectal cancer (CRC) ranks third in tumour type frequency, but second in terms of mortality [1]. Despite the decrease in CRC mortality achieved over the last few years, it remains high, especially in advanced stage cases, as a result of relapse after treatment and resistance to therapy [2]. The latter two factors have been associated with the presence of cancer stem cells (CSCs), a small subset of cancer cells also considered as the root for cancer initiation and metastasis [3]. These cells are characterized by the presence of several surface markers such as CD44, CD133 and CD326, an elevated ALDH1 activity, the overexpression of multidrug resistance associated proteins, and overactivation of DNA repair pathways [4,5]. Since these cells play a key role in cancer biology, in the last few years a deep effort has been made to find new features of the CSCs or specific targets with the aim of achieving more suitable therapies. Poly(ADP-ribose) polymerase-1 (PARP-1) is the most known and abundant member of the PARP enzyme family consisting of 17 ADP-ribosylating enzymes. PARP-1 is mainly involved in the detection and repair of DNA damage through a process in which PARP-1 catalyses the polymerization of ADP-ribose monomers, PARylation or poly(ADP-ribosyl)ation, on target proteins [6]. In addition, PARP-1 is also involved in other molecular and cellular processes, including transcriptional regulation, DNA remodelling, hypoxic response, epithelial mesenchymal transition, angiogenesis, autophagy, and inflammation [7]. PARP-1 is overexpressed in several human cancers, including CRC [8,9,10,11], in which the expression of PARP-1 seems not to be homogenous in tumour cells, but it is highly expressed in CSCs compared with non-CSCs [12], indicating that PARP-1 could regulate CSC programming, as previously reported [7]. However, its relevance or function remains unknown. The fact that PARP-1 is overexpressed in tumour cells and contributes to vital processes for them suggest the idea of the use of PARP-1 inhibitors (PARPi) to fight cancer. In fact, olaparib, a PARPi approved by the FDA, is clinically used for the treatment of BRCA-mutated breast and ovarian cancers [13]. However, no clear effect has been found in clinical trials in which PARP-1 inhibitors have been used in monotherapy regimen in CRC [14]. Recently, a study revealed that PARP-1 seems to protect against the carcinogenic process but once it is initiated, PARP-1 expression facilitates tumour progression [11,15], indicating that this protein acts as a double-edged sword in CRC, suppressing tumour initiation but promoting inflammation-driven tumour progression [15]. Therefore, more research is needed to determine the parameters that influence the implication of PARP-1 in tumour development to potentially know the response to the treatment with PARPi in CRC. The function of PARP-1 in a tumoral context is complex since there is a crosstalk between it and the dynamic of tumour microenvironment [7]. p53, a key protein involved in the response to oncogenic stress that is mutated in a large number of human cancers, seems to interact with PARP-1, although the potential effects are still unclear [16]. PARP- 1 does not seem to regulate p53 transactivation functions itself, but modulates them by promoting p53 nuclear localization [17]. Interestingly, p53 poly(ADP-ribosyl)ation occurs independently of p53 status, wild-type or mutated, as long as the mutations do not arise in specific regions at the C-terminus [17,18]. Recently, it has been described that parylated p53 might affect protein–protein interaction, and therefore p53 activity [18]. This interaction is bi-directional, and PARP-1 forms a complex with p53, but those with mutant p53 result in PARP-1 sequestered in the cytoplasm [19]. However, this arrest is determined by mutant p53 levels, being associated with chromatin and increased poly-ADP-ribosylated proteins in the nucleus when mutant p53 expression is high, but redistributed to the cytosol when its expression is low [20,21]. This PARP-1/p53 interaction is complex and its effects remain elusive. Given the potential relationship between PARP-1 and p53, the aim of this study is to analyse the expression of PARP-1 in colorectal tumours with different p53 status in order to evaluate its influence on the CSC phenotype. PARP-1 mRNA expression was measured in paired tumour and non-tumour tissues from CRC patients. Considering all cases analysed, PARP-1 was highly expressed in the tumoral tissue compared to their paired non-tumoral mucosa (p < 0.0001) (Figure 1a). Stratifying patients according to their p53 status, this tendency was maintained, the overexpression of PARP-1 in the tumoral tissue being significantly increased compared to their corresponding non-tumoral tissue in cases with p53 wild-type (p < 0.0001) and with p53 mutated (p < 0.002) (Figure 1b). To examine the role of PARP-1 in CRC progression, the related clinicopathological parameters were investigated through statistical analysis (Table 1). The results showed that increased expression of PARP-1 was significantly associated with advanced dedifferentiation in the total of the tumours analysed (p = 0.042), and in tumours harbouring a wild-type p53 (p = 0.002). Nonetheless, no significant relationships were found between PARP-1 and age, gender, tumour location, pTNM stage, tumour stage, lymph node metastasis, and p53 mutations. To further evaluate the potential correlation between PARP-1 expression and patient outcome, we generated Kaplan–Meier survival curves from our cohort of patients by sorting out patients based on PARP-1 levels (high and low expression groups) and p53 status (wild-type or mutated). High PARP-1 expression was associated with significantly better OS (χ2 = 0.01, p = 0.031) and DFS (χ2 = 0.01, p = 0.040) in patients with mtp53 tumours (Figure 2c,d). OS and DFS were not significantly related to PARP-1 expression in wtp53 tumours (χ2 = 0.001, p = 0.972 and χ2 = 0.01, p = 0.998, respectively) (Figure 2a,b). Further, multivariate Cox regression for survival analysis showed PARP-1 expression as an independent prognostic factor for survival in CRC patients harbouring mutations in p53 (Table 2). Since the PARP-1 mRNA level correlated with the differentiation grade, it was compared with CSC markers expression (Table 3). As can be seen, high expression of PARP-1 significantly correlated with high expression of CSC markers in the total and in p53 wild-type tumours. No significant correlation between these parameters was found in p53 mutated tumours. To further decipher the role of PARP-1 in CRC, we transiently overexpressed PARP- 1 in two human colorectal cancer cell lines with different status of p53: HCT-116 (p53 wild-type) and HCT-116 p53 null (p53 −/−) (Figure 3). We then compared the behaviour of the control cell lines (expressing only vector, V) with that of PARP-1-overexpressed isogenic daughter (pCMV6-PARP1) cells. PARP-1 overexpression increased ALDH1+ activity in HCT-116 at 96 h after transfection and decreased it in HCT-116 p53 null cells at 72 and 96 h (Figure 4a,b). In HCT-116, transfection with pCMV6-PARP1 induced higher percentage of CD44high/CD326high cells versus mock-transfected cells at 72 h although this difference was not significant at 96 h. On the contrary, we found a similar percentage of CD44high/CD326high/CD133high cells at 72 h after transfection with pCMV6-PARP1 versus mock-transfected cells, and an increase in this percentage at 96 h. As previously reported, we did not find triple-labelled subpopulation cells in the HCT-116 p53 null cells [22]. In these cells, PARP-1 overexpression induced no changes in the percentage of CD44high/CD326high cells versus mock-transfected cells at 72 h, whereas a decrease was found at 96 h after transfection. Altogether, these results indicate a phenotypic change induced due to PARP-1 overexpression in both HCT-116 and HCT-116 p53 null cells. To further study whether such phenotypic changes affect the stemness of the cells, we analysed the self-renewal ability through the sphere formation assay after PARP-1 overexpression in HCT-116 and HCT-116 p53 null cells. As shown in Figure 5, after transfection with pCMV6-PARP1 the sphere number was higher (Figure 5a), although its size decreased in HCT-116 (Figure 5b). However, HCT-116 p53 null cells showed a minor sphere formation ability (Figure 5c) with similar size (Figure 5d) after PARP-1 overexpression. The contribution of PARP-1 in cancer initiation and progression is not totally understood despite it being overexpressed in several human cancer types [8,9,10]. Regarding CRC, little is known about the importance of PARP-1 expression in this tumour, beyond its overexpression in regard to the normal mucosa and being correlated with disease progression [11,15,23]. However, PARP-1 seems not to be the cause of carcinogenesis, since it may act as a tumour suppressor, likely due to its role in DNA repair. However, once a tumour occurs, PARP-1 facilitates its progression [15]. Similar to other studies, in this work we found PARP-1 was overexpressed in CRC samples compared to their healthy mucosa, however, in our cohort, the expression of PARP-1 was correlated to differentiation grade, specifically with dedifferentiated tumours, and not with tumour progression. Additionally, in our work, patients were stratified according to their p53 status, into wild-type p53 or mutated p53, which is not usually performed despite it being mutated in around 43% of cases, which may affect multiple pathways involved in tumour development [24]. In fact, once patients were stratified, the association between PARP-1 and the differentiation grade was only found in the wild-type p53 group, revealing PARP-1 may have different effects on clinicopathological characteristics depending on p53 context. These data suggest that PARP-1 may also exert a different clinical outcome in terms of survival in CRC. In several solid tumours, including breast, ovarian, lung, liver, brain, oesophagus, pancreas, skin, stomach, and acute myeloid leukaemia, high expression of PARP-1 has been related to poorer survival rates [25,26]. On the contrary, there are studies in pancreatic and breast cancer that differ in how PARP-1 affects survival, suggesting that high PARP-1 expression is associated with better survival [27,28]. This indicate that PARP-1 could have different roles depending on the type of cancer, or even that, PARP-1 function could be highly influenced by the heterogeneity context. As far as we know, our work is the first in linking PARP-1 expression with survival rates, OS and DFS, in CRC. According to our data, a high PARP-1 expression is related to better survival rates, but only in those patients with mutated p53 status. The difference in terms of survival among studies might be also attributed to differences in tumour type, study design, ethnicity, or different genomic context. Regarding the latter, we took into consideration differences in p53, which in fact influenced our study depending on its state [25,26,27,28]. The differences between the wild-type p53 and mutated p53 regarding PARP-1 expression could be a direct consequence of the mutation on p53, since those tumours are presumed to be more prone to genetic instability, which triggers DNA repair mechanism [29]. As the mutations on TP53 are diverse, being multiple associated with gain-of-function, some of them lead to damped repair mechanisms, which ultimately may make cells dependent on PARP-1 repair mechanisms to survive [30]. In fact, mutated p53 has been associated with increase of poly-ADP-ribosylated proteins in the nucleus [21]. PARP-1 hyperactivation may result in ATP depletion, toxic levels of poly-ADP-ribose, and mitochondria dysfunction, which cause a new programmed death mode called parthanatos [31]. Although this mechanism could be responsible for the differences observed in survival, this hypothesis requires more studies, since the knowledge about parthanatos is still naïve, and how the different p53 isoforms influence the different oncogenic gain-of-function still represents a black box. Our data suggest that those differences could be due to the relationship of PARP-1 with the differentiation grade depending on p53. In wild-type p53 tumours, PARP-1 was associated with dedifferentiated tumour, and given the link between dedifferentiated tumours and the presence of CSCs cells [32], we next associated PARP-1 levels with different CSC marker levels. In our cohort of study, a high expression of PARP-1 is correlated with a high expression of CSC markers. In this sense similar results have been described in other studies, pointing out that CSCs express higher levels of PARP-1 than non-CSCs [33,34,35]. Once patients were stratified according to their p53 status, we found that in those patients with wild-type 53, a high PARP-1 expression was related to higher expression of CD133 and CD44 markers. The lack of association in the mutated p53 groups suggests that the preferential PARP-1 overexpression in CSCs may not be a direct consequence of the tumoral process per se, but a feature of the stem phenotype, as this type of cells possesses a DNA repair system which is very robust in order to maintain genomic integrity [36,37]. This has opened up the idea of the use of PARPi to target CSCs, but they failed to accurately treat cancer as a single therapy approach [14,33], and are required in combination with other chemotherapeutic agents [12]. However, the benefits of PARPi deteriorate due to the resistance appearance [38]. Additionally, it is important to note that the adverse effect of PARP-1 is in part assumed by its high baseline expression in CSCs, but most studies do not study how PARP-1 expression contributes or influences tumour progression during the disease [33]. In fact, our data suggest that PARP-1 only exerts detrimental activities in patients harbouring tumours with wtp53, but a high PARP-1 expression in mutated p53 tumours seems to be beneficial as it is an independent prognostic factor for survival, and these differences may be a result of a differential regulation of CSCs by PARP-1 depending on p53. To prove it, we developed an in vitro model with two cell lines with different p53 status and overexpressed PARP-1 transiently. In this model, we found that the overexpression of PARP-1 increased the percentage of ALDH1+ population in HCT-116 cell line, meanwhile it was not altered or reduced in the HCT-116 p53 null line. This data was in concordance with the patient data. Since CSCs actually comprise a different subpopulations of stem-like cells [39], we also analysed by flow cytometry the percentage of cells expression in the CSC markers we used in the patient analysis. Similar results were found, and PARP-1 produces an increase of CD44highCD326high and CD44highCD326highCD133high population in the wild-type p53 cell line, and no changes or a decrease in the HCT-116 p53 null one. Additionally, the ability to form spheres was increased upon PARP-1 overexpression in the cells harbouring a wild type p53, meanwhile the lack of p53 resulted in a decrease of spheres when PARP-1 was overexpressed. These in vitro results confirmed the ones obtained from CRC patients and suggest that PARP-1 may exert benign or detrimental effects depending on the p53 status, conceivably via differential regulation of CSC phenotype in CRC. This data may be cancer-type specific, since this study is the first one to shed light on the potential dual effect of PARP-1 in CRC, and the previous PARP-1 works have been mainly focused on the fact of PARP-1 being overexpressed in CSCs, discussed previously, and in the combination of PARP-1 inhibitors with other therapies [12,40]. Nonetheless, results supporting our data have been described in several cancers but have not been put in the context of p53, leading to controversy among some studies. A study using a pancreatic mouse cell line found that the inhibition of PARP-1 reduces CSC features, similar to the results found in neuroblastoma in which PARP-1 promotes stemness [41,42]. Both studies used cell lines harbouring a wild-type form of p53. On the other hand, a recent study in ovarian cancer showed that PARP inhibition reduces tumour size, although it enriches the CSC population within the tumour [43]. This result initiates controversy among studies and may lead to the idea that enrichment may be characteristic of the subpopulation studied or it is an implication of the research design, however, the latter study uses cell lines with mutated p53. This controversy may not be a controversy itself, but a consequence of the differential CSC regulation depending on p53 described in our study. This apparent relationship between PARP-1 and p53 would be supported by the fact that in CRC, patients with wild-type p53 may respond better to PARP inhibition than those with mutated p53 [44]. All these results are in line with our finding, indicating that PARP-1 regulates differentially stemness depending on p53 status, with the subsequent impact on patient prognosis. However, more research is required to confirm our results, and to study in depth the pathways that could induce this differential regulation. Tumour samples and normal mucosa from 201 patients with primary sporadic CRC were prospectively provided by the Andalusian Tumour Bank Network (RBTA). The Ethics Committee of San Cecilio University Hospital approved the study (project code: PI-067/2013; date of approval: 24 January 2014), and all patients gave written informed consent for the post-surgery storage of samples and their use in biomedical research. The exclusion criteria were as follow: (1) Patients aged 18 or less and patients over 85; (2) patients receiving neoadjuvant chemotherapy; (3) patients diagnosed of hereditary CRC; (4) patients with a history of any other cancer. Clinicopathological data of patients were also provided by the RBTA, including age, sex, tumour site, cell differentiation, clinical TNM stage, and lymph node metastasis. Patients were eligible for data collection if they were histologically diagnosed with CRC and previously treated with primary surgery (Table S1). The survival of the cohort was followed for 120 months. Genomic DNA was extracted from tissue samples using QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) following the commercial datasheet. The DNA was quantified using a NanoDrop ND-1000 (Implen GmbH, Munich, Germany) and its integrity was assessed by electrophoresis in agarose gel. P53 mutations were determined as previously reported [45]. Briefly, mutations of TP53 in exons 2–10 of the tissue samples were analysed by PCR using specific primers (Table S2). PCR products were purified using Wizard SV gel and PCR clean-up system (Promega, Madison, WI, USA). Automated sequencing of the PCR products was performed using the 3130 XL Applied Biosystems, Foster City, CA, USA). The results were analysed using the software Chromas Lite 2.1.1 (St South Brisbane, QLD, Australia). Total RNA from tissue samples was isolated using the TRIzol reagent (Invitrogen, Life Technologies, Carlsbad, CA, USA). The amount of total RNA was determined by UV spectrophotometry, and RNA integrity was assessed by agarose gel electrophoresis. First-strand cDNA was prepared by reverse transcription with oligo-dT primers using a commercial cDNA synthesis kit (qScript™ cDNA Synthesis kit, Quanta Biosciences, Gaithersburg, MD, USA). Once retrotranscription was completed, 5 μL of the cDNA was amplified for 40 cycles, employing PerfeCTa SYBR Green SuperMix Kit (Quantabio, Beverly, MA, USA), and using specific primers for: PARP-1, CD44 and CD133, UBC, TBP, and RPS13 (Table S3). UBC, TBP, and RPS13 were used to normalize mRNA levels. For each target, a standard curve representing Ct values versus log cDNA dilution was produced. Additionally, PCR products were verified by melting profile and agarose gel electrophoresis to rule out nonspecific products and primer dimers. Two CRC cell lines with different p53 status were used: HCT-116 (p53 wild type) and HCT-116 p53 null (p53 −/−) obtained from Horizon Discovery (Cambridge- UK). Both cell lines were growing using RPMI 1640 medium (Gibco, Carlsbad, CA, USA) supplemented with 2 mM L-glutamine, and completed with 10% FBS and 1% antibiotic–antimycotic cocktail containing penicillin (100 U/mL), streptomycin (100 μg/mL), and amphotericin B (250 ng/mL) (Gibco, Carlsbad, CA, USA) under standard conditions (37 °C and 5% CO2 in a humidity atmosphere). HCT-116 and HCT-116 p53 null cell lines were transfected with the plasmid pCMV6 containing the human PARP-1 gene and the corresponding empty vector (Origene Technologies, Rockville, MA, USA). For this purpose, 200.000 cells were seeded per well in a 6 well-plate. Once 60–70% confluence was reached, they were transfected with Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA, USA) as a transfection reagent, with a lipofectamine/DNA ratio of 2.5, according to the manufacturer’s instructions. Proteins from whole cell homogenization were isolated using lysis buffer (RIPA buffer supplemented with phosphatase and protease inhibitors). After a 30 min incubation at 4 °C and the subsequent centrifugation at 16,000× g for 30 min, the amount of proteins was quantified through Bradford assay. Fifty micrograms of proteins was loaded into SDS-polyacrylamide gels and transferred to PVDF membrane using a Bio-Rad Trans-Blot Turbo Transfer System (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The blots were probed with the appropriate antibodies for PARP-1 (Abcam, dilution 1:5000), ꞵ-Actin (Santa Cruz Biotechnology, Dallas, TX, USA, dilution 1:1000). As a secondary antibody, HRP-conjugated anti-rabbit or anti-mouse secondary antibody (Bio-Rad, dilution 1:50.000) was used. For detection, Amersham ECL Select Western Blotting Detection Reagent (GE Healthcare, Hatfield, UK) was applied before luminography). The aldefluor assay was performed following the manufacturer’s protocol (STEMCELL technologies, Vancouber, BC, Canada) with a few modifications. After treatments, cells were incubated with BODIPY-aminoacetaldehyde (BAAA), a fluorescent non-toxic substrate for ALDH, which was converted into BOD-IPY-aminoacate (BAA) and retained inside the cells. Viable ALDH1+ cells were quantified by flow cytometry on an FACS Aria III (BD Biosciences, San Jose, CA, USA). The specific inhibitor of ALDH, diethylam inobenzaldehyde (DEAB), was used to control for background fluorescence. Cell surface marker levels of CSCs were determined with human anti-CD44-PE, anti-CD326-FITC, and anti-CD133-APC antibodies (Biolegend, San Diego, CA, USA). After 30 min of incubation at darkness and 4 °C, the samples were analysed using a BD FACSAria III flow cytometry (Becton Dickinson, BD Biosciences) at the Cytometry and Microscopy Research Service of the Biosanitary Research Institute of Granada (ibs.GRANADA). For self-renewal analysis, total population 3000 cells were resuspended in sphere culture medium (DMEM:F12, 1% penicillin/streptomycin, B27, 10 μg/mL ITS, 1 μg/mL hydrocortisone, 4 ng/mL heparin, 10 ng/mL EGF, 20 ng/mL FGF) in ultra-low attachment 24-well plates (Corning). Spheres greater than 25 μM in diameter were counted after 4 days by light microscopy at the Cytometry and Microscopy Research Service of the Biosanitary Research Institute of Granada (ibs.GRANADA). For each patient, mRNA levels of genes in tumour samples were normalized to mRNA levels in normal mucosa. Descriptive statistics was reported as medians with inter-quartile range (IQR) for continuous variables and as whole numbers and percentages for categorical variables. Low and high levels of genes were obtained through the median of the mRNA expression levels in our cohort of patients. Associations between clinicopathological features of CRC patients and gene expression were analysed with the Kruskal–Wallis and Mann–Whitney non-parametric tests. The Pearson’s test was used for the correlation analysis after transforming the variables applying natural logarithms. The Kaplan–Meier method was used to determine the cumulative probability of overall survival (OS) and disease-free survival (DFS), and the differences were evaluated using Log-rank tests. Prognostic factors were evaluated using univariate and multivariate analysis (Cox proportional hazards regression model). p values lower than 0.05 were considered significant. All confidence intervals (CIs) were stated at the 95% level. All statistical calculations were performed using SPSS software version 15.0 for Windows (IBM, Chicago, IL, USA). PARP-1 is overexpressed in the tumour tissue compared to healthy mucosa, being associated with the differentiation grade. Surprisingly, this association is only maintained in tumours harbouring a wild type p53. In fact, in those patients with mutant p53, PARP- 1 is an independent prognostic factor for survival in CRC. The differences observed are due to the differential regulation of stemness exerted by PARP-1, promoting or not, depending on p53 status, wild-type or mutant, respectively. These results imply that PARP-1 could be used as a potential clinical tool for personalized medicine, since patients with elevated expression of PARP-1 and wild-type p53 could benefit from PARPi therapies, meanwhile it could be adverse for those carrying tumours with mutant p53 and high PARP-1 expression.

PMC10002522

Tomás J. Steeman,Andrea M. J. Weiner,Aldana P. David,Andrés Binolfi,Nora B. Calcaterra,Pablo Armas

G-Quadruplexes Regulate miRNA Biogenesis in Live Zebrafish Embryos

02-03-2023

miR-150,G-quadruplex,microRNA,zebrafish,embryonic development

RNA guanine quadruplexes (G4s) regulate RNA functions, metabolism, and processing. G4s formed within precursors of microRNAs (pre-miRNAs) may impair pre-miRNAs maturation by Dicer, thus repressing mature miRNA biogenesis. As miRNAs are essential for proper embryonic development, we studied the role of G4s on miRNA biogenesis in vivo during zebrafish embryogenesis. We performed a computational analysis on zebrafish pre-miRNAs to find putative G4 forming sequences (PQSs). The precursor of the miRNA 150 (pre-miR-150) was found to contain an evolutionarily conserved PQS formed by three G-tetrads and able to fold in vitro as G4. MiR-150 controls the expression of myb, which shows a well-defined knock-down phenotype in zebrafish developing embryos. We microinjected zebrafish embryos with in vitro transcribed pre-miR-150 synthesized using either GTP (G-pre-miR-150) or 7-Deaza-GTP, a GTP analogue unable to form G4s (7DG-pre-miR-150). Compared to embryos injected with G-pre-miR-150, embryos injected with 7DG-pre-miR-150 showed higher levels of miRNA 150 (miR-150) and lower levels of myb mRNA and stronger phenotypes associated with myb knock-down. The incubation of pre-miR-150 prior to the injection with the G4 stabilizing ligand pyridostatin (PDS) reverted gene expression variations and rescued the phenotypes related to myb knock-down. Overall, results suggest that the G4 formed in pre-miR-150 functions in vivo as a conserved regulatory structure competing with the stem-loop structure necessary for miRNA biogenesis.

G-Quadruplexes Regulate miRNA Biogenesis in Live Zebrafish Embryos RNA guanine quadruplexes (G4s) regulate RNA functions, metabolism, and processing. G4s formed within precursors of microRNAs (pre-miRNAs) may impair pre-miRNAs maturation by Dicer, thus repressing mature miRNA biogenesis. As miRNAs are essential for proper embryonic development, we studied the role of G4s on miRNA biogenesis in vivo during zebrafish embryogenesis. We performed a computational analysis on zebrafish pre-miRNAs to find putative G4 forming sequences (PQSs). The precursor of the miRNA 150 (pre-miR-150) was found to contain an evolutionarily conserved PQS formed by three G-tetrads and able to fold in vitro as G4. MiR-150 controls the expression of myb, which shows a well-defined knock-down phenotype in zebrafish developing embryos. We microinjected zebrafish embryos with in vitro transcribed pre-miR-150 synthesized using either GTP (G-pre-miR-150) or 7-Deaza-GTP, a GTP analogue unable to form G4s (7DG-pre-miR-150). Compared to embryos injected with G-pre-miR-150, embryos injected with 7DG-pre-miR-150 showed higher levels of miRNA 150 (miR-150) and lower levels of myb mRNA and stronger phenotypes associated with myb knock-down. The incubation of pre-miR-150 prior to the injection with the G4 stabilizing ligand pyridostatin (PDS) reverted gene expression variations and rescued the phenotypes related to myb knock-down. Overall, results suggest that the G4 formed in pre-miR-150 functions in vivo as a conserved regulatory structure competing with the stem-loop structure necessary for miRNA biogenesis. MicroRNAs (miRNAs) are the most studied small noncoding RNAs (ncRNAs) due to their pivotal function in gene expression control, regulating up to 60% of the human protein-coding genes by repressing translation, promoting degradation of the target mRNA, or enhancing translation at the post-transcriptional level [1]. MiRNAs are single-stranded RNAs of ≈23 nucleotides in length [1], synthesized by the transcription of longer primary miRNAs (pri-miRNAs) and subsequent multi-step maturation. In animals, pri-miRNAs adopt a long hairpin-like structure and are cleaved in the nuclei by the microprocessor complex composed of Drosha, DiGeorge critical region 8 (Dgcr8), and other proteins, rendering the precursor miRNAs (pre-miRNAs) structured as a stem-loop [2]. The pre-miRNAs are then exported to the cytoplasm, where the stem-loop is cleaved by the Dicer exonuclease, generating a short RNA duplex. One of the two strands of the duplex is degraded while the other one, the mature miRNA, is loaded onto the Ago2 protein within the RNA induced silencing complex (RISC), enabling the miRNA to bind and repress target mRNAs [3] through the induction of deadenylation of the poly(A) tail, mRNA destabilization and decay, and/or inhibition of translation [4]. The onset of various human diseases, including cancer [5,6], neurodegenerative diseases [7], and cardiovascular diseases [8], is due to aberrations in the regulatory processes involving miRNA expression and processing. In addition, miRNAs are essential for proper developmental embryogenesis, cell differentiation, organogenesis, growth, and programmed cell death [9,10]. During embryonic development, miRNAs are expressed in distinct spatial and temporal patterns with functions in the coordination of cell replication timing and cell fate transitions [11]. Mis-expression of miRNA leads to aberrant phenotypes [12]. MiRNAs biogenesis depends on structure-based mechanisms through the formation of complex secondary and tertiary structures, including bulges, hairpins, stem-loops, and duplex, triplex, and G-quadruplex motifs, which allow interactions with both proteins and other nucleic acids [13]. Stem-loops in pre-miRNA are key structural players for miRNA processing [2]. Secondary structures competing with stem-loops can inhibit Dicer cleavage and, consequently, the biogenesis of miRNAs. Among others, G-quadruplexes (G4s) structures can mold the biogenesis and function of miRNAs [14,15,16,17]. G4s are thermodynamically stable four-stranded secondary structures that can be formed by the folding of single-stranded guanine (G)-rich DNA or RNA sequences [18]. Putative G-quadruplex sequences (PQSs) present at least four contiguous tracts of two or more guanine nucleotides interspersed with short nucleotide sequences forming the G4 loops. G4s structure is characterized by the stacking of two or more planar arranges of four Gs (called G-tetrads) stabilized by lateral Hoogsteen-type hydrogen bonds, π–π interactions between the Gs in stacked G-tetrads, and by the coordination of monovalent cations. K+ is considered the main intracellular G4-stabilizing cation, while Li+ is considered as a non-stabilizing or neutral cation in G4 folding and stability [19]. G4s were found in all the taxonomic phyla [20] and were reported involved in the regulation of the nucleic acids metabolism [18]. RNA G4s (rG4s) have been validated as regulators of transcription termination, pre-mRNA processing, mRNA targeting, mRNA translation and maintenance of telomere homeostasis [21,22]. Emerging reports have also informed the presence of rG4s in some types of ncRNAs, mainly in long ncRNAs (lncRNAs) and miRNAs biogenesis and functions [14,15,16,17]. Several reports show that rG4s play regulatory roles in every step of the miRNA biogenesis [14,15,16,17]. A few reports suggest that formation of rG4s in pri-miRNAs impact on the processing of the pri-miRNAs by preventing Drosha-DGCR8 binding and processing, eventually suppressing the biogenesis of the pre-miRNAs and ultimately lading to lower levels of miRNAs [23,24,25]. More abundant evidence has been achieved for the role of rG4s formed in pre-miRNAs. All the studies indicate that rG4s in the pre-miRNAs compete with the stem-loop structures recognized by Dicer, thus inhibiting Dicer-mediated maturation and consequently reducing the mature miRNAs levels [26,27,28,29,30,31,32,33]. In addition, a recent report demonstrated that G4s bind to Dicer and inhibit its activity [34]. The formation and function of rG4s in mature miRNAs has been also reported in several works, suggesting a role in impairing miRNA binding to target mRNAs [35,36,37]. MiRNAs expression and functions are also mediated by G4s present in other molecules than the miRNAs or their precursors, e.g., DNA G4s were reported regulating the transcription of miRNA genes [38], and rG4s in mRNAs (mainly in 3′ UTRs) were described as regulators of miRNAs accessibility to their target sequences [39,40,41,42]. In recent years, the knowledge about the biological role of the rG4s in miRNAs biogenesis and functions has made significant progress. However, most of the data have been achieved by bioinformatic predictions, in vitro structural and biochemical assays, or by performing a few assays in cultured cells, but only a couple of works have assayed the function of rG4s in the miRNA biology in vivo using live, complex, and multicellular organisms. One of these works reported the presence of a conserved rG4 within the pri-miRNA of miR-23b/27b/24-1 cluster, which prevents in vivo the processing by Drosha-Dgcr8 and thus suppresses the biogenesis of the three miRNAs involved in regulation of cardiac function in rats [25]. The other work identified an rG4 in the pre-miR-26a-1 that impairs pre-miR-26a-1 maturation, resulting in a decrease in the miR-26a levels. This leads to an increase in the miR-26a targets important for hepatic insulin sensitivity and lipid metabolism in mice [32]. We focused our study on the role of rG4s in the biogenesis of miRNAs during vertebrate embryonic development. During embryonic development, gene expression is orchestrated by specific and highly evolutionarily conserved mechanisms that take place accurately, both at spatial and temporal levels. The alteration of the fine-tuning of gene expression at any of the different developmental stages may set up specific and well defined phenotypes [43]. In this context, we wondered whether rG4s may contribute to the regulation of miRNAs functions that control genes required for the proper vertebrate embryonic development. We used zebrafish (Danio rerio), a vertebrate model ideal for studying embryonic development that shows high genetic similarity with humans [44], and that contains conserved miRNAs in its genome [12,45]. In this work, we present evidence gathered by using combined computational and experimental analyses showing that an evolutionarily conserved rG4 found in the precursor of miRNA 150 regulates in vivo the miRNA 150 biogenesis, thus modulating the expression of the specific target gene myb during zebrafish embryonic development. To our knowledge, this is the first work reporting the function of an rG4 as a regulatory switch to fine-tune the biogenesis of a miRNA during vertebrate embryonic development. We performed a computational analysis on zebrafish pre-miRNAs to search for PQSs within the sequences of zebrafish pre-miRNAs (Figure 1a). A total of 1114 zebrafish sequences predicted as pre-miRNAs were downloaded from miRBase [46] and Ensembl [47] databases (346 sequences shown in Supplementary Table S1 and 768 sequences shown in Supplementary Table S2). Only 350 of the downloaded sequences were annotated and associated with an identified miRNA name (the 346 from miRBase and 4 additional from Ensembl). Then, we used the downloaded sequences to search for PQSs displaying the canonical consensus G2+N1–7G2+N1–7G2+N1–7G2+, i.e., G4s containing two or more G-tetrads, by using Quadparser program [22]. We found that 174 sequences (33 from miRBase, and 141 from Ensembl) contained two-G-tetrads PQSs (Supplementary Table S3). Among them, ≈52% of the sequences from miRBase (17/33) and ≈19% of the sequences from Ensembl (27/141) contained PQSs predicted with high probability to form G4s (Supplementary Table S3). Four sequences containing the canonical three G-tetrads PQSs (G3+N1–7G3+N1–7G3+N1–7G3+) were found, all of them displaying high probability to form G4s (Supplementary Table S4). Three of them were only retrieved from the Ensembl database and had no associated miRNA name, while the fourth was retrieved from both databases and was annotated as the precursor of miRNA 150 (pre-miR-150). We focused our study on the PQS found in pre-miR-150 sequence as a potential regulator of the miRNA 150 (miR-150) biogenesis mainly because it showed the highest scores in all the rG4 predictors (Supplementary Tables S3 and S4). miR-150 is a hematopoietic cell-specific miRNA with essential regulatory roles in both normal and malignant hematopoiesis, becoming a relevant potential therapeutic target in treating various types of hematopoietic malignancies [48], and is involved in a variety of solid tumors, including breast, lung, and gastric cancer [49]. One of the best described and conserved targets of miR-150 is myb (also known as c-myb), a proto-oncogene encoding a transcription factor with an evolutionarily conserved role in vertebrate hematopoiesis controlling the proliferation, differentiation, and survival of hematopoietic progenitors [50,51], as well as being involved in leukemia and certain solid tumors [48,49]. Mis-regulation of myb causes well-defined phenotypes in developing and adult zebrafish [50,51,52]. The PQS found in zebrafish pre-miR-150 is located at the 3′ end in the stem of the predicted stem-loop structure and is partially complementary to the mature miR-150 sequence (Figure 1b). Therefore, the formation of a G4 might interfere with the formation of the stem-loop structure, thus potentially impeding the pre-miR-150 processing by Dicer. The alignment of the sequences of pre-miR-150 orthologues showed that miR-150 sequence is highly conserved and is found only in vertebrates (Supplementary Figure S1). All the assessed pre-miR-150 sequences contain PQSs located at the 3′ ends, which do not overlap with the mature miRNA-150 sequences (Supplementary Figure S1 and Supplementary Table S5). This finding suggests a functional evolutionary conservation of the G4 forming sequences and a role in miR-150 biogenesis. In agreement, a previous computational prediction found that Danio rerio pre-miR-150 (dre-pre-miR-150) was the unique pre-miRNA containing the PQS identified here, responding to an extended canonical PQS with four tracts of three consecutive guanines and loops ranging from 1 to 12 nucleotides (G3+N1–12G3+N1–12G3+N1–12G3+) [28]. The formation of G4 by the PQS of human pre-miR-150 (hsa-pre-miR-150) was studied in vitro as a putative probe for the detection of Nucleolin in liquid biopsies for lung cancer diagnosis, prognosis, and patient response. This work reported the formation of a stable parallel G4 in the presence of KCl or when complexed with the G4 ligand PhenDC3 [53]. However, the formation of the G4 in dre-miR-150 was not previously explored, nor was the function of the G4 in regulating the pre-miR-150 processing. Here, we used a synthetic RNA oligoribonucleotide (Figure 1b and Supplementary Table S6) to assay in vitro G4 formation by the PQS identified in the dre-pre-miR-150 by using four different spectroscopic approaches: circular dichroism (CD) spectroscopy (Figure 2a), thermal difference spectroscopy (TDS) (Figure 2b), thioflavin T (ThT) fluorescence (Figure 2c), and 1 dimension (1D) 1H nuclear magnetic resonance (NMR) (Figure 2d). The CD spectra showed the typical pattern of peaks associated with the parallel G4 structure, showing an increase of a positive peak around 264 nm and a negative peak around 240 nm in response to the presence of increasing concentrations of K+ (Figure 2a). The CD spectra observed in the presence of Li+ did not increase the characteristic G4 peaks. Thermal stability calculated by CD melting (Supplementary Figure S2) showed that the G4 is highly stable in the presence of 100 mM K+, since melting temperature (Tm) could not be estimated due to incomplete melting, while an estimated Tm of 59.5 °C was observed for the G4 folded in the presence of 1 mM K+. In addition, TDS spectra showed the typical G4 signature with two positive peaks around 243 and 273 nm and a negative peak at 295 nm (Figure 2b), and ThT fluorescence assays showed that the folded PQS markedly enhanced the ThT fluorescence by nearly 70-fold (Figure 2c). Consistent with these results, 1D 1H NMR showed defined signals around 11–12 ppm (Figure 2d), confirming the presence of Hoogsteen bonds and G4 structures. Moreover, the absence of signals at 13 and 15 ppm indicates that there was no significant formation of stable Watson–Crick or i-motif structures [54]. These results indicate that the PQS found in dre-pre-miR-150 folds in vitro as a highly stable G4. The miR-150/myb regulatory pathway has been described in zebrafish, for which specific phenotypes in developing embryos overexpressing miR-150 or knocked-down in myb expression were reported [52,55]. Overexpression of hsa-miR-150 in zebrafish embryos caused a reduction of myb mRNA levels in 24 h post-fertilization (hpf) staged embryos and distinctly abnormal phenotypes in 48 hpf staged larvae, with the most obvious and common phenotypes including shortened trunk and reduced eye sizes, together with a slower heartbeat, and sluggish blood flow. Similar abnormal phenotypes were observed when myb expression was knocked-down by microinjecting specific morpholinos. These reports make miR-150 an interesting case for studying the regulation of miRNAs biogenesis and function by G4s during embryonic development. So, we studied the role of the G4 present in the dre-pre-miR-150 in living zebrafish embryos by microinjecting dre-pre-miR-150 into one-cell staged zygotes. Transcripts containing dre-pre-miR-150 sequence were synthesized in vitro using GTP in the ribonucleotide mixture (obtaining G-pre-miR-150) or alternatively the GTP analog 7-Deaza-GTP (obtaining 7DG-pre-miR-150). 7-Deaza-GTP prevents Hoogsteen base-pairing and G4 formation but does not affect the formation of Watson–Crick base-pairing [56], thus allowing only the formation of the stem-loop structure (Figure 3a). Injection of G-pre-miR-150 caused a significant increase of miR-150 levels at 24 hpf staged embryos (Figure 3b), which was dose-dependent to the amount of injected pre-miR-150 (Supplementary Figure S3a). Notably, injection of 7DG-pre-miR-150 caused a significantly higher increase of miR-150 levels (even significantly higher than those observed for G-pre-miR-150 overexpression, Figure 3b), which was also dose-dependent to the amount of injected pre-miR-150 (Supplementary Figure S3a). These results suggest that a higher proportion of the stem-loop structure is recognized by Dicer in 7DG-pre-miR-150 than in G-pre-miR-150, likely as a consequence of the absence of G4 structure in the first one. Injection of either G-pre-miR-150 or 7DG-pre-miR-150 caused a significant decrease of myb mRNA (Figure 3c), which was also dose-dependent to the amount of pre-miR-150 injected (Supplementary Figure S3b). To discard unspecific effects of 7-Deaza-GTP on pre-miRNA processing by Dicer, we injected 7DG- and G-pre-miR-133a, a pre-miRNA that does not contain PQSs in its sequence. No differences were observed in the miR-133a overexpression levels (Figure 3d), indicating that both transcripts were equally processed by Dicer. Moreover, injection of 7DG- and G-pre-miR-133a did not significantly alter the levels of miR-150 (Figure 3b) nor of myb mRNA (Figure 3c). Then, we analyzed the phenotypes of 48 hpf staged larvae injected 7DG-pre-miR-150 and G-pre-miR-150 and observed that both the trunk length and eye size were significantly reduced mainly in larvae injected with 7DG-pre-miR-150 (Figure 3e–h) in the highest amount injected (400 pg, Supplementary Figure S3c,d). This may indicate that the characteristic phenotypes associated with miR-150 overexpression and myb mRNA knock-down need threshold levels of miR-150 overexpression. This fact reinforces the notion that 7DG-pre-miR-150 produces a higher miR-150 overexpression due to a higher proportion of stem-loop and no competing G4 structure. Data gathered so far show that the impairment of the formation of the G4 in the pre-miR-150 favors miR-150 biogenesis. Therefore, we wondered if the stabilization of the G4 in the pre-miR-150 has an opposite effect. Thus, we tested the effect of pyridostatin (PDS) on the pre-miR-150 processing. PDS is a widely used G4 stabilizer molecule with high binding affinity to DNA and RNA G4s [57,58]. First, we tested in vitro the capability of PDS to stabilize the pre-miR-150 G4. Increasing PDS concentrations caused an increase in thermal stability of the G4 formed by the PQS present in pre-miR-150 (Figure 4). A strong stabilization was observed above 5 µM concentrations of PDS, evidenced by flattened melting curves (Figure 4a), as well as high values of CD peaks at 95 °C (Figure 4b,c). So, we used 5 µM PDS for incubation of G- and 7DG-pre-miR-150 previous to their injection in zebrafish embryos. Pre-incubation of G-pre-miR-150 with PDS caused a significant reduction in miR-150 overexpression levels (Figure 5a) as well as a significant reduction in myb mRNA levels, although it was significantly smaller than in the absence of PDS (Figure 5b). As expected, pre-incubation of 7DG-pre-miR-150 with PDS had no effect in miR-150 overexpression levels (Figure 5a) nor in the depletion of myb mRNA levels (Figure 5b). These results are consistent with the stabilizing effect of PDS on the G4 formed in G-pre-miR-150, which reduces both the stem-loop structure folding and the efficiency of Dicer processing, thus leading to lower mature miR-150 levels and eventually causing lower repression of myb expression. On the contrary, no effect of PDS was observed on 7DG-pre-miR-150, probably due to the absence of a G4 structure. In agreement, the developmental phenotypes observed after G-pre-miR-150 overexpression were reverted in the case of trunk length (Figure 5c) and partially reverted in the case of eye size (Figure 5d) by the pre-incubation of G-pre-miR-150 with PDS. Instead, no phenotypic reversion was observed in embryos injected with 7DG-pre-miR-150 pre-incubated with PDS. Here we have used two different strategies, one inhibiting G4 formation by using 7-Deaza-GTP for pre-miRNA synthesis and the other promoting G4 formation by the incubation of miRNA with PDS prior to embryos injection, that enabled to show that the G4 formed by pre-miR-150 functions in vivo as a structural switch preventing the folding of the stem-loop structure and the miR-150 biogenesis. Here we focused our interest in pre-miR-150 that was the only zebrafish pre-miRNA annotated in miRBase containing a three-G-tetrads PQS and displaying the highest scores in G4RNAscreener, a robust bioinformatics tool developed for identification of RNA PQSs that combines three different non-motif-based G4 predictors [59]. However, in the search for PQS capable of forming rG4 with possible regulatory roles in zebrafish miRNAs, we observed that several pre-miRNAs (33/346, ≈10%) annotated in miRBase contain two-G-tetrads PQS, being ≈52% of them (17/33, ≈5% of the total) predicted with a high probability of forming G4 by G4RNAscreener (Supplementary Table S3). In agreement, G4RNAscreener predicted that 2% of human pre-miRNA contains PQSs, all of them overlapping with a processing site [60]. The G4s folded in RNA are intrinsically more stable than those formed in DNA [61] and can be stable enough to be considered as structures formed transiently in the cellular context. rG4s fit perfectly with the intrinsic nature of RNA molecules as a flexible molecule that undergoes dynamic changes between conformational states in response to environmental and physical factors (interaction with proteins, ions, temperature changes, ligands and/or other nucleic acids), thus exerting their function through a structure-based mechanisms [14,62]. Therefore, although the two-G-tetrads G4s have been reported as less stable than G4s containing three or more G-tetrads [63], pre-miRNAs containing two-G-tetrads PQSs with high G4RNAscreener scores are interesting for further studies. The formation of mildly stable rG4s in pre-miRNAs and other RNAs may be relevant for zebrafish development, considering that fishes are sensitive to the effects of environmental conditions during early development, which can significantly impact adult morphology, performance, and survival [64,65]. Because the zebrafish is an ectothermic organism, it needs to adjust physiological performance in response to environmental temperature changes [66]. Thus, mildly stable rG4s might act as natural switches regulating gene expression in response to temperature change. The inspection of pre-miR-150 orthologues sequences showed that miR-150 sequence is highly conserved and suggests that it is exclusive of vertebrate species (38 available sequences in miRBase). The analysis of the presence of PQSs within these pre-miR-150 orthologues revealed that, although the PQS is less conserved in sequence identity than the miRNA sequence, all the pre-miR-150 orthologues contain PQSs within the 3′ ends and in regions not overlapping with the mature miR-150. This suggests a functional conservation of the G4 as a structure able to compete with the formation of the stem-loop and with the processing by Dicer. Moreover, the myb oncogene is a conserved target of miR-150 [52], thus making the regulatory mechanism of miR-150 biogenesis described here also feasible in other organisms and/or other processes involving myb oncogene. Here, the repressor function of the rG4 found in pre-miR-150 on the biogenesis of miR-150 was demonstrated by the overexpression the pre-miR-150 in vivo in developing zebrafish embryos. Two different approaches were used, one inhibiting and another promoting the formation of the G4 in the pre-miR-150. To inhibit the G4 formation, we used the GTP analog 7-Deaza-GTP for preventing Hoogsteen base-pairing and G4 formation, while allowing the formation of Watson–Crick base-pairing. A similar strategy was previously used for the study of the role of rG4s in the processing of pre-mRNAs during splicing, evidencing the structural function of rG4s co-existing with other competing secondary structures [56,67]. An alternative strategy for preventing G4 formation is to mutate specific Gs in the G-tracts of the PQS. However, the alteration of the PQS may in turn alter the interaction of the pre-miRNA with other biomolecules, such as proteins or other nucleic acids. In the particular case of pre-miR150, since the PQS is located in the stem of the stem-loop structure (Figure 1b), mutations of the PQS may also affect the stem-loop folding, which may require compensatory mutations in the complementary strand to maintain the structural moiety. Moreover, as the PQS is complementary with the mature miRNA sequence (Figure 1b), compensatory mutations may change the mature miRNA sequence, with consequences on the recognition of targets. Therefore, the use of 7DG-pre-miR-150 for overexpression and the comparison with results using G-pre-miR-150 turns into a robust and elegant strategy for evidencing the effect of the rG4 in the pre-miRNA processing, avoiding mutagenesis strategies. On the other hand, to promote the G4 formation, we incubated the pre-miRNA with PDS, a widely used G4 stabilizer molecule with high binding affinity to most DNA and RNA G4s [57,58]. So, the strategy of incubating the in vitro synthesized pre-miRNAs with PDS previous to the microinjection in zebrafish embryos reduced the chances of PDS to interact with other DNA and RNA G4s likely present in embryonic cells, allowing to target the effect of the ligand mainly to the injected pre-miRNA molecule. The two strategies used here represent original approaches, since most of the assays formerly used to assess the function of rG4s in pre-miRNAs processing were performed in cultured cells or in vivo by using classical approaches of mutagenesis and incubation of cells with G4-stabilizing or G4-destabilizing ligands. Our results indicate that the evolutionarily conserved rG4 formed in the pre-miR-150 may function in vivo as a conserved regulatory structure competing with the stem-loop structure necessary for miR-150 biogenesis, thus regulating its levels and the expression of its target genes. This evidence supports the idea that rG4s in pre-miRNAs exist in a dynamic equilibrium with the stem-loop structure, which may be sensitive to subtle changes in the intracellular conditions and, ultimately, allow proper regulation of gene expression. Moreover, our results contribute to the knowledge about miRNA expression control and allow us to delineate novel strategies to regulate miRNA function by targeting rG4s with specific ligands or antisense strategies or by intervening in the action of proteins that bind and stabilize or unfold rG4s. The sequences of zebrafish (Danio rerio) pre-miRNAs were downloaded from the Ensemble (www.ensembl.org, Release: 89, Assembly: GRCz10, accessed on 30 November 2015) [47] and miRBase (www.mirbase.org, Release 22.1, accessed on 6 June 2022) [46] databases. Full names, ID numbers and sequences are shown in Supplementary Tables S1 and S2. Putative G4 sequences (PQSs) were identified with Quadparser [22], searching for the consensus sequence G2+N1–7G2+N1–7G2+N1–7G2+. For the spectroscopic studies, a synthetic single-stranded desalted oligoribonucleotide (pre-miR-150-PQS, ST1) was purchased from Invitrogen (Carlsbad, CA, USA) resuspended in nuclease-free water and stored at −20 °C until use. Concentration was determined by absorbance at 260 nm (NanoVue Plus, Biochrom, Holliston, MA, USA) using the molar extinction coefficient provided by the manufacturer. For all experiments, the oligoribonucleotide was diluted to the final experimental concentration in 10 mM Tris-HCl pH 7.5 containing varying KCl or LiCl concentrations, as indicated in each figure and folded by heating for 5 min at 95 °C and slowly cooling to 20 °C. For cloning and RT-qPCR experiments, DNA oligonucleotides were purchased from Macrogen (Seoul, Republic of Korea), resuspended in nuclease-free water, and stored at −20 °C until use. All sequences are shown in Supplementary Table S6. Circular dichroism spectra (ellipticity, θ) were acquired at 20 °C between 220–300 nm using a Jasco-1500 spectropolarimeter (Jasco, Easton, MD, USA, 10 mm quartz cell, 100 nm/min scanning speed, 1 s response time, 1 nm data pitch, 2 nm bandwidth, average of four scans), as described elsewhere [68]. An oligoribonucleotide concentration of 2 µM was used, in the absence or in the presence of varying KCl (1, 10, or 100 mM) or 100 mM LiCl. The spectral contribution of buffers, salts, and ligands was appropriately subtracted. The CD melting curves were recorded by ellipticity measurements between 20 and 95 °C, at the wavelength corresponding to the maximum observed at 20 °C for the positive peak around 264 nm, as described elsewhere [68]. For melting temperature (Tm) calculation, data was analyzed with Prism 9.5.0 (GraphPad, Boston, MA, USA) with a non-linear least square fitting procedure assuming a two-state transition of a monomer from a folded to an unfolded state with no change in heat capacity upon unfolding. For ligand experiments, PDS (pyridostatin trifluoroacetate salt, Sigma-Aldrich SML0678, St. Louis, MO, USA) was added at different concentrations (0, 1, 2, 5, and 10 µM) after oligoribonucleotide folding and incubated for 30 min at 37 °C before CD spectra and melting recording. NMR spectra were acquired at 20 °C on a 700MHz Bruker Avance III spectrometer (Bruker Biospin, MA, USA) equipped with a triple resonance inverse NMR probe (5 mm1H/D-13C/15N TXI). Experiments were performed on 50 µM RNA oligonucleotide samples folded in the presence of 1 mM KCl. 1D 1H NMR spectra were acquired using the zgesgp pulse sequence for efficient water suppression [69]. We used 8K points, 1024 scans, a recycling delay of 1.4 s and a sweep width of 22 ppm. Experimental time for each NMR spectrum was 29 min. Spectra were processed by exponential multiplication (line broadening of 10 Hz) and baseline correction. NMR acquisition, processing, and analysis was performed using Topspin 3.5 (Bruker, Biospin, MA, USA). The oligoribonucleotide folded at 2 µM in the absence or in the presence of 1 mM KCl or 1 mM LiCl was scanned to measure absorbance from 220 to 320 nm using a scan speed of 100 nm/min and a data interval of 1 nm in a 10 mm quartz cell. Spectra were measured at 20 and 70 °C using a Jasco V-630BIO spectrophotometer (Jasco, Easton, MD, USA) with Peltier temperature control. The difference between the spectra at these temperatures (Abs 70 °C–Abs 20 °C) was calculated and plotted to obtain the TDS as previously described [68]. ThT (Sigma-Aldrich T3516) fluorescence assays were performed as previously described [70]. Fluorescence emission measurements were performed using a microplate reader (Synergy 2 MultiMode Microplate Reader, BioTek, VT, USA) with an excitation filter of 485 ± 20 nm and an emission filter of 528 ± 20 nm. Each sample was tested in triplicate and fluorescence values were relativized to ThT fluorescence in absence of oligonucleotides (F0). A threshold of a 10-fold increase was used for considering G4 formation. Zebrafish specimens were handled according to relevant national and international guidelines and ethically authorized by the Internal Committee for the Care and Use of Laboratory Animals of the Facultad de Ciencias Bioquímicas y Farmacéuticas-Universidad Nacional de Rosario, which has been accepted by the Ministerio de Salud de la Nación Argentina (expedient 6060/374, resolution 207/2018). Adult zebrafish were maintained at 28 °C on a 14–10 h light/dark cycle as previously described [71]. Mating was carried out by crossing three males with four females in the same spawning tank. Embryos were staged according to morphological development in hours or days post-fertilization at 28 °C as described elsewhere [72], considering the head-trunk angle (HTA) as well as visual inspection of the onset of heart beating (for 24 hpf, prim-5, staged embryos) and circulation in segmental vessels (for 48 hpf, long-pec, staged larvae). For the overexpression experiments, the genomic regions of zebrafish pre-miR-150 (miRBase accession code MI0002016; chr3: 32571216-32571517 Ensembl Release: 89, Assembly: GRCz10) and pre-miR-133a (miRBase accession code MI0001993; chr2: 4113916-4114439 Ensembl Release: 89, Assembly: GRCz10) were amplified by PCR and cloned into a pSP64T+dsRED vector [73], using EcoRI (ThermoFisher Scientific, Waltham, MA, USA) and XhoI (Promega, Madison, WI, USA) restriction sites included in the primers (Supplementary Table S6). The plasmid containing no cloned pre-miRNA was used as control. RNA for microinjection was in vitro synthesized using as a template the plasmids previously linearized with BamHI (Promega) and transcribed using mMESSAGE mMACHINE SP6 kit (Invitrogen, Carlsbad, CA, USA). The NTP-CAP mix included with the transcription kit was replaced with a custom mix containing CAP 4 mM, ATP, CTP, UTP, and either GTP or 7-deaza-GTP 10mM each (7-Deazaguanosine-5′-Triphosphate, TriLink, San Diego, CA, USA). One-cell staged embryos were microinjected with 200 and 400 pg of the transcripts and incubated at 28 °C until they reached the stage for collection indicated for each experiment. PDS was added at a final concentration of 5 µM to the transcripts and incubated for 30 min at 25 °C before microinjection. For the RT-qPCR experiments, 40 embryos staged as 24 h post-fertilization (hpf) (prim-5 stage, considering morphological development) were collected for each treatment and frozen in liquid nitrogen and used immediately or stored at −80 °C. RNA was extracted by TRIzol-Chloroform extraction (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions, followed by isopropyl-alcohol precipitation and pellet resuspension in nuclease-free water. RNA concentration was assessed by absorbance at 260 nm. Reverse transcription was performed with M-MLV RT Enzyme (Promega, Madison, WI, USA) and oligo-dT (T12 and T16 mix) primers (Supplementary Table S6). Quantitative PCR was performed with HOT FIREPol EvaGreen qPCR Mix Plus (Solis Biodyne, Tartu, Estonia) on RealPlex4 thermocycler (Eppendorf, Hamburg, Germany). For miRNA detection, specific stem-loop primers were added to the retrotranscription reaction, with a generic reverse and specific forward primers used in qPCR [74] (Supplementary Table S6). The amplification of zebrafish rpl13 and eef1a1l1 genes cDNAs was used as reference [75], and relative expressions were normalized to controls microinjected with the RNA transcribed from pSP64T+dsRED plasmid containing no cloned pre-miRNA. Three technical replicates were performed for each experimental condition. All primer sequences are detailed in Supplementary Table S6. Data analysis was performed with REST2009 (Qiagen, Hilden, Germany) [76], and statistical analysis was performed with Prism 9.5.0 (GraphPad, Boston, MA, USA) (see Supplementary Table S7 for p values of pairwise comparisons), following MIQE guidelines [77]. For each treatment, 50 zebrafish larvae were collected at 48 hpf (long-pec stage, considering morphological development) and fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) pH 7.4 at 4 °C overnight, washed 3 times with PBS, and placed in 100% glycerol. The embryos were orientated laterally and photographed with an Olympus MVX10 stereoscopic microscope and Olympus C-60 ZOOM digital camera. Embryo length was measured by drawing a spline from the head to the end of the tail along the notochord, and eye size was by drawing a freehand selection around the eye using FIJI software (National Institute of Health, Bethesda, MD, USA). Statistical analysis was performed with Prism 9.5.0 (GraphPad, Boston, MA, USA) (see Supplementary Table S7 for p values of pairwise comparisons).

PMC10002527

Mirta Tokić,Dunja Leljak Levanić,Jutta Ludwig-Müller,Nataša Bauer

Growth and Molecular Responses of Tomato to Prolonged and Short-Term Heat Exposure

24-02-2023

ABA,ACC,DREB,HSF,HSP70,HSP90,heat stress induced transcription factors,IAA,NAC,root architecture

Tomatoes are one of the most important vegetables for human consumption. In the Mediterranean’s semi-arid and arid regions, where tomatoes are grown in the field, global average surface temperatures are predicted to increase. We investigated tomato seed germination at elevated temperatures and the impact of two different heat regimes on seedlings and adult plants. Selected exposures to 37 °C and heat waves at 45 °C mirrored frequent summer conditions in areas with a continental climate. Exposure to 37 °C or 45 °C differently affected seedlings’ root development. Both heat stresses inhibited primary root length, while lateral root number was significantly suppressed only after exposure to 37 °C. Heat stress treatments induced significant accumulation of indole-3-acetic acid (IAA) and reduced abscisic acid (ABA) levels in seedlings. As opposed to the heat wave treatment, exposure to 37 °C increased the accumulation of the ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC), which may have been involved in the root architecture modification of seedlings. Generally, more drastic phenotypic changes (chlorosis and wilting of leaves and bending of stems) were found in both seedlings and adult plants after the heat wave-like treatment. This was also reflected by proline, malondialdehyde and heat shock protein HSP90 accumulation. The gene expression of heat stress-related transcription factors was perturbed and DREB1 was shown to be the most consistent heat stress marker.

Growth and Molecular Responses of Tomato to Prolonged and Short-Term Heat Exposure Tomatoes are one of the most important vegetables for human consumption. In the Mediterranean’s semi-arid and arid regions, where tomatoes are grown in the field, global average surface temperatures are predicted to increase. We investigated tomato seed germination at elevated temperatures and the impact of two different heat regimes on seedlings and adult plants. Selected exposures to 37 °C and heat waves at 45 °C mirrored frequent summer conditions in areas with a continental climate. Exposure to 37 °C or 45 °C differently affected seedlings’ root development. Both heat stresses inhibited primary root length, while lateral root number was significantly suppressed only after exposure to 37 °C. Heat stress treatments induced significant accumulation of indole-3-acetic acid (IAA) and reduced abscisic acid (ABA) levels in seedlings. As opposed to the heat wave treatment, exposure to 37 °C increased the accumulation of the ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC), which may have been involved in the root architecture modification of seedlings. Generally, more drastic phenotypic changes (chlorosis and wilting of leaves and bending of stems) were found in both seedlings and adult plants after the heat wave-like treatment. This was also reflected by proline, malondialdehyde and heat shock protein HSP90 accumulation. The gene expression of heat stress-related transcription factors was perturbed and DREB1 was shown to be the most consistent heat stress marker. Temperatures vary geographically and are predicted to rise with global warming, presenting serious threats to agricultural productivity. In the upcoming years, heat stress will become a major abiotic stress factor for many crop species. Not only global temperatures increase, but also the more frequent and severe heat waves will have a considerable impact on ecosystem changes and crop loss worldwide. As sessile organisms, plants are mercilessly exposed to environmental conditions. The stress response in plants is a complex trait regulated by many factors and can decrease plant performance. Mild stress retards plant growth, activates defense mechanisms and may change growth and developmental patterns. Intense stress, however, stops growth, causes the accumulation of harmful metabolites and may even cause plant death. A rich repertoire of flexible mechanisms that control gene expression enables plants to exhibit a quick response to external signals and to readily adapt to a plethora of environmental conditions [1]. Tomato (Solanum lycopersicum) is a fruit vegetable crop from the Solanaceae family cultivated worldwide, and as such, it is frequently exposed to extreme temperature fluctuations. After the juvenile stage, in adult plants vegetative and reproductive development occur simultaneously. Tomatoes originate in sub-tropical areas but can be grown in greenhouses throughout the whole year if the temperature is appropriately regulated. Temperature significantly impacts tomato growth and development. The optimum temperature for growth and a high fruit yield ranges from 20 °C to 25 °C during the day and from 18 °C to 25 °C during the night [2]. The effects of varying temperatures on tomato cultivation have been meticulously described [2,3]. Heat stress is defined as a condition where the ambient temperature is between 10 °C and 15 °C higher than the optimum temperature range for plant cultivation. Whereas the rise in temperature could stimulate growth, heat stress causes negative effects on plant morphology, physiology and biochemistry [4]. Heat stress can significantly affect cellular homeostasis, including changes in photosynthesis, protein misfolding and/or aggregation, the accumulation of reactive oxygen species and cell membrane damage [5]. The plant growth regulators play a role in stress response and integrate environmental stimuli and endogenous signals to regulate plant growth and development. Heat stress, especially under water deficit, elicits a rapid and transient increase in endogenous abscisic acid (ABA), which then suppresses growth and coordinate adaptation to stressful conditions. Auxin (especially its most prominent form, indole-3-acetic acid, IAA) is significantly increased in seedlings grown under heat stress and influences thermomorphogenesis by inducing stem elongation and leaf hyponasty, while the role of ethylene in plant response to heat stress varies, and additional studies are needed to clarify its role in heat stress responses [6,7,8,9]. High temperatures induce the synthesis of heat shock proteins (HSPs) that are molecular chaperones and play a vital role in protecting the stability and functional conformation of cellular proteins. HSPs recognize and bind exposed hydrophobic regions of misfolded proteins and prevent protein aggregation. In addition, HSP70 and HSP90 negatively regulate heat stress transcription factors (HSFs). Under normal growth conditions the inactive state of HSFs is maintained by the HSP70/HSP90 complex, while under heat stress partially denatured cellular proteins compete with HSFs for HSP70/HSP90 binding. As a consequence, HSFs are released from the complex and translocated to the nucleus [10,11,12]. They bind to heat shock elements in target gene promoters and regulate their transcription [10,13]. Here, we aimed to evaluate the impact of different heat regimes, partially mimicking global warming, on tomato growth and development. We determined the maximum permissive germination temperature and the maximum exposure time at 45 °C that still allowed tomato seedlings to survive. We also analyzed changes in physiological parameters, including proline; malondialdehyde (MDA); ABA; IAA; and 1-aminocyclopropane-1-carboxylic acid (ACC), the immediate precursor of ethylene, as well as molecular markers, HSP70 and HSP90 protein accumulation and heat stress-related gene expression levels, upon prolonged exposure of tomato seedlings and adult plants to a moderately elevated temperature or a short-term heat wave-like exposure. Our results indicate important physiological and molecular differences between the two stress regimes. To determine the effect of elevated temperatures on tomato seed germination, stratified seeds were incubated for 8 days at a constant temperature ranging from 24 °C to 37 °C. Germination was completely blocked at 36 °C (Figure 1A). Higher temperatures (≥28 °C) correlated with a reduced germination rate and influenced seedling development. Cotyledon size was reduced at temperatures higher than 24 °C, while hypocotyl length increased (Figure 1B,C). The average hypocotyl length of 8-day-old seedlings germinated at 28.5 °C and 31.5 °C was 4.1 cm and 2.6 cm, respectively, representing a significant difference compared to the length of 1.9 cm in seedlings germinated at 24 °C. Further, 12-day-old tomato seedlings germinated at 24 °C were exposed to 37 °C or 45 °C, and the survival rate was determined after a 7-day-long recovery (Figure 2 and Figure S1A). Whereas 24 h of exposure to 37 °C and a 1-hour-long heat wave treatment did not compromise seedling survival rate, a 3-hour heat wave exposure at 45 °C caused cotyledon and leaf bleaching and chlorosis. Exposure to 45 °C for 6 or 12 h resulted in seedling desiccation and an inability to recover for further growth (Figure 2). Therefore, the 3-hour-long 45 °C heat wave treatment was used in subsequent experiments on seedlings. Both heat stress treatments, the continuous exposure to the 37 °C and the 3-hour-long heat wave at 45 °C had a significant impact on seedling growth and development. The average biomass accumulation of tomato seedlings exposed to 37 °C (39.3 mg) and 45 °C (58.4 mg) showed an obvious reduction compared to the control (70.8 mg). Heat treatments disturbed root development by significantly inhibiting primary root growth and disturbing lateral root development (Figure 3). At 37 °C, lateral root formation was extremely reduced, and primary root growth was inhibited. By contrast, the 45 °C treatment compromised lateral root initiation less but arrested primary root growth (Figure 3A–C). Both treatments significantly reduced total root length (Figure 3D). Finally, 11-week-old adult tomato plants were heat treated at 37 °C for 24 h or at 45 °C for 5 h (Figure S1D). After the 5 h treatment, plants exhibited symptoms equivalent to seedlings treated at 45 °C for 3 h (wilting leaves and bending stems). Plants treated at 37 °C had mildly bent leaves and showed no apparent differences compared to control plants. Although both treated plant groups were revived after 7 days at control conditions, the wilted yellow leaves caused by the exposure to 45 °C never recovered. Due to changes in root growth and development provoked by heat stress, we examined the effects of prolonged heat exposure and heat wave on ABA, ACC (the precursor of ethylene) and IAA accumulation. These plant hormones were selected based on their well-known roles in stress responses and regulation of root development. Both types of heat stress treatments significantly induced IAA accumulation in seedlings (Figure 4A), while the heat wave caused a slight but not significant elevation of IAA in the leaf tissue of adult plants (Figure 4A). The different heat treatments had opposite effects on ACC accumulation in seedlings. At 37 °C, ACC was induced, while it decreased at 45 °C (Figure 4B). Both heat treatments significantly reduced ACC levels in adult plant leaves (Figure 4B). ABA content decreased in both heat-treated seedlings and adult plant leaves Figure 4C), although significance was only observed in seedlings exposed to 37 °C. To estimate the severity of heat treatments on tomato seedlings and adult plants, proline and MDA accumulation were measured. Seedlings and adult plants were heat treated at 37 °C for 24 h or 45 °C for 3 or 5 h, respectively. In seedlings, proline and MDA levels were enhanced upon exposure to both heat regimes, significantly increasing only after the heat wave treatment (Figure 5). Adult plants exposed to 37 °C showed no change in either proline or MDA levels, while the heat wave treatment significantly induced proline accumulation (Figure 5). To analyze the response of heat-treated tomato, relative transcript abundance of heat stress-related transcription factors HSFB1, HSFA3, DREB1, NAC4 and NAC6 was measured by qPCR. In seedlings, the heat wave-like treatment significantly induced the expression of all tested heat stress-related transcription factors, while only DREB1 expression was significantly induced at 37 °C (Figure 6A). In adult plant leaves at 37 °C, HSFB1, NAC4 and NAC6 expression were significantly induced. Exposure to 45 °C induced HSFA3 and DREB1 but reduced NAC6 expression (Figure 6B). The exposure to 37 °C slightly reduced the expression of HSFA3 in both seedlings and adult plants. HSPs are molecular chaperones assisting protein stabilization and refolding under heat stress. To further investigate the heat stress effects on tomato, immunodetection of HSP70 and HSP90 was performed. In contrast to HSP90, which scarcely accumulated in control conditions, large amounts of HSP70 were present in seedlings and adult plant leaves (Figure 7). Heat treatments induced the accumulation of HSP70 and HSP90 proteins in seedlings but did not change HSP70 accumulation in adult plant leaves. In seedlings, HSP70 more readily accumulated at 45 °C, while HSP90 accumulation was more obvious at 37 °C (Figure 7). Planet Earth is witnessing significant climate changes characterized by a gradual increase in environmental temperature and pronounced heat fluctuations during which the temperatures can exceed 45 °C. Though plants have evolved various mechanisms to overcome changing environmental conditions [14], increasing temperatures have adverse effects on plant morphology, physiology and biochemistry, affect biodiversity, reduce crop yields and impact the quality of agriculturally important species [15]. High temperature inhibits the vegetative growth of tomatoes [16,17], causes flower drop, reduces fruit set [12] and negatively impacts fruit ripening [2,3,18,19]. The genetic and molecular mechanisms in tomato plants underlying heat stress have been recently reviewed [16,20], but there is still a large gap in understanding the complex network of processes involved in the tomato heat stress response. Many experiments have focused on the fruit development of tomato under heat stress [21,22]. Seedling and vegetative growth, which occur in parallel with reproduction, significantly contribute to the performance of the crop. Seed germination and seedling growth are indeed the most vulnerable stages in a plant’s life cycle. Light and soil temperature are key environmental factors affecting seed germination [23,24]. We exposed tomato seeds, seedlings and adult plants to elevated temperatures and heat wave-like treatments and monitored plant morphology as well as different biochemical and molecular parameters. We demonstrated that the most suitable temperatures for tomato germination ranged between 24 °C and 28 °C (Figure 1), while germination rates and seedling vigor significantly declined at temperatures higher than 28.5 °C. Only 50% of tomato seeds germinated at 31.5 °C, while a total lack of germination occurred at 36 °C. The results we obtained are in accordance with previous studies [23,25], where the highest possible temperature for tomato germination is 34 °C. Tomato seedlings developed at 28.5 °C and 31.5 °C significantly elongated hypocotyls. This is a well-known phenomenon caused by IAA abundance and signaling by which young seedlings move far from the heat-absorbing soil to reach a better environment for growth and development with lower temperatures [26,27]. We further investigated the impact of heat stress on tomato seedling growth and root morphology (Figure 3). Either a prolonged exposure to 37 °C, a temperature mimicking the summer conditions in areas with a continental climate, or a short-term exposure to 45 °C, simulating a heat wave, were applied. The prolonged exposure reduced primary root growth and obstructed lateral root initiation. Exposure to 45 °C, however, blocked primary root growth completely, but seedlings were still able to develop lateral roots. The root system architecture is a major determinant of agronomic productivity and is influenced by changing environmental conditions [28]. The availability of plant hormones and their crosstalk in response to environmental stimuli play a major role in the root system’s development [29]. Therefore, we measured IAA, ABA and ACC (the immediate ethylene precursor) content in heat-treated tomatoes (Figure 4). In accordance with previously reported results [9], both heat stress regimes used in this study induced significant accumulation of IAA in seedlings. It has long been known that auxin positively regulates lateral root formation in most plant species [30,31], although this was not the case here. Despite the significantly increased concentration of IAA, exposure to 37 °C, but not to 45 °C, reduced the tomato seedlings’ capacity to develop lateral roots. The likely cause for this phenomenon may therefore be attributed to the accumulation dynamics of ACC. The twofold increase in ACC levels at 37 °C likely suppressed lateral root induction, which fits the previous description of ACC’s role in tomato [32]. The authors reported enhanced lateral root formation in ethylene-insensitive mutants and inhibited lateral root development when ACC was applied. However, examinations of heat stress effects on ethylene synthesis in different plant species and tissues showed adverse responses [9]. Heat-generated reactive oxygen species [33] indirectly induce ethylene synthesis, which, in turn, participates in stress alleviation [34,35], while excess ethylene production under severe stress suppresses growth and induces senescence [34]. After production, ACC can also be conjugated with malonate, glutamate or jasmonic acid to produce ACC conjugates that are temporarily unavailable for ethylene production [36,37]. Thus, the low amounts of free ACC observed and subsequent retained ability of lateral root initiation under the heat wave treatment may be the result of ACC conjugate production. Furthermore, the plant hormone ABA has a role in the modulation of root architecture and can induce both root elongation and lateral root initiation [38]. The heat treatments applied here reduced ABA in tomato seedlings. In addition to elevated ACC and IAA levels, lower ABA likely contributed to inhibited tomato root growth and lateral root initiation in seedlings at 37 °C. Elevated ABA is considered a good stress marker, and recent work in grapes [39] has associated ABA as a suitable marker for drought but not heat stress. Previous studies on 5-week-old tomato plants showed free ABA levels to increase during heat exposure to 35 °C and 45 °C compared to control plants cultivated at 25 °C [40]. In that experiment, ABA levels were the highest (1.5-fold) 12 h after heat exposure and continued to decrease for 48 h. Compared to [40], the different tomato response observed here could have been the result of the different experimental setups, including different cultivar types and plant ages used, as well as a difference in applied temperature regimes. Proline and MDA are considered reliable indicators of environmental stress severity in tomato [41,42]. To further assess stress severity in heat-treated seedlings and adult plants, proline and MDA accumulations were measured (Figure 5). Proline acts as an osmolyte or molecular chaperone [43]. Proline has been shown to negatively affect ABA and ethylene biosynthesis in Arabidopsis seedlings in particular during heat stress [44]. As a result, the observed significant proline induction could be involved in the reduction of ABA and ACC upon the heat wave treatment in tomato. MDA is a byproduct of lipid peroxidation under environmental stress. Although MDA was shown to function as a protector [45], excess amounts often point to impaired cellular function. Compared to the prolonged treatment at 37 °C, higher proline and MDA levels at 45 °C indicated stronger stress severity in both seedlings and adult plants. Stress perception, signaling and response are highly regulated at the transcriptional level and lead to the accumulation of different stress-responsive factors. These processes are governed by stress-related transcription factors such as those from the HSF, DREB and NAC families. Plant-specific NAC transcription factors are involved in a multitude of biological processes, from plant growth and development to stress response [46,47]. In tomato, NAC4 and NAC6 are specifically known to be stress-responsive [46,48,49]. Onset, early response and long-term acclimation to heat stress are controlled and regulated by HSFs [50]. In the tomato genome, 26 HSFs are present [51], among which HSFA1, HSFA2 and HSFB1 have been described as master regulators of the heat stress response [52,53,54,55]. Moreover, HSFA3 was shown to be important for heat stress memory [56]. DREB transcription factors play vital roles during heat and water stress responses by influencing the transcription of, among others, HSF genes [57,58]. Since neither ABA nor proline or MDA could be considered reliable stress markers in plants exposed to 37 °C, heat stress-related gene expression and HSP protein abundance were analyzed to further investigate the tomato heat stress response (Figure 6 and Figure 7). In line with proline and MDA accumulation in tomato seedlings grown at 45 °C, the expression of all tested genes (HSFB1, HSFA3, DREB1, NAC4 and NAC6), as well as the accumulation of HSP70 and HSP90 proteins, were significantly induced. Only DREB1 expression was significantly induced also at 37 °C, indicating its possible role as a master regulator of heat perception and its use as an early and sensitive heat stress marker in tomato seedlings. Although HSFA3 was previously shown to be heat-responsive [59], an induction was only observed in seedlings exposed to the heat wave treatment, possibly triggering physiological memory formation. Significant gene expression changes were observed in the tomato leaf tissue. The induction of HSFB1, NAC4 and NAC6 levels at 37 °C points to a possible role of these genes in the heat response. In contrast, NAC6 was strongly reduced at 45 °C, comparable to results reported by [49] who found a strong reduction of NAC6 expression in the leaves of 35-day-old tomato plants exposed to 40 °C. The results therefore indicate a complex, stress-specific NAC6 gene expression pattern. DREB1 showed elevated transcript amounts in heat-treated leaves at 45 °C, highlighting the gene yet again as a possible heat stress marker. Lastly, heat stress exposures caused a notable accumulation of HSP70 and HSP90 in tomato seedlings. HSP70 accumulated more prominently at 45 °C, while HSP90 accumulated more prominently at 37 °C. Accordingly, and in contrast to HSP90, the high level of HSP70 correlated with induced HSFB1 and HSFA3 expression in seedlings. HSPs release HSFs in response to heat stress which regulate target gene transcription in the nucleus, among which are HSP genes [10,13]. Consequently, HSPs and HSFs are in constant interaction and interdependence jointly relieving the negative effects of heat stress. Other clients for HSP90 include the auxin receptors from the TIR1 family and might therefore be directly connected to auxin induced growth during heat stress, since a positive effect on Arabidopsis seedling growth was reported through stabilization of the receptor by HSP90 [60]. Whether HSP90 might exert a similar function in tomato has yet to be investigated. In conclusion, our work has shown a complex response pattern of tomatoes to heat stress. The global climate change and expected temperature rise in the future will assuredly impede tomato seed germination and plant development, and consequently affect not only commercially important fruit yield but also tomato biodiversity and its geographical distribution [61]. Our results indicate that the tomato heat stress response is a complex, developmental stage- and heat stress type-dependent process. The heat waves, which are expected to appear more frequently by climate change, caused more pronounced deviations than prolonged exposure to 37 °C. In Figure 8, the results of all biochemical and molecular parameters are summarized. Both types of heat stress caused an increase in proline, IAA, HSP70 and HSP90 proteins and HSFB1, DREB1 and NAC4 gene expression, but reduced ABA levels in seedlings. Expectedly, adult plants were more resilient to heat stress. Both treatments significantly reduced ACC levels in adult plants. At 45 °C, proline, HSP90 and DREB1 increased outstandingly and can therefore serve in the future examination as stress markers for the evaluation of heat stress effects on tomato or be considered targets during the generation of thermotolerant tomato plants either by bioengineering or by molecular breeding. Finally, we would like to emphasize the negative effect of heat stress on root development and growth in seedlings, which appears to be connected to ACC production in tomato and needs to be addressed in more details in the future. Tomato (Solanum lycopersicum L.) cultivar Ailsa Craig (Seed Megastore, Nuneaton, UK) was chosen based on its heat tolerance known from previous studies and its sequenced genome [62,63,64]. For all in vitro assays, seeds were surface sterilized by soaking in 70% EtOH for 1 min and then in 2.5% NaOCl and 0.02% Triton X-100 for 30 min. After a fivefold rinsing step with sterile distilled water, seeds were sown on Murashige and Skoog (MS) medium supplemented with 2% sucrose and 1% agar. Plated seeds were stratified at 4 °C for two days before transferring them to growth chambers in a light/dark cycle of 16/8 h (90–100 µmol/m2 s) and 24 °C. Regarding adult plant stress treatments, plants were grown in plastic pots containing steam sterilized commercial soil (Einheitserde Classic Pikiererde CL P, Gebrüder Patzer GmbH & Co. KG, Sinntal, Germany), in a greenhouse at 24 °C, under natural illumination conditions with a photoperiod of 13–15 h (April to June) and a relative humidity of 60%. The 11-week-old plants were exposed to heat treatments. The effect of elevated temperatures on tomato seed germination was examined at different heat regimes, ranging from 24 °C to 37 °C. Seeds (25–30 seeds per 120 × 120 mm Petri dish) were exposed to continuous temperatures for 8 days in a light/dark cycle of 16/8 h. Radicle emergence was taken as a criterion for germination. The germination percentage was calculated from the ratio of germinated and total seeds. On the last day of the experiment, germinated seedlings were photographed, and hypocotyl lengths measured. The survival rate of 12-day-old seedlings was evaluated by exposing seedlings to 37 °C for 24 h or 45 °C for 1, 3, 6 and 12 h. The percentage of surviving seedlings was estimated after a 7-day recovery period at 24 °C (Figure S1A). Seedlings with continued epicotyl elongation and newly developed true leaves were scored as viable. All molecular and physiological analyses were performed on 12-day-old seedlings germinated at 24 °C on MS medium in Magenta vessels (Sigma-Aldrich) and on 11-week-old adult plants cultivated in long day conditions. Seedlings and adult plants were exposed to 37 °C continuously for 24 h or to a heat wave treatment at 45 °C, for 3 h or 5 h, respectively (Figure S1). Both heat treatments were set up at roughly 10 a.m. After the heat wave treatment, seedlings and adult plants were cultivated at 24 °C until sampling. Whole seedlings (15 per one biological replicate) or young leaves pooled from 6 adult plants (third to fifth leaf from the top of the stem) were sampled 24 h after the beginning of the heat stress treatments. The samples were frozen in liquid nitrogen and stored at −80 °C. During all heat treatments, temperatures were monitored and recorded by a data logger Testo 174H (Testo GmbH & Co., Lenzkirch, Germany). To assess the effect of heat treatments on root development, seeds were sown in square Petri dishes (120 × 120 mm), stratified at 4 °C for 2 days and cultured vertically under control conditions for 7 days. Seedlings were either exposed to 45 °C for 3 h and then returned to control conditions until analysis or continuously cultivated at 37 °C. Control seedlings were continuously cultivated at 24 °C (Figure S1). All groups were analyzed simultaneously after 5 days from the start of the treatment when the root tips reached the bottom of the dish in control conditions. At this time point, the plates were photographed, and root growth was assessed. Primary root length, lateral root number and total root length per seedling were measured. Primary root growth rate was expressed as the difference between final primary root length and the root length at the beginning of heat treatment. The average fresh mass per seedling was calculated by weighing 10 whole seedlings at the end of the experiment for each treatment and control. Hypocotyl and root lengths were measured using ImageJ software (NIH, Bethesda, MD, USA). Endogenous IAA, ABA and ACC were quantified by gas chromatography–mass spectrometry (GC–MS) according to adapted protocols originally described in [65,66,67]. Briefly, 100 ng of labelled standards of 13C6-IAA (Cambridge Isotope Laboratories, Andover, MA, USA), 200 ng of 2H6-ABA (Cambridge Isotope Laboratories, Andover, MA, USA) and 100 ng of 2H4-ACC (Euriso-top GmbH, Saarbrücken, Germany) were added directly to 100 mg of frozen homogenized plant tissue (whole seedlings or leaves). The downstream process and combined derivatization proceeded as described by [66]. After adding anhydrous sodium sulfate and a brief centrifugation at 16,000 g for 1 min, solutions were transferred to GC vials and evaporated to dryness in a stream of nitrogen. Following evaporation, an additional step of derivatization was added [68]. Methanol and trimethylsilyl diazomethane (TMSD; diluted 1:100 in diethyl ether) were added in a 1:1 ratio to the dried samples and incubated at room temperature before repeating the evaporation step. In the final step of sample preparation, the dried samples were dissolved in 50 μL ethyl acetate for analysis performed by GC–MS (Varian Saturn 2100T, 3800 GC and 8400 Autosampler). Hormone levels were measured by increasing the temperature from 70 to 280 °C at a rate of 20 °C/min. Three biological (15 seedlings or leaves pooled from 6 adult plants per replicate) and three technical replicates were analyzed per treatment or control. Phytohormone content was determined using the principles of isotope dilution [69] from diagnostic ion ratios of endogenous and labelled hormones at a m/z of 190/194, 130/136 and 141/145 for ABA, IAA and ACC, respectively. Fifty mg of frozen homogenized tissue (whole seedlings or leaves) was extracted with 1 mL 70% EtOH and centrifuged at 10,000 g and 4 °C for 10 min. Supernatants were used for proline [70] and MDA content determination [71,72]. Three biological (15 seedlings or leaves pooled from 6 adult plants per replicate) and two technical replicates were analyzed per treatment or control sample. RNA was isolated from 50 mg of frozen homogenized tissue (whole seedlings or leaves) using the MagMAx Plant RNA Isolation Kit (Thermo Scientific) according to the manufacturer’s instructions. After elution, RNA was quantified by NanoDropTM 1000 Spectrophotometer (Thermo Scientific). Then, cDNA was synthesized from 1 µg of isolated RNA using 200 U of RevertAid H Minus Reverse Transcriptase and 2.5 µM Oligo(dT)18 primer (Thermo Scientific). Quantitative real-time PCR (qPCR) was performed on the MIC platform (Bio Molecular Systems). The reactions included 1× GoTaq® qPCR Master Mix reagent (Promega), 200 nM of forward and reverse primers (Table S1) and 20 ng cDNA in a total reaction volume of 10 µL. The run profile of the PCR reaction was as follows: 95 °C for 5 min, followed by 40 cycles of 95 °C for 5 s and 60 °C for 10 s. In addition, melting curves were generated to check for specific amplification by increasing the temperature from 55 °C to 95 °C at 0.5 °C/s. Relative expression of heat stress-related genes DREB1, HSFA3, HSFB1, NAC4 and NAC6 was calculated by the ΔΔCq method [73,74] using ACT [75] and EFI-α [76] genes as endogenous controls. Three biological (each consisting of 15 whole seedlings or leaves pooled from 6 adult plants) and two technical replicates were analyzed per treatment and control. Genes, accession numbers and primer sequences are from [48,49,51,75,76,77] and listed in Table S1. Soluble proteins were extracted from 150 mg frozen homogenized tissue (whole seedlings or leaves) in 0.5 mL of protein extraction buffer (92.5 mM TRIS-HCl, 500 mM sucrose, 6.48 mM DTT, pH 7.6; [78]). Protein concentrations were determined using the Bradford reagent [79]. Proteins (25 μg per lane) were separated on 12%-SDS-polyacrylamide gels and transferred to a PVDF or nitrocellulose membrane. Membranes were blocked in 2% (w/v) non-fat dry milk in 1× tris-buffered saline buffer (TBS) overnight at 4 °C. Primary antibodies, anti-HSP90-1 (Agrisera AS08346) or anti-HSP70 (Agrisera AS08371) diluted 1:3000 in 1× TBS, secondary antibody (Anti-Rabbit IgG HRP goat antibody, EMD Millipore diluted 1:50,000), and Immobilon® Forte Western HRP substrate, (Millipore) were used for HSP90 and HSP70 protein detection. Finally, to assess protein quantity, membranes were stained with either Coomassie brilliant blue or Ponceau S. Images were analyzed in ImageJ as described in [80]. To calculate the changes in HSP70 and HSP90 quantities for each treatment, the respective control values were taken as one (fold change). In all experiments, at least 3 biological replicates per treatment or control were analyzed. Statistical analysis was performed in the TIBCO Statistica 13.5.0.17 software package (TIBCO Software, Palo Alto, CA, USA). Data were validated with regard to distribution (Shapiro–Wilk test) and variance (Levene’s test) before proceeding with the analysis. One-way ANOVA and post-hoc Tukey’s test (p < 0.05) were used to determine the significance. The data were represented as means with standard deviations or box plots. Significant differences are denoted by different letters.

PMC10002529

Qi Zhang,Jingjing Zhang,Fei Wei,Xiaokang Fu,Hengling Wei,Jianhua Lu,Liang Ma,Hantao Wang

The CCCH-Type Zinc-Finger Protein GhC3H20 Enhances Salt Stress Tolerance in Arabidopsis thaliana and Cotton through ABA Signal Transduction Pathway

06-03-2023

cotton,GhC3H20,salt stress,GhPP2CA and GhHAB1,ABA signaling pathway

The CCCH zinc-finger protein contains a typical C3H-type motif widely existing in plants, and it plays an important role in plant growth, development, and stress responses. In this study, a CCCH zinc-finger gene, GhC3H20, was isolated and thoroughly characterized to regulate salt stress in cotton and Arabidopsis. The expression of GhC3H20 was up-regulated under salt, drought, and ABA treatments. GUS activity was detected in the root, stem, leaves, and flowers of ProGhC3H20::GUS transgenic Arabidopsis. Compared with the control, the GUS activity of ProGhC3H20::GUS transgenic Arabidopsis seedlings under NaCl treatment was stronger. Through the genetic transformation of Arabidopsis, three transgenic lines of 35S-GhC3H20 were obtained. Under NaCl and mannitol treatments, the roots of the transgenic lines were significantly longer than those of the wild-type (WT) Arabidopsis. The leaves of the WT turned yellow and wilted under high-concentration salt treatment at the seedling stage, while the leaves of the transgenic Arabidopsis lines did not. Further investigation showed that compared with the WT, the content of catalase (CAT) in the leaves of the transgenic lines was significantly higher. Therefore, compared with the WT, overexpression of GhC3H20 enhanced the salt stress tolerance of transgenic Arabidopsis. A virus-induced gene silencing (VIGS) experiment showed that compared with the control, the leaves of pYL156-GhC3H20 plants were wilted and dehydrated. The content of chlorophyll in pYL156-GhC3H20 leaves was significantly lower than those of the control. Therefore, silencing of GhC3H20 reduced salt stress tolerance in cotton. Two interacting proteins (GhPP2CA and GhHAB1) of GhC3H20 have been identified through a yeast two-hybrid assay. The expression levels of PP2CA and HAB1 in transgenic Arabidopsis were higher than those in the WT, and pYL156-GhC3H20 had expression levels lower than those in the control. GhPP2CA and GhHAB1 are the key genes involved in the ABA signaling pathway. Taken together, our findings demonstrate that GhC3H20 may interact with GhPP2CA and GhHAB1 to participate in the ABA signaling pathway to enhance salt stress tolerance in cotton.

The CCCH-Type Zinc-Finger Protein GhC3H20 Enhances Salt Stress Tolerance in Arabidopsis thaliana and Cotton through ABA Signal Transduction Pathway The CCCH zinc-finger protein contains a typical C3H-type motif widely existing in plants, and it plays an important role in plant growth, development, and stress responses. In this study, a CCCH zinc-finger gene, GhC3H20, was isolated and thoroughly characterized to regulate salt stress in cotton and Arabidopsis. The expression of GhC3H20 was up-regulated under salt, drought, and ABA treatments. GUS activity was detected in the root, stem, leaves, and flowers of ProGhC3H20::GUS transgenic Arabidopsis. Compared with the control, the GUS activity of ProGhC3H20::GUS transgenic Arabidopsis seedlings under NaCl treatment was stronger. Through the genetic transformation of Arabidopsis, three transgenic lines of 35S-GhC3H20 were obtained. Under NaCl and mannitol treatments, the roots of the transgenic lines were significantly longer than those of the wild-type (WT) Arabidopsis. The leaves of the WT turned yellow and wilted under high-concentration salt treatment at the seedling stage, while the leaves of the transgenic Arabidopsis lines did not. Further investigation showed that compared with the WT, the content of catalase (CAT) in the leaves of the transgenic lines was significantly higher. Therefore, compared with the WT, overexpression of GhC3H20 enhanced the salt stress tolerance of transgenic Arabidopsis. A virus-induced gene silencing (VIGS) experiment showed that compared with the control, the leaves of pYL156-GhC3H20 plants were wilted and dehydrated. The content of chlorophyll in pYL156-GhC3H20 leaves was significantly lower than those of the control. Therefore, silencing of GhC3H20 reduced salt stress tolerance in cotton. Two interacting proteins (GhPP2CA and GhHAB1) of GhC3H20 have been identified through a yeast two-hybrid assay. The expression levels of PP2CA and HAB1 in transgenic Arabidopsis were higher than those in the WT, and pYL156-GhC3H20 had expression levels lower than those in the control. GhPP2CA and GhHAB1 are the key genes involved in the ABA signaling pathway. Taken together, our findings demonstrate that GhC3H20 may interact with GhPP2CA and GhHAB1 to participate in the ABA signaling pathway to enhance salt stress tolerance in cotton. Cotton is an important oil crop and fiber crop in China, which plays an important role in the national economy [1]. Soil salinization affects not only cotton yield but also cotton quality. It is essential to study the mechanism of upland cotton in response to salt stress, which is closely linked to cotton production and the agricultural economy in China [2]. Transcription factors can activate the expression of downstream genes and form a transcriptional regulatory network in response to stress [3,4]. Transcription factors such as the AP2/ERF family members related to plant stress can regulate the expression of multiple functional genes related to plant stress. It is more effective to use transcription factors to study the stress resistance of crops and cultivate good stress-resistant varieties than to use single genes for the genetic improvement of crops [5]. The zinc-finger transcription factor family, as one of the largest transcription factor families in plants, plays an important role in multiple biological processes, such as morphogenesis, signal transduction, and environmental stress responses [6,7]. Zinc-finger transcription factors contain zinc-finger motifs in which cysteines and/or histidines coordinate with a few zinc atoms to form the local peptide structures that are essential for their specific functions [8]. Several plant zinc-finger families, such as the RING-finger, ERF, WRKY, DOF, and LIM families, regulate gene expression with the aid of DNA-binding, protein-binding proteins, or RNA-binding proteins [9,10,11,12,13]. According to their structural diversities, the zinc-finger transcription factor family has been classified into nine types: C2H2, C8, C6, C3HC4, C2HC, C2HC5, C4, C4HC3, and CCCH [6,8,14,15,16]. The CCCH family contains a typical C3H-type motif, and members of this family have already been identified in organisms from yeast to humans [8,9,10]. In plants, multiple CCCH zinc-finger proteins were found to be involved in abiotic and biotic stresses. For example, as a nuclear CCCH zinc-finger protein, overexpressed GhZFP1 enhanced salt stress and fungal disease tolerance in transgenic tobacco plants by interacting with GZIRD21A and GZIPR5 [17]. The CCCH zinc-finger members participated in salt stress via the ABA signaling pathway. Overexpressed AtOZF2 Arabidopsis plants were insensitive to salt stress, and the mutant atozf2 Arabidopsis plant significantly reduced salt stress tolerance. Further study showed that AtOZF2 regulated salt stress via the ABA signaling pathway mediated by ABI2 [18]. Under salt stress, the expression of three salt stress-responsive genes (RAB18, COR15A, and RD22) in the ABA-dependent pathway was significantly higher in AtC3H17 OXs than in the WT. The results indicated that the CCCH zinc-finger gene AtC3H17 may be involved in salt stress by the ABA-dependent pathway [19]. As a nuclear transcriptional activator, IbC3H18 interacted with IbPR5 and enhanced salt and drought stress tolerance in transgenic tobacco plants by regulating the expression of a range of abiotic stress-responsive genes involved in reactive oxygen species (ROS) scavenging, ABA signaling, photosynthesis, and ion transport pathways [20]. Pieces of evidence show that CCCH zinc-finger genes are involved in response to salt stress through the ABA-dependent pathway. In this study, based on the transcriptome data related to salt stress [21], an up-regulated CCCH zinc-finger gene GhC3H20 was identified. qRT-PCR results showed that GhC3H20 was up-regulated under NaCl, PEG, and ABA treatments. Compared with the control, the GUS activity of ProGhC3H20::GUS transgenic Arabidopsis seedlings under NaCl treatment was stronger. Overexpression of GhC3H20 in Arabidopsis could promote salt stress tolerance, and silencing of GhC3H20 in cotton could decrease salt stress tolerance. In addition, GhC3H20 could interact with GhPP2CA and GhHAB1 to regulate salt stress by the ABA signaling pathway. This study lays the foundation for future studies of GhC3H20 in the improvement of salt stress tolerance in cotton. The coding sequences of the GhC3H20 (GH_D08G2771) gene were cloned from the upland cotton material TM-1. The open reading frame of GhC3H20 was 1062 bp and encoded 353 amino acid residues. The estimated molecular mass (Mw) of the GhC3H20 protein was 39.3 kDa, and the isoelectric point (pI) was 6.23. Subcellular prediction analysis results showed that GhC3H20 might be located in the nucleus of the cell. The result of gene structure analysis showed that GhC3H20 contained an exon (Figure 1A). The cotton GhC3H20 gene is homologous to the Arabidopsis AtC3H20 (AT2G19810) gene, and the protein sequence similarity is 56%. In a previous study, phylogenetic analysis result revealed that AtC3H20 belonged to the CCCH zinc-finger family group IX [22]. The neighbor-joining tree result of GhC3H20 and CCCH family group IX members in Arabidopsis showed GhC3H20, AT4G29190, AT2G19810, and AT2G25900 divided into one branch (Figure 1B). The protein sequence alignments of GhC3H20, AT4G29190, AT2G19810, and AT2G25900 showed that all four proteins contained two CCCH motifs and most amino acid residues were conserved (Figure 1C). The transcriptome data of salt treatment showed that the GhC3H20 gene responded to salt stress [21]. Therefore, the GhC3H20 gene was selected to do qRT-PCR under salt and PEG treatments. qRT-PCR results revealed that the expression levels of GhC3H20 were the highest at 48 h and 24 h of NaCl and PEG treatments (Figure 2A,B). The results showed that the GhC3H20 gene was up-regulated under salt and drought stresses. The expression levels of the GhC3H20 gene in eight tissue samples (vegetative organs: roots, stems, leaves, and buds; reproductive organs: petals, stamens, pistils, and fibers) of upland cotton were determined by qRT-PCR. As shown in Figure 2C, in reproductive organs, the expression levels of the GhC3H20 gene were highest in stamens, and, among vegetative organs, the expression levels of the GhC3H20 gene were highest in stems. The PlantCare website was employed to analyze cis-acting elements in the 2000 bp promoter of GhC3H20. The result revealed that this region included stress-response, light-response, and hormone-response elements. Among hormone-response elements, ABA-response elements occur in the largest number as six elements (Supplementary Table S1). ABA (100 μM) was sprayed on the leaves of upland TM-1 cotton seedlings and the expression levels of GhC3H20 were measured by qRT-PCR. The results showed that after ABA treatment, the expression levels of GhC3H20 were increased significantly from 0 h to 6 h and decreased from 6 h to 12 h (Figure 3A). The results indicated that the GhC3H20 gene had responded to ABA. To understand the tissue expression specificity of GhC3H20, the ProGhC3H20:: GUS vector was constructed and genetically transformed into Arabidopsis thaliana. The roots, stems, leaves, flowers, and fruit pods of T1 generation transgenic Arabidopsis plants were taken for GUS staining, respectively. The results showed GUS staining was found in all tissues of transgenic Arabidopsis plants, except fruit pods (Figure 3B). To further understand the promoter of GhC3H20 in response to salt stress, the GUS staining of transgenic seedlings was detected under NaCl (150 mM) treatment. GUS activity was more strongly expressed in the leaves and stems of NaCl-treated transgenic Arabidopsis seedlings than in the control. (Figure 3C,D). To further understand the relationship of GhC3H20 in response to salt stress, the 35S-GhC3H20 vector was constructed and transformed into Arabidopsis. The transgenic Arabidopsis were detected by 1/2 MS (+kana) and PCR. The T3 generation of transgenic Arabidopsis was used for further analysis under salt stress. qRT-PCR results indicated that three transgenic Arabidopsis lines were significantly overexpressed relative to the WT (Figure 4A). To further understand the function of the GhC3H20 gene in response to salt and osmotic stresses during seedling stages, the root length of ten Arabidopsis seedlings were measured under 0, 150 mM NaCl, and 200 mM mannitol treatments (Figure 4B). Under NaCl and mannitol treatments, the roots of the transgenic lines were significantly longer than those of the WT (Figure 4C). The results indicated that overexpression of GhC3H20 could enhance the root growth of transgenic Arabidopsis seedlings under salt and osmotic stresses. To analyze the function of the GhC3H20 gene in response to salt stress at the seedling stage, WT and transgenic Arabidopsis were planted in nutrient soil and treated with 400 mM NaCl to observe the phenotype. After salt treatment for 5 days, the leaves of the WT wilted and turned yellow. While the transgenic lines still had green rosette leaves (Figure 5A). To better understand the physiological changes in plants, after salt treatment, the leaves of WT and transgenic Arabidopsis were taken, and the activities of CAT were measured. The activities of CAT in the transgenic Arabidopsis lines were significantly higher than those in the WT (Figure 5B). All results indicated that overexpression of GhC3H20 could enhance the salt stress tolerance of Arabidopsis plants. To further analyze the putative function of the GhC3H20 gene, a VIGS strategy was used to knock down the expression levels of the GhC3H20 gene in cotton. Cotton standard line TM-1 plants were grown until the third true leaf expanded and was treated with 400 mM NaCl for five days. Figure 6A shows that PYL156-GhPDS plants (positive control) had an obvious leaf-whitening phenotype. The leaves of the PYL156-GhC3H20 (silenced plants) plants wilted and turned yellow, while the leaves of the PYL156 (control plants) plants only slightly shrank due to water loss (Figure 6A). The expression levels of GhC3H20 determined by qRT-PCR indicated that the gene was effectively silenced (Figure 6B). To understand the physiological changes in plants, after salt treatment, the leaves of the control and the silenced plants were taken, and the content of chlorophyll was measured. The content of chlorophyll in the leaves of the control plants was significantly higher than that of the leaves of the silenced plants (Figure 6C). These results indicated that silencing of the GhC3H20 gene reduced salt stress tolerance in cotton. A transcriptional activation assay was performed in the yeast cells. The yeast cells containing pGADT7+pGBKT7-GhC3H20 (experimental group), pGADT7-large T+pGBKT7-lamin C (negative control), and pGADT7-large T+pGBKT7-p53 (positive control) were transformed into the yeast cells and cultured on SD/−Trp/−Leu and SD/−Trp/−Leu/−His/−Ade medium. The positive control and the experimental group grew well on SD/−Trp/−Leu and SD/−Trp/−Leu/−His/−Ade medium, while the negative control could not, demonstrating that GhC3H20 could autonomously activate the reporter genes in the absence of a prey protein (Figure 7A). Previous studies have reported that CCCH zinc-finger genes participate in salt stress through the ABA signaling pathway [18,19,20]. Therefore, four key genes (GhPP2CA, GhHAB1, GhABF3, and GhABI1) of ABA signal transduction were selected to do the Y2H assay. The coding sequences of GhPP2CA, GhHAB1, GhABF3, and GhABI1 were cloned into the pGADT7 vector. The constructive vectors with pGBKT7-GhC3H20 were co-transformed into the yeast strain Y2HGold. The results showed that the yeast cells containing the positive control pGBKT7-p53+pGADT7-largeT and the experimental group pGBKT7-GhC3H20 +pGADT7-GhPP2CA and pGBKT7-GhC3H20 +pGADT7-GhHAB1 could grow well on SD/−Trp/−Leu/-His/−Ade+40 mM 3AT medium and turned blue on SD/−Trp/−Leu/−His/−Ade/X-a-Gal/AbA+40 mM 3AT medium. Yeast containing the negative control, the control group pGADT7+pGBKT7-GhC3H20, and the experimental group pGBKT7-GhC3H20+ pGADT7-GhABF3 and pGBKT7-GhC3H20+ pGADT7-GhABI1 could not grow on SD-TLHA+40 mM 3AT medium and turned blue on SD-TLHA+X-α-Gal +40 mM 3AT medium. The results showed that GhC3H20 could interact with GhPP2CA and GhHAB1 but did not interact with GhABF3 and GhABI1 (Figure 7B). qRT-PCR results indicated that the expression of GhC3H20 was up-regulated under ABA and PEG treatments. Overexpression of GhC3H20 enhanced the osmotic stress tolerance in Arabidopsis seedlings. Therefore, two interaction genes (PP2CA and HAB1) of GhC3H20 and two osmotic stress-related genes (AtNHX1 and GhNHX2) were used to do qRT-PCR in Arabidopsis and cotton, respectively. Under salt stress, the expression levels of AtPP2CA, AtHAB1, and AtNHX1 in transgenic Arabidopsis were higher than those in the WT, except GhHAB1 in Line 3 (Figure 8A,B), and the expression levels of GhPP2CA, GhHAB1, and GhNHX2 in the silenced plants were lower than those in the control plants (Figure 8C,D). The expression of GhPP2CA and GhHAB1 were the highest at 6 h of NaCl treatment (Figure 8E). PP2CA and HAB1 are key genes in ABA signaling transduction. NHX protein is a Na+/H+ antiporter, which regulates osmotic stress by transporting intracellular Na+ to be extracellular under high salt concentrations. It is most likely that GhC3H20 could enhance salt stress tolerance through the ABA signal transduction pathway and osmotic stress pathway. Having been cloned in various plants such as Arabidopsis [23], rice [24], and sweet potato [20], zinc-finger transcription factors play an important role in the regulation of plant abiotic stress responses. However, the regulatory mechanism of the zinc-finger gene in response to salt stress in cotton remains poorly understood. In our study, a salt-induced gene (GhC3H20) from the zinc-finger family was identified by transcriptome data of salt stress [21]. The gene encodes 353 amino acids, with two C3H-type motifs and no intron, which indicated that the C3H20 gene belongs to the CCCH zinc-finger subfamily [22]. Phylogenetic analysis also revealed that the cotton GhC3H20 gene belongs to a novel CCCH-type zinc-finger protein subfamily. All members of CCCH zinc-finger group IX in Arabidopsis have some characteristics in common, consisting of two CCCH-type motifs. In addition, gene structure analysis indicated that they are all intronless genes [22]. The CCCH zinc-finger family has been reported to be involved in salt stress [25]. More recently, the rice TZF1 gene has been reported to regulate the expression of many abiotic stress tolerance genes to enhance salt stress tolerance in transgenic Oryza sativa L. plants [26]. The first CCCH zinc-finger protein GhZFP1 in cotton has been identified and functionally characterized. Further research showed that GhZFP1 interacted with GZIRD21A and GZIPR5 to regulate salt stress in cotton [17]. Our study lays the foundation for future studies of the CCCH zinc-finger members in the improvement of salt stress tolerance in cotton. In the present study, the GhC3H20 gene was induced by salt, drought, and ABA treatments. The homologous genes of GhC3H20 in Arabidopsis, AtOZF1, and AtOZF2 were also induced by salinity and ABA [18]. The cotton GhZFP1 gene was induced by salt and drought stresses [17]. The result indicated that GhC3H20 is likely to be involved in salt stress. To understand the relationship between the GhC3H20 gene and salt stress, we obtained three transgenic Arabidopsis lines through genetic transformation. Overexpressed GhC3H20 enhanced salt stress tolerance in transgenic Arabidopsis seedlings. Silencing of GhC3H20 decreased the salt stress tolerance in cotton. Plants produce some reactive oxygen species (ROS) under salt stress, such as O2, H2O2, O2−, and HO, which are highly active molecules and can cause oxidative damage to proteins, DNA, and lipids [27,28]. CAT is the most important antioxidant enzyme to help remove excess ROS in plants, maintain a low level of ROS, and improve the tolerance of plants under stress [29,30]. In this work, we found that under salt treatment, transgenic Arabidopsis had higher CAT activity compared with the WT, suggesting that the GhC3H20 gene can increase the relevant antioxidant enzymes in transgenic Arabidopsis in response to salt stress. Chlorophyll is an important material for the photosynthesis of plants. Under stress, the content of chlorophyll decreases and photosynthesis decreases, thus causing damage to plants. Under salt stress, the control cotton plants had a higher content of chlorophyll compared with silent plants. The expression of genes (AtNHX1 and GhNHX2) related to osmotic stress was higher in transgenic Arabidopsis than in the WT and was lower in the silenced plants than in the control plants. It is likely that stress-inducible genes were induced by the expression of GhC3H20. The increased abundance of GhC3H20 transcripts probably up-regulated the transcription of several stress-inducible genes, which in turn contributed to increased endurance under the stress conditions. It has been reported that overexpression of the CCCH-tandem zinc-finger protein OsTZF1 promotes salt stress tolerance by inducing the expression of some stress-related genes in Ubi:OsTZF1 OX plants [31]. These results demonstrated that GhC3H20 might participate as a positive transcript factor in the regulation of salt stress. As an important signaling molecule, ABA plays a key role in abiotic stresses such as salt stress [32,33,34]. Under stress, ABA can improve plant tolerance by altering stomatal closure, modulating root architecture, and osmolyte biosynthesis [35,36]. Some genes enhance plant tolerance through the ABA signaling pathway or the ABA biosynthesis pathway [37,38]. What is more, the expression levels of the stress-related genes were decreased in mutants associated with defects in ABA biosynthesis or response. For example, under salt stress, the expression levels of the RD29B gene in ABA1 and ABI1 mutants are extremely low or do not expressed [39]. All evidence suggests that ABA is a key hormone in response to stress. Multiple studies have reported that CCCH zinc-finger proteins participate in salt stress via the ABA signaling pathway [18,19,20]. In our study, the cotton GhC3H20 gene was up-regulated under ABA treatment. To elucidate the precise molecular mechanism of GhC3H20 increasing salt stress tolerance in transgenic plants and cotton, two proteins–GhPP2CA and GhHAB1–that interact with GhC3H20 were isolated and identified by Y2H assay. The expression of GhPP2CA and GhHAB1 was induced by salt stress. Under salt stress, the expression levels of AtPP2CA and AtHAB1 in transgenic Arabidopsis plants were higher than those in WT Arabidopsis, except AtHAB1 in Line 3, and the expression levels of GhPP2CA and GhHAB1 in silenced plants were lower than that in control plants. GhPP2CA and GhHAB1 genes were likely induced by the expression of GhC3H20. The increased abundance of GhC3H20 transcripts probably up-regulated the transcription of GhPP2CA and GhHAB1 genes, which in turn contributed to increased endurance under salt stress. AtOZF2 was reported to enhance salt stress tolerance by increasing and decreasing the expression of the AtABI2 gene in transgenic Arabidopsis plants. Further research found AtOZF2 participated in the ABA and salt stress responses through the ABI2-mediated signaling pathway [18]. Those pieces of evidence indicated that the GhC3H20 might interact with GhPP2CA and GhHAB1 to regulate salt stress by the ABA signaling pathway. As negative regulators participated in the ABA signaling pathway, PP2CA and HAB1 were also involved in the regulation of stresses. PP2CA together with ABI1 and SnRK2.4 regulate root length under salt stress [40]. Located close to PP2CA, Type 2C Protein Phosphatases SlPP2C1 gene RNAi plants displayed delayed root growth [41]. In this research, overexpression of GhC3H20 enhanced root length in Arabidopsis seedlings under salt stress. Therefore, we speculated that GhC3H20 was involved in ABA signaling to regulate root growth by binding to GhPP2CA and GhHAB1 under salt stress. G. hirsutum standard line TM-1 was used in this study. Cotton seeds were planted in the sand and grown in a plant growth chamber at 25 °C with a 16 h light/8 h dark photoperiod condition. Until the third true leaf expanded for salt or drought treatment, seedlings were soaked in 200 mM sodium chloride (NaCl) or 20% polyethylene glycol 6000 (PEG6000) solution, respectively. Simultaneously, seedlings watered with the deionized water was used as a control. The three roots of the control and salt-treated seedlings were collected for qRT-PCR assay at each time point of 0, 1, 3, 6, 12, 24, and 48 h. The three leaves of the control and drought-treated seedlings were harvested for qRT-PCR assay at each time point of 0, 1, 3, 6, 12, 24, and 48 h. For ABA treatment, the 100 μM abscisic acid (ABA) solution was sprayed onto leaves. Samples from three leaves were collected at each time point of 0, 3, 6, 9, and 12 h. For the tissue-specific test, G. hirsutum standard line TM-1 was planted in the field. The root, stem, leaves, and buds were taken during the third leaf time. The petals, stamens, and pistils were taken during flowering. A total of 15 days of fiber were used to do this study. Then, samples were immediately frozen in liquid nitrogen and stored at −80 °C in a refrigerator for subsequent experiments. The cetyl-trimethylammonium bromide (CTAB) method was used to extract genomic DNA as in the previous study [42]. The total RNA of leaves and roots was extracted and purified using an EASY Spin plus plant Total RNA Extraction kit (SunYa, Henan, China) or an RNAprep Pure Plant Kit (Tiangen, Beijing, China) according to the manufacturer’s instructions. First-strand synthesis of cDNA was performed by the PrimeScript™ RT Reagent kit with gDNA Eraser (Takara, Japan) or the HiScriptII Q RT SuperMix Vazyme for the qPCR (+gDNA wiper) (Vazyme, Nanjing, China) kit according to the manufacturer’s instructions. A cotton actin gene (GhActin) and an Arabidopsis thaliana polyubiquitin gene (AtUBQ10) were used as standard controls. All the primers used for qRT-PCR were designed using a primer database (http://biodb.swu.edu.cn/qprimerdb, last accessed on 26 February 2023) and are listed in Supplementary Table S2. The expression of genes was analyzed using a 7500 Real-Time PCR System (Applied Biosystems, Waltham, MA, USA) with the UltraSYBR Mixture (Low ROX) Kit. The 10 μL reaction volume contained the following components: 0.4 μL of the PCR forward primer (10 μM), 0.4 μL of the PCR reverse primer (10 μM), 1 μL of cDNA, 5 μL of SYBR Premix Ex Taq (2×), and 3.6 μL of ddH2O. The amplification parameters were as follows: 95 °C for 10 min, 40 cycles of 95 °C for 10 s, 60 °C for 30 s, and 72 °C for 32 s. The 2−ΔΔCt method was applied to calculate the relative expression levels with three technical replicates [43]. The full-length coding sequences (CDS) and the 2000 bp upstream sequences of the GhC3H20 (GH_D08G2771) gene from G.hirsutum TM-1 were cloned by specific primers (Supplementary Table S3). For the GUS assay, the GhC3H20 promoter sequences were inserted into the PBI121 binary vector with HindIII and XbaI restriction sites to generate the ProGhC3H20::GUS vector. For overexpression studies, the GhC3H20 CDS was inserted into the pBI121 binary vector with XbaI and SacI restriction sites to generate the 35S-GhC3H20 vector. The vectors were transferred into the Agrobacterium tumefaciens strain GV3101 by the freezing and thawing method [44]. Multiple sequence alignments were performed using DNAMAN software, and a phylogenetic tree was conducted by MEGA 7.0 software using a neighbor-joining method. The GhC3H20 gene structure was predicted by the online website Gene Structure Display Server (GSDS 2.0) (http://gsds.cbi.pku.edu.cn/). The molecular weight (Mw) and isoelectric point (pI) of the GhC3H20 protein were predicted by the online website ExPASy (http://web.expasy.org/compute_pi/). The online website ProtComp 9.0 (http://linux1.softberry.com/berry.phtml?topic=protcompan&group=programs&subgroup=proloc) was used to predict the subcellular localization of the GhC3H20 protein. The online website PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) was employed to predict the cis-elements. The Arabidopsis thaliana Columbia ecotype was used for this study. The floral dip method was used to generate transgenic Arabidopsis plants [45]. Positive transformants were selected on 1/2 MS medium containing kanamycin (50 mg/l) and grew until maturation. Genomic DNA was extracted via the TPS method for PCR to test positive plants. The homozygous overexpressed transgenic lines of T3 generations were used for phenotypic analysis. The GUS staining kit (HuaYueYang, Bejing, China) was used for this experiment. For salt treatment, seeds of transgenic plants were placed on the 1/2 MS medium containing 150 mM NaCl. Samples from the roots, stems, leaves, inflorescences, and fruit pods of the T1 generation transgenic Arabidopsis were placed in the 5 mL tube to which the GUS staining solution was added. After 24 h at 37 °C in the dark, 75% alcohol was added to the tubes for decolorization. For length experiments, seeds of transgenic and WT Arabidopsis were germinated and grown on 1/2 MS medium supplemented with 0 (control), 150 mM NaCl, and 200 mM mannitol. After dark treatment at 4 °C for 2 days, the 1/2 MS medium plates with the seeds were placed in an incubator with a photoperiod of 16 h light/8 h dark at 22 °C. Ten seedlings were grown for two weeks. The root length in WT and transgenic lines were measured. Seedlings were transplanted in the nutrition soil and grown in a plant growth chamber at 25 °C with a 16 h light/8 h dark photoperiod condition. One-month-old plants were subjected to salt treatment. For salt treatment, the transgenic and WT plants were watered with 400 mM NaCl and then kept in the NaCl-contained soil for 5 days. The growth status of the transgenic and WT plants under high salinity (NaCl) was observed. For the VIGS assay, the online website SGN VIGS Tool (https://vigs.solgenomics.net/?tdsourcetag=s_pcqq_aiomsg) was applied to obtain the silenced fragments of GhC3H20. The gene-specific primers were designed by the NCBI Primer-BLAST tool and are listed in Supplementary Table S3. The silenced fragments were cloned by PCR and integrated into the PYL156 binary vector to construct the PYL156-GhC3H20 vector. The constructed vector and the PYL156 (negative control), PYL-GhPDS (positive control), and PLY192 (helper vector) vectors were then introduced into the Agrobacterium tumefaciens strain LBA4404. The LBA4404 cells harboring the vectors were collected and resuspended in filtration buffer (10 mM MgCl2, 10 mM MES, and 200 μM acetosyringone). The LBA4404 cells harboring PYL156 or PYL156-GhC3H20 were equally mixed with PYL192 and co-injected into two fully expanded cotyledons of TM-1 plants. Until the third true leaves expanded, twenty seedlings were treated with 400 mM NaCl solution. The growth status of the PYL156 and PYL156 derivative plants under high salinity (NaCl) was observed. After salt treatment, approximately 0.1 g of samples from leaves were collected and immediately frozen in liquid nitrogen. The CAT content detection kit (Solarbio, Beijing, China) was used to measure the content of CAT. Samples were placed in the 15 mL tube to which 15 mL of mixed liquid (acetone: absolute ethanol = 1:1) was added and treated for 24 h at room temperature in the dark. When the leaves turned white, the absorbance was measured at 645 nm and 663 nm. The CDS sequences of GhC3H20, GhPP2CA, GhHAB1, GhABF3, and GhABI1 were cloned and amplified by PCR. The fragments of CDS sequences were inserted into the pGBKT7 binary vector with EcoRI and BamHI sites to create pGBKT7-GhC3H20, pGBKT7-GhPP2CA, pGBKT7-GhHAB1, pGBKT7-GhABF3, and pGBKT7-GhABI1 plasmids. The pGBKT7-GhC3H20 with pGADT7, pGBKT7-GhPP2CA, pGBKT7-GhHAB1, pGBKT7-GhABF3, or pGBKT7-GhABI1 plasmids were co-transformed into Y2HGold cells. The yeast cells containing transformed products, pGBKT7-p53+pGADT7-largeT (positive control), and pGBKT7-laminC+pGADT7-largeT (negative control) were cultured and detected on SD-Trp-Leu, SD-Trp-Leu-His-Ade containing 0, 20, 40, 60, 80, 100 mM 3AT, and SD-Trp-Leu-His-Ade+X-a-Gal+40 mM 3AT medium at 30 °C for 3–5 days. As one of the transcription factors widely existing in plants, the CCCH zinc-finger protein not only plays pivotal roles in plant growth and development but also participates in stress response. In this study, a CCCH zinc-finger protein, the GhC3H20 gene, was identified. qRT-PCR results showed that GhC3H20 was up-regulated under NaCl, PEG, and ABA treatments. GUS activity was detected in the root, stem, leaves, and flowers of ProGhC3H20::GUS transgenic Arabidopsis. Compared with the control, the GUS activity of ProGhC3H20::GUS transgenic Arabidopsis seedlings under NaCl treatment was stronger. Overexpressed GhC3H20 enhanced the salt stress tolerance in Arabidopsis, and silencing of GhC3H20 decreased the salt stress tolerance in cotton. In the Y2H assay, two interact genes (GhPP2CA and GhHAB1) of GhC3H20 were identified. GhPP2CA and GhHAB1 are the key genes involved in the ABA signaling pathway. The expression levels of PP2CA and HAB1 in transgenic Arabidopsis were higher than those in the WT, and pYL156-GhC3H20 plants were lower than those in the control. GhC3H20 is likely to participate in the ABA signaling pathway to regulate salt stress by interacting with GhPP2CA and GhHAB1. This study lays the foundation for future studies of GhC3H20 in the improvement of salt stress tolerance in cotton. For further study, transgenic cotton of GhC3H20 will be constructed to investigate the biological function in response to salt stress. What is more, the downstream and upstream genes of GhC3H20 will be further studied to form a regulatory network in response to salt stress.

PMC10002531

Cong-Hua Feng,Meng-Xue Niu,Xiao Liu,Yu Bao,Shujing Liu,Meiying Liu,Fang He,Shuo Han,Chao Liu,Hou-Ling Wang,Weilun Yin,Yanyan Su,Xinli Xia

Genome-Wide Analysis of the FBA Subfamily of the Poplar F-Box Gene Family and Its Role under Drought Stress

02-03-2023

F-box protein,P. trichocarpa,FBA genes,drought stress

F-box proteins are important components of eukaryotic SCF E3 ubiquitin ligase complexes, which specifically determine protein substrate proteasomal degradation during plant growth and development, as well as biotic and abiotic stress. It has been found that the FBA (F-box associated) protein family is one of the largest subgroups of the widely prevalent F-box family and plays significant roles in plant development and stress response. However, the FBA gene family in poplar has not been systematically studied to date. In this study, a total of 337 F-box candidate genes were discovered based on the fourth-generation genome resequencing of P. trichocarpa. The domain analysis and classification of candidate genes revealed that 74 of these candidate genes belong to the FBA protein family. The poplar F-box genes have undergone multiple gene replication events, particularly in the FBA subfamily, and their evolution can be attributed to genome-wide duplication (WGD) and tandem duplication (TD). In addition, we investigated the P. trichocarpa FBA subfamily using the PlantGenIE database and quantitative real-time PCR (qRT-PCR); the results showed that they are expressed in the cambium, phloem and mature tissues, but rarely expressed in young leaves and flowers. Moreover, they are also widely involved in the drought stress response. At last, we selected and cloned PtrFBA60 for physiological function analysis and found that it played an important role in coping with drought stress. Taken together, the family analysis of FBA genes in P. trichocarpa provides a new opportunity for the identification of P. trichocarpa candidate FBA genes and elucidation of their functions in growth, development and stress response, thus demonstrating their utility in the improvement of P. trichocarpa.

Genome-Wide Analysis of the FBA Subfamily of the Poplar F-Box Gene Family and Its Role under Drought Stress F-box proteins are important components of eukaryotic SCF E3 ubiquitin ligase complexes, which specifically determine protein substrate proteasomal degradation during plant growth and development, as well as biotic and abiotic stress. It has been found that the FBA (F-box associated) protein family is one of the largest subgroups of the widely prevalent F-box family and plays significant roles in plant development and stress response. However, the FBA gene family in poplar has not been systematically studied to date. In this study, a total of 337 F-box candidate genes were discovered based on the fourth-generation genome resequencing of P. trichocarpa. The domain analysis and classification of candidate genes revealed that 74 of these candidate genes belong to the FBA protein family. The poplar F-box genes have undergone multiple gene replication events, particularly in the FBA subfamily, and their evolution can be attributed to genome-wide duplication (WGD) and tandem duplication (TD). In addition, we investigated the P. trichocarpa FBA subfamily using the PlantGenIE database and quantitative real-time PCR (qRT-PCR); the results showed that they are expressed in the cambium, phloem and mature tissues, but rarely expressed in young leaves and flowers. Moreover, they are also widely involved in the drought stress response. At last, we selected and cloned PtrFBA60 for physiological function analysis and found that it played an important role in coping with drought stress. Taken together, the family analysis of FBA genes in P. trichocarpa provides a new opportunity for the identification of P. trichocarpa candidate FBA genes and elucidation of their functions in growth, development and stress response, thus demonstrating their utility in the improvement of P. trichocarpa. In eukaryotes, the ubiquitin/26S proteasome system (UPS) is responsible for the selective degradation of most intracellular proteins [1]. There are three major enzymes involved in ubiquitin degradation, including the ubiquitin-activating enzyme (E1), ubiquitin-binding enzyme (E2), and ubiquitin-ligase (E3). E3 are divided into the following four categories based on their mechanism of action and subunit positions: the HETC structural domain type, RING-finger structural domain type, U-box structural domain type, and the SCF (Skp1-Cullin -F-box) domain type [2]. F-box proteins play a key role in recruiting substrates to UPS [3,4]. F-box proteins carry at least one 40–50 residue F-box/F-box-like domain at its N-terminus, which is responsible for binding to Skp1/Skp1-like proteins. At the same time, one or more conserved domains can also be found at the C-terminus, which are associated with substrate-specific recognition, including F-box-associated (FBA), kelch repeat and leucine-rich repeat (LRR) domains, etc. [5]. To date, F-box proteins have been discovered in almost all plants, and these F-box members form a larger protein family, of which the FBA subfamily is one of the largest subgroups [6]. According to previous studies, researchers found 94 F-box proteins containing FBA domains in Triticum aestivum [7], 46 in Gossypium hirsutum [8], 25 in Cicer arietinum [9], 14 in Dioscorea esculenta [10], 34 in Juglans regia [11], 278 in Brassica rapa [12], 17 in Zea mays [5], 196 in Arabidopsis thaliana [6], and 133 in Malus domestica [13]. A large number of F-box genes in plants are involved in the regulation of many biological processes, including hormone responses, lateral root formation, meristem formation, photomorphogenesis, senescence and stress response (abiotic and biotic processes) [9,14]. For example, the F-box protein TIR1 is an auxin receptor in Arabidopsis thaliana. Auxin-induced interaction between the Aux/IAA transcriptional repressor proteins and the ubiquitin–ligase complex SCFTIR1 mediates Aux/IAA degradation and auxin-regulated transcription [15]. The Arabidopsis F-box protein UFO is more likely to be a co-regulator that functions together with LFY in controlling organ-identity genes [16]. EID1 possibly operates by targeting activated components of the phyA signaling pathway to ubiquitin-dependent proteolysis [17]. TaFBA1-OE plants improve drought tolerance by increasing antioxidant competence and decreasing ROS accumulation [18,19]. The information on plant F-box genes was mainly obtained from the research on Arabidopsis, rice and wheat, while the research on poplar F-box genes (PtrFBXs) is quite limited, and the research on F-box/FBA proteins is even more sparse. Hua et al. conducted a phylogenetic comparison of F-box proteins in 18 plants including poplar in 2011, and the results revealed the expansion, evolutionary selection, and functional relatedness of the F-box gene family. The results also indicate that the diversification of F-box genes may be achieved through genome drift [4]. In 2008, Yang et al. performed a comparative genomics analysis of the woody perennial plant poplar with the herbaceous annual plants Arabidopsis thaliana and rice, and elucidated the functional significance of this gene family [6]. Nonetheless, genome-wide analysis of PtrFBXs in poplar needs to be further studied. Drought severely restricts plant growth and crop yields, and poplars are ideal materials for studying the response of trees to drought stress [20]. Some studies have been carried out on stress resistance genes in poplars. Previous studies have shown that plants have some intrinsic mechanisms to alleviate drought stress, including closing stomata, reducing transpiration, producing abscisic acid (ABA) and increasing surface wax. For example, PeCHYR1 enhanced drought tolerance by promoting H2O2-mediated stomatal closure in poplar [21]. The overexpression of PePYL4 led to higher water-use efficiency (WUE) and drought tolerance in poplar [22]. PeABF3 improves drought tolerance by promoting ABA-induced stomatal closure by regulating the expression of PeADF5 under drought conditions [23]. PeSHN1 regulates both WUE and drought tolerance in poplar by modulating wax biosynthesis [24]. In recent years, whole genome sequencing of P. trichocarpa has been carried out for the fourth generation of resequencing, providing a valuable reference resource for the analysis of whole gene families, such as the PPO polyphenol oxidase gene family and the gibberellin-stimulated Arabidopsis (GASA) protein under abiotic stress [25,26]. However, the F-box family and its subfamily features have not been reported in poplar. In this study, we aimed to identify and characterize the functions of PtrFBAs from the P. trichocarpa genome. We searched the P. trichocarpa genome using the HMMER website and identified 74 PtrFBA candidate genes. We predicted and analyzed the chromosomal localization, gene amplification, gene structure and upstream promoter cis-acting elements of the candidate PtrFBA genes. The chemical properties, subcellular localization, motifs and phylogenetic relationships of the proteins they encode were also predicted and analyzed. Focusing on identifying PtrFBAs associated with drought stress, the subcellular localization of the representative gene PtrFBA60 and its germination rate and physiological phenotype under drought stress were analyzed. In this study, the PtrFBA gene was comprehensively analyzed by genomics, transcriptomics, and qRT-PCR for the first time, and the representative gene PtrFBA60 was screened. We performed subcellular localization of PtrFBA60 in tobacco and drought stress experiments in Arabidopsis thaliana, and analyzed the seed germination rate and physiological phenotype under drought stress to further study PtrFBA’s family properties and its possible role in drought tolerance in poplar. We searched the genome-wide protein database for the conserved F-box (PF00646) and FBA1(F-box associated 1) (PF07734) domains. The motif maps (Supplementary Figure S1) and hmm model files of the two domains were downloaded; 459 sequences of the whole genome of P. trichocarpa were screened. Redundant sequences without F-box structural domains were then manually removed, resulting in a total of 337 F-box genes. These genes were named according to the order of corresponding chromosomal positions identified in the NCBI database. Different alternative splice forms from the same locus were assigned under the same gene names, such as PtrFBA16 and PtrFBA16.1 (Supplementary Table S1). The domain analysis showed that 101 out of 337 F-box genes contained only F-box domains (accounting for about 30%), 74 genes contained FBA domains (accounting for about 21%) and the other genes showed FBD, PP2, WD40, Kelch-type, LRR, FIST, NBD, DUF, AMN, TUB, RBD, PLN, PP1, or other domains (Figure 1). In different plants, the types and numbers of C-terminus domains are different, which indicates that the evolutionary history of different domains at the C-terminus may be more complex. Previous studies have found that the number of clans in the F-box family and the FBA subfamily is different in different plants. We performed an evolutionary developmental analysis of the FBA subgroup in P. trichocarpa and divided it into four subgroups (Figure 2a). We next compared the number of F-box family and FBA subfamily genes in ten species; the results showed that the gene number of FBA subfamilies and F-box family varied between species. The highest number of FBA genes was found in Brassica rapa with 278 genes, and the number of F-box genes in wheat reached 1013 (Figure 2b). F-box motifs and FBA domains usually contain about 50 amino acids, and several residues are conserved. We screened PtrFBA60 as a representative gene from poplar’s FBA family. Malus domestica FBA protein MdFBX3, Antirrhineum hispanicum FBA protein AhSLF4, Zea mays FBA protein FBX230.1, Arabidopsis thaliana FBA protein AtFOA2 and Triticum aestivum FBA protein TaFBA1 for sequence alignment by using CLUSTALX (Supplementary Figure S2). The results showed that they were conserved in F-box and FBA motifs. We next analyzed the physicochemical properties of the 74 PtrFBAs genes in the en-tire PtrFBAs gene family. Except for PtrFBA34, which encodes about 191 amino acids, the PtrFBAs genes in P. trichocarpa generally encode 277–466 amino acids, with an average of 396 amino acids. The molecular weights of PtrFBAs proteins were both larger than 30 KDa, except for PtrFBA34. The theoretical pI of PtrFBAs was 4.74–9.51; 31 genes belonged to acidic proteins and 43 belonged to basic proteins. In addition, 30% of these FBA proteins were stable with the instability index <40. The overall hydrophilicity of the proteins is relatively strong (Supplementary Table S2). According to the prediction of the Plant-mPLoc database (http://www.csbio.sjtu.edu.cn/bioinf/plant-multi/, accessed on 15 July 2021), most of the PtrFBAs proteins are located in the nucleus, and a few in the membrane system. The 74 PtrFBA genes were unevenly distributed on poplar chromosomes 1–19, with the largest distribution on chromosome 1 and none on chromosome 9 (Figure 3). PtrFBA repeat gene clusters were found on chromosomes 1, 3, 5, 6, 8, 10, 11, 12, 13, 15, 16, and 17 of P. trichocarpa. The largest repeat gene cluster had five genes, and is located on chromosome 6. To illustrate the expansion pattern of PtrFBA genes, we analyzed duplication events in the P. trichocarpa genome. PtrFBAs have undergone multiple repetitive events during evolution. PtrFBA2 and PtrFBA15, PtrFBA7 and PtrFBA68, PtrFBA12 and PtrFBA24, PtrFBA12 and PtrFBA21, PtrFBA11 and PtrFBA60, PtrFBA20 and PtrFBA29, PtrFBA26 and PtrFBA41, PtrFBA27 and PtrFBA42, PtrFBA25 and PtrFBA65, PtrFBX118 and PtrFBA71 all form an entire genome duplication (WGD) pair, indicating that they all share a common ancestor with each other (Supplementary Table S3). The Ka/Ks values of PtrFBA7, PtrFBA68, PtrFBA65 and PtrFBA25 were all greater than 1, indicating that these genes had positive selection effects (Supplementary Table S4). The Ka/Ks values of the remaining three groups of genes are all less than 1, indicating that these genes have the effect of purification selection. In poplar, we found 19 genes that formed 10 segmental repeat events, which were distributed on 11 chromosomes, with two additional tandem repeat regions on chromosomes 6 and 10 (Figure 4). Next, we constructed comparative syntenic maps of the F-box gene family of P. trichocarpa and four typical plants (Arabidopsis, wheat, maize and apple) (Supplementary Figure S3). The syntenic maps showed that there were 129 pairs of FBX homologous genes between P. trichocarpa and Arabidopsis thaliana, 48 pairs of FBX homologous genes with maize, 124 pairs of FBX homologous genes with wheat, and 172 pairs of FBX homologous genes with apple. We then constructed comparative syntenic maps of the FBA gene families of P. trichocarpa and four species (Figure 5); it was found that P. trichocarpa has 8 pairs of FBA homologous genes with Arabidopsis, 1 pair of FBA homologous genes with maize, no FBA homologous genes with wheat, but 26 pairs of FBA homologous genes with apple. In addition, it was found that some PtrFBA genes are related to at least two synonymous gene pairs, especially FBA genes between poplar and apple, indicating that these genes may play an important role in the evolution of plants. The results of this study show that PtrFBAs have fewer homologs in wheat, Arabidopsis and maize and more homologs in the woody plant apple. This suggests that multiple FBA homologs may have been formed in woody plants during the differentiation process. To more clearly explore the evolutionary relationship between different members of the PtrFBA gene family, their gene structures, conserved protein motifs, and conserved structures were evaluated (Figure 6b). Most PtrFBA genes (about three fifths of the genes) have no introns, six genes have two introns (PtrFBA19/24/26/27/37/53) and twenty-one genes have one intron. Moreover, 15 genes did not contain UTR, and 8 genes contained neither introns nor UTRs (PtrFBA5/16.1/23/43/46/47/56/73). At the same time, 10 conserved motifs were found in the PtrFBA protein sequence using MEME (Figure 6a), of which 2 motifs (1 and 8) were associated with the N-terminal F-box domain and 6 motifs (2, 5, 6, 9, 4, and 10) were associated with the C-terminal FBA domain (Supplementary Table S5). To determine the expression pattern of PtrFBAs, the sequence upstream of the PtrFBA family promoter (~2000 bp) was extracted from the P. trichocarpa genomic DNA sequence. The cis-elements of the PtrFBA promoter were analyzed using the PlantCARE database (Figure 7). We visualized the promoter elements in promoter positions and species (Supplementary Figure S4). The specific functions of these elements (cis-elements) are annotated (Supplementary Table S6). These elements are involved in abiotic and biotic stresses, phytohormonal responses, and plant growth and development. Elements of the PtrFBA promoter are mainly involved in abiotic and biotic stresses, as well as phytohormone response factors. Stress-related cis-elements (Myb, Myc, ARE and STRE) were extracted, suggesting that PtrFBAs may play a key role in coping with adverse conditions. In addition, some PtrFBA promoters are enriched in ABRE (involved in the ABA response), such as PtrFBA42/PtrFBA61/PtrFBA/65, and these genes may be responsive to ABA hormones. PtrFBA22/PtrFBA37 have more LTR elements (involved in low-temperature stress responses) in their promoters, suggesting that these genes may be responsive to low-temperature induction. To further investigate the role of PtrFBA genes in growth and development, we examined the tissue expression profiles of 30 PtrFBA genes from transcriptome data (Supplemental Table S7). We created a heat map of the samples and genes clustered in both directions to explore the expression characteristics of PtrFBA genes in 15 poplar tissues (Figure 8a). Most of the genes were highly expressed in mature sites, while a few were highly expressed in young sites (e.g., PtrFBA1/4/6). Some genes were highly expressed in phloem (e.g., PtrFBA49/50/55/72), which may be related to the growth and development of poplar. We downloaded transcriptomic datasets to visualize the expression of 30 PtrFBAs genes after drought, beetle and mechanical damage (Supplemental Table S8) and performed transcriptomic data analysis and visualization to analyze the effects of PtrFBA genes on stress response in poplar (Figure 8b). Most of the genes in the PtrFBA gene family are sensitive to drought stress and their expression is up-regulated under drought stress. Some genes were up-regulated under leaves beetle damage stress (e.g., PtrFBA11/26/65), and some were up-regulated under leaves mechanical damage stress (e.g., PtrFBA26/40/42). PtrFBA26, PtrFBA42 and PtrFBA65 were up-regulated under both types of stress, indicating that these genes may respond to leaf injury stress. To further explore the function of PtrFBAs, we investigated the response of some PtrFBAs to drought stress by qRT-PCR in P. trichocarpa leaves (Figure 9). Three different gradients were set according to soil moisture content. PtrFBA15/45/60/72 were up-regulated under drought stress, while PtrFBA6/11/13/36/50 were down-regulated, and PtrFBA61 showed no significant change. We also found that PtrFBA60 was significantly sensitive to ABA (Figure S5a), and weakly sensitive to ethylene and NaCl (Figure S5b,c). We selected the PtrFBA60 gene for follow-up experiments. To determine the subcellular localization of PtrFBA60, the Super:: PtrFBA60-GFP (green fluorescent protein) fusion protein was transiently transfected into tobacco leaves. It was found to be likely located in the cytoplasm; however, it was not uniformly distributed in the cytosol, but rather formed punctate aggregates (Figure 10a). The PtrFBA60 aggregates were found to correspond to P-bodies by co-localization experiments between PtrFBA60-GFP and the P-body markers AtTZF1-mCherry and AtAGO-mCherry (Figure 10b). The results indicate that PtrFBA60 is localized in the P-bodies [27,28]. Since drought stress can induce the expression of PtrFBA60 (Figure 9), we hypothesized that PtrFBA60 plays a vital role in the drought response. The seed germination rate experiments showed that there was no significant difference in the germination rate of the OE-PtrFBA60, WT, and mutants fba60 on 1/2 MS medium, but the germination rate of OE-PtrFBA60 was significantly higher for the 200 mM mannitol cultured medium (Figure S6). The OE-PtrFBA60, mutant fba60, and WT Arabidopsis were cultured under the same conditions and then subjected to drought treatment. After 20 days, the leaves of mutant fba60 were severely wilted and the leaves of WT turned yellow and wilted, while those of OE-PtrFBA60 lines remained green and turgid. Furthermore, OE-PtrFBA60 trangenic plants recovered normally after being rewatered, while the mutant and WT Arabidopsis plants did not fully recover (Figure 11a). The survival rate of OE-PtrFBA60 was 85.8%, while the survival rate of WT and mutant fba60 was 51.6% and 36.6%, respectively (Figure 11b). The REC (relative electrical conductance) characterizes the degree of damage to plant cell membranes and the REC of OE-PtrFBA60, mutant fba60, and WT both increased after drought treatment, but the values for fba60 and WT were three times larger than those for OE-PtrFBA60 (Figure 11c). These results indicated that OE-PtrFBA60 transgenic Arabidopsis had better drought stress tolerance. As a component of the SCF E3 ligase complex, the F-box protein family plays an im-portant role in plant growth, development and stress resistance [29,30,31,32,33,34]. The FBA subfamily is a ubiquitous and abundant subfamily in the F-box family, consisting of an N-terminal F-box conserved domain and a C-terminal FBA conserved domain [18,19]. The FBA gene family is related to many plant physiological activities, such as plant growth and development, hormone signal transduction, abiotic and biotic stresses and plays a significant role [3,15,17,35,36,37]. In many plants (wheat, maize, Arabidopsis, etc.), the biological function and bioinformatic analysis of the FBA gene family has been widely reported, but this is not the case for the woody plant poplar [6]. In this study, 74 candidate FBA genes were identified in P. trichocarpa, and bioinformatics and expression patterns of these genes were analyzed. As is the case with maize, apple and Arabidopsis, the FBA genes in P. trichocarpa are unevenly distributed, and distributed on several chromosomes (Figure 3). These 74 proteins are all high-molecular-weight proteins, except PtrFBA34. The experimental results show that in agreement with other species, the predicted subcellular localization varies, but most are located in the nucleus. However, the subcellular localization of our target gene, PtrFBA60, showed that PtrFBA60 was localized in the P-body (Figure 10) [27,28]. Due to the gap between their occurrence positions and distances, the FBA gene families were discretely distributed in the phylogenetic tree, and were mainly divided into four clusters (Figure 2). The origin and evolution of the PtrFBAs genes were also further determined. The increase in gene family members and the mechanism of genome evolution mainly depend on gene duplication events, including tandem duplication and segmental duplication [38]. In the present study, 74 PtrFBAs genes were unevenly distributed on 19 poplar chromosomes, and this phenomenon of uneven distribution of chromosomes suggested that this change occurred before species differentiation (Figure 3). A total of 10 pairs of whole genome duplicate genes and 12 tandem duplicate genomes were detected in poplar. This indicates that both tandem repeat sequences and whole genome repeats are involved in the evolution of poplar PtrFBAs genes (Figure 4). Previous gene family studies have suggested that tandemly repeated genes may have similar functions and expression patterns. Most PtrFBAs have no introns, while some have up to two introns (Figure 6). In order to quickly respond to stress, organisms need to stimulate the expression of genes, and gene structures with few or no introns contribute to the rapid expression of mRNA [39,40]. Genes such as PtrFBA3/4/7 lack introns but contain UTRs, so they can be transcribed faster to form mRNA. A large number of genes for FBAs were present in each species (Figure 2b). Covariance analysis with four different species revealed that the F-box gene family and FBA gene family of poplar are more closely related and homologous to the woody plant apple than other species (Supplementary Figure S3, Figure 5) [11,13,31,41,42,43,44,45]. This suggests that considerable FBA divergence and doubling may have occurred during the evolution of perennial woody plants. The expression patterns of FBA genes in various tissues have been confirmed in many species [10]. Due to differences in the number of FBA genes in different species, there is no uniform FBA gene expression profile in plants. According to the RNA-seq data of poplar, some genes were preferentially expressed in the phloem and shoots (Figure 8a), indicating that these genes are essential for plant growth [3]. Some genes are highly expressed throughout the growth and development process, indicating that the FBA gene family plays an important role in the entire plant growth and development process [14,18,32,33,35]. The expression of the FBA gene family in plants is closely related to stress and adversity, mainly in response to drought, insects and mechanical damage [5,18]. Transcriptome data showed that drought significantly induced some FBA genes in poplar (PtrFBA2/15/60) (Figure 8b), which was verified by quantitative reverse transcription polymerase chain reaction (qRT-PCR) (Figure 9). By analyzing the promoters of the FBA gene family, we found that there are many different elements in their promoters, especially ABRE and ARE, which are involved in biotic and abiotic stresses (Figure 7) [14,46,47,48]. Furthermore, ABRE elements play an important role in response to abiotic stress, which ensures that FBA genes can be rapidly induced under stress [49]. In order to verify whether the FBA gene can provide plants with the function of drought resistance, we overexpressed the PtrFBA60 gene in the model plant Arabidopsis thaliana and purchased Arabidopsis thaliana seeds with mutant fba60. It was found that overexpression of PtrFBA60 significantly improved the drought resistance of Arabidopsis (Figure 11), and the expression of PtrFBA60 also significantly increased under ABA treatment, which may improve plant drought resistance through the ABA pathway (Supplementary Figure S5) [19,46,50,51,52]. In summary, plants can adapt to various complex environments by regulating their expression level of FBA genes. The identification and analysis of the FBA gene family were based on the whole genome of P. trichocarpa. Relevant poplar genomic and protein data were downloaded from the Phytozome13.0 database (https://phytozome-next.jgi.doe.gov/, accessed on 12 July 2022). We downloaded the Hidden Markov model of the FBA1 structural domain (PF07734) and F-box structural domain (PF00646) from the Pfam database (http://pfam.xfam.org/, accessed on 30 January 2023), and searched the genome of P. trichocarpa using the Simple HMM Search program in TBtools software (TBtools_windows-x64_1_098685) and predicted the IDs of the PtrFBAs gene family. These sequences were further confirmed by the Protein BLAST function of the NCBI database (https://blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 30 January 2023) [53,54]. The P. trichocarpa 4.1 genome, CDS, transcripts, polypeptides and 2000 bp upstream of the promoter region of the translation initiation site (ATG) were extracted from the Phytozome 13.0 database. The amino acid sequences of PtrFBX and PtrFBA target sequences were analyzed by ClustalX. In addition, using the neighbor-joining (NJ) method parameter on MEGA7.0 with pairwise deletion and 1000 bootstrap replicates, multiple alignments were applied to construct a phylogenetic tree among PtrFBAs [54,55]. PtrFBAs and PtrFBXs are named according to their positions on the scaffold and chromosomes, respectively. The amino acid sequences of each species were obtained from the NCBI database, the ClustalW (https://www.genome.jp/tools-bin/clustalw, accessed on 30 January 2023) website was used, the Clustal algorithm was used for sequence alignment, and the ENDscript/ESPript (https://espript.ibcp.fr/ESPript/cgi-bin/ESPript.cgi, accessed on 30 January 2023) website was opened to map the sequence alignment results. The physicochemical properties of the PtrFBAs family were analyzed using the TBtools Protein Paramter Calc function. Subcellular localization was predicted by the site Plant-mPLoc (http://www.csbio.sjtu.edu.cn/bioinf/plant-multi/, accessed on 30 January 2023) [56,57] and the predicted subcellular localization of the PtrFBAs family was obtained by entering their amino acid sequences into the website. Gene positions and chromosome sizes of PtrFBAs and PtrFBXs were obtained from the NCBI database and visualized by TBtools (South China Agricultural University, Guangzhou, China) [58]. The PtrFBX and PtrFBA duplication events in the poplar genome were evaluated using multiple collinear scanning toolkits (MCScanX) [59]. The poplar genome and gff3 files were applied to investigate the putative duplication events using TBtools with one-step MCScanX with default parameters. In addition, the ratio of non-synonymous (Ka) and synonymous (Ks) substitutions in PtrFBA gene pairs were measured to evaluate the selection pressure of the PtrFBAs evolutionary process. To further evaluate the collinearity relationship among poplar, Arabidopsis, apple, maize and wheat, the MCScanX with default parameters was also applied to identify the putative orthologs. In a similar manner to intra-poplar collinearity, the genome and gff3 files of poplar, Arabidopsis, apple, maize and wheat were aligned to achieve the three important files, including the control file (ctl), gff and collinearity formats. We manually removed unnamed chloroplasts and mitochondrial chromosomes from the ctl files and reordered them, and finally visualized them with TBtools. We uploaded gff3 annotation files that contained gene information to the TBtools (https://github.com/CJ-Chen/TBtools, accessed on 30 January 2023) software (TBtools_windows-x64_1_098685)to reveal gene conserved motifs and gene structures. In addition, the PtrFBA protein sequences were submitted to Motif’s Multiple Expectation Maximization (MEME) web version (https://meme-suite.org/meme/tools/meme, accessed on 30 January 2023) [60]. In addition, gff3 files and patterns stored in the Extensible Markup Language (XML) format were uploaded to the TBtools software for displaying conserved motifs and gene structures. The cis-acting elements of promoters were predicted by PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/, accessed on 30 January 2023), and classified and analyzed according to the literature, and the results were visualized by EXCEL and TBtools. To analyze the expression profile of the PtrFBA family, we searched and downloaded the transcriptome data of 30 PtrFBAs genes from the PopGenIE (https://plantgenie.org/, accessed on 30 January 2023) public database [61,62]. In this database, expression data were collected for three abiotic stresses and one biotic stress, as well as expression data for 15 different plant tissues. Heat maps of the relative expression quantification of PtrFBAs were created using TBtools, and Row Scale was chosen for homogenization, while all the other parameters were set as default parameters. It was modified to: Annual poplar seedlings were planted in the nursery of Beijing Forestry University, with two poplar seedlings planted in each pot. P. trichocarpa seedlings were grown for about 3 to 4 months under the control of a greenhouse environment. Under different abiotic stresses, 5 potted seedlings with the same growth conditions were selected for each abiotic stress treatment. All collected samples were immediately frozen in liquid nitrogen and stored in a −80 °C freezer. After returning to the laboratory, RNA extraction (kit from Aidlab biotechnologies CO.LTD., Beijing, China) and reverse transcription (kit from TIANGEN BIOTECH CO.LTD.) were performed. Arabidopsis ecotype Colombia (Col-0) was used in this study. The fba60 (SALK_019628.33.80.x) mutant was obtained from the Arabidopsis biological resource center ABRC (ABRC (osu.edu)). fba60 is a T-DNA insertional mutant. Plants were grown in soil or 1/2 MS medium and cultured in a constant temperature incubator (16 h light/8 h dark cycle, 22 °C). qRT-PCR analysis-specific primers were designed and tested by NCBI Primer BLAST (https://www.ncbi.nlm.nih.gov/tools/primerblast/index.cgi?LINK_LOC=BlastHome, accessed on 30 January 2023). Real-time PCR was run with the CFX96 Touch™ instrument (Bio-Rad, Hercules, CA, USA) to detect the chemical SYBR Green. The reaction system established was as follows: 10 µL of 2× SuperReal Premix Plus (TIANGEN), 0.5 µL of each forward and reverse primer, 1 µL of diluted cDNA template (100 ng/µL), and finally, RNase-free ddH2O was added until the total was 20 µL. The reaction procedures were as follows: 95 °C for 15 min, 45 cycles of 10 s at 95 °C and 30 s at 60 °C. We used two internal reference genes to conduct the experiment. The selection of internal reference genes and analysis of the experimental results were based on the research of Wang et al. [63,64]. SPSS (IBM, Armonk, NY, USA) was used for statistical analysis, and the LSD test was used to calculate the p value. All the primers used are listed in Supplemental Table S9. To determine the subcellular localization of PtrFBA60 in plant cells, the full-length CDS of PtrFBA60 was cloned into the pSuper-1300 vector, generating a PtrFBA60-GFP fusion protein, and then introduced into A. tumefaciens GV3101. A plasmid that contained GFP alone served as a control. Similarly, the full-length CDS of the P-body markers gene AtAGO and AtTZF1 was cloned into the pSuper-1300 vector to generate AtAGO-mCherry and AtTZF1-mCherry fusion protein, and then introduced into A. tumefaciens GV3101. Transient expression assay in tobacco leaf epidermal cells was performed as previously described [65]. The GFP or mCherry signal was detected by laser confocal microscopy SP8 after two days of infestation (Leica Microsystems TCS SP8, Wetzlar, Germany). The PtrFBA60 gene was cloned using P.trichocarpa as a template, and PtrFBA60 was inserted into the pBI121-GUS vector with XbaI and XmaI restriction sites. After fusion with the GUS in the pBI121 vector, the OE-PtrFBA60 vector was obtained and transformed into GV3101 Agrobacterium. The inflorescence method was used to transform Arabidopsis thaliana. We chose the activated monoclonal positive Agrobacterium containing the target gene and placed it into 10 mL fresh YEB medium (containing 100 mg/L Kan, 50 mg/L Rif), shaken overnight in a constant temperature incubator at 28 °C and 200 rpm cultivated for 12 h. The overnight cultured bacterial solution was added to 100 mL of YEB culture solution (containing 100 mg/L Kan, 50 mg/L Rif), shaken again to achieve the OD600 = 0.8 and centrifuged at 5000× g rpm for 10 min. The supernatant was discarded and we used the resuspension solution (containing 2.18 g 1/2 MS, 0.3–0.4% foaming agent, 50 g sucrose; 0.5 g MES per liter), resuspending the agrobacterium until the OD600 was 0.8. Arabidopsis thaliana inflorescence was invaded in the infestation solution for 2–3 min.After completion, the solution was incubated in the dark for 24 h. The second infection experiment was performed after 3–4 days. For the histochemical GUS analysis, the prepared material was immersed in GUS staining solution and incubated overnight at 37 °C. Wild-type and transgenic Arabidopsis seeds were sown simultaneously on 1/2 MS solid medium at a mannitol concentration of 0 or 250 mmol/L. Fifty-six seeds of each line were sown equidistantly on each medium, and three technical replicates were made. After 2 days of dark culture, the cells were placed in a constant temperature incubator at 22 °C for 16 h in the light, 8 h in the dark, and the germination rate was recorded every 12 h. Wild-type and transgenic Arabidopsis seeds were also sown simultaneously in soil and grown in a climate-controlled chamber with a light intensity of 180 μ mol/m2/s, a light duration of 14 h/dark duration of 10 h, and a temperature of 25 °C. Arabidopsis were subjected to natural drought treatment for 20 days after one month of growth. Wild-type and mutant Arabidopsis leaves were photographed after water loss and yellowing, and finally treated with rehydration for three days and counted for viability and REC. Statistical analyses were conducted using Microsoft Excel 2016 (Microsoft Corporation, Redmond, WA, USA) and SPSS (version 25, IBM Corporation, Armonk, NY, USA). One-way ANOVA with the LSD multiple comparisons test (* p < 0.05, ** p < 0.01, *** p < 0.001.) was performed for the gene relative expression. Before applying the ANOVA test, the data were tested for normality and homogeneity of variance. Student‘s t test (* p < 0.05; ** p < 0.01) was performed for the survival rates and REC. We believe that both tandem and WGD repeats were involved in the evolution of PtrFBAs genes in poplar, and that perennial woody plants may have undergone considerable FBA gene differentiation and doubling during evolution. Moreover, PtrFBA genes play a significant role in the process of plant growth and development, and also play a regulatory role in adversity stress. Among them, PtrFBA60 plays a positive regulatory role in plant drought resistance, which may allow plants to acquire drought resistance through the ABA pathway. Future experiments will aim to investigate the nature of these results from a genetic, biochemical and physiological perspective.

PMC10002532

Gabriela Ortiz-Soto,Natalia S. Babilonia-Díaz,Mercedes Y. Lacourt-Ventura,Delmarie M. Rivera-Rodríguez,Jailenne I. Quiñones-Rodríguez,Mónica Colón-Vargas,Israel Almodóvar-Rivera,Luis E. Ferrer-Torres,Ivette J. Suárez-Arroyo,Michelle M. Martínez-Montemayor

Metadherin Regulates Inflammatory Breast Cancer Invasion and Metastasis

28-02-2023

metadherin,inflammatory breast cancer,invasion,metastasis

Inflammatory breast cancer (IBC) is one of the most lethal subtypes of breast cancer (BC), accounting for approximately 1–5% of all cases of BC. Challenges in IBC include accurate and early diagnosis and the development of effective targeted therapies. Our previous studies identified the overexpression of metadherin (MTDH) in the plasma membrane of IBC cells, further confirmed in patient tissues. MTDH has been found to play a role in signaling pathways related to cancer. However, its mechanism of action in the progression of IBC remains unknown. To evaluate the function of MTDH, SUM-149 and SUM-190 IBC cells were edited with CRISPR/Cas9 vectors for in vitro characterization studies and used in mouse IBC xenografts. Our results demonstrate that the absence of MTDH significantly reduces IBC cell migration, proliferation, tumor spheroid formation, and the expression of NF-κB and STAT3 signaling molecules, which are crucial oncogenic pathways in IBC. Furthermore, IBC xenografts showed significant differences in tumor growth patterns, and lung tissue revealed epithelial-like cells in 43% of wild-type (WT) compared to 29% of CRISPR xenografts. Our study emphasizes the role of MTDH as a potential therapeutic target for the progression of IBC.

Metadherin Regulates Inflammatory Breast Cancer Invasion and Metastasis Inflammatory breast cancer (IBC) is one of the most lethal subtypes of breast cancer (BC), accounting for approximately 1–5% of all cases of BC. Challenges in IBC include accurate and early diagnosis and the development of effective targeted therapies. Our previous studies identified the overexpression of metadherin (MTDH) in the plasma membrane of IBC cells, further confirmed in patient tissues. MTDH has been found to play a role in signaling pathways related to cancer. However, its mechanism of action in the progression of IBC remains unknown. To evaluate the function of MTDH, SUM-149 and SUM-190 IBC cells were edited with CRISPR/Cas9 vectors for in vitro characterization studies and used in mouse IBC xenografts. Our results demonstrate that the absence of MTDH significantly reduces IBC cell migration, proliferation, tumor spheroid formation, and the expression of NF-κB and STAT3 signaling molecules, which are crucial oncogenic pathways in IBC. Furthermore, IBC xenografts showed significant differences in tumor growth patterns, and lung tissue revealed epithelial-like cells in 43% of wild-type (WT) compared to 29% of CRISPR xenografts. Our study emphasizes the role of MTDH as a potential therapeutic target for the progression of IBC. Inflammatory breast cancer (IBC) is a very aggressive and rare subtype of breast cancer (BC) with a poor prognosis and accounts for 1–5% of all cases of BC. The 5-year survival rate for people with IBC is 41% compared to non-IBC (nIBC) [1,2,3]. However, a recent study in Puerto Rico reports that the five-year survival rate for IBC patients is 0%, while the three-year survival is 39%, highlighting the urgency of IBC studies [4]. The lethality of IBC is due to the efficient ability of cancer cells to invade the vascular and lymphatic systems and often the absence of a distinct tumor mass [1]. The uncommon nature of this BC subtype, the rare clinical presentation, and the overlap of these atypical signs and symptoms with other diseases complicate its diagnosis and treatment [5]. Although the introduction of multimodal therapy has improved overall survival, it remains poor compared to nIBC [6]. Therefore, studies directed toward finding targets and understanding IBC biology are essential to improve therapeutic strategies and overall patient outcome. Previously, our group characterized the cell surface proteome of IBC. We identified metadherin (MTDH) overexpression in IBC cell lines vs. non-cancerous mammary epithelial cells, and nIBC cell lines [7]. Subsequently, the presence of MTDH was validated in IBC tissues and within tumor emboli [7], a hallmark of IBC metastasis [1]. MTDH is a single-pass transmembrane protein expressed mainly in the endoplasmic reticulum and the perinuclear space and has been found to colocalize with tight junctions [8,9,10,11,12,13]. MTDH is a cell adhesion molecule and functions as a scaffold protein that interacts with multiple cell networks [13,14,15,16]. The localization of MTDH varies depending on the state of the cell; it is found in the nucleus in non-malignant tissue and in the cytoplasm and cell membrane in malignant tissue, where it interacts with various oncogenic pathways. Some of the pathways associated with MTDH interactions are NF-κB, PI3K/AKT, MAPK, and Wnt/β-catenin [17,18,19,20,21,22,23]. Dysregulation of these pathways involves cell regulatory mechanisms, such as proliferation, migration, and cell survival; which promotes chemoresistance, invasion, and metastasis [8]. Metastasis is the leading cause of death in BC patients [24]. IBC patients have a higher risk of distant metastasis and locoregional recurrence than nIBC patients [25]. More than 30% of IBC patients have metastases at the time of diagnosis [26,27]. Understanding the metastatic processes that influence the aggressiveness and lethality of IBC is crucial to support the multidisciplinary discussion of potential approaches to diagnose and treat IBC. Furthermore, various groups confirm that MTDH plays an important role in BC metastasis [15,28]. The pro-metastasis function of MTDH is due to the interaction of a lung-homing domain of MTDH with an unknown receptor in endothelial cells [15,29] and has been shown to increase expression of adhesion molecules by activating NF-κB [21]. However, the role of MTDH in the invasion and metastasis of IBC remains unknown. This study investigates the role of MTDH in IBC by assessing its function in cell proliferation, migration, tumor spheroid formation, signaling of oncogenic pathways, tumor progression, and metastasis. We develop isogenic models of the SUM-149 and SUM-190 IBC cell lines using MTDH-CRISPR/Cas9 base knockout for in vitro and in vivo studies. To validate the oncogenic role of MTDH, we also overexpressed MTDH in the non-cancerous mammary epithelial cell line MCF-10A. Our findings show that the absence of MTDH significantly decreases IBC cell proliferation, migration, invasion, and tumor spheroid formation and decreases the expression of various signaling molecules. In addition, we demonstrate that MTDH affects invasion and metastasis. To our knowledge, this study is the first to assess the role of MTDH in the invasion and metastasis of IBC. Thus, this study provides significant new evidence for the field of IBC, contributing to studying possible new targets involved in IBC progression. In a previous study, we demonstrated that MTDH is overexpressed in IBC cells and patient tumor tissues compared to normal tissues and in the tumor emboli. Furthermore, we reported that MTDH overexpression was found in the plasma membrane of IBC cells, specifically in SUM-149 cells, compared to the non-cancerous mammary epithelial cell line, MCF-10A [7]. Moreover, IHC data showed strong cytoplasmic expression in IBC tissues compared to normal tissues [7]. To validate that MTDH is not solely overexpressed in the plasma membrane of IBC cells, whole cell lysates of SUM-149 (triple-negative (TN) IBC) and SUM-190 (HER2+ IBC) cells were immunoblotted. As shown in Figure 1A,B, MTDH is significantly overexpressed in SUM-149 with a 2.6 fold-change (f.ch.) and 3.0 f.ch. for SUM-190 IBC cell lines compared to MCF-10A cells. These findings confirm our results and correlate with previous findings from Zhang et al., where high ratios of HER2 transcripts increased proteomic levels of MTDH [30]. Several publications have highlighted MTDH as an important target in the treatment of BC, including its role in cancer onset and progression [31,32]. Therefore, our objective is to elucidate the cellular and molecular functions that MTDH plays in IBC. We hypothesized that knocking out MTDH in IBC cells would decrease proliferation and their capacity to form colonies. Here, we developed a MTDH CRISPR/Cas9-based knockout cell line (sg-MTDH) using SUM-149 and SUM-190 IBC cells. To test our hypothesis, we compared their expression with wild-type (WT) cells. The knockout of the MTDH gene in SUM-149 and SUM-190 was validated by qPCR and immunoblotting (Figure 2A,B,E,F). IBC sg-MTDH SUM-149 cells show a significant decrease in gene expression levels (−3.44 f.ch.) and protein expression levels (59%) compared to WT. Similarly, sg-MTDH SUM-190 IBC cells showed a significant reduction in gene expression (−3.56 f.ch.) and protein expression levels (67%) compared to WT. To test differences in cell proliferation capacity, we monitored SUM-149 and SUM-190 (WT and sg-MTDH) for 24, 48, and 72 h. The results showed that the proliferative capacity of SUM-149 (26%) and SUM-190 (51%) was significantly reduced at 72 h in the sg-MTDH cell model (Figure 2C,G). Consistent with the proliferation results, the clonogenicity of MTDH-depleted cells was significantly reduced compared to WT cells (Figure 2D,H). Furthermore, we hypothesized that the overexpression of MTDH in normal cells would promote malignant alterations. To investigate the malignant effects of MTDH in normal cells, we overexpressed MTDH in the non-cancerous mammary epithelial cell line, MCF-10A. We confirm the overexpression of MTDH in MCF-10A cells by performing qPCR and immunoblotting for both control cells (Ctrl) and overexpressed cells (OE-MTDH) (Figure 2I,J). The results showed a significant increase in gene expression (8.33 f.ch.) and protein expression (96%) in OE-MTDH cells compared to Ctrl. As predicted, the proliferative capacity of OE-MTDH increased significantly (72%) compared to Ctrl cells. These results suggest that the depletion of MTDH in IBC cells is directly related to the proliferative capacity of IBC cells in short and long periods of time. Previous studies have shown that IBC cells can spontaneously form 3D tumorspheres in vitro [33,34,35]. To study whether the decrease in cell proliferation in SUM-149 and SUM-190 was translated into a 3D culture, we tested the ability of cells to form tumor spheroids after 96 h in culture. Representative images of the tumor spheroids of SUM-149 and SUM-190, WT, and sg-MTDH are shown in Figure 3A,D. We observed that the SUM-149 and SUM-190 cells of the sg-MTDH cell model significantly formed the smaller tumor spheroids (Figure 3B,E) derived from IBC cells. However, there were no significant differences in the number of spheroids (Figure 3C,F). These findings suggest that silencing MTDH affects the capacity of WT IBC cells to form larger tumor spheroids, suggesting a potential role of MTDH in the maintenance of tumor spheroid integrity in IBC cells. MTDH overexpression has been associated with increased cancer cell metastasis [10]. Therefore, we evaluated the effects of MTDH knockout on the migration and invasion potential of IBC cells. We performed a 24 h wound healing assay using SUM-149 and SUM-190 cell lines (WT and sg-MTDH). The results showed a significant reduction in the percent of wound closure of the sg-MTDH cell model in both SUM-149 (79% reduction) and SUM-190 (83% reduction) cells (Figure 4A–D), suggesting that there is a reduction in the IBC cell migration capacity. We also investigated whether MTDH overexpression promotes the migration of non-cancerous cells MCF-10A. The results showed a significant alteration in the ability of OE-MTDH (92% increase) cells to close the wound compared to Ctrl MCF-10A cells (Figure 4E,F). Next, we assessed the effects of the absence of MTDH on IBC cell invasion using a transwell assay with SUM-149 WT and sg-MTDH cells. We observed a significant reduction in sg-MTDH invasion (46%) compared to WT cells (Figure 4G,H). This evidence suggests that MTDH affects the migratory and invasive capacity of IBC cells, confirming a key role in cell motility. Since we showed that MTDH depletion plays a key role in proliferation, tumor spheroid formation, and IBC cell motility, we then investigated how key signaling pathways related to these functions are affected. MTDH has been associated with the NF-κB, AKT, and ERK pathways, affecting cancer cell function [8,17,18,36]. Moreover, NF-κB has been found to be hyperactive in IBC when compared to nIBC models [37,38]. Furthermore, because our group and others have demonstrated the importance of the JAK2/STAT3 pathway in the regulation of IBC cellular identity [39], stemness, and tumor progression [40], herein, we also investigated the modulation of STAT3 expression after MTDH knockout. First, we evaluated the expression of the STAT3 and NFKB1 genes in both IBC cell models. For SUM-149, there is a trend (p = 0.10) of STAT3 downregulation (−2.00 f.ch.) in sg-MTDH cells. Moreover, the expression of the NFKB1 gene is significantly reduced by −2.11 f.ch. (Figure 5A) in sg-MTDH cells. In contrast, there were no significant differences in the expression of the STAT3 or NFKB1 gene expression in SUM-190 (Figure 5B). We also evaluated the protein expression of these signaling molecules by immunoblotting whole cell lysates of SUM-149 and SUM-190 (WT and sg-MTDH) and MCF-10A cells (Ctrl and OE-MTDH) cells (Figure 5C). Densitometry analysis demonstrates that, in SUM-149 sg-MTDH cells, there is a significant downregulation of STAT3, AKT and NF-κB protein expression. No significant differences were observed for JAK2 and ERK (Figure 5D). Similarly, there was a significant reduction in AKT and NF-κB in SUM-190 sg-MTDH cells. However, we also saw a significant reduction in ERK expression. Interestingly, there were no significant differences in STAT3 expression (Figure 5E), which could be related to HER2+ overexpression and changes in the downstream signaling in SUM-190 IBC cells. These findings suggest a possible crosstalk in the regulation of STAT3 and NF-κB total protein expression through MTDH in TN SUM-149 sg-MTDH cells but not in HER2+ SUM-190 sg-MTDH cells. Furthermore, in MCF-10A OE-MTDH cells, there was an upregulation in the protein expression of STAT3 and NF-κB signaling molecules (Figure 5F). These findings validate the possible interaction of a regulation between STAT3 and NF-κB in SUM-149 IBC cells. A crosstalk between STAT3 and NF-κB signaling pathways has previously been described where both proteins regulate distinct and overlapping groups of genes during tumorigenesis [41]. We also verified whether MTDH knockout or overexpression affected the phosphorylation of probed proteins. However, there were no significant differences in the activation of any of the signaling molecules evaluated (Figure 5G–I). MTDH has been correlated with BC progression and poor overall survival in patients [42], but its role in regulating tumor cell proliferation remains controversial [14]. Therefore, we investigated the role of MTDH in the regulation of SUM-149 WT and sg-MTDH tumor growth. We hypothesized that a reduction in MTDH would affect tumor development and progression. First, the results showed that the health of the mice was not compromised by the absence of MTDH (Figure 6A). A reduction in tumor size was observed at weeks 1 and 2 in sg-MTDH tumors compared to WT (Figure 6B), and, contrary to what we expected, the absence of MTDH did not reduce tumor volume or tumor weight (Figure 6C). These results demonstrate that the reduction in cell proliferation detected in vitro was not observed in vivo, at least in the primary tumor. Furthermore, H&E staining of lung tissues from WT and sg-MTDH mice revealed the presence of epithelial-like cells. For the WT group, 43% of the mice showed aggregates of epithelial-like cells in the alveolar parenchyma (Figure 6D), as Zhang et al. also described [43]. On the other hand, in the sg-MTDH group, 29% of the mice showed aggregates of epithelial-like cells near the hilum and between the lung arterioles. These results are consistent with the in vitro findings in which MTDH silencing demonstrates a reduction in the invasive capacity of IBC cells. Although no significant changes in primary tumor volume were observed, metastatic lesions were found in a higher percentage of total WT mice than in the sg-MTDH mice group. These findings suggest that what is affected is the invasion and metastatic potential of tumor cells instead of the progression of the primary tumor in vivo. To validate that SUM-149 sg-MTDH cells remained with a low expression of MTDH after 10 weeks of study, we performed a qPCR for MTDH gene expression. A significant reduction (−5.52 f.ch.) was observed (Figure 6E) in mouse tumors of the sg-MTDH group compared to WT. We also investigated whether MTDH depletion affected the expression of signaling molecules in sg-MTDH tumors observed in in vitro models. MTDH depletion showed a significant decrease in STAT3 expression, consistent with what we observed in SUM-149 sg-MTDH cells (Figure 6F,G). We also verified whether the reduction of MTDH in sg-MTDH tumors affected the activation of the indicated proteins and there were no significant differences detected (Figure 6H). High expression of MTDH has been associated with tumor metastasis, decreased survival outcomes, and higher mortality in female reproductive cancers such as breast, ovarian, and cervical [44]. We previously reported that MTDH is overexpressed in the plasma membrane of IBC cells compared to non-cancerous mammary cells and nIBC cell lines, in addition to its presence in IBC tissues and the tumor emboli [7]. Many studies have shown that lethality of IBC is due to its ability to invade the vascular and lymphatic systems through the tumor emboli (non-adherent clusters of cancer cells), causing the inflammatory phenotype, breast edema, and lymph node metastases [45]. Therefore, studies are needed to identify potential proteins associated with IBC invasion and metastasis. To our knowledge, this is the first study to evaluate the functional role of MTDH in IBC invasion and metastasis. Our results demonstrated that MTDH depletion decreased the proliferation of the TN and HER2+ IBC models and their ability to form colonies. At the same time, MTDH overexpression promoted the proliferation of non-cancerous mammary cells. Although MTDH has been correlated with BC proliferation by showing high levels of Ki-67 in tissues, previous studies demonstrated that knocking down of MTDH did not affect the proliferation of MDA-MB-231-LM2 cells, a subcell line of nIBC with a high propensity to lung metastasis [28,46,47], in contrast to our unique findings in IBC cells. We also demonstrate that depletion of MTDH reduced the size of tumor spheroids, suggesting that MTDH could play an important role in their integrity, possibly affecting cell-to-cell interactions. These findings may have a correlation with studies that have demonstrated MTDH as a cell membrane protein important in tight junctions (TJ) during cell–cell adhesion interactions [12]. TJ proteins fundamentally influence cell processes that regulate polarity, differentiation, and migration, all of which are critical steps to cancer progression [48]. Herein, we also established that depletion of MTDH decreased the migration and invasion of IBC cells while overexpression of MTDH promoted the migration of non-cancerous cells. Studies demonstrate that MTDH is relevant in cancer cell invasion, since Matrigel invasion assays have shown that different cancer cells, such as hepatocellular carcinoma and glioma cells, display an increased invasive ability through MTDH overexpression [36]. Additionally, the upregulation of MTDH increased the invasion of cancer cells by upregulation of matrix metalloprotease enzymes (MMPs), specifically MMP-2 and MMP-9 [49,50]. Interestingly, IBC is known to ubiquitously express E-cadherin in primary tumors, tumor emboli, and is associated with metastasis [51,52]. Studies have shown that E-cadherin increases the invasion capacity of the SUM-149 cell line by increasing the levels of MMPs [53]. E-cadherin promotes the dissemination of IBC cells by maintaining embolus integrity through cell-to-cell interactions. Studies suggest that E-cadherin plays a key role in a passive metastatic mechanism by which IBC cells invade the circulatory system in clusters, rather than spreading as single cells [33,54]. Therefore, our findings can have a potential correlation with the cellular and molecular functions of E-cadherin, as the migration and invasion capabilities of MTDH-depleted cells were impaired, as well as the integrity of the spheroids. More studies could be conducted to better understand the relationship between MTDH depletion and E-cadherin function in IBC since MTDH was also found present in tumor emboli in our previous studies [7]. In the current study, we also assessed the expression of signaling pathways that play an important role in the regulation and interactions of MTDH, such as AKT, ERK, and NF-κB [36]. We demonstrate that the depletion of MTDH negatively regulates the genetic expression of NF-κB (NFKB1) and the expression of NF-κB, AKT, ERK, and STAT3 in a TN IBC model. In a HER2+ IBC model, no changes in gene expression were observed and only a down-regulation of NF-κB, AKT, and ERK proteins was detected. The contrasting results in gene expression between the IBC models could be explained by previous studies which have demonstrated that the amplification or overexpression of HER2 promotes the activation of NF-κB and STAT3 in contrast to TNBCs [55,56]. Interestingly, no changes in the phosphorylated proteins were observed for both isogenic models. Therefore, we propose that these pathways are being affected via post-transcriptional regulation. Previous studies have demonstrated that MTDH interacts with the cyclic AMP-responsive element binding protein (CBP), which is a NF-κB coactivator that serves as a bridge element for NF-κB, CBP, and the basal transcription machinery, enhancing migration and invasion processes [17,57]. MTDH has also been shown to interact with NF-κB by translocating into the nucleus and combining with the p65 subunit of NF-κB and promoting the expression of downstream genes as cell adhesion molecules (i.e., ICAM-2, ICAM-3, selectin P ligand, selectin E, selectin L), toll-like receptor TLR4 and TLR5, FOS, JUN and cytokines IL-8 that are involved in tumor progression and metastasis [17]. Furthermore, we evaluated the STAT3 pathway due to its importance in the regulation of IBC growth and tumor progression described by us [40] and others [39]. In IBC, STAT3 has been implicated to play a crucial role, because IL-6, an inflammatory cytokine that activates the STAT signaling pathway, is up-regulated in IBC tumors compared to nIBC [58]. It is known STAT3 collaborates with NF-κB to promote cancer development and progression [41]. STAT3 and NF-κB, as two important transcription factors, bind cooperatively in a subset of gene promoters to collaboratively induce gene expression during tumorigenesis. More specifically, members of the NF-κB family such as RelA can physically interact with STAT3, and their association can modify their transcriptional activity, promoting tumorigenesis [59]. Currently, there are no studies demonstrating a direct relationship between MTDH and STAT3, but a recent study showed that treating lung cancer cells with a mushroom extract decreased the expression of both MTDH and STAT3, inhibiting proliferation and metastasis [60]. Based on our results, we suggest that MTDH could have a potential effect on STAT3 expression through the regulation of the NF-κB in this TN-IBC model as it has been suggested in other models [61,62]. On the other hand, we did not report any changes in STAT3 expression in the HER2+ IBC model. According to a study by Du et al., up-regulation of MTDH is associated with reduction of the phosphatase and tensin homolog deleted on chromosome 10 (PTEN) and resistance to trastuzumab in HER2+ BC [63]. The same study concludes that MTDH modulates the PTEN-PI3K/AKT pathway through a NF-κB-dependent pathway [63]. These findings can potentially correlate with our results, which suggest that both AKT and NF-κB are affected by MTDH depletion but not STAT3 in HER2+ IBC cells. Additional studies are needed to understand the role of MTDH in NF-κB and STAT3 for IBC models. Given that our in vitro results suggested that MTDH knockout would decrease tumor growth, we developed IBC xenograft models using WT and MTDH knockout cells. Our results show that MTDH depletion only reduced tumor volume in the early stages of the study, however the reduction was not sustained. These discrepancies suggest the involvement of interactions between the cancer cells and additional components of the tumor microenvironment in the in vivo model that might compensate the growth response [64]. Nonetheless, our results demonstrated that upon MTDH silencing there was reduced metastasis to lung tissue of mice. Interestingly, studies have described that MTDH contains an extracellular lung-homing domain that mediates BC cells to form lung metastasis [15]. In addition, another study demonstrated that MTDH knockdown reduced the adhesion of nIBC cells to lung microvascular endothelial cells, as well as to the bone marrow [28]. In our study, we show that MTDH knockout decreases lung metastasis of MTDH depleted TN IBC cells (29%) compared to WT (43%). Based on these findings, we propose that a lack of MTDH delays the proliferative capacity of tumor initiating cells, which affects the initial stage of tumor development and does not affect primary tumor growth but regulates invasion and metastasis. Furthermore, we evaluated NF-κB and STAT3 in tumor xenograft models where MTDH remained depleted after 10 weeks. We found that there is a reduction in NF-κB and STAT3 proteins, which was observed in the in vitro model using the same TN IBC cells. In this study, we confirm and validate that MTDH promotes metastasis as suggested by other studies [15,28]. Therefore, our results suggest a functional role for MTDH in the aggressiveness of IBC by affecting cell proliferation, tumor spheroid formation, migration, and invasion. Furthermore, we showed for the first time how MTDH depletion affects the STAT3 and NF-κB signaling pathways in IBC models and a potential crosstalk between pathways. We also demonstrated the effects of MTDH depletion on early tumor development and metastasis in xenograft models. In summary, this study shows the potential of MTDH as a possible target to better understand the progression and metastasis of IBC. The SUM-149 and SUM-190 cell lines were obtained from BioIVT (Westbury, NY, USA) and cultured in Ham’s F-12 Nutrient Medium (Gibco/Life Technologies, Waltham, MA, USA) according to the manufacturer’s instructions. SUM-149 cells were supplemented with 10% fetal bovine serum (FBS) (Corning®, Corning, NY, USA) and SUM-190 cells were supplemented with 2% FBS (Corning®), 1 g/L bovine serum albumin (BSA) (Rockland Immunochemicals, Pottstown, PA, USA) and insulin-transferrin-sodium (ITS) (Sigma-Aldrich, St. Louis, MO, USA). HEK-293 and MCF-10A cells were obtained from ATCC® (Manassas, VA, USA). The human embryonic kidney cell line HEK-293 (ATCC® CRL-1573TM) was cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Gibco/Life Technologies) and supplemented with 10% FBS. The human non-cancerous mammary epithelial cell line MCF-10A (ATCC® CRL-10317TM) was cultured in DMEM/Ham’s F-12 (Gibco/Life Technologies) with 10% horse serum (HS) (Sigma-Aldrich) supplemented with 20 ng/mL of epidermal growth factor (EGF), Cholera Toxin B, Hydrocortisone solution, HEPES and insulin (Sigma-Aldrich) as described in [65,66]. Cells were regularly tested to ensure they were free of mycoplasma infection using the Mycoplasma Detection Kit (ASB-1310001, Nordic BioSite AB, Sweden). All cell lines were genotyped for authenticity using the short tandem repeat (STR) profile and interspecies contamination testing services from IDEXX BioResearch (Columbia, MO, USA). Plasmids: The MTDH-CRISPR/Cas9 vectors were constructed in the Duke Functional Genomics Shared Resource at Duke University (Durham, NC, USA) using the lentiCRISPR/Cas9 lentiCRISPRv.2.0 backbone plasmid 52961 from Addgene (Watertown, MA, USA). The MTDH overexpression vector (LV-h-MTDH ORF-GFP) and the empty vector (LV-GFP) were constructed and purchased from VectorBuilder (Chicago, IL, USA). Vectors for packaging (PCMV delta R8.2) and envelopes (pMD2.G) of viral particles were purchased from AddGene. Lentiviral Particles Generation: HEK-293 (2 × 106) cells were seeded in 60 mm culture plates and transiently transfected with the 1 μg of lentiCRISPR/Cas9 MTDH sgRNA or LV-MTDH-ORF, 900 ng of PCMV delta R8.2 and 100 ng of pMD2.G. for 18 h in Opti-MEM medium (Gibco/Life Technologies). Fugene (Promega, Madison, WI, USA) was used as a transfection reagent according to the manufacturer’s instructions. After the transfection period, cells were refreshed with DMEM supplemented with 30% FBS to maintain the stability of the viral particles. The supernatant containing the particles were collected at 48 and 72 h post-transfection. Ultimately, the particles were concentrated with the Lenti-XTM Concentrator (Takara, Kusatsu, Shiga, Japan) for 72 h at 4 °C, filtered, and stored at −80 °C. Mammalian Cell Transduction: Parental cells SUM-149, SUM-190, and MCF-10A were transduced with the respective viral particles for each cell line. Briefly, cells (1 × 105) were seeded in 6-well plates and allowed to grow until ~60% confluence. The transduction reaction was carried out with the viral particles of interest and 8 μg/mL Polybrene (Sigma-Aldrich) in Opti-MEM media (Gibco/Life Technologies) for 24 h. After transduction, the transduced cells were selected with 1 μg/mL of Puromycin (Sigma-Aldrich). Total RNA from MTDH-edited and parental cell lines (SUM-149, SUM-190, and MCF-10A) and tumors from xenografted mice were extracted using the RNeasy Plus mini kit (Qiagen, Germantown, MD, USA) according to the manufacturer’s instructions. DNase digestion was performed after RNA isolation with DNase I according to the manufacturer’s protocol (Millipore Sigma, St. Louis, MO, USA). The samples were quantified with Nanodrop 1000 (Thermo Fisher Scientific, Waltham, MA, USA). cDNA was synthesized (1 µg RNA) using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA). All reactions for cDNA synthesis were performed with the presence and absence of reverse transcriptase to exclude any possibility of DNA contamination. A quality control PCR with B2M primers was performed on an MJ Mini thermocycler (Bio-Rad) as follows: 12.5 µL 2× MM Jumpstart Taq (Millipore Sigma) or 2× MM Eco-Taq (Midsci, St. Louis, MO, USA), 1 µL 5 µM B2M of forward and reverse primers (Table 1), 2 µL 50 ng of cDNA or RT-minus reaction. The thermocycler program for the Jumpstart MM was utilized as follows: 94 °C for 2 min 1×, 94 °C for 30 s, 59 °C for 30 s, and 72 °C for 2 min 30× and a final extension of 72 °C for 5 min. The thermocycler program for the Eco-Taq MM was done as follows: 95 °C for 10 min 1×, 94 °C for 1 min, 59 °C for 30 s and 72 °C for 45 s 30×, and a final extension of 72 °C for 10 min. The PCR reactions were electrophoresed in a 2% agarose gel stained with SmartGlow safe green stain (Midsci) and photodocumented with the BioSpectrum Imaging System (UVP LLC, Upland, CA, USA). Additionally, amplification efficiency curves were performed for all primers. Those samples with a strong amplification band on RT-plus with its matched RT-minus reaction with no amplification band were selected to perform the qPCR. Quantitative PCR reactions were performed as follows 12.5 µL SSO SYBR Green Supermix (Bio-Rad), 300 nM of each primer, and 100 ng of cDNA in a 25 µL reaction. qPCR was performed on a CFX96 (Bio-Rad), and the thermocycler program was utilized as follows: an initial 95 °C for 3 min 1X, then forty cycles were run at 95 °C for 10 s, 60 °C for 30 s. The changes in gene expression were calculated using the 2−ΔΔCt method as described in [65,67,68] using triplicate cDNA samples from two individual experiments. All primers were designed using the following websites https://www.idtdna.com/pages, https://primer3.ut.ee, http://biotools.nubic.northwestern.edu/OligoCalc.html, and https://blast.ncbi.nlm.nih.gov/Blast.cgi (all accessed on 3 June 2022) [69,70,71,72]. Primers were synthesized by Millipore Sigma, and their information can be found in Table 1. Cells and flash-frozen primary tumors from xenografted mice were lysed on ice and proteins were extracted using a lysis buffer containing 50 mM HEPES pH 7.0, 250 mM NaCl, 2 mM EDTA, 0.5% Igepal, 2 mM Na3VO4, 25 mM β-glycerol phosphate, 50 nM NaF, and cOmpleteTM Mini Protease Inhibitor Cocktail (Sigma-Aldrich). Total protein was quantified using the Precision Red protein assay kit (Cytoskeleton, Inc. Denver, CO, USA). Equal total protein amounts (10 μg) were subjected to separation by SDS-PAGE gels and transferred onto a PVDF membrane. After blocking with 5% milk, the membrane was incubated with the indicated primary antibodies in 5% bovine serum albumin (BSA) at 4 °C overnight at a dilution of 1:1000. The primary antibodies used were Anti-LYRIC/AEG1 (#ab124789, Abcam, Cambridge, MA, USA), AKT (#9272, CST, Danvers, MA, USA), p-AKT (Ser473) (#4060, CST), p44/42 MAPK (ERK1/2) (#9102, CST), phospho-p44/42 MAPK (p-ERK1/2 Thr202/Tyr204) (#4370, CST), STAT3 (#4904, CST), p-STAT3 (Tyr705) (#9145, CST), p-STAT3 (Ser727) (#9134, CST), JAK2 (#3230, CST), phospho-JAK2 (Tyr1007/1008) (#3771, CST), β-tubulin (#86298, CST) and β-actin (#A1978, Sigma-Aldrich). The membranes were then incubated with a secondary antibody according to their respective antibody species at a dilution of 1:10,000 or 1:20,000 for 1 h at room temperature (RT). The membrane was then developed with the Pierce TM ECL Western Blot Substrate kit (Thermo Fisher, Waltham, MA, USA) and visualized using the BioSpectrum Imaging System (UVP LLC). The integrated density of the bands of interest was quantified using ImageJ software (NIH, Bethesda, Maryland). Quantification of each protein was ensured by normalizing the integrated densities of bands of interest for all antibodies to the integrated density of the same immunoblotted lysate for β-acting or β-tubulin as described by us [40,65,73,74]. Arbitrary units are equal to the normalized integrated density of each protein relative to the control or parental cell line. Parental or MTDH-edited cells (SUM-149, SUM-190 and MCF-10A) were seeded in a 96-well plate with a density of 2.0 × 103 cells/well for 24, 48 and 72 h. Proliferation was evaluated using the CyQUANT® NF Cell Proliferation Assay Kit (Invitrogen, Waltham, MA, USA). Fluorescence was measured using a GloMax® Explorer microplate reader (Promega) in the 500–550 nm range. The experiments were carried out in triplicate at least three times. SUM-149 MTDH edited and parental cells were seeded in triplicate at 200 cells/well in a 24-well plate containing Ham’s F-12 with 10% FBS and incubated for 10 days. For SUM-190, MTDH edited and parental cells were seeded at 1 × 103 cells/well in triplicate in a 6-well plate containing Ham’s F-12 with 2% FBS and incubated for 14 days changing the media every 4–5 days as described by [75]. After the incubation period at 37 °C, cells were fixed with methanol, washed with 1X phosphate saline buffer (PBS), stained with crystal violet for 5 min at RT, washed with water, and left to dry overnight. Colonies containing >50 cells were counted and analyzed using the Cytation 10 Confocal Imaging Reader (Agilent/BioTek, Santa Clara, CA, USA). The cell capacity to migrate was evaluated by seeding parental or MTDH-edited cells (SUM-149, SUM-190, and MCF-10A) in two-well silicone inserts with a defined cell-free gap wound plate (Ibidi USA Inc., Madison, WI, USA). For SUM-149, 4 × 104 cells/well were seeded, 8 × 104 cells/well for SUM-190, and 1 × 105 cells/well for MCF-10A. All cell lines were cultured in their respective complete culture medium for 24 h. After the incubation period, the insert was removed and the cells were allowed to migrate for 24 h at 37 °C as described by us in [74]. For MCF-10A cells, a complete culture medium with 2% HS was used to perform the assay. The cells were then fixed with 4% paraformaldehyde for 15 min, washed with 1X PBS, permeabilized with 0.1% Triton X-100 for 15 min at RT, and blocked with 1% BSA. Cells were stained for 1 h with a 1X rhodamine-phalloidin solution (InvitrogenTM/Life Technologies) to visualize actin filaments (F-actin). After washing three times with 1X PBS, cells were incubated with 1 µg/mL of DAPI (Life Technologies) for nuclear staining. Cell migration was quantified by measuring the distance (µm) between the wound edges using Olympus CellSense Imaging Software (Center Valley, PA, USA) on micrographs at a magnification of 4×. Fluorescence images were obtained at a magnification of 20× using the Cytation 10 Confocal Imaging Reader (Agilent/BioTek). Cell invasion was measured using the BD BioCoat MatrigelTM invasion assay (BD Biosciences, San José, CA, USA). SUM-149 wild-type (WT) and MTDH-edited quiescent cells (1 × 105) were seeded in the upper chambers and incubated at 37 °C for 24 h to allow invasion into 10% FBS medium (chemoattractant). After the incubation period, cells in the top chambers were removed with a cotton swab and cells attached to the bottom surface of the membrane were fixed and stained with propidium iodine (Sigma-Aldrich) as previously described in [65,74]. Cells were quantified with Gen5 Data Analysis Software (Agilent/BioTek) with a montage of micrographs obtained using the Cytation 10 Confocal Imaging Reader (Agilent/BioTek) at a magnification of 20×. SUM-149 and SUM-190 (WT and MTDH-edited) cells were seeded in triplicate in 6-well ultralow attachment plastic plates (Corning®). For SUM-149, a density of 4 × 104 cells/well was used and 8 × 104 cells/well for SUM-190. All cell lines were cultured in their respective complete media supplemented with 2.25% polyethylene glycol (PEG-800) (Sigma-Aldrich) as described by [34,76]. Cells were incubated at 37 °C for a period of 96 h. After the incubation period, the micrographs were captured using the Cytation 10 Confocal Imaging Reader (Agilent/BioTek). Gen5 Data Analysis Software (Agilent/BioTek) was used to calculate the quantity and area of tumor spheroids. The study was approved by the Universidad Central del Caribe (UCC) Institutional Animal Care and Use Committee (IACUC) (Animal Welfare Assurance #D16-00343) and was carried out following IACUC guidelines. Female severe combined immunodeficient mice (SCID) (Charles River Laboratories International, Wilmington, MA, USA) between 21 and 28 days of age were housed under specific pathogen-free conditions. Mice received an irradiated AIN 76-A phytoestrogen-free diet (Tek Global, Harlan Teklad, Madison, WI, USA) and water ad libitum. To test the effects of MTDH silencing in IBC tumor formation and progression, we injected 1.5 × 106 WT SUM-149 and MTDH-edited cells into the lower right mammary fat pad of female mice in a 100 µL volume dilution (1:1) of reduced growth factor Matrigel (BD Biosciences) and serum-free media, as previously described in [67,73]. Group allocation was made randomly: (a) mice injected with WT cells (n = 10) and (b) mice injected with MTDH-edited cells (n = 10). One week after injection, the weight and tumor volume of the mice were measured weekly for 10 weeks. Tumor volume (mm3) was measured with a caliper and calculated as follows: [(width)2 × length)/2] as described in [77]. At the end of the study, the tumors and both lungs were excised and kept in optimal conditions for future experiments. Sectioning: The excised lungs were left in 10% neutral-buffered formalin (NBF) overnight and then immersed in a solution of 20% sucrose in 1X PBS. The lungs were then placed in optimal cutting temperature compound (OCT) (TFMTM, General Data Company Inc., Cincinnati, OH, USA). Sectioning was performed using a Leica CM 1860 cryostat (Leica Byosystems, Deer Park, IL, USA) by cutting 8 μm sections of the left and right lung of mice injected with WT SUM-149 or MTDH-edited cells (n = 7/group). The sections were mounted on slides with a subbing solution as described in [78]. Hematoxylin & Eosin (H & E) staining: First, tissues were stained with Mayer hematoxylin solution (Sigma-Aldrich) for 10 min at RT and then rinsed with running tap water for 10 min until the water was colorless. The tissues were then rinsed in 95% alcohol for 10 s. Then, the staining was performed with Eosin Y with Phloxine B (Sigma-Aldrich) solution for 30 s and rinsed in ascending series of ethanol (70%, 95%, and 100%). Finally, tissues were processed in xylene 1 min twice and mounted using VectaMount® permanent mounting medium (Vector Labortories, Newark, CA, USA). A pathologist analyzed H&E-stained slides and the micrographs were captured at a magnification of 20×. The p-value for the in vitro studies was calculated using Student’s t-tests, ordinary-one way or two-way analysis of variance (ANOVA) with the Bonferroni multiple comparison test. Gene expression studies for each cell line or tumor were evaluated using the 2−ΔΔCt formula by comparing their relative gene expression to the reference genes. Data are expressed as Mean ± S.E.M. Each experiment was carried out in three or more independent biological replicates. Statistical analyses were performed using Graph Pad Prism v.9.0 (San Diego, CA, USA), and differences were considered significant when p < 0.05. In vivo studies: Summary statistics were performed to describe the tumor volume and weight. The normality evaluation was performed using the Shapiro–Wilk test. The presence of outliers was verified using a generalized extreme studentized deviation (ESD) test. Mean changes were evaluated over time using a general linear model with repeated measures. Model diagnostics were performed as well, i.e., Cooks distances and DF betas were ascertained to access influential values. A pairwise comparison was performed to determine the difference between the groups and each week. Only if the data did not follow a normal distribution, Wilcoxon–Mann–Whitney tests were performed. The significance level (α) was set to 0.05. Analyses were performed using the statistical software R version 4.1.0. Our findings identify for the first time a potential role of MTDH in promoting IBC proliferation, migration, invasion, and metastasis by regulating NF-κB and STAT3. Furthermore, we showed that MTDH plays a role in the invasive capacity of IBC promoting metastasis to the lung. Altogether, these findings open the possibility of a new target to better understand the molecular biology of IBC. MTDH has the potential to serve as a diagnostic target to diagnose and treat the progression of this deadly disease in IBC patients.

PMC10002533

Simon J. S. Cameron,Arwyn Edwards,Robert J. Lambert,Mike Stroud,Luis A. J. Mur

Participants in the Trans-Antarctic Winter Traverse Expedition Showed Increased Bacterial Load and Diversity in Saliva but Maintained Individual Differences within Stool Microbiota and Across Metabolite Fingerprints

02-03-2023

salivary microbiota,stool microbiota,metabolomics,metabolite fingerprinting,isolation,longitudinal,microbiota dynamics,microbiota-metabolome integration

Understanding the impact of long-term physiological and environmental stress on the human microbiota and metabolome may be important for the success of space flight. This work is logistically difficult and has a limited number of available participants. Terrestrial analogies present important opportunities to understand changes in the microbiota and metabolome and how this may impact participant health and fitness. Here, we present work from one such analogy: the Transarctic Winter Traverse expedition, which we believe is the first assessment of the microbiota and metabolome from different bodily locations during prolonged environmental and physiological stress. Bacterial load and diversity were significantly higher during the expedition when compared with baseline levels (p < 0.001) in saliva but not stool, and only a single operational taxonomic unit assigned to the Ruminococcaceae family shows significantly altered levels in stool (p < 0.001). Metabolite fingerprints show the maintenance of individual differences across saliva, stool, and plasma samples when analysed using flow infusion electrospray mass spectrometry and Fourier transform infrared spectroscopy. Significant activity-associated changes in terms of both bacterial diversity and load are seen in saliva but not in stool, and participant differences in metabolite fingerprints persist across all three sample types.

Participants in the Trans-Antarctic Winter Traverse Expedition Showed Increased Bacterial Load and Diversity in Saliva but Maintained Individual Differences within Stool Microbiota and Across Metabolite Fingerprints Understanding the impact of long-term physiological and environmental stress on the human microbiota and metabolome may be important for the success of space flight. This work is logistically difficult and has a limited number of available participants. Terrestrial analogies present important opportunities to understand changes in the microbiota and metabolome and how this may impact participant health and fitness. Here, we present work from one such analogy: the Transarctic Winter Traverse expedition, which we believe is the first assessment of the microbiota and metabolome from different bodily locations during prolonged environmental and physiological stress. Bacterial load and diversity were significantly higher during the expedition when compared with baseline levels (p < 0.001) in saliva but not stool, and only a single operational taxonomic unit assigned to the Ruminococcaceae family shows significantly altered levels in stool (p < 0.001). Metabolite fingerprints show the maintenance of individual differences across saliva, stool, and plasma samples when analysed using flow infusion electrospray mass spectrometry and Fourier transform infrared spectroscopy. Significant activity-associated changes in terms of both bacterial diversity and load are seen in saliva but not in stool, and participant differences in metabolite fingerprints persist across all three sample types. Stress that impacts a person’s normal lifestyle can lead to changes in the human microbiota [1]. Given this, it is likely that extreme stress may have an impact on the human microbiota that affects physiology and endurance in a range of scenarios. It has been suggested, for example, that the stress associated with military combat training increases intestinal permeability and symptoms of gastrointestinal disease [2], which may have important implications for the human microbiota. These factors could also be a major consideration in long-duration space travel by humans and could be important factors in health and fitness screening of potential participants in long-term exploration journeys. The human stress response centres on the endocrine system to produce hormones such as corticosteroids, which exert rapid non-genomic effects on neurons in the hypothalamus to help maintain homeostasis and adapt to stressful stimuli [3]. Stress can also increase the permeability of the gut, potentially allowing the gut microbiota to indirectly influence the human body through the hypothalamic–pituitary–adrenal axis (HPA). The human microbiota has been shown to modulate the host’s stress response through the production or alteration of key neurotransmitter systems, such as the serotonergic (5HT), norepinephrine (NE) and endorphin systems [4]. Indeed, work using animal models has suggested that the microbiota could influence depression [5] or anxiety and stress [6]. Further, the microbiota could play a role in various pathologies, including those involving the central nervous system [7], and could affect CNS auto-immune diseases, such as multiple sclerosis, Devic’s disease, and Guillain–Barré [8]. There have been limited human-focused studies investigating the link between the microbiota and stress and they have been confined to assessing the impact of probiotics [9]. In this current study, our motivation was to examine the impact of prolonged stressful conditions on the human microbiota and metabolome, which may prove relevant to long-duration space travel. To date, the field of medicine that is interested in the effect of long-duration space travel has focused on improving the provision of artificial life-support systems, though the human microbiota and metabolome have received some attention [10]. Nevertheless, the combined effects of isolation and extreme conditions on these human systems are unknown. Considering practical limitations, a substantial amount of work has been completed on the impact of space travel on the environmental microbiota [11,12]. Directly studying the impact of long-duration space travel on the human microbiota is difficult; it is confounded by issues such as very small participant cohorts, sample collection, stable long-term storage, and the intact return of materials to Earth. Effective studies have been completed on animal models and astronaut twin pairings that show that long-term space travel is marked by a mitochondrial stress response [13]. Terrestrial analogues present potential alternative models for space travel, allowing for the study of the human microbiota and metabolome under conditions representative of those that may be encountered. Isolation studies attempting to mimic the effects of prolonged space travel have measured changes in the structure and function of the human gut microbiota. In the Mars-500 study, the crew of volunteers lived and worked in a mock-up spacecraft to simulate a 520-day manned return mission to Mars. Changes in the taxonomic composition of the gut microbiota were seen within the first 14 days but no such changes were seen in its core functional capacity, with a reversion to the initial microbiota composition beginning to be seen within two weeks of the study ending [14]. More recently, additional work has highlighted the potential role of targeted probiotics for the promotion of astronaut health and to alleviate symptoms of illness [15]. Beyond the human microbiota, the broader implications of long-term space travel on health and metabolism are increasingly being explored, alongside potential nutritional counter-measures [16]. A broader appreciation and utilisation of multi-omic investigation a also evident, but these cases have focused more on genomics and transcriptomics [17]. The Coldest Journey Trans-Antarctic Winter Traverse (TAWT) expedition team aimed to complete a 2000-mile crossing of Antarctica in the winter (March to September 2013). This period is characterised by extreme cold, with ambient temperatures below −50 °C and three months of complete darkness. Initially, the TAWT expedition was to be undertaken from March 2013 to September 2013, with several months on either side to prepare for the expedition and subsequent uplift. The progression of the TAWT expedition was halted in June 2013, and the expedition team held firm in the interior of Antarctica. The five-man expedition team endured extreme environmental and physiological conditions for the remainder of the period in Antarctica, serving as a useful and realistic situation analogous to prolonged space travel. Here, we present what we believe is the first study to explore the multi-site response of the human microbiota (saliva and stool) and metabolome (saliva, stool, and plasma) to long-term environmental and physiological stress, which may be features of space travel. The TAWT expedition did not complete its initial intention to cross the Antarctic during the winter months of March to September. Due to the dangers posed by unexpectedly large crevasse fields, the expedition was halted during the fourth month. This was preceded by three months of traversing, where a total distance of 300 km was travelled. After three months of remaining in their stationary position, indicated in Figure 1a, the TAWT expedition undertook a further two months of traversing back to their starting position. This activity is detailed in Figure 1b and used as the framework for statistical analysis of data sets generated under the experimental workflow shown in Figure 1c. The TAWT expedition team consisted of five males, with a mean age of 37 (range 28 to 54), as detailed in Supporting Data Matrix S1, alongside individualised physiological information collected throughout the expedition. Metataxonomic sequencing of the V3 to V4 regions of the 16S rRNA gene was used to obtain compositional data on the salivary and stool microbiota (Figure 2). The human microbiota displays spatially different responses to the physiological and environmental stressors of the TAWT expedition. For saliva, activity-related differences could be observed in beta-diversity analysis, shown in Figure 2a, between the baseline and samples collected during the expedition (R2 = 0.31, p < 0.001). It should also be noted that participant differences in the salivary microbiota were also observed (R2 = 0.40, p < 0.001), with Participant D showing maintained individual differences through the expedition, as seen in Figure 2b. Within stool microbiota metataxonomics, no significant (R2 = 0.09, p = 0.186) activity-related differences were observed, but significant (R2 = 0.56, p < 0.001) individual differences were maintained throughout the expedition. These patterns are supported by significant differences between alpha-diversity measures using the Shannon diversity metric (Figure 2e) in saliva between baselines and expedition activity periods and in both saliva and stool between at least two participants. With regard to the 16S rRNA gene copy number (Figure 2f), which has previously been suggested as a marker of immunity [19], significant (p < 0.001) differences were only seen between baseline and expedition saliva samples. At the individual OTU level, seen in Figure 3, significant differences were observed across both sample (saliva and stool) and comparison (activity and participant) types. These were, however, dominated by significant differences between participants in both saliva and activity, with only Streptoccous and Prevotella melaninogenica in saliva and Ruminococcaceae family in stool significantly altered by expedition activity. As with diversity measures, these differences were seen between the baseline and samples collected during the expedition. For participant comparisons, significant differences in 32 and 33 OTUs, for saliva and stool, respectively, were observed. The lowest taxonomic identifications for these OTUs are listed in Supplementary Tables S1 and S2 for saliva and stool, respectively. We have previously shown that the salivary microbiota is stable over a one-year period in terms of microbiota composition but there were increases in bacterial load, as measured by the 16S rRNA gene copy number, in winter months [20]. We have previously suggested that the bacterial load in saliva is indicative of immune response in athletes following periods of extreme energy expenditure [19]. Here, we observe that both bacterial load and diversity were significantly higher during the TAWT expedition when compared with baseline samples. This may be indicative of an impaired immune status that allows expansion of the ecological niche in saliva and potentially the wider oral cavity. Significant reductions in salivary IgA have been previously demonstrated in athletes after periods of intensive physical activity, which is further associated with an increased risk of upper respiratory tract infections [21]. Due to logistical limitations, analysis of immunity markers was not possible in this study. Nevertheless, these results suggest that the salivary microbiota significantly increased in terms of both bacterial load and diversity during the expedition period. As these metrics were not only higher during traversing periods, it suggests that they are not caused solely by environmental stressors and may be impacted by stresses related to isolation and reduced environmental interaction. Although the level of taxonomic identification was not sufficient to identify the presence of pathogens in the saliva, this may indicate an increased risk of infection during periods of extreme stress and/or isolation. Interestingly, acidification of saliva was also observed during the later stages of the TAWT expedition, shown in Supplementary Figure S1, which could be associated with increased pathogen load in dental caries [22]. Interestingly, a related study on the effect of isolation during the Mars-500 experiment suggested that individual taxonomic features of the salivary microbiota remained stable over the study’s duration [23]. With stool samples, no change in bacterial diversity or load was observed throughout the expedition, indicating the maintenance of prior individual differences. This is in line with observations of the Mars-500 isolation study, although reduced temporal fluctuations were observed at the OTU level [24]. The resilience of the stool microbiota is particularly noteworthy when considering the changes in dietary intake in terms of calorific content and type, such as freeze-dried meals, associated with the TAWT expedition. The one OTU that showed a significant decrease from baseline from the beginning of the expedition period was assigned to the Ruminococcaceae family. Although our taxonomic resolution prevented detailed insight, members of the Ruminococcaceae family have been associated with perceived health benefits of dietary fibre [25]. This may suggest that dietary changes, rather than environmental or physical stresses, during the TAWT expedition had a greater influence on stool microbiota composition. Stool water content, as determined by weight loss during lyophilisation, also showed significant participant differences and none associated with activity, as seen in Supplementary Figure S2. Stool consistency has previously been shown to be a major determinant of the composition of the stool microbiota, likely associated with transit time through the gastrointestinal tract [26]. FTIR fingerprinting can provide a wide-ranging bioanalytical fingerprint of a sample. The mid-infrared wavelength (4000–600 cm−1) is divided into regions that are indicative of absorbance for key functional groups of biomolecules, including fatty acids, amides, and polysaccharides. Considering the entire spectral fingerprint, PCA modelling, seen in Figure 4a–f, showed no significant differences between activity and participants across all three analysed biofluids (saliva, stool, and plasma). At the univariate level (i.e., each cm wavelength division) of analysis, significant differences between at least two participants were observed across the absorbance spectra in both saliva and stool and between activity periods in saliva, shown in Figure 4g. Activity-related differences were shown to be abundant within the alcohol and primary amide region of 3500 to 3400 cm−1, the primary amide region of ≈1690 cm−1, and the N-O stretching functional group region of 1550 to 1500 cm−1. Although various applications of FTIR fingerprinting have shown promise in clinical diagnostics [27], one of its limitations is the lack of analytical resolution for specific biological entities. Within this study, wavelengths associated with amide functional groups appear to be the main source of statistically significant features related to the TAWT expedition. This may indicate that at the proteome/peptidome level, the TAWT expedition participants experienced a response to the physiological stress. This has been observed in plasma in response to functional overreaching exercises [28] and in the responses of mice [29] and humans [30] to space travel. Although not directly comparable, particularly with regard to the effect of zero gravity, it cannot be excluded that the TAWT expedition had an impact on participants at the proteome/peptidome level, which may have important implications for endurance and health. Using time series analysis, a significant interaction (p < 0.001) was observed in FTIR fingerprinting between timepoint and participant variables in plasma samples (Supplementary Table S3). This potentially indicates a variable response between participants to the physiological stress of the TAWT expedition. This may have important biological implications, but due to the limitations of experimental design and subsequent statistical power, this aspect is not possible to explore further. Nevertheless, based on the limited significantly different features observed through other analytical methods, it is likely that the interaction effect is statistically significant but with limited biological significance. Greater insights into metabolite variation can be provided by flow-infusion electrospray mass spectrometry, and this was used to fingerprint saliva, stool, and plasma samples. This provided a snapshot of each participant’s metabolism at the time of sampling and can suggest the identification of key metabolites. For logistical reasons, baseline plasma samples were not collected from expedition participants. In line with metataxonomic results, individual differences substantially outweighed the minimal activity-associated differences across all three biofluids. Unsupervised PCA of negative ion detection mode data was completed to prevent the overfitting of models based on small participant numbers (Figure 5a–f). These indicated no large-scale shift in metabolite fingerprints associated with activity across all three biofluids, and only participant differences were observed in stool samples. At the univariate metabolite fingerprint level, as seen in Figure 5g, 282 features were significantly different between at least two participants. Of these, three were detected in plasma and 279 were detected in stool samples. In stool, a total of eight features significantly differed between at least two activity periods. Tentative identifications of metabolites are given in Supplementary Data Matrix S2. A complementary analysis of positive ion detection mode data is shown in Supplementary Figure S3 and shows similar patterns, but fewer features were significantly different between participants. Time series analysis was completed and showed no significant (p > 0.05) interaction between timepoint and participant variables in all three biofluids and in both ion detection mode data sets (Supplementary Table S3). In negative ion detection mode data, six out of the eight significantly different features were not identified through a comparison of detected mass-to-charge features against the HMDB. Homovanillic acid sulphate was tentatively identified as a [M-2H]- ion and has been previously detected in human stool samples [31] and in plasma, and concentrations have been associated with levels of dopamine in the brain [32]. Homovanillic acid sulphate has not only been suggested as a marker of renal disease in adults [33] but also as a marker of bacterial biotransformation of dietary phenols [34]. It was not possible to link this with the microbiotas, as homovanillic acid sulphate metabolism is not a feature of the only bacterial taxon significantly differing in the stool, Ruminococcaceae. However, changes in the functional capacity and/or activity of the stool microbiota during the TAWT expedition would not be detected using metataxonomic sequencing. The second metabolite, tentatively identified as the [M+K-2H]- adduct of 6′-[(carboxymethyl)-C-hydroxycarbonimidoyl]-2′,3′,4,4′,5,6-hexahydroxy-[1,1′-biphenyl]-2-carboxylic acid, may have been a product of biotransformation from dietary intake. In positive ion detection mode data, four of the 12 features that significantly differed between at least two activity periods were tentatively identified: trans-3-feruloylcorosolic acid, coumesterol, acutilobin, and a pyranoflavonoid. None of these metabolites have a reported link with stress, isolation, or physical activity in the published literature. Considering the extent of individual participant differences, the minimal effect of the TAWT expedition on the detected metabolome suggests that participants were minimally impacted in terms of metabolic function at the point of host metabolism and host/microbiota interaction, considering the range of biofluids sampled in this study. Metabolomics in endurance athletes has been studied as a route to a mechanistic understanding of performance [35] and appears to be particularly associated with endurance capacity [36]. The TAWT expedition did not have a non-stressed control group and thus, direct comparisons are not experimentally possible. Temporal stability of the saliva, stool, and plasma metabolite fingerprints observed at the individual level suggests that expedition participants possess a high degree of physiological endurance. Thus, metabolomic profiles of potential participants in long-term expeditions, including space travel, may act as a useful predictor of endurance potential and act as a baseline against which perturbations may be indicative of impaired function. As a result of the combination of multi-omic analyses across saliva and stool samples, possible interactions or relationships between host and microbiota metabolism could be explored. To accomplish this, a correlation analysis between metataxonomic and metabolite fingerprints for saliva and stool biofluids was completed, shown in Figure 6. Firstly, metataxonomic feature correlations between matched timepoint/participant saliva and stool samples were completed (Figure 6a). This explored a potential link between oral and gastrointestinal microbiota as a route for the mechanistic understanding of systemic disease from changes in the oral microbiota [37]. Here, we found no correlation between any same taxon within both time- and participant-matched saliva and stool samples. We did, however, find correlations between different taxonomic groups between matched samples. It has been previously suggested that changes in either the oral cavity or gastrointestinal tract may cause systemic microbiota changes in other systems due to impaired immunological function [38]. We did not see a significant biological effect of the TAWT expedition in either saliva or stool associated with the activity period, so it can be considered unlikely to be an important factor in this analysis. A greater number of correlations of higher statistical significance were seen in both saliva and stool samples, as shown in Supplementary Figures S4 and S5, respectively. Correlations between microbiota taxa have been shown in disease states, such as periodontal disease state and severity [39]. In this study, because of technical limitations of sample numbers, it was not possible to look at whether the strength of correlation was associated with the activity period. It may be that these correlations represent multi-species communities within each ecological niche, which may be differentially affected by physiological and endurance stressors. In stool samples, many microbial taxa identified in these clusters, such as Bifidobacterium species, are associated with gut health. It may be possible to provide probiotic supplementation tailored to each baseline microbiota composition during future expeditions to maintain composition and potentially host health. Within each sample type, metataxonomic and metabolite fingerprinting data were combined to assess if there were correlations that may be indicative of host-microbiota interaction or potential prebiotic compounds (Figure 6b,c) for saliva and stool, respectively. Positive ion detection mode correlations are shown in Supplementary Figures S6 and S7 for saliva and stool, respectively. In both ion detection modes, stool samples showed the greatest number of correlating features that are expected when considering the greater microbial and metabolite abundance and complexity within the gastrointestinal tract [40]. Of note is the large number of metabolites within the stool, which are significantly correlated with a metataxonomic OTU and significantly different between participants, as detailed in Supplementary Data Matrix S2. As participant differences were the greatest separating factor modelled in both metataxonomic and metabolomic fingerprinting analyses, this raises the potential that these features are by-products of host–microbiota interactions or dietary prebiotics that assist in modulation of the gastrointestinal microbiota [41]. Due to the observational nature of this study, it is not possible to explore this further. Further work is needed to better understand the relationship between prebiotics and probiotics in the gut and salivary microbiotas and how they could be utilised to maximise long-term health during future long-term endurance journeys. The White Mars study protocol, from which samples were received for this study, formed part of the TAWT expedition that took place from December 2012 to September 2013. The White Mars protocol was hosted by King’s College London’s Centre of Human and Aerospace Physiological Sciences, which gained the necessary ethical approval to collect samples. This component of the White Mars project received additional ethical approval from the Aberystwyth University Research Ethics Committee. Informed consent was obtained from all study participants, and all participants’ information was anonymised by the expedition doctor. The funders of this work had no input into the design or reporting of this study. A summary of the study protocol is given in Figure 1. A total of five participants took part in the White Mars study protocol undertaken during TAWT. All sampling was completed by the expedition doctor. Stool samples were collected by depositing a small amount (≈10 g) of fresh faeces, meaning after a regular bowel movement, using a polystyrene faecal storage device (Greiner Bio-One, Frickenhausen, Germany). Saliva samples were collected by stimulating saliva production through chewing and depositing into a wide-neck 30 mL universal tube, after which ≈4 mL of sample was transferred to a smaller storage container. A baseline sample (except blood plasma) was taken from all participants prior to the start of the expedition and stored at −80 °C and analysed in parallel with the expedition samples. During the expedition, stool and saliva samples were taken at monthly intervals over an eight-month period. Stool samples were donated at a time convenient to expedition members’ daily activities, whereas saliva samples were donated by participants in the morning before any food or drink was consumed and prior to any oral hygiene regimen. No restrictions were placed on participants with regard to behaviour prior to sample donation. Baseline samples were frozen at −25 °C throughout the duration of the expedition. Expedition samples were stored at outside ambient temperatures (≈ −25 to −40 °C) in an uninsulated box until external temperatures exceeded approximately −20 °C, at which point they were stored in the expedition freezer set at −40 °C. Following the end of the expedition, samples were stored at −25 °C in freezers at the Princess Elizabeth base until transferred to the UK via ship transit in cold storage where temperatures did not exceed −25 °C. Upon arrival to the UK, samples were stored at University College London (London, UK) until transfer on dry ice to Aberystwyth University laboratories, where they were stored at −80 °C until sample processing. From storage, the 4 mL raw saliva samples were thawed at 4 °C and placed on a vortex mixture for 30 s to homogenise the mixture. Each saliva sample was split into two 2 mL aliquots. One aliquot was frozen at −80 °C until thawed for pH measurement and metabolite fingerprinting, and the remaining aliquot underwent centrifugation at 13,000× g for ten minutes at 4 °C. The resulting supernatant was removed, and the remaining pellet was also immediately frozen at −80 °C until used for DNA extraction, which was completed within seven days of receiving the sample. Blood plasma samples were thawed at 4°C and underwent centrifugation at 13,000× g for ten minutes at 4 °C to ensure no separation (which would be indicative of a mixed blood sample) and 2 mL transferred to a sterile 2 mL microcentrifuge tube and immediately frozen at −80 °C until thawed for pH measurement and metabolite fingerprinting. Stool samples were transferred to sterile, pre-weighed glass vials and weighed to determine wet weight. Stool samples were frozen at −20 °C and transferred to a freeze-drier set at −50 °C. Stool samples were then weighed every 24 h until they had reached a stable weight, determined as a weight reduction, over a 24-h period of less than 5%. After freeze-drying, stool samples were stored at −80 °C. For saliva samples, total genomic DNA was extracted using a FastDNA SPIN kit for Soil (MP Biomedical, Strasbourg, France) from the saliva pellet and 100 mg of freeze-dried stool as previously described [20] with an additional ethanol wash step included for stool samples. Separate DNA extraction controls were included for saliva and stool extraction batches. Measurements of plasma and saliva pH were completed in duplicate using a B-212 Twin pH meter (Horiba, Japan) after two-point calibration using a pH 7 and pH 4 buffer. For each measurement, 200 µL of sample was deposited on the sensor, ensuring full coverage of both sensor points. After stabilisation of the reading, the pH value was recorded, and the sensor was washed with ultrapure water and blotted dry. Standards for quantitative PCR were created through amplification of the entire 16S rRNA gene for each of the five baseline samples for stool and saliva separately using a protocol described previously [20]. After creation of standards, quantitative PCR was completed in duplicate on neat extracted DNA for each of the saliva and stool samples, as described previously [20]. Reactions were run using a C100 thermal cycler (BioRad, Hercules, CA, USA) and CFX96 optical detector (BioRad), with data captured using CFX Manager software (version 3.1, BioRad), under conditions of 95 °C for 10 min, 40 cycles of 95 °C for 15 s and 60 °C for 60 s, followed by a melt curve consisting of a temperature gradient of 60 °C to 95 °C in 0.5 °C increments, each for five seconds. The V3 to V4 region of the 16S rRNA gene was amplified through duplicate PCR alongside negative extraction controls, as previously described [20]. In brief, an initial PCR was completed to amplify the V3 to V4 region of the 16S rRNA gene using primers with Illumina overhang adapter sequences. After clean-up of combined PCR reactions, a second limited cycle PCR was completed to attach Illumina Nextera XT Index Primer 1 and 2 to allow multiplex sequencing. Following gel-based excision and purification, PCR products were quantified, and equimolar pools of sample libraries were sequenced, along with 20% PhiX DNA as a control for low diversity, on the Illumina MiSeq platform using MiSeq v3 reagents for a 2 × 300 bp run at the IBERS Translational Genomics Facility, Aberystwyth University. Sample reads were analysed using the QIIME 2 [42] pipeline as previously described [43]. In brief, reads were demultiplexed and truncated to 250 bp based on quality scores using FastQC [44] and underwent denoising, filtering, trimming, and chimera removal using DADA2 [45]. Taxonomic assignment of consensus operational taxonomic units (OTUs) was determined through classification against the Greengenes database (version 13.8) [46] using the open-reference vsearch at 97% sequence identify function. Resulting data matrices were analysed using MicrobiotaAnalyst [47], with samples C5 and A6 removed from the stool data set and samples D1 and B4 removed from the saliva data set due to low (<500) read number indicating failure of sequencing. OTUs were filtered based on a minimum count of 4 in 10% of samples and rarefied to minimum library size and scale using total sum scaling. The Shannon diversity measure was used for alpha-diversity analysis and principal coordinate analysis based on Bray–Curtis index was used for beta-diversity. For saliva and blood plasma, samples were thawed at 4 °C and 200 µL transferred to a sterile 2 mL microcentrifuge tube, to which 30 mg of <106 µM acetone-washed glass beads (Sigma-Aldrich) was added. To this, 1520 µL of HPLC-grade methanol and chloroform (4:1 v/v) was added. Freeze-dried stool samples were reconstituted at a concentration of 100 mg/mL in HPLC-grade water, methanol, and chloroform (2:5:2 v/v), and 30 mg of <106 µM acetone-washed glass beads was added. Samples were homogenised through vortex mixing for 5 s and then milling for 30 s at 5 Hz, after which they were shaken for 20 min at 4 °C and then stored at −20 °C for 20 min to precipitate protein. Samples then underwent centrifugation at 13,000× g for 6 min at 4 °C and the supernatant was transferred to a clean 2 mL microcentrifuge tube. Stool samples were diluted by 50% in the water, methanol, and chloroform mixture used in extraction in a clean 2 mL microcentrifuge tube. Samples were analysed using an LTQ linear ion trap (ThermoFisher Scientific, Waltham, MA, USA) as previously described [48], with 20 µL of sample injected into a water–methanol (7:3 v/v) solvent mix running at a flow rate of 60 µL/min. Data matrices were analysed using MetaboAnalyst [49] using total ion count normalisation, log transformation, and Pareto scaling. Tentative metabolite identifications were assigned using accurate mass matches to the Human Metabolome Database [50] with a 10 ppm cut-off. Unprocessed saliva and plasma and re-suspended stool samples in the solvent mix used for FIE-MS analysis were thawed at 4 °C from storage at −80 °C. Within each group, samples were randomised and 5 µL spotted onto duplicate 96 well silica transmission plates and dried at 55 °C for 30 min to remove residual solvent and water. Each plate was analysed using an Equinox 55 instrument (Bruker, Coventry, UK) with an attached HTS-XT microplate reader (Bruker) in transmission mode. The background level for each plate was determined by a blank well, and average values from over 32 readings were taken for each detected wavelength over the 4000–600 cm−1 region. The average value for each wavenumber for each duplicate reading was calculated and subjected to background subtraction and total intensity count normalisation. Data matrices were analysed using MetaboAnalyst [49] using log transformation and Pareto scaling. Univariate data sets (16S rRNA qPCR and pH) were analysed using Prism 8 software (GraphPad Software, San Diego, CA, USA) after means of duplicate readings and using one-way ANOVA with post-hoc Tukey’s test corrected for multiple comparisons. Univariate analysis completed in MetaboAnalyst was exported and formatted on a mass-to-charge/wavelength axis for visualisation in the R environment. A false discovery rate corrected p value threshold of less than 0.05 was used for the identification of significant univariate features in both MicrobiotaAnalyst and MetaboAnalyst analyses. Multi-omic correlations were completed between normalised data sets in the R environment using Bonferroni correction for multiple hypothesis correction. The effect of the TAWT expedition on the human microbiota and metabolome has been shown to be spatially different across bodily sites. Significant activity-associated changes in terms of both bacterial diversity and load were seen in saliva but not in stool, and participant differences in metabolite fingerprints persisted across all three sample types. Although the TAWT expedition was associated with considerable environmental and physiological stress, the observations noted in this work may also be associated with changes in the day-to-day conditions of the participants, in terms of close quarters and small group living. This may be of particular interest when considering the living conditions for long-term space travel. Further work beyond simulated exercises is needed to understand the overall interaction of these biological systems and how they impact host health and fitness and should form part of future studies in tracking responses to long-term space travel or their terrestrial proxies. It is further justified by this work to expand analysis beyond microbiota profiling to include broader measures of system function, including metabolomic profiling.

PMC10002535

Ioannis A. Voutsadakis

Characteristics and Prognosis of 8p11.23-Amplified Squamous Lung Carcinomas

21-02-2023

lung cancer,squamous,NSCLC,8p11 amplifications,NSD3,FGFR1,WHSC1L1

Background: Copy number alterations are common genetic lesions in cancer. In squamous non-small cell lung carcinomas, the most common copy-number-altered loci are at chromosomes 3q26-27 and 8p11.23. The genes that may be drivers in squamous lung cancers with 8p11.23 amplifications are unclear. Methods: Data pertaining to copy number alterations, mRNA expression and protein expression of genes located in the 8p11.23 amplified region were extracted from various sources including The Cancer Genome Atlas, the Human Protein Atlas and the Kaplan Meier Plotter. Genomic data were analyzed using the cBioportal platform. Survival analysis of cases with amplifications compared to nonamplified cases was performed using the Kaplan Meier Plotter platform. Results: The 8p11.23 locus is amplified in 11.5% to 17.7% of squamous lung carcinomas. The most frequently amplified genes include NSD3, FGFR1 and LETM2. Only some of the amplified genes present concomitant overexpression at the mRNA level. These include NSD3, PLPP5, DDHD2, LSM1 and ASH2L, while other genes display lower levels of correlation, and still, some genes in the locus show no mRNA overexpression compared with copy-neutral samples. The protein products of most locus genes are expressed in squamous lung cancers. No significant difference in overall survival in 8p11.23-amplified squamous cell lung cancers versus nonamplified cancers is observed. In addition, there is no adverse effect of mRNA overexpression for relapse-free survival of any of the amplified genes. Conclusion: Several genes that are part of the commonly amplified locus 8p11.23 in squamous lung carcinomas are putative oncogenic candidates. A subset of genes of the centromeric part of the locus, which is amplified more commonly than the telomeric part, show high concomitant mRNA expression.

Characteristics and Prognosis of 8p11.23-Amplified Squamous Lung Carcinomas Background: Copy number alterations are common genetic lesions in cancer. In squamous non-small cell lung carcinomas, the most common copy-number-altered loci are at chromosomes 3q26-27 and 8p11.23. The genes that may be drivers in squamous lung cancers with 8p11.23 amplifications are unclear. Methods: Data pertaining to copy number alterations, mRNA expression and protein expression of genes located in the 8p11.23 amplified region were extracted from various sources including The Cancer Genome Atlas, the Human Protein Atlas and the Kaplan Meier Plotter. Genomic data were analyzed using the cBioportal platform. Survival analysis of cases with amplifications compared to nonamplified cases was performed using the Kaplan Meier Plotter platform. Results: The 8p11.23 locus is amplified in 11.5% to 17.7% of squamous lung carcinomas. The most frequently amplified genes include NSD3, FGFR1 and LETM2. Only some of the amplified genes present concomitant overexpression at the mRNA level. These include NSD3, PLPP5, DDHD2, LSM1 and ASH2L, while other genes display lower levels of correlation, and still, some genes in the locus show no mRNA overexpression compared with copy-neutral samples. The protein products of most locus genes are expressed in squamous lung cancers. No significant difference in overall survival in 8p11.23-amplified squamous cell lung cancers versus nonamplified cancers is observed. In addition, there is no adverse effect of mRNA overexpression for relapse-free survival of any of the amplified genes. Conclusion: Several genes that are part of the commonly amplified locus 8p11.23 in squamous lung carcinomas are putative oncogenic candidates. A subset of genes of the centromeric part of the locus, which is amplified more commonly than the telomeric part, show high concomitant mRNA expression. Lung cancer is the most prevalent cancer worldwide and the leading cause of cancer death [1]. Two main types of lung cancers are distinguished histologically: non-small cell lung cancers (NSCLCs), which are the most common, and small cell lung cancers (SCLCs), which represent about 15% to 20% of the total lung cancers. NSCLC is divided into two main subtypes, adenocarcinomas and squamous lung cancers, which have distinct molecular pathogenesis. Based on molecular abnormalities, some adenocarcinomas are currently treated with targeted therapies against EGFR, ALK or ROS kinases [2]. Immunotherapies inhibiting CTLA-4 or the PD-L1/PD-1 ligand/receptor pair are also effective for subsets of patients with both subtypes of NSCLC [3,4]. Besides immunotherapies, no other targeted therapies currently exist for squamous cell lung cancers. Thus, there is a need for further rationally developed targeted therapies for this type of lung cancer based on molecular defects. The molecular landscape of squamous cell lung cancer has been elucidated by TCGA and consists of recurrent mutations in 11 genes and a mean of 360 exonic mutations per tumor as well as a mean of 323 copy number alterations [5]. Copy number alterations (CNAs), both gains and losses, are common molecular lesions in cancer and complement mutations and epigenetic changes in neoplastic pathogenesis [6]. Recurrent CNAs may be promoted by increasing the expression of oncogenes or leading to tumor suppressor losses. However, often, the gained or lost region contains several genes and most of them are passenger alterations with no pathophysiologic benefit to the cancer cell. In many cases of recurrent CNAs, the possible driver oncogene(s) or tumor suppressor(s) is not well defined. In squamous NSCLC, the most commonly recurrently amplified area, in about 40% of cases, is at chromosome locus 3q26-27, which includes the oncogene SOX2, a transcription factor and member of a panel of factors that are able to reprogram differentiated cells to pluripotent stem cells [7]. SOX2 is involved in lung organogenesis and squamous differentiation [8]. Other oncogenes located in this locus include PIK3CA and MECOM, encoding for EVI1. Another amplified area in squamous NSCLC is at chromosome 8p11.23, which is the focus of this report. This locus is amplified in about one in six squamous NSCLCs and has been also reported to be amplified in breast cancers and urothelial bladder carcinomas [9,10,11]. Publicly available genomic data pertaining to the squamous subtype of NSCLC from The Cancer Genome Atlas (TCGA) were extracted using the cBioCancer Genomics Portal (cBioportal, http://www.cbioportal.org, last accessed: 12 November 2022), a site that allows for interrogation of data for genetic alterations such as mutations and CNAs as well as mRNA expression of any gene of interest [5,12]. CNAs are computed in TCGA using the Genomic Identification of Significant Targets in Cancer (GISTIC) algorithm. The algorithm assigns a putative amplification status to genes with a score of 2 or higher. A global aneuploidy status of each case is also provided in TCGA by an aneuploidy score (AS) representing the sum of chromosomal arms with gains or losses in Affymetrix 6.0 SNP arrays. The definition of a CNA of a whole chromosome arm for the calculation of AS consists of somatic copy number alterations in more than 80% of the total arm length as determined by the ABSOLUTE algorithm [13]. In contrast, chromosomal arms with alterations in less than 20% of their total length are defined as not copy-number-altered, while chromosome arms with alterations involving 20% to 80% of their length are not called, and thus considered not altered, according to the algorithm. The RSEM algorithm was used for normalization of mRNA expression [14]. The Human Protein Atlas (www.proteinatlas.org, last accessed: 19 November 2022), a publicly available database of protein expressions in human normal and neoplastic tissues, was interrogated for the expression of proteins of genes located at chromosome location 8p11.23 in squamous lung cancer [15]. The Human Protein Atlas contains data from a semi-quantitative immunohistochemistry-based evaluation of the expression of proteins in human tissues. On many occasions, evaluations have been carried out with several different commercially available antibodies for each protein. The overall survival (OS) of 8p11.23-amplified squamous lung cancers (defined as NSD3 amplified per the GISTIC algorithm definition) and nonamplified squamous lung cancers was determined in the TCGA cohort. The prognosis of squamous lung cancer patients according to the mRNA expression level of each of the 8p11.23 genes and association with relapse-free survival (RFS) was tested using data from a series contained in the online publicly available platform Kaplan Meier Plotter (www.kmplot.com, last accessed: 18 November 2022) [16]. The cut-off of amplified and nonamplified samples for each gene was set at the higher quartile of amplification, as the closer cut-off to the percentage of lung cancer cases with 8p11.23 locus amplifications, provided by the kmplot platform. Statistical comparisons of categorical and continuous data were carried out with Fisher’s exact test or the χ2 test and the t test. The log-rank test was used to compare Kaplan–Meier survival curves. All statistical comparisons were considered significant if p < 0.05 except for the RFS survival analysis according to mRNA expression levels of different genes, which was considered significant at a p < 0.0005 level to account for multiple comparisons. Squamous lung cancers with amplification of the 8p11.23 locus, as defined by NSD3 amplification, do not differ from non-8p11.23-amplified carcinomas in the mean age of patients at presentation, the percentage of patients older than 65 years old or in the sex and race distribution (Table 1). The stage at diagnosis is also similar between the two groups. Genes located at the 8p11.23 locus are amplified in 11.5% to 17.7% of squamous NSCLCs (Table 2). A higher frequency of amplification is observed in the most centromeric parts of the locus with genes NSD3, FGFR1 and LETM2 showing a higher number of amplified cases in TCGA (Figure 1). Genes in the central and more telomeric parts of the locus display progressively lower frequencies of amplification, with the most telomeric genes ERLIN2 and ZNF703 being amplified in 60% to 70% of NSD3-amplified cases (Figure 1). Conversely, in NSD3 nonamplified cases, the rest of the genes of the locus are amplified only rarely, in isolated cases (not shown). The global copy number alteration burden of 8p11.23-amplified and nonamplified squamous NSCLC as measured by the AS was not different, with most cases in both groups having an intermediate AS between 4 and 24 (Figure 2). No cases in the amplified group had an AS below 4, but even in the nonamplified group, only 4% of cases had an AS below 4. The mean AS of the amplified group was 16.29 (SD: 6.45), and the mean AS of the nonamplified group was 16.08 (SD: 6.64, unpaired t test p = 0.78). Individual chromosome arms with more frequent gains in the 8p11.23-amplified group compared to the nonamplified group included 8q (45.3% in the 8p11.23-amplified group versus 31.4% in the nonamplified group, p = 0.01) and 11q (16.3% versus 8%, respectively, p = 0.02) (Figure 3A). In contrast, arm 8p showed no gains in any of the cases in the 8p11.23-amplified group and it was gained in 6% of the nonamplified group. Chromosome arms with losses occurring more frequently in 8p11.23-amplified squamous NSCLC included 8p (72.1% versus 48.6% in the nonamplified group, p = 0.0001) and 5q (84.9% versus 68.6% in the nonamplified group, p = 0.003) (Figure 3B). Chromosome arm 11q was more frequently (but not statistically significantly) lost in the 8p11.23 nonamplified group (26.4% versus 16.3% in the amplified group, p = 0.06). The tumor mutation burden (TMB) was similar in 8p11.23-amplified and nonamplified squamous NSCLC, with about half of the patients in both groups having a TMB between 200 and 500 mutations and about 5% of the amplified group and a slightly higher percentage of the nonamplified group presenting a TMB above 500 (Figure 4). The frequency of the two categories with mutation numbers above 200, which display a higher probability of responses to immunotherapy with immune checkpoint inhibitors, did not differ significantly between the 8p11.23-amplified and nonamplified groups (Fisher’s exact test p = 0.9). The mean mutation number of the 8p11.23-amplified group was 271.9 (SD: 152.8) and did not differ from the mean mutation number of the nonamplified group, which was 270.9 (SD: 199.6, unpaired t test p = 0.9). Among individual oncogene mutations, TP53 mutations were more common in the 8p11.23-amplified group (90.7% versus 81.9% in the nonamplified group, Fisher’s exact test p = 0.05), while the master hypoxia response transcription factor NFE2L2 was more often mutated in the nonamplified group but not statistically significant (16.1% versus 9.3% in the 8p11.23-amplified group, Fisher’s exact test p = 0.13). The prevalence of PIK3CA mutations in the amplified group (8.1%) was also not different from the nonamplified group (11.6%, Fisher’s exact test p = 0.44). Figure 5 shows the percentage of mutations in the most frequently mutated oncogenes in squamous NSCLC in the two groups. Another chromosome area that is most frequently amplified in squamous NSCLC is 3q26, which harbors oncogenes SOX2, PIK3CA and MECOM. An evaluation of 8p11.23-amplified and nonamplified samples disclosed that SOX2, PIK3CA and MECOM genes are coamplified in similar percentages of cases (40.7% versus 39.7% for SOX2, 38.4% versus 35.2% for MECOM and 38.4% versus 37.7% for PIK3CA, respectively). Increased mRNA expression of the amplicon genes in amplified cases correlates with gene amplification in several genes but not in others. The most frequently amplified gene of the amplicon NSD3 shows higher mRNA overexpression in over 90% of the amplified samples. In contrast, the two neighboring genes FGFR1 and LETM2, which are almost invariably coamplified, are overexpressed at the mRNA level in about half or less of the amplified cases (Figure 6). Four other genes that show high mRNA expression (over 70% of amplified cases with a z score above 2 compared with diploid samples) include PLPP5, DDHD2, LSM1 and ASH2L. Four genes, ADGRA2, GOT1L1, ADRB3 and STAR, are not overexpressed in any amplified cases. Besides ADGRA2, ADRB3 and STAR, which are not expressed, protein products of genes of 8p11.23 are in general expressed in squamous NSCLC, at least with one antibody checked (Table 3). However, there was significant variability depending on the antibody used. Although the OS of the 8p11.23-amplified squamous lung carcinoma group in TCGA was better than the OS of the nonamplified group, this difference did not reach statistical significance (log-rank p = 0.12, Figure 7). Relapse-free survival (RFS) of squamous NSCLC patients in the higher quartile of mRNA expression of NSD3 was no different from counterparts in the three lower quartiles (not shown). Surprisingly, the RFS of patients in the higher quartile of mRNA expression of FGFR1 expression was improved compared with patients in the three lower quartiles (not shown). All other genes besides PLPP5, which showed better RFS in the mRNA overexpressed group, showed no significant differences in RFS between groups (not shown). Chromosomal locus 8p11.23 is the second most frequently amplified locus in the squamous histology of lung cancers, and squamous NSCLC is a type of cancer with a higher frequency of amplifications in this locus among all cancer histologies and primary locations. In contrast, adenocarcinomas of the lung display amplifications in this chromosomal locus at a lower frequency (about 2.5% to 3% of cases in TCGA lung adenocarcinoma study). Genes located at 8p11.23 include receptor tyrosine kinase FGFR1; two methyl-transferases, ASH2L, which is part of the mixed lineage leukemia (MLL) complex, and NSD3; two phospholipid phosphatases, DDHD2 and PLPP5; and proteins ZNF703 and BRF2, which are transcription regulators (Table 1). Two proteins, EIF4EBP1 and LSM1, located at 8p11.23 are involved in mRNA translation and metabolism. A list of additional genes amplified in squamous NSCLC is shown in Table 1. Previous studies have examined the implications of some 8p11.23 genes in lung cancer. ZNF703 is a transcription factor with roles in development and in ER-positive breast cancers where it is associated with more aggressive subsets [17]. In lung cancer, samples with ZNF703 amplification displayed variable mRNA overexpression, suggesting an imperfect correlation [18]. Another transcription regulator from 8p11.23, BRF2, is a subunit of transcription factor TFIIIB [19]. TFIIIB co-operates in transcription guided by RNA polymerase III, the polymerase transcribing tRNA genes. Thus, BRF2 plays an important role in the regulation of protein synthesis, with implications for proliferating cancer cells. In lung cancer, pathways upregulating BRF2 have been shown to favor cancer progression [20,21]. BAG4, a protein transcribed from a gene at 8p11.23 with a role in the inhibition of apoptosis, has been shown to transform breast cells and may have functional implications for lung cancer, being commonly coamplified with FGFR1 and NSD3 [22,23]. The current study examines the 8p11.23-amplified area and the nineteen genes that are located at this chromosomal locus in squamous NSCLC. The analysis based on published genomic data shows that the genes of the locus are amplified en bloc in the majority of amplified cases, while in fewer cases, only a subset of genes at 8p11.23 are amplified. The higher frequency of amplification among the genes of the 8p11.23 locus is observed in the most centromeric genes including NSD3, LETM2 and FGFR1. The amplification of 8p11.23 has no significant influence on the TMB or the AS of the cases, suggesting that genes in the locus are not involved in aneuploidy or DNA repair mechanisms of the cancer cells. In addition, no influence of 8p11.23 amplification on the prevalence of the other frequent CNA in squamous NSCLC, the amplification of 3q26 is observed. mRNA expression of the amplified genes is variable, with a higher correlation of amplification and overexpression observed in several genes in the most centromeric part of the locus (NSD3, PLPP5, DDHD2, LSM1 and ASH2L) and also in a few genes that are located toward the telomeric end of the area (BRF2 and PLPBP). Several other genes show lower or no increase in mRNA expression in amplified cases. Interestingly, increased mRNA expression was not associated with worse patient RFS for any of the genes amplified at 8p11.23. The caveat of this survival analysis is that, in the increased-mRNA-expression group, up to one-third of patients may be nonamplified, as the cut-off was at the highest quartile of expression. These data suggest that genes amplified at 8p11.23 do not confer survival or other cancer-related process benefit in squamous NSCLC but may be amplified as part of an underlying defect that makes the locus prone to repeated DNA replication. However, given the limitations of the survival analysis, it cannot be totally excluded that a survival benefit of increased copy numbers of one or more 8p11.23 genes exists for squamous NSCLC cells. The centromeric area of the 8p11.23 amplicon that contains genes NSD3, LETM2, FGFR1 and TACC1 and presents a higher amplification frequency in squamous NSCLC is homologous to an area at human chromosome 4p that contains genes related to each of the 8p11 genes. These include NSD2, LETM1, FGFR3 and TACC3. This 4p area, although rarely amplified, it is the site of fusions between FGFR3 and TACC3 in a small percentage (1.2%) of squamous lung carcinoma cases, while FGFR1 and FGFR2 fusions are observed in isolated cases [5]. In bladder cancer, which is also a type of cancer that harbors 8p11.23 amplifications, the same fusions are present in the homologous 4p location. In addition, bladder cancers present mutations of FGFR3, which make them susceptible to treatment with FGFR inhibitors [24]. The presence of genetic abnormalities in these homologous regions harboring genes with similar functions in the same cancers would argue for a functional benefit conferred by one or more genes. The prime candidate in this case is obviously FGFR family members for which clinical data for the efficacy of their inhibition in other cancers exist, implying functional dependence of cancers carrying FGFR genetic lesions, at least in some cases. Related to FGFR1 targeting in 8p11.23-amplified squamous lung carcinomas, it is worth noting that, as shown here, amplification is not always associated with mRNA overexpression. This is important when considering FGFR1 amplifications as potential biomarkers for the development of treatments with FGFR inhibitors or inhibitors of the downstream cascades such as PI3K inhibitors [25,26]. An improved understanding of recurrent copy number alterations in cancer and the genes that are affected by them as well as the characterization of candidate driver genes in altered areas present therapeutic opportunities. The proof of principle has already been provided decades ago when amplifications of 17q in a subset of breast cancer genes were revealed to harbor ERBB2 amplifications and became the basis for effective therapies that have changed the outcomes of HER2-positive breast cancers [27,28]. A similar opportunity may arise with 8p11.23 amplifications and the development of FGFR inhibitors. Clinical translation will require a better grasp of driver genes in the locus and clarification of additional drivers. Methyltransferase NSD3 is a serious candidate and it has been shown to promote squamous lung carcinoma proliferation in experimental models in vitro and in vivo by promoting the transcription of oncogenes through lysine methylation of histone 3 at position 36 [29,30]. Pathways activated by driver genes creating tumor dependency would be important to elucidate. For FGFR receptors, for example, transactivation of additional tyrosine kinase receptors may be at play and could have implications for inhibition effectiveness in cancers with defects in these other tyrosine kinase receptors such as HER2 amplifications [31]. The development of companion diagnostics for the measurement of the amplification with the most effective cut-off for clinical efficacy will also be of paramount importance as the case of HER2 has shown.

PMC10002540

Chiara Puricelli,Elena Boggio,Casimiro Luca Gigliotti,Ian Stoppa,Salvatore Sutti,Mara Giordano,Umberto Dianzani,Roberta Rolla

Platelets, Protean Cells with All-Around Functions and Multifaceted Pharmacological Applications

26-02-2023

platelets,inflammation,platelet derivatives,extracellular vesicles

Platelets, traditionally known for their roles in hemostasis and coagulation, are the most prevalent blood component after erythrocytes (150,000–400,000 platelets/μL in healthy humans). However, only 10,000 platelets/μL are needed for vessel wall repair and wound healing. Increased knowledge of the platelet’s role in hemostasis has led to many advances in understanding that they are crucial mediators in many other physiological processes, such as innate and adaptive immunity. Due to their multiple functions, platelet dysfunction is involved not only in thrombosis, mediating myocardial infarction, stroke, and venous thromboembolism, but also in several other disorders, such as tumors, autoimmune diseases, and neurodegenerative diseases. On the other hand, thanks to their multiple functions, nowadays platelets are therapeutic targets in different pathologies, in addition to atherothrombotic diseases; they can be used as an innovative drug delivery system, and their derivatives, such as platelet lysates and platelet extracellular vesicles (pEVs), can be useful in regenerative medicine and many other fields. The protean role of platelets, from the name of Proteus, a Greek mythological divinity who could take on different shapes or aspects, is precisely the focus of this review.

Platelets, Protean Cells with All-Around Functions and Multifaceted Pharmacological Applications Platelets, traditionally known for their roles in hemostasis and coagulation, are the most prevalent blood component after erythrocytes (150,000–400,000 platelets/μL in healthy humans). However, only 10,000 platelets/μL are needed for vessel wall repair and wound healing. Increased knowledge of the platelet’s role in hemostasis has led to many advances in understanding that they are crucial mediators in many other physiological processes, such as innate and adaptive immunity. Due to their multiple functions, platelet dysfunction is involved not only in thrombosis, mediating myocardial infarction, stroke, and venous thromboembolism, but also in several other disorders, such as tumors, autoimmune diseases, and neurodegenerative diseases. On the other hand, thanks to their multiple functions, nowadays platelets are therapeutic targets in different pathologies, in addition to atherothrombotic diseases; they can be used as an innovative drug delivery system, and their derivatives, such as platelet lysates and platelet extracellular vesicles (pEVs), can be useful in regenerative medicine and many other fields. The protean role of platelets, from the name of Proteus, a Greek mythological divinity who could take on different shapes or aspects, is precisely the focus of this review. The assumption that platelets are involved only in primary hemostasis and regulation of blood flow was put into question a long time ago, and now there is evidence that these cell fragments have a distinctive role in the immune response [1]. The description of platelet-leukocyte micro-aggregates dates back to the 1990s [2], and all nine Toll-like receptors (TLRs) described in humans are on platelets [3,4], supporting the hypothesis of platelet involvement at least in innate immunity. Moreover, platelets display Fcγ immunoglobulin G receptors (FcγR) [5,6] and CD40 ligand (CD40L), mediating the interaction with CD40 on B cells, dendritic cells, and macrophages, and they are able to sample the surrounding environment and present foreign pathogens and molecules to T cells [4,7]. All these attributes imply a bridge between innate and adaptive immunity so that platelets are not only involved in first protection against foreign antigens but also take part in the entire immune response in a much more comprehensive pattern. When induced by pathogen-associated molecular patterns (PAMPs) during infections or by cell-damage-associated molecular patterns (DAMPs), TLRs can initiate intracellular signaling leading to platelet activation, to promote not only primary hemostasis but also the immune response. For instance, lipopolysaccharide (LPS)-mediated induction of TLR4 can promote the formation of platelet-neutrophil aggregates where platelets play a role as inducers of neutrophil extracellular traps (NETs). NETs consist of a network of histones, chromatin, and degradation enzymes released by neutrophils during a unique type of cell death (NETosis), and their main goal is pathogen entrapment and elimination through oxidative and non-oxidative mechanisms, thus limiting their diffusion in the bloodstream [1,8]. The complex formed by platelets plus NETs thus serves as a scaffold to bind and eliminate pathogens on one side and amplify platelet activation on the other [1]. Furthermore, the phenomenon of autophagy has been detected also in activated platelets [9]. Several platelet surface molecules and receptors have been associated with their ability to interact with immune cells, including P-selectin recognizing P-selectin glycoprotein ligand-1 (PSGL-1) on lymphocytes, neutrophils, and monocytes [1,10]. Even in the absence of foreign pathogens, platelets can take part in sterile inflammatory reactions underlying several pathological processes, such as the multi-step evolution of an atherosclerotic plaque. One of the most interesting features of platelets is the wide number of biologically active molecules contained in their granules. Platelets contain two main types of secretory organelles, α granules, and dense bodies (δ granules), and most effector functions depend on their secretion. As a consequence of granule fusion with the platelet plasma membrane, several granule molecules may be expressed on the platelet surface or released as soluble molecules (e.g., coagulation factors, mitogenic factors, angiogenic mediators, and chemokines) acting locally at sites of vascular injury or even systemically [1]. Proteomic studies indicate that α-granules release more than 300 soluble proteins acting in processes such as blood coagulation, inflammation, immunity, cell adhesion and growth, and possibly other less-known activities [11]. First of all, α-granules contain many mediators of blood coagulation such as fibrinogen, von Willebrand factor (VWF), and adhesive proteins that mediate platelet-platelet and platelet-endothelial interactions. VWF of α-granules constitutes 20% of the total VWF protein, mainly in the high-molecular-weight forms [12]. Moreover, α-granules also contain components of the VWF receptor complex (GPIbα-IX-V), the main receptor of fibrinogen (integrin αIIbβ3), and the collagen receptor (GPVI) [13,14]. These receptors are constitutively expressed in resting platelets both on the plasma membrane and in α-granules (containing two-thirds of the whole αIIbβ3 and one-third of GPVI) and are upregulated on the membrane of activated platelets upon α-granule secretion. The key molecules in α-granules are the basis of several coagulation pathways, secreted by activated platelets and involved in secondary hemostasis: Factor V, endocytosed from the plasma and stored in α-granules as activated factor V (FVa) complexed to the carrier protein; Factor XI and XIII, synthesized in megakaryocytes and stored in α-granules; Factor II prothrombin; high molecular weight kininogens (HMWK), involved in the intrinsic clotting cascade; and plasminogen activator inhibitor-1 (PAI-1) and α2 antiplasmin, which are protease inhibitors limiting plasmin-mediated fibrinolysis [15]. α-Granules may also contribute to hemostatic balance by releasing several proteins limiting the progression of coagulation: antithrombin III, inhibiting both the intrinsic and extrinsic pathways, C1-inhibitor, degrading plasma kallikrein, factor XIa, and factor XIIa, and protein S and tissue factor pathway inhibitor (TFPI). Moreover, they store the fibrinolytic proteinase plasmin and its inactive precursor plasminogen [15]. Activated platelets externalize the anionic phospholipid phosphatidylserine (PS) required to support all coagulation reactions, and produce P-selectin and CD40L (see below) stimulating monocyte production of tissue factor (TF), mostly bound to microvesicles, which bind and fuse with platelets to initiate coagulation on the platelet surface [16]. Therefore, platelets can contribute to both anti- and pro-coagulant activities, but it is not known whether these opposite functions are ascribable to distinct platelet subsets. Moreover, α-granules contain several chemotactic factors, including platelet factor-4 (PF4), β-thromboglobulin, epithelial neutrophil-activating peptide 78 (ENA-78), growth-related oncogene-α (GROα), regulated upon activation, normal T-cell expressed and secreted chemokines (RANTES), monocyte chemotactic protein 1 (MCP-1), macrophage inflammatory protein 1α (MIP-1α), and thymus- and activation-regulated chemokine (TARC), playing key roles in inflammation by their ability to recruit and activate several types of leukocytes. PF4 induces neutrophil activation and β2-integrin-mediated adhesion to endothelial cells [17,18]. ENA-78 is a potent chemokine for neutrophils, GROα and MCP-1 for monocytes, RANTES for eosinophils, monocytes, and T lymphocytes, and MIP-1α for monocytes, T and B lymphocytes, NK cells, basophils, and eosinophils. TARC is a selective chemoattractant for T cell subsets expressing a class of receptors binding TARC with high affinity and specificity [19,20,21,22]. MIP-1α and TARC may play a role in atherosclerotic plaque destabilization and atopic dermatitis, respectively. Besides chemoattractants, α-granules also store other types of immunomodulatory molecules, such as CD40L, triggering receptor expressed on myeloid cells-1 (TREM-1) ligand, and transforming growth factor (TGF)-β1. CD40L (CD154) is known as a costimulatory receptor expressed on activated T cells, binding to CD40, and expressed on the surfaces of B cells, endothelial cells, and dendritic cells. The CD40L/CD40 interaction plays a bidirectional role in the costimulation of both T cells and their CD40-expressing partner and in inducing immunoglobulin (Ig) isotype switching in activated B cells [1]. However, CD40L can also be produced in a soluble form (sCD40L) due to metalloproteinase-dependent cleavage of the membrane form or alternative splicing of the CD40L RNA. Besides CD40, CD40L can also bind to the integrins GPIIb/IIIa, α5β1 (CD49e/CD29), and αMβ2 (CD11b/CD18, Mac-1). Platelets are the main source of circulating sCD40L, which is involved in the inflammatory and prothrombotic responses: plasma levels of sCD40L are routinely used as a systemic marker of platelet activation. In several cohorts, plasma sCD40L predicts clinical cardiovascular adverse events. sCD40L levels are increased in plasma from patients with sickle cell disease, likely reflecting platelet activation [23]. TREM-1, a member of the V-type immunoglobulin superfamily, is constitutively expressed by neutrophils and monocytes [24]. Engagement of TREM-1 by its ligand expressed by platelets stimulates an oxygen burst and IL-8 production in neutrophils [25]. TGF-β is stored in large amounts in platelet α granules, and platelet storage seems to be important for maintaining circulating levels of this potent immunoregulatory factor. All of the several α-granule mediators described above influence inflammation by exerting proinflammatory and immunomodulatory activities, mainly by recruiting and activating several leukocyte types. Platelets also contribute to the inflammatory process by expressing receptors that facilitate the adhesion of platelets to other cells. Most α-granule membrane-bound proteins are also present on the plasma membrane of resting platelets, including integrins (e.g., αIIb, α6, β3), immunoglobulin family receptors such asglycoprotein VI (GPVI), Fc receptors, platelet endothelial cell adhesion molecules (PECAMs), leucine-rich repeat family receptors (e.g., GPIb-IX-V complex), tetraspanins (e.g., CD9), and other receptors (CD36, Glut-3) [26]. Platelet activation induces upregulation of these molecules on the membranes of activated platelets. Several other α-granule membrane-associated proteins are not expressed on the surfaces of resting platelets and are expressed only upon activation, including the integral membrane proteins fibrocystin L, CD109, and P-selectin, so they are markers of activated platelets. In particular, P-selectin mediates platelet interaction with endothelial cells, monocytes, neutrophils, and lymphocytes by binding to PSGL-1, and it promotes their recruitment to sites of inflammation. Then, the platelet α-granule proteins fibrinogen, fibronectin, vitronectin, and VWF contribute to stabilize platelet-endothelial adhesion by forming cross-bridges between GPIIb-IIIa and the endothelial αVβ3 integrin or ICAM-1 [12]. Dense granules are nearly 10-fold less abundant than α-granules in human platelets. They contain many small molecules and comparatively fewer proteins. Among the former, key roles are played by ADP and ATP (650 mM and 440 mM, respectively), uracil and guanine nucleotides, calcium and potassium, and bioactive amines such as serotonin and histamine [27]. Platelet dense granules contain high concentrations of polyphosphates, whose release activates factor FXII and, in turn, the kallikrein-kinin system, resulting in the generation of bradykinin, inducing increased vascular permeability and edema in vivo [28,29]. The dense granule membrane proteins include CD63 (granulophysin) and LAMP-2. Several platelet plasma membrane proteins have also been identified in dense granule membranes, including GPIb and αIIbβ3. Although platelets are traditionally recognized for their central role in hemostasis, the presence of chemotactic factors, chemokines, adhesion, and costimulatory molecules in their granules and membranes indicates that they may play an immunomodulatory role in the immune response besides their capacity to trigger blood coagulation and inflammation [30]. Therefore, platelets perform a sentinel role at the sites of vascular injury, which may be crucial in regulating not only the inflammatory response but also the adaptive immune response, taking part in pathogen clearance and tissue repair [1]. Platelets themselves possess therapeutic properties that have been exploited in recent years, especially in the field of regenerative medicine, and that show promising potential for the future. Moreover, increased life expectancy has made the aging population grow in the last few decades, and chronic disorders to become more and more common, so the regenerative potential of platelets and platelet derivatives has started to be investigated more deeply. Platelet granules are a powerful source of platelet growth factors (PGFs), ranging from more platelet-specific molecules to bioactive mediators shared by other cells involved in the regenerative cascade (endothelial cells, macrophages, neutrophils, and fibroblasts). Their action is mediated by the interaction with tyrosine kinase receptors and subsequent activation of multifunctional intracellular signaling cascades [31,32,33]. A list of the main growth factors released by platelets upon their activation is shown in Table 1. Briefly, their broad functions can be summarized in four categories: (1) mitogenesis and cell differentiation; (2) chemotaxis and migration, with important implications in angiogenesis and re-epithelialization; (3) regulation of the inflammatory response; and (4) extracellular matrix (ECM) formation and remodeling [34,35,36,37,38,39,40,41,42,43]. With these premises, it is not surprising that transfusion and regenerative medicine have come to an agreement characterized by the advent of hemocomponents “not for transfusion use”, such as platelet concentrates (PCs), exploiting the properties of what has been addressed as the “secretome” of platelets [36]. PCs have been explored in several clinical fields, including wound healing, treatment of osteoarticular lesions, dry-eye syndrome, corneal ulcers, and many others. PCs are autologous or allogeneic platelet derivatives with a platelet concentration higher than the blood baseline (1.50–3.50 × 1011/L) [34]. Data are conflicting regarding the PC’s optimal platelet concentration, and it must be argued that the platelet count is not necessarily an index of efficacy unless platelets are also of good quality. Moreover, excessive concentrations of platelets may even be detrimental since very high concentrations of some PGFs, such as TGF-β, may have antiproliferative effects or induce counterproductive proteolysis in the ECM [44,45]. Weibrich et al. suggested an optimal concentration of around 1 × 106/uL [46], which is now the working definition of platelet-rich plasma (PRP) [39]. In general, a PC platelet count that is four to five times higher than the basal count in whole blood seems to be the best compromise [35]. Platelet concentration is also influenced by the size of donor platelets and the donor hematocrit, since approximately 20% of platelets remain adsorbed in the red blood cells (RBC) pellet during PRP preparation [39]. PCs have been classified depending on their leukocyte, platelet, and fibrin content, and properties such as gelification, processing techniques, and possible applications. The main terms used include pure platelet-rich plasma (P-PRP), leukocyte and platelet-rich plasma (L-PRP), pure platelet-rich fibrin (P-PRF) and leukocyte and platelet-rich fibrin (L-PRF), platelet gel (PG), platelet lysate (PL), and, when PCs are used in the form of collyrium, serum eye drops (E-S), and PRP eye drops (E-PRP) [34,35,37,40,47,48]. The main advantage derived from their use is that, instead of single recombinant growth factors, PCs offer a broad range of bioactive molecules often acting synergistically and improving the overall treatment efficacy [35,40]. The processing steps and the peculiar characteristics of each PC are presented in detail in Table 2, while Figure 1 shows a series of images exemplifying the technical preparation of P-PRP, L-PRP, and PRF underlying the differences between the three. The choice between autologous or donor-derived PCs mainly depends on ethical and safety issues as well as patient-related conditions. Besides being better-accepted by patients, autologous PCs are devoid of the risk of infection related to contamination of blood products. On the other hand, autologous PCs suffer from large variability in the quality of platelets since patients frequently have underlying comorbidities compared with healthy donors whose blood components are also prepared through standardized procedures [37]. Moreover, autologous PC use might not be feasible if the patient is being treated with drugs affecting platelet function that cannot be suspended, such as aspirin or other anti-aggregating agents often used in older subjects with underlying cardiovascular disorders [63]. Autologous PRP was first used in cardiac surgery by Ferrari et al. in 1987, but at that time only to avoid the use of homologous blood products during intraoperative blood salvage techniques [64]. Cardiac surgery was again a field of application in the early 1990s as a fibrin sealant [65,66]. Most studies have been conducted in the fields of maxillofacial surgery, dentistry [53,54], wound care [67,68], dermatology and esthetic medicine [69], orthopedics [70], ophthalmology [71], and neurology [72,73] (Figure 2). Chronic wound healing was probably one of the first “atypical” applications of platelets as a treatment. While acute wounds restore the anatomical and functional integrity of the lesion in a relatively short time and well-organized process, chronic and complicated wounds deviate from the traditional healing cascade, resulting in chronic inflammation and incomplete or prolonged reconstitution of the lesion’s integrity due to disturbing local and/or systemic factors such as concomitant infection, diabetes, underlying malignancies, and chronic inflammatory disorders [74]. Platelets participate in wound healing from the very beginning of the four-step process (coagulation-inflammation-proliferation-remodeling [74]), i.e., from the formation of the platelet plug, representing the first step in the hemostatic cascade. However, their role is not limited to blood loss prevention. Instead, the release of platelet granule content represents a key step influencing whole-tissue remodeling. In the first few days of wound healing, platelets modulate inflammatory events at the wound site mainly by promoting leukocyte chemotaxis. Angiogenesis, fibroblast proliferation, and ECM deposition through collagen synthesis and regulation of collagenase production characterize the next steps, and it is especially in this scenario that PGFs play a major role. The first clinical application of PRP was in 1991 in the setting of chronic leg ulcers, where the new formation of vascularized connective tissue could be demonstrated [75]. Since then, several trials have been conducted on human subjects, and a 2011 meta-analysis fully investigating the results obtained in the previous ten years agreed on the key role of PRP in accelerating the healing process, in both difficult-to-treat wounds and chronic ulcers such as those affecting diabetic patients. Furthermore, it seems that PRP also bears antimicrobial activity, which certainly accelerates the healing process by resisting local infections [76]. Just as in wound healing, bone repair also implies a tripartite process consisting of inflammation, proliferation, and remodeling. Here, the synergistic effect of platelet-derived TGF-β, PDGF, EGF, and b-FGF creates the most adequate microenvironment to sustain bone healing. In particular, they are primarily involved in osteoinduction by promoting differentiation of osteogenic precursors and mitogenesis, and, when bone grafts are used, they favor the process of osteoconduction by promoting vascular and cellular migration along a proper scaffold [37]. Starting from one of the most common orthopedic complaints, especially of older patients, i.e., osteoarthritis, PRP can support a cost-effective, low-intensity but high-quality therapeutic program where PRP shows beneficial effects on joint physiology, including chondrocyte proliferation, ECM production, and suppressed catabolism as well as an anti-inflammatory effect [70]. In addition, the osteoinductive properties of PCs have extended their applications beyond the field of orthopedics [77], broadly covering also maxillofacial surgery and dentistry [53,54,78]. Indeed, the osteoinductive properties of platelet derivatives improve the integration of dental implants, providing better grafting results. The versatility of PCs has allowed the introduction of collyrium preparations that have proved to be beneficial in the treatment of ocular surface diseases such as dry eye syndrome or corneal ulcers [71], and even macular holes [79]. E-PRP functions as a lubricant and was shown to decrease inflammation in patients affected by disorders of the tear film and to accelerate the healing of corneal ulcers, and it is likely a promising alternative to topical steroids, which can quickly relieve symptoms, but whose long-term usage may result in side effects such as an increased risk of infection, cataract formation, and increased intraocular pressure [71]. Furthermore, compared to pure autologous serum, whose usage dates back to the 1980s [80], E-PRP seems to provide a slightly higher benefit, probably due to its platelet content, which allows prolonged release of growth factors ensuring a longer-lasting effect [71]. In addition to their regenerative potential, platelet derivatives may have analgesic properties as well. Data on the topic are conflicting [81,82,83,84,85,86], but studies showing a benefit in pain control suggest that the underlying mechanisms might be either the release of dense granule-derived serotonin, a well-known anti-nociceptive mediator in pain pathways, or an indirect analgesic effect due to the enhancement of nerve repair and re-myelination, thus eliminating the source of neuropathic pain [37,84,85]. A discussion on all the other possible clinical applications of PCs is beyond the scope of this review. However, Supplementary Table S1 shows a complete list of the interventional and observational studies conducted since 1999. These results have been obtained after a search on clinicaltrials.gov using the keywords “platelet gel”, “platelet lysate”, “platelet derivatives”, “platelet-rich plasma”, “platelet-rich fibrin”, “serum eyedrop”, and “platelet-rich plasma eyedrop”. A total of 920 interventional and 41 observational studies have been identified, mostly in the fields of orthopedics, dentistry, and maxillofacial surgery. The greatest hindrance to the use of PCs is the scarce harmonization among studies, already starting from the terminology used to address the different platelet derivatives and encompassing a lack of standardization in preparation protocols, which may deeply influence the product’s quality in terms of platelet count, growth factor type and concentration in the final product, and, eventually, therapeutic efficacy [47,48]. An important issue is the centrifugation protocol, since high centrifugation forces may prematurely activate platelets resulting in a low final content of PGFs. The centrifugation speeds mentioned in the literature range from 180 g to 3000 g for 10–15 min depending on the platelet derivative to be prepared [37,39], and nowadays most procedures are still homemade and based on a “trial and error” approach. Comparison of results is thus difficult due to a lack of reproducibility and quality controls [87], and many open questions still remain in this intriguing field, ones regarding not only technical aspects but also the effective benefits that platelet derivatives could provide to patients in terms of healing efficacy, pain management, and anti-inflammatory effects. Even more, autologous donations require close logistic planning, and not all patients may be good candidates for this kind of approach, such as the cases of heavy smokers, cancer patients, or subjects taking anti-platelet drugs, whose platelet function is too altered to be of therapeutic relevance [63]. However, platelet extracellular vesicles (pEVs) might represent a promising alternative to overcome many limitations of PCs (see later). It is currently well-established that atherosclerosis is an inflammatory disease whose development largely depends on endothelial dysfunction. Rather than an anatomical disruption of the vessel wall, the starting point of atherogenesis is the intrinsic alteration of endothelial cell function, which in turn may be caused by many risk factors, including shear stress-related injury. The formation of the atherosclerotic plaque involves several steps, starting from endothelial damage, and subsequently encompassing leukocyte transmigration through the vessel wall and proinflammatory cytokine production. In this process, the innate immunity contributes through macrophages that engulf oxidized low-density lipoprotein (LDL) particles, thus accumulating foam-like lipid droplets in their cytoplasm, turning into foam cells, and triggering a sterile proinflammatory intracellular signaling cascade. The involvement of the adaptive cell-mediated immune response through the major histocompatibility complex (MHC)-II-mediated antigen presentation to T helper cells completes the process and ensures an ongoing amplification of inflammation [88,89]. In this complex scenario, the contribution of platelets in atherosclerosis has been known since Fitzgerald et al. first reported increased thromboxane A2 levels as an index of platelet activation in unstable coronary artery disease in 1986 [90]. Platelet activation is likely the result of multiple contributing factors that, however, can be summarized with, again, the concept of endothelial dysfunction. Indeed, dysfunctional endothelial cells lose their anti-thrombotic properties (nitric oxide release, prostacyclin synthesis, or ecto-ADPase CD39 expression) and acquire a pro-thrombotic phenotype characterized by increased expression of tissue factor (TF), adhesion molecules, chemokines, and proinflammatory cytokines that eventually promote platelet tethering to an intact but dysfunctional vessel wall. In this way, even in the absence of a true break that needs to be repaired, the hemostatic cascade is triggered, starting from primary hemostasis, where platelets play a pivotal role. Among the molecules involved in the platelet contribution to the atherosclerotic process, P-selectin appears to be indispensable, since it not only mediates platelet adhesion to the endothelium but also induces platelet-leukocyte aggregates to interact with the endothelium. Leukocytes, especially monocytes, express PSGL-1, and P-selectin binding to this receptor activates nuclear factor-κB (NFκB) and induces the expression of monocyte chemotactic protein-1 (MCP-1) together with a wealth of proinflammatory cytokines, primarily tumor necrosis factor α (TNF-α). In turn, platelet activation also leads to their release of cytokines synthesized de novo, such as interleukin-1β (IL-1β), and pre-formed molecules stored in platelet granules, especially α-granules, such as CD40L, the chemokine RANTES, MIP-1α, and PF4, which is crucial in promoting the uptake and the esterification of oxidized LDL by macrophages and their transformation into foam cells. On the whole, platelets are really at the core of atherogenesis, and they may be seen as facilitators and leukocyte chaperones at the atherosclerotic site. In the already-formed atherosclerotic plaque, they tend to form an adherent monolayer, which might be the forerunner of a future thrombotic process. Moreover, even when they adhere sparsely to the dysfunctional endothelium in the initial stages of atherogenesis, they can still mediate the delivery of proinflammatory and chemotactic factors or facilitate leukocyte-endothelium interactions, thus initiating and/or amplifying the whole process [89]. Beyond these initial stages, platelets also contribute to the subsequent remodeling of the atherosclerotic microenvironment characterized by vascular smooth muscle cell migration and proliferation and arterial intimal hyperplasia. In fact, α-granules release PDGF with chemotactic and mitogenic properties. Therefore, platelets contribute to the instability of the plaque and are directly involved in the sites of rupture of the atherosclerotic plaque, where they can cause, in situ, occlusion of the vessel and/or formation of thrombi that can occlude distant vessels [89]. It is therefore not surprising that the efficacy of antiplatelet agents such as aspirin might be related to a wide effect, perhaps including inhibition of the proinflammatory activity of platelets. Indeed, the antiplatelet treatment may affect the platelet-mediated inflammatory cascade in addition to playing a role in preventing aggregation and subsequent thrombotic events at the site of unstable plaques [91]. Acetylsalicylic acid (ASA) inhibits the enzyme cyclooxygenase (COX), of which two isoforms exist: COX-1 and COX-2. COX-1 is expressed constitutively in all tissues, while COX-2 expression is induced in inflammatory states. ASA can inhibit COX-1 by acetylating a serine residue at position 529 (Ser 529), and its antiaggregant effect is due to the reduction of thromboxane A2, a potent vasoconstrictor and inducer of platelet aggregation, whose production is COX-1-dependent. In contrast, ASA is 170 times less potent against COX-2 (through acetylation of Ser516), which plays a major role in inflammation by mediating the conversion of arachidonic acid into prostaglandin H2. As a result, the low doses of ASA usually recommended for cardiovascular prevention achieve a sufficient antiplatelet effect but an inadequate anti-inflammatory effect. Indeed, ASA has been shown to have greatest anti-inflammatory effects at doses above 1.2 g [92]. Among the other antiaggregant agents, thienopyridines such as clopidogrel have been extensively studied for their role extending beyond the antagonism of platelet aggregation. Different anti-inflammatory mechanisms have been proposed for this class of drugs, including decreased expression of adhesion molecules, chemokines, and TF, and a reduction in leukocyte aggregates formation [93]. These drugs antagonize the ADP P2Y12 receptor by irreversibly binding to a cysteine residue. Since ADP plays an essential role in platelet activation, the main result is an anti-aggregating effect. However, by doing so, thienopyridines also indirectly reduce platelet-mediated inflammation and interaction with leukocytes, for instance by reducing the P-selectin expression [94]. In addition, unlike ASA, which does not have any effect on leukocyte-platelet aggregates [95], thienopyridines can attenuate leukocyte function directly since the P2Y12 receptor has been found also on leukocytes [96]. Always dealing with the remodeling of the vascular microenvironment, another scenario where platelets are pathologically involved is the local evolution of vascular stents used to treat acute ischemic events. Basically, there are two main types of stents: bare metal stents and drug-eluting stents, with the latter being superior due to their release of antiproliferative drugs such as sirolimus, paclitaxel, or rapamycin, which prevent smooth muscle cell hyperplasia and hence restenosis [97,98]. On the other hand, they may also seriously injure the endothelium, contributing to the vicious cycle of endothelial dysfunction-platelet aggregation that has already been described in the pathogenesis of atherosclerosis. The consequence is a high risk of thrombosis and delayed endothelialization of the stent surface [99] and it is thus not surprising that antiplatelet drugs belong in the current therapy of patients receiving stent implants despite carrying an increased bleeding risk [100]. In this sense, several attempts have been made to optimize the engineering of stents in order to solve the urgent problem of obtaining biomaterials with antithrombotic, antiproliferative, anti-inflammatory, and pro-endothelialization properties within the same device [101]. The detailed achievements are beyond the scope of this narrative review. However, just to cite a few examples from the biomedical engineering field, Han and colleagues have developed a new coating of cardiovascular stents where the traditionally used magnesium alloy is combined with plant-derived ferulic acid, a highly biocompatible natural compound with anti-inflammatory properties and able to inhibit platelet aggregation and smooth muscle cell hyperplasia and to promote endothelial cell proliferation [102]. Even more interestingly, Li et al. have proposed a stent coated with a new subtype of chondroitin sulfate that, in addition to retaining the already-known anti-aggregating properties, shows what the authors call a “spatiotemporal orderliness of function”, i.e., a process in which the biomaterials direct cell fates in time and space sequence by influencing the surrounding microenvironment and inducing phenotype changes not only in vascular wall cells (smooth muscle and endothelial cells) but also in inflammatory cells, such as M1 and M2 macrophages [101,103]. These approaches demonstrate that a mindful choice of the biomaterials used in intravascular devices is of paramount importance for the patient outcome, and it must consider several variables regarding the different cell types involved in the complex process of device tolerance, including platelets. Platelets are also involved in cancer, by regulating several aspects of tumorigenesis and metastasis. The interaction between cancer cells and platelets is very complex and bidirectional, both in the blood and in the tumor microenvironment. Platelets contain more than 300 bioactive molecules in their granules (e.g., chemokines, platelet-derived growth factors) and express numerous receptors on their surfaces (e.g., P-selectin, integrin αIIbβIII, P2Y12, protease-activated receptor-1 (PAR-1)) directly involved in inflammation, cancer progression, and metastasis. First, platelets are involved in the outcome of cancer patients. Elevated platelet counts are significantly correlated with a worse progression and lower overall survival in many cancers (breast, colon, lung, kidney, and pancreatic cancers). One mechanism that could explain thrombocytosis in many tumors could be tumor-derived interleukin-6 (IL-6), which stimulates thrombopoietin (TPO) production in the liver, thereby promoting megakaryopoiesis and thrombocytosis [104]. Moreover, a high percentage of cancer patients suffer from vascular thromboembolism (pulmonary embolism and deep venous thrombosis). Armand Trousseau first described, writing in 1865, that cancers can induce venous thrombus formation. Thrombosis is one of the most common clinical manifestations in cancer patients and is associated with worse prognosis and survival: the principal cause of high thrombotic risk is platelet activation and aggregation through direct and indirect mechanisms induced by tumor cells [104]. An important mechanism of tumor-induced platelet aggregation is the secretion by cancer cells of thrombin, a serine protease that converts fibrinogen to fibrin, activates many coagulation factors (FV, FVIII, FXI, and FXIII), and activates platelets by PAR. TF, the main activator of the coagulation cascade, is also expressed by cancer cells. High serum levels of TF have been found in several types of cancer. Activated platelets also express TF on their membrane, contributing to thrombosis [105]. Another mechanism of platelet activation by cancer cells is mediated by ADAM9 (a disintegrin and metalloproteinase 9), which is found in several cancer types and has been correlated with tumor aggressiveness and poor prognosis. ADAM9 can bind to the platelet laminin receptor (α6β1), a key platelet receptor for laminins. In this way, it supports both the adhesion and the activation of platelets and enhances platelet activation and tumor cell extravasation [106]. Secondly, platelets are directly involved in the progression and evolution of the tumor, playing an important role in metastases, whose formation is strongly inhibited by the cytotoxic activity of natural killer (NK) cells against the circulating tumor cells. Tumor cells and platelets are able to produce microthrombi in the circulation by the binding of the platelet P-selectin to the ligands of tumor P-selectin and thanks to the interaction between platelet TLR4 and the released High Mobility Group Box 1 (HMGB1) protein from the tumor [106]. The interaction between cancer cells and platelets is also mediated by the binding of GPIb-IX-V and GPIIb-IIIa to tumor cell integrin αvβ3 and via P-selectin, which can bind to mucins on the tumor cell membranes. Moreover, platelet-derived TGF-β diminishes NK cell activity by downregulating the NK activatory receptor NKG2D and increases tumor cell survival by activating the TGF-β/Smad and NF-kB pathways [104]. In this way, the metastases present in the circulation are protected/shielded by the platelets in a platelet aggregate, evading the immune response by their natural killers [107]. The activation of platelets by cancer cells and the subsequent “release reaction” of platelet secretome has many other pro-cancerous effects that modulate the tumor microenvironment, stimulate tumor growth and help metastases. After platelet activation, the platelet secretome is sequestered into the local microenvironment, where it can promote and support tumor cell proliferation and tumor growth. Activated platelets also release microvesicles [108] that can further promote disease progression through multiple mechanisms. For example, to disseminate, circulating cancer cells need to adhere to endothelial cells and must infiltrate through the vessel wall. ATP secreted from dense granules activates P2Y2 on endothelial cells, increases endothelial permeability and promotes the diapedesis of cancer cells. In this way, platelets are directly involved in increasing endothelial permeability, promoting tumor cell transendothelial migration. The significant crosstalk between tumor cells and the endothelium, mediated by platelets, may then improve tumor metastasis. Angiogenesis is another essential process in tumor growth and metastasis and, indeed, angiogenesis-based targeted therapy is considered a cornerstone for cancer treatment. PDGF and VEGF, secreted from α-granules, promote, respectively, tumor growth and angiogenesis [104]. Moreover, the proinflammatory cytokines released by platelets are powerful recruiters and activators of leukocytes; IL-8 and chemokines secreted by platelets attract hematopoietic cells to the tumor site, stimulating tumor growth and angiogenesis [109]. In this way, platelets participate directly in several steps of cancer metastasis and affect disease burden and treatment efficacy in cancer patients (Figure 3). The strong evidence of an association between platelets and cancer, described above, has led to the hypothesis that antiplatelet therapy could be used in antitumor therapeutic strategies. It has been observed that, in adenoma, platelets that are activated by COX-2-mediated signal transduction pathways increase the aggressive phenotype of cancer cells. Low-dose aspirin, by inhibiting COX-1, may exert antimetastatic effects, decreasing the cancer-mediated activation and aggregation of platelets [110]. In a pancreatic tumor mouse model, daily administration of the antiplatelet drug clopidogrel decreased tumor growth rate and the number of metastatic foci significantly [111]. Similarly, the administration of antiplatelet drugs that bind glycoprotein complex αIIbβIII reduced the proliferation of melanoma cells injected in rat subcutis [107]. Platelets also play a role in protecting cancer cells against chemotherapy-induced apoptosis and in maintaining the integrity of tumor vasculature. In colon and ovarian cancer cell lines, platelets increase the resistance to 5-fluorouracil and paclitaxel. Moreover, low blood platelet counts in mouse models of breast and lung cancer significantly increase sensitivity to paclitaxel. In cancer patients and murine tumor models, high platelet counts have been associated with a poor response to chemotherapy [104]. Because of these pro-tumor effects, anti-platelet drugs have been introduced into cancer treatment strategies. Many pieces of evidence indicate that aspirin is useful in cancer to reduce metastasis and mortality, especially in colorectal cancer [106]. In 2016, the US Preventive Services Task Force approved the prophylactic use of low-dose aspirin in colorectal cancer patients [112]. Several other antiplatelet drugs are in the preclinical stage or are being tested in clinical trials, such as GPVI and GPIba antagonists [113]. GPVI is a platelet receptor for collagen on the subendothelial matrix and can bind also to fibrinogen and fibrin. GPVI promotes metastasis in mice [114] and may be involved in “tumor cell-induced platelet aggregation”. GPVI antagonists have broad inhibitory effects on tumor–platelet aggregation. Xu and colleagues observed that anti-GPIba antibodies decrease thrombopoietin generation and inhibit tumor-induced thrombocytosis [115]. Other antiplatelet agents can have antitumor effects, but their use in cancer therapy can be limited by their interference with hemostasis, increasing the risks of bleeding and gastrointestinal toxicity. In the near future, however, it will be important to further characterize the mechanism of action of antiplatelet agents in tumors, because it would certainly improve the antitumor therapeutic strategies. The brain has the ability to reorganize itself throughout life to adapt to environmental changes via the continuous generation of new functional neurons derived from neural precursor cells. This process occurs in specialized neurogenic niches, predominantly in the subgranular zone of the hippocampal dentate gyrus and in the subventricular zone of the lateral ventricles [116]. The brain is well vascularized by a dense network of fine microvasculature, and molecular exchanges between the blood and the nervous system are finely controlled and influence neurogenesis during life. Platelets are a link between the blood and the brain and can promote and modulate neurogenesis by secreting bioactive molecules from granules. Platelets can also rapidly respond to environmental changes in the brain by modifying their proteome via translation of stable mRNAs [116]. It has been recently demonstrated that platelets can cross the inflamed brain microcapillaries [117] and exert local actions through their surface receptors and released factors. Platelets express several surface molecules, which allow them to directly interact with glial cells, endothelial cells, and neurons [118]. In particular, CD62P, ALCAM, Siglec-H, and Siglec-15 platelet receptors can bind sialylated gangliosides present in the lipid rafts of neuronal processes [119]. These bindings would promote the formation of new dendritic spines on neurons and neural precursor cells [120] (Figure 4). Platelets carry several neurotransmitters that are essential for the communication between neurons. In particular, they can promote and modulate neurogenesis by influencing neural precursor cells, and they can have neuroprotective effects via α-granules bioactive molecules, such as VEGF, EGF, FGF-2, IGF-1, PF4, TGF-β, and stromal cell derived factor 1 (SDF-1), and via dense granule neurogenesis-promoting molecules, such as serotonin, histamine, epinephrine, and dopamine [121] (Figure 4). It must be emphasized that platelets and neural cells are comparable in their intracellular storage compartments: platelet dense granules and small dense-core synaptic vesicles of neurons store serotonin and adenosine triphosphate contents, whereas the large dense-core vesicles of neurons and platelet α-granules contain neuropeptides, neurohormones, and neurotransmitters. PRP treatment enhances the recovery of peripheral nerves following injury, including cavernous nerve injuries [121] and damage to the facial and sciatic nerves [122]. Moreover, PRP injections into the injured spinal cord of rats have been shown to promote locomotor recovery, local angiogenesis, and neuronal regeneration [123]. Another study in mice suggested the therapeutic use of PRP in neuroinflammatory central nervous system diseases [124]. In rats, PL injection promotes proliferation, neurogenesis, and survival and reduces apoptosis in neural precursor cells of the subventricular zone of the lateral ventricles [116]. Moreover, platelet-derived serotonin increases the expression of genes involved in synaptic plasticity [120]. Platelet microparticles and platelet exosomes can promote neural precursor cell proliferation, survival, and differentiation in vitro [125]. Human platelet lysates have been investigated as a novel biotherapy for amyotrophic lateral sclerosis (ALS) and Parkinson’s disease (PD) patients. In a cell-based model of ALS, human platelet lysates confer a neuroprotective effect against apoptosis and oxidative stress, inhibiting neuronal loss [126]. In a human mesencephalic cell-based model of PD, pre-treatment of the cells with human platelet lysates also protects against ferroptotic cell death [126]. Platelets express also glutamate receptors and exhibit glutamate uptake activity and carry considerable amounts of γ-aminobutyric acid (GABA) in dense granules. Glutamate, the most abundant excitatory neurotransmitter in the brain, and GABA, the major human inhibitory neurotransmitter, are crucial for healthy brain function; abnormalities in glutamate and GABA signal transduction pathways are associated with many neurodegenerative conditions such as PD, Alzheimer’s Disease (AD), and ALS (Figure 4). Another similarity between platelets and neurons is the production and secretion of amyloid-ß. Platelets are silos of the amyloid precursor protein (APP), containing about 90% of the circulating APP in their plasma membrane and α-granules. APP is cleaved by ß-secretase into amyloid-ß and secreted in the blood by activated platelet. APP acts as a platelet receptor and it is involved in thrombosis and coagulation, whereas amyloid-ß promotes platelet aggregation. Amyloid-ß induces platelet activation by binding to the scavenger receptor CD36 and GP1bα and activating the p38 MAPK/COX1 pathways. These pathways induce the release of TXA2, triggering platelet activation, adhesion, and aggregation (Figure 4) [127]. In AD, a neurodegenerative disease that progressively leads to the loss of neurons and consequent dementia, deposition of amyloid-ß in the brain tissue and cerebral vessels is one of the most important neuropathological mechanisms. A recent study has shown that platelets are hyperactive in AD patients and a transgenic mice model of AD. Moreover, the early development of platelet inclusions in cerebral blood vessels in AD mice suggests a role of platelets in amyloid-ß plaque formation. Another work showed that platelets promote the formation of amyloid-ß aggregates in the brain vasculature and that amyloid-ß itself is able to activate platelets [128]. Platelet dysfunction is associated with several other neurodegenerative diseases. In Huntington’s disease, a hereditary autosomal dominant neurodegenerative disorder, platelets display many abnormalities, including aberrant amplification of adenosine A receptor signaling, nitric oxide metabolism dysregulation, and elevated monoamine oxidase activity (MAO). MAO is a mitochondrial enzyme that catalyzes the oxidative deamination of dopamine and presents two different isoforms, A and B, and MAO-B is expressed by platelets Increased platelet MAO-B activity has been positively correlated with Huntington’s disease progression (Figure 4). Several studies suggest that elevated platelet MAO-B activity has been associated with neuronal damage in many other degenerative conditions, such as PD. In PD patients, many other platelet alterations have been observed, including increased mean platelet volume and decreased glutamate uptake [128]. In conclusion, platelets can regulate neural cells, contribute to brain plasticity and carry pro-neurogenic factors, and multiple mechanisms are involved in platelet-neural cell communication. For their neuroactive effects, platelets could represent a potential target in neurodegenerative diseases. On the one hand, antiplatelet drugs could be introduced in treatment strategies in order to reduce platelet activity and/or inhibit the overproduction, for example, of amyloid-β in AD patients [129]. Thanks to cell-specific interactions mediated by their receptors, platelets may be used also as a drug delivery system to target specific cells that are difficult to access, such as neurons. For example, platelets have been recently suggested as a model system of glutamate and GABA transport in patients suffering from neurodegenerative conditions [128]. The most recent research in the field of regenerative medicine has brought to light a new interest in the translational and therapeutic potential of extracellular vesicles (EVs). EV is an umbrella term encompassing cell membrane-derived structures differing in their size and including exosomes (30–100 nm), microvesicles (100 nm–1 µm), and apoptotic bodies (>1 µm) [130]. The nomenclature is somewhat confusing and used to be misused [131] until the International Society of Extracellular Vesicles (ISEV) clarified the terminology by reclassifying them into small EVs (sEVs), less than 200 nm in diameter, and medium/large EVs (m/lEVs), more than 200 nm in diameter. Alternatively, the EV density, the biochemical composition, or the isolation technique can be used as additional classification criteria [132]. In general, EVs can be defined as heterogeneous structures delimited by a lipid bilayer and unable to replicate, since they do not contain a functional nucleus [132]. EVs have been implicated in several physiological and pathological functions including immune surveillance, oncogenesis, cardiometabolic, and neurologic disorders and their detailed description is beyond the scope of this review. Here, we provide only a brief overview of three main clinical areas where EVs have a demonstrated involvement. In general, their role can be recapitulated primarily by their ability to mediate intercellular communication [133]. In oncogenesis, EVs have been implied in microRNA transfer between malignant cells to transmit chemotherapeutic resistance [134], or from a malignant to a nonmalignant cell to selectively silence gene expression and induce transformation to cancer cells [135]. Some authors have proposed EV involvement in the metastatic spread, by its harboring molecules necessary for the epithelial-to-mesenchymal transition [136] and/or by acting at a distance and prime a metastatic niche to welcome the migrating metastatic cells appropriately or to locally promote cancer cell growth [137,138]. Regarding their clinical applications, they have been proposed as biomarkers of tumor monitoring [139,140] and as therapeutic vectors for small molecule delivery [141,142]. In cardiology, Emanueli et al. have shown that the concentration of circulating exosomes in plasma correlates with the levels of cardiac troponin, increasing 24 to 48 h after coronary artery bypass surgery [143]. Furthermore, EVs have been implicated in cardiac hypertrophy and remodeling in general, as well as in cardiometabolic disorders where they could mediate vascular damage in metabolic diseases such as diabetes and obesity [144,145]. In neurology, recent studies suggest a role of the EV content in neurodegenerative disorders such as AD or PD, where a correlation between the clinical manifestations and altered expression of miRNA or synaptic proteins has been described in blood and cerebrospinal fluid (CSF) EVs [146,147]. Similarly, a study showed that the circulating levels of EVs containing tau protein, a general marker of neuronal damage, were more elevated in American football players compared with healthy controls, and were probably related to a chronic subclinical traumatic encephalopathy already described in these subjects [148]. Finally, the therapeutic use of EVs has been explored in AD through the use of engineered EVs containing small interfering RNA (siRNA) to alter the expression of beta-amyloid [149] or EVs derived from stem cell-derived EVs to mitigate the clinical manifestations [150,151]. Despite the promises of the different diagnostic and therapeutic applications of EVs, most studies are still preclinical, and clinical use is hindered by the lack of clarity regarding EVs biogenesis, a necessary prerequisite to fully understanding their potential, in the absence of standardization in the isolation and characterization techniques, and limited information about the influence of age, gender, and ethnicity in humans [133]. Platelets are one of the main bodily sources of EVs [152,153]. Their ability to release EVs was already reported in 1967 by Peter Wolf, who coined the expression “platelet dust” to describe the microscopic lipid-rich particulate he obtained after ultracentrifugation of plasma and serum samples [154]. Later on, further studies prompted the evolution of this term to the more accurate one of “platelet microparticles”, which eventually entered the umbrella definition of pEVs according to the ISEV’s recommendations [132]. Surprisingly, the attention of most studies on the therapeutic role of EVs has been focused on ex vivo sources, such as mesenchymal stromal cells (MSCs), with many protocols being already in place in the field of regenerative medicine [155,156]. Nevertheless, using blood cells as a source of EVs would represent a great advantage in terms of cost, manufacturing techniques, and safety. Indeed, ex vivo sources of EVs require at least two steps for their production, i.e., isolation followed by expansion and differentiation in a growth medium, whereas body-cell-derived EVs are already available after collection from autologous or allogeneic donations. Despite the dependence on blood donors, in vivo sources of EVs require little manipulation and eliminate the concerns regarding potential contamination by the growth medium, thus circumventing regulatory issues related to the manufacturing process [157]. Their inherent biocompatibility, compared to MSC-derived EVs or synthetic nanoparticles, also makes them “immunologically transparent” and better tolerated by the recipients [158]. Even more, unlike platelets, pEVs can cross tissue barriers, including the blood-brain barrier (BBB), especially when it is inflamed [159], also thanks to their expression of adhesive receptors such as integrins or Siglec molecules [119,160], thus greatly expanding their use beyond the blood compartment [153]. pEVs bear a phenotype that reflects the typical characteristics of the original platelet, including expression of phosphatidylserine (PS), CD31, CD41, CD42, CD61, CD62, and CD63. Their cargo, too, reflects their parental origin since p-EVs contain growth factors, cytokines, chemokines, lipids, neurotransmitters such as serotonin, and nucleic acids (messenger RNA and microRNA) that mediate a delicate balance between proinflammatory and anti-inflammatory, pro-coagulant and anti-coagulant, and pro-angiogenic and anti-angiogenic events. Furthermore, at least some p-EVs have been found to include mitochondria [161,162]. Since pEVs are true biological products, reporting their isolation and characterization techniques used is paramount in getting formal approval as therapeutic agents. Unfortunately, as in the case of PCs, different preparation methodologies have been used without reaching a unanimously shared protocol. Moreover, the isolation technique utilized greatly affects the nature of pEVs [87]. The most common isolation methods are based on ultracentrifugation and density gradient [163] even though size exclusion chromatography, the use of antibody affinity columns, or filtration techniques should be considered as well. A possible hurdle in pEV isolation is the overlap in their size and/or density with other particles such as lipoproteins. Beyond the impact of their purity on the effective number of pEVs obtained, a great number of lipoproteins might also have undesirable biological impacts, since they may behave as proinflammatory effectors [87]. A comparison of different isolation techniques has shown that the best approach is a combination of methods, which decreases the risk of co-isolating lipoproteins but also decreases the pEVs’ yield. In particular, size exclusion chromatography is worth being included in the process, since it is highly efficient in removing free and potentially interfering blood biomolecules. Nevertheless, so far, there has been no proof of therapeutic superiority of either isolation technique [164,165]. An important point is that all these methods require several pre-analytical manipulations representing potentially stressful conditions that might promote platelet damage or alter the pEV’s morphology and functional characteristics [108]. The next step in EV manipulation is their characterization in order to correctly identify their source and to determine their specific phenotype. The most-used techniques are electron microscopy and nanoparticle tracking analysis (NTA) to analyze size and morphology, while Western blot, flow cytometry, and mass spectroscopy are used to identify the expression of surface proteins and the cytosolic cargo [163]. Regarding the source, a recent review [87] identified PRP as the most frequent source used to prepare pEVs. Even though pEVs are already generated under physiological conditions, their use for therapeutic purposes might prompt the need to further enhance their production beyond their constitutive release, and therefore a still-controversial issue is whether platelets should be triggered by an agonist before pEV isolation. Active PCs may perform better not only in quantitative terms (absolute number of pEVs obtained) but also concerning the qualitative properties of pEVs, such as PS exposure on the outer membrane layer or release of growth factors with therapeutic potential [87]. Induction of pEV release can be achieved using several agonists such as collagen, thrombin, ADP, and arachidonic acid as well as calcium ionophores and LPS [157]. Interestingly, the agonist used does not seem to have a significant effect on the pEV’s size. However, in quantitative terms, calcium ionophores seem to ensure the greatest yield, even though also thrombin is a potent inducer, especially if compared with the weaker activity of LPS. Conversely, calcium ionophore-induced pEVs bear a smaller protein cargo, reflecting a poor packaging capacity [166]. On the whole, the choice of the best inducing technique reflects a delicate balance between the absolute yield of pEVs from a single platelet and the amount of cargo and physical properties of the pEVs obtained. In other words, manufacturing techniques should take into account both quantitative and qualitative issues, both of which are greatly influenced by the chosen activation pathway [165,167,168,169] and the storage time before induction of pEV release. Indeed, old stored platelets seem to give a greater yield of pEVs compared to fresh platelets [87]. Storage is another important caveat that must be solved to harmonize pEV processing and preservation techniques. It is noteworthy that, differently from PCs, pEVs can tolerate freezing [170,171,172], thus eliminating the concerns regarding special storage and transport requirements or the need to be used within a few days. Additionally, in this case, the storage temperature may affect both the morphology and function of pEVs [173], but it is not clear whether the best condition is −80 °C or −20 °C [87]. It would be worth examining and comparing different storage conditions, the use of cryopreservatives, and their impact on the viability and therapeutic potential of pEVs. So far, many therapeutic applications of pEVs have been investigated, even though most studies are still limited to in vitro or animal experiments. The main rationale behind their therapeutic potential is the vast array of molecules expressed on their membrane and their rich content in signaling molecules and growth factors, which allow them to be perfect candidates for intercellular interaction and for acting at a distance on different targets (Table 1 and Figure 2). Dealing with hemostasis, pEVs have a proven effect on vascular permeability [174] and seem to be even more pro-coagulant than activated platelets [172,175,176]. In the field of regenerative medicine, pEVs might represent a promising approach to wound repair, since the growth factors in their cargo promote fibroblast and keratinocyte migration and proliferation [177,178]. In muscle [179] and bone regeneration, they can enhance stem cell intra-articular engraftment and differentiation [180], as well as promote expansion and prevent apoptosis of local cells, also through the activation of intracellular signaling cascades including the Wnt/β-cateninin [181] or the Akt/Bad/Bcl-2 pathway [182]. In addition, their mitogenic effect could be attributed also to their genetic content (miRNA) [183]. In neurodegenerative disorders, pEVs can promote the proliferation of neural stem cells thanks to their growth factors, more than the use of the same growth factors alone [116,125,128,184]. Their success is also supported by their proven angiogenic potential, primarily derived from their content rich in PDGF, VEGF, and bFGF [184,185,186]. The few studies on human subjects are mainly focused on pEV as an outcome measure, for instance as an index of therapeutic efficacy of antiplatelet drugs (study identification numbers NCT02931045; NCT04578223), or dietary supplements (NCT03203512), or as diagnostic biomarkers (NCT05530330). However, at the time of this review (January 2023), there are a few ongoing or recently completed interventional studies on human subjects using pEVs as a therapeutic approach, in particular for the surgical treatment of chronic tympanic membrane perforations (NCT04761562), the nonsurgical treatment of chronically inflamed post-surgical temporal bone cavities (NCT04281901), the treatment of patients with acute myocardial infarction undergoing percutaneous coronary intervention (NCT04327635), and to promote healing of skin grafts (NCT04664738). Platelet EVs can be exploited both directly due to their intrinsic properties, and indirectly as delivery vehicles whose cargo can be adjusted according to different therapeutic needs. Their peculiar structure, i.e., an outer lipidic bilayer and an inner aqueous environment, makes them capable of housing both hydrophobic and hydrophilic drugs [157,187]. Platelets may be preloaded with the drugs and then induced to release pEVs using different agonists. Alternatively, the post-loading of already formed pEVs can be chosen. In both cases, loading can be achieved passively through incubation in a drug-containing medium or actively through sonication, electroporation, uptake after saponin treatment or freeze-thaw cycles, and transfection [157]. The most straightforward clinical application of drug-loaded pEVs is probably in the field of oncology, since the interaction between platelets and cancer cells is a consolidated phenomenon, and cancer cells are capable of internalizing pEVs [188,189]. Therefore, pEVs might be used as Trojan horses for anticancer drug delivery. Intriguingly, Michael et al. demonstrated an anticancer effect already for native pEVs by showing the inhibition of lung and colon carcinoma growth in mice transfused with platelet-derived microparticles (PMPs) and identified miR-24 as the main factor responsible for the induction of tumor cell apoptosis in vivo [190]. Furthermore, when synthetic nanoparticles (NPs) are coated with platelet membranes, targeting of tumor tissues is improved, which proves that the molecules displayed on the NP surface are the major determinants of the NP fate [191,192]. A pioneering approach has been tried to facilitate the delivery of antiviral therapies. In an in vitro study, pEVs entrapping the anti-HIV drugs lamivudine and tenofovir comparatively increased the inhibitory effects on HIV-1 replication, while decreasing cytotoxicity, likely due to the slower release by pEVs [193]. Evidence of pEV implication in viral disease emerged during the COVID-19 pandemic with findings of higher levels of pEVs in hospitalized SARS-CoV-2-positive patients compared with uninfected hospitalized controls [108]. Finally, the pEV role in cardiovascular disease is also well established [187]. pEVs have been shown to correlate with the size of the myocardium at risk after an acute coronary syndrome (ACS) [194] and to act as biomarkers of vascular inflammation, perhaps because they contain some proinflammatory isoforms of C-reactive protein (CRP) [187]. This also explains their putative involvement in atherogenesis, where pEVs may interact with endothelial cells and leukocytes, acting as functional bridges to mediate monocyte recruitment to the vascular walls [195], and induce macrophage apoptosis [196], which contributes to the formation of foam cells. Surprisingly, pEVs have not been found within the atherosclerotic plaque itself but only in the blood of atherosclerotic patients. Atherosclerotic plaques contain microvesicles, but not of platelet origin [197], which instills the doubt that they might be simple bystanders of the atherosclerotic process instead of active players in its pathogenesis [187]. Nevertheless, the fact that they can be engineered to modify their cargo and used as drug-delivery systems has inspired new therapeutic platforms, such as the one developed by Pawlowski and colleagues, where pEV-like NPs loaded with a thrombolytic drug were successful in obtaining targeted fibrinolysis in preclinical models [198]. In addition, not all pEVs are associated with cardiovascular risk. Indeed, also natural pEVs have shown therapeutic potential. For instance, transfusion of pEVs from rats after hind limb ischemia-reperfusion conditioning into rats with middle cerebral artery occlusion was able to reduce the infarct area [199], and a similar result was obtained in rat models of limb ischemia [200], suggesting that the transfer of conditioned or “educated” pEVs might be protective and partially reverse the injury already occurred. The precise mechanism underlying this cardiovascular protective role is still poorly understood but likely derives from the rich biological cargo of pEVs as well as from their capability to deliver it to several cell types in a finely targeted manner [187]. In summary, from their humble origin as “platelet dust”, pEVs have certainly made huge steps forward, and they have been established not only as biomarkers with diagnostic, prognostic, and predictive significance but also as promising therapeutic strategies. If until now the discussion has been focused on platelet-derived biomaterials and fragments such as extracellular vesicles as a therapeutic opportunity, it would also be worth considering the potential ability of platelets themselves to become drug delivery vehicles. Indeed, their multifaceted characteristics make them good candidates for therapeutic drug delivery. First, they are the second-most abundant blood cell types after erythrocytes, and they are easier and quicker to purify [201]. Second, their biological properties allow them to encapsulate both hydrophobic and hydrophilic drugs [157] and to store them mainly in their open canalicular system [159]. In this way, they can cloak the drug and hide it from the body, decreasing its clearance rate and meanwhile allowing a more targeted delivery of the drug with fewer systemic effects [159]. Third, their biocompatibility allows them to travel through biological structures almost undisturbed and with little or no immunogenicity [201]. The best example is probably their ability to cross the BBB [159]. The fact that platelets release the encapsulated drug only when they are activated and undergo degranulation [159] allows us to define their mechanism of targeted drug delivery as a true deceit for the body. For instance, in the field of oncology, tumor cell-induced platelet aggregation can be exploited to let platelets release the chemotherapeutic drug directly into the tumor mass. In other words, a process that normally would contribute to tumorigenesis can be converted into a therapeutic attempt at the expense of the tumor itself, as has already been demonstrated for doxorubicin-loaded platelets in adenocarcinoma [202] or lymphoma cell lines [203]. Similarly, in the case of occluding strokes, where platelet aggregation plays a pivotal pathological role at the site of an unstable atherosclerotic plaque, they have been hypothesized as effective carriers of an anticoagulant drug. The rationale behind this promising hypothesis is that, since platelets will likely be activated at the ischemic site, this will be also the site where the drug will be released, ensuring a targeted therapy [159]. Several steps forward have been made in the field of platelet engineering, i.e., a method of modifying platelets through chemical and/or physical mechanisms favoring their engulfing of the desired drug, coating them with biomolecules, or changing their properties to improve their drug-delivery ability. One of them consists in exploiting the abundant thiol and amine residues on their membrane to chemically link them to biomolecules [201]. Another technique is electroporation, which allows molecules to penetrate the cell through pores created in its membrane [204]. For example, Rao et al. have used this method to load platelets with nanorods to facilitate phototherapy in the treatment of head and neck squamous cell carcinoma in mice. Conveying light energy through an external irradiating source, the nanorods converted it into heat energy, injuring the tumor cells [205]. The limited lifespan of platelets and their relatively low stability in response to external stimuli have prompted the need to think about valid alternatives. The advent of biomimetic platelets, i.e., artificial platelets that mimic the biological properties of their natural counterparts, represents a recent achievement in biomedical engineering [206]. In practice, biomimetic platelets must bear at least three key platelet properties: flexibility, discoid morphology, and ability to aggregate upon activation. This can be achieved by using bi-layered nanoparticles coated with platelet-mimicking peptides [207]. Nevertheless, platelet-mimicking biomaterials eliminate the greatest advantage of their natural counterparts, i.e., the evasion of the immune system. Being synthetic, they are characterized by high immunogenicity, and they often have lower biological efficacy. For these reasons, they are still not used in clinical practice [201]. There are still many unanswered questions that must be investigated, not only in terms of posology issues (e.g., the amount of drug encapsulated by the platelet, the platelet concentration to be administered, the best timing and way of administration) but also referring to the possible, and probably unavoidable, alteration of platelet characteristics induced by the drug itself. To cite just one of these concerns, it cannot be excluded that an anticoagulant drug taken up by a platelet to treat stroke will alter platelet features, for instance by changing its morphology, reducing its capability as a drug delivery vehicle, or increasing the immunogenicity. Furthermore, the release of the drug precisely at the desired site might still seem utopian [159], and further in vivo studies are deemed necessary to better disentangle the several doubts that remain in this fascinating but still pioneering area of research. Platelets are multifunctional blood components, capable of changing shape and function, responding rapidly to environmental changes, and fulfilling distinct context-dependent functions throughout life, just as did Proteus.Activated platelets release numerous bioactive molecules from α-granules, δ-granules, and lysosomes. The platelet secretome influences many physiological and pathophysiological processes beyond hemostasis, such as inflammation, immunity, neurogenesis, and oncogenesis. Platelets also exhibit complex interactions with many different cells, beyond endothelial immune cells in the circulation [1]. Based on their multifaceted functions and multiple cell interactions, we assume that the platelet population is probably composed of subgroups with different functions: hemostatic platelets, immune platelets, sentinel platelets, helper platelets, and scavenger platelets. We intend to test this hypothesis in the future. Certainly, the in-depth knowledge of all their functions will help to understand the pathogenesis of numerous pathologies and to reveal new therapeutic properties with pleiotropic effects. Nowadays, antiplatelet drugs are the first-choice therapy for the treatment of cardiovascular disease and the prevention of atherothrombosis. However, growing evidence indicates that antiplatelet drugs are also effective in cancer treatment strategies and neurodegenerative disease therapy. Interestingly, not only drugs targeting platelet function but also platelets themselves show potential as therapeutic strategies. Attention has also been focused on the use of platelet concentrates in regenerative medicine and pEVs as a new approach to develop functional nanoparticles for disease-targeted delivery. Many clinical trials have been carried out in the last two decades, mainly in the fields of orthopedics, maxillofacial surgery and dentistry, ophthalmology, and cardiac surgery. Furthermore, despite being at its onset, the study of platelets as drug delivery vehicles is showing promising results, which are predominantly derived from high biocompatibility, versatility and minimal immunogenicity. Platelets, apparently simple cellular fragments, hide a world, mostly still unknown, with enormous diagnostic and therapeutic potential, now summarized in this review.

PMC10002541

Yidi Wang,Lili Li,Pingyao Cai,Rikke Heidemann Olsen,Shuai Peng,Hecheng Meng

Antimicrobial Activity of Sertraline on Listeria monocytogenes

28-02-2023

Listeria monocytogenes,sertraline,antimicrobial activity,biofilm,virulence

We explored the antimicrobial activity of sertraline on Listeria monocytogenes and further investigated the effects of sertraline on biofilm formation and the virulence gene expression of L. monocytogenes. The minimum inhibitory concentration and minimum bactericidal concentration for sertraline against L. monocytogenes were in the range of 16–32 μg/mL and 64 μg/mL, respectively. Sertraline-dependent damage of the cell membrane and a decrease in intracellular ATP and pHin in L. monocytogenes were observed. In addition, sertraline reduced the biofilm formation efficiency of the L. monocytogenes strains. Importantly, low concentrations (0.1 μg/mL and 1 μg/mL) of sertraline significantly down-regulated the expression levels of various L. monocytogens virulence genes (prfA, actA, degU, flaA, sigB, ltrC and sufS). These results collectively suggest a role of sertraline for the control of L. monocytogenes in the food industry.

Antimicrobial Activity of Sertraline on Listeria monocytogenes We explored the antimicrobial activity of sertraline on Listeria monocytogenes and further investigated the effects of sertraline on biofilm formation and the virulence gene expression of L. monocytogenes. The minimum inhibitory concentration and minimum bactericidal concentration for sertraline against L. monocytogenes were in the range of 16–32 μg/mL and 64 μg/mL, respectively. Sertraline-dependent damage of the cell membrane and a decrease in intracellular ATP and pHin in L. monocytogenes were observed. In addition, sertraline reduced the biofilm formation efficiency of the L. monocytogenes strains. Importantly, low concentrations (0.1 μg/mL and 1 μg/mL) of sertraline significantly down-regulated the expression levels of various L. monocytogens virulence genes (prfA, actA, degU, flaA, sigB, ltrC and sufS). These results collectively suggest a role of sertraline for the control of L. monocytogenes in the food industry. Listeria monocytogenes (L. monocytogenes) is a Gram-positive pathogen and the causative agent of listeriosis [1] in which the typical clinical symptoms include meningitis, septicemia and stillbirth with a mortality rate of up to 30%, especially in immunocompromised hosts [2]. L. monocytogenes inhabits a broad ecologic niche and can contaminate various food products and food processing environments [3]. As this bacterium is psychrophilic; resistant to desiccation, acid and heat; and tolerant to increased sub-lethal concentrations of disinfectants or resistant to lethal concentrations [4], it can persist in food processing environments and poses a challenge to the food production industry. L. monocytogenes can adhere to food-processing surfaces and form a biofilm on these surfaces [5]. The biofilm formation ability is essential for L. monocytogenes to survive and persist in food processing environments [5]. Biofilm is composed of microbial cells and self-produced extracellular polymeric substances (EPS), including polysaccharides, nucleus acids, proteins and lipids [6]. The EPS forms unique three-dimensional (3D) spatial structures and provides mechanical stability to the biofilm [6]. Notably, biofilm is closely related to functional properties, such as decreasing the efficiency of cleaning treatments and providing in-biofilm located bacteria with high resistance to antimicrobial agents [6]; for these reasons, bacterial biofilm is a severe threat to food safety. Thus, to secure public health, it is important for the food industry to find effective approaches to control the biofilm formation of L. monocytogenes. Sertraline, a selective serotonin reuptake inhibitor originally commercialized as an antidepressant drug, is reported to possess antimicrobial activity against a wide range of bacteria, such as Salmonella spp., Shigella spp., Staphylococcus spp., Staphylococcus epidermidis, Streptococcus spp., Vibrio spp., etc. [7,8]. Although the antimicrobial activity of sertraline has attracted attention, it is not fully elucidated how this compound performs its functions, and little is known about its effects on biofilm and virulence gene expression. In this study, we examined the antimicrobial activity of sertraline against L. monocytogenes and further analyzed its effects on biofilm formation and the virulence gene expression of L. monocytogenes strains. The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) values of sertraline for the L. monocytogenes strains were 16–32 μg/mL and 64 μg/mL, respectively (Table 1). Interestingly, the treatment of 64 μg/mL sertraline sharply decreased the bacterial concentration of both L. monocytogenes strains, which reached 0 CFU/mL at 12 h (Figure 1), indicating the bactericidal effect of sertraline on L. monocytogenes. Fluorescent staining can distinguish living, damaged and dead cells, which can indicate the viability of the bacteria [9]. The results depicted by the standard curves illustrated a good linear relationship (R2 = 0.99) between the green fluorescence intensity at 530 nm and the percentage of viable bacteria of L. monocytogenes ATCC 11915 and L. monocytogenes 001 (Figure S1). When exposed to 16, 32 and 64 μg/mL of sertraline, the green fluorescence intensity of L. monocytogenes 11915 decreased by 29.1%, 38.7% and 46.9%, respectively, and the green fluorescence intensity of L. monocytogenes 001 decreased by 45%, 47.9% and 50.9%, respectively (Figure 2). ATP is essential for many cell functions including substance transportation across the cell membrane [10]. When the cell membrane is impaired, the intracellular ATP decreases as a result of the decreased synthesis and increased hydrolysis of the ATP and the loss of inorganic phosphate through the highly compromised permeable cell membrane [11]. Compared with the control group, the intracellular ATP concentrations (Figure 3) and the intracellular pH (pHin) (Figure 4) of L. monocytogenes were significantly reduced when the concentration of sertraline increased from 16 to 64 μg/mL. The pHin of normal L. monocytogenes ATCC 11915 was 7.87. After exposure to 16, 32 and 64 μg/mL sertraline, the pHin decreased to 7.55, 7.46 and 6.95, respectively. The pHin of normal L. monocytogenes 001 was 7.13. After adding 16, 32 and 64 μg/mL sertraline, the pHin decreased to 7.09, 6.91 and 6.82, respectively. Compared with the control group, the percentage of biofilm formation of L. monocytogenes ATCC 11915 decreased by 67.9%, 50.4%, 12.8% and 9.4% after treatment with 64, 32, 16 and 8 μg/mL sertraline for 12 h, respectively; similarly, the biofilm formation percentage of L. monocytogenes 001 decreased by 86.0%, 50.9%, 23.3% and 5.2%, respectively (Figure 5). Normal L. monocytogenes cells have a smooth surface, a complete structure and an elongated rod shape (Figure 6). After treatment with sertraline concentrations higher than 16 μg/mL, the cell surface of L. monocytogenes became irregular and showed varying degrees of contraction and intercellular aggregation. The percentage of damaged cells and the degree of damage increased as the concentration of sertraline increased. The microstructural observations were in accordance with the findings on the impaired cell membrane integrity, decreased intracellular ATP concentration and reduced pHin under the application of sertraline as observed in this study. The results showed that sertraline had different effects on the 10 virulence genes (hly, argA, prfA, degU, actA, flaA, sigB, ltrC, sufS and sufU) of L. monocytogenes. Compared with the control group, sertraline at 0.1 and 1 μg/mL significantly inhibited the expression of these virulence genes in the L. monocytogenes ATCC 11915 strain (Figure 7). However, when the concentration was higher at 2–8 μg/mL, sertraline exposure weakened the inhibition effect on actA, flaA, sigB, ltrC, sufS and sufU gene expression. Especially when the sertraline concentration exceeded 4 μg/mL, the expression of hly, argA, prfA and degU genes was up-regulated. Sertraline had slightly different inhibition effects on the virulence gene expression of L. monocytogenes 001 (Figure 7). Sertraline at 0.1 μg/mL resulted in down-regulation in the expression of prfA, degU, actA, flaA, sigB, ltrC and sufS genes. When the concentration was higher than 1 μg/mL, the inhibitory effect of sertraline on virulence gene expression was weakened. Sertraline at 4 and 8 μg/mL increased the expression of all evaluated virulence genes. Previous studies have reported on the antimicrobial activity of sertraline with a MIC value of 4–128 μg/mL for S. aureus, 8–128 μg/mL for B. subtilis, 128 μg/mL for Candida albicans and 4–256 μg/mL for E. coli [12,13]. The antimicrobial activity of sertraline is highly species dependent. In this study, the intrinsic antibacterial activity of sertraline against the L. monocytogenes strains was similar to the activity reported in S. aureus and B. subtilis [14]. In addition, consistent with previous studies, sertraline exhibited a bactericidal effect against L. monocytogenes at 2 × MIC concentrations. Although not completely elucidated, the intrinsic antimicrobial activity of sertraline may be due to the benzene rings in the structure [12]. The bacterial cell membrane keeps the internal environment of the cell stable and maintains the normal metabolic function and energy transfer of the cell [10,15]. When the cell membrane is damaged, phosphate bonds and ion gradients as well as the energy (such as pH and ATP) transfer will change [16]. Thus, intracellular ATP and pHin are good indicators of the integrity of the cell membrane. In the present study, we observed that sertraline can interact with the cell membrane of L. monocytogenes as reflected by the decreased intracellular ATP and pHin. These findings are in accordance with previous studies, which demonstrated that sertraline can destroy the cell membrane of H. pylori [17] and, when combined with polymyxin, can significantly affect the ability of Acinetobacter baumannii, K. pneumoniae and P. aeruginosa to reshape their outer membranes [18]. In addition, sertraline is reported to cause Candida cell death by blocking the mitochondrial respiration and significantly decreasing transmembrane potential [19]. These results together indicate that sertraline most likely interacts with the bacterial cell membrane. L. monocytogenes is a consistent source of cross-contamination, both in housing storage and food processing environments, and biofilm is easily formed on contaminated surfaces [5,20]. Previously, sertraline was shown to reduce the biofilm formation in different species, e.g., in Candida spp. [21]. In the present study, sertraline was found to have strong inhibitory effects on the biofilm formation ability of L. monocytogenes in food processing environments. The virulence gene expression of L. monocytogenes involves several key steps: host cell adhesion and invasion, intracellular proliferation and motility and intercellular diffusion [22]. Specific bacterial factors are involved in each stage. For example, the hly gene, an important pathogenic factor, can help the bacteria escape the vacuole and interact with the host [23,24]. The prfA gene can regulate the expression of other virulent genes and control the biofilm formation [25]. The actA gene is required for the intracellular movement of L. monocytogenes in host cells [26]. The agrA gene affects biofilm formation and assists bacteria in intracellular invasion [1,27]. The degU and flaA genes are thought to stimulate the synthesis of bacterial flagella [28]. The sigB and ltrC genes have been shown to be involved in the low temperature adaptation of L. monocytogenes [29,30]. The sufS and sufU genes are associated with bacterial pathogenicity and virulence [31]. In this study, we found sertraline regulated the expression of the virulence genes of L. monocytogene; however, the regulation was not in a dose-dependent manner. In concentrations as low as 0.1 and 1 μg/mL, the expression levels of most virulence genes were down-regulated. However, high concentrations of sertraline increased the virulence gene expression. In a previous in vivo study, high concentrations of sertraline exacerbated pathological outcomes in chickens infected with resistant E. coli [30]. These results suggest sertraline might have multi-effects on virulence gene expression. Nonetheless, the inhibitory effect of a low concentration of sertraline on virulence gene expression indicates the potential application of sertraline on modulating the virulence of pathogens. Further modification of sertraline or synthesis of its structural analogue is expected to improve its inhibitory effect on pathogens. Sertraline (purity ≥ 98%) was purchased from Shanghai Macklin Biochemical Co., Ltd. (Shanghai, China). The L. monocytogenes strains that were used are shown in Table 1. The strains were inoculated with brain heart infusion broth (BHI) and cultured at 37 °C for 16 h. The MIC values of sertraline against the L. monocytogenes strains were determined with the broth microdilution method [32]. Briefly, the L. monocytogenes strains were grown aerobically overnight at 37 °C on BHI broth. Then, the colonies were suspended in 0.9% NaCl and adjusted to 0.5 McFarland standard with a SensititreTM nephelometer (Thermo-Fisher Scientific, Eugene, OR, USA). Subsequently, the suspensions were diluted 100-fold in Mueller–Hinton broth (MHB), and 100 μL of the dilution was transferred to the wells of a sterile 96-well plate that had different concentrations of sertraline in MHB (100 μL). The final concentrations ranged from 2 to 128 μg/mL. The positive controls contained bacteria inoculum only, whereas the negative controls contained MHB only. The lowest concentration of the compounds that resulted in no visible growth of the test organisms was determined as the MIC. The MBC was the lowest concentration at which microbial growth could not be observed on the medium [33]. All experiments were determined in biological triplicates. The growth and viability assays were performed as previously described [9]. Briefly, L. monocytogenes ATCC 11915 and a selected L. monocytogenes isolate (L. monocytogenes 001), previously obtained from a food product, were applied as test strains. The assay was prepared at 37 °C for 12 h with continuous shaking. The strains were grown overnight, resuspended in BHI broth to reach an OD600 = 0.1 and then exposed to 0, 16, 32 and 64 μg/mL of sertraline. The samples (100 μL) were collected at seven different time points during the 12 h period. Then, the samples were serially diluted, spread on MHA plates and incubated at 37 °C for 18 h followed by successive counting. The experiment was performed in triplicate, and the results were expressed as the average Log10 CFU/mL. The influence of the sertraline treatments on the membrane integrity of L. monocytogenes ATCC 11915 and L. monocytogenes 001 was assessed using the LIVE/DEAD BacLightTM Bacterial Vitality Kit (Thermo-Fisher Scientific, Eugene, OR, USA) as previously reported [34]. Briefly, standard samples were first prepared to construct a standard curve. The strain cultures were grown to the late exponential phase and then centrifuged, washed two times and re-suspended in 0.85% NaCl or 70% isopropyl alcohol (for the killed bacteria). Subsequently, the suspensions were incubated at room temperature for 1 h with mixing every 15 min. After incubation, the samples were pelleted two times with centrifugation (10,000× g, 10 min) and resuspended in NaCl to reach an OD600 = 0.5. Different viable cell proportions (0%, 10%, 50%, 90% and 100%) were utilized as the standard samples. A working stain solution (2×) was prepared by adding 6 μL of SYTO 9 and 6 μL of propidium iodide (PI) to 2 mL of filter-sterilized water. The cultures of the strains were grown overnight and then adjusted to an OD600 = 0.5 followed by treatment with sertraline at 0, 16, 32 and 64 μg/mL for 15 min at 37 °C. After treatment, each culture was centrifuged and resuspended in 0.85% NaCl. Then, an equal volume of 100 μL of cell suspension and working stain solution (2×) was added to the black opaque 96-well microtitration plates (Corning, New York, NY, USA) and mixed thoroughly. The mixture was cultured in darkness at 25 °C for 15 min. The fluorescence was determined using a multifunctional enzyme marker (BioTek, Winooski, VT, USA). The green (530 nm) and red (630 nm) emission integral intensities of the suspension excited at 485 nm were obtained three times via wavelength measurement. The influence of sertraline on the intracellular ATP concentrations of L. monocytogenes ATCC 11915 and L. monocytogenes 001 was assessed as described previously [10]. Briefly, the overnight cultures of the strains were harvested with centrifugation (5000× g, 5 min). Then, the cells were washed three times with 0.1 mol/L of phosphate-buffered saline (PBS, pH 7.0) and resuspended in PBS to achieve an OD600 = 0.5. Subsequently, sertraline was added to each tube to achieve the final concentrations of 0, 16, 32 and 64 μg/mL and cultured at 37 °C for 30 min. The ATP was ultrasonically extracted on ice and centrifuged (5000× g, 5 min). The supernatant was kept on ice to avoid loss of ATP. The ATP content was determined using an adenosine triphosphate detection kit with a multifunctional enzyme marker (BioTek, Winooski, VT, USA) following the manual’s instructions (Beyotime Biotechnology, Shanghai, China). A good linearity was found between intracellular ATP content and luminescence (R2 = 0.99). The influence of sertraline on the pHin of L. monocytogenes ATCC 11915 and L. monocytogenes 001 was determined according to a modified method of Wang et al. [35]. First, the calibration curve was constructed by measuring a series of fluorescence intensities of the pH buffers with values in the range of 3 to 10. The buffers consisted of 50 mmol/L KCl, 50 mmol/L Na2HPO4·2H2O, 50 mmol/L glycine and 50 mmol/L citric acid, and they were adjusted with either NaOH or HCl. The pHin and pHout were equilibrated by adding 10 μmol/L valinomycin and 10 μmol/L nigericin. A total of 300 μL of the overnight-cultured strains was transferred into 30 mL BHI broth and cultured at 37 °C for 8 h. After centrifugation (5000× g, 10 min), the cells were washed two times with 50 mmol/L HEPES buffer (containing 5 mmol/L EDTA, pH = 8) and resuspended in 20 mL buffer. Then, 3 μmol/L of the probe (carboxyfluorescein diacetate succinimidyl ester; cFDA-SE) (Meilunbio, Dalian, China) was added and cultured at 37 °C for 20 min. The cells were subsequently washed with 50 mmol/L potassium phosphate buffer added with 10 mmol/L MgCl2 (pH = 7.0), resuspended in 10 mL buffer and subsequently added with 10 mmol/L glucose and cultured at 37 °C for 30 min to eliminate the unbound cFDA-SE. The obtained particles were washed two times using the above mentioned method and suspended in 50 mmol/L potassium phosphate buffer on ice. Sertraline was added to the treated cell suspension to obtain the final concentrations of 0, 16, 32 and 64 μg/mL. Then, the mixture was transferred into a black opaque 96-well microtiter plate. After treatment for 20 min, the fluorescence intensity was measured under the excitation wavelengths of 440 and 490 nm with an emission wavelength of 520 nm at 25 °C by using a multifunctional enzyme marker (BioTek, Winooski, VT, USA). The pHin was determined as the ratio of the fluorescence signals at the pH sensitive wavelength of 490 nm and pH insensitive wavelength of 440 nm. The fluorescence of the cell-free controls was measured and deducted from the fluorescence of the samples. The effect of sertraline on biofilm formation was investigated according to the method described previously [6,36]. Briefly, the concentration of the bacterial solution was adjusted to an OD600 = 0.5. Then, 100 μL of the bacterial solution was added into a 96-well plate and incubated at 37 °C for 6 h to form biofilms. Sertraline was added to the treated cell suspension to obtain the final concentrations of 0, 16, 32 and 64 μg/mL. MH broth and the bacterial solution were used as the negative and positive controls, respectively. After treatment for 16 h at 37 °C, the plate was washed 3 times with normal saline and dyed with 125 μL 1% crystal violet solution. The optical density at 595 nm was determined with a multifunctional microplate analyzer (BioTek, Winooski, VT, USA). The biofilm formation ability was the value after deducting background staining. The FESEM assays were prepared as previously reported with minor modifications [9]. Briefly, the cells (OD600 = 0.5) were treated with sertraline at 0, 16, 32 and 64 μg/mL and then cultured at 37 °C for 4 h. The cultured cells were centrifuged (5000× g, 10 min), washed two times with 0.85% NaCl, resuspended in 2.5% glutaraldehyde solution and cultured at −4°C for 10 h for fixation. Then, the cells were centrifuged (5000× g, 10 min) and dehydrated in gradient concentrations of ethanol (30%, 50%, 70%, 80%, 90% and 100%). Finally, the samples were fixed onto the FESEM support, sputter-coated with gold under vacuum and examined using a FESEM apparatus (Bruker, Berlin, Germany). The effect of sertraline at different concentrations on the expression of 10 virulence genes (hly, argA, prfA, actA, degU, flaA, sigB, ltrC, sufS and sufU) of L. monocytogenes was detected with real-time quantitative PCR using the primers listed in Supplementary Table S1 [23]. First, the total RNA of L. monocytogenes cultures treated with sertraline at 0, 16, 32 and 64 μg/mL for 4 h was extracted with a commercial kit (Magen, GuangZhou, China) following the manufacturer’s instructions. The concentration and quality of the RNA were assessed with agarose gel electrophoresis and a Nanodrop 2000® spectrophotometer (ThermoFisher Scientific, Waltham, MA, USA), respectively. Then, the cDNA was synthesized with a commercial kit (Vazyme Biotech Co., Ltd., Nanjing, China) following the manufacturer’s instructions. The real-time PCR system was performed with the 2×Ultra SYBR Green qPCR Mix kit (CISTRO, GuangZhou, China) with the CFX96TM Real-Time System (Bio-rad, Hercules, CA, USA). The 16S rRNA gene was used as the reference gene [14]. The relative expression levels of the virulence genes were calculated according to the Ct values. All experiments were conducted in triplicate. The statistical analyses were performed in GraphPad Prism ver. 8.0.1 (San Diego, CA, USA). The data were presented as the mean ± standard deviation (n = 3), and the differences between the mean values were tested via the one-way ANOVA. The differences were considered to be significant at p < 0.05. This study investigated the antimicrobial activity of sertraline against L. monocytogenes. Sertraline caused damage of the cell membrane and decreased the intracellular ATP and pHin of L. monocytogenes. Moreover, sertraline significantly inhibited biofilm formation and regulated the virulence gene expression of L. monocytogenes. These results suggest that sertraline may potentially be used to control L. monocytogenes in food-processing environments, thereby reducing the risk of food contamination.

PMC10002545

Chiara Agostinis,Miriam Toffoli,Andrea Balduit,Alessandro Mangogna,Hadida Yasmin,Chiara Ragazzon,Silvia Pegoraro,Giuseppina Campisciano,Guglielmo Stabile,Gabriella Zito,Uday Kishore,Manola Comar,Federica Scrimin,Roberta Bulla,Giuseppe Ricci

Anti-Spike Antibodies Present in the Milk of SARS-CoV-2 Vaccinated Mothers Are Complement-Activating

23-02-2023

COVID-19 vaccines,antibodies,breast milk,complement system,classical pathway

Although only 0.8–1% of SARS-CoV-2 infections are in the 0–9 age-group, pneumonia is still the leading cause of infant mortality globally. Antibodies specifically directed against SARS-CoV-2 spike protein (S) are produced during severe COVID-19 manifestations. Following vaccination, specific antibodies are also detected in the milk of breastfeeding mothers. Since antibody binding to viral antigens can trigger activation of the complement classical - pathway, we investigated antibody-dependent complement activation by anti-S immunoglobulins (Igs) present in breast milk following SARS-CoV-2 vaccination. This was in view of the fact that complement could play a fundamentally protective role against SARS-CoV-2 infection in newborns. Thus, 22 vaccinated, lactating healthcare and school workers were enrolled, and a sample of serum and milk was collected from each woman. We first tested for the presence of anti-S IgG and IgA in serum and milk of breastfeeding women by ELISA. We then measured the concentration of the first subcomponents of the three complement pathways (i.e., C1q, MBL, and C3) and the ability of anti-S Igs detected in milk to activate the complement in vitro. The current study demonstrated that vaccinated mothers have anti-S IgG in serum as well as in breast milk, which is capable of activating complement and may confer a protective benefit to breastfed newborns.

Anti-Spike Antibodies Present in the Milk of SARS-CoV-2 Vaccinated Mothers Are Complement-Activating Although only 0.8–1% of SARS-CoV-2 infections are in the 0–9 age-group, pneumonia is still the leading cause of infant mortality globally. Antibodies specifically directed against SARS-CoV-2 spike protein (S) are produced during severe COVID-19 manifestations. Following vaccination, specific antibodies are also detected in the milk of breastfeeding mothers. Since antibody binding to viral antigens can trigger activation of the complement classical - pathway, we investigated antibody-dependent complement activation by anti-S immunoglobulins (Igs) present in breast milk following SARS-CoV-2 vaccination. This was in view of the fact that complement could play a fundamentally protective role against SARS-CoV-2 infection in newborns. Thus, 22 vaccinated, lactating healthcare and school workers were enrolled, and a sample of serum and milk was collected from each woman. We first tested for the presence of anti-S IgG and IgA in serum and milk of breastfeeding women by ELISA. We then measured the concentration of the first subcomponents of the three complement pathways (i.e., C1q, MBL, and C3) and the ability of anti-S Igs detected in milk to activate the complement in vitro. The current study demonstrated that vaccinated mothers have anti-S IgG in serum as well as in breast milk, which is capable of activating complement and may confer a protective benefit to breastfed newborns. The complement (C) system is an innate immune surveillance system consisting of a complex network of plasma and membrane-associated proteins that are designed for the recognition and clearance of pathogens, neoplastic antigens, immune complexes, and apoptotic bodies, in order to maintain physiological homeostasis [1,2,3,4]. The C system mainly acts as an enzyme lytic cascade through three broad effector functions following the recognition of activating surfaces or ligands: opsonisation and enhancement of phagocytosis, contribution to inflammatory responses mainly via anaphylatoxins C3a and C5a, and target-cell lysis by C5b-9 membrane attack complex [5,6]. Antibody-dependent C-system activity is one of the most efficient mechanisms against infectious pathogens (e.g., bacteria, viruses, fungi, parasites). Secreted antibodies, such as IgM and/or IgG, bind to pathogens, which are recognized by the classical pathway initiation molecule C1q via the globular head (gC1q) region [7,8]. Secretory dimeric IgA (but not monomeric IgA) can also activate the C system via the lectin pathway, through the binding of mannose binding lectin (MBL), ficolins, or collectin-11 [9,10]. A number of recent studies have suggested that C-system hyperactivation, directly or indirectly by SARS-CoV-2 proteins, contributes to the pathogenesis in COVID-19 [11,12,13,14]. Specifically, reports highlighted the ability of MBL, ficolin-2, and collectin-11 to bind spike glycoprotein (S) and nucleocapsid protein (N) of SARS-CoV-2, and promote C3b and C4b deposition [15]; IgG and IgM antibodies directed against the receptor-binding domain (RBD) of S are considered as key players for the classical pathway activation [16]. An increased risk of hypertensive disorders in pregnant women who are exposed to SARS-CoV-2 in their early stage of pregnancy has been recently reported who are likely to develop pre-eclampsia [12,17]. Despite the higher risk of severe disease, the administration of COVID-19 vaccines during pregnancy was initially limited, as pregnant women were excluded from pre-authorization clinical trials due to safety concerns. Once a lack of adverse effects was demonstrated, COVID-19 vaccination was gradually extended to pregnant and breastfeeding women [18,19]. Four COVID-19 vaccines initially received authorization for emergency use by the EMA: two mRNA vaccines from Pfizer-BioNTech (BNT162b2) and Moderna (mRNA-1273), and two adenovirus-vectored vaccines from Johnson & Johnson/Janssen (Ad26.CoV2.S) and Oxford-AstraZeneca (AZD1222, Vaxzevria) [20,21]. COVID-19 vaccines incorporated S to elicit robust T-cell responses, along with high anti-viral neutralizing antibody production by B cells. Although the majority of the trials have examined the antibody responses in the blood of vaccinated and infected individuals, few studies have assessed the possible presence of anti-SARS-CoV-2 antibodies in breast milk. Notably, mother’s milk is not only the gold standard for providing nutrients (including carbohydrates, lipids, proteins, vitamins, and minerals, as well as bioactive molecules, such as cytokines, growth factors, and oligosaccharides) [22], but also the first source of antibody-mediated immune protection to the immunologically naïve and immature new-borns [7,23,24]. Human milk contains a variety of Igs, IgA being the most abundant (>90%) [25], followed by IgM and IgG [26]. Various studies evaluating S-specific Igs in the breast milk have shown high levels of IgA and IgG, and negligible IgM levels [27,28,29]. Vaccination against SARS-CoV-2 during the lactation period has been shown to significantly augment the level of antibodies in breast milk [30,31,32,33]. Interestingly, it has been reported that maternal vaccination with an mRNA-based vaccine during lactation resulted in higher SARS-CoV-2 antibody response in human milk compared to vector-based vaccines [31,34,35]. Since the possible contribution of the C system to COVID-19-related maternal immunity has not yet been examined, the present study aimed to investigate not only the presence of IgG and IgA against S in the serum and in the milk of SARS-CoV-2-vaccinated breastfeeding healthcare and educational workers, but also their capability to activate the C system. First, we tested for the presence of anti-S IgG and IgA in serum and milk using a cohort of 22 SARS-CoV-2-vaccinated women (n = 17 with Pfizer–BioNtech, n = 4 with Oxford–AstraZeneca, n = 1 with both vaccines; Table S1). In order to consistently compare IgG with IgA levels, as well as the values obtained for serum and milk samples, we used an immune serum pool (COVID-19-recovered and -vaccinated patients) as a standard calibrator. The calibrator was previously titred, obtaining a 1:75,000 titre for both IgG and IgA (Table S2). To determine this titre, plates were coated with S, then serum (1:50–1:200 dilution) and milk samples (1:2–1:4 dilution) were incubated and the binding of Igs was detected using specific anti-human IgG, anti-human serum IgA, or anti-human secretory IgA. A milk pool showed high positivity for IgA antibodies, thus it was used as a calibrator for secretory IgA. As shown in Figure 1, patients presented a wide range of IgG and IgA levels in serum as well as milk, although the Ig levels were 200–300 times higher in serum than in breast milk. The correlation (Pearson test) between serum and milk Ig levels revealed a statistically significant value (p < 0.0001, r2 = 0.62) for IgG, whereas no statistical significance was obtained for IgA levels. Surprisingly, no correlation was observed between Ig levels and days elapsed between vaccination and sample collection (Figure S1). By identifying patients with low IgA levels, we noticed that almost all were vaccinated with Oxford–AstraZeneca (patients #2, #4, #6, #21, and #22), while patients displaying very high levels of IgA had previously contracted SARS-CoV-2 (patients #18, #19, and #20). Since the main aim of this study was to understand whether the presence of anti-S antibodies in breast milk after anti-SARS-CoV-2 vaccination may be responsible for C-system activation, we first analyzed the levels of C1q, MBL, and C3, the first recognition subcomponents of the classical, lectin, and alternative pathways, respectively (Figure 2). Thus, we measured their levels in milk and sera of our cohort of patients using commercial ELISA kits. Our results indicated that the C1q level in milk was about 500–1000 times lower than that in serum; MBL, when physiologically present, was about 3000–5000 times lower in milk than in serum; and C3 appeared to be only 20–70 times lower in the milk than in serum. For all three subcomponents, we could not establish a statistically significant correlation between their serum and milk levels (Figure 2B,D,F). Then, we evaluated complement C1q, MBL, and C3 levels in the milk of our cohort of patients via Wieslab ELISA assay, a test designed for analyzing C-system functionality in the serum. Milk samples failed to validate activation of all the three C-system cascades, whilst all the sera were found to be C-sufficient (Figure S2). Finally, we investigated the ability of anti-S Igs present in the milk of vaccinated women to activate the C system in vitro. Following the binding of serum or milk IgG to S, a pool of AB Rh+ sera was added to the wells as a source of C-system components, and subsequently, the deposition of C1q, C3, and C9 neo-antigens was detected by adding specific antibodies (Figure 3). In addition, a wide variability of C-system activation components was also observed; the milk samples identified for their ability to strongly activate the C system were obtained independently from patients having received Pfizer–BioNTech or Oxford–AstraZeneca vaccine or being COVID-recovered. Almost all the milk samples were able, even though at low levels, to induce C1q (Figure 3A) and C3 (Figure 3D) deposition, but only few samples led to the activation of the whole cascade until the formation of the C9 polymer (Figure 3G). We observed a significant correlation between C1q or C3 deposition and IgG presence, both in serum and in milk (Figure 3B,C,E,F), whereas no correlation for C9 neoantigen deposition was found in milk (Figure 3H,I). Since the activation of the classical pathway, induced by the binding of specific anti-S antibodies to their target, could be potentially blocked by the activity of C1-inhibitor, we also evaluated the presence of this protein in the serum and milk of the enrolled women. We found high levels of this C-system inhibitor in breast milk compared to the other C-system components previously measured, although levels were around 100–200 times lower than in the serum (Figure 4). During the COVID-19 pandemic, a number of studies aimed to characterize the protective antibody-mediated viral neutralization in response to SARS-CoV-2 infection in order to unveil the mechanisms of SARS-CoV-2–host interaction, mainly focusing on host immune response and C-system activation. Garred and co-workers reported that the anti-SARS-CoV-2 (RBD) IgG response in convalescent plasma was mainly driven by IgG1 and IgG3 subclasses, the main ligands for C1q-mediated activation of the classical pathway [36]. Furthermore, they observed that C4b, C3bc, and C9 polymer deposition due to antibodies specifically directed against SARS-CoV-2 could be significantly correlated with both IgG levels and disease severity, suggesting that patients with high IgG levels or severe symptoms may exhibit a more powerful C-system activation during viral infection [36]. Moreover, Cunningham and colleagues demonstrated that C1q binding to SARS-CoV-2 Igs in vitro strongly correlated with antibody responses, whereas the detection of downstream C-system components (C4b, C3b, and C5b) showed some variability depending on the group analyzed. In particular, the deposition of C3b–C5b on S was consistently observed in convalescent hospitalized patients, but not in the non-hospitalized group [37]. Interestingly, a few studies also investigated the presence of anti-S specific Igs in the milk of vaccinated mothers [27,30,31,32,33,34,35,38,39,40,41], but they did not assess if these antibodies were complement-activating. A recent study demonstrated a robust secretion of SARS-CoV-2-specific IgA and IgG in the breast milk for 6 weeks following vaccination [42]; similar findings were reported in women who had recovered from COVID-19 [33,43]. In the current study, we initially confirmed the presence of anti-S specific IgG and IgA in serum as well as in milk samples and observed a range of antibody titres among vaccinated women. Consistent with an earlier study [44], we found that almost all women/participants with low IgA levels had received the Oxford–AstraZeneca vaccine, whereas individuals with very high levels had previously contracted COVID-19. As expected, only serum and milk IgG levels showed a statistically significant correlation, since they are both produced by the same plasma cells in secondary lymphoid organs and bone marrow whilst serum and secretory IgA are characterized by different plasma-cell origins (i.e., bone marrow for serum IgA and mucosa-associated lymphoid tissue for secretory IgA) [45,46,47,48], as summarized in Figure 5. Surprisingly, no correlation was observed between Ig levels and number of days post-vaccination, a result assuring us that samples collected at different timepoints could be comparable. In fact, the persistence of neutralizing antibodies following COVID-19 vaccination is currently under investigation. With respect to IgA, this may be explained by studies showing that IgA levels in milk did not rise further when measured after 18 days following the second dose of vaccine, in contrast to the significant increase in IgG levels after the second immunization [49]. Antibodies elicited by mRNA-1273 vaccine were detectable in the serum until six months [50] and by AZD1222 until three months [51], whilst Ad26.COV2.S and BNT162b2 vaccines have been shown to give shorter duration protection [52]. Despite also providing consistent T-cell mediated immune responses, BNT162b2 vaccine induced anti-S IgG production 11 days after the first dose administration, showing the peak at day 21, whereas the AZD1222 elicited a neutralizing effect 22 days after the first dose [53]. Historically, very little information is available about the importance of the C system and its contribution to mucosal immunity in human breast milk, mainly due to the low levels of C-system components detected in mature milk [26]. Moreover, the relative contribution of C-system components transported from the serum and those that are locally produced is still poorly understood [54]. In order to explore potential activation of the C system in breast milk due to immunization against COVID-19, we subsequently examined the amounts of C1q, MBL, and C3, the first recognition subcomponents of the classical, lectin, and alternative pathways, respectively. Our findings showed that C1q levels in the milk were 500–1000 times lower than those in serum, and MBL levels, when physiologically present, were 300–5000 times lower. C3 was the most abundantly detected among C-system activators, being only 20–70 times less abundant in the milk than in serum samples. Despite the detection of C-system activators, even though at very low levels, we failed to demonstrate the capability of C-system components present in the milk samples to activate C cascades using Wieslab assay. This may have been due to the relatively small amount of C components present in breast milk compared to serum, highlighting the need for a more sensitive assay. Moreover, some C-component concentrations in human milk seemed to be quite similar to those detectable in the serum when considering colostrum and early lactational milk, but they significantly fell during the first few months of breastfeeding [55]. To overcome these technical limitations, we investigated the ability of anti-S Igs, previously detected in the milk of vaccinated women, to activate the C system in vitro. Even though at low levels, nearly all milk samples caused C1q and C3 deposition, but only a limited number of samples induced the formation of C9 polymer. Since in vitro activation of the C system leading to C3b deposition on killed bacteria has been already documented in human milk [56], it is tempting to hypothesize that C-system activation may mainly exert effector functions of opsonization. Interestingly, we observed a significant correlation between C1q or C3 deposition and IgG presence, both in serum as well as milk, whereas no correlation was noticed for C9 deposition in milk. In accordance with Lamerton et al. [37], we observed that patients presenting antibodies able to induce C9-deposition in the milk belonged to the COVID-19 recovered group (#18 and #19) and, surprisingly, to the AstraZeneca-vaccinated group (#4, #21, and #22). Even though an excessive or uncontrolled C-system activation has already been established as a well-known pathogenic player in severe COVID-19, the potential involvement of C-system factors in protective immunity against SARS-CoV-2 has been largely neglected. Since human milk represents a vehicle to transfer maternal immunity against infections to infants via bioactive factors, it is likely that Igs contained in the milk of vaccinated women maintain their ability to activate the C system along respiratory mucosa and the gastrointestinal tract of the breast-fed newborns, ensuring protection against COVID-19. This study sheds light on the possible physiological and protective significance of the C system in vaccination-driven maternal immunity. The study cohort comprised twenty-two breastfeeding healthcare professionals and school workers who received the COVID-19 vaccine and were randomly selected from the cohort previously described by Scrimin et al. [41] (Table 1). Between 1 February and 30 July 2021, women experiencing physiological pregnancies and normal early postpartum in and around Trieste (Italy) were recruited by a perinatal study group at the Institute for Maternal and Child Health IRCCS “Burlo Garofolo” (Trieste, Italy). Nasopharyngeal swabs for SARS-CoV-2 testing were performed on all the enrolled mothers and newborns at the time of enrolment, one week prior to enrolment, and one week following enrolment. At the time of sample collection, the participants had no symptoms, and all the tests were found to be negative. Serum and milk samples were collected from each woman (maximum 75 days after vaccination). After nearly complete breast expression, milk samples were obtained, including the foremilk and the hindmilk. Serum and milk samples were both immediately delivered to the laboratory for storage and analysis. To separate the cells from the fat content of the milk, samples of breast milk were centrifuged at 800× g for 10 min at 4 °C. The study was approved by the Internal Review Board of the Institute for Women and Child Health IRCCS “Burlo Garofolo” (IRB 06/2021). All participants signed detailed, informed consents and were over the age of 18. For the Elisa assay, 96-well plates were first coated with 2 µg/mL SARS-CoV-2 S recombinant protein (S1/S2) (RP-87671, Invitrogen, Waltham, MA, USA) diluted in phosphate-buffered saline (PBS) and incubated overnight at 4 °C. Plates were then washed three times with PBS + 0.1% Tween 20 (PBS-T) and blocked with 3% w/v skimmed milk in PBS-T for 1 h at RT. Patient sera (1:200) and milk (1:2) diluted in PBS-T + 1% w/v skimmed milk were added and incubated 2h at RT. After washing three times, antibodies for Ig detection were added for 1 h at RT: anti-human IgG-alkaline phosphatase (AP)-conjugate (1:50,000, Sigma Merck, St. Louis, MO, USA), anti-IgA antibody (1:700, Sigma Merck) for serum samples and anti-IgA secretory component antibody (1:250, Sigma Merck) for milk samples. Anti-rabbit IgG AP-conjugate (1:10,000, Sigma Merck) and anti-mouse polyvalent Igs (G,A,M) AP-conjugated (1:30,000, Sigma Merck) were incubated 1 h at RT for the detection of anti-IgA and anti-secretory IgA, respectively. The binding of secondary antibodies was revealed using p-nitro phenyl phosphate (pNPP) as a substrate. The absorbance was read at 405 nm by PowerWave X Microplate Reader (Bio-Tek Instruments, Winooski, VT, USA). A standard curve was also included. Commercial ELISA kits were used to quantify C1q (Hycult Biotech, Uden, The Netherlands, #HK356-02), MBL (Hycult Biotech, #HK323), C3 (Abcam, Cambridge, UK, #ab108823), and C1 inhibitor (R&D Systems, Minneapolis, MN, USA, #DY2488-05) in serum and milk samples, following the instructions provided by the manufacturer. The absorbance was read by a PowerWave X Microplate Reader (Bio-Tek Instruments) spectrophotometer. The functionality of classical, alternative, and lectin pathways was determined using the commercial kit Wieslab® (Technogenetics, Milan, Italy). The assay was performed following the manufacturer’s instructions, with the exception of milk dilution (1:2). Patient sera (1:50) and milk (1:2) were incubated in microplate wells coated with SARS-CoV-2 S recombinant protein (S1/S2), as described above. In order to assess the capability of anti-SARS-CoV-2 antibodies to activate the C system, wells were then incubated with AB Rh+ pooled sera (1:100 in PBS + 2% w/v BSA + 0.7mM Ca++Mg++) for 30 min at 37 °C. After washing with PBS-T, the binding of C1q was evaluated using rabbit anti-human C1q polyclonal antibody (1:2000, Dako, Santa Clara, CA, USA) for 1 h at 37 °C and anti-rabbit IgG AP-conjugated (1:10,000, Sigma-Aldrich, St. Louis, MO, USA) as a secondary antibody. Simultaneously, the deposition of C3 was detected using goat anti-human polyclonal antibodies (1:5000, Quidel, San Diego, CA, USA) for 1 h at 37 °C and anti-goat IgG AP-conjugated (1:30,000, Sigma-Aldrich). The formation of the terminal C complex C5b-9 was assessed using anti-human C5b-9 (1:50, clone: aE11, Dako) for 1 h at 37 °C and anti-mouse polyvalent Igs (G,A,M)-AP (1:30,000, Sigma Merck). The binding was revealed with pNPP and the absorbance was read at 405 nm using a PowerWave X Microplate Reader (Bio-Tek Instruments). The experiments were run in duplicate and data are expressed as mean of values. The correlation was analyzed using the Pearson test and p-values < 0.05 were considered statistically significant. All statistical analyses were performed using GraphPad Prism software 9.0 (GraphPad Software Inc., La Jolla, CA, USA). After infection and vaccination, the presence of anti-SARS-CoV-2 antibodies in breast milk may offer a potentially protective benefit for the nursing infant, not only for the direct presence of anti-S antibodies but also for their C-system activation capability. In general, the level of IgG antibodies in serum was higher than that in breast milk but the C-system activation potential was retained in the case of milk antibodies. Furthermore, our study raises a cardinal point that the remit of vaccine efficacy should not exclusively rely on the antibody titers and highlights the importance of describing the complex immune response in its entirety.

PMC10002546

Anukriti Singh,Brian P. Hermann

Conserved Transcriptome Features Define Prepubertal Primate Spermatogonial Stem Cells as Adark Spermatogonia and Identify Unique Regulators

01-03-2023

spermatogonia,stem cells,prepubertal testis,non-human primate,human,Adark,Apale

Antineoplastic treatments for cancer and other non-malignant disorders can result in long-term or permanent male infertility by ablating spermatogonial stem cells (SSCs). SSC transplantation using testicular tissue harvested before a sterilizing treatment is a promising approach for restoring male fertility in these cases, but a lack of exclusive biomarkers to unequivocally identify prepubertal SSCs limits their therapeutic potential. To address this, we performed single-cell RNA-seq on testis cells from immature baboons and macaques and compared these cells with published data from prepubertal human testis cells and functionally-defined mouse SSCs. While we found discrete groups of human spermatogonia, baboon and rhesus spermatogonia appeared less heterogenous. A cross-species analysis revealed cell types analogous to human SSCs in baboon and rhesus germ cells, but a comparison with mouse SSCs revealed significant differences with primate SSCs. Primate-specific SSC genes were enriched for components and regulators of the actin cytoskeleton and participate in cell-adhesion, which may explain why the culture conditions for rodent SSCs are not appropriate for primate SSCs. Furthermore, correlating the molecular definitions of human SSC, progenitor and differentiating spermatogonia with the histological definitions of Adark/Apale spermatogonia indicates that both SSCs and progenitor spermatogonia are Adark, while Apale spermatogonia appear biased towards differentiation. These results resolve the molecular identity of prepubertal human SSCs, define novel pathways that could be leveraged for advancing their selection and propagation in vitro, and confirm that the human SSC pool resides entirely within Adark spermatogonia.

Conserved Transcriptome Features Define Prepubertal Primate Spermatogonial Stem Cells as Adark Spermatogonia and Identify Unique Regulators Antineoplastic treatments for cancer and other non-malignant disorders can result in long-term or permanent male infertility by ablating spermatogonial stem cells (SSCs). SSC transplantation using testicular tissue harvested before a sterilizing treatment is a promising approach for restoring male fertility in these cases, but a lack of exclusive biomarkers to unequivocally identify prepubertal SSCs limits their therapeutic potential. To address this, we performed single-cell RNA-seq on testis cells from immature baboons and macaques and compared these cells with published data from prepubertal human testis cells and functionally-defined mouse SSCs. While we found discrete groups of human spermatogonia, baboon and rhesus spermatogonia appeared less heterogenous. A cross-species analysis revealed cell types analogous to human SSCs in baboon and rhesus germ cells, but a comparison with mouse SSCs revealed significant differences with primate SSCs. Primate-specific SSC genes were enriched for components and regulators of the actin cytoskeleton and participate in cell-adhesion, which may explain why the culture conditions for rodent SSCs are not appropriate for primate SSCs. Furthermore, correlating the molecular definitions of human SSC, progenitor and differentiating spermatogonia with the histological definitions of Adark/Apale spermatogonia indicates that both SSCs and progenitor spermatogonia are Adark, while Apale spermatogonia appear biased towards differentiation. These results resolve the molecular identity of prepubertal human SSCs, define novel pathways that could be leveraged for advancing their selection and propagation in vitro, and confirm that the human SSC pool resides entirely within Adark spermatogonia. Cancer survivors are often faced with negative side-effects of their life-saving treatments, which impact their quality of life and one of the most devastating is the loss of spermatogenesis and the resulting permanent infertility [1]. Spermatogenesis is the process by which the progeny of diploid spermatogonial stem cells (SSCs) terminally differentiate to produce mature spermatozoa in the testis. For adult patients, this infertility risk of chemotherapy and radiotherapy can be mitigated by cryopreserving sperm recovered from an ejaculate for future use in medically-assisted reproductive technologies, like IVF and ICSI [2,3,4]. Prepubertal patients who are not yet producing sperm, however, cannot take advantage of this standard of care [5]. One experimental fertility preservation option that may address the challenges of prepubertal boys is cryopreservation of testicular tissue that contains SSCs prior to commencing gonadotoxic treatments [6,7,8]. The proof of principle that cryopreserved testicular tissue can be used to generate sperm using SSC transplantation has been established in a variety of species, including non-human primates [9,10]. While harvesting testicular tissue for future SSC transplantation into the testis of infertile recipients is a promising approach, utilizing this technology in humans is limited by the lack of exclusive biomarkers that can unequivocally identify therapeutic SSCs, and the lack of knowledge about their development. Moreover, testicular biopsies from prepubertal patients are of limited size and may not contain sufficient SSCs to produce robust spermatogenesis after transplantation. It has been estimated that a 1300-fold increase in SSCs is required to achieve sufficient colonization in a clinical human application [11], and thus, it may be necessary to expand SSCs in culture. Initially, attempts to culture human SSCs were derived from advances in rodent models [12,13]. Subsequently, studies have tested feeder-based conditions utilizing human embryonic stem cells-derived fibroblasts, human Sertoli cells, or an extracellular matrix such as laminin, gelatin, and hydrogel [14,15,16,17,18]. However, the lack of the ability to functionally identify SSCs and the lack of exclusive biomarkers have made it difficult to assess and repeat these experiments. Therefore, future attempts to propagate human SSCs in vitro would benefit from a better understanding of their fundamental characteristics and in vivo kinetics. Functionally, SSCs balance self-renewal with the initiation of differentiation to maintain fertility throughout the male reproductive lifespan [19]. Clermont et al. proposed one of the earliest models of SSC self-renewal called the A0/A1 model based on 3H-thymidine-labeled whole mount preparations of rat seminiferous tubules [20,21,22]. According to this model, rodent testes contain two types of stem cells, reserve (A0) and active (A1). A0 reserve stem cells exist as largely quiescent singles and pairs of type-A spermatogonia that do not contribute to ongoing spermatogenesis and only become active upon the loss of spermatogenesis due to gonadotoxic insult [23]. The active stem cell pool (A1) undergoes clonal expansion, giving rise to the differentiating A1 to A4 spermatogonia and stem cell renewal is accomplished by fragmentation of A2, A3, and A4 spermatogonial clones [23]. Although the A0/A1 model was largely supplanted by three new models describing the identity and kinetics of rodent SSCs [22,24,25,26,27,28], concepts reminiscent of Clermont’s ideas are still incorporated in our current understanding. The prevailing “Asingle model” categorizes subsets of rodent undifferentiated spermatogonia based on clone size (Asingle, Apaired and Aaligned) [25,29]. Stem cell capacity is considered to reside within the Asingle subset [30,31]. The renewal of Asingle SSCs occurs by either asymmetric cell division, where only one daughter cell of a dividing SSC retains stemness, or population-level asymmetry, where half of the stem cell divisions produce only stem cells. An alternate “fragmentation model” challenges this view, proposing that the SSC compartment is comprised of the entire undifferentiated spermatogonia pool (Asingle, Apaired, and Aaligned spermatogonia), and these clones are interconvertible and essentially equipotent [27,28]. This model is consistent with the fragmentation proposed by Clermont’s A0/A1 model and supported by the results of in vivo live imaging studies using reporter mice, which appear to show the separation of clones. This model has been criticized for a lack of definitive identification of the intercellular cytoplasmic bridges defining clones and a failure to confirm cell fate after fragmentation, and thus, remains controversial. Lastly, the “revised Asingle model” was proposed based on experiments using transgenic Id4-eGFP reporter mice, in which undifferentiated spermatogonia were sorted based on the intensity of eGFP epifluorescence into Id4-eGFPBright (SSC-enriched) or Id4-eGFPDim (SSC-depleted) subsets, which have more and less regenerative capacity upon transplantation, respectively [32,33,34]. Given the observation of phenotypic and functional heterogeneity within the Id4-eGFPBright population, which are exclusively arranged as Asingle spermatogonia, this model holds that a subset of Asingle are self-renewing SSCs (SSCultimate), while the remainder are in an intermediate state (SSCtransitory), poised to transition to a progenitor state and likely only contribute to the SSC pool upon perturbations of steady-state conditions [34]. Primate testes contain two subtypes of undifferentiated spermatogonia, Adark and Apale, identified based on the differences in their nuclear architecture and staining intensity with hematoxylin [24,35,36]. Initially, Clermont’s group proposed a linear model in which Adark SSCs produce Apale progenitor spermatogonia, which give rise to differentiating B-type spermatogonia [36]. Subsequently, this model was modified based on the observation in Vervet monkeys (Chlorocebus pygerythrus) that Adark spermatogonia did not label with a pulse of 3H-thymidine, indicating a failure to go through the S-phase and self-renew, while roughly 1/3 of Apale were labeled [37]. Moreover, Apale were cleared following cytotoxic insult, but Adark were not and subsequently began proliferating and regenerated spermatogenesis [37]. Consequently, both Adark and Apale spermatogonia were considered to have stem cell capacity with Adark taking the role of “reserve” stem cells, which only serve as a regenerative backup, while Apale spermatogonia are considered “active” stem cells, which sustain ongoing spermatogenesis in a steady-state [35,38]. Alternatively, others have posited that the low mitotic index of Adark more closely reflects the expected behavior of SSC, while the high turnover of Apale reflects their activity as transit-amplifying progenitor spermatogonia that are actively proliferating to increase the total number of germ cells [39]. More recently, based on molecular phenotyping in nonhuman primates, it has been suggested that Adark and Apale both comprise the SSC pool, but are simply in different phases of the cell cycle [40,41]. Regardless, the lack of tools to functionally and quantitatively measure SSC activity in primate species has precluded a definitive identification of the primate SSC pool. Recently, single-cell mRNA profiling studies of immature and adult human testicular cells have begun to address the challenge of identifying human SSCs by facilitating subclassification of undifferentiated spermatogonia and defining the molecular signatures for these subsets [33,42,43,44,45,46,47,48]. In human testes, spermatogonia have been classified into undifferentiated (UTF1+) and differentiating spermatogonia (KIT+, MKI67+). Undifferentiated spermatogonia have been further subdivided into at least four distinct groups with varied levels of key markers including FGFR3, GFRA1, NANOS2, NANOS3, PIWIL4, TSPAN33, and UTF1, highlighting the phenotypic heterogeneity within the undifferentiated spermatogonial compartment. There also appears to be a gradual transition in transcriptome states rather than binary on/off programs [33,44,46,49]. Despite these reports, functional evidence of human SSC capacity is not available to support SSC identity in any of these studies [33,50]. To reduce the uncertainty about the identity of prepubertal human SSCs and understand their developmental relationship with more advanced germ cells, we hypothesized that a comparative analysis of germ cells from the testes of prepubertal nonhuman primates (baboon and rhesus macaque), humans, and mice would allow us to define the most highly conserved phenotype of prepubertal human SSCs. To test this hypothesis, we generated testis single-cell transcriptomes from two prepubertal baboons along with two prepubertal rhesus macaques and defined the spermatogonial transcriptome signatures in both species. To identify a conserved primate SSC gene expression signature diagnostic of the immature human SSC population, we compared the nonhuman primate spermatogonia with their prepubertal human [42,43,44] and mouse [33] counterparts. By determining conservation in gene expression profiles across mammals and unique features in primates, we derived a more refined phenotype for SSCs in the prepubertal human testis, identified pathways that are overrepresented uniquely in primates, and confirmed that SSCs are Adark spermatogonia. To investigate the cellular diversity during prepubertal testis development in primates, we re-analyzed published human testis single-cell datasets from two newborns [42], two infants [44], and two juveniles [43] (Figure 1A and Table S1), and we performed single-cell RNA-seq (10× Genomics) on unselected testis cells from two prepubertal baboons (newborn and 26 months old; Figure 1B and Table S1) and two prepubertal rhesus monkeys (15 and 20 months old; Figure 1C and Table S1). In addition, we mined existing germ cell transcriptomes from postnatal day (P) 6 mice [33] in order to define the evolutionarily conserved and divergent programs (Figure S2 and Table S1). A total of 92,867 cells (30,752 human, 23,219 baboon, 25,342 rhesus, and 13,554 mouse cells) passed quality control and were used for downstream analyses (Table S1). Using an unsupervised analysis approach for testis cells from each species, we identified 16, 21, and 23 distinct clusters of cells for baboon, rhesus macaque, and human testis cells, respectively (Figure 1A–C). We observed no bias among the replicates for each species based on cell distributions among the clusters (Figure S1A and Table S2). The clusters were subsequently annotated using established cell-type markers for testicular somatic and germ cells (Figure S1B–D). In the baboon testes, the major cell types consisted of germ cells (DDX4+), Sertoli cells (SOX9+, WT1+, DHH+), Leydig cells (IGF1+, INSL3+, STAR+), peritubular myoid cells (ACTA2+, MYH11+), and endothelial cells (PECAM1+) (Figure 1D, Figure S1B, Table S2). Similarly, the rhesus testis cells were comprised of germ cells (DDX4+), Sertoli cells (SOX9+, DHH+), Leydig cells (INSL3+, STAR+), peritubular myoid cells (ACTA2+, MYH11+) endothelial cells (PECAM1+), and macrophages (AIF1+) (Figure 1E, Figure S1B, Table S2). From the human testis cells, we found germ cells (DDX4+), Sertoli cells (SOX9+), Leydig cells (INSL3+, STAR+, IGF1+), peritubular myoid cells (ACTA2+, MYH11+), endothelial cells (PECAM1+), macrophages (AIF1+), peripheral glia (S100B+, LGI4+), and monocytes (HSPA6+, RND3+, TCMI+) (Figure 1F, Figure S1B, Table S2). Notably, we found interspecies differences in terms of the cell types present. Specifically, we identified a testicular macrophage population in the human and rhesus testis cells, which was absent from the baboon testis cells. We also noted differences in expression of somatic cell type specific markers, including an absence of WT1 mRNAs in rhesus Sertoli cells and a lack of DHH in human Sertoli cells. To gain a better understanding of prepubertal germ cell heterogeneity in humans, we performed a focused analysis of the 1451 germ cells identified in the human testis cell cluster 11 (Figure 1A). Unsupervised re-clustering of the human germ cells visualized in a UMAP projection revealed 11 unbiased cell clusters (Figure 2A). The identity of the cells within these clusters were scrutinized by examining 5289 genes differentially expressed between all the clusters (Figure 2D and Table S2), as well as expression of genes known to distinguish undifferentiated spermatogonia (BCL6B, DUSP6, EGR2, ETV5, FGFR3, GFRA1, ID4, ITGA6, PIWIL4, RET, UTF1, and ZBZTB16) from early differentiating spermatogonia (DMRT1, KIT NANOS3, SOHLH1, SOHLH2, STRA8, and UPP1) and late differentiating spermatogonia that have activated the meiotic gene program (HORMAD1, MEIOB, SYCP2, and SYPC3) (Figure 2D). The cells in clusters 0, 1, and 6 expressed elevated levels of genes consistent with SSCs, including FGFR3, ID4, PIWIL4, TSPAN33, and UTF1 (Figure 2D,G and Table S2). Along with these prototypical SSC genes, enhanced expression of novel genes was also observed in cluster 0 (DUSP5, EGR4, FBXW5, TCF3, and TSPAN4), cluster 1 (GNAS and MMP2), and cluster 6 (BNIP3, HES4, and HRAS) (Figure 2G and Table S2). Furthermore, although the cells in cluster 1 exhibited lower levels of FGFR3 and UTF1 mRNA, they also had low levels of DUSP6 and ZBZTB16, possibly indicating that these cells are transitioning towards a progenitor state. In addition to expressing markers of undifferentiated spermatogonia (DUSP6, ETV5, ITGA6, and ZBZTB16), the cells in clusters 3, 4, 7, 9, and 10 were also enriched for progenitor markers (NANOS3 and UPP1), thus representing progenitor spermatogonia (Figure 2D,G and Table S2). In contrast, clusters 2 and 8 were designated as early differentiating spermatogonia based on the elevated expression of DMRT1, KIT, MKI67, and SOHLHL2 (Figure 2D,G and Table S2). Lastly, the cells in cluster 5 were considered late differentiating spermatogonia because their distinct transcriptome featured elevated levels of meiotic genes known to be activated during spermatogonial differentiation, including MEIOB, SYCP2, SYCP3, and HORMAD1 (Figure 2D,G and Table S2) [33]. Among baboon testis cells, cluster 11, comprising a total of 233 cells, expressed the germ cell marker DDX4 (Figure 1B,E). When reanalyzed in isolation, the cells in this cluster further resolved into three cell clusters (clusters 0–2), which were distinguished based on 1256 genes differentially expressed between all the clusters (Figure 2B,E and Table S2). However, unlike human germ cell clusters, which appeared as distinct groups on the UMAP projection, the baboon germ cells largely appeared to cluster together. The cell in Cluster 0 expressed lower levels of both undifferentiated (ETV5, PIWIL4, and TSPAN33) and differentiated (DMRT1 and SOHLH1) spermatogonial markers (Figure 2E,H). Both clusters 1 and 2 displayed similar patterns of elevated gene expression for the undifferentiated spermatogonial markers FGFR3, ID4, and UTF1 (Figure 2H). Cluster 1 uniquely expressed the undifferentiated spermatogonial markers DUSP6, ETV5, GFRA1, RET, TSPAN33, and ZBZTB16, whereas cluster 2 uniquely expressed EGR2 and PIWIL4 (Figure 2H). In addition, there was also expression of differentiating spermatogonial markers in cluster 1 (DMRT1, KIT, and NANOS3) and cluster 2 (SOHLH1) (Figure 2H). Moreover, meiotic markers SYCP2 and SYCP3 were expressed at a higher level in cluster 1 compared to cluster 2 (Figure 2H), suggesting some of these cells are more advanced. Since the baboon germ cell clusters expressed varying levels of the markers indicative of both undifferentiated and differentiating spermatogonia, it was not possible to annotate the cell types like we did in humans (Figure 2E,H). Using the same approach as we did for human and baboon germ cells, we identified a total of 599 rhesus germ cells based on the expression of DDX4 and performed iterative re-clustering of these cells, yielding six spermatogonial clusters (clusters 0–5) distinguished based on 1382 differentially expressed genes between all clusters (Figure 2C,F and Table S2). Similar to baboons, though, rhesus germ cells clustered together and did not exhibit distinct spermatogonial groups based on gene expression (Figure 2F,I). Specifically, cluster 0 mostly lacked expression of undifferentiated and differentiating marker genes, with few cells expressing NANOS3 (Figure 2I). Clusters 1–5 were characterized by varying levels of ID4 and SOHLH1 expression. Interestingly, Clusters 1, 2, and 5 largely lacked expression of any other undifferentiated and differentiating marker genes. In contrast, cluster 4 had elevated expression of markers of both undifferentiated (ZBZTB16, DUSP6, EGR2, ETV5, and ID4) and differentiating spermatogonia (DMRT1, KIT, NANOS3, and SOHLH1). We also observed species-specific features, including the absence of FGFR3, PIWIL4, and TSPAN33 expression in rhesus spermatogonia. Furthermore, the expression of meiosis-related genes (HORMAD, MEIOB, SYCP2, and SYCP3) was essentially absent in rhesus spermatogonia, indicating the absence of late differentiating spermatogonia. Thus, rhesus spermatogonia were clustered together and exhibited gradual transitions in marker expression, making it difficult to distinguish spermatogonial subtypes. (Figure 2F,I). To align the undifferentiated spermatogonial clusters/subsets across these three primate species, we first performed a joint analysis of the germ cells from each species by comparing primate germ cell subtypes with transplant-validated mouse SSCs (Figure S3A–C). For this purpose, we used 1-1-1-1 orthologous genes (n = 9068) to merge the gene expression data for 1451 human, 233 baboon, 599 rhesus, and 10,012 mouse germ cells [33]. To identify analogous spermatogonial subsets across the species, we defined the whole-transcriptome correlation from each of the three primate species to those from mice (Figure S3A). Surprisingly, cells in human cluster 5, which expressed transcriptomes consistent with late differentiating spermatogonia (see Figure 3), were most highly correlated with mouse SSCs (Pearson coefficient = 0.34) compared with the remaining human germ cell clusters (Pearson correlation coefficients ranging 0.01–0.33; Figure S3A). Among all human germ cells clusters, cluster 5 was also the most highly correlated with mouse progenitors (Pearson correlation coefficient = 0.40; Figure 3A) and mouse differentiating spermatogonia (Pearson correlation coefficient = 0.60; Figure S3A). Identical comparisons of baboon germ cell clusters with mouse spermatogonial subtypes revealed that baboon cluster 0 was most similar to mouse SSCs (Pearson correlation coefficient = 0.34; Figure S3C), while cluster 2 was the most highly correlated with both mouse progenitors (Pearson correlation coefficient = 0.33; Figure S3C) and mouse differentiating spermatogonia (Pearson correlation coefficient = 0.38; Figure S3C). Likewise, rhesus cluster 5 was the most highly correlated with mouse SSC (Pearson correlation coefficient = 0.35; Figure S3D), mouse progenitors (Pearson correlation coefficient = 0.36; Figure S3D) and mouse differentiating spermatogonia (Pearson correlation coefficient = 0.40; Figure S3D). Thus, although functionally informed, a comparison of the mouse spermatogonial subtypes to the unbiased groups of primate undifferentiated spermatogonia indicates that the degree of transcriptome similarity between primate and rodent germ cells is relatively low and may not predict cell type identity in primates, likely due to evolutionary distance. Therefore, we assigned human cell identity based on the relative abundance of mRNA for proposed and established markers of undifferentiated spermatogonia (FGFR3, ID4, PIWIL4, TPSAN33, and UTF1) versus progenitor, differentiating the spermatogonia and meiotic markers (Figure 2H). In addition to the use of these markers, global differential gene expression analysis between the unbiased human germ cell clusters facilitated assignment of cell type identities, SSC (Human germ cell clusters C0, C1, and C6), progenitor (Human C3, C4, C7, C9, and C10), differentiating (Human C2 and C8), and late differentiating (Human C5) cells, which are represented as a UMAP projection (Figure 2D, Figure 3A and Table S1). A comparison of the human germ cells bearing these cell type assignments with mouse spermatogonia revealed that the mouse SSCs were still most highly correlated with human differentiating spermatogonia (Pearson correlation coefficient = 0.38; Figure S3B). In contrast, human progenitors were most correlated with mouse progenitors and human late differentiating spermatogonia were most correlated with mouse differentiating spermatogonia, indicating that more advanced germ cells share a more similar transcriptome across species (Figure S3B). To identify the analogous spermatogonial subsets in baboons and rhesus macaques, we compared those cells identified as prepubertal human SSCs with clusters of baboon and rhesus macaque germ cells. We found that the cells in baboon cluster 2 were the most highly correlated with human SSCs (Pearson correlation coefficient = 0.93; Figure 3C), while the cells in rhesus cluster 1 were the most correlated with human SSCs (Pearson correlation coefficient = 0.86; Figure 3D). Human progenitors were most highly correlated with baboon cluster 2 (Pearson correlation coefficient = 0.52; Figure 3C) and rhesus macaque cluster 4 (Pearson correlation coefficient = 0.45; Figure 3D). Human differentiating spermatogonia were most highly correlated with baboon cluster 2 (Pearson correlation coefficient = 0.35; Figure 3C) and rhesus macaque cluster 4 (Pearson correlation coefficient = 0.45; Figure 3D). In addition, human late differentiating spermatogonia were most highly correlated with baboon cluster 0 (Pearson correlation coefficient = 0.51; Figure 3C) and rhesus macaque cluster 2 (Pearson correlation coefficient = 0.58; Figure 3D). A subsequent comparison of these putative primate SSCs with transplant-validated mouse SSCs demonstrated a low correlation with human (correlation coefficient = 0.25), baboon (correlation coefficient = 0.20), and rhesus (correlation coefficient = 0.24) SSCs (Figure 3E). Reciprocally, primate SSCs were much more highly correlated to each other, which likely reflects the evolutionary distance between primates and rodents (pairwise correlation coefficients 0.86 and 0.93 Figure 3E). Having identified the SSCs in each of the primate species, we next sought to identify the conserved components of the primate SSC gene expression signature (i.e., markers of primate SSCs). For this purpose, differential expression analysis was performed between human SSCs, progenitors, and differentiating spermatogonia, revealing 508 genes with higher expression in human SSC compared to other human spermatogonial subtypes. Similarly, putative baboon SSCs (Baboon C2) uniquely expressed higher levels of 347 genes and Rhesus SSCs (Rhesus C1) expressed higher levels of 54 genes (Table S2). Thirty-two of these genes were markers of SSCs in all three species (Figure 3F and Table S2). Among these genes, only PMAIP1 and RPL22L1 were also conserved in mouse SSCs. Next, we determined the expression of these 32 genes in publicly available datasets using the Mammalian Reproductive Genetics Database (MRGDv2; https://orit.research.bcm.edu/MRGDv2, accessed on 23 December 2022) (Figure S4) [51]. Congruent with our data, a majority of these 32 primate SSC markers were expressed in SSEA4+ human spermatogonia (undifferentiated), but absent from KIT+ differentiating spermatogonia (Figure S4). However, the majority of the primate SSC markers were expressed by both mouse SSCs (ID4 + high) and progenitors (ID4 + low) (Figure S4). In contrast to primates, mouse spermatogonia lacked expression of a number of these markers (Sh2b2, Rpl32) (Figure S4). Gene ontology analyses indicated that the 32 conserved primate SSC markers were enriched for genes involved in “Cytoplasmic Ribosomal Proteins” (RPL32, RPL18, RPS12; Figure 4A and Table S2). This same pathway was overrepresented in conserved markers between both human and baboon SSCs (RPL32, RPL11, RPSA, RPL18, RPS12; Figure 4B and Table S2), as well as human and rhesus SSCs (RPS4X, RPS9, RPL32, RPL18, RPS12; Figure 4C). Like primate SSCs, “Cytoplasmic Ribosomal Proteins” (Rpl4, Rps4x, Rps28, Rps27, Rps29, Rpl37, Rpl39, Uba52; Figure 4D and Table S2) were overrepresented among mouse SSC markers, pointing to translational control as a common feature of mammalian SSCs. However, although the same pathway was identified in the GO analysis of primate SSCs, the specific marker genes were different in primates and rodents, indicating possible differences in the precise mechanisms that regulate translation. Three genes (MAP2K2, RAC3, and PAK4) that were conserved in the SSCs of all three primates were also involved in numerous pathways, including “Pancreatic adenocarcinoma pathway”, “Integrin-mediated Cell Adhesion”, “Regulation of Actin Cytoskeleton”, and “Ras signaling” (Figure 4A–C). These pathways and genes were absent from the mouse SSC markers, indicating a species-specific role in SSC biology among primates. The genes involved in IL-1 signaling pathway (MAP2K2 and TOLLIP; Figure 4A and Table S2) were also enriched in the SSCs from all three primate species. Curiously, the genes annotated as involved in KIT receptor tyrosine signaling (MAP2K2 and SH2B2; Figure 4A and Table S2), ordinarily associated with spermatogonial differentiation, were also observed in primate SSCs despite the lack of KIT expression (Figure 2H,I and Figure 3B), which may suggest unique primate-specific signaling cascades that converge at these factors. To use our data, we sought to relate our primate SSC signatures with the histological definitions of human Adark and Apale spermatogonia based on nuclear staining intensity with hematoxylin. We examined our data for markers that are known to discriminate Adark and Apale by immunostaining. Previous studies demonstrated that FGFR3 and EXOSC10 are restricted to rarefaction zone-containing Adark spermatogonia, while MKI67 and DMRT1 were absent from these cells [50,52]. We found that FGFR3 was expressed by SSCs and progenitors, EXOSC10 by progenitors and differentiating spermatogonia, and both DMRT1 and MKI67 were restricted to differentiating spermatogonia (Figure 5) [53]. These data indicate that the SSCs and progenitors are all Adark, while cells that have initiated differentiation are Apale (Figure 5). The field of spermatogonial biology has struggled to propagate primate SSCs in culture, which would enable their routine use in experimentation, as well as exploiting their therapeutic potential to restore spermatogenesis. Although attempts at in vitro human SSC culture have been made, a robust culture system for human SSCs is yet to be developed [12,13,18,54]. Indeed, supporting evidence proving spermatogonial identity (let alone SSCs) is weak or circumstantial and none of the reported conditions have been robustly repeated by external groups. Thus, the field still lacks the ability to maintain primate (including human) SSCs in culture. By refining the molecular characteristics of SSCs in humans, baboons, and rhesus macaques, we have identified potential culture considerations that may finally permit the expansion of human SSCs. The major objective of our study was to refine the identity and phenotype of SSCs in primate species. To this end, we employed a single-cell RNA-seq to define the germ and somatic cell types in prepubertal human, baboon, and rhesus testes, and focused our analyses on determining the degree of gene expression conservation among putative primate SSCs relative to functionally defined cells in mice. Specifically, we found 32 genes conserved and uniquely enriched in primate SSCs. Three genes in particular (MAP2K2, PAK4, and RAC3) exhibited a conserved expression pattern in the SSCs from all three primates and are also involved in integrin-mediated cell adhesion, the regulation of the actin cytoskeleton, and cell migration. Among these genes, MAP2K2, or mitogen-activated protein kinase kinase 2, is already known to be involved in cell fate determination, cell growth, and differentiation [55]. PAK4 or p21 (RAC1) activated kinase 4 is required for SSC homing and transmigration through the blood-testis barrier in mice [56]. RAC3 is a small GTPase that interacts with the integrin binding protein CIB1 (calcium and integrin binding 1) to promote cell adhesion [57]. In addition, the genes implicated in the positive regulation of cell migration (CIB1 and PRR5) were enriched in the primate SSCs. Overall, these results suggest that genes involved in cell adhesion, regulation of the actin cytoskeleton, and cell migration may play an outsized role in SSC function in primates. In contrast to primates, genes enriched uniquely in mouse SSCs were involved in the GDNF/RET signaling axis (Gfra1, Ret, Foxc2, and Lhx1), which is known to be important for mouse SSC self-renewal. In mice, the addition of GDNF and FGF2 to culture medium is essential for long-term propagation of SSCs [58]. Absence of the GDNF/RET signaling axis in primate SSCs may explain why attempts to culture and expand human SSCs using this factor have not been fruitful. Instead, unique features of primate SSCs that are not prominent in rodents may need to be investigated in order to propagate human SSCs in vitro. The genes implicated in ribosome biosynthesis and the regulation of translation were enriched in the SSCs from all four species of mammals. Specifically, genes encoding structural components of ribosomes and translational regulation were observed. These data raise the interesting concept that post-transcriptional gene regulation is involved in establishing the stem cell state in mammals. Decades of work have set the clear precedent for the concept of post-transcriptional regulation of germ cell development in invertebrates, lower vertebrates (C. elegans, Drosophila, Xenopus) [59,60,61], and in mammals [62,63]. However, the precise role for translational regulation in maintenance of the SSC state has never been firmly determined. Future studies may leverage this knowledge to promote enhanced translation of specific messages and prevent translation of others in hopes of driving enhanced SSC renewal. Importantly, the correlation of human spermatogonial single-cell transcriptomes (SSC, progenitors, differentiating, and late differentiating) with markers of Adark (expressing EXOSC10 and FGFR3) and Apale (expressing MKI67 and DMRT1) spermatogonia provided a prediction of the likely histological phenotype of molecularly-defined SSCs. These data indicated that human SSCs are a subset of Adark and lack MKI67 expression, consistent with the notion of a slow-cycling stem cell population [35,37,38]. Progenitor spermatogonia were also found in the Adark population, arguing for a single pool of SSCs that is comprised entirely of Adark spermatogonia. Intriguingly, the Apale spermatogonial population appear to consist of cells that are proliferative and initiate differentiating, which is incongruent with the designation of Apale as active SSCs, but supportive of their role as a transit-amplifying progenitor poised to differentiate. Since the generation of single-cell suspensions for scRNA-seq necessitates the dissociation of the three-dimensional testis tissue architecture, we are unable to directly associate the SSC transcriptome phenotypes with their native context within seminiferous tubules. Previous studies have shown that the self-renewal and differentiation of SSCs is influenced by their microenvironment [64,65]. Therefore, spatial transcriptomics could potentially overcome this key limitation of traditional single-cell profiling and provide crucial insights into the regulatory mechanisms of spermatogenesis. Using Slide-seq2, Chen et al. performed spatial analysis of adult human and mouse testis using ~29 K beads to capture transcripts. This analysis revealed differences in the spermatogonial compartments and the cellular compositions of the spermatogonial microenvironment between humans and mice [66]. In addition, the study described spatially resolved expression patterns. For instance, Smcp expression was enriched in round spermatids near the tubule lumen, while Lyar expression was elevated in the spermatocytes near the basement membrane. Future studies may overcome the relatively low resolution of this method, which limits investigation of cell–cell interactions, such as ligand–receptor relationships [66]. Precise analysis of expression patterns of the ligand receptor in SSCs and their supporting somatic cells has been possible using scRNA-Seq and led to predictions of the interactions between distinct cell types. Several studies have used this approach to suggest that FGF, GDNF, KIT, retinoic acid, and activin/inhibin signaling as the most relevant to human spermatogenesis [42,43,45,49]. Our results fail to corroborate many of these predictions, and instead suggest that prototypical examples, like GDNF (GFRA1 and RET) and FGF (FGFR3), are false positives, likely the result of flawed cell type designation. Our data do support the potential for activin/inhibin signaling insofar as we identified expression of ACVR1/2 receptors in primate SSCs, suggesting this pathway may be important for the SSC state. Future studies may help corroborate ligand–receptor analysis using protein data. In conclusion, scRNA-seq identified SSCs in humans, macaques, and baboons highlighted a surprising phenotypic discordance between primate SSCs and functionally defined mouse SSCs. These results may explain the difficult adapting conditions for the propagation of mouse SSCs to primates. Importantly, primate SSCs were enriched for transcripts encoding components and regulators of the actin cytoskeleton and regulation of cell-adhesion, which may be the missing pieces necessary to devise a robust protocol for in vitro propagation of human SSCs. Addressing a nearly half-century debate, our data confirm that SSCs are a subset of Adark spermatogonia, putting to rest the concept of a dual stem cell system in primate testes. Taken together, these results refine the identity of prepubertal human SSCs and elucidate molecular features that may be leveraged for their use in the clinic. The testicular tissues were recovered at necropsy at the Southwest National Primate Research Center (SNPRC, Texas Biomedical Research Institute) as biomaterials collections from two immature baboons (postnatal day 0 and 26 months) and two rhesus macaques (15 months and 20 months) after they had been euthanized for another purpose. Therefore, the research conducted with these tissues was not considered to be animal research. The donor cells were recovered from olive baboon and rhesus macaque testes using a two-step enzymatic digestion approach as previously described [67]. Briefly, testicular parenchyma was digested with 1 mg/mL Collagenase Type IV (Worthington Biochemicals) for 2–3 min at 37 °C, washed with Hank’s Buffered Salt Solution (HBSS) to remove the interstitial cells, digested with 0.25% trypsin/EDTA containing 1.4 mg/mL DNase I (Sigma) for 7–9 min at 37 °C, and quenched with 10% FBS. The testis cell suspensions were filtered through a nylon mesh to generate a single-cell suspension. The cell yield was determined by counting the cells using a hemocytometer and viability was assessed by trypan blue exclusion. Single-cell transcriptomes were generated using the 10× Genomics Chromium Cell Gene Expression kit (v2 chemistry) per manufacturer recommendations [68] at the UTSA Genomics Core as previously described [33]. For each replicate, we targeted a collection of 5000 single-cell gel bead emulsions (GEMs) containing single cells and generated libraries as per manufacturer recommendations [68]. The libraries were sequenced at the Genome Sequencing Facility (GSF) at Greehey Children’s Cancer Research Institute at UT Health San Antonio (UTHSA) on a HiSeq2500 (Illumina) instrument. The 10× Genomics Single-cell 3’ libraries were prepared according to manufacturer recommendations (Single Cell 3’ v3 chemistry for 37979, Single Cell 3’ v2 chemistry for all others). Illumina base-calling and demultiplexing was performed with BCL-2-Fastq. The trimmed FASTQ files were generated using the Cell Ranger v3.1.0 (10× Genomics) mkfastq command using the default parameters. The alignment of the trimmed reads, filtering, and UMI counting were performed using the CellRanger v3.1.0 count function using mouse GRCm38 (mm10), human GRCh38 (hg38), Papio anubis (olive baboon) Panubis1.0 (Ensemble GCA_008728515.1), and Macaca mulatta Mmul_10 (Ensemble GCA_003339765.3) references. For each replicate, a single gene-barcode matrix file (MTX) was produced. Each sparse matrix is stored in Market Exchange Format (MEX), which contains TSV files with features (genes) and barcode sequences corresponding to the row and column indices, respectively. Our baboon and rhesus macaque scRNA-seq datasets are publicly available (GSE222105). We compared the nonhuman primate datasets generated from these libraries we produced to the published datasets representing testis cells from prepubertal humans at PD2 and PD7 (GSE124263) [42], 12 mo, 13 mo (GSE120508) [44], and 7 yo, 11 yo GSE134144 [43], along with previously published single-cell datasets from postnatal day 6 (P6) mouse testes [33] (see Table S1). The gene expression matrices for each dataset were imported into Seurat (v3.2.3) [69] using the Read10X function and merged by species using the merge function. The merged datasets were subsequently filtered for cells expressing ≥ 200 detected genes and genes expressed in 3 ≥ cells. SCTransform normalization was performed on each of the merged datasets using the command sctransform, which performs normalization, variance stabilization, and selection of variable genes in one step. We subsequently filtered low-quality cells exhibiting >20% reads mapping to mitochondrial genes using the vars.to.regress argument. Integration was performed by calculating the Pearson residuals and identifying anchors. Cell clustering and visualization was performed using the FindNeighbors, FindClusters, and RunUMAP functions, using a resolution of 0.5 based on the principal components selected using the JackStraw function. The FindMarkers function was used to run differential expression analysis for clusters or using cell type annotations. For the iterate clustering of germ cells, the subset function was used to extract the germ cell clusters. Orthologues were retrieved from Ensembl Biomart (Ensembl Gene Version 94, Sept 2021). The pairwise orthologues for human vs. baboon (n = 18,131), human vs. rhesus (n = 21,616), and human vs. mouse (n = 20,390) were retrieved separately. These pairwise orthologues were used to subset each of the baboon, rhesus, and mice datasets separately in Seurat using the human Ensembl gene ID as reference. Subsequently, a merged dataset of all 4 species was produced using the merge function in Seurat, with expression data for one-to-one-to-one-to-one orthologues only for further analysis. Subsequently, the correlation coefficients were calculated between the clusters or cell types using the AverageExpression function in Seurat and the R package tidyr (v1.1.4) [70]. The R package gpplot2 (v3.3.5) was used to visualize the correlation heatmaps [71]. Gene ontology enrichment analysis was performed using EnricherR using the pathway database Wikipathways [72,73].

PMC10002553

Jarrod Moore,Jourdan Ewoldt,Gabriela Venturini,Alexandre C. Pereira,Kallyandra Padilha,Matthew Lawton,Weiwei Lin,Raghuveera Goel,Ivan Luptak,Valentina Perissi,Christine E. Seidman,Jonathan Seidman,Michael T. Chin,Christopher Chen,Andrew Emili

Multi-Omics Profiling of Hypertrophic Cardiomyopathy Reveals Altered Mechanisms in Mitochondrial Dynamics and Excitation–Contraction Coupling

01-03-2023

hypertrophic cardiomyopathy,human induced pluripotent stem-cell-derived cardiomyocytes,myectomy,liquid chromatography–tandem mass spectrometry,phosphoproteomics,metabolomics,network systems biology

Hypertrophic cardiomyopathy is one of the most common inherited cardiomyopathies and a leading cause of sudden cardiac death in young adults. Despite profound insights into the genetics, there is imperfect correlation between mutation and clinical prognosis, suggesting complex molecular cascades driving pathogenesis. To investigate this, we performed an integrated quantitative multi-omics (proteomic, phosphoproteomic, and metabolomic) analysis to illuminate the early and direct consequences of mutations in myosin heavy chain in engineered human induced pluripotent stem-cell-derived cardiomyocytes relative to late-stage disease using patient myectomies. We captured hundreds of differential features, which map to distinct molecular mechanisms modulating mitochondrial homeostasis at the earliest stages of pathobiology, as well as stage-specific metabolic and excitation-coupling maladaptation. Collectively, this study fills in gaps from previous studies by expanding knowledge of the initial responses to mutations that protect cells against the early stress prior to contractile dysfunction and overt disease.

Multi-Omics Profiling of Hypertrophic Cardiomyopathy Reveals Altered Mechanisms in Mitochondrial Dynamics and Excitation–Contraction Coupling Hypertrophic cardiomyopathy is one of the most common inherited cardiomyopathies and a leading cause of sudden cardiac death in young adults. Despite profound insights into the genetics, there is imperfect correlation between mutation and clinical prognosis, suggesting complex molecular cascades driving pathogenesis. To investigate this, we performed an integrated quantitative multi-omics (proteomic, phosphoproteomic, and metabolomic) analysis to illuminate the early and direct consequences of mutations in myosin heavy chain in engineered human induced pluripotent stem-cell-derived cardiomyocytes relative to late-stage disease using patient myectomies. We captured hundreds of differential features, which map to distinct molecular mechanisms modulating mitochondrial homeostasis at the earliest stages of pathobiology, as well as stage-specific metabolic and excitation-coupling maladaptation. Collectively, this study fills in gaps from previous studies by expanding knowledge of the initial responses to mutations that protect cells against the early stress prior to contractile dysfunction and overt disease. Hypertrophic cardiomyopathy (HCM) is among the most common inherited cardiomyopathies and, historically, a leading cause of sudden cardiac death in young adults [1,2]. Roughly 60% of patients have a defined genetic disease, with the majority of mutations mapping to genes encoding thick and thin myofilament proteins [3,4]. Of these defined loci, mutations in myosin heavy chain 7 (MYH7) and myosin binding protein C3 (MYBC3) are most common and are clinically characterized by asymmetric left ventricular thickening, diastolic dysfunction, and fibrosis [3]. For instance, the Arg403Gln (R403Q) mutation in the MYH7 gene can result in a severe HCM phenotype with progressive myocardial dysfunction and increased incidence of sudden cardiac death [5,6]. Despite profound insights into the genetics of HCM, however, there is low correlation between mutation and clinical prognosis [7]. Moreover, there remains a need to define the earliest cell intrinsic responses to mutation prior to frank pathology. Studies on isolated cardiac tissue have highlighted changes in contraction mechanics as potential drivers of the observed phenotype. HCM cardiomyocytes have increased tension costs (i.e., force generation per ATP utilized) due to changes at the biophysical level [8,9]. For example, both MYBC3 and MYH7 mutants exhibit a decreased number of super-relaxed-state myosin heads, a configuration with low ATPase cross-bridge utilization, and MYH7 mutants also show faster cross-bridge kinetics that directly increase tension costs [8,10,11]. Moreover, HCM exhibits marked Ca2+ handling dysfunction (e.g., increased Ca2+ sensitivity and intracellular diastolic Ca2+), further exacerbating tension costs and ATPase activity [4,12]. Although the exact mechanism remains unclear, this high energy demand is hypothesized to increase mitochondrial workload and oxidative stress, ultimately resulting in maladaptive cardiac remodeling [13]. Insight into cardiomyocyte remodeling has come from large-scale molecular (i.e., omics-type) profiling studies, whose findings imply that increased energy demand damages mitochondria via augmented oxidative stress [9]. Mutant cardiac cells exhibit increased mitochondrial ultrastructure damage and evidence of oxidative damage (presumably resulting from increased reactive oxygen species generation) [14,15]. Excessive oxidative stress can impair mitochondrial components, such as the electron transport apparatus and mitochondrial DNA, which can exacerbate ATP generation deficits [9,16]. Notably, patient myectomies exhibit marked metabolic deficits associated with mitochondrial dysfunction, such as significantly decreased levels of fatty acid metabolic enzymes and lower ratio of phosphocreatine-to-ATP levels, which are presumably secondary to mitochondrial dysfunction [9,17,18]. Precision mass spectrometry is a powerful tool for elucidating biomolecular networks and associated signaling cascades driving pathobiology [19]. Recently, we applied a quantitative phosphoproteomic profiling workflow to study the impact of cardiac fibrosis in a three-dimensional biomimetic in vitro co-culture model system (heart-on-a-chip) and in myectomy specimens from HCM patients, which revealed pathway-level alterations associated with altered energetics and calcium handling [20]. However, this model lacked associated HCM mutations and major outstanding questions remained regarding molecular cascades driving mitochondrial dysfunction and associated excitation–contraction coupling perturbations. To address these gaps, we have now performed a comparative multi-omics (proteomic, phosphoproteomic, and metabolomic) analysis to illuminate the early and direct consequences of mutations in MYH7 in engineered human induced pluripotent stem-cell-derived cardiomyocytes (hiPSC-CM), in comparison to isogenic control cells, versus human HCM-patient-derived cardiac myectomy tissue, representing the spectrum of very early to clinically advanced stages of pathology. While some of the changes observed were co-ordinate, major differences were also detected. For instance, whereas we noted parallel increases in mitochondrial dynamics, favoring fission, in the early model and myectomy specimens, the myectomy specimens showed a complete abrogation of the major protective mechanisms against oxidative stress. Integrated metabolic pathway analysis also highlighted increases in glutamine anaplerosis in the mutant hiPSC-CMs, which apparently replenished Krebs cycle intermediates and glutathione. Conversely, advanced-stage disease showed greatly diminished use of this pathway. We also characterized pronounced shifts in the post-translational (phosphorylation) states of the sarcoplasmic reticulum (SR) ATPase and functionally related excitation–contraction coupling proteins between the early and late disease stages, which correlated with alterations to Ca2+ handling, likely secondary to the metabolic defects. In summary, our comprehensive assessment highlights metabolic, mitochondrial, and Ca2+ handling dysfunction across different stages of HCM pathology. To investigate metabolic status and mitochondria stress information in a model of the earliest pathobiological changes preceding overt disease, we first performed liquid chromatography–tandem mass spectrometry (LC/MS)-based global proteomics on 14 paired, isogenic hiPSC-CM samples (7 WT and 7 MYH7R403Q+/− replicate cultures) (Figure 1A). The hiPSC-CM cultures were differentiated from CRISPR/Cas9-edited MYH7R403Q+/− hiPSCs via small-molecule, monolayer manipulation of the Wnt signaling pathway (Figure 1B) [21,22]. MYH7R403Q+/− mutations led to hypertrophy, heightened metabolic activity, and increased force output at the single-cell level [23]. To ensure quantitative accuracy, we subjected each proteolytically digested sample to stable isotope labeling using isobaric tandem mass tag reagents and augmented the depth of proteome coverage by applying extensive peptide-level offline prefractionation prior to precision Orbitrap mass analysis (see Section 4, Figure 1C) [24]. From our initial proteomic analysis, we identified and quantified 7829 proteins between the mutant and control (WT) hiPSC-CM samples (Figure 2A). Differential proteome analysis revealed 648 statistically significant proteins (moderated Student t-test, false discovery rate (FDR) adjusted p < 0.05) (Figure 2B, Supplementary Table S1A, Supplementary Figure S1) [25]. From this set, we captured differential components linked to myofibrillary function and stability, including elevated levels of troponin I (TNNI3), filamin-C (FLNC), and four and a half LIM domains protein 1 (FHL-1) [26,27]. Moreover, we observed significantly altered expression of factors involved in Ca2+ handling, including downregulation of L-type calcium channel subunits and upregulation of calsequestrin (CASQ2). Since phospho-signaling responses provide additional functional information complementary to differential protein expression, we next performed large-scale phosphopeptide affinity capture and a secondary LC/MS analysis, which allowed us to map 4362 phosphorylation sites on 1357 distinct phosphoproteins in our hiPSC-CM model system (Supplementary Table S1B). We observed 129 statistically significant (FDR < 0.05) differential phosphosites (Figure 2A), which included altered phosphorylation patterns on key sarcomeric proteins, such as titin (TTNSer1423), myosin-binding protein C (MYBPC3Ser427), filamin-C (FLNCSer2233), and leiomodin-2 (LMOD2Ser392) (Figure 2B). Collectively, these twin data sets offer a rich resource for analyzing biochemical processes altered in the mutant (early-stage) samples. To systematically assess the pathway-level changes in the mutant cells, we performed gene set enrichment analysis (GSEA) with the proteomics data to parse annotated molecular pathways altered as a result of the MYH7R403Q+/− mutation (Supplementary Table S2) [25,28]. We noted significant enrichment (FDR < 0.05) of the mitochondrion organization and mitophagy gene sets in the mutated cells (Figure 3A), which pointed to potential modulation of morphology and function. Within the mitochondrion organization pathway, increased expression of key mitochondrial fusion and fission effectors was detected. Most notably, fusion proteins upregulated in MYH7R403Q+/− mutant cells included the primary GTPases driving fusion of the outer mitochondrial membranes, mitofusion-1 and -2 (MFN1 and −2) [29]. We also measured increased levels of Dynamin-1-like protein (DNM1L), the main GTPase conferring mitochondrial constriction during fission, as well as significant upregulation of clustered mitochondria protein homolog (CLUH) and mitochondrial Rho GTPase 2 (RHOT2), important regulators of mitochondrion trafficking that also mediate mitochondrial dynamics [29,30,31,32,33]. In particular, CLUH directly binds the mRNA transcripts encoding DNM1L receptor proteins (i.e., mitochondrial fission factor (MFF) and mitochondrial dynamics protein MID51 (MIEF1)) to facilitate fission [34,35]. Consistent with the upregulation of fission mediators, our phosphoproteomics data also showed an increasing trend (just short of statistical significance) in the phosphorylation levels of these DNM1L recruitment factors, MFFSer155 and MIEF1Ser55,59 (Supplementary Table S1B). Mitochondrial fission primes cells for mitophagy, a selective form of autophagy during oxidative stress, which fragments the mitochondrial network and degrades damaged mitochondria via autophagosome engulfment [36,37]. Consistent with this, we observed overexpression of a number of key modulators of mitophagy and autophagosome pathways in the mutant cells (Figure 3A), suggesting increased engagement of mitochondrial quality control mechanisms. Notably, microtubule associated protein 1 light chain 3 (MAP1LC3B) and its ubiquitin binding adaptor protein sequestosome-1 (SQSTM1) were upregulated, as were the related ubiquitin-like proteins ATG5 and -12 [38,39,40]. MAP1LC3B is important for maturation of the autophagosome, and the MAP1LC3B-SQSTM1 interaction is crucial for targeting mitochondria to autophagosome and initiating degradation [41]. Taken together, these findings provide evidence of increased fission and mitophagy in mutant CMs, highlighting these as potential determinants of mitochondrial health in early disease. Oxidative stress and mitochondrial damage are well-known features of cardiac disease [15,42]. In HCM, mitochondria are especially vulnerable to oxidative damage due to augmented reactive oxygen species (ROS) production [9,18,43,44]. An important consequence of oxidative stress is a change in mitochondrial morphology and function, including key macromolecule modifications [9,15]. Accordingly, we found the ROS sensing response as significantly enriched in the mutant hiPSC-CM samples (Figure 3A). This pathway includes components of the proteasome complex and assembly, which function in clearance of accumulated oxidized proteins, such as proteasome activator complex subunit 4 (PSME4) [45,46]. Moreover, we noted differential expression of thiol-based antioxidants, which protect against excessive ROS (Figure 3B). These included upregulation of components of the glutaredoxin and thioredoxin systems, which assist in reversing protein glutathionylation, a consequence of increased oxidative stress (i.e., cysteine residue thiol oxidation) (Supplementary Table S1A) [47,48]. In particular, we noted significantly (log2FC = 0.791) elevated expression of glutaredoxin-2 (GLRX2). However, we also detected decreases in other major components of these pathways, such as thioredoxin reductase 2 (TXNRD2) and glutathione peroxidase 7 (GPX7). TXNRD2 and GLRX2 serve similar roles in maintaining redox homeostasis by reducing mitochondrial-specific redox proteins, suggesting moderate antioxidant protection in the mutant hiPSC-CM [47]. Mitochondrial DNA is a principal target of oxidative stress (i.e., oxidized bases and strand breaks) [49,50]. Consistent with this, we noted significant decreases in key components of mitochondrial base excision repair (BER) (Figure 2B), the main repair mechanism of oxidative damage in the mitochondria [51,52]. For example, the mutant hiPSC-CM samples significantly downregulated endonuclease-III-like protein 1 (NTHL1), a bifunctional glycosylase that excises oxidized DNA bases and generates abasic sites [53]. We also noted decreases in Poly (ADP-ribose) polymerase 1 and 2 (PARP1/2), which detect abasic sites and recruit the DNA repair protein XRCC1 (XRCC1), forming a scaffolding complex for other repair factors [54,55]. While PARP1/2 are predominately localized to the nucleus, PARP1 migrates to the mitochondria via interactions with Mitofilin (IMMT), where it plays a protective/repair role for mtDNA by interacting with DNA ligase 3 (LIG3) [56,57]. Notably, then, is the observation that LIG3 also downregulated in our MYH7R403Q+/− samples, which catalyzes the last ligation step of BER. Taken together, these data suggest that mtDNA is particularly vulnerable to oxidative damage, which could further facilitate mitochondrial dysfunction in mutant hiPSC-CMs. To examine the persistence of the defects we observed in the mutant hiPSC-CM cells, we next measured the proteome and phosphoproteome of human donor biospecimens using the same workflow. We sampled explants from 10 sex-matched patients, 5 donor WT and 5 presenting with MYH7 mutant alleles (Supplementary Table S3, Table 1). As with the hiPSC-CM model, we subjected the peptide samples to stable isotope labeling, followed by extensive prefractionation before in-depth quantitative LC/MS analysis. In total, we identified and quantified 6463 proteins along with 9697 phosphorylation sites on 2766 phosphoproteins in the human donor samples (Figure 2A). Comparative statistical analysis demonstrated hundreds of statistically significant (FDR < 0.05) differences in proteins and phosphosites between the HCM patient and case controls (Figure 2A). Notably, we measured significantly elevated phosphorylation-based regulation of key mitochondrial fission factors via RAS/MEK/MAPK1 pathway activation (Figure 3C). For example, the HCM donors exhibited increased phosphorylation of serine-616 on Dynamin-1-like protein (DNM1LSer616), which promotes activity and dimerization for constriction-based mitochondrial fission [58,59]. Further upstream, we noted evidence of activation of the mitogen-activated protein kinase (MAPK) cascade via hyperphosphorylation of the protein kinase domain on mitogen-activated protein kinase 1 (MAPK1Thr185), a conserved threonine/glutamate/tyrosine motif whose phosphorylation induces kinase activity [60]. MAPK1 increases mitochondrial dynamics by directly phosphorylating and activating DNM1LSer616 [61,62]. Consistent with this, we detected significant phosphorylation of the mitochondrial outer membrane docking receptors for DNM1L, including MIEFSer59, that are recruited during fission [34,63]. Hence, although not recapitulating the fusion and mitophagy alterations seen in the hiPSC-CM early model of pathology, these findings demonstrate a persistent upregulation of mitochondrial fission in advanced disease. In contrast to the hiPSC-CM model, we observed profound decreases in factors linked to the oxidative stress response, such as thiol-based peroxidases (Figure 3D). The HCM patient samples exhibited significant decreases in the cell redox homeostasis pathway from GSEA, which included decreased expression of peroxiredoxin-6 (PRDX6) and mitochondrial thioredoxin (TXN2), important enzymes for reducing cellular peroxide levels (Supplementary Table S4, Supplementary Figure S2) [64]. From this pathway-level analysis, we detected decreased levels in thioredoxin reductase-1 and -2 (TXNRD1/2), which function to reduce thioredoxin, suggesting reduced thioredoxin antioxidant capacity [47,65]. Moreover, our analysis revealed concomitant alterations in key components of the double-strand DNA repair pathway (Supplementary Table S4), an important response that normally counters chronic oxidative stress [66]. Overall, in only partial accordance with the hiPSC-CM results, the advanced HCM specimens showed decreased expression in thiol-based oxidative response elements that are integral for maintaining mitochondrial integrity. Metabolic remodeling is associated with increased mitochondrial dynamics in cardiomyocytes [67]. Thus, we applied an integrated metabolic pathway evaluation strategy on the hiPSC-CM samples to determine early metabolic alterations arising from the mutation. First, we performed metabolic enrichment network analysis (MOMENTA) to discover differentially expressed metabolic pathways predicted from our proteomic data, which we then validated by direct metabolomic analysis [68]. Among the many metabolic pathways altered in the early-stage mutant model (Figure 4A), we noted significant increases in enzyme levels mediating glutamate degradation, as well as 2-oxoglutarate (α-ketoglutarate) decarboxylation to succinyl-CoA, suggesting upregulation of glutaminolysis in the early HCM samples (Supplementary Table S5). To confirm these predictions further, we performed two independent phases (global and targeted) of small-molecule mass-spectrometry-based metabolomics analysis to directly confirm changes in metabolic profiles and flux. Consistent with our proteomics predictions, our global metabolomics survey found direct evidence for a significant increase in glutamate levels in the MYH7R403Q+/− samples (Figure 4A, Supplementary Table S6). Among the hundreds of changes in metabolite levels that were detected, glutamate was among the top differential features (log2FC = 2.20, FDR < 0.001). Given the evidence for glutaminolysis, we subsequently performed targeted metabolomic analyses to find further support of this pathway. For this analysis, we accurately quantified select glycolytic/citric acid cycle intermediates using 13C stable-isotope glucose tracer to determine whether unlabeled glutamate was supplementing the citric acid cycle (Figure 4B). As expected, the mutant cells showed elevated levels of labeled glucose-derived TCA metabolites, including 13C malate (Supplementary Table S7). In contrast, we observed a significant decrease in 13C succinate, the TCA produced following 2-oxoglutarate decarboxylation (Figure 4B). We additionally saw a significant increase in 13C alanine. Viewed from the vantage of current models of metabolism, these findings are consistent with the transamination of unlabeled glutamate (a downstream metabolite of glutaminolysis) with labeled pyruvate, resulting in increased labeled alanine and decreased succinate. To verify that the enhanced glutamine is directly entering the citric acid cycle, we treated the WT and mutant hiPSC-CM cultures with isotopically (13C5) labeled glutamine (Supplementary Table S8). As expected, we observed a significant increase in labeled glutamine preferentially in the mutant cells (Figure 4C). Concomitantly, we detected increases in the labeled form of related TCA intermediate α-ketoglutarate in the HCM samples, further supporting the role of increased flux of glutamine into the TCA in the mutant cells. To confirm the glutamine dependency of the MYH7R403Q+/− cells, we treated the hiPSC-CMs with the selective glutaminase inhibitor BPTES. Strikingly, we saw a rapid and significant decrease in contraction over the treatment period in the mutant cells (Figure 4B), leading to a complete loss of contraction in the MYH7R403Q+/− CMs by 90 min. Intriguingly, we also detected a trend towards elevated glutathione-based synthesis from glutamine from our targeted metabolomics analysis (log2FC = 1.21), though slightly beyond statistical threshold (p-value = 0.0521) (Figure 4C, Supplementary Table S8). Our proteome data further supported this finding, as we observed significant decreases in glutathione catabolic components (Figure 3B). These included the putative catabolic factor gamma-glutamyl transpeptidase 3 (GGT3P). Moreover, MOMENTA revealed enrichment in the mutant cells of the pentose phosphate pathway, which produces NADPH (Figure 4A). NADPH is a cofactor for glutathione reductase that enables reduction of glutathione to alleviate oxidative stress [69]. We likewise detected increased levels of the fructose-2,6-bisphosphatase enzyme TIGAR, which increases the activity of the pentose phosphate pathway [70,71]. Glutathione is the primary intracellular thiol-based redox buffer, further supporting the protective engagement of the thiol-based oxidative stress response machinery in the early-stage disease model [48]. Taken together, our comprehensive metabolic analysis of the hiPSC-CM model strongly supports glutamine anaplerosis as an important mechanism for replenishing TCA metabolites as an early adaptive response to altered contractile function. In striking contrast to the results from our cell culture model, our proteomics analysis of the myectomy samples indicated decreased metabolic supplementation through the TCA/glutamine shunt. Beyond significant reductions in the components of the glutamate degradation II pathway from MOMENTA (Supplementary Table S9), our GSEA of the clinical samples also revealed a significant (FDR < 0.05) decrease in mitochondrial fatty acid beta-oxidation and citric acid cycle/respiratory electron transport gene sets in the affected patient specimens (Figure 4A, Supplementary Table S4). Within the citric acid cycle/respiratory electron transport set, the HCM samples showed markedly decreased expression of electron transport complex proteins, such as NADH dehydrogenase (ubiquinone) 1 (NDUFV1). Moreover, we noted decreases in several enzymes crucial in fatty acid metabolism, such as ACAT1 (acetyl-CoA acetyltransferase) and ACAA2 (3-ketoacyl-CoA thiolase) from the fatty acid beta-oxidation gene set (Supplementary Table S4). Given the importance of fatty acid oxidation in cardiac cells, these findings suggest poor metabolic capacity in the affected tissue [72]. Lastly, again in contrast to our early model, we noted significantly decreased enrichment of the pentose phosphate pathway in the advanced disease with respect to myectomy controls (Figure 4A). This pathway is crucial for its production of NADPH, which is used by glutaredoxin and thioredoxin systems to reduce key antioxidant enzymes. Taken together with the decreases seen in mitochondrial-specific metabolic pathways, compounded with the decrease in glutamine anaplerosis, these findings demonstrate the potential for the persistence in impaired bioenergetic replenishment and antioxidant protection in advanced HCM. Given the profound differences in mitochondrial function and metabolism noted in both the mutant model and advanced HCM specimens, we anticipated persistent dysregulation of calcium handling and contractility mechanisms in both contexts. First, examining the hiPSC-CM model, we observed significantly altered phosphosites on key calcium handling factors localizing to the sarcoplasmic reticulum, mitochondria, sarcomere, and plasma membrane between the mutant and control samples in both the early model and advanced disease stages. For instance, in the hiPSC-CM mutant cells, we measured significantly increased phosphorylation events on a number of key SR proteins (Figure 5A, Supplementary Table S1B). These included hyperphosphorylation of cardiac phospholamban (PLNSer16,Thr17), ryanodine receptor 2 (RYR2Ser2814), sarcoplasmic/endoplasmic reticulum calcium ATPase 2 (ATP2A2Ser663), and junctophilin-2 (JPH2Ser241). The RYR2Ser2814 and PLNSer16,Thr17 events are linked to increased flux of calcium through the SR, with RYR2Ser2814 associated with channel opening and augmented outward flux, while PLNSer16,Thr17 has been shown to increase inward flux by releasing PLN inhibition of ATP2A2 (Supplementary Table S10A) [73,74]. We additionally measured persistent changes in the regulation of key sarcomeric proteins modulating contractility. These included increased phosphorylation of filamin-C (FLNCSer2233,2236) and leiomodin-2 (LMOD2Ser392), along with decreased levels of major phosphosites on titin (TTNSer1423,1418) and myosin-binding protein C (MYBPC3Ser424,427). While M-domain phosphorylation on MYBPC3 is crucial for modulating the ATPase activity of myosin via stability of its super relaxed state and its interaction with F-actin, we captured events in the adjacent C2 domain, an F-actin binding domain [75,76]. Given the decrease in super-relaxed-state myosin heads, this event could represent a novel mode of phosphorylation events increased in HCM regulation for this process and energetics. We also noted increased phosphorylation LMOD2, another actin binding protein that regulates contractility, and the protein phosphatase 1 regulatory subunit 12B (PPP1R12BSer842) and rho-associated protein kinase 1 (ROCK1Ser1105), upstream modulators of calcium sensitivity of the contractile machinery (Figure 5B) [77,78,79]. From our proteomics analysis of the early disease model, we observed decreases in the abundance of key plasma membrane calcium channels, as well as electrochemical gradient-forming ATPases (Supplementary Table S1A). These included decreased expression of the β-2/-3 and α2/δ3 subunits of the L-type calcium channel and plasma membrane calcium-transporting ATPase 1 (PMCA1), which directly modulate calcium flux at the plasma membrane [80,81]. Moreover, the MYH7R403Q+/− hiPSC-CM samples had decreased levels of the sodium/potassium-transporting ATPase subunit alpha-1/2 (catalytic) and beta-1/2, which are essential for cardiac excitability by assisting in calcium extrusion at the membrane [82]. Furthermore, they displayed reductions in CASQ2 and cardiac junction (ASPH), which play key roles in SR Ca2+ storage and the magnitude of release during excitation-contraction coupling buffering via interacting with ryanodine channels [83,84]. Overall, these findings suggested increased SR Ca2+ storage, with concurrent increases in Ca2+ flux throughout the SR in the MYH7R403Q+/− cells. To independently validate our prediction of defects in Ca2+ handling in the mutant cells, we directly measured Ca2+ flux in paced hiPSC-CMs via live-cell Ca2+ imaging. As summarized in Figure 5C, we observed a significant increase in SR Ca2+ storage in the HCM cells relative to WT following caffeine treatment, in line with our proteome measurements of CASQ2 expression. Surprisingly, however, we also detected a significant increase in time of Ca2+ release and re-uptake into the SR, despite the increased flux predicted from our phosphoproteome analysis. We note that this counterintuitive result is potentially explained by the increase in overall SR Ca2+, by which the additional Ca2+ must be pumped against a higher concentration gradient, thus increasing ATP2A2 (SERCA) energetic requirements. This would be expected to exacerbate ATP production demands on the mitochondria and to further thermodynamically limit SERCA function [85]. Focusing on pronounced changes in the HCM tissue samples, we again consistently measured differential phosphosites belonging to the excitation contraction coupling mechanism in advanced disease, as was seen in the early-stage model, though we noted a markedly different phosphorylation landscape (Figure 5D, Supplementary Table S3B). Top differential excitation contraction coupling phosphosites included downregulation of RYR2Ser2811 and JPH2Ser247 and increases in RYR2Ser2363 AKAP-12Ser283,286,T285/-13Ser2398,2728, ankyrin-1 (ANK1Ser1686). Yet, in contrast to the early HCM model, we detected decreases in PLNSer16,Thr17, RYR2Ser2814, and ATP2A2Ser663, which were increased in the early disease model samples (Figure 5E, Supplementary Table S10B). We also noted increases in effectors that directly phosphorylate SR ATPases (e.g., RYR2), calcium channels, and sarcomere proteins to modulate Ca2+ homeostasis and sensitivity, which dictate overall ATP utilization. For example, protein phosphatase 1 and its regulatory subunits decrease PLNThr17 phosphorylation, allowing PLN to increase its inhibitory effects on ATP2A2 [74]. In the HCM tissue, we captured differential phosphorylation on regulatory subunits PPP1R12BSer711, PPP1R12CSer509, 453, PPP1R12AThr443, Ser445, PPP1R3ASer649, and PPP1R3DSer46. Other effectors seen in the clinical specimens included key upstream modulators of Na+ and Ca2+ homeostasis regulation, such as calcium/calmodulin-dependent protein kinase type II subunit beta (CAMK2BSer367,276). Other important regulators of sarcomeric calcium sensitivity showing persistent changes in advanced disease included differential phosphorylation of TTN, FLNC, synaptopodin-2 (SYNPO2), MYBPC3, and myosin regulatory light chain 2 (MYL2) (Figure 5D). In particular, the MYL2Ser15,19,Thr24 decreased phosphorylation levels in advanced HCM, as well as decreased phosphorylation of its upstream kinase MYLK3Ser152,355,Thr359, having important regulatory implications, since these phosphosites function in cross-bridge kinetics [86]. For instance, the serine-15 modification increases lever arm stiffness and myosin attachment, enhancing contraction and protecting against hypertrophy-related stress [87,88]. Among the main plasma membrane channels detected in the clinical samples, we found decreased phosphorylation of voltage-dependent L-type calcium channel subunits (CACNB2Ser514,550 and CACNA1CSer1784) and potassium voltage-gated channels (KCNH2Ser255) in the HCM tissue. These components regulate the influx of calcium into the cytoplasm and the rectifying potassium current, respectively, and their phosphorylation provides precise regulation of open probability [89]. The overall phospho-pattern of SR proteins and modulators of contractility suggests differing signaling-based regulation in advanced disease compared to the early specimens. While previous molecular studies by our group and others have elucidated important mechanisms of HCM pathobiology, few have explored consequences associated with MYH7 mutations over different disease-state phenotypes [9,90,91]. Through an integrated quantitative multi-omics analysis, we captured hundreds of additional differential features which mapped to distinct molecular mechanisms dictating mitochondrial homeostasis in both early and advanced models of HCM, as well as stage-specific metabolic pathways and excitation-coupling mechanistic adaptations. Although there are well-known limitations associated with hiPSC-based models, in particular the immature fetal-like maturation states, our mutant system provides a tractable means to explore the direct molecular and phenotypic consequences of defective MYH7 function at the early stages of pathology. Conversely, despite substantial clinical heterogeneity in HCM presentation, our comparative analysis of patient myectomy specimens, belonging to a single defined mutation subpopulation, allowed us to elucidate perturbations associated with both early and late stages of pathology. Moreover, while we analyzed an array of MYH7 mutations in this subpopulation, we were able to identify consistent findings amongst late-stage HCM specimens. Oxidative stress and mitochondrial damage are commonly described features of HCM [9,18]. In physiological conditions, the deleterious effects of ROS (e.g., oxidative modification) are balanced by endogenous antioxidant systems [92]. However, excess production can impair the function of mitochondrial components [5,93,94]. We observed a mixed antioxidant response in the mutant hiPSC-CM, followed by a near total loss in this pathway signature in the advanced myectomy specimens. The early-stage in vitro model suggests increased protection against protein glutathionylation through thiol-based systems, the primary intracellular redox buffer for protecting protein cysteine residues from oxidative modification. We found evidence in both our proteomics and metabolomics data for increased antioxidant protection via expression of glutaredoxin and thioredoxin enzymes and upregulation of the pentose phosphate pathway. The latter provides NADPH for reducing oxidized glutathione, allowing another redox cycle [95]. From our targeted metabolomics experiments, we confirmed that glutamine moved faster towards GSH synthesis in the mutant cells, consistent with a crucial role of glutaminolysis for cardiomyocytes under oxidative stress [96]. Thus, there appears to be a relatively robust intracellular antioxidant protection early on, which likely serves to counteract the known overproduction of ROS reported before in in vitro HCM models [14,15]. Conversely, our myectomy data revealed a profound loss of these protective mechanisms in advanced stage disease, including TXN2 and its reductase TXRN2. These expression deficits are likely associated with the reduced ROS scavenging capabilities noted by previous studies of HCM myectomy samples, particularly the increased ratio of GSSG:GSH (i.e., reduced GSH antioxidant capacity) [15,97]. Interestingly, Wang et al. recently reported increased evidence of pentose phosphate pathway in advanced HCM, but our own patient cohort data suggest the opposite trend [90]. In contrast to their KEGG pathway analysis, our integrative MOMENTA framework detected significant downregulation of transaldolase and the NAD-dependent master regulatory deacetylase enzyme sirtuin-2. One potential explanation for this discrepancy is the clinical heterogeneity of HCM, which varies greatly from mutation to mutation. While we exclusively utilized MYH7 mutants, Wang et al. analyzed samples representing an array of mutations. Overall, our observation of poor thiol-based antioxidant protection in the clinical HCM specimens suggests exacerbation of the mitochondrial damage and dysfunction observed before in myectomy specimens [9]. Given the importance of double-strand break response in oxidative stress, the decreased expression of DNA repair mechanisms in our analysis of advanced HCM specimens further demonstrates poor oxidative stress adaptation [66]. Mitochondrial DNA is especially vulnerable to oxidative modification given its proximity to ROS production [98]. Hence, the decreased expression of BER enzymes in the mutant hiPSC-CM model implies the potential accumulation of damaged mtDNA at the early stages of pathology. Mitochondrial damage is known to result in reduced bioenergetics and could be the cause of the overall decreased ATP generation and mitochondrial metabolic pathways noted before in advanced HCM [9,99]. Further supporting this, a number of cardiomyopathies linked to mtDNA mutations exhibit pronounced mitochondrial dysfunction [100]. Mitochondrial dynamics are an important mechanism for maintaining mitochondrial homeostasis and function [23,101]. Mitochondrial fusion ensures optimal respiratory function by exchanging contents (e.g., proteins and DNA) after merging [102,103]. Conversely, fission is the division of a single mitochondrion into two daughter organelles, which assists in mitochondrial distribution but can also serve in quality control during oxidative stress via activation of mitophagy. Imbalances between these processes can result in malfunctioning mitochondria and metabolic disturbances [101]. There is substantial evidence that fission and fusion processes are active in HCM pathology, especially given the increases in mitochondrial number and cristae disorganization noted before in both clinical and in vitro culture models of HCM [9,23]. Accordingly, we noted an increase in mitochondrial dynamics in response to mutation in both the early and advanced pathology. In our advanced specimens, we found significant phosphorylation of DNM1LSer616 and its recruitment factors MFFSer157 and MIEFSer59, which are the principal components for fission [104]. This echoes the increased evidence from Ranjbarvaziri et al., in which they found this ultrastructural mitochondrial remodeling from imaging [9]. Perplexingly, despite increased oxidative stress (e.g., elevated 4-hydroxy-2-nonenal-modified proteins) and the mitochondrial damage noted in their study, they failed to find molecular activation of the fusion/fission mechanisms. In contrast, we provide direct phosphorylation level regulation of this pathway, which helps explain those ultrastructural changes in HCM. Though we found that the regulators of fission and fusion were upregulated in both early and advanced stages, we only found direct evidence of mitophagy in the in vitro culture model. Mitophagy is important for clearing impaired mitochondria via the autophagosome–lysosome pathway following fission/fusion and would provide another means of combating oxidative stress [105]. We failed to find direct molecular evidence to support elevated mitophagy in advanced specimens. Mitochondrial dynamics and mitophagy are dependent on cardiolipin lipid species for recruiting fusion and fission mediators [106]. Consistent with this, our MOMENTA of the hiPSC-CM samples showed increased enrichment of cardiolipin biosynthesis in the mutant cells (Figure 4A), as well as increased expression of cardiolipin synthase (CRLS1), which catalyzes the last step in de novo cardiolipin synthesis (Supplementary Table S1A). In contrast, the advanced specimens showed decreased expression of CRLS1 and the lysocardiolipin acyltransferase 1 (LCLAT1) (Supplementary Table S3A), both crucial enzymes for cardiolipin production in vivo [107]. Given the previously reported decreases in cardiolipin species in advanced HCM, the failure to upregulate these mechanisms could be associated with failure to upregulate mitophagy in advanced specimens [9]. The metabolism of cardiomyocytes is dynamic and cardiac tissues are able to adapt their preferences for carbon sources depending on metabolic demand [96,101,108]. Our discovery of increased glutamine anaplerosis in early mutant samples is one such demonstration, noting that glutamine catabolism may serve as an important contributor to preservation of cardiac function in early-stage pathology. We demonstrated that increased glutaminolysis serves a biosynthetic role in the mutant CMs, replenishing the TCA intermediate 2-oxoglutarate via transamination of glutamate, as well as providing a major precursor for glutathione synthesis. In contrast, the advanced samples exhibited decreased glutamate degradation, a pathway downstream of glutamine anaplerosis. Recently, Watanabe et al. found that glutaminolysis improves cell viability in healthy cardiomyocytes under oxidative stress [96]. We provide compelling evidence that this mechanism provides a dual protective role in HCM, while its loss in advanced stage disease likely represents a further maladaptation to chronic oxidative stress. Excitation contraction coupling, particularly ATPase-dependent components, rely on readily available ATP for contiguous function. Unsurprisingly, our global proteome and phosphorylation landscape captured distinct calcium handling and calcium sensitivity between HCM and normal controls in both the in vitro and in vivo contexts. Our analysis of the mutant cell cultures, which showed increased metabolic activity supplemented by glutaminolysis, revealed significant activation of SR calcium handling proteins in HCM. We noted increases in PLNSer16,Thr17, events that remove its reversible inhibition of ATP2A2 [74]. Importantly, these sites are regulated by distinct kinases (e.g., CAMK2A vs. PKA), demonstrating multiple levels of regulation. Since this latter enzyme accounts for the majority of Ca2+ reuptake, reducing the degree of inhibition is predicted to increase SR Ca2+ reuptake activity. This regulation is likely necessary given the increased SR Ca2+ storage in MYH7R403+/− cells, revealed from Ca2+ imaging, particularly when coupled to the decrease we observed in PMCA1 and supporting Na+/K+ ATPases, which assist ATP2A2 in decreasing cytosolic Ca2+ for relaxation. This high Ca2+ reuptake burden on SR ATPases is expected to enhance ATP demands over time, and may be a predominant initial source of mitochondrial ATP demand. Conversely, we noted an opposite trend on SR handling in the advanced disease samples, with decreased phosphorylation events at the same sites. This suggests a lower burden of ATPase-dependent activity, perhaps due to decompensated ATP production. Other novel findings included phospho-regulation of contractility, such as decreased MYBPC3 phosphorylation in both systems. Collectively, this study fills in gaps from previous studies of HCM and expands knowledge of the initial responses to mutations that protect against early-stage cardiac pathology that eventually are overwhelmed, leading to irreversible advanced disease. Beyond the molecular insights highlighted, all the data are being made publicly available to serve as a community resource for future mechanistic studies. Human induced pluripotent stem cell (hiPSC)-derived cardiomyocyte cell culture—Human iPSC-derived cardiomyocytes were generated from the Harvard Personal Genome Project line 1 (PGP1, GM23338), generously provided by the Seidman Lab at Harvard Medical School [23,109]. Heterozygous mutations were introduced into the MYH7 allele in hiPSC as described [23,110]. hiPSCs were cultured with mTESR1 media (STEMCELL Technologies, Vancouver, BC, Canada) on Matrigel (Thermo Fisher Scientific, Waltham, MA, USA)-coated plates and differentiated into cardiomyocytes (day 0) at 80–100% confluency via activation of the WNT pathway with 12 μM CHIR 99021 (Tocris, Bristol, UK) in RPMI + GlutaMAX media supplemented with B27 minus insulin (RPMI and B27 minus, Thermo Fisher Scientific). After 24 h, the cells were washed with PBS and given RPMI and B27 minus. On day 3, the WNT pathway was inhibited via 5 μM IWP-4 (Tocris) in fresh RPMI and B27 minus media for 48 h. On day 5, the media was replaced with fresh RPMI and B27 minus and changed every other day with the use of B27 plus insulin, starting on day 9. On day 11 and 13, the cardiomyocyte population was purified by metabolic selection via RPMI glucose-free media (Gibco, Waltham, MA, USA) supplemented with 4 mM of DL-lactate (MilliporeSigma, Burlington, MA, USA). On day 15, the media was replaced with RPMI + GlutaMAX media supplemented with B27 plus insulin. Following 48 h, cells were replated onto fibronectin-coated tissue culture plastic plates using 0.25% Trypsin-EDTA (Thermo Fisher Scientific) and 10 μg/mL deoxyribonuclease I (STEMCELL Technologies). They were then placed in RPMI B27+ supplemented with 5 μM Y-27632 (Tocris) and 2% fetal bovine serum (MilliporeSigma) for seeding. Cultures were maintained on RPMI + GlutaMAX media supplemented with B27 plus insulin until harvesting at 30+ days post-differentiation. For harvesting, cells were trypsinized, centrifuged, and snap frozen in liquid nitrogen for proteomics analysis. Patient Sample Acquisition—Surgical myectomy tissue was obtained from HCM patients with symptomatic LVOT obstruction. Patients gave informed consent for their myectomy tissue to be used in research. MYH7 pathogenic variants were identified by whole exome sequencing of peripheral blood mononuclear cell DNA and confirmed by commercially available Gene Panels (GeneDx, Stamford, CT, USA). Myectomy sample processing was conducted as previously described [111,112,113,114]. A total of 100 mg of collected myectomy tissue was minced into 1 mm3 pieces, placed in 0.5 mL of CryoStor CS10 Freeze Media (STEMCELL Technologies), and stored in a MrFrosty (ThermoFisher) at 4 °C for 10 min, and then transferred to −80 °C overnight. Sample collection was approved by the Tufts University/Medical Center Health Sciences Institutional Review Board under IRB protocol # 9487. All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki. Tissue from organ donor patients without underlying cardiac disease was obtained and processed as described previously [111,112,113,114]. Homogenization, Metabolite/Protein Extraction, and Trypsin Digestion—The workflow is summarized in Figure 1C. For hiPSC-CMs, frozen samples were resuspended in a 4 vol. mixture of ice-cold methanol/acetonitrile/water (MeOH/ACN/H2O, 40/40/20, v/v; MS-grade, Fisher) in chemically resistant microcentrifuge tubes (Eppendorf, Hamburg, DEU). The samples were flash frozen in liquid nitrogen for 1 min, allowed to thaw, and sonicated on ice (Ultra Autosonic, Pune, IND) for 5 min three times. The extracts were then incubated at −20 °C for 1h and centrifuged at 12,000× g at 4 °C for 15 min to pellet the protein precipitate. The supernatants (with the metabolites) were transferred to new microtubes, dried under vacuum at 30 °C. The crude metabolite extracts were then subjected to solid-phase micro-extraction (SPME) for cleaning up. In brief, the coated SPME blades were preconditioned in MeOH/H2O (50:50, v/v) for 30 min. The samples were resuspended in 2% MeOH and incubated for 1 h with the resins (blades). After the blades were rinsed for 20s using water, the bound metabolites were desorbed using ACN/H2O (50:50, v/v) for 1 h. The eluted metabolites were dried under speedvac (Eppendorf) and kept at −80 °C prior to LC-MS analysis. The protein precipitate was transferred to 200 µL of lysis buffer containing 6 M GuHCl, 100 mM Tris pH 8.5, 1 mM CaCl2, 10 mM TCEP, and 40 mM CAM, with EDTA-free protease inhibitor (Sigma) and phosphatase inhibitor (PhosSTOP Roche, MilliporeSigma), and heated for 6 min at 95 °C. Human samples were placed in 5 vol. lysis buffer containing 8M urea, 5mM dithiothreitol (DTT), 50 mM ammonium bicarbonate (NH4HCO3), with EDTA free protease inhibitor (Sigma), and phosphatase inhibitor (PhosSTOP Roche). Afterwards, they were mechanically homogenized. All samples were sonicated on ice (Branson, Brookfield, CT, USA). Protein quantity was assessed by Bradford protein assay (Bio-Rad, Hercules, CA, USA), followed by digestion overnight at 37 °C with sequencing-grade Trypsin (1:50 enzyme to protein ratio, w/w, ThermoFisher). After adding trifluoroacetic acid to 0.1% v/v, peptide digests were desalted using a C18 Sep-Pak (Waters, Milford, MA, USA) according to the manufacturer’s instructions, resuspended in 100 mM TEAB, and quantified by Quantitative Colorimetric Peptide Assay (Pierce) prior to TMT labeling. Stable-Isotope Labeling and Offline Reverse-Phase High-Performance Liquid Chromatography Fractionation—For each sample, 100 µg of peptide digest (adjusted to 100 µL with 100mM TEAB) was mixed with a unique amine-reactive isotope-coded isobaric tandem mass tag (TMT-16-plex) reagent (Thermo Fisher Scientific) prior to sample multiplexing and precise quantification by LC/MS. After pooling, labeled peptide was desalted, dried, and suspended in 300 µL buffer containing 0.1% ammonium hydroxide and 2% acetonitrile (ACN). The pooled sample mixture was pre-fractionated by high pH reverse-phase HPLC on a XBridge Peptide BEH C18 column (130Å, 3.5 μm, 4.6 mm × 250 mm, Waters) using an Agilent 1100 HPLC system. Peptides were eluted using a gradient of mobile phase A (0.1% NH4OH −2% ACN) to B (0.1% NH4OH −98% ACN) over 48 min and collected as 12 pooled fractions. For phosphoproteomics, the bulk (95%) of each sample was subject to phospho-peptide enrichment using FeO2 metal-chelate resin (PureCube Fe-NTA MagBeads, Cube Biotech, Rhein, DEU) [24], while the remaining (5%) portions were analyzed directly by nanoflow LC/MS as bulk proteome measurements (a total of 24 injections, 12 for proteomics and 12 for phosphoproteomics). Mass Spectrometry Analysis of Peptides and Identification—Isotope-labeled peptides were reconstituted in mobile phase A (0.1% formic acid, 2% ACN) prior to LC/MS analysis on a Thermo-Fisher Exploris 480 hybrid quadrupole-Orbitrap mass spectrometer interfaced with Thermo-Fisher FAIMS Pro with integrated Proxeon EASY-nLC 1200 system. After loading onto a C18 reverse-phase pre-column (75 μm i.d. × 2 cm, 3 μm, 100Å, Thermo Fisher Scientific), peptides were gradient separated on an EASY-Spray C18 nanocolumn (75 μm i.d. × 50 cm, 2 μm, 100Å; ES803A, Thermo Fisher Scientific) using 2–35% mobile phase B (0.1% formic acid, 80% ACN) over 120 min (proteome) or 180 min (phosphoproteome), and electro-sprayed at ~250 nL/min into the Exploris instrument operated in positive ion mode (capillary temperature 275 °C, 2100 V potential). Data-dependent spectra were acquired automatically via high-resolution (60,000) precursor ion scan (350–1500 m/z range) to select the 12 most intense peptides for MS/MS fragmentation by high energy dissociation (normalized collision energy of 33 at 45,000 resolution). The resulting RAW files were searched by MaxQuant (1.6.7.0) using default settings against the human proteome (SwissProt Taxonomy ID: 9606, downloaded September, 2021), allowing for two missed cleavage sites and variable modifications (Ser/Thr/Tyr phosphorylation, N-terminal acetylation, and Met oxidation) and carbamidomethylation of cysteine and TMT labels as a fixed modification. Peptide- and protein-level matches were filtered to high confidence (1% FDR), with a minimum phosphosite localization probability of 0.7. TMT quantification involved label correction (lot values provided by ThermoFisher). Phosphoproteomics Statistical Analysis and Pathway Enrichment—Bioinformatic analysis was performed using R (language and environment for Statistical Computing; http://www.R-project.org, accessed on 12 May 2021.). Peptide feature intensities were log transformed and loess normalized. LIMMA R package was used for differential analysis (moderated Student t-tests), and to generate ranked lists for subsequent enrichment analysis using the Benjamini–Hochberg FDR correction [25,115]. Statistical enrichment analysis was performed using fgsea R package [28,68]. Volcano plots were created with the EnhancedVolcano R package [116]. All figures were created with BioRender.com. Mass Spectrometry of Metabolites and Identification—Metabolites were reconstituted in mobile phase A (2% ACN) prior to metabolite nanoflow (nLC) LC/MS analysis on a Hybrid Quadrupole-Orbitrap Q-Exactive HF (Thermo Scientific). After loading onto a C18 reverse-phase pre-column (75 mm i.d. × 2 cm, 3μm, ThermoScientific), metabolites were separated on a capillary column (75 mm i.d. × 25 cm, 2 μm, 100 Å, ThermoFisher Scientific) using a gradient of 2% to 60% mobile phase B (80% ACN) for 20 min, increased to 95% mobile phase B over 10 min, and maintained at 95% mobile phase B for 15 min, and electro-sprayed at 300 nL/min into the Q-Exactive HF. Data-dependent spectra were acquired automatically (automated switching ESI mode) via high-resolution (60,000) precursor ion scan over a full mass scan range of m/z 67−1000. The source ionization parameters were optimized for a transfer temperature at 300 °C and a spray voltage set to 2.1 kV and −1.8 kV for the positive and negative modes, respectively. MS2 scans were performed at 15,000 resolution, with a maximum injection time of 64 ms, using stepped normalized collision energies (NCEs) of 10, 20, and 40. Dynamic exclusion was enabled using a time window of 10s. Raw data (switching mode) were split into positive and negative files and subject to OmicsNotebook (R script) for peak detection, deconvolution, retention time alignment, and metabolite identification against open database [25]. Intensities of features (putative metabolites) were normalized prior to differential analysis (moderated Student t-test). Live-Cell Calcium Imaging—On day 30+ of differentiation, WT and mutant hiPSC-CMs were plated at 80k cells/well in a 24-well plate. Three days after plating the hiPSC-CMs, they were washed three times in Tyrode’s buffer (Thermo Fisher Scientific) and incubated in 10 μM Rhod-3 AM calcium indicator, PowerLoad, and Probenecid for 45 min covered from light, following the protocol from the Rhod-3 AM calcium imaging kit (ThermoFisher). Cells were washed in Tyrode’s buffer and incubated in Probenecid for an additional 45 min. Cells were then washed in Tyrode’s buffer three additional times and kept in Tyrode’s during live imaging. Calcium transients were acquired at 30 frames per second at 6X on a Nikon Eclipse Ti (Nikon Instruments, Tokyo, JP) with an Evolve EMCCD Camera (Photometrics, Tucson, AZ, USA), equipped with a temperature and CO2 equilibrated environmental chamber. hiPSC-CMs were electrically stimulated at 1 Hz using a C-Pace EP stimulator (IonOptix, Westwood, MA, USA). Videos were acquired at 15 locations per experimental group with a 560 nm laser illumination wavelength. Calcium transients of each cell were calculated using a custom Matlab script tracking intensity change over time within the cell. Calcium release was calculated as the time it took the calcium intensity in the cytoplasm to reach 50% maximum intensity, while calcium reuptake was calculated as the time it took the calcium intensity in the cytoplasm to decrease to 50% of its maximum intensity. This was repeated over three differentiations (N = 3), with n > 23 cells per differentiation. SR storage was determined from the ratio of the calcium transient immediately following and preceding incubation with the addition of 20 mM caffeine to Tryrode’s buffer (final concentration of 10 mM). All data were normalized to the average WT values for the matched differentiation. Data were assessed with a two-way ANOVA with a mixed-effect model that was corrected for multiple comparisons using a Tukey test with a 95% confidence level. Targeted C13 Flux and BPTES application—C13 metabolomics flux metabolomics analyses were performed according to Yuan and collaborators [117]. Briefly, hiPSCs-derived cardiomyocytes, at day 30, were cultured during 24 h in RPMI glucose free + B27+ and 13C6 glucose (CLM-1396) added at a final concentration of 11.1 mM for glucose flux or cultured in RPMI glutamine free + B27+ and 13C5 glutamine (CLM-1822) added at a final concentration of 2.05 mM for glutamine flux. Metabolites were extracted with −80 °C cold 80% methanol (LC-MS grade), centrifuged, and supernatant was dried on speedvac over 18 h. Metabolite pellets were stored at −80 °C up to analysis (no more than 7 days) and protein pellets were used to normalize metabolites levels. Targeted metabolomics was performed using SRM LC-MS (13C glucose: N = 5 and 13C glutamate: N = 8, where N = number of differentiations). In cases of no analyte measurement (i.e., below detection limit) in a given sample, analyte values were excluded for that sample. All possible 13C labeled transitions were monitored. BPTES inhibition of glutamine oxidation was performed according to the manufacturer’s instructions (Agilent Seahorse XF Mito Fuel Flex Test Kit), 2 differentiations. Contractile parameters were assessed according to Toepfer and collaborators [118]. Data were assessed via unpaired, two-tailed t-test with Welch’s correction. The schematic in panels (B) and (C) was created in reference to Figure 1 of Fan et al. and Mcdonald et al. [119,120].

PMC10002556

Milica Jaksic Karisik,Milos Lazarevic,Dijana Mitic,Nadja Nikolic,Maja Milosevic Markovic,Drago Jelovac,Jelena Milasin

Osteogenic and Adipogenic Differentiation Potential of Oral Cancer Stem Cells May Offer New Treatment Modalities

28-02-2023

oral cancer,cancer stem cells,CD44,osteogenic differentiation,adipogenic differentiation,miRNA-21,miRNA-133 and miRNA-491

(1) Treatment failure of oral squamous cell carcinoma (OSCC) is generally due to the development of therapeutic resistance caused by the existence of cancer stem cells (CSCs), a small cell subpopulation with marked self-renewal and differentiation capacity. Micro RNAs, notably miRNA-21, appear to play an important role in OSCC carcinogenesis. Our objectives were to explore the multipotency of oral CSCs by estimating their differentiation capacity and assessing the effects of differentiation on stemness, apoptosis, and several miRNAs’ expression. (2) A commercially available OSCC cell line (SCC25) and five primary OSCC cultures generated from tumor tissues obtained from five OSCC patients were used in the experiments. Cells harboring CD44, a CSC marker, were magnetically separated from the heterogeneous tumor cell populations. The CD44+ cells were then subjected to osteogenic and adipogenic induction, and the specific staining was used for differentiation confirmation. The kinetics of the differentiation process was evaluated by qPCR analysis of osteogenic (Bone Morphogenetic Protein—BMP4, Runt-related Transcription Factor 2—RUNX2, Alkaline Phosphatase—ALP) and adipogenic (Fibroblast Activation Protein Alpha—FAP, LIPIN, Peroxisome Proliferator-activated Receptor Gamma—PPARG) markers on days 0, 7, 14, and 21. Embryonic markers (Octamer-binding Transcription Factor 4—OCT4, Sex Determining Region Y Box 2—SOX2, and NANOG) and micro RNAs (miRNA-21, miRNA-133, and miRNA-491) were also correspondingly evaluated by qPCR. An Annexin V assay was used to assess the potential cytotoxic effects of the differentiation process. (3) Following differentiation, the levels of markers for the osteo/adipo lineages showed a gradual increase from day 0 to day 21 in the CD44+ cultures, while stemness markers and cell viability decreased. The oncogenic miRNA-21 also followed the same pattern of gradual decrease along the differentiation process, while tumor suppressor miRNA-133 and miRNA-491 levels increased. (4) Following induction, the CSCs acquired the characteristics of the differentiated cells. This was accompanied by loss of stemness properties, a decrease of the oncogenic and concomitant, and an increase of tumor suppressor micro RNAs.

Osteogenic and Adipogenic Differentiation Potential of Oral Cancer Stem Cells May Offer New Treatment Modalities (1) Treatment failure of oral squamous cell carcinoma (OSCC) is generally due to the development of therapeutic resistance caused by the existence of cancer stem cells (CSCs), a small cell subpopulation with marked self-renewal and differentiation capacity. Micro RNAs, notably miRNA-21, appear to play an important role in OSCC carcinogenesis. Our objectives were to explore the multipotency of oral CSCs by estimating their differentiation capacity and assessing the effects of differentiation on stemness, apoptosis, and several miRNAs’ expression. (2) A commercially available OSCC cell line (SCC25) and five primary OSCC cultures generated from tumor tissues obtained from five OSCC patients were used in the experiments. Cells harboring CD44, a CSC marker, were magnetically separated from the heterogeneous tumor cell populations. The CD44+ cells were then subjected to osteogenic and adipogenic induction, and the specific staining was used for differentiation confirmation. The kinetics of the differentiation process was evaluated by qPCR analysis of osteogenic (Bone Morphogenetic Protein—BMP4, Runt-related Transcription Factor 2—RUNX2, Alkaline Phosphatase—ALP) and adipogenic (Fibroblast Activation Protein Alpha—FAP, LIPIN, Peroxisome Proliferator-activated Receptor Gamma—PPARG) markers on days 0, 7, 14, and 21. Embryonic markers (Octamer-binding Transcription Factor 4—OCT4, Sex Determining Region Y Box 2—SOX2, and NANOG) and micro RNAs (miRNA-21, miRNA-133, and miRNA-491) were also correspondingly evaluated by qPCR. An Annexin V assay was used to assess the potential cytotoxic effects of the differentiation process. (3) Following differentiation, the levels of markers for the osteo/adipo lineages showed a gradual increase from day 0 to day 21 in the CD44+ cultures, while stemness markers and cell viability decreased. The oncogenic miRNA-21 also followed the same pattern of gradual decrease along the differentiation process, while tumor suppressor miRNA-133 and miRNA-491 levels increased. (4) Following induction, the CSCs acquired the characteristics of the differentiated cells. This was accompanied by loss of stemness properties, a decrease of the oncogenic and concomitant, and an increase of tumor suppressor micro RNAs. Oral squamous cell carcinoma (OSCC) is a frequent and aggressive malignancy in the group of head and neck tumors, with a prevalence of over 630,000 new cases per year worldwide [1]. Despite advances in multimodal therapeutic strategies, the main problem remains the resistance to chemotherapy and biologic agents, resulting in a poor 5-year survival rate of only 50% [2,3]. Recent evidence suggests that one of the reasons for OSCC therapy failure is the presence of a small pluripotent cell subpopulation identified as “cancer stem cells” (CSCs), which are considered to have a tumor-initiating and self-renewal ability [4]. Studies also indicate that CSCs possess a remarkable differentiation capacity [5,6], and this characteristic could potentially be exploited for novel OSCC therapeutic modalities, as conventional therapy regimens may effectively treat the bulk tumor mass yet leave CSCs behind as a source for tumor recurrence upon treatment [7]. The transmembrane glycoprotein CD44 has been recognized as a characteristic CSC surface marker that may be used independently or in combination with other markers for the identification of CSCs in various cancers [6]. CSCs show a high level of expression of embryonic (Octamer-binding Transcription Factor 4—OCT4, Sex Determining Region Y Box 2—SOX2, and NANOG) stem cell markers and different micro RNAs (miRNAs) as well [7,8]. More specifically, the overexpression of CD44 has been shown to increase the expression of embryonic transcription factors OCT4, SOX2, and NANOG [9]. Several miRNAs have been involved in cancer development, acting either as tumor suppressors or as oncogenes [10]. MiRNAs are short (20 to 24 nucleotide) non-coding RNA molecules that regulate the expression of up to 50% of all protein-coding genes at the post-translational level [11]. They are involved in a variety of biological processes, such as cell proliferation, apoptosis, immunological response, etc. [12,13]. As oncogenes or tumor suppressors, miRNAs play a critical role in cancerogenesis via the regulation of self-renewal and apoptosis across CSC signaling pathways [14]. MiRNA-21 is an oncogenic miRNA that is overexpressed in a variety of malignancies, and in OSCCs, its overexpression has been correlated with poor prognosis [15]. The NANOG signaling axis has been shown to stimulate the overexpression of miRNA-21 and regulate cell growth and self-renewal in CD44+ cells [16,17]. However, the present literature did not investigate the difference in the miRNA-21 expression between CSCs and the heterogenic cancer cell population or whether CSC differentiation affects the expression of miRNA-21, or other miRNAs, either with oncogenic or tumor suppressor roles. Thus, the aims of the study were to: (a) compare the multipotency of CD44+ and CD44− cells isolated from the same primary OSCC cultures by estimating their osteogenic and adipogenic differentiation capacity, (b) assess the effect of osteo- and adipo-induction on stemness-related genes’ expression, (c) analyze the influence of differentiation on apoptosis, and (d) investigate miR-21, miR-133, and miR-491 expression levels in CSCs during differentiation. CD44+ positive cells were isolated from primary tumor cell cultures (Figure 1a) (i.e., separated from the remaining CD44− cells) by a magnetic-activated cell sorting (MACS) system, and the further expanded (Figure 1b,c) MACS isolated cells were validated by flow-cytometry, resulting in over 92% of the population expressing CD44 (the data are not shown). Additionally, the cancer stemness properties of the isolated CD44+ cells were confirmed by the formation of tumor spheroids and colonies (Figure 2). CD44+ and CD44− cells, grown in an osteogenic differentiation medium for 7, 14, and 21 days, were stained with Alizarin Red S (Figure 3a–f), and the color was quantified (Figure 3g). The formation of mineralized nodules was significantly higher in the CD44+ compared to the CD44− cell cultures after 7 (p = 0.0009), 14 (p = 0.00001), and 21 days (p = 0.00001) of cultivation (Figure 3g), pointing to a marked osteogenic differentiation potential of the CD44+ cells. In addition, to confirm the osteogenic differentiation after 21 days, BMP4, RUNX2, and ALP expression levels were assessed by qPCR (Figure 4). There were no major variations in the expression of the different osteogenic markers between the patients, as suggested by the relatively small standard deviations. The population of the CD44+ cells cultivated in the osteogenic medium showed a significantly higher expression of BMP4 (a 3.3-fold increase, p ≤ 0.0001), RUNX2 (a 1.3-fold increase, p ≤ 0.0001), and ALP (a 2.3-fold increase, p ≤ 0.0001) compared to the un-induced cells. Following the same conditions, the osteogenic induction was done in the CD44− cell population. Compared to CD44− cancer cells, the CD44+ cancer cell population has a substantially higher osteogenic differentiation potential, as judged from the relative expression levels of all the analyzed osteogenic markers (p = 0.0001). To evaluate the kinetics of the osteogenic induction, we examined the relative gene expression of the BMP4, RUNX2, and ALP at the beginning (0 days) and after 7, 14, and 21 days of osteo-differentiation. Only the CD44+ cells were used in this experiment, and the levels of osteogenesis-related genes generally increased throughout time (Figure 5). CD44+ and CD44− cells cultivated for 14 and 21 days in an adipogenic differentiation medium were stained using Oil Red O (Figure 6a–d), and the color was quantified (Figure 6e). The accumulation of neutral triglycerides and lipids was significantly higher in the CD44+ compared to the CD44− cell cultures after 14 (p ≤ 0.0001) and 21 days (p ≤ 0.0001) of cultivation (Figure 6e). This demonstrated that the CD44+ cells were capable of adipogenic differentiation. In addition, to confirm the adipogenic differentiation after 21 days, FAP, LIPIN, and PPARG expression levels were assessed by qPCR (Figure 7). Generally, there were no major variations in the expression of the different adipogenic markers between the patients, as suggested by the relatively small standard deviations. Populations of the CD44+ cells cultivated in the adipogenic medium showed a significantly higher expression of FAP (a 23.2-fold increase, p = 0.0001), LIPIN (a 7.6-fold increase, p = 0.002), and PPARG (a 1.2-fold increase, p = 0.002) compared to the un-induced cells. Following the same conditions, the adipogenic induction was conducted in the CD44− cell population. Comparing the relative expression levels of all the adipogenic markers, a statistically significant difference was noted between the CD44+ and CD44− cells (p = 0.0001), confirming their greater adipogenic differentiation potential based on the expression of specific markers after the adipogenic induction. To evaluate the kinetics of the adipogenic induction, we examined the relative gene expression of the FAP, LIPIN, and PPARG at the beginning (0 days) and after 7, 14, and 21 days of the adipo-differentiation. As for the osteogenic differentiation, only the CD44+ cells were used. The levels of adipogenesis-related genes generally increased throughout time (Figure 8). The relative gene expression of the stem cell markers OCT4 (Figure 9a), SOX2 (Figure 9b), and NANOG (Figure 9c) was evaluated to confirm the cancer stem cell features of the CD44+ cells. The gene expression of all three markers was significantly higher in the CD44+ cells compared to the CD44− cells (p ≤ 0.001). To assess whether the cell differentiation led to alterations in the stem cell markers’ expression levels (OCT4, SOX2, and NANOG), the RNA was isolated from the cells grown in the osteogenic, adipogenic, or complete growth mediums for 21 days. The expression levels of the stem cell markers in the CD44+ cells were significantly lower (p ≤ 0.05) after the osteogenic (OCT4- 1.8, SOX2- 2.4, NANOG- 4.7–fold decrease) and adipogenic differentiations (OCT4- 6.5, SOX2 -4.3, NANOG -6.1-fold decrease) compared to the un-induced, control cells (Figure 10a–c). The expression levels of the stem cell markers in the CD44− subpopulation remained unchanged compared to the control (Figure 10d–f). To evaluate the kinetics of cancer stemness changes during the osteogenic and adipogenic inductions, we examined the relative gene expression of the OCT4, SOX2, and NANOG markers in the CD44+ cells at the beginning (0 days), and after 7, 14, and 21 days of differentiation. The levels of all the stemness-related genes decreased throughout time at a different pace (Figure 11). Upon the osteogenic and adipogenic differentiation inductions, considerable cell death by apoptosis was noted, but it was less pronounced in the CD44+ than the CD44− cells (Figure 12). The percentage of apoptotic (early+late apoptosis) CD44− cells was 34.88 after the OI and 64.01 after the AI (Figure 12g) and was almost 2 times higher compared to the CD44+ cells (15.52 after the OI and 27.99 after the AI) (Figure 12g). A significantly higher expression level of miRNA–21 was observed in the CD44+ compared to the CD44− cell cultures (Figure 13a) (p = 0.0008). The CD44+ cell differentiation influenced the expression level of miRNA-21. On the contrary, the miRNA-133 and miRNA-491 expression levels were significantly (p < 0.05) higher in the CD44− compared to the CD44+ cell culture. To evaluate the kinetics of the miRNA expression pattern during the osteogenic and adipogenic inductions, we examined the relative expression of miRNA-21, miRNA-133, and miRNA-491 in the SCC-25 cell line at the beginning (0 days) and after 7, 14, and 21 days of differentiation. The level of oncogenic miRNA-21 decreased throughout time, while the expression of miRNA-133 and miRNA-491 increased, suggesting their role as tumor-suppressors (Figure 14). CSCs represent a subpopulation of pluripotent cells in cancer that possess a high proliferative and self-renewal capacity. In vitro, CSCs grow faster than normal cells and have a high colony formation ability. They can be separated by using specific biomarkers, mostly located on the cell surface, such as CD44, CD133, EpCAM, etc. [18]. In this study, the characteristic marker, CD44, and magnetic-activated cell sorting were used to isolate the CSCs (CD44+ cells) from primary OSCC cell cultures. The main objective of our study was to investigate the capacity of CSCs to undergo differentiation into specific lineages and, to the best of our knowledge, this is the first report dealing in parallel with the osteogenic and adipogenic potential of CSCs originating from OSCC. In the present study, it was clearly shown that the CD44+ subpopulation exhibited a much higher potential for osteogenic and adipogenic differentiation compared to CD44−, as judged from the specific staining and expression of the respective markers. This finding is consistent with the stemness characteristics of CD44+ cells. Namely, the expression of embryonic stem cell markers, OCT4, SOX2, and NANOG, which play a crucial role in stemness maintenance and cell migratory capacity, was significantly higher in the CD44+ than in the CD44− cell populations. Concomitantly, with the increase of specific, either osteogenic or adipogenic markers, there was a sharp decrease in the embryonic stemness markers, a logical phenomenon accompanying the process of cell differentiation. The process of differentiation was monitored throughout time in the CD44+ cells, and a relatively gradual decrease of embryonic markers followed the similarly even increase of lineage-specific markers. Patil et al., in a recent study, demonstrated the capacity of OSCC CD44+ cells to undergo adipogenic differentiation and showed increased levels of adipogenic markers, concomitantly with the decrease of stemness markers (SOX2, NANOG and Kruppel-like factor 4—KLF4), which is fully in line with our findings [19]. Our findings are also in line with the findings reported by Zhau et al., who observed a decreased expression of stemness markers upon adipogenic induction in the human prostate cancer cell line (PC–3 cells) [6]. Milosevic et al. also noted a decrease of OCT4, SOX2, and NANOG expressions following the osteo- and chondro-induction of basal cell carcinoma CSCs [5]. The concept of CSC differentiation as an antineoplastic therapy is becoming increasingly appealing. Indeed, the process of differentiation induction results ultimately in the terminal differentiation of CSCs, which in turn irreversibly abolishes their tumorigenic potential and proliferative ability, causing them to become sensitive to conventional anticancer treatments [20,21]. Interestingly, the osteogenic and adipogenic inductions affected cell viability, implying that mature cells have a limited lifespan and eventually perish through programmed cell death mechanisms, mostly through apoptosis [22]. In the present study, adipogenic induction led to a 64% and 29% apoptosis rate in CD44− and CD44+ cell subpopulations, respectively, while osteogenic differentiation had a lesser impact on apoptosis. Zhau et al. noted that adipogenic induction led to the apoptosis of 77% of the PC–3 cell population [6]. New evidence, especially regarding miRNAs and CSCs, has completely changed our understanding of carcinogenesis. Small noncoding molecules (miRNAs) are considered to control the expression of more than 60% of human genes. The aberrant expression of miRNAs has been linked to the development of human malignancies and the regulation of stemness properties of CSCs [23,24]. The current findings indicate that changes in the expression of miRNAs are related to tumor functions [25]. In our study, we examined the levels of three micro RNAs: one oncogenic (miR–21) and two tumor suppressors (miRNA-133 and miRNA-491) in CD44+ and CD44− cells. It appeared that CD44+ cells had a significantly higher miRNA-21 expression and significantly lower miRNA-133 and miRNA-491 than the CD44− cells. Then, we examined the levels of all three micro RNAs during the process of differentiation, both osteo- and adipogenic in the CD44+ cells. We established that changes in the levels of micro RNAs followed the expected kinetics. Namely, in parallel with the differentiation process and accompanied by the loss of stemness properties, a significant decrease of miRNA-21 was registered. This is the first report indicating that miRNA-21 expression levels decrease in the OSCC CSCs’ subpopulation during osteo/adipo-differentiation. Our findings support the concept of miRNA-21 involvement in the maintenance of oral cancer CSCs’ stemness. Namely, this oncomiRNA has consistently been linked to colon and pancreatic CSCs’ regulation [26,27]. In anaplastic thyroid carcinoma therapy, the knockdown of miRNA-21 has significantly impacted the expression pattern of genes involved in the control of stemness, tumor growth, differentiation, and apoptosis [28]. On the contrary, levels of both the micro RNAs with tumor suppressor activity (miRNA-133 and miRNA-491) increased over time and the course of differentiation. Our findings are in agreement with those of Huan et al., who established that miRNA-491 was a modulator of OSCC behavior and that lower levels of this micro RNA were related to poorer survival, confirming its tumor suppressor role [29]. Similarly, He et al. showed that miRNA-133 was able to restrain OSCC proliferation and invasion, again corroborating its tumor suppressor role, which is fully in agreement with our findings [30]. Altogether, these findings suggest that differentiation leads to the loss of malignant phenotypes and the appearance of terminally differentiated cells with seemingly “normal” adult cell phenotypes. However, there are several limitations to our study. All the experiments have been conducted in vitro; therefore, it was not possible to assess how the microenvironment would affect the CSCs’ differentiation and the micro RNAs’ expression. In the near future, it will be necessary to examine in more detail the signaling pathways involved in OSCC CSC stemness and their relationship with different micro RNAs’ expression during differentiation. In conclusion, we have successfully isolated CD44+ cells from primary OSCC cell cultures and successfully induced osteogenic and adipogenic differentiation. The process of differentiation has affected the CD44+ stemness properties, their viability, and the levels of both oncogenic and tumor suppressor micro RNAs, implying that CSC differentiation might be used as a novel therapeutic modality. In the present study, primary cell cultures were generated from tumor tissue samples of five patients diagnosed with OSCC (3 males and 2 females, aged 59.2 ± 8.04 years, localization—tongue, the floor of the mouth, and gingiva; three patients had T2N0M0, and two had T4aN0M0 status; two patients had bone infiltration, and all of them had an HG2 NG2 tumor grade) obtained immediately after surgery from the Clinic of Maxillofacial Surgery of the School of Dental Medicine, University of Belgrade. All patients were informed of the study and signed a written informed consent form. All samples were examined by a pathologist, and the diagnosis of oral squamous cell carcinoma was confirmed. None of the patients recruited in this study received any preoperative chemotherapy or radiotherapy. The study was approved by the institutional Ethical Committee (No 36/6) of the University of Belgrade, Republic of Serbia, in accordance with the Declaration of Helsinki. Tissue samples were cut with blades into small pieces and washed three times with phosphate-buffered saline (PBS), then seeded into T25 cell culture flasks in a complete growth medium (Dulbecco’s Modified Eagle Medium (DMEM), supplemented with 10% of Fetal Bovine Serum (FBS), 100 U/mL of a penicillin–streptomycin solution, and 400 ng/mL hydrocortisone, all chemicals purchased from Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) under standard conditions in a humidified atmosphere with 5% CO2 at 37 °C. The complete growth medium was changed every 2–3 days, and after the cells reached 80% of confluence, the heterogenic cell population was magnetically sorted. In addition, the SCC–25 (ATCC®, CRL– 1628™) cell line was also used for miRNA kinetics experiments. The culturing conditions of the SCC–25 cell line were the same as for primary cultures. CD44+ cell separation from primary cultures and the SCC–25 cell line was performed using a magnetic-activated cell sorting (MACS) system (Miltenyi Biotec, San Francisco, CA, USA) according to the manufacturer’s protocol. Total populations of adherent cells (2 × 106) were enzymatically detached and counted. The cells were incubated with 100 μL of CD44 magnetic microbeads (Miltenyi Biotech) at 4 °C for 30 min. Upon incubation, the cell suspension was passed through the MACS MS column and placed in the magnetic field of a MACS separator. CD44-positive (CD44+) cells were retained in the column, and the unlabeled cells were eluted as a suspension known to consist of CD44-negative (CD44−) cells and seeded onto a new T25 flask [31]. When the column was removed from the magnetic field, the magnetically retained cells (CD44+) were also seeded into another flask for further experiments. As previously described [9], CD44+ cells were seeded at a density of 103/mL on 24-well culture plates that had been coated with 1ml poly-HEMA (poly 2−hydroxyethyl methacrylate, Sigma−Aldrich, Taufkirchen, Germany) to inhibit cell attachment. Cells were cultured in DMEM supplemented with B-27, N2, an epidermal growth factor, and antibiotics following germination (Sigma−Aldrich). Following 7 and 10 days of incubation, the size of spheroids was determined using ImageJ software 1.48 version (NIH, Bethesda, MD, USA) (Java 1.8.9_66). To determine the clonogenic potential of CD44+ cells, a colony-forming assay was performed, as previously described [32]. Briefly, CD44+ cells were seeded in a 32mm Petri dish at a density of 1000 cells/plate. After 7 days, the colonies were stained with 0.1% of a Coomassie Blue solution (Sigma−Aldrich). CD44+ and CD44− cells were seeded onto 24-well culture plates (1 × 104 per well) and cultured in the complete growth medium. After reaching 80% of confluence, the cells were cultivated in an optimized medium for osteogenic and adipogenic differentiation, respectively (StemMACS™, Miltenyi Biotec) for 7, 14, and 21 days. Cells were incubated under standard conditions (5% CO2 at 37 °C), and the medium was replaced every 3rd day. Cells were washed twice with PBS, then fixed with a 4% paraformaldehyde solution for 30 min at room temperature. An Alizarin Red S 2% solution (Centrohem, Belgrade, Serbia) for osteogenic differentiation and a 0.5% Oil Red O solution (Sigma−Aldrich) for adipogenic differentiation were poured over the cells, and the samples were incubated for 30 min and washed with distilled water. To quantify the calcium deposits and lipid droplets in the matrix, 10% acetyl pyridinium chloride was added for de-staining. The absorbance of the solution was measured at 450 nm OD using a microplate reader. The quantification was normalized against the stained cells grown in the complete growth medium [33]. CD44+ and CD44− cells were seeded into 24-well plates (1 × 105 per well) and cultured in respective differentiation media (1 mL of medium per well). After 14 days of induction, Annexin staining for detecting apoptosis was performed with an Annexin V–FITC Apoptosis Detection Kit (Invitrogen, Thermo Fisher Scientific) according to the manufacturer’s instructions. Annexin V–FITC staining was analyzed by flow cytometry, and the results were presented in a two-dimensional dot plot of propidium iodide (PI) versus Annexin V–FITC. PI was used to detect necrotic or late apoptotic cells. The plots were divided into four regions corresponding to: (a) viable cells, negative for both probes (PI/FITC −/−; Q3); (b) apoptotic cells, PI-negative and Annexin-positive (PI/FITC −/+; Q1); (c) late apoptotic cells, PI- and Annexin-positive (PI/FITC +/+; Q2); (d) necrotic cells, PI-positive and Annexin-negative (PI/FITC +/−; Q4). The cells cultured in the complete growth medium under the same conditions were used as controls. Total RNA was extracted from the 21-day-old osteo- and adipo-induced cells using a TRIzol Reagent (Invitrogen, Thermo Fisher Scientific), according to the manufacturer’s recommendations. The RNA concentration was measured using a microvolume spectrophotometer (BioSpec–nano Microvolume UV–Vis Spectrophotometer; Shimadzu Scientific Instruments, Columbia, MD, USA). An oligo d(T) primer and RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA) were used to synthesize cDNA from 2 µg of total RNA [34]. For assessing the miRNA–21 expression level in OSCC tissues, RNA was isolated from CD44+ and CD44− cell cultures. Real-time quantitative polymerase chain reaction (qPCR) was performed using the first strand cDNA, 0.2 μM forward and reverse primers, and a SensiFAST SYBR Hi–ROX Kit (Bioline, London, UK). The expression of the following markers was analyzed: ALP, RUNX2, OCN, and BMP2 (osteogenic) and PPARG, FAP, and LIPIN (adipogenic). In addition, the expression of embryonic stem cell markers (OCT4, SOX2, and NANOG) was also examined, after the CSCs’ separation and differentiation. The housekeeping gene, glyceraldehyde-3-phosphate dehydrogenase—GAPDH, was used as a reference. Relative gene expression values were calculated using the 2−ΔCt method [35]. The sequences of all primers used in this study are given in Table 1. Reverse transcription was accomplished using 15 ul reactions that consisted of 10× a Reverse Transcription Buffer, an RNase inhibitor, 100 mM deoxyribonucleotide triphosphate (dNTP), and a Multi Scribe Reverse Transcriptase, and containing 3 μL of a 5× concentrate miRNA-21-, miRNA-133-, and miRNA-491-specific primers. Thermal cycler settings were 16 °C for 30 min, 42 °C for 30 min, and 85 °C for 5 minutes, followed by cooling to 4 °C. Quantitative polymerase chain reaction (qPCR) was accomplished in 20 μL reactions using a TaqMan 20× concentrate of miRNA-21, miRNA-133, and miRNA-491 assays (all from Applied Biosystems, Thermo Fisher Scientific, Waltham, MA, USA), Universal PCR Master Mix, and the product from the reverse transcription reaction. Thermal cycler settings were 50 °C for two minutes, 95 °C for 10 minutes, then 40 cycles of 95 °C for 15 s and 60 °C for 60 s. The fold change was calculated based on the threshold cycle (Ct) value using the formula: Relative Quantity (RQ) = 2−ΔΔCT, using RNU44 as the internal control. The data analysis was performed using the statistical software GraphPad Prism version 9.0 (GraphPad Software, Inc., La Jolla, CA, USA). The normality of the distribution was confirmed by a Kolmogorov–Smirnov test. To identify statistical differences between groups, a one-way ANOVA test was applied, followed by Dunnett’s multiple comparison test. The difference was considered statistically significant when p ≤ 0.05. All experiments were conducted in triplicate and repeated at least two times.

PMC10002557

Yongfu La,Xiaoming Ma,Pengjia Bao,Min Chu,Ping Yan,Chunnian Liang,Xian Guo

Genome-Wide Landscape of mRNAs, lncRNAs, and circRNAs during Testicular Development of Yak

23-02-2023

testis,yak,development,non-coding RNA,reproduction

Testicular development is a tightly regulated process in mammals. Understanding the molecular mechanisms of yak testicular development will benefit the yak breeding industry. However, the roles of different RNAs, such as mRNA, lncRNA, and circRNA in the testicular development of yak, are still largely unclear. In this study, transcriptome analyses were performed on the expression profiles of mRNAs, lncRNAs, and circRNAs in testis tissues of Ashidan yak at different developmental stages, including 6-months-old (M6), 18-months-old (M18), and 30-months-old (M30). A total of 30, 23, and 277 common differentially expressed (DE) mRNAs, lncRNAs, and circRNAs were identified in M6, M18, and M30, respectively. Furthermore, functional enrichment analysis showed that the common DE mRNAs during the entire developmental process were mainly involved in gonadal mesoderm development, cell differentiation, and spermatogenesis processes. Additionally, co-expression network analysis identified the potential lncRNAs related to spermatogenesis, e.g., TCONS_00087394 and TCONS_00012202. Our study provides new information about changes in RNA expression during yak testicular development, which greatly improves our understanding of the molecular mechanisms regulating testicular development in yaks.

Genome-Wide Landscape of mRNAs, lncRNAs, and circRNAs during Testicular Development of Yak Testicular development is a tightly regulated process in mammals. Understanding the molecular mechanisms of yak testicular development will benefit the yak breeding industry. However, the roles of different RNAs, such as mRNA, lncRNA, and circRNA in the testicular development of yak, are still largely unclear. In this study, transcriptome analyses were performed on the expression profiles of mRNAs, lncRNAs, and circRNAs in testis tissues of Ashidan yak at different developmental stages, including 6-months-old (M6), 18-months-old (M18), and 30-months-old (M30). A total of 30, 23, and 277 common differentially expressed (DE) mRNAs, lncRNAs, and circRNAs were identified in M6, M18, and M30, respectively. Furthermore, functional enrichment analysis showed that the common DE mRNAs during the entire developmental process were mainly involved in gonadal mesoderm development, cell differentiation, and spermatogenesis processes. Additionally, co-expression network analysis identified the potential lncRNAs related to spermatogenesis, e.g., TCONS_00087394 and TCONS_00012202. Our study provides new information about changes in RNA expression during yak testicular development, which greatly improves our understanding of the molecular mechanisms regulating testicular development in yaks. The yak (B. grunniens) is an iconic symbol of the herbivore living on the “roof of the world”, found in and around the Himalayas and north at high altitudes of 2500 to 5500 m [1]. Yak production plays an important role in the Tibetan life of the people in high-altitude areas [2]. Survey reports reveal that the majority of yaks suffer from various reproductive problems like late maturity, long calving interval, poor oestrous expression, and repeat breeding [3]. Therefore, it is of practical significance to understand and master the reproductive and physiological characteristics of yaks. However, very little is known about molecular roles in the development and reproduction of Ashidan yaks, especially in yak testis. Advances in science and technology have shown that mRNAs, lncRNAs, and circRNAs are involved in the regulation of biological processes in animals. Although lncRNAs were considered to be by-products of RNA polymerase II transcription and will not be translated into proteins, many lncRNAs are involved in nuclear transport, transcription activation and interference, and chromosomal and genome modification, thereby driving more researchers to explore how lncRNAs affect human biology [4,5]. Many lncRNAs were identified from testis at different developmental stages, and they were predicted to play key roles in testis development and spermatogenesis in rats, mice, and cattle [6,7,8]. Currently, there were few studies on the effect of lncRNAs on testicular development and spermatogenesis. For example, LOC102176306 and miR-1197-3p regulate testosterone production and cell proliferation by regulating PPARGC1A expression in goat Leydig cells [9]. Although spermatogenesis and testicular development are partially regulated through the action of lncRNAs, the functions of most of them have not been determined. CircRNA biosynthesis is regulated by cis- and trans-factors, and its expression is tissue and cell-specific [10]. Some circRNAs are evolutionarily conserved, and they play important biological functions as miRNA inhibitors or by regulating protein function. CircRNAs play key roles in cell migration, proliferation, and differentiation [11]. To date, there were few studies on the expression patterns of lncRNAs and circRNAs in yak testis. To investigate the effect of ncRNAs on yak testis development, we performed transcriptome analysis on the expression profiles of mRNAs, lncRNAs, and circRNAs in yak testis tissues at different developmental stages. Our study will provide a good model for studying the mechanisms that regulate testicular development and spermatogenesis and provides newer insights regarding the regulation of male yak reproduction. A total of 274.25, 285.88, and 299.67 M clean reads were obtained in M6, M18, and M30, respectively. The Q30 values were 94.68–95.11% while the GC content ranged between 46.47% and 48.87%. About 94% of the clean reads were successfully mapped in the yak genome. Of the successfully mapped reads, 83.69% of the uniquely mapped reads were used for transcript construction (Table 1). A total of 20,953 mRNAs, 10,591 lncRNAs, and 16,185 circRNAs were identified from the 6, 18, and 30-month yak testis. Through genome alignment, these lncRNAs transcripts were classified into exon antisense, intron antisense, intergene downstream antisense, intergene upstream antisense, exonic sense, intron sense, intergene downstream sense, and intergene upstream sense, including 786, 838, 786, 1228, 705, 1463, 1283 and 1039, respectively (Figure 1A). However, sense overlapping circRNAs for 87.35%, intergene circRNAs for 8%, and the least located in introns (Figure 1B). Most lncRNAs with lengths greater than 2000 bp contain two exons (Figure 1C). Similarly, most circRNAs are longer than 2000 bp and contain less than 6 exons. (Figure 1D). To evaluate the differences in gene expression patterns during three developmental stages, we performed pairwise comparisons between the three developmental stages. For M18 vs. M6, 3037 mRNAs, 3076 lncRNAs, and 1787 circRNAs were upregulated, 2685 mRNAs, 602 lncRNAs, and 2245 circRNAs were downregulated (Figure 2A and Tables S1–S3). For M30 vs. M6, 3317 mRNAs, 3397 lncRNAs, and 1934 circRNAs were upregulated, 2870 mRNAs, 697 lncRNAs, and 2371 circRNAs were downregulated (Figure 2B and Tables S4–S6). For M30 vs. M18, 65 mRNAs, 27 lncRNAs, and 947 circRNAs were upregulated, 47 mRNAs, 28 lncRNAs, and 923 circRNAs were downregulated (Figure 2C and Tables S7–S9). There were 30 mRNAs, 23 lncRNAs, and 277 circRNAs were differentially expressed in all three developmental stages (Figure 2D–F and Table S10). Between M30 and M6, significantly up-regulated GO terms were mainly involved in the spermatogenesis, cell division, and spermatid development for DE mRNAs and DE lncRNAs, and DNA recombination, positive regulation of DNA-templated transcription, termination, cellular response to fibroblast growth factor stimulus and transposition, RNA-mediated for DE circRNAs (Figure 3A and Table S11). Furthermore, significantly down-regulated GO terms were mainly involved in the angiogenesis, regulation of cell shape, and positive regulation of cell migration for DE mRNAs, DE lncRNAs, and DE circRNAs (Figure 3A and Table S12). Between M18 and M6, significantly up-regulated GO terms were mainly involved in the spermatogenesis, spermatid development for DE mRNAs and DE lncRNAs, and DNA recombination, negative regulation of endocytosis and animal organ regeneration for DE circRNAs (Figure 3B and Table S13). Furthermore, significantly down-regulated GO terms were mainly involved in the angiogenesis, regulation of cell shape for DE mRNAs, and regulation of cell shape and maintenance of DNA methylation for DE lncRNAs, and cytokine production negative regulation of cell proliferation for DE circRNAs (Figure 3B and Table S14). Between M30 and M18, significantly up-regulated GO terms were mainly involved in the negative regulation of endopeptidase activity and G protein-coupled receptor signaling pathway for DE mRNAs, and spermatid development, cell differentiation, DNA recombination for DE lncRNAs, and negative regulation of androgen receptor signaling pathway and cellular response to fibroblast growth factor stimulus for DE circRNAs (Figure 3C and Table S15). Furthermore, significantly down-regulated GO terms were mainly involved in the cell cycle, positive regulation of cell proliferation for DE mRNAs, protein ubiquitination and DNA recombination for DE lncRNAs, and regulation of DNA repair and transnational initiation for DE circRNAs (Figure 3C and Table S16). Compared with 6 months of age, 30 mRNAs, 23 lncRNAs, and 277 circRNAs were identified as common DE genes during testicular development. GO analysis showed that the terms were mainly involved in the erythrocyte development, gonadal mesoderm development, cell differentiation and spermatogenesis for DE mRNAs (Figure 4A and Table S17), and negative regulation of signal transduction, cell differentiation, and positive regulation of meiotic cell cycle for DE lncRNAs (Figure 4B and Table S17), and sperm axoneme assembly, positive regulation of secretion, positive regulation of cell cycle and positive regulation of cell proliferation for DE circRNAs (Figure 4C and Table S17). The top 30 KEGG pathways are shown in Figure 5. KEGG analysis of the common DE mRNAs indicated that these genes were involved in the PI3K-Akt signaling pathway, parathyroid hormone synthesis, secretion and action, chemokine signaling pathway, and thyroid hormone synthesis (Figure 5A). KEGG analysis of the common DE lncRNAs indicated that target genes were involved in the AMPK signaling pathway, wnt signaling pathway, aldosterone-regulated sodium reabsorption, and PI3K-Akt signaling pathway (Figure 5B). Similarly, the common DE circRNAs parent genes were associated with the AMPK signaling pathway, MAPK signaling pathway, steroid hormone biosynthesis, FoxO signaling pathway, adipocytokine signaling pathway, and steroid biosynthesis (Figure 5C). To better understand the relationship between male reproduction and testicular development, 21 common DE mRNAs related to spermatogenesis and cell differentiation processes and 22 common DE lncRNAs targeting them were selected to construct the mRNA-lncRNA co-expression network. The results showed that the co-expression network comprised 255 connections and each lncRNA may be related to multiple mRNAs (Figure 6). Noticeably, DE lncRNAs TCONS_00087394, TCONS_00066011, and TCONS_00012202 were involved in the regulation of several DE mRNAs such as INVS, TLR7, PRSS51, F2R, and NBDY, indicating that these lncRNAs might play an important role in regulating spermatogenesis and cell differentiation processes. A total of twelve genes, including five mRNAs (CYP11A1, VDR, NR5A1, GATA4, and RGS1) related to testicular development, and Leydig cells growth and development and seven random lncRNAs (TCONS_00061526, TCONS_00094557, TCONS_00063420, TCONS_00048982, TCONS_00017364, TCONS_00037735, and TCONS_00036145) for qRT-PCR verification. The comparison found that the expression trends of RNA-Seq and qRT-PCR were similar, which confirmed the reliability of the sequencing data (Figure 7). The main function of the testis is to produce sperm and synthesize hormones. Testicular development and spermatogenesis are tightly regulated process that primarily involves the localization and differentiation of Leydig cells, Sertoli cells, and germ cells, requiring precise control of numerous genes and networks that act synergistically or antagonistically at the transcription and post-transcription levels [12]. Therefore, it is very important to determine the regulatory mechanism of testicular development and spermatogenesis for the study of yak breeding. In particular, the regulatory roles of lncRNAs and circRNAs on target genes have been extensively studied in organ development. In our study, we investigated the mRNAs, lncRNAs, and circRNAs expression profiles of 6, 18, and 30 months yak testis, which included juvenile, pubertal, and sexual maturation of testicular development [13]. Although previous studies have shown that lncRNAs and circRNAs are involved in testis development, the dynamic process of expression profiles of lncRNAs and circRNAs in yak testis is rare, and our study provides a theoretical basis for future new explorations. Normal testicular development and spermatogenesis are the basis for ensuring the reproductive ability of male animals, which are regulated by different genes or different expression levels of the same gene at different developmental stages [14]. LncRNAs and circRNAs have received increasing attention as the most popular ncRNAs, which participate in the regulation of different biological processes in different ways [15,16]. LncRNA has become a major regulatory factor in animal reproductive processes such as sex hormone responses, gonadogenesis, spermatogenesis, sex determination, and meiosis [17]. In this study, 6-, 18- and 30-month-olds correspond to infant, adolescent, and adult stages in male yaks presenting the dynamic process of testis development. A total of 5722, 6187, and 112 DE mRNAs, 3678, 4094, and 55 DE lncRNAs, and 4032, 4305, and 1870 DE circRNAs were identified between M18 and M6, between M30 and M6, and between M30 and M18, respectively. Only 30 DE mRNAs, 23 DE lncRNAs, and 277 DE circRNAs were identified from the testis of yak at the age of 6, 18, and 30 months old. At the same time, we discovered many novel lncRNAs and circRNAs during the research process, which may indicate that compared with other animals, the research on lncRNAs and circRNAs in yak testis tissue is still limited. Since the novel lncRNAs and circRNAs were obtained through the yak genome alignment and their characteristics, and the identification criteria were strict, they are of high value for future research on the molecular mechanisms of male yak testis development and spermatogenesis. Furthermore, lncRNAs and circRNAs have lowered lengths, exon numbers, and expression levels compared to mRNAs. The characteristics and differences of lncRNAs and circRNAs have also been found in other animals, which may suggest that there is also some conservation in the regulation of lncRNAs and circRNAs in mammals [18]. We selected five mRNAs and seven lncRNAs related to testis development, Leydig cell growth, and development for qRT-PCR to validate the accuracy of RNA-seq data. Through analysis, it was found that qRT-PCR and RNA-Seq data have similar expression trends, indicating that our RNA-Seq data is of high quality and sequencing data can be used for in-depth analysis. Generally speaking, the main function of the testis is to produce sperm and androgen, which depends on the normal development of both testicular somatic cells and germ cells. They first guide fetal germ cell differentiation toward spermatogenic destiny and then take care of the full service to spermatogenic cells during spermatogenesis. The number of Sertoli cells sets the limits of sperm production. Leydig cells secrete androgens that determine masculine development. Testis development does not depend on germ cells, testicular somatic cells also develop in the absence of germ cells, but spermatogenic cell development is dependent on somatic cells [19]. In this study, we performed GO enrichment analyses that can help to elucidate the functions and pathways involved in candidate target genes. Our results showed that differentially expressed RNAs in 6-, 18- and 30-month-old testis were mostly enriched in GO terms related to spermatogenesis, cell proliferation, positive regulation of cell proliferation, and cell differentiation. At the same time, more differentially expressed RNAs were observed in the M30 vs. M6 and M18 vs. M6 comparison groups compared with M30 vs. M18, suggesting that more RNAs were involved in cell differentiation and proliferation processes in the early stages of testis development while indicating that Cell proliferation and differentiation in the testis of early animals is more vigorous. However, further studies are needed to confirm this speculation. Based on the above studies, we suggested that the transcription differences were caused by changes in cell types in the yak testis during testicular development and that lncRNAs might play important regulatory roles in male yak sexual maturation. Previous studies have demonstrated that certain lncRNAs play important regulatory roles during testicular development and spermatogenesis in male animals [20,21]. In this study, the differentially expressed lncRNAs TCONS_00061526, TCONS_00094557, TCONS_00063420, TCONS_00048982, TCONS_00017364, TCONS_00037735 and TCONS_00036145 targeted to regulate the expression levels of CYP11A1, VDR, NR5A1, GATA4, and RGS1 genes, and GO analysis found that these genes were enriched in biological processes such as gonadal mesoderm development, cell differentiation, spermatogenesis, cell cycle, and DNA recombination. The CYP11A1 gene encodes the CYP11A1 enzyme, which is located in the inner mitochondrial membrane and catalyzes the conversion of cholesterol to pregnenolone in the first and rate-limiting step of the steroid hormone synthesis [22]. Studies of the CYP11A1 gene suggested that this gene plays an important role in gene regulation, testosterone secretion, and male reproductive organ development [23]. Similarly, in this study, the CYP11A1 gene was significantly upregulated in the 30-month-old group compared with the 6-month-old, suggesting that the CYP11A1 gene might regulate the yak testis development by promoting steroid hormone synthesis. GATA4 is expressed in Leydig and Sertoli cells, is required for mouse fetal testis development, and is a key transcription regulator of Sertoli cell function in adult mice [24]. Studies of chimeric mice derived from Gata4-/- embryonic stem cells show that GATA4 plays an integral role in the development of fetal Leydig cells [25]. In this study, GATA4 has the highest expression in the testis of sexually mature yak, and we speculated that this gene might promote spermatogenesis by regulating the functions of the Leydig and Sertoli cells, thereby achieving precocious puberty. Furthermore, the constructed lncRNA-mRNA co-expression network shows that lncRNAs (such as TCONS_00087394, TCONS_00066011, and TCONS_00012202) were regulated by multiple lncRNAs, suggesting that these lncRNAs and their target mRNAs might also play an important role in yak testicular development. However, the underlying mechanisms of these interacting lncRNA and mRNA regulatory activities need to be further investigated. All yaks were handled in strict accordance with the Animal Ethics Procedures and Guidelines of the People’s Republic of China. The present study was approved by the Animal Administration and Ethics Committee of the Lanzhou Institute of Husbandry and Pharmaceutical Sciences of the Chinese Academy of Agricultural Sciences (Permit No. 2019-002). All animals used in this study were from the Datong Breeding Farm of Qinghai province. The nine selected Ashidan yak were healthy and fed in similar conditions. Furthermore, the animals were separated into three groups (6 months, M6; 18 months, M18; and 30 months, M30). Every group contained three male yaks. The nine male yaks were slaughtered and tissues from the left testis were collected. All samples were immediately stored at −80 °C for total RNA extraction. Total RNA was extracted from testis tissue using TRIzol (Invitrogen, Carlsbad, CA, USA). Genomics DNA was removed using DNase I. A total of 9 cDNA libraries were constructed in this study using the NEB Next Ultra Directional RNA LibraryPrep Kit for Illumina (NEB, Ipswich, MA, USA). Sequencing libraries were then sequenced on an Illumina HiseqTM 2500 (Illumina Corp., San Diego, CA, USA) instrument to generate 150-nt paired-end reads. Libraries were constructed and sequenced using the Illumina platform by OE Biotech Co. (Shanghai, China). Then, the raw data was checked using the FASTQC (version 0.11.9) tools. Trimmomatic (version 0.36) software was first used for removing adapters, and then low-quality bases and N-bases or low-quality reads were filtered out. The quality of the trimmed reads was rechecked with the FASTQC tool, after which the clean reads were aligned with the yak genome using HISAT2 (version 2.0.5) [26]. The genome sequence of yak (BosGru_v3.0) and the annotation file was downloaded from Ensemble. Transcripts were assembled by Stringtie (version 1.3.4) [27]. Raw data for RNA-seq has been documented in the SRA public database (Accession number: SRP367128). Use the following procedure to identify lncRNAs: (1) Transcripts annotated as “i (a transfrag falling entirely within a reference intron)”, “u (unknown, intergenic transcript)”, “x (exonic overlap with reference on the opposite strand)” and “o (generic exonic overlap with a reference transcript)” were retained by the cuffcompare software [28]. (2) Transcripts longer than 200 bp and containing more than 2 exons were retained. (3) Finally, four approaches of Coding-Non-Coding-Index (CNCI, score < 0), Coding Potential Calculator (CPC, score < 0), Pfam (E value < 0.001), and k-mer scheme (PLEK, score < 0) were used to predict coding potential, and transcripts without coding potential were candidate lncRNAs [15]. Then, CIRI software was used to scan for PCC signals (paired chiastic clipping signals), and circRNA sequences were predicted based on junction reads and GT-AG cleavage signals [29]. Briefly, paired chiastic clipping, paired-end mapping, and GT-AG splicing signals were found by scanning the obtained slicing alignments. Next, the alignment files were scanned again using a dynamic programming algorithm to detect additional junction reads and eliminate false-positive circRNA candidates. Final circRNAs were obtained by retaining sequences with ≥2 junction reads. The fragments per kilobase of transcript per million read mapped (FPKM) and spliced reads per billion mappings (SRPBM) values were used to detect the expression levels of mRNA, lncRNA, and circRNA, respectively [30]. Differentially expressed mRNAs, lncRNAs, and circRNAs were detected using the DESeq2 software package [31], and differentially expressed genes were defined as |log2 (fold change)| ≥ 1 and FDR < 0.001 between any comparison groups. Meanwhile, genes differentially expressed in three comparisons (M6 vs. M18, M6 vs. M30, and M18 vs. M30) were defined as common DE genes. The cis-target mRNAs were screened by the genomics location 50 Kb upstream and downstream of the lncRNA, trans-target mRNAs were identified by the Pearson correlation coefficient of the lncRNA-RNA pairs being ≥ 0.9, and then the cis- and trans-target mRNAs were subjected to GO analysis. The parental genes of DE circRNAs were mapped to GO terms using the Gene Ontology database (http://www.geneontology.org) (accessed on 14 May 2022), followed by enrichment analysis with the Omicshare (version 3.0) tools. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was also performed to annotate the signaling pathways associated with these common DE genes [32]. Terms with p-values less than 0.05 were identified as significant or enriched terms. To study the functions of key lncRNAs, Cytoscape software (version 3.1.1) was used to construct the co-expression network of common DE lncRNAs and common DE mRNAs [33]. We used qRT-PCR to verify the gene expression levels. The PCR reaction was performed on the LightCycler 480 II (Roche, Basel, Switzerland) using the SYBR Green Real-time PCR Master Mix (TOYOBOCO, Ltd., Osaka, Japan) with different cycling conditions as 95 °C for 10 min, followed by 45 cycles for 15 s at 95 °C, annealing for 60 s at 55 to 60 °C, extension for 30 s at 72 °C, final extension for 5 min at 72 °C, and storage at 4 °C. GAPDH was used as an internal reference to normalize target gene expression. All experiments were performed in triplicate. The primers (Table S18) were produced by Sheng gong Biotech Co., Ltd. (Shanghai, China). For gene expression levels, each experiment was repeated in at least 3 replicates, and the threshold cycles were calculated using the 2-ΔΔCt method [34]. All data were expressed as “means ± SD”. A p-value < 0.05 was established as a significant difference. In conclusion, a genome-wide view of the mRNAs, lncRNAs, and circRNAs expression profiles during yak testis development was explored in our study. In addition, we identified several DE genes that may contribute to testicular development and spermatogenesis. Our study provides a comprehensive basis for the expression levels of mRNAs, lncRNAs, and circRNAs during the yak testicular development, thus providing new clues for our understanding of the molecular regulatory mechanism of yak testis development.

PMC10002558

Antonio Franco,Francesc Moreso,Eulàlia Solà-Porta,Isabel Beneyto,Núria Esforzado,Francisco Gonzalez-Roncero,Asunción Sancho,Edoardo Melilli,Juan Carlos Ruiz,Cristina Galeano

Outcome of Kidney Transplants from Viremic and Non-Viremic Hepatitis C Virus Positive Donors into Negative Recipients: Results of the Spanish Registry

23-02-2023

kidney transplantation,hepatitis C virus,viremic donor,graft outcome,hepatocellular carcinoma

Historically, donor infection with hepatitis-C virus (HCV) has been a barrier to kidney transplantation. However, in recent years, it has been reported that HCV positive kidney donors transplanted into HCV negative recipients offer acceptable mid-term results. However, acceptance of HCV donors, especially viremic, has not broadened in the clinical practice. This is an observational, multicenter, retrospective study including kidney transplants from HCV positive donors into negative recipients reported to the Spanish group from 2013 to 2021. Recipients from viremic donors received peri-transplant treatment with direct antiviral agents (DAA) for 8–12 weeks. We included 75 recipients from 44 HCV non-viremic donors and 41 from 25 HCV viremic donors. Primary non function, delayed graft function, acute rejection rate, renal function at the end of follow up, and patient and graft survival were not different between groups. Viral replication was not detected in recipients from non-viremic donors. Recipient treatment with DAA started pre-transplant avoids (n = 21) or attenuates (n = 5) viral replication but leads to non-different outcomes to post-transplant treatment with DAA (n = 15). HCV seroconversion was more frequent in recipients from viremic donors (73% vs. 16%, p < 0.001). One recipient of a viremic donor died due to hepatocellular carcinoma at 38 months. Donor HCV viremia seems not to be a risk factor for kidney transplant recipients receiving peri-transplant DAA, but continuous surveillance should be advised.

Outcome of Kidney Transplants from Viremic and Non-Viremic Hepatitis C Virus Positive Donors into Negative Recipients: Results of the Spanish Registry Historically, donor infection with hepatitis-C virus (HCV) has been a barrier to kidney transplantation. However, in recent years, it has been reported that HCV positive kidney donors transplanted into HCV negative recipients offer acceptable mid-term results. However, acceptance of HCV donors, especially viremic, has not broadened in the clinical practice. This is an observational, multicenter, retrospective study including kidney transplants from HCV positive donors into negative recipients reported to the Spanish group from 2013 to 2021. Recipients from viremic donors received peri-transplant treatment with direct antiviral agents (DAA) for 8–12 weeks. We included 75 recipients from 44 HCV non-viremic donors and 41 from 25 HCV viremic donors. Primary non function, delayed graft function, acute rejection rate, renal function at the end of follow up, and patient and graft survival were not different between groups. Viral replication was not detected in recipients from non-viremic donors. Recipient treatment with DAA started pre-transplant avoids (n = 21) or attenuates (n = 5) viral replication but leads to non-different outcomes to post-transplant treatment with DAA (n = 15). HCV seroconversion was more frequent in recipients from viremic donors (73% vs. 16%, p < 0.001). One recipient of a viremic donor died due to hepatocellular carcinoma at 38 months. Donor HCV viremia seems not to be a risk factor for kidney transplant recipients receiving peri-transplant DAA, but continuous surveillance should be advised. Hepatitis C virus (HCV) is an RNA virus from the Flaviridae family with seven different genotypes [1]. Parenteral transmission of HCV has been well-documented and renal transplants recipients from HCV positive donors can acquire the infection and develop acute and chronic hepatitis [2]. For this reason, until recently, renal transplantation from HCV positive donors has not been done in non-infected recipients. The report of the American Society of Transplantation consensus conference on the use of HCV positive donors in solid organ transplantation states that, in general, donors with positive HCV antibodies without viremia do not transmit the infection; thus, historical data of HCV “positive” donors must be viewed as limited since it does not specifically differentiate the presence or absence of viremia. In this document, it is recommended to abandon the term HCV “positive” donor and to include the term HCV viremic donor which requires a nuclear acid testing (NAT) result and allows us to differentiate the presence of viremia from its absence, improving the infection transmission risk. Thus, the term HCV viremic donor should be adopted to replace the term HCV positive donor. Additionally, despite this document’s support of the proposition that the policy of kidney transplantation from HCV viremic donors into non-viremic recipients should be conducted under Institution Review Board (IRB)-approved protocols with a rigorous, multi-step informed consent process [3], transplantation from HCV viremic donors into non-viremic recipients was adopted as the standard of care in many centers from US Importantly, different prospective studies have documented the safety and efficacy of the direct-acting antivirals (DAA) drugs to transplant kidneys from HCV viremic donors into negative recipients (reviewed in [4,5]). Finally, the pangenotype efficacy of the combination glecaprevir/pibrentasvir allows the initiation of treatment before HCV genotype identification [6,7]. In 2019, we report the first European experience with HCV viremic/non-viremic donors into HCV negative recipients conducted at three Spanish hospitals [8]. We describe four recipients from HCV viremic donors receiving an 8-week course of the combination glecaprevir/pibrentasvir started pre-operatively, and seven patients receiving kidneys from HCV non-viremic donors which were managed conservatively. Our data suggest that renal transplantation from HCV positive donors into HCV negative recipients is safe when only recipients of organs from HCV viremic donors are treated [9]. This early experience inspired the Spanish consensus document coordinated by the National Transplant Organization (ONT) [10]. In this document, it is realized that between 2011 and 2017, from 246 HCV positive potential donors, only 31% were finally effective donors, and it was estimated that 5–10 HCV viremic donors can be obtained yearly. A specific patient informed consent was provided for this type of donor and universal treatment with DAA was covered by our national health system. For renal transplants recipients, it was suggested to employ DAA with a pangenotype efficacy from the pre-operative period. However, currently, only 24 out of 43 adult renal transplant units in Spain accept kidneys from HCV non-viremic donors and 9 accept HCV viremic donors. The aim of the present study is to compare clinical outcomes of HCV negative renal transplant recipients receiving organs obtained from HCV viremic/non-viremic donors, reported to the Spanish group. This is a retrospective, observational study conducted in 24 renal transplant units from Spain. All donors with a positive HCV serology between 2013 and 2021 were identified and NAT testing was done before organ retrieval (XPERT HCV/HIV load test, Cepheid Inc., Sunnyvale, CA, USA). Donors were identified as HCV viremic or non-viremic and this information was available for the clinicians before transplantation. Donors who were HIV positive, active intravenous drug-abusers, and institutionalized persons during the last year were discarded. Recipients from HCV viremic donors signed a written informed consent form defined by the Spanish consensus document [10] and were treated with a combination of glecaprevir 300 mg/day and pribentasvir 120 mg/day for 8–12 weeks regardless of the presence of recipient viremia. Treatment can be started pre-operatively to complete an 8-week course or until 10 days after surgery once HCV replication was confirmed to complete a 12-week course. HCV viral load was determined approximately at +7, +14, +21, +30, +45, +60, +90, +120, and +180 days. Sustained virologic response (SVR) was defined as a negative NAT result 12 weeks after ending treatment with DAA. For HCV non-viremic donors, NAT monitoring was done at the same intervals and treatment was started in the case of a positive result. Recipient liver enzymes (AST and ALT) were recorded as per local practice at each visit. Induction and maintenance immunosuppression was guided according to local practices. The following variables were recorded: donor and recipient demographics (age and sex); recipient’s weight; donor/recipient blood group; brain death donor or donor after circulatory death; first transplant vs. re-transplant; last calculated panel reactive antibodies (cPRA) by Luminex assay; ABDR HLA matching; and cold ischemia time. The present study was performed in accordance with the Declaration of Helsinki and is consistent with the Principles of the Declaration of Istanbul on Organ Trafficking and Transplant Tourism. The following primary outcome variables were recorded: patient and graft survival at the end of follow up as well as HCV viral load 1 week after transplantation and 12 weeks after completing DAA treatment (sustained virologic response: SVR). The following secondary outcome variables were recorded: primary non-function including early vascular thrombosis; delayed graft function defined as dialysis requirement during the first week after transplant once vascular and/or urinary tract complications were ruled out; biopsy-proven acute rejection; estimated glomerular filtration rate according to CKD-EPI formula at the end of follow up. The rate of HCV seropositive recipients at the end follow-up and the evolution of liver enzymes (AST and ALT) after transplant were also analyzed. Finally, adverse events related with DAA and treatment interruptions of DAA were also recorded. Variables were described as frequencies, median and interquartile range, or mean and standard deviation for categorical, non-normally distributed continuous variables and normally distributed continuous variables, respectively. To compare data between groups, Fisher’s exact test, Mann–Whitney U test, Kruskal–Wallis test and Student’s t-test were employed according to the variable distribution. Graft and patient survival were analyzed by means of Kaplan–Meier curves and comparison between groups was done by the log-rank test. All analysis were two-tail and a p-value < 0.05 was considered significant. During the study period, 69 HCV positive donors were used to perform 116 kidney transplants into seronegative recipients in 24 Spanish renal transplant units. A total of 75 kidney transplants were done from 44 non-viremic donors and 41 transplants from 25 viremic donors. From non-viremic donors, 5 kidneys were employed to transplant seropositive recipients while 8 were discarded by surgeons (inadequate perfusion, vascular damage, or macroscopical aspect). From viremic donors, three kidneys were employed to transplant seropositive recipients and six kidneys were discarded by surgeons. This means that the efficacy rate in this pool of donors was 91% for non-viremic donors (80 out of 88) and 88% for viremic donors (44 out 50). Donor, recipient, and transplant-related variables in kidney transplants from non-viremic and viremic donors were not different (Table 1). In transplant recipients from viremic donors, DAA treatment was started pre-transplant (n = 26) or during the initial 10 days after confirming viral transmission (n = 15), depending on the renal transplant unit preferences. Patient and graft survival were not different between groups (Figure 1 and Figure 2). In the non-viremic group, 5 patients out of 75 died because of sudden death (2 recipients), acute pancreatitis, sepsis, and COVID-19 infection at 1, 2, 2, 12, and 24 months, respectively. In the viremic group, 4 patients out of 41 died because of sepsis, coronary disease, COVID-19 infection, and hepatocellular carcinoma (HCC) at 1, 12, 14, and 38 months, respectively. Two renal transplants from non-viremic donors failed due to early vascular thrombosis and there was one never functioning kidney. Five renal transplants from non-viremic donors experienced late failure due to recurrent primary disease (n = 1), non-treatment compliance (n = 1), chronic rejection (n = 2), and chronic allograft dysfunction (n = 1). Three renal transplants from viremic donors failed due to early vascular thrombosis (n = 1), chronic rejection (n = 1), and chronic allograft dysfunction (n = 1). Patient and graft survival were also not different in recipients from a viremic donor starting DAA treatment before or after surgery (Figure 2). Noticeably, there was a 72-year-old male recipient dying with a functioning graft at 38 months due to HCC. He had adult polycystic kidney disease and received a first deceased donor transplant in July 2018 from an expanded criteria donor who experienced early vascular thrombosis. Later, he received a second graft from a brain-dead 50 years-old male viremic donor (5.6 log). The recipient received an 8-week course of glecaprevir/pribentasvir started before transplant. Immunosuppression was based on the combination of anti-thymocyte globulin, tacrolimus, sirolimus, and corticosteroids. Clinical course after transplantation was uneventful and he reached a nadir serum creatinine of 1.3 mg/dL. HCV viral load was negative at all time points and a mild increase of liver enzymes was detected at 2 weeks (AST 56 and ALT 190 UI/L) which normalized thereafter. At 30 months, he was admitted to another hospital due to progressive weight loss, diarrhea, and acute renal failure. Liver ultrasound showed multiple nodular hypoechogenic lesions (8 cm involving segments V-VI, 3 cm in segment V, and 2 nodules of 1 cm in segment III). A liver biopsy of the largest lesion yielded the diagnostic of HCC. Immunohistochemistry was positive for cytokeratin 8/18 and negative for cytokeratin 7 and alpha-fetoprotein. The viral load in the liver biopsy was negative. Hepatitis B virus surface antigen (HBsAg) and antibody to core antigen were negative, and alpha fetoprotein serum levels were ×10 times the upper normal limit. Despite a short treatment with sorafenib, the clinical condition of the patient grew progressively impaired, and he died few months after diagnosis. Recipients from non-viremic donors showed a negative HCV viral load at all time points since the beginning of transplant. Recipients from a viremic donor starting DAA treatment pre-transplant did no showed viral replication at any time (n = 21) or a very low replication (<2 log) at 7 days (n = 5). Recipients from a viremic donor starting DAA after transplantation (n = 15) showed transmission in all cases with a viral load at approximately 7 days of 5.1 ± 1.0 log. From 14 days after transplant to the end of follow up, all recipients from viremic donors, either starting DAA treatment before or after transplant, had a negative HCV viral load. Thus, the SVR for treated patients was 100%. Primary non-function, delayed graft function, biopsy-proven acute rejection, and eGFR at the end of follow up were not different between groups (Table 2). The rate of HCV seroconversion was 100% (14 cases) in recipients from viremic donors starting DAA treatment after transplantation, 56.2% (13 out of 23) in recipients from viremic donors starting DAA treatment before transplantation, and 16.4% (12 out of 73) in patients receiving grafts from HCV non-viremic donors (p-value < 0.001). There were 6 cases in whom serology for HCV was not done. Evolution of liver enzymes (AST, ALT) according to the presence of donor viremia and DAA treatment is shown in Figure 3. ALT tended to be higher at 30 days in patients receiving a graft from a viremic donor starting treatment after transplantation, but this difference did not reach statistical significance (p = 0.069). Seroconversion rate and evolution of liver enzymes were not different according to induction therapy at the time of transplant. All patients completed treatment with DAA without reported adverse events and without treatment interruptions. We report a large experience of HCV seronegative recipients receiving kidneys from HCV positive donors, either non-viremic or viremic. In recipients from non-viremic donors, active surveillance of viral transmission was pursued; in recipients from viremic donors, peri-transplant treatment with DAA was started. Both cohorts were not different in their baseline characteristics and patients and graft outcomes were not significantly different. Noticeably, we report the first case of an HCV seronegative renal transplant recipient receiving a graft from an HCV viremic donor who develops an HCC after transplantation. Despite a link between transplantation with a viremic donor and development of HCC, it cannot be stablished in our case; this observation may recommend continuous liver surveillance of these patients. Undoubtedly, new reports of large series using HCV positive viremic donors will contribute to establish whether this association was pathogenic. Since the initial reports in the US and Europe, HCV positive donors have been progressively employed as kidney donors in different countries [11,12]. In 2019, we report the first series in Europe using HCV non-viremic and viremic donors, and we conclude that renal transplantation from HCV positive donors into HCV negative recipients is safe when only the recipients of organs from viremic donors are treated [8]. This preliminary experience led the Spanish National Organization (ONT) to establish a protocol to guide patient-informed consent and treatment with DAA for recipients of viremic donors [10]. Despite this protocol, there are a significant number of renal transplant units in Spain not accepting HCV positive donors, especially viremic donors. Thus, this registry will contribute to defining the outcomes of these transplants and to encourage other renal transplant units to employ them. Both cohorts, viremic and non-viremic, are homogenous and like the populations of donors and recipients in Spain. We decided not to include a matched population of HCV seronegative donors and recipients, since national registries are available for comparisons. In fact, patient and graft survival were not significantly different to the last report of the Catalan registry of renal transplants [13]. Patient survival at 1 and 5 years in this registry is 94.9% and 85.1%, while in our cohorts it was 94.6% and 89.5%. Similarly, graft survival including patient’s death in the Catalan registry is 87.1% and 68% at 1 and 5 years, respectively, while in our cohorts these figures were 90.3% and 74.3%. Other outcomes included in our study such as delayed graft function, biopsy-proven acute rejection rates or renal function, were also non-different. One of the most intriguing observations of our study is that one recipient develops a fatal HCC early (30 months) during follow up. This is the first case reported in the literature of this scenario. In a recent review of HCV-viremic donors into HCV-negative recipients [4], such a serious event was not reported in over 300 kidney transplants. Until now, the pathogenesis of HCV-induced HCC and the HCC risk after DAA cure were incompletely understood. HCV is an RNA virus with little potential for integrating its genetic material into the host genome and it has been proposed that HCV contributes to hepatocarcinogenesis through direct and indirect ways. HCV-mediated liver disease and carcinogenesis are considered multistep processes that include chronic infection-driven hepatic inflammation and progressive liver fibrogenesis with the formation of neoplastic clones that arise and progress in the carcinogenic tissue microenvironment. A gene expression signature in the liver tissue of HCV-infected patients has been associated with HCC risk and mortality, suggesting that virus-induced transcriptional reprogramming in the liver could play a functional role in hepatocarcinogenesis [14]. Moreover, epigenetic changes associated with liver cancer persist after sustained virologic response [15]. From the clinical point of view, it has been widely described that although the viral cure decreases the overall HCC risk in HCV-infected patients, it does not eliminate virus-induced HCC risk, especially in patients with advanced fibrosis [16,17]. In our case, there was no known preexisting liver disease and the patient only presented mild transient elevation of liver enzymes after transplantation, while viral replication in blood was not detected at any time since DAA treatment was started pre-transplant. Unfortunately, once the diagnostic was reached, the disease was too advanced, with multiple large nodules in the liver. Since our patient received chronic immunosuppression, we cannot discard that this fact increases the risk of this serious complication. Despite the patient receiving induction treatment with thymoglobulin, in our study, there were other patients receiving a kidney from a viremic donor treated with thymoglobulin who did not develop liver complications. Similarly, in the studies conducted in the US a significant proportion of patients were also treated with thymoglobulin [18]. Thus, whether surveillance by liver enzymes, serum biomarkers (alpha-fetoprotein), and ultrasound are advisable for these patients should be decided by their treating physicians. Our study confirms some data already reported. Kidney transplant from HCV non-viremic donors is a safe procedure and does not require treatment since viral transmission is exceptional [19,20]. It is important to emphasize that in our cohort, we exclude active intravenous drug-abusers and institutionalized persons to reduce the risk of transmission of other viruses such as HIV; thus, the probability of being in the window period after HCV transmission is further reduced [21,22]. Kidney transplantation from HCV viremic donors is associated with a very high rate of transmission. Despite the Spanish consent document suggesting starting treatment with DAA before transplant, some centers decided to start treatment during the first week once viral replication was detected. Pre-transplant DAA treatment prevents early viral replication in a lot of cases (21 out of 26) or reduced its intensity to a very low viral load (5 out of 26). Conversely, kidney transplantation from HCV viremic donors into seronegative recipients led to universal transmission without DAA treatment as it has been previously described [23,24,25,26]. In any case, early treatment with DAA is associated with rapid clearance of viral replication and, in our study, it was associated with non-different outcomes to those which have been previously described [24,27]. We observed a high rate of seroconversion in patients experiencing HCV viremia (100%), but also in recipients from HCV viremic donors who never had viremia (56%) and even in recipients from non-viremic donors (16%). This observation has been previously described and the rate of seroconversion of recipients from HCV non-viremic donors under close monitoring may be as high as 44% [20]. In this study the authors concluded that the reason(s) for seroconversion is(/are) unclear, but this phenomenon does not appear to indicate HCV transmission. Later, Porrett et al. conducted a nice study in HCV naïve kidney and heart transplant recipients from HCV viremic donors to address this question. They concluded that the IgG isotype of this antibody (not IgM) and the kinetics of its appearance and durability suggest that the anti-HCV antibody is donor derived and likely produced by a cellular source. They suggest that transfer of donor humoral immunity to a recipient may be much more common than previously appreciated [28]. The evolution of liver enzymes, AST and ALT, was not significantly different among groups despite the trend to higher ALT levels at 30 days in transplants receiving a graft from a viremic donor and starting DAA treatment after transplantation. In kidney transplants, increase of AST/ALT is a frequent finding, affecting up to 24% of patients depending on the series. The main causes during the first months after transplant are attributed to hemodynamic alterations during the surgical procedure, viral infections, previous liver disease, and pharmacological hepatotoxicity (immunosuppressants including thymoglobulin, antibiotics, and antiviral agents). Additionally, the initial kidney injury and the kidney–liver crosstalk may cause hepatic damage and might be responsible for this increase. Recently, it has been described that the increase of AST/ALT is a frequent and transient event related to the kidney donor type, being more frequent in recipients from uncontrolled DCD that normalizes one month after transplant [29]. Thus, we cannot discard that the mild elevation of ALT in transplants receiving a graft from a viremic donor and starting DAA treatment after transplantation reflects hepatocellular damage related to HCV infection. However, we did not observe cases of fibrosing cholestatic hepatitis in recipients from viremic donors who started treatment either before or after transplantation. Noticeably, in our country, the scenario is different from the US It should be noted that HCV-viremic donors are younger in the US than in our cohort (32 vs. 55 years). Additionally, in the US cohorts, HCV-viremic donors were persons who injected drugs and died due to drug-overdose [30], while these kinds of donors were rejected in our country and in Germany [8,12]. For these reasons, in Europe, the number of non-viremic HCV donors is higher than the number of viremic ones. This different criterion to accept viremic donors is related to the fact that the rate of fatal drug-related overdose is much lower in Europe than in the US (20.6 vs. 2.3 per 100.000 in 2017) [31]. Our study has some limitations related to a multicenter registry, the heterogenicity in the immunosuppression and, especially, in the timing (before or after transplantation) of the initiation of DAA treatment. Unfortunately, histological assessment of pre-implantation kidney biopsies to evaluate organ quality from HCV positive donors was not done, despite new tools to characterize it impending [32]. Additionally, we cannot rule out a selection bias in our cohort, since a contemporary control group receiving a kidney transplantation from HCV negative donors was not included. However, it allows us to confirm non-different outcomes in HCV seronegative renal transplant recipients receiving grafts from viremic or non-viremic donors. Additionally, all patients were treated with the same combination of DAA (glecaprevir/priventasbir), which are pangenotypic and can be safely employed in patients with chronic renal failure [7]. In summary, we reported the analysis of HCV naïve recipients of a renal graft from non-viremic and viremic HCV donors and we described that patient and graft outcomes were not different between groups. Recipients from viremic donors treated with DAA before or early after transplantation also showed non-different outcomes. Despite one recipient of a viremic donor developing an early HCC, we cannot establish a link between both events.

PMC10002560

Yu-Huei Huang,Lun-Ching Chang,Ya-Ching Chang,Wen-Hung Chung,Shun-Fa Yang,Shih-Chi Su

Compositional Alteration of Gut Microbiota in Psoriasis Treated with IL-23 and IL-17 Inhibitors

26-02-2023

psoriasis,interleukin-23 inhibitor,guselkumab,interleukin-17 inhibitor,secukinumab,ixekizumab,gut microbiota,metabolic pathway

Alterations in the gut microbiota composition and their associated metabolic dysfunction exist in psoriasis. However, the impact of biologics on shaping gut microbiota is not well known. This study aimed to determine the association of gut microorganisms and microbiome-encoded metabolic pathways with the treatment in patients with psoriasis. A total of 48 patients with psoriasis, including 30 cases who received an IL-23 inhibitor (guselkumab) and 18 cases who received an IL-17 inhibitor (secukinumab or ixekizumab) were recruited. Longitudinal profiles of the gut microbiome were conducted by using 16S rRNA gene sequencing. The gut microbial compositions dynamically changed in psoriatic patients during a 24-week treatment. The relative abundance of individual taxa altered differently between patients receiving the IL-23 inhibitor and those receiving the IL-17 inhibitor. Functional prediction of the gut microbiome revealed microbial genes related to metabolism involving the biosynthesis of antibiotics and amino acids were differentially enriched between responders and non-responders receiving IL-17 inhibitors, as the abundance of the taurine and hypotaurine pathway was found to be augmented in responders treated with the IL-23 inhibitor. Our analyses showed a longitudinal shift in the gut microbiota in psoriatic patients after treatment. These taxonomic signatures and functional alterations of the gut microbiome could serve as potential biomarkers for the response to biologics treatment in psoriasis.

Compositional Alteration of Gut Microbiota in Psoriasis Treated with IL-23 and IL-17 Inhibitors Alterations in the gut microbiota composition and their associated metabolic dysfunction exist in psoriasis. However, the impact of biologics on shaping gut microbiota is not well known. This study aimed to determine the association of gut microorganisms and microbiome-encoded metabolic pathways with the treatment in patients with psoriasis. A total of 48 patients with psoriasis, including 30 cases who received an IL-23 inhibitor (guselkumab) and 18 cases who received an IL-17 inhibitor (secukinumab or ixekizumab) were recruited. Longitudinal profiles of the gut microbiome were conducted by using 16S rRNA gene sequencing. The gut microbial compositions dynamically changed in psoriatic patients during a 24-week treatment. The relative abundance of individual taxa altered differently between patients receiving the IL-23 inhibitor and those receiving the IL-17 inhibitor. Functional prediction of the gut microbiome revealed microbial genes related to metabolism involving the biosynthesis of antibiotics and amino acids were differentially enriched between responders and non-responders receiving IL-17 inhibitors, as the abundance of the taurine and hypotaurine pathway was found to be augmented in responders treated with the IL-23 inhibitor. Our analyses showed a longitudinal shift in the gut microbiota in psoriatic patients after treatment. These taxonomic signatures and functional alterations of the gut microbiome could serve as potential biomarkers for the response to biologics treatment in psoriasis. Psoriasis is an inflammatory skin disease that is associated with many other medical conditions, and affects adults and children worldwide [1]. Overall prevalence ranges from 0.1% in east Asia to 1.5% in western Europe, and is highest in high-income countries [2,3]. Most patients with psoriasis have some detriment to their quality of life attributable to the disease, and many feel a substantial, negative effect on their psychosocial wellbeing. It has been regarded that psoriasis involves the interplay between predisposing genetic and environmental (e.g., infection and antibiotics treatment) factors [1,4,5,6]. Studies have shown that skin and the gut microbiome play a role in modulating the development of chronic plaque psoriasis [7]. Recent evidence revealed a combined increase in Corynebacterium, Propionibacterium, Staphylococcus, and Streptococcus in psoriatic plaque sites [7,8]. Gut microbiota is known to play a critical role in the regulation of metabolism, the immune system, and intestine permeability [9]. A disturbed intestinal microbiome was shown to be involved in a number of autoimmune diseases including type 1 diabetes, rheumatoid arthritis, multiple sclerosis, celiac disease, and inflammatory bowel disease (IBD) [10,11]. In psoriasis, similar evidence demonstrated gut dysbiosis with lower diversity and altered relative abundance for certain bacteria [9,12]. Several studies have found the relative abundance of Bacteroidetes was lower and that of Firmicutes was higher in patients with psoriasis compared to healthy controls [12,13,14]. However, an inconsistent result reported by Huang et al., revealed an increased abundance of Bacteroidetes and decreased Firmicutes in psoriasis [15]. These changes in gut microbiota are considered to be crucial causes for initiating or exacerbating psoriasis in humans and animal models [16,17]. Treatment for psoriasis may change the composition of the skin and gut microbiota [18,19,20]. A change in lesional skin microbiota has been associated with a clinical response after balneotherapy [18] and phototherapy [19]. A reduced mean rate of Staphylococcus aureus on psoriatic plaques, reaching a nadir at weeks 16–20 after treatment, was noted in our previous research [20]. Regarding the gut microbial change after psoriasis treatment, the relative abundance of Pseudomonadaceae and Enterobacteriaceae increased significantly following secukinumab therapy, while no significant change was noted in gut microbiome composition following ustekinumab treatment [21]. In the past 20 years, findings from immunological and genetic studies have highlighted causal immunological circuits of psoriasis that converge on adaptive immune pathways involving interleukin (IL)-17 and IL-23 [1,22,23]. The suppression of psoriasis-related, proinflammatory, and Th17-associated cytokines, such as tumor necrosis factor (TNF)-α, IL-17A, and IL-23, was observed in mice fed with Lactobacillus pentosus [24]. The clinical significance of the interaction between microbiota and the immune system is of importance. Although guselkumab, a selective IL-23 inhibitor, and secukinumab and ixekizumab, monoclonal antibodies targeting IL-17A, were highly effective in treating psoriasis, their treatment results in IBD were not consistent. Clinical trials for biologics blocking either IL-17A or its receptor have contributed to the exacerbation of IBD [25,26]. This raised the possibility that blockade of IL-17 could interfere with the microbiota composition and homeostasis in the intestine that might predispose susceptible individuals to develop IBD [27,28]. Moreover, in a phase 2 trial, guselkumab demonstrated a greater efficacy than a placebo in patients with Crohn’s disease [29]. These findings indicated a sophisticated interaction between gut microbiota composition and biologic therapies. Yet, how gut microbiota in psoriasis react to the IL-17 and IL-23 blockers has scarcely been investigated. Therefore, this study aimed to investigate the dynamic alteration of gut microbiota in psoriasis patients before and after receiving IL-17 and IL-23 antagonists. A total of one hundred and ninety-two fecal samples were obtained from 48 patients with 30 cases receiving the IL-23 inhibitor (guselkumab) (mean age 45.2 years) and 18 cases receiving IL-17 inhibitors (secukinumab and ixekizumab) (mean age 52.8 years). There was no significant difference in gender, weight, psoriatic arthritis, baseline PASI score, and baseline CRP level between the two groups. Patients treated with an IL-17 inhibitor were older than patients treated with an IL-23 inhibitor (Table 1). The mean PASI scores decreased at weeks 4, 12, and 24 after either IL-23 or IL-17 inhibitor therapy. All these changes from baseline were significant (Figure 1A). In addition, the CRP level was significantly reduced after 12 weeks and 24 weeks of treatment (Figure 1B). Moreover, we found recruited patients did not change their eating habits during the study. We studied the temporal alteration of microbial diversity in patients treated with IL-23 or IL-17 inhibitors. Calculation of the weighted-UniFrac distance matrix (β diversity) displayed a significantly altered distance in microbial community structures among samples from patients receiving an IL-23 or IL-17 inhibitor during 24-week treatment, while no significant difference in the α diversity was observed among the groups (Figure 2A). Moreover, Bray–Curtis distance was used to measure β diversity at week 0 and 24 among the responders (R) and non-responders (NR) (Figure 2B). The results showed that β diversity of gut microbiota in the responders to the IL-23 inhibitor was significantly higher than that in non-responders both at baseline and week 24 (p < 0.05), while there was no significant difference in β diversity between responders and non-responders treated with IL-17 inhibitors. We then sought the most relevant taxa whose abundance altered after the treatment (week 24) to explore the effect of biologics on the composition of gut microbiota. In patients treated with the IL-23 inhibitor, we identified five taxa whose levels were significantly different from the baseline (Figure 3). The relative abundance of Roseburia, Lachnoclostridium, Bacteroides vulgatus, Anaerostipes, and Escherichia–Shigella increased over the time course of the treatment. In patients treated with IL-17 inhibitors, levels of Bacteroides stercoris and Parabacteroides merdae were significantly increased at week 24, while those of Blautia and Roseburia were significantly reduced (Figure 3). Furthermore, we assessed the association between the therapeutic outcome and changes in relative abundance of individual taxa from the baseline to 24 weeks post-treatment. We found that among patients treated with the IL-23 inhibitor for 24 weeks, the relative abundance of Lachnospiraceae and Romboutsia significantly decreased from the baseline in the responders compared to non-responders (Figure 4). Meanwhile, the relative abundance of Fusicatenibacter in patients treated with IL-17 inhibitors for 24 weeks significantly increased compared to non-responders, whereas that of Lachnospiraceae NK4A136 and Roseburia significantly decreased (Figure 4). Considering the pathways related to metabolism, we found a number of pathway modules associated with lipid metabolism, inositol phosphate metabolism, and glutathione metabolism enriched in patients treated with the IL-23 inhibitor at week 24. In contrast, bacterial genes assigned to energy metabolism, arginine biosynthesis, cysteine and methionine metabolism, fructose and mannose metabolism, and carbapenem biosynthesis were less abundant. In patients treated with IL-17 inhibitors, the abundance of pathway modules associated with indole alkaloid biosynthesis increased, while that with lysine biosynthesis decreased (Table 2). In addition, we investigated alterations in microbial functions at week 24 from the baseline between responders and non-responders. Among patients treated with the IL-23 inhibitor, the pathway of taurine and hypotaurine metabolism was enriched in the responders compared to non-responders (Table 3). Among patients treated with IL-17 inhibitors, 13 metabolism pathways were significantly enriched and 3 decreased in responders after 24 weeks of treatment compared with non-responders (Table 3). The pathways involved in amino acids metabolism, biosynthesis of antibiotics, and carbohydrate metabolism were differentially enriched from the baseline after the treatment with IL-17 inhibitors. In the present study, we analyzed the gut microbial diversities and taxonomies in patients with psoriasis at weeks 4, 12, and 24 after the treatment with IL-23 or IL-17 inhibitors. This is the first study to demonstrate a significant increase in β diversity of gut microbial communities and altered abundance of certain bacteria in patients receiving the IL-23 inhibitor for 24 weeks. In addition, we identified microbial taxa and functional pathways associated with the therapeutic options and treatment responses. Changes in gut microbiota composition due to therapeutic agents and their influence on clinical response have been reported in patients with inflammatory bowel disease (IBD) [30]. Common types of gut microbiota change after biologics treatment encompassed an increased abundance of short-chain fatty acids (SCFAs)-producing bacteria, which are considered beneficial commensal bacteria [30]. An improvement in intestinal dysbiosis was reported with an increment in the abundance of SCFAs-producing bacteria such as Anaerostipes, Blautia, and Roseburia from IBD patients after receiving infliximab [31]. Moreover, similar findings were also demonstrated in IBD patients receiving ustekinumab [32]. In this study, we found that the relative abundance of Anaerostipes and Roseburia increased in patients after IL-23 inhibitor treatment, which may increase the production of SCFAs and consequently restore the immunomodulatory function and intestinal epithelial barrier [33,34]. Conversely, the abundance of Blautia and Roseburia was reduced in those receiving IL-17 inhibitors. One previous study investigating the impact of secukinumab on gut microbial composition [21] showed a reduction in the abundance of the SCFAs-producing bacteria Firmicutes, consistent with our findings. The Bacteroides genus constitutes 30% of the total colonic bacteria and Bacteroides vulgatus is one of the most commonly encountered Bacteroides species in the human gut [35]. The role of B. vulgatus in modulating the immune system has been investigated in animal experiments. Supplementation with B. vulgatus attenuated symptoms of colitis in mice and decreased the expression of TNF-α, IL-1β, and IL-6 in the colon [36]. Moreover, suppression of the systemic and intestinal immune response was observed in mice gavaged with Bacteroides vulgatus [37,38]. The present study demonstrated that the relative abundance of Bacteroides vulgatus increased after anti-IL-23 inhibitor treatment, which might further imply the beneficial effect of gut immunomodulation by the IL-23 inhibitor in psoriasis. The gut is considered to be a major immune organ, with gut-associated lymphoid tissue (GALT) being the most complex immune compartment [39]. It is well known that changes in the gut microbial composition may promote both health and disease [40]. Strong evidence has indicated that intestinal dysbiosis is clinically relevant to psoriasis [41]. The importance of the gut–skin axis in the pathogenesis of psoriasis has recently been documented in humans, as well as in animal models. [42]. In imiquimod-induced psoriasis-like mice, gut microbiota promoted intestinal and cutaneous inflammations by enhancing the IL-23/IL-17 axis [42,43]. In addition, a gut microbial genus, Romboutsia, increased in mice with imiquimod-induced psoriasis [43], suggesting that IL-23/IL-17-axis-related psoriasis may be associated with levels of gut Romboutsia. Intriguingly, our study revealed that the abundance of Romboutsia significantly decreased at week 24 in the responders to the IL-23 inhibitor when compared with non-responders. However, there was no significant difference in the gut Romboutsia level between responders and non-responders treated with IL-17 inhibitors. Based on these findings, we speculate that blocking IL-23 may ameliorate Romboutsia-mediated psoriasis by improving IL-23/IL-17-axis-related skin inflammation. At the genus level, an enriched Lachnospiraceae NK4A136 group was detected in patients with ankylosing spondylitis [44] and IBD [45]. Recently, a study on the gut microbiome demonstrated an increase in the abundance of gut Lactobacillaceae in psoriatic patients [13]. Our results further revealed that the abundance of Lachnospiraceae NK4A136 at week 24 significantly decreased in responders to IL-17 inhibitors compared to non-responders. It has been shown that the Lachnospiraceae NK4A136 group is correlated with elevated levels of intestinal IL-17 and IL-6 in mice with diabetes mellitus, resulting in intestinal inflammation [46]. Thus, we hypothesize that responders to IL-17 inhibitors might benefit from reduction in the gut Lachnospiraceae NK4A136 group, which likely contributes to declined skin inflammation. Further investigation should be conducted to address the causal relationship of these findings. In our study, sixteen KEGG pathways were found to be significantly enriched in responders to IL-17 inhibitors, such as the biosynthesis of amino acids, energy metabolism, and biosynthesis of antibiotics including vancomycin, validamycin, and novobiocin. Previously, dramatic changes in glucose metabolism, amino acid metabolism, and energy metabolism have been shown in psoriasis [47,48]. Metabolic regulation of cell proliferation and apoptosis was thought to be critical for dysregulated keratinocyte hyperproliferation in psoriasis [49,50]. Altogether, these findings suggest that altered gut-microbiota-mediated biosynthesis of amino acids and energy metabolism may also contribute to specific phenotypes in patients with psoriasis, such as uncontrolled keratinocyte hyperproliferation. It was reported that treatment with broad-spectrum antibiotics in mice with imiquimod-induced psoriasis reduced proinflammatory IL-17-producing T cells and skin thickness [16,42]. Moreover, Actinobacteria, isolated from the gut of freshwater fish, exhibited antimicrobial activities by producing antibiotic compounds [51]. Our data showed that gut microbiome-encoded metabolic KEGG pathways enriched in the responders to IL-17 inhibitors were concentrated in the biosynthesis of antibiotics. According to these findings, we suggest that IL-17 inhibitors may partially improve psoriasis-related skin inflammation by enhancing gut-microbiota-mediated biosynthesis of antibiotics. In addition, reduction in the abundance of the taurine and hypotaurine metabolism pathway in patients with severe psoriasis has been observed in one recent study [52]. Our results demonstrated that the abundance of the taurine and hypotaurine metabolic pathway was significantly enhanced in the responders to the IL-23 inhibitor, as compared with that in non-responders. Taurine, an abundant amino acid in leukocytes, is found in high concentrations in inflammatory lesions and tissues exposed to oxidative stress. [53]. Collectively, these findings and our data imply that a shift in gut bacterial composition due to the IL-23 inhibitor could lead to significant changes in taurine metabolism, which may correlate with an improvement in the inflammatory status in patients with psoriasis. Our results should be considered in the context of several limitations. First, sample sizes were limited, and larger cohorts should be assessed in future studies. Second, due to the relatively limited resolution of the 16S rRNA sequencing technique [54], shotgun metagenomic sequencing methods are needed to identify specific bacterial strains in psoriasis. Third, based on the gut-microbiota-mediated metabolic pathways related to the response to the IL-23 inhibitor or IL-17 inhibitors identified in psoriatic patients, it is necessary to explore their key regulatory targets. Finally, we did not investigate inflammatory markers collected from the peripheral blood, gut, and stool so we could not explain the association of inflammatory changes with the microbial composition. In summary, treatments with IL-23 and IL-17 inhibitors were associated with distinct shifts in gut microbial composition in patients with psoriasis. Significant differences in the relative abundance of bacteria taxa between the responders and non-responders suggested that IL-23 and IL-17 inhibitors may functionally interact with gut microbiota to reduce cutaneous inflammation. Moreover, we demonstrated the association between the treatment response and gut microbial function, which might serve as potential biomarkers in the treatment response. This prospective study enrolled forty-eight patients with psoriasis, including 30 cases treated with an IL-23 inhibitor (guselkumab) and 18 cases with IL-17 inhibitors (ixekizumab or secukinumab) in the Chang Gung Memorial Hospital (Taoyuan, Taiwan) from September 2020 to March 2022. None of the included cases had taken systemic antibiotics, systemic immunosuppressant agents, oral corticosteroids, and probiotics one month before each sample collection other than guselkumab, ixekizumab, or secukinumab. The anti-IL-23 medication group received guselkumab 100 mg at week 0, 4, and every 8 weeks thereafter. The anti-IL-17 medication group received either ixekizumab 160 mg at week 0, 80 mg at week 2, 4, 6, 8, 10, and 12, and 80 mg every 4 weeks thereafter, or secukinumab 300 mg at week 0, 1, 2, 3, and 4, and every 4 weeks thereafter. The demographics and clinical data of the patients, including their age, gender, weight, and psoriatic arthritis (PsA), were collected at baseline. Psoriasis Area and Severity Index (PASI) score and serum C-reactive protein (CRP) level were collected at week 0, 4, 12, and 24. Responders were defined as those having a PASI improvement of ≥90% after 24 weeks of treatment and non-responders as having a PASI improvement of <90%. Information about food intake during the study was collected at week 0, 12, and 24 through a food-frequency questionnaire (FFQ) [55]. Stool specimens were collected using the Longsee Fecalpro Kit (Longsee Medical Technology Co., Guangzhou, China) at baseline and 4 weeks, 12 weeks, and 24 weeks after treatment. As described previously [56], DNA was isolated by using the QIAamp PowerFecal Pro DNA Kit for Feces (Qiagen, Germantown, MD, USA) following the manufacturer’s instructions. Around 0.25 g of the sample in the Bead Tube was added with 750 μL of PowerBead Solution and 60 μL of Solution C1, which was then heated at 65 °C for 10 min. The mixture was vortexed by using a PowerLyser Homogenizer at 1000 RPM for 10 min. After the steps of cell lysis, removal of contaminating matters, washing and eluting with DNA-free, PCR-grade water, DNA was extracted. The concentrations and qualities of the extracted DNA were measured by using Qubit 4 Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). The variable regions 3 and 4 (V3–V4) of 16S rRNA gene were PCR (polymerase chain reaction)-amplified by using the primer set (the Illumina V3 forward 5′-CCTACGGGNGGCWGCAG-3′ and V4 reverse 5′-GACTACHVGGGTATCTAATCC-3′) [57]. The Illumina sequencing adapters ligated to the purified amplicons by a second-stage PCR using the TruSeq DNA LT Sample Preparation Kit (Illumina, San Diego, CA, USA) were performed to construct a library. Purified libraries were quantified, normalized, pooled, and applied for cluster generation and sequencing on a MiSeq instrument (Illumina). Paired-end reads were processed by using DADA2 [58] to filter out noisy sequences, correct errors in marginal sequences, remove chimeric sequences, and eliminate singletons to infer amplicon sequence variants (ASVs). Bacterial taxonomy was assigned by applying a pre-fitted QIIME2 classifier built with the Scikit-lean package [59] based on the information collected from the SILVA database [60]. Arrangement of multiple sequences were performed by the PyNAST software v.1.2 [61] for assessment of the phylogenetic relationship of various ASVs, and a phylogenetic tree was constructed with the FastTree 2.1.0 [62]. Functional composition of metagenomes was predicted from 16S rRNA data by the Tax4Fun2 software [63]. To predict functional profile of the microbial community, the taxonomic abundance transformed from the SILVA-based 16S rRNA and normalized by the 16S rRNA copy number acquired from the NCBI annotations were applied to incorporate the precomputed functional profiles of KEGG pathways [63]. KEGG analysis was only focused on “Metabolism” pathways. Demographic and clinical characteristics were presented as n (%) for categorical variables and mean ± standard deviation (SD) or median with range for continuous variables. For estimating alpha diversity, species richness was evaluated by inverse Simpson’s index. Beta diversity was analyzed by Bray–Curtis or unweighted-UniFrac distance matrix. In order to investigate the association of treatment effect and bacteria in the fecal specimens, we further identified differentially abundant bacterial taxa among groups. Statistically significant biomarkers were analysed by the non-parametric Kruskal–Wallis test, Wilcoxon rank-sum test, and linear discriminant analysis (LDA) to identify differentially abundant taxa. Change in relative abundance after treatment from the baseline was compared between responders and non-responders by fitting a linear mixed model, measured on a continuous scale to identify longitudinal biomarkers. All statistical tests are two-tailed, and a p-value less than 0.05 was considered statistically significant.

PMC10002563

Laura Koll,Désirée Gül,Manal I. Elnouaem,Hanaa Raslan,Omneya R. Ramadan,Shirley K. Knauer,Sebastian Strieth,Jan Hagemann,Roland H. Stauber,Aya Khamis

Exploiting Vitamin D Receptor and Its Ligands to Target Squamous Cell Carcinomas of the Head and Neck

28-02-2023

gender-specific effects,nuclear receptors,calcitriol,3D tumor spheroids

Vitamin D (VitD) and its receptor (VDR) have been intensively investigated in many cancers. As knowledge for head and neck cancer (HNC) is limited, we investigated the (pre)clinical and therapeutic relevance of the VDR/VitD-axis. We found that VDR was differentially expressed in HNC tumors, correlating to the patients’ clinical parameters. Poorly differentiated tumors showed high VDR and Ki67 expression, whereas the VDR and Ki67 levels decreased from moderate to well-differentiated tumors. The VitD serum levels were lowest in patients with poorly differentiated cancers (4.1 ± 0.5 ng/mL), increasing from moderate (7.3 ± 4.3 ng/mL) to well-differentiated (13.2 ± 3.4 ng/mL) tumors. Notably, females showed higher VitD insufficiency compared to males, correlating with poor differentiation of the tumor. To mechanistically uncover VDR/VitD’s pathophysiological relevance, we demonstrated that VitD induced VDR nuclear-translocation (VitD < 100 nM) in HNC cells. RNA sequencing and heat map analysis showed that various nuclear receptors were differentially expressed in cisplatin-resistant versus sensitive HNC cells including VDR and the VDR interaction partner retinoic acid receptor (RXR). However, RXR expression was not significantly correlated with the clinical parameters, and cotreatment with its ligand, retinoic acid, did not enhance the killing by cisplatin. Moreover, the Chou–Talalay algorithm uncovered that VitD/cisplatin combinations synergistically killed tumor cells (VitD < 100 nM) and also inhibited the PI3K/Akt/mTOR pathway. Importantly, these findings were confirmed in 3D-tumor-spheroid models mimicking the patients’ tumor microarchitecture. Here, VitD already affected the 3D-tumor-spheroid formation, which was not seen in the 2D-cultures. We conclude that novel VDR/VitD-targeted drug combinations and nuclear receptors should also be intensely explored for HNC. Gender-specific VDR/VitD-effects may be correlated to socioeconomic differences and need to be considered during VitD (supplementation)-therapies.

Exploiting Vitamin D Receptor and Its Ligands to Target Squamous Cell Carcinomas of the Head and Neck Vitamin D (VitD) and its receptor (VDR) have been intensively investigated in many cancers. As knowledge for head and neck cancer (HNC) is limited, we investigated the (pre)clinical and therapeutic relevance of the VDR/VitD-axis. We found that VDR was differentially expressed in HNC tumors, correlating to the patients’ clinical parameters. Poorly differentiated tumors showed high VDR and Ki67 expression, whereas the VDR and Ki67 levels decreased from moderate to well-differentiated tumors. The VitD serum levels were lowest in patients with poorly differentiated cancers (4.1 ± 0.5 ng/mL), increasing from moderate (7.3 ± 4.3 ng/mL) to well-differentiated (13.2 ± 3.4 ng/mL) tumors. Notably, females showed higher VitD insufficiency compared to males, correlating with poor differentiation of the tumor. To mechanistically uncover VDR/VitD’s pathophysiological relevance, we demonstrated that VitD induced VDR nuclear-translocation (VitD < 100 nM) in HNC cells. RNA sequencing and heat map analysis showed that various nuclear receptors were differentially expressed in cisplatin-resistant versus sensitive HNC cells including VDR and the VDR interaction partner retinoic acid receptor (RXR). However, RXR expression was not significantly correlated with the clinical parameters, and cotreatment with its ligand, retinoic acid, did not enhance the killing by cisplatin. Moreover, the Chou–Talalay algorithm uncovered that VitD/cisplatin combinations synergistically killed tumor cells (VitD < 100 nM) and also inhibited the PI3K/Akt/mTOR pathway. Importantly, these findings were confirmed in 3D-tumor-spheroid models mimicking the patients’ tumor microarchitecture. Here, VitD already affected the 3D-tumor-spheroid formation, which was not seen in the 2D-cultures. We conclude that novel VDR/VitD-targeted drug combinations and nuclear receptors should also be intensely explored for HNC. Gender-specific VDR/VitD-effects may be correlated to socioeconomic differences and need to be considered during VitD (supplementation)-therapies. In the last three decades, there have been tremendous attempts to undercover the role of vitamin D (VitD) in the prevention, prognosis, and treatment of cancer. Unfortunately, the results have been contradictory, and until now, no general recommendations or standard treatment options considering VitD for cancer patients exist [1,2]. However, the majority of the observational studies supported a benefit of higher vitamin D intake concerning the reduction in cancer incidence (e.g., colon and breast cancer) [3,4]. Other studies showed a correlation between high serum VitD levels and lower cancer risk [5,6]. Nevertheless, for many entities, the clinical relevance of VitD as well as its molecular mechanisms of action requires further investigation. Head and neck squamous cell carcinomas (HNC) are among the top ten most common cancers worldwide, frequently exhibiting limited treatment response [7,8,9,10]. Here, reasons for unsatisfactory treatment success and the long-term survival of HNC patients can be found in the development of resistance toward established treatments as well as the lack of novel therapeutic targets. A promising approach to potentially increase the success of established treatment options, which are surgery, chemo-, radiotherapy as well as targeted (Cetuximab) therapy, is the use of combinational treatments. Here, the use of functional foods such as VitD offers an alternative, cost-effective cancer care regimen harboring the potential to improve treatment success. Functional foods and food components affect the body, reaching beyond a basic nutritional effect. Among the group of functional foods, VitD is one of the most important members for which anti-tumoral effects have already been suggested [11,12]. Interestingly, little is known about the relevance of VitD for HNC patients in general and the potential clinical benefit of combinational VitD therapies in particular. Previous studies and meta-analyses have already demonstrated the need to determine and evaluate the VitD influence on cancer pathogenesis and patient prognosis [13,14]. VitD, which is rather a (steroid) hormone than a vitamin, has a variety of functions in health and disease [15]. Approximately 90% of the VitD requirement is produced in the skin in response to ultraviolet-B (UV-B) light from sun exposure. The biologically active form of VitD, 1α,25-dihydroxyVitD3 (1,25(OH)2D3), also called calcitriol, is produced by enzymatic conversions of its precursor calcidiol (25-hydroxy VitD, 25(OH)D3) via hydroxylation in the liver and kidneys [16]. 25-Hydroxy VitD is the most single reliable marker of VitD concentration in the body due to its relatively long half-life time (three weeks) compared to the active form of 1,25-OH2D (approximately 4 h) [15,16,17]. It is also an indication of the availability of the substrate for tissue production and the auto/paracrine action of 1,25-OH2D. On the other hand, although 1,25-OH2D is the active form, it is regulated by several enzymatic and physiological inputs [16]. The fact that 1,25-OH2D concentrations in the blood may not decrease or decrease at a late stage even in presence of VitD deficiency make 25-(OH)D3 a better marker for the assessment of VitD supply [18]. Notably, VitD deficiency is widespread and can cause various diseases such as rickets in children and osteoporosis [19]. Therefore, VitD food fortification is practiced in some countries. Moreover, VitD deficiency has been correlated to multiple systemic and physiological conditions such as insulin resistance and diabetes, autoimmune disease, cardiovascular disease, and all-cause mortality [16,20]. Importantly, VitD deficiency seems to be (in)directly correlated with the occurrence of cancer [21,22]. Furthermore, previous studies have suggested that sufficient levels of VitD can reduce the risk of many cancer types such as colon and breast cancer [3,4]. Meta-analyses show that patients with serum levels of the VitD pre-cursor calcidiol (25-hydroxyVitD) ranging from 20–40 ng/mL show a significantly reduced risk of about 35% for breast and colorectal cancer [23,24]. Furthermore, there is evidence for some entities such as breast, colorectal, lung, bladder cancer, and prostate cancer that higher serum levels of calcidiol at the time of diagnosis are correlated with improved survival rates [25,26,27,28,29]. VitD executes its biological functions via binding to the VitD receptor (VDR), a member of the nuclear receptor family [30,31]. Nuclear receptors (NRs) are key regulators of health and disease including cancer, and thus represent important targets for anti-cancer therapies [32,33,34]. NRs are transcription factors involved in a wide range and extremely complex spectrum of physiological and pathophysiological processes, hence they are interesting therapeutic targets [32,35]. However, despite intensive research, the detailed mechanisms of homo-/heterodimerized nuclear receptors including VDR are still not resolved (for more details, see also following reviews [3,4,32,34]. Upon the binding of its ligand VitD and nuclear translocation, VDR is able to form heterodimers with the retinoid X receptor (RXR). By binding to specific VDR-responsive elements on the DNA, the nuclear receptors are able to activate various transcriptional programs [3,4,30,31,32,34,36]. Importantly, for VDR, anti-tumoral effects have been already suggested [3,4,32,34,37,38]. Hence in this study, we investigated the (pre)clinical and potential therapeutic relevance of the VDR/VitD-axis to assess the association between the VitD level and VDR expression for HNC. Aside from analyzing the HNC dataset of The Cancer Genome Atlas (TCGA), a case-control study was analyzed. To mechanistically uncover VDR/VitD’s pathophysiological relevance, we further combined the evaluation of clinical data with comprehensive dry and wet lab systematic studies of innovative HNC cell models. Besides the use of the 2D tumor cell model, there is increasing evidence that advanced 3D tumor spheroids react differently compared to conventional 2D cultures when exposed to drugs, radiation, or signaling ligands [39,40,41,42]. Hence, we established a 3D cell culture model aiming to approach the tumor situation in vivo. In comparison to the 2D culture systems, 3D spheroids exhibit a number of advantages, for example, they mimic a more realistic 3D architecture of a tumor including the supply of nutrients, oxygen, and anti-cancer drug. Another advantage is the development of polarity in the spheroid culture due to neighboring cell-to-cell contacts [39,40]. Collectively, cells in 3D tumor spheroids seem to preserve key morphological and signaling patterns closely associated with tumor development and drug resistance in animal models and patients [39,40,41,42]. As knowledge of the VDR/VitD-axis for HNC is limited, we first investigated the VDR expression and VitD serum levels in a cohort of newly diagnosed HNC patients (n = 40) compared to the healthy individuals (n = 40) (details see Table 1, Supplementary Tables S3 and S4). The most common site of occurrence was the tongue (60%) and the least was the lip (Figure 1a). Notably, regarding gender, there was a significant difference in the male-to-female ratio (Figure 1b, Supplementary Tables S3 and S5) which is often observed in the Middle East and North Africa (MENA) region [43,44,45,46]. Histopathologically, the most common differentiation subtype was moderately differentiated HNC (60%, n = 24; Figure 1c, Supplementary Table S5). In order to correlate the VitD serum levels with VDR expression in the tumor tissues, peripheral blood samples were taken from the patients before or during surgery. Total serum VitD (25-hydroxyVitD3) was quantified by using fully validated, modified high-performance liquid chromatography (HPLC) [47]. The VitD serum levels were lowest in patients with poorly differentiated cancers (4.1 ± 0.5 ng/mL), increasing from moderate (7.3 ± 4.3 ng/mL) to well-differentiated (13.2 ± 3.4 ng/mL) (Figure 1d, Table 2). The mean serum VitD level was 7.4 ± 4.5 ng/mL in cancer patients in comparison to 28.7 ± 4.6 ng/mL in healthy individuals (Figure 1e, Table 2). Notably, females showed higher VitD insufficiency compared to males, correlating with poor tumor differentiation (Table 1). Additionally, the VDR protein expression was analyzed by immunofluorescence and immunohistochemical staining in tumor biopsies classified as poorly, moderately, and well-differentiated (Figure 1g–i). Here, a significant inversely proportional correlation between the VitD levels and VDR expression was found (Figure 1f). As shown in Figure 1g–i, all studied cases showed immunofluorescence reactivity to the VDR antibody with varying intensities. Moreover, we found that the VDR levels correlated with the patients’ clinical and pathobiological tumor parameters. Particularly, poorly differentiated tumors showed high VDR and Ki67 expression (Figure 1g), whereas VDR and Ki67 levels decreased from moderate to well-differentiated tumors (Figure 1h,i). To independently confirm the relevance of VDR expression in the HNC patients, we bioinformatically analyzed the PANCAN dataset acquired from The Cancer Genome Atlas (TCGA), encompassing more than 12,000 samples of cancer patients of various entities and clinical backgrounds. Moreover, upon binding of its ligand VitD and nuclear translocation, VDR is also able to form heterodimers with the retinoid X receptor (RXR), thereby activating various cancer-relevant transcriptional programs (Supplementary Figure S1) [23,25,32,34,48,49,50]. As VDR/RXR expression has not been studied for HNC, we also studied the expression of RXR in the datasets. We found VDR overexpressed in the primary tumors. Comparing the different entities, the highest expression of VDR was found in rectal and colon adenocarcinoma and kidney cancer, directly followed by HNC (Supplementary Figure S2), supporting our conclusions obtained from the analyses of our cohort (see also Figure 1). Thus, in the second step, we focused on the analysis of the TCGA HNC cohort (n = 604) showing upregulation of VDR in tumor versus non-tumor tissues (Figure 2a, n = 564, p = 0.0059 **). Interestingly, RXRα expression showed no correlation with the disease markers (Figure 2b, n = 520 p = 0.4931). VDR expression highly correlated with the histological differentiation of the tumor, in contrast to the RXRα levels (Figure 2c,d, n = 540, p = 0.0002 ***/p = 0.0056 **). Since the HPV status affects the therapy outcome and prognosis of HNC patients, we analyzed HPV-negative versus HPV-positive patients. VDR expression was significantly increased in HPV-negative HNC patients (Figure 2e, n = 114, p < 0.0001 ****). Again, changes in the RXRα levels were less significant (Figure 2f; n = 114; p < 0.0108). Moreover, high VDR expression correlated with perineural invasion (Figure 2g, n = 393, p = 0.0006 ***) in contrast to the RXRα levels (Figure 2h, n = 393, p = 0.4154), underlining again the relevance of VDR but not of RXRα as a biomarker and/or therapeutic target for HNC. It is accepted by the field that the superfamily of nuclear receptors are key regulators in many pathologies including cancer [32,33,34]. Thus, we used ‘omics’ approaches to profile nuclear receptor expression and the potential pathobiological relevance in HNC tumor cell models. As HNC treatment is often complicated by recurrence due to resistance to cisplatin-based treatments, we analyzed the chemoresistant HNC cells. The cisplatin-resistant cell line, Picares, was established by selecting HNC Pica cells with sub-toxic concentrations of cisplatin (3–5 µM) for six months. Hence, Picawt and Picares allow for the comparison of cisplatin-sensitive and resistant HNC cells. Here, next-generation RNA sequencing transcriptomics was used to analyze the expression of various nuclear receptors (Figure 3a, Supplementary Table S6). As illustrated in the heat map analysis (Figure 3a; green: downregulated, red: upregulated), VDR and several other receptors such as Nuclear Receptor Subfamily 4 Group A Member 2 (NR4A2) or RXRα were differentially expressed in therapy-resistant (res) versus sensitive (wt) Pica cells. These data also suggest investigating the pathobiological relevance of other nuclear receptors for HNC in comprehensive follow-up studies. When studying the impact of nuclear receptors, it is also the key to control if the respective receptor is expressed and indeed capable of cytoplasmic to nuclear trafficking upon ligand binding in the relevant cell model. Nuclear translocation is required to activate ligand-dependent transcriptional programs [3,4,30,31,32,34,36]. The activation of VDR by ligand binding typically involves VDR-RXRα dimerization and the initiation of downstream signaling (Supplementary Figure S1). The immunofluorescence staining of endogenous VDR and RXR receptors demonstrated that VitD triggered nuclear accumulation of the receptors, in contrast to retinoic acid (RA) treatment alone (Figure 3b). When referring to VitD, the active form calcitriol (1,25(OH)2D3) was used if not indicated otherwise. Hence, although both receptors are capable of cytoplasmic to nuclear trafficking, VitD and VDR seem more relevant in HNC cells. We also confirmed and quantified the VDR expression in different HNC cell lines (Figure 3c,d). To further study the kinetics of VDR translocation in real-time, we established HNC cell lines stably expressing VDR fused to GFP. Therefore, the VDR reading frame was cloned from primary HNC tumor cells and stably expressed VDR-GFP in the HNCUM-02T or HNC FaDu cells (Figure 3e). An important question regarding VDR’s nuclear translocation is the determination of the most effective ligand dose and the time kinetics of the process. Using the high-content screening microscopy platform, Array Scan VTI, we automatically quantified VDR translocation. Here, cells were treated with different clinically relevant doses of VitD (0–100 nM) for 30 min (Figure 3f,g). Fluorescence microscopy showed dose-dependent VDR translocation into the nucleus by VitD, which was most effective at a VitD concentration of 100 nM. Importantly, RA alone did not trigger the nuclear translocation of VDR (Supplementary Figure S3). Chemoresistance is not only one of the main causes influencing cancer progression, but it is also strongly correlated to the cancer mortality rates. Hence, developing strategies for enhancing chemo sensitivity, potentially also by functional food supplementation with VitD, is expected to benefit patients. Indeed, such efforts have been made to correct VitD deficiency in cancer patients [11,12,51,52]. However, the success of VitD/VDR targeting therapies requires mechanistic knowledge and experimental investigation in vitro. To examine the effect of combination therapy on HNC, we thus measured the cell viability after the VitD/cisplatin treatments. To also mimic the pathophysiological conditions of high and low VitD serum levels, cells were seeded in the presence or absence of 100 nM VitD, which we found to trigger efficient VDR nuclear translocation, and thus biological activation (see Figure 3f,g). After 24 h of VitD pre-treatment, cells were additionally treated with physiological concentrations of VitD (100 nM), 15–20 µM cisplatin, or a combination (Figure 4). As expected, VitD alone did not affect cell viability. However, the combination treatments significantly enhanced tumor cell death compared to cisplatin alone in the three HNC cell lines tested (Figure 4a; Supplementary Figure S4). To objectively uncover a potential synergistic effect of VitD/cisplatin combinational treatments, we performed the Chou–Talalay method. The calculation of the combination index (CI) using the Chou–Talalay algorithm allowed us to uncover additive (CI = 1), synergistic (CI < 1), or antagonistic effects (CI > 1) of the drug combinations [53]. As shown in Figure 4b–d, all calculated indices were less than 1, revealing a synergistic effect on tumor cell killing for the VitD/cisplatin combinations in the tested HNCUM 02T, FaDu, and Pica cell lines. Conventional 2D tumor cell models are well-established tools to assess various aspects of tumor pathobiology. However, there is increasing evidence that advanced 3D tumor spheroids react differently compared to conventional 2D cultures when exposed to drugs, radiation, or signaling ligands [39,40]. The architecture of spheroids leads to a gradient of nutrition and oxygen from the outer surface to the core, and drug delivery to parts of the 3D cell cluster also seems to differ. Additionally, cells in 3D tumor spheroids seem to preserve certain distinct signaling patterns that are closely associated with drug resistance in animal models and patients. In order to closely approach the tumor situation in vivo, we next established HNC 3D tumor spheroids to investigate the effects of VitD/VDR targeting in an experimental setting, more closely mimicking the patients’ tumor microenvironments. Here, cells were cultured in ultra-low adhesion cell culture vials that promoted the formation of 3D spheroid-shaped tumor cell clusters. As summarized in Figure 5a, various pathobiological relevant properties of the established 3D spheroids were subsequently analyzed by fully automated high-content microscopy, allowing for an objective assessment of the tumor spheroids’ growth, morphology, and vitality. First, we found that the synergistic killing effect of the VitD/cisplatin combinations observed in the 2D cultures was also relevant for the 3D spheroids (Figure 5b). Cotreatment significantly reduced the mean objective area and viability (Figure 5b,c). Interestingly, although VitD alone did not affect the vitality of the 2D cultures, it already affected the 3D spheroid formation and induced morphological and architectural changes. As shown in Figure 5b–e and Supplementary Figure S5, automated high-content microscopy revealed that spheroid formation and growth were significantly impaired, suggesting that the expression of epithelial surface markers may be reduced. Notably, the effect was more prominent for the cisplatin-resistant cell line Picares (Figure 5d,e, Supplementary Figure S5), although the molecular details are not known. In conclusion, these data uncover a novel effect of VitD and also demonstrate that 3D tumor spheroids are a valuable experimental tool to uncover aspects of tumor pathobiology potentially occluded in conventional 2D tumor cell models. To further investigate how VitD or VitD/cisplatin combinations inhibit the proliferation and clonogenic survival of HNC cells, we examined the cancer-relevant signaling pathways. First, bioinformatics analyses employing the Ingenuity Pathway Analysis software (Version v01-04) revealed multiple molecular mechanisms involved in cancer pathogenesis and treatment resistance (Supplementary Figure S6). Subsequently, we focused on potential VDR-RXR activation pathways (Supplementary Figure S1) and further explored the literature [54,55,56,57]. As summarized in Figure 6a,b, VitD has been suggested to regulate several pathways including the cancer-relevant mTOR/PI3K-Akt pathways. Here, key regulatory proteins are (in)directly affected by VitD overlap such as the Akt kinase (Figure 6a,b). Under ‘healthy’ conditions, the mTOR/PI3K-Akt pathways are important players in development, cellular homeostasis, and health control. However, in cancer, abnormally activated mTOR/PI3K-Akt signaling stimulates tumor cells to grow, metastasize, and become resistant to treatment [54,55,56,57,58]. Notably, when we examined the impact of VitD and VitD/cisplatin treatment combinations in HNC cell models, we found that expression of the active, phosphorylated forms of mTOR and Akt (i.e., of pmTOR and pAkt) was particularly decreased upon VitD/cisplatin cotreatment (Figure 6c,d). No significant reduction in pmTOR and pAkt was detected upon cisplatin treatment alone. These findings not only provide a potential molecular explanation for the enhanced cisplatin-killing effect on the cancer cells by VitD, but also suggest the further experimental exploitation of additional cotreatment combinations such as using mTOR and Akt inhibitors in combination with VitD. The VDR/VitD-axis has been intensively investigated for more than a decade for the prevention and/or treatment of many cancers. Such (pre)clinical studies range from VitD food supplementation and cancer-prevention trials to different combination therapies [3,4,32,34,37]. Indeed, various anti-tumoral effects have been suggested for this member of the nuclear receptor superfamily, and VitD deficiency is often observed in cancer patients [21,22,59,60]. However, the underlying mechanisms of the VitD/VDR-mediated effects are not understood in detail and sometimes conflicting reports underline that its role, especially in specific cancer types, remains to be dissected [3,4,34,37]. Our clinical and experimental data support a significant role of the VDR/VitD-axis in the prognosis and clinical outcome of HNC patients. First, by analyzing our cohort of n = 40 HNC patients compared to healthy controls (n = 40), we demonstrated that both the VitD serum levels and VDR expression correlate with clinical parameters such as histopathological tumor classification. Although we could not provide specific data on patient prognoses such as survival curves, in general, the HNC patients’ overall survival correlates with histopathological differentiation of the tumor (see Supplementary Figure S7). Our finding that patients with poorly differentiated tumors and thus poor prognosis exhibited the lowest VitD levels is in line with previous studies of other entities. For example, Yao et al. found that low serum 25OHD levels at diagnosis were associated with poorer survival and worse prognosis in breast cancer patients [61]. Additionally, there have been studies observing an inverse relationship between cancer mortality and serum VitD level [59,62], suggesting that VitD supplementation therapy was most effective in patients with VitD deficiency at diagnosis [62]. However, in contrast to other clinical studies, we here paid attention to recruiting an age-sex-matched control group of non-cancer patients, allowing us to draw conclusions about a potential (gender-specific) correlation between the serum VitD level and HNC. This study’s confinement of cases to 40 patients due to the complexity of the subject matter could be seen as a potential limitation. It also has to be mentioned that the study cohort includes tumors of different sites such as the tongue and lip, which can differ in their prognosis. Nevertheless, the cohort is suitable to represent the commonly observed distribution of subsites and histopathological differentiation. Of note, our study cohort was recruited in Egypt, exhibiting socio-economical characteristics, which we feel worth discussing. First, the study cohort differed in its gender composition from typical Northern European and American study groups because it consisted mainly of women (male–female ratio 1:4). This is often observed in the Middle East and North Africa (MENA) region [43,44,45,46], which among other factors such as increased smoking [63,64] could be explained by differences in VitD supply. A normal VitD supply is defined as when the 25(OH)D serum concentrations ranged between 30 and 50 ng/mL, whereas levels <20 ng/mL were classified as VitD deficiency [65,66]. The mean level of serum VitD (25(OH)D) in the healthy population differs depending on the geographical residence, whereas mean VitD levels in adults in North America, Asia Pacific, and Europe range between 20.4 and 28.9 ng/mL, and thus could be classified as insufficient, but not yet deficient. Interestingly, in the Middle East and North Africa region (MENA) the mean VitD levels seem to be significantly lower with 13.6–15.2 ng/mL (applies to the same age group, does not take differences in sunshine duration into account) [67]. Different reasons may explain lower VitD levels in the MENA region such as increased air pollution, reducing the amount of UVB rays available for VitD production in the skin [17,68,69]. Another explanation could be a physiological de-toxification mechanism of VitD, which is activated after longtime sunlight exposure to prevent the toxic effects of very high VitD levels in the human body [15,16]. While the analyzed healthy patients of our study cohort lay above the statistic MENA value with a mean VitD concentration of 28.7 ng/mL, the HNC patients exhibited very low VitD levels (mean 7.4 ng/mL), classified as severe VitD deficiency (<12 ng/mL). Here, especially the female patients exhibited very low VitD levels (5.3 ng/mL), which is supported by other studies describing the female gender as a risk factor for hypovitaminosis [67]. Aside from the general reasons for VitD deficiency in the MENA region described above, additional circumstances such as veiling and/or reserved clothing style, lower socio-economic standard, and predominant indoor activity may contribute to VitD deficiency in women [67,70,71]. These factors come along with the lack of awareness about the importance of VitD to the human body [67]. Assuming a significant role of VitD in the pathogenesis of HNC, this could partly explain the increased incidence of HNC in females. Such a correlation has also been suggested for colorectal cancer. Here, the VitD levels were inversely proportional to the risk of cancer in women, but not statistically significant in men [24]. Furthermore, it has been proposed that VitD supplementation could be protective against breast cancer in menopausal women, underlining its effect on tumorigenesis [72]. Again, for the gender-specific conclusions also drawn from our study, the restricted sample size of n = 34 females should be considered, suggesting further larger studies focusing on the gender-specific relevance of VitD in HNC. Since VitD executes its biological functions via nuclear receptor binding, we analyzed the clinical relevance as well as expression and ligand-dependent activation of VDR and its heterodimerization partner RXRα. Here, we showed that VDR, but not RXRα, was significantly overexpressed in the primary HNC patients, which also correlated with clinically relevant disease markers such as HPV status, perineural invasion, and histopathological differentiation. However, there are some conflicting studies about the clinical relevance of VDR overexpression for tumorigenesis [73,74,75]. For example, Choi et al. correlated the VDR overexpression with negative prognosis in thyroid cancer [73], supporting our data showing that VDR overexpression in poorly differentiated, highly proliferative tumor tissue. Other studies have correlated the high VDR expression with an improved prognosis of patients [74,75,76]. RXRα expression and its clinical relevance in HNC have also been controversially discussed. RXR agonists such as bexarotene can benefit HPV-negative HNC patients [77]. Bexarotene combination therapy was also effective in a preclinical trial [78]. For breast cancer, there are studies demonstrating a concurrent overexpression of VDR and RXRα [79], partially also describing a worse disease-free survival when RXRα is overexpressed [80,81,82]. Hence, RXR might be worth investigating in future experimental and clinical VitD/VDR studies in general. Chemoresistance is a major cause of cancer progression and impacts the mortality of cancer patients, particularly for HNC [9,10,39,83]. Hence, developing strategies for enhancing chemosensitivity, potentially also by food supplementation with VitD, is needed and may benefit patients. Indeed, such efforts have been made to correct VitD deficiency in cancer patients [11,12,51,52]. Through our comprehensive in vitro studies applying established 2D as well as 3D spheroid HNC cell models, we could show that VitD treatment improves chemotherapeutic killing, especially of therapy-resistant HNC tumor cells, suggesting VitD supplementation during the primary (radio)chemotherapy of HNC patients. Of course, the serum VitD levels of respective patients should be carefully monitored during therapy, and other clinically relevant factors also have to be considered. Previous observational studies and clinical trials have partially reported improved survival of cancer patients after VitD supplementation, but the findings are not conclusive yet, and further studies combining clinical with wet lab investigation are needed [84]. Our nuclear receptor profiling by next-generation RNA sequencing transcriptomics provides the first data suggesting that other nuclear receptors may also be relevant for cisplatin-chemoresistance in HNC. Furthermore, VDR or RXRα investigated here, differentially expressed receptors such as Nuclear Receptor Subfamily 4 Group A Member 2 (NR4A2) seem to be relevant for various aspects of HNC pathobiology including HPV status and mTOR/Akt signaling, underlining the value of our datasets [85,86,87]. Due to the complexity of this area, we did not explore other nuclear receptors in this study, which might be considered as a potential limitation. Hence, we refer the reader to the literature regarding the specific receptor of interest. We conclude that the data provided here may stimulate the field to further explore the relevance of the nuclear receptor superfamily for therapy resistance in HNC. In cancer, abnormally (de)activated signaling pathways such as mTOR/PI3K-Akt and NFκB signaling stimulate tumor cells to proliferate aggressively, metastasize, and become even more resilient to therapy [54,55,56,57,58]. Here, we found that the VitD/VDR-axis enhances the chemotherapeutic effect via mTOR-PI3K/Akt downregulation in HNC. The potential relevance of the Akt- and mTOR pathways in VitD/VDR signaling is supported by reports in other tumor types [56,88]. It has to be mentioned that VitD executes its biological functions by various cellular pathways, and thus is likely that additional proapoptotic pathways may contribute to cancer-associated VitD effects. Of note, bioinformatic modeling and predictions, as performed in our study, will aid in hypothesis building, but detailed investigations are needed to confirm the candidates’ relevance. Our findings not only suggest an additional molecular mechanism for the observed beneficial effects of VitD supplementation, but also suggest further exploitation of additional cotreatment combinations such as using mTOR and Akt inhibitors (e.g., ICSN3250, LY3023414, AZD8055, or rapamycin) [58,89]. However, these preliminary results give the first molecular evidence for further co-treatment options, and detailed analyses have to be performed in future (pre-)clinical studies. Collectively, we can conclude that novel VDR/VitD-aided drug combinations should be intensely investigated in (pre)clinical studies. Here, gender-specific VDR/VitD-effects impacted by country-specific socioeconomic differences may need additional attention. Moreover, nuclear receptors should be further explored not only for breast or colon cancer, but also for HNC. Unless stated otherwise, chemicals were purchased from Sigma Aldrich/Merck (Darmstadt/Munich, Germany) or MSC (MSC UG and CoKG, Mainz, Germany). Cell culture media and reagents were sourced from Gibco/Thermo Fisher Scientific (Dreieich, Germany). Disposables were purchased from Greiner Bio-One (Frickenhausen, Germany). Ab used: α-VDR (sc-13133; Santa Cruz Biotechnology, Heidelberg, Germany), α-VDR (ab3508, Abcam, Erlangen, Germany), α-RXRα (5388, Cell Signaling, Leiden, The Netherlands), α-phospho-mTOR (5536, Cell Signaling, Leiden, The Netherlands), α-phospho-Akt (3787, Cell Signaling, Leiden, The Netherlands), and α-actin (A2066; Sigma Aldrich, Munich, Germany). Appropriate HRP-, Cy3-, or FITC-conjugated secondary antibodies (Sigma Aldrich, Munich, Germany; Santa Cruz Biotechnology, Heidelberg, Germany) were used (Supplementary Table S1). Reagents such as cisplatin were from Sigma (Sigma Aldrich, Munich, Germany) or MSC (MSC UG & CoKG, Mainz, Germany). 1α,25-Dihydroxy VitD3 (Calcitriol) was purchased from Sigma and Santa-Cruz (D1530, Sigma Aldrich, Munich, Germany and CAS 32222-06-3, Santa Cruz Biotechnology, Heidelberg, Germany). Ki67 (IR626, DAKO Agilent, Santa Clara, CA, USA), and DakoEnVision Flex (Linker) (DM824, DAKO Agilent, Santa Clara, CA, USA) were also used. The investigation was conducted following the ethical standards according to the Declaration of Helsinki of 1975 and according to the local, national, and international guidelines. Tissue samples were obtained from patients undergoing surgical resection of HNC at the Department of Oral and Maxillofacial Surgery at the Faculty of Dentistry of Alexandria University from December 2017 to November 2018. In that period, the cases were consecutively enrolled in the study. The study protocol was approved by the local ethics committee (#0008839) after obtaining the patient’s informed consent to participate in the study and was processed anonymously. Patients undergoing simultaneous chemo- or radio-treatment before or during the surgery were excluded from the study. All cases were diagnosed histopathologically as HNC and staged according to the TNM classification of malignant tumors recommended by the ‘Union International Contre le Cancer UICC (8th edition). All experiments were performed in accordance with the relevant laws and the Alexandria University Guidelines and approved by the institutional ethics committee at the Faculty of Dentistry, Alexandria University. In this study, tumor specimens and corresponding non-malignant tissue were analyzed, different tumor sizes (T1–T4), lymph node status (N0-2), and grading G1–G3. Upon resection, samples were immediately fixed in formaldehyde. Histological analyses were performed to ensure that each specimen contained >70% tumor tissue and <10% necrotic debris. Samples not meeting these criteria were rejected. Specimens were handled as usual (i.e., paraffin-embedded, sectioned, and H&E staining). The H&E stain was implemented by staining the specimens with Harris’ hematoxylin as described [83,90]. The interpretation was performed by oral pathologists at Alexandria University. Peripheral blood samples were taken from patients before or during surgery. The total serum 25-hydroxyVitD concentration (sum of D3 & D2 forms) is regarded as the best single marker of VitD status in the human body. Total serum VitD (25-hydroxyVitD3) was quantified by using fully validated, modified high-performance liquid chromatography (HPLC) [47]. HNC tissue samples were included from The Cancer Genome Atlas (TCGA) Research Network (http://cancergenome.nih.gov/, accessed on 1 October 2022). The TCGA Research Network included patients following the guidelines of the Declaration of Helsinki of 1975 and all patients provided signed informed consent. Publicly available gene expression and survival datasets were obtained from The Cancer Genome Atlas (TCGA) Research Network (http://cancergenome.nih.gov/, accessed on 1 October 2022), filtering for patients with HNCs (TCGA HNC). Of note, the expression values were not detectable for all genes of interest for every patient in the TCGA database. Here, VDR and RXR expression was found in n = 604 patients and analyzed as described [39]. Data were assessed via the USCS Xena server and patients were grouped according to the indicated phenotypic or clinical characteristics as described [37]. Authenticated and characterized cell lines FaDu and SCC-4 were purchased from the ATCC repository, expanded, stocks prepared at early passages, and frozen stocks kept in liquid nitrogen. SCC-4 cells were established from a tongue squamous cell carcinoma. HNCUM-01T and -HNCUM-02T were established from tongue squamous cell carcinoma as described by Welkoborsky et al. [91]. The Pica cell line was established from laryngeal squamous cell carcinoma and maintained as described [39]. The FaDu cell line was established from a hypopharyngeal squamous cell carcinoma [92]. Thawed cells were routinely monitored by visual inspection and growth-curve analyses to keep track of the cell-doubling times, and were used for a maximum of 20 passages for all experiments. Depending on the passage number from purchase, cell line authentication was further performed at reasonable intervals by short tandem repeat (STR) profiling. We cultured the HNCUM-01T, HNCUM-02T, and SCC-4 cells in Dulbecco’s modified Eagle’s F-12 medium. Pica and FaDu cells were cultured in Dulbecco’s modified Eagle’s medium. We added 10% fetal bovine serum (FBS), and 1% penicillin-streptomycin to all medium types. Cells were cultured under a 5% CO2 atmosphere at 37 °C and subcultured every 3 days as described [39]. We checked the absence of mycoplasma regularly via the Venor GeM Advance Detection Kit (Minverva Biolabs, Berlin, Germany) according to the manufacturer’s instructions. The cell numbers were determined using Casy Cell Counter and Analyzer TT (OMNI Life Science GmbH & Co KG, Bremen, Germany). To treat the cells, Hy-clone fetal bovine serum (FBS) (Sigma Aldrich, Munich, Germany) was used instead of standard FBS to ensure the absence of VitD in the controls and the control VitD treatment doses in the treated samples. We generated constantly selected cell lines by treatment with sub-toxic doses of cisplatin corresponding to IC90 (5 µM) and then constant treatment (3 µM). We used the resistant cell line for experiments 6 months after constant exposure to cisplatin and the re-establishment of relatively regular proliferation. To probe cell viability, we seeded the cells in 96-well plates (5000 to 15,000 cells/well) depending on the cell line and the treatment duration and treated them with the indicated substances and concentrations (n = 3) starting 24 h after seeding. After 48/72 h treatment, we performed a commercially available assay CellTiter-Glo® 2.0 (Promega, Walldorf, Germany) according to the manufacturer’s instructions and recorded the luminescent signals using a Tecan Spark® (Tecan Group Ltd., Männedorf, Switzerland). Later, we normalized the signals to the untreated control samples. In order to objectively determine the pharmacological effect of the proposed drug combination, we used the combination index equation described by Chou–Talalay [53]. In this algorithm, synergy is defined as combination index values < 1.0, antagonism as values > 1.0, and additivity as a value = 1.0. Fluorescence images were acquired, analyzed, and quantified using an Axiovert 200 M fluorescence microscope (Zeiss, Oberkochen, Germany) or an automated high-content screening microscope Array Scan VTI (Thermo Fisher, Dreieich, Germany) as described [39,93,94]. We seeded cells in microscopic dishes (35 mm, MatTek, Ashland, MA, USA) or clear-bottom 96-well plates (Greiner, Kremsmünster Austria) and fixed them with 4% PFA (20 min, RT). For immunofluorescence staining, we additionally permeabilized the cells via incubation with Triton-X 100 (0.1%, 10 min, RT). Antibodies were diluted in 10% FBS/PBS and incubated with samples for 1 h at RT. We washed the cells (n = 3) in PBS and then incubated the samples with fluorophore-labeled antibodies for 1 h at RT. Finally, we stained the nuclei by adding Hoechst 33342 (50 ng/mL in PBS) for 30 min at RT. For automated high-content screening, regions of interest were created using the nucleus signal and each sample was acquired in triplicate, imaging at least 5000 events per sample according to [39]. RNA sequencing was then performed as described in [95] and the visualizations were achieved with the help of GraphPad Prism. Ingenuity Pathway Analysis (Qiagen, Hilden, Germany) was used to visualize the mTOR and PI3K-AKT signaling pathways. To construct a VDR expression plasmid, cDNA was isolated out of the HNC cancer cell lines, and the full open reading frame of human VDR cDNA was cloned into the pcDNA3.1 mammalian expression vector (Invitrogen, Karlsruhe, Germany) with the C-terminal GFP-tag (for primer sequences, please see Supplementary Table S2). Colony PCR was performed to check for positive clones [93,94,96]. For cellular transfection, plasmid DNA and Lipofectamine 3000 (Fisher Scientific, Schwerte, Germany) were mixed according to the manufacturer’s instructions and added to the cells, which were cultured in Opti-MEM medium as described [97].To mark VDR-expressing cells, plasmid pC3 coding for GFP expression was co-transfected. To exclude artifacts, a control transfection of empty plasmid pC-DNA3 and the GFP-coding plasmid was conducted in parallel. The medium was changed 5 h post-transfection to a normal cell culture medium. We confirmed the VDR overexpression of cell lines via Western blot analysis, and positively transfected cells were selected by the addition of puromycin (1 µg/mL; Sigma Aldrich, Munich, Germany). To establish a uniform expression of the VDR transfected cells, the cells were sorted into low, medium, and high fluorescence using FACS as previously described [96]. Whole-cell lysates were prepared using low salt lysis RIPA buffer (50mM Tris pH8.0, 150 mM NaCl, 5 mM EDTA, 0.5% NP-40, 1 mM DTT, 1 mM PMSF, Complete EDTA-free from Roche Diagnostics, Mannheim, Germany) and samples were separated on 8–12% SDS gels, as has previously described [96,98,99]. Blotting onto activated PVDF membranes was achieved with Trans-Blot Turbo (Bio-Rad, Munich, Germany) and blocking and antibody incubations (1 h/RT or 16 h/4 °C depending on antibody) were performed in 5% milk powder or BSA in TBST or PBS. The detection of the luminescence signal of HRP-coupled secondary antibodies after the addition of Clarity Western ECL Substrate was performed using the ChemiDocTM imaging system (Bio-Rad). Equal loading of lysates was controlled by reprobing blots for housekeeping genes (Actin). At least n = 2 biological replicates were performed and representative results are shown. Results of the densitometric analyses of all Western blots can be found in the Supplementary Materials. Statistical analyses were performed using GraphPad Prism (version 9.3.1) as described [39]. Survival data were obtained from the USCS Xena server, visualized, and analyzed by GraphPad Prism (Log-rank/Mantel-Cox test; Hazard Ratio (Mantel-Haenszel)). For two groups, a paired or unpaired Student’s t-test, for more group analysis of variance (ANOVA) was performed. Unless stated otherwise, p values represent data obtained from two independent experiments conducted in triplicate. Statistical significance is represented in the figures as follows: * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, and n.s. indicates not significant. A p-value that was less than 0.05 was considered statistically significant.

PMC10002567

Erika Richter,Thangiah Geetha,Donna Burnett,Tom L. Broderick,Jeganathan Ramesh Babu

The Effects of Momordica charantia on Type 2 Diabetes Mellitus and Alzheimer’s Disease

28-02-2023

bitter melon (M. charantia),type 2 diabetes mellitus,Alzheimer’s disease,glucose-lowering effects,hypoglycemic,medicinal plants,natural,neuroprotective,prevalence,therapeutic,bioactive

T2DM is a complex metabolic disorder characterized by hyperglycemia and glucose intolerance. It is recognized as one of the most common metabolic disorders and its prevalence continues to raise major concerns in healthcare globally. Alzheimer’s disease (AD) is a gradual neurodegenerative brain disorder characterized by the chronic loss of cognitive and behavioral function. Recent research suggests a link between the two diseases. Considering the shared characteristics of both diseases, common therapeutic and preventive agents are effective. Certain bioactive compounds such as polyphenols, vitamins, and minerals found in vegetables and fruits can have antioxidant and anti-inflammatory effects that allow for preventative or potential treatment options for T2DM and AD. Recently, it has been estimated that up to one-third of patients with diabetes use some form of complementary and alternative medicine. Increasing evidence from cell or animal models suggests that bioactive compounds may have a direct effect on reducing hyperglycemia, amplifying insulin secretion, and blocking the formation of amyloid plaques. One plant that has received substantial recognition for its numerous bioactive properties is Momordica charantia (M. charantia), otherwise known as bitter melon, bitter gourd, karela, and balsam pear. M. charantia is utilized for its glucose-lowering effects and is often used as a treatment for diabetes and related metabolic conditions amongst the indigenous populations of Asia, South America, India, and East Africa. Several pre-clinical studies have documented the beneficial effects of M. charantia through various postulated mechanisms. Throughout this review, the underlying molecular mechanisms of the bioactive components of M. charantia will be highlighted. More studies will be necessary to establish the clinical efficacy of the bioactive compounds within M. charantia to effectively determine its pertinence in the treatment of metabolic disorders and neurodegenerative diseases, such as T2DM and AD.

The Effects of Momordica charantia on Type 2 Diabetes Mellitus and Alzheimer’s Disease T2DM is a complex metabolic disorder characterized by hyperglycemia and glucose intolerance. It is recognized as one of the most common metabolic disorders and its prevalence continues to raise major concerns in healthcare globally. Alzheimer’s disease (AD) is a gradual neurodegenerative brain disorder characterized by the chronic loss of cognitive and behavioral function. Recent research suggests a link between the two diseases. Considering the shared characteristics of both diseases, common therapeutic and preventive agents are effective. Certain bioactive compounds such as polyphenols, vitamins, and minerals found in vegetables and fruits can have antioxidant and anti-inflammatory effects that allow for preventative or potential treatment options for T2DM and AD. Recently, it has been estimated that up to one-third of patients with diabetes use some form of complementary and alternative medicine. Increasing evidence from cell or animal models suggests that bioactive compounds may have a direct effect on reducing hyperglycemia, amplifying insulin secretion, and blocking the formation of amyloid plaques. One plant that has received substantial recognition for its numerous bioactive properties is Momordica charantia (M. charantia), otherwise known as bitter melon, bitter gourd, karela, and balsam pear. M. charantia is utilized for its glucose-lowering effects and is often used as a treatment for diabetes and related metabolic conditions amongst the indigenous populations of Asia, South America, India, and East Africa. Several pre-clinical studies have documented the beneficial effects of M. charantia through various postulated mechanisms. Throughout this review, the underlying molecular mechanisms of the bioactive components of M. charantia will be highlighted. More studies will be necessary to establish the clinical efficacy of the bioactive compounds within M. charantia to effectively determine its pertinence in the treatment of metabolic disorders and neurodegenerative diseases, such as T2DM and AD. Diabetes mellitus is considered one of the five leading causes of death in the world affecting an estimated 537 million adults in the year 2021 [1,2]. By 2045, this number is expected to rise to 783 million worldwide [2,3,4]. There are two major types of DM, the first being type 1 diabetes (T1DM), characterized by hyperglycemia due to the autoimmune destruction of pancreas beta cells, resulting in the overall decreased production of insulin [3,4,5]. The more common version of this metabolic disorder is type 2 diabetes mellitus (T2DM), described by increased insulin release to compensate for insulin resistance and progressive decline in islet secretory function within the pancreas, thus causing overall insulin deficiency [3,4]. Several complications can result from diabetes, including nephropathy, retinopathy, neuropathy, and atherosclerosis [3,4]. About 60–70% of diabetes patients will eventually develop peripheral neuropathy if proper care and treatment do not take place [6]. If left untreated, these complications may become life-threatening and lead to heart disease, kidney disease, blindness, and in some cases amputations. Recent epidemiological research proposes the likely association between the development of Alzheimer’s disease (AD) and diabetic neuropathy arising from T2DM [7]. In fact, AD is now being referred to as “type 3 diabetes”, as it is considered a form of diabetes that selectively involves the brain and has molecular and biochemical features that intersect with both T1DM and T2DM [8]. AD is characterized as the gradual deterioration of the brain, gravely impairing neurological function leading to symptoms of early onset dementia and impacting one’s ability to carry out simple everyday tasks [9]. An estimated 5.8 million Americans were living with AD in 2020 [5]. Due to rampantly increasing rates of AD, coupled with high economic and social stress, it is imperative to secure successful, yet practical treatment options to effectively delay the symptoms and development of both T2DM and AD [3]. Antidiabetic drugs have various adverse effects; however, there has been no success in finding a treatment that effectively prevents or reverses AD progression [7,9]. As AD has no cure, it is essential to determine whether any antidiabetic drugs or therapies hold any significant improvement toward the progression or prevention of the disease, which is where nutrition plays a vital role [10,11]. Considering the underlying biochemical associations between AD and T2DM, it is important to thoroughly investigate preventative measures from a nutritional standpoint, as there may be a common therapeutic target for AD and T2DM within the bioactive components of certain medicinal foods [12,13]. These components found within nature can be found in several plants, many of which are consumed in the everyday diet across the globe [12,13]. Through the consumption of a variety of different fruits, vegetables, plants, and animal proteins, an array of vitamins, minerals, and phytochemicals are provided for the body. A recent study has unveiled that an estimated 30% of patients with DM use alternative medicine as a way of controlling their metabolic state [3,14]. Therefore, identifying the bioactive components within these plants that aid in the control or delay of diseases such as AD and T2DM is the next step in finding affordable and preventative medicine. With food functioning as medicine, the prevention of these diseases is attainable from both an economic and scientific standpoint. Identifying the biochemical elements within these nutriments which hold such curative power is the first step in establishing nutrition as a modern, effective medicine. This tactic is otherwise known as complementary and/or alternative medicine and utilizes herbs and other dietary subsistence from various plants and animals secondary to traditional medical treatments [3]. Through further research on the bioactive components within food, modern-day medicine can expand its treatment approach towards metabolic disease states such as T2DM and AD. Further, the natural bioactive compounds within foods can be utilized as an accessible, promising dietary treatment option for DM and AD due to the efficacy, few side effects, and availability of the nutritional components [15]. M. charantia, otherwise known as bitter melon, karela, balsam pear, or bitter gourd, is a popular plant commonly used to treat diabetes-related conditions amongst the indigenous populations of Asia, South America, India, the Caribbean, and East Africa [3,16,17]. This odd gourd is known for its distinguishing bitter taste, which becomes more distinct through maturation [3]. Through several biochemical and animal model experiments, an ample amount of data and hypotheses have been procured regarding the antidiabetic effects of M. charantia, although very few clinical human studies with adequate design quality have been published [3]. However, the effects of M. charantia on the treatment of AD have yet to be adequately explored. As the connection between DM and AD grows seemingly more intertwined, nutritional preventative care needs to be at the forefront of research. Further clinical research with a heightened focus on the nutritional aspect of medicinal foods such as bitter melon cannot only help expand preventative treatment options for the population but also help alleviate the economic burden these metabolic diseases create today. Gaining specific knowledge regarding the bioactive components found in M. charantia would provide the necessary insight for the proper strategic consumption and sustainable use needed to enhance the treatment of both T2DM and AD [3]. By further understanding the capabilities of the nutritive elements within various plant species, their potential role in overall health strengthens allowing nutrition to work in congruence or replace common T2DM and AD pharmaceuticals. Thus, utilizing the antidiabetic and hypolipidemic effects of M. charantia in congruence with modern medicine may be an effective option to best treat DM as well as delay further complications of the disease, including the progression of AD [3]. Furthermore, investigating the various antioxidant, anti-inflammatory, and antiapoptotic properties of M. charantia may help shed light on the importance of incorporating nutrition as part of preventative care as a whole [1,15,18]. Through the assessment of the various beneficial effects of bioactive compounds found in M. charantia and other medicinal plants, their impact on overall health can be better understood. By doing so, medical nutrition therapy increases the number of effective methods readily available for the treatment of metabolic disease states making large strides in advancing medicine for the benefit of the overall health and well-being of the population [1]. Throughout this paper, the overall medicinal impact of M. charantia will be further understood through a discussion of the pathophysiology of T2DM and AD, and how they relate to one another, as well as the underlying mechanisms of the bioactive components that make up M. charantia. Through this detailed review, the potential antidiabetic, and hypoglycemic effects of M. charantia and its medicinal potency in effectively treating T2DM and AD will be thoroughly discussed. To fully understand the beneficial effects of the bioactive compounds of bitter melon on T2DM and AD, we must first discuss the pathophysiology of both diseases [3]. Insulin resistance coupled with an overproduction of hepatic glucose, along with dysfunctional β-cell activity, thus leading to β-cell deficiency characterizes the onset of T2DM [19,20]. Peripheral insulin resistance is attributed to the subsequent failure of cells to efficiently react to changing insulin levels within the various target tissues within the body [21]. In the body’s normal state, insulin will suppress the production of glucose from the liver during both fasting states and after a meal is consumed [22]. However, in the case of insulin resistance, glucose levels tend to increase after a meal rather than becoming corrected back down to a normal postprandial state causing hyperglycemic episodes indicating T2DM [22]. Another major factor contributing to elevated hepatic glucose production is heightened lipolysis within fat cells [23]. Initially, insulin resistance will trigger compensatory ß-cell hypertrophy and a rise in insulin production; however, continued subjection to hyperglycemic-induced oxidative stress, inflammatory markers, and immune-suppressing cell behaviors may add to the subsequent cell death and diminished cell expansion leading to overall β-cell destruction [24,25]. This continuous decline of β-cell performance eventually reduces insulin secretion, thus disrupting glucose homeostasis [25]. There are several pharmacotherapy treatment options available to control T2DM; however, recent studies also suggest that healthy lifestyle choices may prove to be more effective at prevention and treatment in the long run [26]. AD is a progressive neurological disease characterized by brain atrophy and subsequent dementia affecting one’s overall quality of life and ability to function independently [8]. Early signs of the disease include short-term memory loss and, as the disease progresses, serious disruptions in the ability to carry out simple everyday tasks ensue [8]. AD can be distinguished from other metabolic disease states by the accumulation of extracellular misfolded amyloid plaques (Aβ peptide) in senile plaques, intracellular neurofibrillary tangles (NFTs), inflammation of neurotransmitters, and the deterioration of specific areas within the cerebrum [8,13,27]. Aβ peptides are made up of 38–43 amino acid residues originating from a chemical change in the amyloid precursor protein (APP) [19,28]. In normal states, APP utilizes an α-secretase to create Aβ products that tend to not produce amyloid deposits [19,28,29,30]. However, in a brain negatively affected by AD, BACE-1 (a β-secretase), and γ-secretase assemble Aβ from APP through several overriding enzymatic steps [19,28,29,30]. The failure to remove excess Aβ, therefore, gives rise to the aggregation of Aβ and the ensuing damage to the nervous system as a whole [8,30]. Following this event, the Aβ’s will then spontaneously manifest into various configurations, including undesirable monomers, oligomers, fibrils, and plaques [19,31]. Another common indicator of an AD diagnosis is the detection of NFT within the brain [29]. Although medications may temporarily improve or slow the progression of symptoms, there has yet to be a successful treatment that fully cures this deadly disease [8]. Therefore, it is imperative to look beyond traditional medications to make meaningful strides in medicine. Although T2DM and AD seem like two completely different disease states with entirely different sets of symptoms and complications, this is not the case. The biochemical, molecular, and cellular abnormalities that precede or accompany AD neurodegeneration, including heightened activation of predeath genes and signaling pathways, impaired energy metabolism, mitochondrial dysfunction, chronic oxidative stress, and DNA damage, are all recognizable detriments with the exact etiologies remaining unknown [8,32,33,34,35,36,37,38,39]. Recent evidence suggests a role of impaired cerebral glucose utilization and energy metabolism, which represent very early abnormalities that coincide with the initial stages of cognitive impairment in AD [8,40,41,42]. This finding has led researchers to believe that impaired insulin signaling has an important role in the pathogenesis of AD, thus its potential representation as “type 3 diabetes” [8,33]. Therefore, both T2DM and AD are intertwined through several characteristics, including chronic inflammation, oxidative stress, impaired insulin signaling, insulin resistance, glucose intolerance, and cognitive impairment [12]. By further understanding what links the two of these disease states, we will be able to better understand the underlying mechanisms of the bioactive components of plants such as M. charantia and their impact on both T2DM and AD. Recent emerging evidence has revealed that the markers of insulin resistance and insufficiency, prevalent in T2DM, are also contributors to the pathology of AD [19,43]. Consequently, AD is denoted as a form of DM targeting the nervous system [13,19]. Insulin receptors (IR) are expressed in both the peripheral and central nervous systems (CNS), which include a region of the brain called the hippocampus [19,44,45]. The hippocampus is largely responsible for learning and memory function and is most often the earliest affected area in AD neurological degradation [19,44,45]. This occurs due to the attachment of insulin to IR leading to the phosphorylation of tyrosine and the subsequent stimulation of insulin receptor substrate (IRS), thus enhancing the activity of both Akt and phosphatidylinositol-3 kinase (PI3 kinase) [19,28]. Akt will then mediate the phosphorylation or deactivation of glycogen synthase kinase 3β (GSK3β) [19,28]. This results in lowered insulin signaling causing an increased GSK3β activity [19,28,46]. This phenomenon causes the hyperphosphorylation of tau proteins, the emergence of NFTs, and heightened assembly of Aβ within the cerebrum, as illustrated in Figure 1 [19,28,46]. In a brain affected by AD, Aβ oligomers accelerate atypical activation of tumor necrosis factor-α (TNF-α)/c-Jun N-terminal kinase pathway (JNK), thus bringing about the inhibition of IRS1 and the overall disturbance in insulin regulation [12,19,47]. Furthermore, the insulin-degrading enzyme (IDE) regulates the breakdown of Aβ and APP, creating competition between insulin and Aβ [19,48], and therefore reducing Aβ destruction altogether regarding insulin resistance [18,19]. T2DM accompanied by IR and hyperglycemia increases the risk of potential metabolic issues within the brain and other tissues that initiates a cascade of pathogenic processes such as oxidative stress, inflammatory responses, advanced glycation products, and autophagic dysfunction. The ROS cultivated by these pathways heightens the process of neuronal death. Simultaneously, the IR disturbs the signaling pathways and causes an increased formation of Aß oligomers and aggregates of hyperphosphorylated tau. The overall effect of all these factors allows neurons to face serious degradation, eventually resulting in the loss of synapses and neuronal death (Figure 1). Chronic inflammation is another pertinent link between T2DM and AD [19]. Elevated erythrogenic cytokines, including interleukin-6 (IL-6), interleukin-1β (IL-1β), and TNF-α are recognized markers of T2DM [19,49]. The emergence of these inflammatory proteins is correlated with β-cell destruction and death, as well as diminished insulin production, all of which negatively affect the pancreas and/or brain [19,50,51]. Certain proinflammatory cytokines also have been found to cross the blood–brain barrier (BBB), causing harmful effects on the CNS, thus leading to the postulated induction and progression of AD [19,52]. In a recent study, higher amounts of IL-1 within the cerebrum led to limited acetylcholine (Ach) delivery, decreased nerve growth factor (NGF) expression, and problems with memory recollection [19,52,53]. Another critical aspect of chronic inflammation involves advanced glycation end products (AGE), produced through spontaneous non-enzymatic reactions of free reducing sugars with free amino groups of various proteins and lipids [19,54]. In cases of chronic hyperglycemia seen in T2DM, AGEs are further generated by interacting with RAGE receptors, prompting the activation of various intracellular erythrogenic pathways such as the nuclear factor-kappa B (NF-κB) and proinflammatory mediators such as IL-6, TNF-α, and C-reactive protein (CRP) [19,54,55]. These AGEs noticeable in T2DM may also be involved in AD pathology due to Aβ attachment to RAGE [3]. This occurrence promotes Aβ crossing through the BBB, causing Aβ inflation linked to AD development [19,56,57]. Additionally, this interaction relates to elevated levels of oxidative stress, magnified inflammatory response, and severe damage to neuronal cells [19,58]. Oxidative stress ensues from a chemical disparity between the making of reactive oxygen species (ROS) and reactive nitrogen species (RNS), as well as antioxidant defense, thus attributing to the rise of T2DM and AD development [19,59]. The disruption of crucial polyunsaturated fatty acids (PUFAs), proteins, and DNA within cell membranes can be attributed to some of these ROS including nitric oxide synthase (iNOS), cyclooxygenase-2 (COX-2), and hydroxyl radicals [19,60]. One of the main factors contributing to the ongoing development of T2DM and its complications includes the overproduction of these ROS/RNS entities [4,19]. In cases of T2DM, elevated glucose levels may promote glucose autoxidation, mitochondrial defects, and heightened levels of ROS [3,53]. Excess ROS promotes lipid peroxidation causing subsequent β-cell failure and weakened biochemical pathways, such as NF-κB, JNK/stress-activated protein kinase (SARK), and p38-mitogen-activated protein kinase (p38-MAPK), thus contributing to IR and metabolic complications [19,61,62]. These factors hold a significant role in the prognosis of AD as well [19,61,62]. Since neurons heavily rely on mitochondria for ATP production and maintenance of Ca2+ homeostasis, any oxidative stress on these mitochondrial pathways may result in neuronal injury and death due to bioenergetic depletion [19,63]. Moreover, mitochondrial damage can also enhance ROS production, which then amplifies NFT development, tau hyperphosphorylation, and Aβ aggregation, which promote the overall generation of oxidative stress [19,61]. Together, these factors accelerate the progression of AD, thus making it imperative to find an effective treatment that prevents oxidative stress [19]. By further investigating the common biochemical abnormalities that T2DM and AD share, a common therapeutic and preventive agent that may be effective in treating both diseases can be found [3]. The medicinal properties of M. charantia may be a potential solution to preventing and slowing down the progression of insulin resistance, chronic inflammation, and oxidative stress that occurs in these metabolic disorders. For thousands of years, M. charantia has been used as an herbal remedy throughout many countries and regions, as its components provide impactful beneficial effects [64]. The plant itself, especially its fruit and seeds, contains significant pharmacological effects that have been utilized in the treatment of DM as well as AD as recent evidence suggests its plausible role in its pathology [64,65]. The phytoconstituents within bitter melon activate antihyperglycemic, antioxidant, antidiabetic, hepatoprotective, antibacterial, and anti-inflammatory activities in tissues and cellular pathways [64,65,66,67,68,69,70]. Plant-based medicine is a cost-effective, safe, and practicable option to treat T2DM and AD [3]. In fact, in many developing parts of the world, this may be the only therapeutic option for medical treatment [3]. There are several studies reviewing the effectiveness of antidiabetic herbal plants in comparison to modern-day forms of medicine [71,72,73,74,75,76,77]. The ancient Indian medical system, otherwise known as Ayurveda, utilizes several plants and herbs for the treatment of disease states such as T2DM and AD, as this system is based on a natural and holistic approach to both physical and mental overall health [3]. Although this method is not considered modern medicine, it is a system that produces few side effects at a low cost [3]. Current diabetes medications utilize the insulin and oral hypoglycemic effects within M. charantia to control T2DM and other metabolic conditions. Metformin, one of the most popular diabetic medications, is one of them. This medication lowers blood sugar levels by improving insulin sensitivity. M. charantia maintains the same hypoglycemic effects as metformin; it promotes insulin secretion, improving glucose uptake by adipose or muscle tissues, and inhibiting glucose absorption from the intestines and glucose production from the liver [78], whereas consuming M. charantia or its extracts can present the same effects with far fewer side effects [3]. Current research exemplifies the potential use of M. charantia as a more recognizable tool for T2DM and AD prevention and treatment, as it is the most accepted hypoglycemic plant [3]. M. charantia maintains hypoglycemic effects through antioxidant effects promoting ß-cell protection, the decrease in glucose absorption from the intestines and glucose production from the liver, and improved glucose uptake by adipose or muscle tissues, thus affecting the brain as well due to glucose being its main energy source as illustrated in Figure 2 [79]. This paper will focus on M. charantia and its bioactive components in relation to its ability to control, treat, and possibly prevent both T2DM and AD, a seemingly interconnected disease state due to impaired insulin signaling. Although bitter melon’s potential as a replacement therapy for traditional medicine still needs further clinical research, the result of current research looks promising [3]. With a high vitamin and mineral content, M. charantia is one of the most promising medicinal plants available for not only the treatment of T2DM and AD but for overall human health and longevity [3,29]. Distinguishable by its unusual taste, bitter melon may be further explained by its various bioactive effects that general fruits and vegetables do not provide [64]. Though there are several herbal remedies that claim to effectively treat T2DM and AD, M. charantia is one that has received a considerable amount of attention [3]. Traditionally, bitter melon has been known for its effectual antidiabetic, anticancer, anti-inflammation, antivirus, and cholesterol-lowering medicinal properties [3,64]. Antioxidant and antimutagen properties have also been identified amongst M. charantia’s of bioactive and phenolic compounds [3,15,44]. M. charantia, otherwise known as bitter melon or bitter gourd, is a member of the Cucurbitaceae family distinctly known for its intensely bitter taste, as shown in Table 1 [3]. This flowering vine is widely cultivated in tropical and subtropical regions of the world such as Asia, India, East Africa, and South America [3,64]. This climbing perennial can grow up to 5 m tall and produces oblong or spindle-shaped fruits with knobby, protruding bumps along the surface [3,28,68,69]. Resembling a small cucumber, bitter melon’s looks deceive. Its exterior starts out a deep emerald green and as it ripens turns bright orange, while the inside of the fruit changes to bright red during maturation [64]. The fruit itself can be incorporated into the diet in every stage of maturation and can be found in several popular dishes [64]. In Asian culture, bitter melon is incorporated into recipes with knowledge of its medicinal capability, whereas in Western culture the notion regarding bitter melon is not as recognized. As the population becomes more aware of the benefits of a well-rounded diet, trends in overall health will reap the benefits. Thus, through a nutritious diet with foods such as M. charantia, certain metabolic disease states such as T2DM and AD can be better controlled, treated, and prevented. M. charantia is a potent nutrient-dense plant composed of an elaborate range of beneficial compounds and elements [3]. The powerful assembly of bioactive compounds, vitamins, minerals, and antioxidants within this gourd all give rise to its remarkable ability in treating a wide range of illnesses [3]. Bitter melon contains large amounts of vitamins C, A, E, B1, B2, and B3, as well as vitamin B9 (folate), making this vegetable a healthful addition to any diet [3]. Regarding caloric content, the values for the leaves, fruit, and seeds are approximately 213, 242, and 177 Kcals per 100 g [80]. Bitter melon is also rich in many minerals including potassium (K), calcium (Ca), zinc (Zn), magnesium (Mg), phosphorus (P), and iron (Fe), and is an excellent source of dietary fiber [3]. A diet rich in vitamins, minerals, and fiber can help promote the overall health and well-being of the general population, especially those at risk of developing metabolic disorders, such as T2DM and AD. The medicinal effects of M. charantia can be explained by its high antioxidant properties due in part to components such as phenols, flavonoids, isoflavones, terpenes, anthraquinones, and glycosylates, all of which contribute to its formidable bitter taste [3,81]. All these properties combined are what give bitter melon such therapeutic potential for treating and preventing T2DM and AD, as well as providing adequate nutrition, vitamins, and minerals to those who incorporate this unusual vegetable into their diet [Figure 3]. M. charantia is made up of several carbohydrates, proteins, and lipids that all possess different bioactive components that prove to be beneficial for overall health and specific ailments for certain metabolic diseases [64]. Within these classifications lie various key bioactive components such as triterpenoids, saponins, polypeptides, flavonoids, alkaloids, and sterols which are the compounds ultimately responsible for these medicinal effects that bitter melon is so well known for shown in Table 2 [3,64,82,83]. M. charantia also consists of glycosides, saponins, reducing sugars, resins, phenolic constituents, fixed oil, and free acids [3,156]. M. charantia consists of the following chemical constituents: alkaloids, charantin, charine, cryptoxanthin, cucurbitins, cucurbitacins, cucurbitanes, cycloartenols, diosgenin, elaeostearic acids, erythrodiol, galacturonic acids, gentisic acid, goyaglycosides, goyasaponins, guanylate cyclase inhibitors, gypsogenin, hydroxytryptamines, karounidiols, lanosterol, lauric acid, linoleic acid, linolenic acid, momorcharasides, momorcharins, momordenol, momordicilin, momordicin, momordicinin, momordicosides, momordin, momordolo, multiflorenol, myristic acid, nerolidol, oleanolic acid, oleic acid, oxalic acid, pentadecans, peptides, petroselinic acid, polypeptides, proteins, ribosome-inactivating proteins, rosmarinic acid, rubixanthin, spinasterol, steroidal glycosides, stigmasta-diols, stigmasterol, taraxerol, trehalose, trypsin inhibitors, uracil, vacine, v-insulin, verbascoside, vicine, zeatin, zeatin riboside, zeaxanthin, zeinoxanthin amino acids-aspartic acid, serine, glutamic acid, thscinne, alanine, g-amino butyric acid and pipecolic acid, ascorbigen, b-sitosterol-d-glucoside, citrulline, elasterol, flavochrome, lutein, lycopene, and pipecolic acid. The pulp within bitter melon contains soluble pectin but no free pectic acid [3]. Recent evidence indicates that the leaves of bitter melon are significant nutritional sources of Ca, Mg, K, P, and Fe. It was also demonstrated that both the edible portion of M. charantia, along with the leaves are excellent sources of B vitamins, which are significant in overall health and longevity [3,79]. Further understanding the phytochemical composition of the bioactive elements within M. charantia will allow the specific role and impact of each constituent to be easily identified. As research continues to isolate the components of M. charantia, the more likely this “superfood” is to be utilized as a recognized method of medical nutrition therapy for patients suffering from various metabolic disorders. By doing so, bitter melon can be used as a cost-effective method in the treatment, prevention, and control of both T2DM and AD. Bioactive compounds are the components within foods that can regulate the metabolic processes in both humans and/or animals and improve overall health [19,62]. These bioactive components are found largely in vegetables, fruits, and whole grains, which can be consumed daily with a healthy diet that consists of a variety of different foods [20,64]. These bioactive components have the potential to regenerate pancreatic β cells, enhance insulin release, and reverse insulin resistance, all of which are desirable in the control and prevention of several metabolic disease states [77]. The beneficial effects of various bioactive compounds within M. charantia have been found to diminish inflammation, target free radicals, and regulate cell signaling pathways in both cell and animal studies [19,63,64]. Due to their rich availability, safety, and low amount of side effects, the use of bioactive compounds has been reported to lessen the occurrence or delay the progression of several diseases, including T2DM and AD [15,18,19]. Some examples of bioactive compounds include polyphenols, carotenoids, phytosterols, polysaccharides, and vitamins [19]. The role of major compounds that have been isolated from bitter melon and identified as hypoglycemic agents include polysaccharides; proteins and peptides such as polypeptide-p, and peroxidase; saponins and terpenoids such as charantin; and flavonoids and phenolic compounds such as quercetin, rutin, kaempferol, and isorhamnetin —[3,64,84,157], Table 2. These bioactive constituents within bitter melon provide it with medicinal properties, thus making it essential to fully assess their mechanisms of action in detail to successfully utilize the plant and its extracts in actual, medicinal practice. For AD in particular, recent research has revealed the neuroprotective impact of bioactive compounds within natural products, such as bitter melon, through its influence on the peripheral metabolism affecting the cognitive function of the brain, whereas traditional drugs must target crossing the BBB to be effective [158]. Additionally, when a drug is ineffective, it is likely due to its inability to pass the BBB [158]. Thus, further research on the bioavailability of natural products, such as M. charantia, is essential in determining its impact on the CNS and its therapeutic potential on AD. The bioactive compounds within natural products have an impact on brain function but work differently than typical medications as they do not necessarily cross the BBB [158]. This next section will discuss the effects of each of these bioactive compounds within the inner workings of metabolism and the CNS, along with their mechanisms of action. Polysaccharides are carbohydrates consisting of small sugar molecules that formulate the starch, cellulose, or glycogen stores of a plant, and are one of the more important bioactive components of M. charantia [64]. Carbohydrates are not only utilized as an energy source for the brain, but also heavily influence biological interactions affecting cell growth, signaling, and differentiation. From this information, research has revealed the potential of the natural carbohydrates from the fruit of M. charantia to possess various antioxidant, antidiabetic, immune, neuroprotective, antitumor, and antimicrobial bioactivities [141,142,143,144,145,146,147]. Polysaccharides within bitter melon are classified as heteropolysaccharides, composed of galactose (Gal), glucose (Glu), arabinose (Ara), rhamnose (Rha), and mannose (Man) [148]. The contents of M. charantia’s polysaccharides were shown to be influenced by several conditions and classified into two main fractions [149]. One fraction of the M. charantia polysaccharide consisted of an acidic, branched heteropolysaccharide (MCBP) mainly composed of Man, galacturonic acid (GalA), Rha, Glu, Gal, xylose (Xyl), and Ara, whereas the rest of the fraction consisted of a pectic polysaccharide (PS) [147,150]. The MCBP fraction possessed antioxidant, α-amylase inhibition, and angiotensin-converting enzyme (ACE) inhibition functions [64,150]. Antioxidants remove potentially damaging oxidative agents, which is pivotal to maintaining optimal health, whereas α-amylase inhibition delays starch digestion by completely blocking access to the active site of the α-amylase enzyme. This action induces weight loss and reduces blood glucose excursions caused by CHO consumption, thus leading to the control and reduction of metabolic comorbidities [151]. This slowed absorption of CHO through the inhibition of enzymes responsible for digestion is one of the many ways in which bitter melon can assist in the control, treatment, and prevention of T2DM and AD [151]. ACE inhibitors aim to lower blood pressure by preventing the production of angiotensin II, a peptide responsible for inducing vasoconstriction of blood vessels. When ACE inhibitors are lacking, the heart contracts against a higher afterload, reducing the stroke volume and causing the release of various hormones known to further increase heart work [152]. Thus, the natural production of ACE inhibitors through the ingestion of biochemically active polysaccharides from bitter melon can potentially achieve cost-effective methods for the treatment and prevention of T2DM and AD [152]. Recently, a water-soluble polysaccharide (MBP) component isolated from M. charantia, mainly composed of Ara, Xyl, Gal, and Rha, demonstrated significant hypoglycemic effects, crucial for efficiently controlling T2DM [150]. By doing so, complications often found in uncontrolled T2DM, such as blurred vision, difficulty concentrating, confused thinking, slurred speech, numbness in extremities, and drowsiness are significantly reduced [152]. Complications in relation to AD need further research. In another study, M. charantia polysaccharides were found to improve oxidative stress, hyperlipidemia, inflammation, and apoptosis in myocardial ischemia through the inhibition of the NF-kB signaling pathway [153,154]. M. charantia polysaccharides were also found to enhance total volatile fatty acids production, modulate the rumen fermentation pathway, and influence the cellulolytic bacteria population, promoting overall healthy digestion [155]. This is due to the plant’s ability to target both Aß and tau proteins present in the brain. Both findings are relevant to T2DM and AD as these properties hold major importance for each of these metabolic disease states; however, more research on the clinical application of bitter melon is needed. Proteins and peptides are other major functional components in both the fruit and seeds of M. charantia [64]. Various types have been isolated from different parts of bitter melon, such as polypeptide-p and peroxisomes [85,86,87,88,89,90,91]. These proteins exhibit RNA N-glycosidase activity, PAG activity, DNase-like activity, phospholipase activity, superoxide dismutase activity, anti-tumor, anti-cancer, and immunosuppressive and anti-microbial activity. The hypoglycemic-related activities of the bioactive proteins and peptides of bitter melon are discussed in the next section [85,86,87,88,89,90,91]. One of the main compounds isolated from the medicinal bitter melon plant is polypeptide-p [3]. This peptide is a carbohydrate-binding protein secreted by plant cells, playing a significant role in cell recognition and adhesion reactions, and lowering blood glucose levels [64]. When injected subcutaneously, polypeptide-p acts as an insulin-like hypoglycemic protein [84]. By mimicking the action of human insulin and binding to the INS receptor, plant-based insulin may be used as a replacement in patients with T1DM [92]. In a recent study, the 498 bp gene sequence coding was cloned for the M. charantia polypeptide-p and when given to alloxan-induced diabetic mice, lowered blood glucose to controlling methods [93]. Through a better understanding of this polypeptide-p’s actions, this knowledge can be applied to the treatment of T2DM and AD. There are several other polypeptides that have been isolated from the fruit, seeds, and tissues of M. charantia, including peroxidase [94,95,96]. Peroxidase plays an important protective function in T2DM complications by decreasing oxidative stress and removing toxicity from peroxides and thus converting them into non-toxic substances [97]. Oxidative stress is involved in the pathogenesis of diabetic nephropathy as free radical production exceeds the antioxidative mechanisms to overcome its detrimental effects [97]. These protective functions provide the same benefits toward the pathology of AD, as the pathogenesis of these two metabolic disease states are seemingly related. There are many other important polypeptides within M. charantia, including Momordica cyclic peptides, trypsin inhibitors, cystine knot peptides, RNase MC2, antifungal protein, and MCha-Pr. Overall, peroxidase is just one of the many polypeptides within bitter melon that can potentially improve both T2DM and AD [94,95,96,97]. Saponins are a class of glycosides that are responsible for reducing blood lipids, lowering the blood glucose response, and decreasing certain cancer risks [64,98]. They are widely distributed in a variety of plants and are also some of the key active ingredients within several different drugs on the market today [64,99]. All saponins are composed of sugar and aglycone, with the difference lying in the structure of aglycones [64]. In M. charantia, saponins can be found within the roots, stems, leaves, and fruit of the plant [64]. Saponins contain a CHO moiety attached to a triterpenoid or steroids, which is another major chemical constituent of M. charantia, often referred to as cucurbitanes, which are known for their bitterness in taste and toxicity [64,98]. Cucurbitacins are highly oxygenated, tetracyclic, triterpenic plant substances derived from the cucurbitane skeleton that demonstrate antidiabetic and hypoglycemic bioactive activity [64,100]. Cucurbitane-type compounds, such as goyaglycosides and goyasaponins, have been isolated from the methanolic extract of M. charantia fruits, as well as triterpenoids that showed blood-glucose-lowering effects in diabetic mice [64,101]. One study found that these compounds of M. charantia have hypoglycemic effects in vivo as well [99,102]. Four of these triterpenoids of M. charantia have been identified to possess AMP-activated protein kinase activity, which is a notable hypoglycemic mechanism [3,103]. This evidence highlights the potential of bitter melon in controlling, preventing, and treating T2DM and AD through its bioactive compounds. One of the major compounds isolated as a saponin from bitter melon and identified as a hypoglycemic agent includes charantin, a cucurbitane-type triterpenoid [93,104]. In recent research, charantin was a viable option to treat diabetes and showed the potential to replace treatment entirely [105]. Some evidence has shown that the compound has the capability to be more effective than the oral hypoglycemic agent, tolbutamide [17]. In recent a study, two aglycones of charantin were isolated and identified as sitosterol and stigmastadienol glycosides. When tested separately for their hypoglycemic effects in vivo, these two components did not induce any significant changes in blood glucose levels [102,106]. This suggests that charantin may contain other specific factors that have yet to be uncovered that could be identified for the hypoglycemic activity observed in diabetes [3]. Future research is necessary to determine the effect on the pathology of AD. Flavonoids and phenolic compounds are vital components of M. charantia that are beneficial to T2DM and AD prevention and treatment [159,160]. Flavonoids can be further classified into six subclasses: flavanols, flavones, flavanones, isoflavones, flavanols, and anthocyanidins [64,132]. They are considered a class of biologically active secondary metabolites of plants responsible for the smell and pigment of flowers and hold several antiviral, anti-allergic, antibacterial, and anti-inflammatory functions [132,161]. Flavonoids can improve the pathogenesis of T2DM, AD, and its complications through the regulation of glucose metabolism, expression of hepatic enzyme activities, and lipid profile [132]. M. charantia contains naturally occurring flavonoids with antidiabetic potential including quercetin, rutin, kaempferol, isorhamnetin, and genistein [64,162]. Quercetin is the most abundant flavonoid in human dietary nutrition and acts as the base for the formation of other flavonoid skeletons, such as naringenin, rutin, and hesperidin [107]. With antioxidant, anti-inflammatory, and antiapoptotic effects, quercetin exhibits the potential to treat both T2DM and AD [102,108,109,110]. Quercetin is involved in several biological actions such as glucose homeostasis, insulin-sensitizing and secretion, glucose utilization in peripheral tissues, and the inhibition of intestinal glucose absorption [111]. In skeletal muscle cells, quercetin increases glucose uptake through the stimulation of GLUT4 translocation by activating the AMPK pathway in glucose homeostasis [108]. In hepatocytes, quercetin activates the AMPK pathway by suppressing glucose-6-phosphatase (G6Pase), which in turn lowers hepatic glucose production [108]. A recent study found that quercetin heightened glucose-induced insulin secretion and preserved β-cell function and viability from H2O2−-induced oxidative damage in INS-1 cells [3,109]. These effects were modulated by phosphorylation of extracellular signal-regulated kinase (ERK1/2), suggesting that ERK1/2 activation was involved in quercetin’s mechanism of action [109]. In another study, quercetin improved glucose and lipid metabolism, alleviated hepatic histomorphological injury, and reduced gluconeogenesis in STZ-induced diabetic rats [110]. This was likely associated with the upregulation of SIRT1 activity, as well as its effect on the Akt signaling pathway, showing therapeutic potential for T2DM and obesity [110,111]. Vascular complications are the main cause of morbidity and mortality rates in diabetes patients [112]. In STZ-induced diabetic rats, the administration of quercetin prevented the advancement of diabetes-induced hypertension and negated diabetes-induced vasoconstriction [79]. These outcomes were likely due to the inhibitory effects of quercetin on inflammatory pathways, via NF-κB and by reducing TNF-α and CRP levels in the aorta of diabetic rats [79]. Quercetin may have neuroprotective effects in diabetic peripheral neuropathy [113]. There have been several in vivo and in vitro studies that demonstrated these neuroprotective effects [113,114,115]. One recent study found that high levels of glucose disrupted the proliferation of rat RSC96 cells and primary rat Schwan cells, including suppression of beclin-1 and light chain (LC3), which are the biomarkers for autophagy, and reducing the number of autophagosomes in both cell types [113]. However, after treatment with quercetin, these effects were attenuated [113]. In another study, the supplementation of quercetin reversed cognitive decline in mice fed a high-fat diet [114]. This was associated with altered signaling of Nrf2, which improved overall cognitive function [114]. Furthermore, it was reported that quercetin has the potential to decrease oxidative stress and diminish inflammation and protein glycation in the brain of diabetic rats [115]. These consequences may be related to the upregulation of glyoxalase, which is a ubiquitous cellular enzyme that participates in the detoxification of the cytotoxic byproduct of glycolysis and is believed to have a role in the pathogenesis of diabetic encephalopathy [115]. The favorable effects of quercetin on AD were also established in both cell and animal studies [116,117,118,119,120]. In cultured neurons, pretreatment with quercetin enhanced Aβ1-42-induced protein oxidation, lipid peroxidation, cytotoxicity, and apoptosis. However, high doses had the opposite effect with non-neuroprotective and toxic effects [116]. In one study, quercetin extended the lifespan and supported the climbing ability of AD flies [117]. Cell-cycle-related proteins were disrupted by Aβ accumulation, therefore allowing quercetin to improve cell-cycle-related signaling pathways [117]. In another study utilizing a triple transgenic AD (3xTg-AD) mouse model, a 3-month treatment with quercetin reduced extracellular β-amyloidosis and enhanced microglial and astroglial activation in the brain, as evidenced by diminished levels of Aβ1-40, Aβ1-42, and BACE1-mediated cleavage of APP [118]. In addition, learning and memory were ameliorated [124]. The administration of quercetin to APPsw/ PS1dE9 mice improved learning and memory deficits and reduced plaque levels in comparison to control mice [119]. Quercetin accomplished these protective effects by reducing mitochondrial inhibition through the activation of AMPK [119]. Furthermore, recent research unveiled the potential anti-inflammatory role of quercetin in AD mice [120]. Quercetin treatment diminished β-amyloid plaque accumulation as well as reduced IL-1β/COX-2/iNOS proinflammatory signaling in the hippocampal CA1 region of 3xTg-AD mice [120]. Further research is needed to determine the exact effects of quercetin on both AD and T2DM to best utilize bitter melon in medical preventative practice. Rutin is another flavanol found in M. charantia. Its many biological effects include antioxidant, anti-inflammatory, antihyperglycemic, and neuroprotection, all of which support a potential use in the prevention and treatment of T2DM, AD, and associated complications [3,121,122]. For example, in nicotinamide- (NA-) STZ-induced diabetic rats, rutin improved glucose tolerance and reduced serum glucose levels. Rutin also improved the lipid profile, including LDL-cholesterol, VLDL-cholesterol, and triglycerides (TGs) [123]. All these changes were accompanied by an improvement in the oxidative status of diabetic rats. The possible mechanisms for the antihyperglycemic and antihyperlipidemic effects of rutin were further investigated [3]. In one study, rutin reduced hepatic glucose output through the decreased activity of G6Pase and glycogen phosphorylase, as well as the increased activity of hepatic hexokinase activity [123]. Furthermore, a reduction in glucose levels was accomplished by improving glucose uptake by tissues [122]. Indeed, a recent study showed that rutin decreased blood glucose levels in insulin-resistant mice through the amplification of insulin-dependent receptor kinase (IRK) activity and GLUT4 translocation [123]. In another study, a protective effect of rutin in the livers of db/db mice was revealed through the activation of the IRS2/PI3K/Akt/GSK3β signal pathway, improving hepatocyte proliferation, and decreasing the generation of AGEs, thus making it a useful component towards the treatment and prevention of metabolic complications [124]. Recent evidence has revealed a role of PPARγ expression in adipose tissue and skeletal muscle. Stimulation of PPARγ expression in these tissues enhances insulin sensitivity and increases glucose uptake [123,125]. Treatment with rutin also increased β-cell viability and reduced glucotoxicity through the activation of AMPK and IRS2 signaling [126]. In one study, rutin demonstrated the capacity to improve insulin secretion in isolated rat pancreatic islets [123]. Through this evidence, rutin was found to successfully lower the formation of ROS, AGE precursors, sorbitol, and pro-inflammatory cytokines, thus making it a viable treatment and preventative option for T2DM [108,127]. Other antidiabetic effects of rutin include the reduction of CHO absorption from the small intestine, suppressing gluconeogenesis, activating insulin secretion from β cells, and protecting the islets of Langerhans from degenerative processes [160]. Rutin was also found to significantly minimize oxidative stress through the suppression of inflammatory cytokines in STZ-induced diabetic rats [127]. Therefore, rutin’s ability to enhance glucose uptake by peripheral tissues, improve insulin resistance, suppress gluconeogenesis in the liver, and stimulate insulin secretion make bitter melon a viable remedial option for naturally controlling blood sugar levels [3,123,124,125,126]. Further research on rutin will be beneficial for both medicinal and pharmaceutical purposes in the future. Rutin has also demonstrated significant therapeutic potential for AD [128]. Through the reduction of inflammatory markers of neurodegeneration, reducing oxidative stress relating to neuronal cell loss, and preventing Aβ aggregation, rutin has the potential to have a significant effect on the prevention and treatment of AD [124]. In a study using APPswe (APP Swedish mutation) cells, rutin prevented Aβ25-35 fibril formation and slowed the activity of BACE [129]. Rutin also improved cell viability, while also lowering GSH levels induced by the overexpression of APP in APPswe cells [129]. In a similar study, rutin inhibited Aβ42 fibrillization and improved Aβ42-induced cytotoxicity in SH-SY5Y cells [32]. Additionally, rutin decreased mitochondrial damage, reduced the formation of ROS, GSSG, NO, iNOS, and proinflammatory cytokines, and promoted the activities of SOD and catalase [124]. In a study using Aβ-injected rats, the administration of rutin significantly improved memory through the activation of the MAPK pathway and brain-derived neurotrophic factor (BDNF) gene expression and declined oxidative stress and neurotoxicity induced by Aβ [130,131]. Kaempferol is a non-toxic flavonoid with several medicinal effects benefitting both T2DM and AD [133]. Some of the major antidiabetic effects of kaempferol include improving AMP-activated cellular protein expression and activation, reducing cellular apoptosis by suppressing caspase 3 activities, and increasing the production and secretion of insulin from β cells [133,134]. Additionally, kaempferol increases glucose uptake by the cells through protein kinase C and PI3K pathways and enhances the synthesis of glucose transporter proteins [135]. Furthermore, kaempferol significantly reduces serum HbA1c levels and fasting blood glucose while enhancing insulin sensitivity when administered orally [132,135]. Kaempferol also has been shown to lessen the expression of PPARγ mediated through regulating AMPK activation [136]. Moreover, kaempferol improved the diabetic state of STZ-induced mice through glucose metabolism in skeletal muscle and suppression of hepatic gluconeogenesis [137]. In another study, researchers found that kaempferol reduced diabetic nephropathy in NRK-52E and RPTEC cells by decreasing RhoA/Rho-kinase mediated pro-inflammatory signaling (i.e., TNF-α, IL-1β, and TGF-β1) [137]. Taken together, the results of these studies highlight the potential of kaempferol as a therapeutic agent for the treatment of diabetes. However, further research is needed to gain a better understanding of the role of kaempferol in AD. Isorhamnetin is another bioactive compound within M. charantia that has anti-obesity and antidiabetic effects that may also be applied to AD prevention and treatment [138]. In a recent study, isorhamnetin was administered orally for 10 days to STZ-diabetic mice at a dose of 10 mg/kg or 20 mg/kg and successfully demonstrated a reduction in oxidative stress and hyperglycemia [139]. In another study, isorhamnetin was not only able to reduce blood glucose levels, but also decreased the aggregation of sorbitol levels on the lens of the oculus, the sciatic nerve, and red blood cells, which are all common complications of uncontrolled T2DM in humans. [140]. Any bioactive effect that prevents the complications of either T2DM or AD is beneficial for the overall treatment and in the understanding of these diseases. Further research is needed on isorhamnetin to best identify the precise effects of its role in bitter melon. There are several other components within bitter melon that have been identified and isolated from the plant including unsaturated fatty acids (FA), alkaloids, amino acids (AA), vitamins, and minerals [3,163,164,165]. There is a relatively high amount of unsaturated FA components with monounsaturated fatty acids (MUFAs) making up about 20.1% of total fatty acid content and polyunsaturated fatty acids (PUFAs) making up roughly 64.3% [64]. Altogether, nine different types of unsaturated FA have been identified in bitter melon [166]. These unsaturated FAs are known to enhance insulin sensitivity, thereby improving metabolic disease states such as T2DM and AD [166]. Through acid hydrolysis and AA analysis, the total content of AA is estimated at 11.99% and 2.36% for free AA [167]. AA are important bioactive components of bitter melon as they provide nutrients to improve blood pressure, hyperglycemia, visceral obesity, and abnormal cholesterol or TG levels. Additionally, M. charantia is a considerable source of vitamins and minerals, including ascorbic acid [168,169,170]. Ascorbic acid appears to lower blood glucose increase insulin synthesis and secretion, and enhance insulin sensitivity, all of which are beneficial in T2DM, AD, and other metabolic diseases [73,170]. The effects of bioactive compounds in bitter melon on AD and T2DM are given in Table 1. To fully understand the impact of bitter melon on T2DM and AD, further research will be necessary to uncover the vast effects of these other components within the plant. Bitter melon can be easily incorporated into one’s diet to maintain control of blood glucose levels. The bioactive components within M. charantia allow for these hypoglycemic effects to occur. Information regarding the proper ways to consume this nutritious plant is necessary to achieve the safest and best results. Clinical nutrition is focused on the prevention, diagnosis, and management of nutritional changes in patients with chronic diseases and conditions. As research reveals the importance of a well-balanced diet filled with a variety of vitamins and minerals, it has become apparent that nutrition plays a much bigger part in overall health. In relation to chronic disease and illness, medical nutrition therapy has become the forefront of treatment, making nutrition critical for successful patient care and recovery. Medical nutrition therapy utilizes the different bioactive components within foods to best treat patients in the most cost-effective and safe way possible. The importance of this therapy is further exemplified by its gravitation towards its individualized care in determining which foods best manage and treat ailments of certain chronic diseases or metabolic disorders. As elements within plants such as M. charantia are unraveled, and their medicinal effectiveness better understood, the more we can utilize their effects in actual practice. As previously discussed, the medicinal benefits of bitter melon extend beyond the fruit itself, but to its stem, leaves, and roots as well [3]. Therefore, the entire plant can be used to help manage the complications relating to T2DM, as well as prevent or slow the progression of the signs and symptoms contributing to the pathogenesis of AD [3,55]. The phytochemical components within bitter melon act as insulin to help reduce blood sugar levels and demonstrate certain neuroprotective effects through its bioactive additives including terpenoids, glycosides, flavonoids, phenolic, and charantin. Furthermore, as AD and T2DM share common etiologies, the consumption of bitter melon carries significant value as a treatment and control option for both diseases. M. charantia’s ability to lower blood glucose levels is a significant representation of the power of food and its impact on the population’s overall health. The fruit itself can either be consumed raw or cooked, however many prefer it cooked to tone down the bitter flavor. One should consult with a primary care provider before incorporating bitter melon into one’s diet, as consuming this fruit may cause very low blood sugar levels when combined with traditional diabetes medicine. Bitter melon is not safe for children or for pregnant or breastfeeding women to consume. Further research is needed regarding any possible interference with the bioavailability of other nutrients as well as its effect on the mechanisms of various medications one may be prescribed. Nevertheless, bitter melon is a safe, effective, and affordable option to naturally control blood glucose levels. As a result, the moderate consumption of this nutrient-rich food can help in the control of T2DM and possibly delay the progression of AD. Overall, T2DM and AD are complex metabolic disorders occurring at alarmingly high rates with substantial social and economic burdens [19]. The lack of successful therapeutic treatment options in the management of AD and long-term diabetes requires the development of safe and effective complementary approaches, such as the use of the M. charantia plant [19]. The various bioactive compounds within bitter melon have piqued the interest of researchers and have led to the exploration of the therapeutic potential of this vine fruit [19]. Recognizing the molecular mechanisms of action underlying the antidiabetic and neuroprotective effects of bioactive compounds in cell cultures and animal models of T2DM and AD is the first step in uncovering M. charantia’s potential action on these metabolic disorders [19]. Published data emphasize the prospective beneficial effects of bioactive compounds (Figure 4) on lowering hyperglycemia, magnifying insulin secretion, amplifying β-cell function, reducing Aβ accumulation, and strengthening cognitive function [19]. In this paper, the role of the major bioactive compounds of bitter melon was explored, including polysaccharides; proteins and peptides such as polypeptide-p, and peroxidase; saponins and terpenoids such as charantin; and flavonoids and phenolic compounds such as quercetin, rutin, kaempferol, and isorhamnetin to fully grasp its significance in the treatment and prevention of both T2DM and AD [3,64,92,157]. As research on the bioactivities of M. charantia continues to develop, a better understanding of the antioxidant, anti-inflammatory, and antiapoptotic properties and their mechanisms is necessary for the advancement of plant-based alternative medicine, the curation of new and effective drugs, and the growth of medical nutrition therapy [19,64]. Thus, clinical studies of the bioactive components should be the focus of future research [64]. In doing so, the relationship between the structure and mechanisms of the various functional components within bitter melon will be further clarified [64]. Most research conducted on the bioactive compounds of M. charantia has produced controversial results, which may be due to several factors including experimental design, dosage, and types of bioactive compounds examined [19]. Additionally, the possible complications from long-term consumption on the human body have not been extensively explored [64]. Therefore, carefully designed clinical trials will be needed to produce relevant evidence for the potential therapeutic utilization of bioactive compounds within M. charantia in the treatment of T2DM and AD [19]. Recognizing the utilization of food as medicine is an important concept in nutrition sciences as reputable research continues to grow within this field [3]. Throughout many centuries, M. charantia has been used as a method of alternative medicine and dietary supplement for treating symptoms and conditions related to what we know today as diabetes [3]. Bitter melon is characterized as a multipurpose plant worthy of treating several diseases known to mankind and has been substantially studied across the globe for its powerful medicinal properties [2,3]. This may be due to the plant’s various medicinal elements that act either separately or congruently to exert their medicinal effects [3,170]. In relation to DM, the hypoglycemic properties of bitter melon are what bring this plant its justified attention [3]. In relation to AD, the various antioxidant properties are what provide protection to both cognitive function and cholesterol levels within the brain. These different bioactive compounds seem to exert their beneficial effects through several mechanisms of action to best control and treat DM and AD [3]. Although this review includes an elaborate discussion of biochemical and animal studies on M. charantia, these studies are flawed by small sample size, lack of control, and poor study designs. This paper advocates the need for certain improvements to be made in study design for clinical trials in relation to adequate sample size and statistical power. By doing so, the safety and efficacy of M. charantia as a natural nutritional treatment for DM and AD can be suggested with scientific evidence [3]. Furthermore, M. charantia may have a greater effect on ethnic minorities who have a higher incidence of diabetes but prefer natural treatment based on cultural beliefs [3]. In conclusion, the application of bitter melon in medicine remains in the initial processing stages, as scientists continue to uncover its numerous health benefits from its bioactive constituents [64]. M. charantia not only has the potential to be a safe and effective therapeutic method for individuals suffering from the complications of T2DM and AD but also as a cost-effective option to ease the economic and social burden these metabolic disorders have on the populations worldwide. Further research is needed to determine the exact effects of bitter melon on humans from a clinical standpoint.

PMC10002572

Sunanda Singh,Hector J. Gomez,Shreya Thakkar,Samara P. Singh,Ashutosh S. Parihar

Overcoming Acquired Drug Resistance to Cancer Therapies through Targeted STAT3 Inhibition

01-03-2023

STAT3,acquired drug resistance,kinase inhibitors,chemotherapy,monoclonal antibodies,immune checkpoint inhibition

Anti-neoplastic agents for cancer treatment utilize many different mechanisms of action and, when combined, can result in potent inhibition of cancer growth. Combination therapies can result in long-term, durable remission or even cure; however, too many times, these anti-neoplastic agents lose their efficacy due to the development of acquired drug resistance (ADR). In this review, we evaluate the scientific and medical literature that elucidate STAT3-mediated mechanisms of resistance to cancer therapeutics. Herein, we have found that at least 24 different anti-neoplastic agents—standard toxic chemotherapeutic agents, targeted kinase inhibitors, anti-hormonal agents, and monoclonal antibodies—that utilize the STAT3 signaling pathway as one mechanism of developing therapeutic resistance. Targeting STAT3, in combination with existing anti-neoplastic agents, may prove to be a successful therapeutic strategy to either prevent or even overcome ADR to standard and novel cancer therapies.

Overcoming Acquired Drug Resistance to Cancer Therapies through Targeted STAT3 Inhibition Anti-neoplastic agents for cancer treatment utilize many different mechanisms of action and, when combined, can result in potent inhibition of cancer growth. Combination therapies can result in long-term, durable remission or even cure; however, too many times, these anti-neoplastic agents lose their efficacy due to the development of acquired drug resistance (ADR). In this review, we evaluate the scientific and medical literature that elucidate STAT3-mediated mechanisms of resistance to cancer therapeutics. Herein, we have found that at least 24 different anti-neoplastic agents—standard toxic chemotherapeutic agents, targeted kinase inhibitors, anti-hormonal agents, and monoclonal antibodies—that utilize the STAT3 signaling pathway as one mechanism of developing therapeutic resistance. Targeting STAT3, in combination with existing anti-neoplastic agents, may prove to be a successful therapeutic strategy to either prevent or even overcome ADR to standard and novel cancer therapies. Cancer encompasses a diverse group of diseases with common features and behaviors [1]. Within each histologic type of malignancy, there is often tremendous heterogeneity, which develops from the highly unstable genome of cancer cells. This genomic instability leads to development of cancer cell variants within the bulk tumor. When selection pressure is applied by exposing cancers to anti-neoplastic treatments such as chemotherapeutics, targeted therapeutics, anti-hormone therapeutics, and monoclonal antibodies, clonal selection may occur in patients, and oftentimes acquired drug resistance (ADR) develops. Drug resistance is a very complex and heterogenous problem that has developed through a variety of many mechanisms. Even within the same patient, several modes of ADR may be present across different tumors [2]. Drug resistance can be intrinsic (i.e., de novo) or conditional to an initial response to an anti-neoplastic agent followed by progression of the cancer—otherwise known as ADR. Signal transducer and activators of transcription 3 (STAT3) has been implicated in the development and maintenance of ADR in multiple cancers in response to various therapies. In this review, we will focus on the role of STAT3 in the development of ADR and clinically relevant drugs that are susceptible to STAT3-mediated ADR. Combination therapy of STAT3 inhibitors with therapies prone to ADR may prove to be synergistic and a compelling strategy to overcome therapeutic resistance in the clinical setting. Human cells have evolved to develop complex regulatory mechanisms, including positive feedback loops and significant crosstalk among oncogenic signaling pathways. In its simplistic form, ADR can develop when the inhibition of one pathway induces the activation of another, which may impair any therapeutic effect. Twelve pathways, critical to ADR, have been identified, and the STAT3 signaling pathway, described as the “master regulator of antitumor immune response” is one of them [3,4]. In general, the anti-neoplastic agents that have limited efficacy as result of the development of ADR can be divided into four groups: (1) traditional chemotherapeutic drugs, (2) targeted therapeutics, (3) anti-hormone therapeutics, and (4) monoclonal antibodies (Table 1) [5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30]. The traditional chemotherapeutic drugs that develop ADR and are discussed here include doxorubicin [31,32,33,34], gemcitabine [35,36,37,38,39,40,41,42,43,44], cisplatin [34,45,46,47,48,49,50,51,52,53,54,55], temozolomide [56,57,58,59,60], and paclitaxel [61,62,63,64,65]. Targeted therapeutics for which the development of ADR has been documented included in this review are afatinib [37,60,66,67], crizotinib [68], dasatinib [69], and erlotinib [70,71,72,73,74]. The anti-hormone therapeutics that develop ADR are flutamide [75,76,77], enzalutamide [78,79], and tamoxifen [80,81,82,83]. ADR is also developed to monoclonal antibodies such as cetuximab [84,85,86], bevacizumab [87], trastuzumab [88,89,90,91], and to the immune checkpoint inhibitors (ICIs) [92,93,94,95,96,97,98,99,100,101,102,103,104,105,106], such as pembrolizumab [101], nivolumab [102,104], and ipilimumab [103]. In response to all these anti-neoplastic agents, cancer cells utilize STAT3 as one mechanism of escaping their therapeutic effects and promoting ADR. There is a very large body of scientific and medical literature to support the use of anti-STAT3 therapeutics to overcome ADR in these cases. While other mechanisms of ADR exist, here we focus on STAT3 as a key mechanism for the development of ADR. It should be stated that STAT3 is present in all mammalian cells and plays an important role in physiological functions. Under normal conditions the duration of STAT3 activity is short and transient but in pathological situations, such as cancer, a stronger activation is maintained over long periods of time [117,118]. Activated STAT3 refers to the phosphorylated STAT3 (tyrosine or serine phosphorylated) and is measured and or quantified and described as p-STAT3. This pathological form is referred to in the literature by many different terms such as aberrant, constitutive, dysregulated, etc. STAT3 present in cancer cells is p-STAT3, the constitutively phosphorylated form responsible for the acquired resistance described in this publication [117,118]. STAT3 is located intracellularly, downstream many kinases at an exchanging point of the most important signaling pathways involved in cancer. Oftentimes, when an administered drug blocks a specific kinase pathway, the STAT3 pathway is triggered as result of the crosstalk amongst upstream pathways, resulting in the aberrantly persistent form of p-STAT3. There are seven STATs (STAT1, STAT2, STAT3, STAT4, STAT5a, STAT5b, and STAT6) that are intracellular proteins which function as signal messengers and transcription factors. They transmit signals from cytokines, growth factors, intracellular kinases, mutated oncoproteins, and other signaling pathways to the nucleus. Tyrosine phosphorylation cascade occurs after ligand binding by many extracellular molecules such as epidermal growth factor (EGF), platelet-derived growth factor (PDGF), fibroblast growth factor (FGF), interleukin-6 (IL-6), IL-5, oncostatin-M (OSM), granulocyte colony stimulating factor (GCSF), colony stimulating factor-1 receptor (CSF1R), leukemia inhibitory factor (LIF), c-kit, c-Met, insulin receptor, angiotensin-II receptor (AgtR2), interferons (IFNs), G-protein coupled receptors (GPCRs), and others. After such ligands bind the extracellular portion of their receptors, their intracellular portion attracts the Janus Kinase family (JAK1, JAK2, JAK3, and Tyk2) of proteins, which become phosphorylated. The JAK protein then phosphorylates STAT3 (pSTAT3) at tyrosine 705 and sometimes serine 727 to activate STAT3. Other intracellular kinases, which can directly activate STAT3 are Src and BCR–ABL, the mutant fusion protein in chronic myelogenous leukemia (CML) [26]. P-STAT3 then forms dimers, which translocate to the nucleus via chaperone proteins. There p-STAT3 dimers bind to specific nine base pair sequences in regulatory genomic regions to regulate transcription of specific genes. The signaling function of p-STAT3 is carefully regulated by inhibitory molecules such as protein inhibitors of activated STAT (PIAS), protein tyrosine phosphatases (PTPases), and suppressors of cytokine signaling (SOCS). Dysregulation of the normal physiologic balance of p-STAT3 and unphosphorylated STAT3 can occur due to upstream mutations or protein overexpression. This results in constitutive expression of p-STAT3 and continuous transcription of pro-oncogenic and anti-apoptotic genes, which promotes cancer growth, proliferation, cell cycle re-entry, angiogenesis, immunosuppression, and, metastasis when anticancer agents apply selective pressure might induce the development of ADR. STAT3 play critical roles within neoplastic cells, immune cells, and other stromal cells, such as cancer-associated fibroblasts (CAFs). Once activated within tumor cells, phosphorylated STAT3 (p-STAT3) regulates the transcription of various immunosuppressive cytokines such as VEGF, IL-10, and TGF-β. p-STAT3 can promote tumor progression by increasing transcription of genes associated with stemness and epithelial to mesenchymal transition (EMT) [119]. Additionally, p-STAT3 is involved in two apoptotic processes, autophagy and anoikis, both contributors to ADR development. Autophagy, a cellular degradation process, is another regulatory mechanism that plays a major role in maintaining homeostasis, and its dysfunction has been implicated in cancer progression and ADR. The signaling pathways that control inducible autophagy and cell death are closely associated and incorporated into the tumor regulatory network of autophagy related proteins, ultimately affecting the fate of tumor cells [120]. The crosstalk between autophagy and other stress response pathways including STAT3, determines the survival or death of a cell. Nuclear STAT3 regulates autophagy in various forms. For instance, STAT3 inhibits autophagy by activating BCL2 or increases it by upregulating and stabilizing HIF1A under hypoxia; however, it has been determined that cytoplasmatic STAT3 regulates autophagy in a more direct way [121]. Autophagy initially prevents cancer progression but under stressful situations improves the survival of cancer cells [122] contributing to ADR and therapy failure. p-STAT3 has been found to be associated with aberrant autophagy activity in many oncological studies [123]. The anti-autophagy action is partly due to STAT3-mediated inhibition of the BEBN1/PIKC3 complex, resulting in reduced Beclin-1 activity. There is a link between ADR to chemotherapeutics, sometimes described as chemoresistance, and autophagy. The autophagic process vary depending on the tumor stage. In some cases, high dosage chemotherapy may induce protective autophagy that leads to ADR. Some proteins such as mTOR, Beclin-1, miRNA, and autophagy-related genes play a role during treatment of some cancers such as osteosarcoma. The use of autophagy inhibitors in combination with chemotherapeutics is being studied as a new treatment of cancer that might avoid chemoresistance [124]. STAT3 inhibition increases autophagy by increasing transcription of key activators of autophagy. [125]. The importance of autophagy in tumor immunity and ADR is now recognized and has been reported that optimal induction or inhibition of autophagy may induce effective treatments when combined with immunotherapy [126]. Anoikis, another type of apoptosis, is triggered by loss of cell adhesion [127]. It might be activated during tumorigenesis to clear off extracellular matrix (ECM) and detached cells. Cancer cells that develop the ability to survive are called anoikis-resistant cells. These cells become very aggressive and drug resistant, developing the capacity to invade and migrate to metastatic sites. Several features have been identified as responsible for modulating anoikis resistance, one of which is STAT3. STAT3-related anoikis-resistance has been reported in cancer cells of human pancreatic cancer, melanoma, cholangiocarinoma, esophageal squamous cell carcinoma, squamous cell carcinoma, nasopharyngeal carcinoma, and lung carcinoma [128,129,130,131,132]. These cancer cells were reported to have enhanced cell migration, invasion capability and high metastatic potential, and inhibition of STAT3 led to sensitization of all those anoikis-resistant cells [133]. Nicotinamide N-methyltransferase (NNMT) participates in the development of ADR. NNMT, a cytoplasmic enzyme that methylates nicotinamide, is regulated by STAT3 and has been shown to be overexpressed in solid tumors. Furthermore, STAT3 activation intensifies the expression of NNMT and stimulates its activity [134]. NNMT has been identified as an oncogene in intrahepatic cholangiocarcinoma [135]. NNMT is upregulated in cutaneous squamous cell carcinoma, induces cellular invasion via EMT-related gene expression [136] and plays critical roles in the incidence and development of various cancers [137]. Evidence that NNMT plays an important role in cancer can be seen by the fact that NNMT knockdown reduces tumorigenesis and chemoresistance and that Yuanhuadine, a natural inhibitor of NNM, reverses EGFR inhibitors ADR [138]. Chemoresistance or ADR to adriamycin and paclitaxel in breast cancer has been also reported by Wang et al., 2019. This group found that reversal of NNMT related ADR can be accomplished by using the SIRT1 inhibitor, EX527 or using siRNA therapy. SIRT1 also represses the activation of STAT3 and NF-κB proteins via deacetylation [139]. The major function of the tumor suppressor p53 is to induce transient cell cycle arrest, cellular senescence, and apoptosis, a significant barrier to the development of tumors; however, p-STAT3 can inhibit these repressive functions of p53 [140]. This crosstalk between STAT3 and p53 also contributes to the development of ADR and the loss of the pharmacologic effects of anticancer agents [141]. STAT3 inhibition upregulates the expression of p53 and increases cellular apoptotic activity, thereby reversing ADR. Another important signaling pathway for growth and proliferation is the RAS/mitogen activated pathway kinase (MAPK). The crosstalk between STAT3 and p53/RAS signaling regulates metastasis and cisplatin resistance in ovarian cancer through the Slug/MAPK/PI3K/AKT-mediated regulation of EMT and autophagy [142]. Therefore, RAS and STAT3 activation promote ovarian cancer growth, metastasis, and cisplatin resistance. Dual inhibition of STAT3 and KRAS, achieved by nano-antibody SBT-100, would be an ideal treatment for this type of cancer to overcome ADR in ovarian and many other types of cancer [143]. As previously mentioned, p-STAT3-mediated ADR occurs in response to anti-neoplastic agent therapy by utilizing multiple intracellular signaling pathways. As illustrated in Figure 1, treatment with a receptor tyrosine kinase inhibitor, which blocks MAPK pathways, results in the cancer cells secreting ligands, which bind to receptors on the cancer cells themselves in an autocrine fashion or to CAFs, intratumor macrophages, and other cells in the tumor microenvironment (TME) in a paracrine fashion. This ligand binding to its cognate receptor results in STAT3 activation, turning on numerous genes that promote growth, proliferation, cell cycle re-entry, anti-apoptosis, angiogenesis, immunosuppression, and metastasis, and ultimately circumventing the anti-neoplastic therapy being used resulting in ADR. Doxorubicin is an anthracycline compound and currently one of the most effective classes of anti-cancer agents in clinical applications; however its use is limited by its chronic and acute toxicities [107]. It binds to topoisomerase I and II, resulting in intercalation of the base pairs of the DNA double helix and inhibition DNA replication. Because of this mechanism of action, doxorubicin has been highly effective in treating a wide variety of malignancies. Its efficacy is unfortunately limited in many cases by ADR due to STAT3 upregulation. A well-known STAT3 inhibitor, Stattic, was formulated in nanostructured lipid carriers to enhance the efficacy of doxorubicin against melanoma cancer cells [108,109]. Doxorubicin induces p-STAT3 in human breast cancer MCF cell line (ER+, non-metastatic) and human triple negative breast cancer MDA-MB-231 cell line (metastatic) [110]. The p-STAT3 was then suppressed by tyrphostin AG490 (an inhibitor of the upstream activating Janus kinases), transfection with a dominant-negative form of STAT3, and with satraplatin (a tetravalent platinum derivate that inhibits STAT3 phosphorylation) [110]. These treatments downregulated p-STAT3 and sensitized the cancer cells to doxorubicin. Alantolactone (ALT), a sesquiterpene lactone component of Inula helenium, has anti-neoplastic effect against a variety of malignancies. Mechanistic research demonstrated that ALT abrogated STAT3 phosphorylation by promoting STAT3 glutathionylation. Reactive oxygen species scavenger NAC reverted ALT-mediated STAT3 glutathionylation and abrogation of STAT3 activation. With lung adenocarcinoma (A549 cell line), STAT3 inhibition by ALT enhanced chemosensitivity to doxorubicin via oxidative stress [111]. In the above three examples, genes transcribed by p-STAT3 dimers that are necessary for malignant cell behavior include BCL2L1 (Bcl-xL), BIRC5 (survivin), HIF1A, HIF1B (HIF-1), and MMP9 had their expression reduced [145]. Human osteosarcoma (SJSA-1) tumors when treated in vivo with doxorubicin undergoes significant growth suppression during a 14-day course of treatment; however, only 28% of the treated mice survived the 3-week xenograft study. The doxorubicin-associated toxicity was killing the mice. It is not clear if doxorubicin’s effect on normal cells caused this, or the induction of p-STAT3 in the SJSA-1 cells or a combination of both may have contributed to the death of the mice. When xenograft mice received doxorubicin with SBT-100, a STAT3 inhibitor, the osteosarcoma tumor growth was significantly suppressed, and survival of the mice increased to 71%. By some mechanism, SBT-100 was protecting the mice from doxorubicin toxicity. SBT-100 is a camelid derived single domain antibody that penetrates the cell membrane and blood brain barrier (BBB) rapidly in vivo and inhibits STAT3. SBT-100 has broad range of efficacy against many human malignancies such as ER + PR+ breast cancer, HER2-amplified breast cancer, triple negative breast cancer (TNBC), pancreatic cancer, prostate cancer, glioblastoma, osteosarcoma, fibrosarcoma, and leukemia [143]. Cisplatin is a platinum-based anti-neoplastic agent that binds DNA and inhibits its replication. It is used to treat ovarian, cervical, testicular, head and neck, colorectal, esophageal, bladder, lung, and breast cancers. Some mechanisms by which cisplatin resistance can develop include decreased drug import, increased drug export, increased DNA damage repair, increased drug inactivation by detoxification enzymes, and inactivation of cell death signaling, which occur within cancer cells [146]. Another mechanism of cisplatin resistance involves STAT3 overexpression. Sun et al have summarized utilization of STAT3 inhibition to reverse cisplatin induced resistance [52]. They summarized a large variety of STAT3 inhibitors which reverse cisplatin resistance in lung cancer, ovarian cancer, cervical cancer, breast cancer, laryngeal cancer, head and neck cancers, esophageal cancer, and hepatocellular carcinoma [52]. Morelli et al found, through network analysis and classification of proteome analysis of A549 cells (lung adenocarcinoma), that there were pathways altered in cisplatin resistant A549 cells. The resistance profile of these A549 cells included STAT3 overexpression. Furthermore, p-STAT3 is a marker of poor prognosis and cisplatin resistance in lung cancer. Generation of A549 STAT3 knockout cells resulted in impairment of clonogenic survival and mesenchymal phenotype in these A549 cells. These STAT3 knockouts do not develop cisplatin resistance nor over activation of mammalian target of rapamycin (mTOR) signaling with cis treatment. Moreover, the A549 knockout cells are more sensitive to mTOR inhibition by rapamycin [147]. Ovarian cancer can be effectively treated with paclitaxel; however, the development of resistance remains a major problem. Sheng et al have shown that STAT3 directly activates the pentose–phosphate pathway to cause pro-oncogenic behavior of paclitaxel resistant ovarian cancer [112]. Furthermore, they discovered that STAT3, p-STAT3, and glucose-6-phosphate dehydrogenase (G6PD) protein levels are increased in paclitaxel resistant cell lines versus paclitaxel sensitive cell lines. Blocking STAT3 decreased G6PD mRNA expression and increased the sensitivity of paclitaxel resistant ovarian cancer cells to paclitaxel. In addition, they demonstrated that STAT3 directly binds to the G6PD DNA promoter region and increases the expression of G6PD at the transcriptional level. In summary, their research reveals overexpression of STAT3 increases ovarian cancer colony formation, proliferation, and resistance to paclitaxel by increasing G6PD expression and pentose–phosphate metabolism flux [112]. With cervical cancer, paclitaxel is an important chemotherapeutic agent, but here too resistance to paclitaxel develops. Fan et al compared microRNA (miRNA) expression in cervical cancer cell lines to their paclitaxel resistant cervical cancer counterparts [113]. They found miR-125a to be a significantly decreased miRNA among paclitaxel-resistant cervical cancer cells and these cells also developed cisplatin resistance. Upregulating miR-125a sensitized resistant cervical cancers to paclitaxel in vitro and in vivo and to cisplatin in vitro. Importantly, they showed miR-125a increased sensitivity of cervical cancers to paclitaxel and cisplatin by decreasing STAT3. MiR-125a improved paclitaxel and cisplatin sensitivity by causing chemotherapy induced apoptosis. Clinically, miR-125a expression was linked to increased responsiveness to cisplatin combined with paclitaxel and this resulted in improved outcome. Their data suggests that miR-125a may provide a method which allows treatment of resistant cervical cancer. In addition, miR-125a may function as a biomarker for predicting resistance to cisplatin and paclitaxel in cervical cancer patients. Temozolomide, a chemotherapeutic that penetrates the BBB is used for the treatment of the heterogenous glioblastoma and anaplastic astrocytoma. Despite treatment with surgery, chemotherapy, and radiation, survival is maximum 15 months. Hyperactivated STAT3 has been demonstrated to modulate the behavior of gliomas and promote ADR and the STAT3 inhibitor pacritinib in combination with temozolimide has been shown to be effective in glioblastoma overwhelming STAT3/miR-21/PDCD4 signaling [114,115,148]. Moreover, the antipsychotic pimozide, a repurposed STAT3 inhibitor, reduces STAT3, triggers an autophagy-dependent, lysosomal type of cell death and improves survival in GBM cells [149,150]. A rational therapy for the treatment of glioblastoma would be the combination of temozolomide with the STAT3 inhibitor SBT-100, two anticancer compounds that penetrate the BBB [143]. Sun et al have described numerous studies that have shown that STAT3 activation can result in the failure of many different types of targeted therapies, especially EGFR targeted therapies [20,52]. For example, afatinib-induced STAT3 activation decreases the suppression of lung cancer cells to afatinib, and inhibiting IL-6R/JAK1/STAT3 signaling reverses the resistance. Blocking STAT3 can prevent ADR induced by EGFR inhibitors. Rhein, a lipophilic anthraquinone, sensitizes pancreatic cancer cells to erlotinib by inhibiting STAT3. Alantolactone, a natural sesquiterpene lactone, also sensitizes pancreatic cancer cells to erlotinib and also to afatinib by inhibiting STAT3 [116]. Silibinin, a polyphenolic flavonoid, is a direct inhibitor of STAT3, and it reverses ADR of crizotinib in ALK-rearranged lung cancer cells [68]. In addition, silibinin synergistically improves the response to sorafenib by hepatocellular carcinoma (HCC) cells by blocking STAT3 [151]. Pancreatic adenocarcinoma (PDAC) is a lethal malignancy that presents in late stages and responds poorly to current therapeutic regimens with an overall 5-year survival of 11%. It is characterized by an extensively dense fibrotic tumor stroma with poor vascularity, which hinder the intratumor delivery of anti-neoplastic agents [152]. Examining the response of PDAC to monotherapy with gemcitabine, dasatinib (Src inhibitor), and erlotinib (EGFR inhibitor) reveals that the upregulation of p-STAT3 is causing ADR. When combination therapy of dasatinib and erlotinib is used p-STAT3 is downregulated. This results in tumor collagen (types 1 and 4) and fibrosis to decrease within the tumor stroma in an orthotopic mouse model of PDAC. Here Dosch et al used PANC-1, which has a KRAS (G12D) mutation and constitutive expression of p-STAT3, and BxPC3, which has wild type KRAS and constitutive expression of p-STAT3. Interestingly, the addition of gemcitabine to combine with dasatinib and erlotinib therapy did not reverse the antifibrotic effects of this drug combination [69]. This two-drug combination inhibits the EGFR and Src signaling pathway and reduces p-STAT3. In turn, an increase in tumor vascularity occurs in vivo, and to determine this, treated PDAC were examined for CD31 (PECAM-1) by immunohistochemical (IHC) staining. CD31 is an endothelial marker that is associated with vascular normalization, maturity, and has been correlated with chemotherapeutic response in PDAC [11,28]. Monotherapy with dasatinib or erlotinib versus control showed no significant increase in CD31 positive staining; however, combined in vivo treatment with dasatinib plus erlotinib resulted in significant increase in CD31 positive endothelial cells. This finding was sustained with gemcitabine added to the two-drug combination. Dosch et al showed levels of gemcitabine is nearly undetectable in tumors treated with erlotinib or dasatinib monotherapy or in combination with gemcitabine [69]. When combination therapy with dasatinib plus erlotinib was administered, a marked increase in gemcitabine levels within PDAC tumors was detected. These findings demonstrated that combined Src and EGFR inhibition decreases p-STAT3 activity, which increases the microvascular density within PDAC tumors, which ultimately results in increased delivery of cytotoxic chemotherapy into the tumor mass. The above orthotopic PDAC studies were conducted on athymic nude mice. Transgenic PKT mice (Ptf1aCre/+; LSL-KrasG12D/+; Tgfbr2flox/flox) were used for in vivo studies to examine tumor volume with the pancreas and overall survival of combined Src and EGFR inhibition. This immunocompetent, spontaneous mouse model of PDAC underwent treatment with dasatinib, erlotinib, and gemcitabine either alone or in combination. This therapy was continued for 4 weeks, after which the mice were sacrificed and the PDAC tumors harvested for histo-pathology evaluation. In PKT tumors, dasatinib plus erlotinib, and dasatinib plus erlotinib with gemcitabine treatments significantly reduced tumor weight at the end of the study. Furthermore, stromal remodeling of the PDAC tumors occurred as it did in the orthotopic tumors with decreased stromal fibrosis, decreased collagen type 1 and 4, increased microvascular density, and increased number of CD31 positive endothelial cells. Moreover, p-STAT3 levels were significantly decreased with combined treatment of dasatinib plus erlotinib, and dasatinib plus erlotinib with gemcitabine in the PKT tumor samples. These two combination regimens also prevented the progression of PDAC tumors in the PKT mice and increased their overall survival. Furthermore, they have previously shown tumor cell-derived IL1α induces stromal-derived IL-6, reciprocally activating tumor cell-autonomous STAT3 signaling, a well-known indicator of chemoresistance and oncogenic signaling in PDAC [126,133,153]. Lee et al performed extensive and elegant work on defining STAT3 as an escape mechanism for many cancers treated with targeted pharmaceutical therapeutics [154]. They discovered that many drug treated “oncogene addicted” malignancies use a positive feedback loop resulting in STAT3 hyperactivation. As a result, promoting cancer cell proliferation, survival, and decreasing response to targeted drug therapy. This was noted in malignant cells driven by different activated kinases such as HER2, EGFR, MET, ALT, and mutant KRAS [10,13]. MEK inhibition resulted in autocrine activation of STAT3 via FGFR and JAK kinases. Importantly, simultaneous drug suppression of MEK, JAK, and FGFR increased tumor regression. Their data implies that blocking the STAT3 feedback loop enhances the response to a wide range of pharmacologic therapeutics that inhibit pathways of oncogene addiction. Monoclonal antibodies (mAbs) to cell surface receptors, [e.g., human epidermal growth factor receptor 2 (HER2)], and to extracellular protein, e.g., vascular endothelial growth factor (VEGF), have greatly improved the treatment of patients with cancer. HER2 is a receptor tyrosine kinase (RTK) that controls differentiation and cell growth signaling pathways. In about 20–25% of breast cancers and in 30% of gastric cancers, HER2 is significantly overexpressed. This results in a very aggressive cancer phenotype and poor prognosis. Trastuzumab is a humanized mAb that binds to an extracellular domain of the HER2 molecule and inhibits its function. It provides significant benefit in patient outcome; however, treatment resistance does develop in some patients. Li et al found that p-STAT3 is hyper-expressed in de novo and acquired trastuzumab-resistant gastric cancer and breast cancer cells [96]. Here, increased STAT3 activation and signaling is caused by elevated levels of IL-6, fibronectin (FN), and EGF in an autocrine manner. This leads to ADR by upregulating the expression of MUC1 and MUC4. Both are downstream targets of p-STAT3. MUC1 and MUC4 can induce trastuzumab resistance by maintaining HER2 activation and masking of trastuzumab to prevent HER2 binding, respectively. Knocking down STAT3 expression and blocking STAT3 function with a small molecule inhibitor abrogated STAT3 activation, which allowed trastuzumab sensitivity of resistant cells in vitro and in vivo [89]. Trastuzumab–emtansine (T-DM1) is an antibody drug conjugate made with the trastuzumab mAb linked to a cytotoxic moiety DM1 (a derivative of maytansine) and it was developed to overcome ADR associated with trastuzumab use. T-DM1 has demonstrated great efficacy clinically; however, ADR to its use has emerged and is a significant problem for patients. Wang et al used BT-474/KR cells, a T-DM1 resistant cell line developed from HER2-positive BT-474 breast cancer cells, to show that STAT3 activation mediated by leukemia inhibitory factor receptor (LIFR) overexpression results in T-DM1 resistance. Furthermore, they demonstrated STAT3 inhibition sensitizes resistant cell to T-DM1 both in vitro and in vivo [90]. Anti-VEGF treatments help several types of cancer patients, but ADR can develop with therapy. There are several VEGF pathway inhibitors, which include bevacizumab (anti-VEGF mAb), aflibercept (decoy receptor that binds VEGF-A), and ramucirumab (anti-VEGF receptor 2 mAb), which inhibit tumor growth in preclinical cancer models and improve cancer patients’ survival. Eichten et al developed cell lines from anti-VEGF resistant tumor xenografts and one called A431-V epidermoid carcinoma developed partial resistance to aflibercept [87]. A431-V tumors secreted much more IL-6 and produced higher amounts of p-STAT3 compared to parental tumors. Combined inhibition of IL-6 receptor (IL-6R) and VEGF resulted in enhanced suppression of A431-V tumors. In addition, inhibition of IL-6R increased the suppression of DU145 prostate cancer cells using aflibercept. These DU145s have high endogenous IL-6R activity. These data indicate that ADR to anti-VEGF therapy is mediated in part by increased IL-6/STAT3 signaling in cancer cells. Inhibition of IL-6 signaling on cancer cells can overcome this ADR. Immune checkpoint inhibitors (ICIs) provide significant benefit to some cancer patients and improve survival in a minority of patients. Moreover, some might even be cured [99]. Tumor-intrinsic resistance is the reason for the lack of response [97]; however, when ICIs are used for the first time, less than 45% respond and most responders eventually develop ADR [101]. ADR has been reported in several types of cancer patients and animal models treated with ICIs due to overactivation of STAT3. This undesirable effect has been observed in the case of anti-PD-1, anti PD-L1, and anti-CTLA-4 antibodies. STAT3 can directly or indirectly regulate these immune checkpoint molecules. There is a clear relation between STAT3 and PD-1, PD-L1, and PD-L2 [6]. STAT3 can increase their expression by direct binding to their promoters [105]. STAT3 binds to the CD274 (PD-L1) gene promoter and is required for CD274 gene expression. The NPM/ALK carrying T cell lymphoma (ALK + TCL) cells strongly express PD-L1 that is regulated by STAT3 [100]. These investigators at that time were already suggesting that the treatment of this lymphoma should combine inhibition of both NPM/ALK and STAT3. Under hypoxic conditions overactivated STAT3 interacts with PD-L1 and enables its nuclear translocation by the importin α and β pathways [95]. It was shown that in nasopharyngeal carcinoma, LMP1 upregulates PD-L1 through STAT3, AP-1, and NF-kB [94]. In gastric cancer, the suppressor gene miR-502-5p reduces PD-L1 expression through inhibition of the CD40/STAT3 pathway [106]. To obtain better results, combinations must be used. Nivolumab plus ipilimumab improved outcomes in 43% of patients with metastatic renal cell carcinoma but ADR was found in the rest of patients receiving this combination. Nonresponders exhibited significant increases in cytokines and higher levels of p-STAT3. The addition of a STAT3 inhibitor to the combination of the two ICIs showed significant tumor growth inhibition in a syngeneic model [92]. These investigators suggest that anti-PD-1 therapy administered along with a STAT3 inhibitor is a rational combination and should be further explored. In patients with melanoma, less than 20% respond to ICIs. Studies in a melanoma mouse model showed that the addition of STAT3 inhibitors to an ICI increases the response to the ICI-resistant tumor. These data suggest that the combination of ICIs with STAT3 inhibitors might be effective in patients with melanoma [96]. In drug-resistant BRAF-mutant melanoma, a combined blockade of STAT3 and PD-1 overcomes resistance [8,14,19,98]. In PDAC, the addition of MEK inhibitors plus STAT3 inhibitors to Nivolumab overcomes ICI resistance [93]. Ipilimumab is efficacious only in a subset of patients with prostate cancer. In a syngeneic prostate cancer mouse model, the combination of an anti-CTLA-4 with a STAT3 inhibitor significantly inhibited tumor growth and enhanced survival possibly by blocking STAT3 mediated resistance mechanisms such as Tregs in the immunosuppressive environment. These investigators raise the possibility that STAT3 inhibition in combination with anti-CTLA-4 could constitute a future novel treatment approach in advanced prostate cancer [103]. In summary, it appears that combining a STAT3 inhibitor with an ICI is an attractive way to prevent the development of ADR and increase their efficacy [92,93,96,106]. Several mechanisms involved in the development of ADR have been studied in animal models of several types of cancer and in patients. As result of the unique cytoplasmatic location of the STAT3 signaling pathway downstream of major pathways involved in cancer and the significant crosstalk that occurs among them, it is activated by the inhibition of many other pathways and in most types of cancers becomes the constitutively activated p-STAT3. This explains why the administration of any anticancer therapeutic and the inhibition of its specific pathway through the crosstalk relations induces the production of p-STAT3 that appears to be responsible for the various changes that end in the development of ADR and the loss of their therapeutic effects. p-STAT3 alters autophagy and anoikis, two apoptotic processes that normally eliminates unwanted cells and the lack or reduction of their actions contribute to ADR and the progression of the tumor or hematological cancer. The physiological relation between STAT3, and p53, or NNMT is altered when STAT3 becomes constitutively activated. These alterations act as factors contributing to the development of ADR. Participation of STAT3 in ARD after administration of any type of anticancer therapy—including the newer targeted agents such as the eight new ICIs and the two KRAS inhibitors—indicates that the use of a STAT3 inhibitor should be part of any rational pharmacological treatment including radiotherapy. The inclusion of a STAT3 inhibitor in anticancer regimens increases their efficacy and most likely prevents the development of ADR. Some natural products, such as curcumin, are STAT3 inhibitors. A limitation of curcumin is the low oral bioavailability; however, new delivery technologies have improved it oral absorption [155]. When ADR is already present, the administration of a STAT3 inhibitor reverses it and restores the efficacy of the anticancer therapeutic. Many investigators have shown that STAT3 inhibition can reverse ADR, restore the efficacy of anticancer agents [118,156], enhance anti-cancer immune responses, and rescue the suppressed immunologic system [157]. The treatment of cancer usually requires combination therapy, and two or more agents might be needed. As soon as direct STAT3 inhibitors reach the market, a rational combination for the pharmacologic treatment of cancer patients should include a STAT3 inhibitor to prevent and or reverse ADR and thereby increase the efficacy and duration of the therapeutic regimen.

PMC10002576

Chen Yang,Mark Stephen Rybchyn,Warusavithana Gunawardena Manori De Silva,Jim Matthews,Katie Marie Dixon,Andrew J. A. Holland,Arthur David Conigrave,Rebecca Sara Mason

The CaSR Modulator NPS-2143 Reduced UV-Induced DNA Damage in Skh:hr1 Hairless Mice but Minimally Inhibited Skin Tumours

03-03-2023

ultraviolet radiation (UV),photoprotection,1a,25-dihydroxyvitamin D3 (1,25D),cyclobutane pyrimidine dimer (CPD),8-hydroxy-2′–deoxyguanosine (8-OHdG),calcium sensing receptor (CaSR),photocarcinogenesis,cyclic AMP response element binding factor (CREB),squamous cell carcinoma (SCC)

The calcium-sensing receptor (CaSR) is an important regulator of epidermal function. We previously reported that knockdown of the CaSR or treatment with its negative allosteric modulator, NPS-2143, significantly reduced UV-induced DNA damage, a key factor in skin cancer development. We subsequently wanted to test whether topical NPS-2143 could also reduce UV-DNA damage, immune suppression, or skin tumour development in mice. In this study, topical application of NPS-2143 (228 or 2280 pmol/cm2) to Skh:hr1 female mice reduced UV-induced cyclobutane pyrimidine dimers (CPD) (p < 0.05) and oxidative DNA damage (8-OHdG) (p < 0.05) to a similar extent as the known photoprotective agent 1,25(OH)2 vitamin D3 (calcitriol, 1,25D). Topical NPS-2143 failed to rescue UV-induced immunosuppression in a contact hypersensitivity study. In a chronic UV photocarcinogenesis protocol, topical NPS-2143 reduced squamous cell carcinomas for only up to 24 weeks (p < 0.02) but had no other effect on skin tumour development. In human keratinocytes, 1,25D, which protected mice from UV-induced skin tumours, significantly reduced UV-upregulated p-CREB expression (p < 0.01), a potential early anti-tumour marker, while NPS-2143 had no effect. This result, together with the failure to reduce UV-induced immunosuppression, may explain why the reduction in UV-DNA damage in mice with NPS-2143 was not sufficient to inhibit skin tumour formation.

The CaSR Modulator NPS-2143 Reduced UV-Induced DNA Damage in Skh:hr1 Hairless Mice but Minimally Inhibited Skin Tumours The calcium-sensing receptor (CaSR) is an important regulator of epidermal function. We previously reported that knockdown of the CaSR or treatment with its negative allosteric modulator, NPS-2143, significantly reduced UV-induced DNA damage, a key factor in skin cancer development. We subsequently wanted to test whether topical NPS-2143 could also reduce UV-DNA damage, immune suppression, or skin tumour development in mice. In this study, topical application of NPS-2143 (228 or 2280 pmol/cm2) to Skh:hr1 female mice reduced UV-induced cyclobutane pyrimidine dimers (CPD) (p < 0.05) and oxidative DNA damage (8-OHdG) (p < 0.05) to a similar extent as the known photoprotective agent 1,25(OH)2 vitamin D3 (calcitriol, 1,25D). Topical NPS-2143 failed to rescue UV-induced immunosuppression in a contact hypersensitivity study. In a chronic UV photocarcinogenesis protocol, topical NPS-2143 reduced squamous cell carcinomas for only up to 24 weeks (p < 0.02) but had no other effect on skin tumour development. In human keratinocytes, 1,25D, which protected mice from UV-induced skin tumours, significantly reduced UV-upregulated p-CREB expression (p < 0.01), a potential early anti-tumour marker, while NPS-2143 had no effect. This result, together with the failure to reduce UV-induced immunosuppression, may explain why the reduction in UV-DNA damage in mice with NPS-2143 was not sufficient to inhibit skin tumour formation. Skin cancers can be categorized into three main types: (i) basal cell carcinoma (BCC); (ii) squamous cell carcinoma (SCC), both of which arise from keratinocytes; and (iii) melanoma. Pathological changes in skin, including ultraviolet radiation (UV)-induced DNA damage [1,2,3], mutagenesis [4], inflammation [5], and immunosuppression [6,7] can ultimately lead to photocarcinogenesis. UV not only directly induces DNA lesions such as cyclobutane pyrimidine dimers (CPDs) and (6–4) photoproducts [8], but also induces indirect biological damage targeting DNA, protein, and lipids via the production of reactive oxygen species (ROS) and nitric oxide products, forming 8-hydroxy-2′-deoxyguanosine (8-OHdG) as a marker of oxidative DNA damage [9]. UV has potent immunosuppressive effects that promote tumour development [7,10,11,12]. Repetitive UV-induced epidermal thickening and pigmentation production together protected mice [13] and humans from subsequent UV challenges, with 75% less erythema and 60% less DNA damage in skin [14,15]. This thickening of the skin as a result of keratinocyte differentiation and may be more protective than melanogenesis (pigmentation production) in response to UV, at least in some populations [16]. Calcium concentration is believed to act as a switcher between proliferation and differentiation of keratinocytes [17]. This is consistent with a well-defined gradient for total calcium that increases from the basal to the outermost layers of the epidermis [18]. The responses of keratinocytes to extracellular calcium ion concentrations (Ca2+o) and the maintenance of systemic calcium homeostasis are mainly controlled by the calcium-sensing receptor (CaSR), a member of family C of the G protein-coupled receptors (GPCR) [19,20]. There are commercially available small molecule allosteric agents, for example, NPS-2143 works as an antagonist that reduces CaSR activity to block the increase of Ca2+i [21,22,23,24,25]. NPS-2143 has been used in an attempt to promote a brief secretion of parathyroid hormone in plasma for treatment of osteoporosis [22]. Previously we reported that CaSR knockdown or exposure to the CaSR negative allosteric modulator NPS-2143 protected human keratinocytes in culture against UV-induced DNA damage at a similar level to the known photoprotective agent, 1,25-dihydroxyvitamin D3 (1,25D) [26]. This photoprotective activity of NPS-2143 was attributed at least in part to enhanced DNA repair and to reduction in ROS [26]. Immunosuppression, along with UV-induced DNA lesions, is a key factor leading to photocarcinogenesis [27]. Topical 1,25D has been shown to protect mice from UV-induced CPDs, apoptotic sunburn cells, and UV-induced immunosuppression, and to reduce UV-induced skin tumours [28,29,30,31] as well as chemically-induced skin tumours [32]. The use of albino hairless (Skh:hr1) mice exposed to chronic UV is accepted as a reliable model of photocarcinogenesis [33,34,35]. While cultured primary keratinocytes provide a powerful approach for studying epidermal biology, they imperfectly model the multi-cell types and structural order of living epidermis [36]. Thus we aimed to investigate, for the first time in a mouse model, whether manipulation of the CaSR by its negative allosteric modulator NPS-2143 would protect against DNA damage in mouse epidermis after acute UV exposure. We also wanted to examine if topical treatment of NPS-2143 would reduce UV-induced skin inflammation and immune suppression, as well as in response to a chronic UV-exposure, whether it would reduce UV-induced skin tumours in comparison with the positive control, 1,25D. Acute UV irradiation generated CPDs, oxidative DNA damage 8-OHdG (Figure 1), and sunburn cells (Figure 2) in mouse skin. The photoprotective hormonal form of vitamin D, 1,25D [29,37,38,39,40,41], was used as the positive control in these experiments. All agents in all experiments were applied topically immediately after exposure to solar-simulated UV (ssUV). Minimal staining in SHAM skin, particularly of 8-OHdG, indicates basal damage of the nuclei. In female mice, topical NPS-2143 at 2 concentrations, 228 pmol/cm2 and 2280 pmol/cm2 effectively reduced both CPD (p < 0.05, F(2.062, 19.25) = 15.64) and 8-OHdG (p < 0.05, F(1.537, 11.78) = 34.58) (Figure 1a–d). Topical NPS-2143 also reduced UV-induced sunburn cells (p < 0.01, F(2.044, 15.67)= 32.04) which are apoptotic keratinocytes with characteristic pyknotic nuclei and eosinophilic cytoplasm [42] (Figure 2a,b). In male Skh:hr1 mice, significant reduction in UV-induced CPD was only seen after treatment with a high dose of NPS-2143, 2280 pmol/cm2 (p < 0.01, F(2.190, 18.98) = 6.738) (Figure 1e–h). Both concentrations of this agent, however, as well as 1,25D, significantly protected against oxidative DNA damage, 8-OHdG (p < 0.01, F(1.284, 14.55) = 8.726), in males (Figure 1f,h), and against sunburn cells (p < 0.01, F(2.663, 33.74) = 53.07) (Figure 2c,d). After exposure to 3 minimal erythemal doses (MED) of solar-simulated UV, where a MED is defined as the lowest dose of UV which produces a mild reddening of the skin at 24 h, the mice developed skin edema [31]. In this study, skinfold thickness increased daily, reaching a maximum at day 4 post-UV, then decreased gradually (Figure 3a). On the 4th day, UV-induced edema was significantly reduced in the presence of 1,25D (11.4 pmol/cm2) (p < 0.05) or NPS-2143 (2280 pmol/cm2) (p < 0.05), compared to the vehicle-treated control mice (F(1.531,9.187) = 8.857). In order to study how UV exposure affects contact hypersensitivity to oxazalone, female mice were exposed to ssUV or SHAM, then treated topically with the various agents. One week later, all the mice were sensitized with 2% oxazolone applied to the non-irradiated abdominal skin. The mice were then challenged one week after this, by topical application of oxazalone to the ears to trigger swelling. Ear thickness measurements were taken before the challenge and again 18 h later. The average ear swelling expressed as the difference between ear thickness measured before and after challenge (at 18 h) in the non-UV exposed (SHAM) vehicle-treated mice was 293 ± 45 microns, and there were no differences among all non-irradiated groups (Figure 3b). In the UV-irradiated vehicle-treated mice, the average ear swelling of vehicle-treated mice was 155 ± 30 microns, indicating significant suppression of the immune response. With topical treatment with 1,25D, average ear swelling after UV was 217 ± 52 microns (Figure 3b), consistent with partial restoration of the contact hypersensitivity response. Though this swelling in response to oxazolone was smaller than in the SHAM with 1,25D-treated mice, the response was significantly better than in the vehicle-treated UV-exposed mice (p < 0.05, F (3.172, 28.55) = 21.73). Mice treated with NPS-2143 and UV had a measured average ear swelling of 177 ± 63 microns, not significantly different from vehicle-treated, UV exposed mice (Figure 3b). When calculated as a percent immune suppression after UV [31], the values were 52% immune suppression in the vehicle-treated group, 26% in the 1,25D-treated group (p < 0.05 vs vehicle-treated), and 39% in the mice treated with NPS-2143 (n.s. vs vehicle-treated mice) (F (1.582, 21.36) = 3.405) (Figure 3c). Albino hairless Skh:hr1 mice develop papillomas and then SCC after 10 weeks of chronic ssUV exposure [31,34,43]. During the 40 weeks of study, tumours normally appeared as small papillomas which gradually increased in diameter (Figure 4a). Papillomas are a benign outgrowth of skin in mice, comparable to the onset of actinic keratoses (AK) in humans [44]. A proportion of these papillomas showed signs of progression towards malignancy. These may be identified grossly and verified histologically as squamous cell carcinomas in later weeks (Figure 4a) [34]. Tumour latency The onset of detectable tumour (papilloma) formation in mice varied between treatment groups. As shown in Figure 4b, the latency in the vehicle-treated group was 24.0 ± 1.0 weeks. A significantly increased latency of 33.6 ± 2.5 weeks (p < 0.0001, F(2, 35) = 2.560) was seen in 1,25D-treated mice (11.4 pmol/cm2). The average latency for NPS-2143-treated mice (2280 pmol/cm2) was 22.4 ± 1.0 weeks, which was not significantly different from the vehicle control (Figure 4b). Tumour multiplicity Tumour multiplicity including both papillomas and SCCs was calculated at each weekly time point, as the average number of tumours per tumour-bearing mouse. Figure 4c shows tumour multiplicity throughout the 40-week study. Vehicle- and NPS-2143 (2280 pmol/cm2)-treated mice showed a steady increase in tumour multiplicity from week 16 to week 40, while 1,25D-treated mice (11.4 pmol/cm2) had remarkably lower tumour multiplicity. Compared to vehicle-treated mice, tumour multiplicity was significantly reduced in the 1,25D-treated group at all week-points assessed (p < 0.05 at week 20, p < 0.01 at week 25, 30 and 35, p < 0.005 at week 40, F(2, 45) = 2.255). However, there was no significant difference between NPS-2143 treated and vehicle-treated mice. Tumour incidence Progressive total tumour incidence including both papillomas and SCCs was calculated each week as the percentage of mice in each group bearing at least one tumour, as shown in Figure 4d. The incidence data were analysed statistically using a Mantel–Haenszel log-rank test (Mantel–Cox test) [45], in which all treatments were compared to vehicle-treated mice at 27 weeks and after (Table 1a). This analysis reveals whether there was a difference in the risk of developing a tumour. Mice treated with 1,25D (11.4 pmol/cm2) had significantly reduced tumour incidence compared with the vehicle-treated group throughout the entire experiment (Figure 4d green dotted line, Table 1a, Mantel–Cox test Chi-square Value = 27.09, df = 1). NPS-2143-treated (2280 pmol/cm2) mice demonstrated a time point-dependent increase in total tumour incidence compared to the vehicle-treated group at 27 weeks after the first irradiation, but by 28 weeks and over the subsequent period until 40 weeks, there was no significant difference (Figure 4d blue dotted line, Table 1a, Mantel–Cox test Chi-square Value = 4.061, df = 1). Mice developed squamous cell carcinomas (SCCs) throughout the study from 18 weeks. The SCC-only incidence is shown in Figure 4e. Mice treated with 1,25D (11.4 pmol/cm2) had significantly reduced SCC incidence compared with the vehicle control group throughout the experiment (Figure 4e green dotted line, Table 1b, Mantel–Cox test Chi-square Value = 5.355, df = 1). Only one mouse in the group of 18 (5.5%) treated with 1,25D developed an SCC at week 32 and this was still present at the end of the study. NPS-2143-treated mice had a significantly lower risk of developing SCC (5 out of 18, 27.8%) compared to the vehicle-treated group (8 out of 18, 44.4%) at the 24th week post-irradiation (Figure 4e Blue dotted line, Table 1b, Mantel-Cox test Chi-square Value = 22.09,df = 1). However, from the 25th week until the end of the experiment, there was no significant difference between the risk of SCC in NPS-2143- and vehicle-treated mice (Table 1b). Phosphorylation of cyclic AMP response element binding protein (CREB) as a predictor of anti-tumour activity After UV exposure, CREB phosphorylation in epidermal cells increases and this has been proposed as a marker of tumour promoting activity [46]. In this study, 1,25D significantly reduced the risk of developing papillomas and SCC compared with the vehicle-treated group, while NPS-2143 had no overall effect on tumour or SCC incidence (Figure 4e, Table 1b). Phosphorylation of CREB after UV, 1,25D, or NPS-2143 was studied in normal human keratinocytes. Negligible basal phospho-CREB (p-CREB) was seen in non-irradiated keratinocytes (SHAM) (Figure 4f). In cultured human keratinocytes, exposure to ssUV increased p-CREB measured 90 min after exposure (Figure 4f). Treatment of the cells immediately after UV with 1,25D significantly reduced p-CREB while treatment with NPS-2143 had no effect whether expressed as a function of tubulin as loading control (Figure 4f, F (3, 8) = 0.8378, and Figure S1a,) or as a function of total CREB (Figure S1b). A summary of the main differences between responses to the positive control 1,25D and NPS-2143 is shown below (Table 2). In this study, the CaSR negative allosteric modifier NPS-2143, like the positive control 1,25D, when applied topically immediately after ssUV, effectively reduced UV-induced DNA lesions of CPD and 8-OHdG in female Skh:hr1 mice. Both NPS-2143 and 1,25D reduced oxidative DNA damage in male mice and at the higher concentration, NPS-2143 also reduced CPD in male mice. These results are consistent with the findings from our study using keratinocytes in primary culture from male human donors [26]. This is a discovery of a photo-protective role for NPS-2143, entirely different from its better recognised role as a therapeutic agent for raising parathyroid hormone levels. NPS-2143 negatively modulates the affinity of the CaSR for extracellular Ca2+, thereby reducing its activity [21,22,23,24,25]. In order to better discriminate the role of the CaSR in this study, it would have been useful to examine a CaSR antagonist (NPS2390 or Calcium-Sensing Receptor Antagonists I), but these studies were beyond the resources available for this work. Sunburn cells and apoptotic keratinocytes were observed as soon as 3 h after acute exposure to UVB [47], despite being cells that undergo programmed cell death as a result of extensive and irreparable DNA damage [42]. It is reasonable to propose that reduced DNA damage, along with increased DNA repair [21] in the presence of NPS-2143, led to fewer apoptotic keratinocytes in mouse skin. In the meantime, we also observed improved survival of human keratinocytes in culture after UV exposure in the presence of NPS-2143 (Figure S2). It is likely that reduced generation of ROS, as previously reported with NPS-2143 after UV [21], contributed to reduced apoptosis. While sunburn cells are an index of apoptosis, analysis of more specific markers of apoptosis such as caspase3/7 or cleaved PARP in the mouse tissue would indicate early stages of apoptotic events and could help to elucidate the mechanism. The reductions in 8-OHdG in male mice with 1,25D or NPS-2143 were similar to those seen in female mice. However, only the higher dose of NPS-2143 reduced CPD in male mice. Resistance in male mice to protection against UV-induced CPD in the presence of 1,25D has been previously reported in a separate study [47]. In that study, we demonstrated that the estrogen receptor-β (ER-β), the only estrogen receptor present in female mouse skin, seemed likely to be involved in reductions in CPD with 1,25D, since treatment with an ER-β antagonist or the use of female ER-β knockout mice reduced the response to 1,25D [47]. The current results indicate less effective protection against UV-induced CPD by a negative allosteric modulator of the CaSR. Whether this is also related to the presence of ER-β in female mouse skin or some other sex-related difference is an interesting question but was beyond the scope of the current study. A simple explanation for the reduced effectiveness of the lower dose of NPS-2143 could be that male mice have approximately 20% thicker skin than female mice, regardless of UV [48]. DNA damage is a major contributor to UV-induced immune suppression [49] and susceptibility of male mice or humans to UV-induced immune suppression is greater than in their female counterparts [50,51]. Male mice are more susceptible than females to photocarcinogenesis [52], whereas incidence and mortality of skin cancers is greater in men [53,54]. Given resource limitations, the increased potential for male mice to fight and scratch, producing skin damage which would interfere with observations [55], meant that longer studies of skin edema, immune suppression, and tumour development after UV were only undertaken in female mice. Topical NPS-2143, like 1,25D, produced a significant decrease in skin edema of mice, reflecting reduced inflammation after ssUV. UVR induces immediate and sustained production in NO in the skin [56,57,58,59] promoting the secretion of inflammatory mediators such as IL-6 [60]. A major limitation of the study is that it was not possible under the circumstances to examine a cytokine profile of mouse skin tissue before and after UV with or without NPS-2143 or 1,25D. Nevertheless, from the literature, there is evidence that NPS-2143 reduces NLRP3 inflammasome activation [61,62,63], overproduction of NO [64,65], the pro-inflammatory cytokine IL-6, and more [66,67,68,69]. These observations could explain the ability of NPS-2143 to reduce inflammation in mouse skin on day four after UV. Somewhat surprisingly, despite a reduction in skin inflammation after UV, treatment with NPS-2143 had no effect on UV-induced immune suppression. Both DNA damage and increased IL-6 are important promoters of UV-dependent immune suppression [70]. Yet NPS-2143, like 1,25D, reduced DNA damage and, from the literature, also reduces IL-6 [66,67,68,69]. UV can directly damage antigen-presenting cells and promote the production of immunosuppressive cytokines such as IL-10 and IL-4 [71,72,73]. IL-10 is an anti-inflammatory cytokine and a potent immunosuppressant [74]. Secreted by UV-irradiated keratinocytes [75,76] and regulatory T cells [77], IL-10 not only prevents T cell expansion and activation but can also suppress other antigen-presenting cells [74]. It has been shown that reductions in CPD using liposomes containing T4N5 endonuclease led to reduced UV-immune suppression due to decreases in UV-upregulated IL-10 and TNF-α at both the mRNA and protein levels [78]. Furthermore, IL-10−/− mice were protected against photocarcinogenesis [79]. Though not tested in this study, IL-10 was increased with NPS-2143 treatment in rats [66,69]. This may explain why NPS-2143 failed to prevent UV-induced immunosuppression, though it protected against CPD and inflammation. Skin tumour development depends on a combination of DNA damage, inflammation, and immunosuppression [49,80]. CPD are a major contributor to UV-induced mutations, so reduced CPDs might lead to fewer UV-induced mutations [80] and thus fewer tumours. Furthermore, enhanced repair of CPDs has been shown to reduce skin cancer incidence in mice [81] and humans [82] and we previously found that NPS-2143 increased DNA repair in keratinocytes [26]. Based on these findings, it seemed possible that NPS-2143 would have some protective capacity at an early stage of photocarcinogenesis due to its ability to reduce DNA damage and inflammatory reactions in vivo. In the photocarcinogenesis study, however, a single concentration of NPS-2143 (2280 pmol/cm2) was not superior to vehicle, either in time to develop the first tumour (including benign papilloma) or in the total number of tumours per mouse. Although NPS-2143 significantly reduced SCC incidence at 24 weeks, the effect did not persist. Although the failure of NPS-2143 to prevent UV-induced immune suppression may explain its failure to prevent tumours in the chronic UV study, other factors may be involved. Cyclic AMP Response Element Binding protein (CREB) is a transcription factor essential for basic cellular function and homeostasis [83]. CREB is activated by phosphorylation at Ser133 by various kinases [83,84]. CREB overexpression supports growth and progression in various cancers [85,86,87,88,89]. CREB activation promotes enhanced cell proliferation, dysregulation of differentiation and reduced sensitivity to apoptosis and metastasis, particularly in melanoma [87,90] and SCC [88,91]. Using human keratinocytes, we observed that UV exposure increased phosphorylation of CREB at Ser 133 (phospho-CREB Ser133). While it would have been useful to verify the CREB and p-CREB changes in mouse skin, this was beyond the scope of the study and is a limitation. Our results in human keratinocytes are consistent with a recent study using reverse phase protein microarray analysis, which reported that p-CREB Ser133 was significantly activated at 1 h, 5 h, and 24 h after a single acute dose of 2MED UV in human skin [92] and with the report of increased p-CREB in mouse skin [46]. It has been argued that p-CREB is important in the initiation of papilloma formation, while other transcription factors such as CCAAT/enhancer binding protein (C/EBP)–B [93,94] control later stages of tumour growth and Activator Protein 1 (AP1) [95,96,97] maintains tumour identity. In a study of SCC, shRNA-mediated knockdown of CREB resulted in a significant increase in G2 phase arrest and a reduction in tumorigenic activity [91]. These authors identified that a key transcription factor complex, CREB and RFX1, which binds in the nucleus and is stabilized by CCAR2, is required to maintain proliferation in SCC [91]. Overexpression of CREB in a human squamous carcinoma cell line SCC13 remarkably increased its colony forming ability via a β-catenin-dependent pathway [88]. These studies suggest critical functions of CREB not only in the initial stage of papilloma formation but also in the development of neoplastic characteristics of SCC. Treatment of keratinocytes with 1,25D reduced UV-induced expression of p-CREB, fitting with its ability to protect mouse skin from developing both papillomas and SCC in the photocarcinogenesis study. NPS-2143, on the other hand, did not reduce UV-upregulated p-CREB. This may be part of the explanation for its inability to reduce tumour incidence, apart from its failure to decrease UV-induced immunosuppression. Bikle et.al reported that double knockout of the vitamin D receptor and CaSR in the epidermis leads to spontaneous SCC formation in mice without any induction by UV, which was not observed in mice with deletion of either gene alone [98,99]. Those studies did not involve UV exposure. This is the first study to investigate whether negative modulation of the CaSR in skin alters responses to UV. NPS-2143 reduced two types of DNA damage in epidermal cells as well as skin inflammation to a similar extent as 1,25D, a known photo-protective agent (Figure 1, Figure 2 and Figure 3). However, NPS-2143 did not ameliorate UV-induced immune suppression (Figure 3). This latter observation, together with the failure of NPS-2143 to reduce post-UV CREB phosphorylation, probably explain the limited effect of this compound on skin tumour formation after ssUV (Figure 4). It is possible that the reduction in UV-induced DNA damage including oxidative damage by NPS-2143 may indicate an anti-aging effect [100]. These novel findings may lead to new research directions on the relationship between UV and the CaSR. The in vivo studies were approved by the Animal Ethics Committee of the University of Sydney (Approval number: 2015/794) and conformed to ARRIVE criteria. Skh:hr1 hairless albino mice, originally from Charles River (Wilmington, MA, USA), were from an in-house colony maintained at the University of Sydney. All Skh:hr1 hairless mice were housed in groups in wire-topped plastic boxes at an ambient temperature of 23–25°C under gold lighting (F40GO tubes, General Electric Co., Hobart, TAS, Australia) that does not emit UV radiation, and fed with Gordon Rat and Mouse Pellets (Yanderra, NSW, Australia) and tap water ad libitum. Male and female Skh:hr1 mice that were aged-matched in groups were used for experiments [31]. Mice were not allowed to be housed singly for this study but were housed in groups. Female mice are less prone to fighting than male mice and the fighting produces skin damage and artefacts [55]. For this reason, it is possible to study both female and male mice for DNA damage within hours after a UV exposure; however, the use of female mice for studies of skin edema or contact hypersensitivity conducted over 7 days and 16 days, respectively, or photocarcinogenesis (over 40 weeks) is preferred (Figure 5). As previously established, the minimum erythemal dose (MED) of UV with this source for Skh:hr1 mice was 1.33 kJ/m2 UVB and 23.7 kJ/m2 UVA [31,101]. UV-irradiated mice were subjected to a single dose of 3 MED of UV (UVB value at 3.99 kJ/m2) for acute and immunosuppression studies. In the chronic photocarcinogenesis study, mice were subjected to 5 days of 0.75 MED followed by 5 days/week of 1 MED, for a total of 10 weeks (Figure 5). Mice were treated topically over approximately 7 cm2 on the irradiated dorsal skin with 100 μL of vehicle only, or vehicle containing 1,25D (Sapphire Bioscience Pty Ltd., Redfern, NSW, Australia), or NPS-2143 2143 (HY-1007 MCE®, Medchem Express, Monmouth Junction, NJ, USA) immediately after irradiation, as previously described [31]. The compounds (1,25D and NPS-2143) were freshly diluted in spectroscopic grade ethanol (Merck, Darmstadt, Germany), combined with propylene glycol (Sigma-Aldrich, St. Louis, MO, USA) and MilliQ water at a ratio of 2:1:1 (v/v/v). Vehicle (base lotion) was combined, ethanol:propylene glycol:water 2:1:1 v/v [31]. The dose of NPS-2143, equivalent to 20× (228 pmol/cm2) and 200× (2280 pmol/cm2) of an effective dose of 1,25D 11.4 pmol/cm2 was determined according to the same ratio of 1,25D doses as determined from in vitro experiments [26,31]. Biopsies of dorsal skin were taken in triplicate from each mouse, 3 h post-UV and paraffin-embedded for immunohistochemistry of DNA damage as previously described [31]. Quantification of positive nuclei as % total nuclei (the percentage of CPD or 8-OHd positive nuclei staining in the selected nuclei in an area of epidermis) was obtained using MetaMorph (Molecular Devices, San Jose, CA, USA) and normalized to SHAM. Routine haematoxylin and eosin staining was carried out by Veterinary Pathology Diagnostic Service (University of Sydney) to visualize sunburn cells. The stained sections were examined under a Zeiss Axioscan light microscope (Oberkochen, Germany) at 20× magnification, and the number of sunburn cells per linear millimetre of skin section recorded, as previously described [31,47]. Non-irradiated samples as SHAM control were obtained from the abdomen. Three areas of each section were analysed. Changes in dorsal skin thickness, a measure of edema, were recorded daily from 24 h onward until the until levels returned close to pre-UV condition on the 7th day after irradiation. The contact hypersensitivity response was tested to investigate the effects of NPS-2143 on UV-induced systemic immunosuppression, as previously described [31]. Briefly, female mice were sensitized 1 week after irradiation and treatments, with 100 μL of 2% oxazolone (Sigma-Aldrich, USA) (w/v) in absolute alcohol applied to the non-irradiated abdominal skin. Sensitization was repeated on the subsequent day. The sensitized mice were challenged 2 weeks after irradiation by application of 5 μL 2% oxazolone to both surfaces of each ear, so that each mouse received 20 μL in total. Ear thickness measurements, taken using a spring micrometre (Interapid, Zurich, Switzerland), were recorded before the challenge and at 18 h after challenge, as previously reported [31]. The difference between pre- and post-oxazolone challenge ear thickness measurements of each mouse was recorded as ear swelling and the means for each group of 5 mice was calculated. Ear Swelling = pre-challenge ear thickness–post-challenge ear thickness. The immune response was then calculated for each mouse, as shown in the formula below: Immunosuppression was calculated as 100% minus this value, ± SEM [31] as in the formula below: For this study, groups of 18 mice were used. Immediately after ssUV irradiation, mice were treated topically with either base lotion [31], 1,25D (11.4 pmol/cm2), or NPS-2143 (2280 pmol/cm2). During the next 30 weeks, the time of appearance, location, and visual identification of tumours with a diameter of at least 1 mm were monitored and mapped for each mouse. As previously described [31], the term “tumour” includes papilloma and SCC. The photocarcinogenic outcomes were reported as tumour latency, tumour incidence, tumour multiplicity, and SCC incidence. At the end of the experiment, all tumours were harvested for histological examination to confirm the classification. Keratinocytes were harvested from skin samples under University of Sydney Human Research Ethics Committee protocol no. 2015/063 and cultured, as previously described [26]. The concentration of NPS-2143 used in these in vitro studies was based on previous experiments where we performed serial dilutions of NPS-2143 to determine the concentration-dependent response in human keratinocytes [26].A total of 500 nM was in the effective concentration range (5 nM~500 nM) and was also 10× IC 50; thus, it was chosen for all the in vitro experiments. Keratinocytes were irradiated with an Oriel 1000 W xenon arc lamp (Newport Corporation, USA) and subsequently treated with vehicle, 1,25D, or NPS-2143, as in our previous study [26]. Western blot was performed, as previously described [102], with α-tubulin as the loading control. Primary antibodies used in this study were anti-phospho-CREB (Ser133) at 1 in 1000 dilution (mouse monoclonal, #9196, Cell Signaling Technology, Trask Lane Danvers, MA, USA), anti-CREB(Total) at 1 in 1000 dilution (mouse monoclonal, #9197, Cell Signaling Technology, Trask Lane Danvers, MA, USA), or anti-tubulin at 1µg/mL (mouse monoclonal, SC-5286, Santa Cruz Biotechnology). The band was imaged with the ChemiDocTM imaging system (Bio-Rad Laboratories, Inc, Hercules, CA, USA) and densitometry was carried out using Image J. SHAM, showing negligible expression of p-CREB which served as a negative control, and the data was normalized to UV+ vehicle to pool experiments. Animals in this study were divided into treatment groups of three for acute study, groups of five for immunosuppression study, and groups of eighteen for chronic photocarcinogenesis [31,47,101,103]. These numbers were determined by power analysis manually calculated from data from previous studies to have an 80% chance showing a 20% difference between treatment group at a significance level of 5% [104,105]. Results are expressed as either as mean + SEM or as indicated. All the data sets passed goodness of fit tests, which determine whether sample data exhibit skewness and kurtosis that matches a normal distribution. All statistical analyses were performed with GraphPad Prism statistical program 9.0 (GraphPad Software Inc.). Unless otherwise stated, analysis of comparisons between treatment groups were made by linear mixed-model analysis, appropriate for dealing with repeated measurement data. The Mantel–Haenszel log-rank test (also called Mantel–Cox test) was used to analyse incidence data in the photocarcinogenesis study [31,45].

PMC10002577

Anyelo Durán,David A. Priestman,Macarena Las Heras,Boris Rebolledo-Jaramillo,Valeria Olguín,Juan F. Calderón,Silvana Zanlungo,Jaime Gutiérrez,Frances M. Platt,Andrés D. Klein

A Mouse Systems Genetics Approach Reveals Common and Uncommon Genetic Modifiers of Hepatic Lysosomal Enzyme Activities and Glycosphingolipids

03-03-2023

metabolism,lysosomal enzymes,glycosphingolipids,systems genetics,modifier genes

Identification of genetic modulators of lysosomal enzyme activities and glycosphingolipids (GSLs) may facilitate the development of therapeutics for diseases in which they participate, including Lysosomal Storage Disorders (LSDs). To this end, we used a systems genetics approach: we measured 11 hepatic lysosomal enzymes and many of their natural substrates (GSLs), followed by modifier gene mapping by GWAS and transcriptomics associations in a panel of inbred strains. Unexpectedly, most GSLs showed no association between their levels and the enzyme activity that catabolizes them. Genomic mapping identified 30 shared predicted modifier genes between the enzymes and GSLs, which are clustered in three pathways and are associated with other diseases. Surprisingly, they are regulated by ten common transcription factors, and their majority by miRNA-340p. In conclusion, we have identified novel regulators of GSL metabolism, which may serve as therapeutic targets for LSDs and may suggest the involvement of GSL metabolism in other pathologies.

A Mouse Systems Genetics Approach Reveals Common and Uncommon Genetic Modifiers of Hepatic Lysosomal Enzyme Activities and Glycosphingolipids Identification of genetic modulators of lysosomal enzyme activities and glycosphingolipids (GSLs) may facilitate the development of therapeutics for diseases in which they participate, including Lysosomal Storage Disorders (LSDs). To this end, we used a systems genetics approach: we measured 11 hepatic lysosomal enzymes and many of their natural substrates (GSLs), followed by modifier gene mapping by GWAS and transcriptomics associations in a panel of inbred strains. Unexpectedly, most GSLs showed no association between their levels and the enzyme activity that catabolizes them. Genomic mapping identified 30 shared predicted modifier genes between the enzymes and GSLs, which are clustered in three pathways and are associated with other diseases. Surprisingly, they are regulated by ten common transcription factors, and their majority by miRNA-340p. In conclusion, we have identified novel regulators of GSL metabolism, which may serve as therapeutic targets for LSDs and may suggest the involvement of GSL metabolism in other pathologies. Hydrolytic enzymes are abundant in lysosomes [1]. In a healthy cell, the biosynthesis and catabolism of macromolecules are subject to regulatory mechanisms that maintain cellular homeostasis [2]. The degradative processes in lysosomes are controlled by their own enzymes [3,4]. Lysosomes play a central role in several biological processes, including energy metabolism, signaling, plasma membrane repair, secretion, and others [3]. Loss-of-function variants in genes encoding lysosomal proteins cause lysosomal storage disorders (LSDs), a group of diseases characterized by intracellular buildup of partially degraded material [5]. Growing evidence suggests that variants in lysosomal genes increase the risk of developing Parkinson’s disease (PD) [6,7]. In the sphingolipidoses, a subset of LSDs, glycosphingolipids (GSLs) accumulate in late endocytic organelles (late endosomes/lysosomes) and participate in their pathological cascades [8]. Current treatments for LSDs include substrate reduction therapy (SRT), which aims to reduce the rate of biosynthesis of stored substrates [5,9,10], and enzyme replacement therapies (ERT) aimed at replacing a deficient enzyme [11,12]. Emerging treatments include gene and cell therapies [13,14,15] and chaperones for improving enzyme folding and trafficking [16]. Although there is a range of therapeutic options for LSDs, they have limitations, such as tissue accessibility [17], antibody-mediated reaction [18], cost [19], and others. So far, therapies aimed at increasing enzyme activity or reducing lipid levels by modulating a second (modifier) gene have not been studied. In this context, a deeper understanding of the regulatory mechanisms that govern GSLs metabolism must be uncovered to fully develop this approximation. Genome-wide association studies (GWAS) in humans and systems genetics strategies, which include gene mapping in model organisms, have identified genetic regulators of physiological and pathophysiological processes [20,21,22]. The Hybrid Mouse Diversity Panel (HMDP) has been a useful tool because genomes and tissue transcriptomes are freely available, allowing the combination of modifier gene mapping by GWAS and pathway analysis [23,24]. In this study, we have analyzed the activities of 11 lysosomal enzymes and several of their natural substrates in 25 strains of the HMDP panel followed by gene mapping and transcript integration. We identified a lack of correlation between most enzyme activities and their mRNA levels. Similarly, most substrates had no association between their levels and the enzyme activity that catabolizes them. Finally, we mapped putative modifier genes of each lysosomal enzyme and GSL by GWAS. We found associations between the mRNA levels of many modifier genes and enzyme activities or GSL levels. We clustered the putative modifiers in pathways and identified common and uncommon genetic regulators between GSLs and lysosomal enzymes, including transcription factors that regulate them. Our discoveries may help develop novel therapeutics for diseases with altered lysosomal enzyme activities and GSLs. We measured hepatic enzyme activity of β-hexosaminidase A and B (defective in Tay-Sachs and Sandhoff disease, respectively), α-neuraminidase (defective in Sialidosis/Mucolipidosis Type I), α-galactosidase A and B (defective in Fabry and Schindler disease), β-D-galactosidase (defective in GM1 Gangliosidosis), α-glucosidase (defective in Pompe), chitotriosidase (elevated in Gaucher disease), α-L-fucosidase (defective in fucosidosis), lysosomal acid phosphatase (elevated in patients with Gaucher), and Tartrate-resistant acid phosphatase (TRAP; altered in Gaucher disease) by fluorimetry in liver samples derived from 25 inbred mice strains using 4-methylumbelliferone (4-MU) based artificial substrates. We observed significant variability in the average enzymatic activity between the different strains (ANOVA p ≤ 0.05) (Figure 1). We did not find changes in α-galactosidase A, lysosomal acid phosphatase, and TRAP activities across the tissues analyzed (Figure 1d,j,k). We observed unique activity distribution patterns across the strains for the other enzymes, suggesting specific modifiers for each enzyme. Advantages of using tissues derived from the HMDP panel of inbred mouse strains include the fact that their genomes are sequenced, and transcriptomic data are available. Thus, we analyzed potential correlations between the genes encoding lysosomal enzymes and their activities. Recently we described the natural variation of hepatic acid β-glucocerebrosidase levels across many different mouse strains and included them in this analysis [20]. We did not identify significant correlations between enzyme activity and its transcript levels (Figure 2), with the only exception being Glb1, the gene encoding for β-D-galactosidase (r = 0.5775; p ≤ 0.002) (Figure 2c). These results indicate that mRNA levels are a poor proxy for enzyme activities. Next, we measured the levels of GSLs in livers of the inbred mice strains in which we had access to enough material for three biological replicates (23/25) by Normal Phase-High-Performance Liquid Chromatography (NP-HPLC). We observed significant variability in GSLs among the strains, especially in total GSLs, GM3-Gc, GM2-Gc, GM1agc, GM3, Gb3, GM1a, GM1b, GD1b, and GD1a (Figure 3). For example, the levels of GM3-Gc were significantly increased (ANOVA p < 0.0001) in NOD/ShiLtJ compared with the other samples (Figure 3b). These results indicate that GSLs levels vary across strains. A possibility is that GSL levels could correlate with their biosynthesis rate. Since we started from frozen tissues, we could not test this directly. Instead, we utilized the transcriptomic data available from the repository GSE16780 UCLA Hybrid MDP Liver Affy HT M430A [24]. We found transcript probes for 21 mRNA of the 21 anabolic enzymes of the GSLs pathway and four GSL transfer proteins. The analyzed gene list of the biosynthetic pathway is presented in the Supplementary Table S1. The expression values were organized according to GSLs levels from lowest to highest and presented as a heatmap. The analysis showed significant correlations for Cgt (r = −0.4263; p = 0.042) with total GSLs (Figure 4a). For GM2-Gc with Cgt (r = −0.4582; p = 0.0279), Galgt1 (r = 0.6078; p = 0.0021), A4galt (r = 0.4903; p = 0.0176), Gltp (r = −0.454; p = 0.0296) (Figure 4b). GM3 levels correlated with Galgt1 (r = −0.579; p = 0.0038), Gltp (r = 0.4151; p = 0.0489) (Figure 4c). GM1a with Col4a3bp (r = 0.4458 p = 0.033) (Figure 4d). GM3-Gc is associated with Galgt1 (r = −0.9591, p ≤ 0.0001) and it was the most significant correlation (Figure 4e). GM1agc levels with Slc17a2 (r = 0.4163; p = 0.0482) (Figure 5f). Gb3 with A4galt (r = 0.6011, p = 0.0024) (Figure 4g) and GM1b with Galgt1 (r = −0.5764; p = 0.004) and St8sia5 (r = −0.4194, p = 0.0046) (Figure 4h). No significant correlations were found between the majority of GSLs and biosynthetic genes (Supplementary Table S2); thus, we analyzed potential correlations between GSL levels and the enzyme activity that catabolizes them across the mouse panel. It is possible to speculate that the strains that present high activity of a particular enzyme should have reduced levels of its natural substrate because the enzyme catabolizes it. Unexpectedly, for most enzymes, we did not find significant correlations between the GSL levels and the enzyme activity that degrades it (Figure 5), except for neuraminidase and GM3-Gc (r = −0.4706; p = 0.0234) (Figure 5g). These results suggest that for most strains, the rate of biosynthesis and/or uptake of GSLs varies along with the catabolic rates which most likely are genetically regulated. To identify genetic regulators, we conducted genome-wide association studies with a quality control analysis that considered the population structure of the HMDP panel strains to reduce false associations [25,26]. We used enzyme activity levels as a trait and included the β-glucosidase activity, which we reported previously in the same and a few other strains [20]. For all the enzymes together, we identified 211 significant Single Nucleotide Variants (SNVs) that passed the empiric threshold of significance p ≤ 4.1 × 10−6 (−log10P = 5.39), previously calculated by permutations [21,22,26], while the Bonferroni threshold was p ≤ 3.9 × 10−7 [26]. These SNVs were located in different genomic regions (exonic, intronic, UTR3, downstream, and intergenic) (Table 1, Supplementary Table S3) in a total of 137 non-redundant genes. Similarly, we identified 3215 SNVs associated with GSLs levels (1744 non-redundant genes) whose variants are located in different genomic regions (Table 1, Supplementary Table S3). These analyses indicated that our strategy has sufficient power to map putative modifier genes. To prioritize the putative modifier genes that could regulate each enzyme, we searched for correlations between the transcript levels of putative modifier genes and their traits (enzyme activity and GSL levels, respectively) (Figure 6). We found transcript probes for 67 mRNA of the 137 putative modifiers of the enzymes. The expression values were organized according to enzyme activity from lowest to highest and presented as a heatmap. The analysis showed significant correlations in Fip1l1 (r = −0.4462; p = 0.0254) with α-L-fucosidase (Figure 6a). For β-D-galactosidase with Lyplal1 (r = −0.702; p = <0.0001), Arrdc4 (r = 0.627; p = 0.0008), Pde2a (r = 0.5306; p = 0.0064), Glb1 (r = 0.5753; p = 0.0026), Bptf (r = 0.5135; p = 0.0087), Oxr1 (r = −0.447; p = 0.0251) (Figure 6b). No significant correlations were found for the other enzymes analyzed. We used SIFT to explore the impact of genetic variants on the genes identified by GWAS (benign or deleterious changes) associated with changes in enzyme activity [27], because the full genomes of the strains are known [28]. This strategy identified 308 predicted deleterious variants (Supplementary Table S4) in 43 of the 67 genes whose functions are related to organelle biogenesis (Chchd6) [29], intracellular signaling (Pde4dip) [30], and tissue development (Fam181b) [31], among others. These results suggest that amino acid substitution could affect protein function and signaling pathways leading to changes in enzyme activity. The same analysis was performed to identify putative modifiers of GSL levels (Figure 6c–f). For 1744 non-redundant SNVs, we found expression values for 994 genes. The analysis identified 45 significant correlations, of which 33 were correlated with GM3-Gc levels, 10 genes with LacCer, and one gene with GD1b and GA2 (Figure 6c–f). Overall, we recorded 4.9% (52/1061) of significant correlations distributed between the two traits. We also explored the impact of genetic variants associated with changes in GSLs with SIFT [27]. This strategy identified 515 deleterious variants predicted to disrupt the protein structure (Supplementary Table S4) in 132 genes related to DNA methyltransferase activity (Setdb1) [32] and synapse (Slitrk1) [33], among others. If there is an orchestrated regulation of GSL levels and the enzymes that degrade them, it would be expected to observe enrichment in common pathways [34]. We therefore utilized gProfiler [35] to perform enrichment analysis using the putative modifier genes lists. For the modifier of enzyme activities, we found significantly associated pathways such as cell periphery (p = 5.9 × 10−4), plasma membrane (p = 2.4 × 10−3), and integral components of the plasma membrane (p = 2.6 × 10−2) (Figure 7b), which could be related to endocytic processes necessary to deliver key molecules to the lysosome, including the lysosomal enzymes that can be recycled from the extracellular space. Significant biological processes analysis included regulation of cellular processes (p = 3.9 × 10−2) (Figure 7d) (Supplementary Table S5). We did not find significant enrichment for the molecular function category. For GSLs, we observed enrichment in terms like cytoplasm (p = 3.5 × 10−28), cell junction (p = 7 × 10−21), synapse (p = 4.6 × 10−19), and 70 other pathways related to cellular components (Figure 7a; Supplementary Table S5). Many of these pathways require cellular membranes, where GSLs play a structural role. Significantly enriched Gene Ontology (GO) terms included protein binding (p = 9.1 × 10−31), ion binding (p = 8.8 × 10−14), binding (p = 2.3 × 10−13), ATP binding (p = 9.4 × 10−13), carbohydrate derivate binding (p = 1.8 × 10−1), and 27 other pathways related to molecular functions (Supplementary Table S5). Biological processes terms revealed 328 pathways, including system development (p = 4.3 × 10−39), anatomical structure development (5.6 × 10−38), and multicellular organism development (p = 1.4 × 10−37). We searched for the overlap between the cellular component domains of modifiers of enzyme activity and GSLs, which resulted in three common pathways (GO:0071944—cell periphery, GO:0005886—plasma membrane, and GO:0005887—integral component of plasma membrane) (Figure 7c) and one pathway associated with biological processes (GO:0050794; regulation of cellular process) (Supplementary Table S5). Common regulators of GSLs and enzymatic activities are relevant for understanding GSL metabolism and may be attractive therapeutic targets for LSDs. Therefore, we examined the overlap between them. We found 30 common and 1821 uncommon genes (Figure 7e). We explored their functions and identified genes involved in mitochondrial biogenesis and dynamics (Tfb1m, Timen135, Chchd6) [29,36,37], cell proliferation (Fstl5, Fzd10, Arhgap18) [38,39,40], platelet function (Cdh6) [41], vesicular trafficking (Vps45) [42], gene expression (Tfb1m, Zfat) [36,43], and regulating levels of the proto-oncogene MYC (Pvt1) [44]. Many of the 30 genes have been linked to diseases, such as Pvt1, Tiam2, Fstl5, Fzd10, Cdh6, Pvt1, Chchd6 in liver, colorectal, nasopharyngeal, and gastric cancer [45,46,47,48,49,50]. Others participate in neurodegenerative conditions; PD, schizophrenia, and intellectual disability (Tenm4, Pde4dip, Grid2, Arhgap18) [51,52]. These results suggest that lysosomal enzymes and GSLs may play a role in their pathophysiology and should be explored further (Table 2). To better understand the molecular regulation of these 30 genes, we analyzed the transcription factors that bind to their promoters and/or enhancers (Figure 7f). We found no information for three of the 30 genes since they are putative (Rik) genes. The following transcription factors can bind to the 27 genes for which we have information: REST, TBP, CEBPB, EP300, POLR2A, FOS, DPF2, CTCF, RAD21, and SP1. Some of these transcription factors are broad regulators of transcription, such as TBP and POLR2A, while others are selective for specific processes, such as CTCF and RAD21. Considering all the promoters/enhancers of the 27 shared genes, we identified a total of 533 transcription factors that can bind them, although some only bind a few genes (Supplementary Table S6). We also searched for potential shared microRNA (miRNA) regulators using miRTarBase, a curated microRNA database [53]. We identified that miR-340-5p can bind to 11 of the 27 known common genes (Tusc1, Fam91a1, Zc3h12c, Adamts5, Tmem135, Tenm4, Grid2, Csnk1g3, Cdh6, Fam181b, and Pde4dip; p = 2.2 × 10−2) (Figure 7g). This result suggests that miRNA-340-5p regulates GSLs metabolism and may be involved in the pathogenesis of LSDs and the disorders described in Table 2. In this study we searched for genetic modulators involved in the regulation of the lysosomal enzyme activities and the levels of substrates related to GSLs, with the idea of finding novel therapeutics targets for disorders in which they participate. By GWASs, we identified common and uncommon genetic regulators, evaluated the associations between modifier gene mRNA levels and each trait, and also clustered them in pathways. We identified 30 shared putative modifiers and described the transcription factors that are predicted to regulate them, and we noted that the miRNA340-5p can bind to 11 of these genes. Our first unexpected finding was that most lysosomal enzyme activities do not correlate with their mRNA levels, nor with most of their substrate levels. Although enzyme activity can decrease with age [92], we used sex and age-matched samples; thus, the variation observed across strains was shown not to be due to any of these factors. Another unexpected finding was that GM2-Gc levels correlate with the mRNA levels of the Cgt gene, which encodes for the UDP-galactose ceramide galactosyltransferase (CGT). CGT is a key enzyme for the biosynthesis of galactocerebrosides. Gangliosides, including GM2 derivates, are built from glucosylceramide and not from the galacto series [93]. However, for most of the biosynthetic genes there were no associations between the amount of lipids and the transcript levels of their anabolic pathways. Altogether, our results suggest that the GSL biosynthesis rate and uptake differ across the mouse strains, suggesting the existence of specific modifier genes for each trait. Our third unexpected finding was that TFEB, the master transcriptional regulator of lysosomal genes [94], did not appear in the list of modifiers of lysosomal enzymes. This may be due to the fact that we screened for enzymatic activity instead of mRNA levels, and we showed a lack of correlation between transcript levels and enzyme activity under physiological conditions, at least for most enzymes. One exception was β-D-galactosidase, for which we found a positive correlation between its transcript levels and activity. Furthermore, the GWAS for this enzyme identified Glb1, the gene encoding for β-D-galactosidase, as a putative modifier of its activity, validating the power of discovery of our population-based strategy [95]. Our study had some limitations: First, we quantified lysosomal traits from liver homogenates that were not in living or isolated organelles, which may have diluted enzyme activity or promoted molecular interactions that might not occur in vivo because of cellular compartmentalization. Second, we could not directly measure GSLs biosynthesis and uptake because we started with mouse liver samples. Third, we used SNV catalogs with imputation, which may lead to false associations, though with increased mapping resolution. Most of the enzymes we assayed are associated with LSDs [5,8]. For many LSDs, no therapies are available, and the few currently available treatments have severe limitations [5]. In this context, targeting a modifier gene could be a novel therapeutic approach. For example, lack of β-D-galactosidase activity triggers GM1 gangliosidosis, a disease with no approved therapies [96]. Our study identified the druggable Lypla1 and Pkm genes as putative modifiers of β-D-galactosidase activity, which can be pharmacologically modulated [97,98]. We found other druggable genes as well for several traits, and with the current gene editing technologies virtually any gene can be targeted. The potential modifying effects of these genes and compounds can be tested in LSDs disease models. A hallmark of the sphingolipidoses is the intracellular buildup of GSLs, so strategies aimed at reducing their levels could lead us to novel therapies [5,8]. GSLs comprise a ceramide moiety with one or more sugar residues linked to it [99]. An approved therapy for Gaucher and Niemann-Pick disease type C is Miglustat [100,101], a small molecule inhibitor of GSL biosynthesis, thus reducing their levels. Our GSLs GWAS identified more than 50 genes previously associated with sphingolipid metabolism, which served as a positive control, including B3gnt5, Cln8, Hexb, Pnpla1, St8sia1, and Cgt. B3gnt5 regulates GSLs metabolism and lung tumorigenesis [102]. Our study also identified Lipc as a modifier of GM3-Gc levels, which has been previously associated with elevated serum levels of liver enzymes (alkaline phosphatase and γ-glutamyl transferase) [103], suggesting a new connection between GM3-Gc and liver damage. Variants in LIPC, CPS1, PABPC4, CITED2, TRPS1, and MVK are associated with changes in plasma lipoprotein levels [104], connecting novel traits to GSLs metabolism. Lysosomal leakage has been associated with Alzheimers’ [105], cancer, and inflammation among other conditions [106]. Recently, the phosphoinositide signaling pathway was implicated in lysosomal repair [107]. Many genes of this pathway appear in our discovery list (Osbpl9, Osbpl6, Pde4dip, Pde2a, Pde1a, Pde7a, Pde7b, Pde4d, Pde8b, Pld5, Pik3r1, Pip4k2a, Pip5k1a, Pip5k1b, Pi4kb, Pdpk1, Atg4c, Atg10), suggesting that integrity of the lysosomal compartment is key to the proper functioning of enzymes and/or that these enzymes and lipids participate in lysosomal repair. Furthermore, this novel lysosomal repair pathway may facilitate the development of novel therapeutics for these diseases with lysosomal leakage. Defects in the 30 shared genes are related to several pathologies, such as vision abnormalities (TMEM135) [60], cancer (CDH6 [49], FZD10 [48], TIAM2 [47]), neuropsychiatric disorders (Tenm4 [51], Pde4dip [52], Grid2 [78]), deafness (TFB1M) [55], neutrophil disorders (VPS45) [74] and others. Lysosomal enzymes and GSLs have been widely studied in cancer and neurodegenerative diseases [46,108,109,110,111]; however, their role in the other identified conditions should be explored. Although not binding the complete list of shared genes, we identified some transcription factors previously known to be involved in lipid metabolism and autophagy-lysosomal functions (PPARγ, SREBF1, HNF1A, YY1, EGR1, SP1 and TFE3, E2F1, CREB1, MYC) [112,113,114,115,116,117,118,119,120,121], and many more that have not been previously linked to GSL metabolism. We also identified miR-340-5p as a putative regulator of many common modifier genes. Changes in miR-340-5p are linked to preeclampsia, neuroinflammation [122,123,124,125,126], adipocyte differentiation [127], as well as obesity and diabetes [128]. GSL metabolism plays a crucial role in the two last-mentioned disorders, and inhibitors of their biosynthesis have shown promising results in animal models of these conditions, validating the relevance of our strategy [129,130]. In conclusion, we described putative regulators of hepatic lysosomal enzymes and GSLs, many of them druggable and associated with diseases where alterations in GSL metabolism have not been previously described and should be assessed. We expect our findings may facilitate the development of novel therapeutics for conditions with alterations in these traits. We used 8 weeks-old mice livers derived from 25 inbred mouse strains, which were kindly donated by Dr. Aldons Lusis (University of California, Los Angeles, CA, USA). (i) 129X1/SvJ, (ii) A/J, (iii) AKR/J, (iv) BALB/cJ, (v) BTBR T<+> tf/J, (vi) BUB/BnJ, (vii) C57BL/6J, (viii) C58/J, (ix) CAST/EiJ, (x) CBA/J, (xi) CE/J, (xii) DBA/2J, (xiii) KK/HlJ, (xiv) LG/J, (xv) LP/J, (xvi) MA/MyJ, (xvii) NOD/ShiLtJ, (xviii) NON/ShiLtJ, (xix) NZB/BlNJ, (xx) NZW/LacJ, (xxi) PL/J, (xxii) RIIIS/J, (xxiii) SEA/GnJ, (xxiv) SM/J, (xxv) SWR/J. Tissues were homogenized and adjusted to 50 mg tissue/mL in deionized water with a Potter-Elvehjem tissue homogenizer (Omni International, Kennesaw, GA, USA). Three or more livers per mouse strain were used to quantify traits (Supplementary Table S7). Lysosomal hydrolase activities were determined using an artificial fluorescent substrate based on 4-methylumbelliferone (4-MU) [131]. For α-glucosidase, 1.47 mM 4-MU α-D-glucopyranoside (Sigma, Dorset, UK) in 100 mM citric acid/100 mM sodium phosphate, 0.1% TritonX-100, pH 4.0 was used as substrate [132]. The substrate for α-galactosidase A and B activities was 5 mM 4-MU α-D-galactopyranoside (Santa Cruz, CA, USA) with and without 250 mM N-acetyl-galactosamine (Sigma, Dorset, UK) in 100 mM citric acid/100 mM tri-sodium citrate, 0.1% TritonX-100, pH 4.0 [133,134]. For measuring β-hexosaminidase A and B activity, 3 mM 4-MU N-acetyl-β-D-glucosaminide (BioChemika, Dorset, UK) in 100 mM citric acid/100 mM sodium phosphate, 0.1% TritonX-100, pH 4.5 was used as substrate. Heat inactivation assay for β-hexosaminidase A was carried out at 50 °C for 3 h [135]. For β-galactosidase activity, 1 mM 4-MU β-D-galactose (Sigma, Dorset, UK) in 200 mM sodium acetate buffer, 100 mM NaCl, 0.1% TritonX-100, pH 4.3 was used as substrate [136]. The substrate for neuraminidase activity was 0.4 mM 4-MU α-D-N-acetylneuraminic acid (Sigma, Dorset, UK) in 0.1 M acetate buffer, 0.1% TritonX-100, pH 4.6 [137,138]. For chitotriosidase activity, 0.013 mM 4-MU chitotrioside (Sigma, Dorset, UK) in 100 mM citric acid/200 mM sodium phosphate, 0.1% TritonX-100, pH 5.2 was used as substrate [139,140]. For total acid phosphatase activity, 5 mM 4-MU phosphate (Sigma, Dorset, UK) with 40 mM NaCl in 200 mM citric acid/200 mM sodium phosphate, 0.1% TritonX-100, pH 4.5 was used as substrate. For tartrate-resistant acid phosphatase (TRAP) activity, 5 mM 4-MU phosphate (Sigma, Dorset, UK) with 40 mM Na Tartrate in 200 mM citric acid/200 mM sodium phosphate, 0.1% TritonX-100, pH 4.5 was used as substrate. The difference between total acid phosphatase activity and TRAP corresponded to lysosomal acid phosphatase (Lys AP) activity [141,142]. The substrate for α-L-fucosidase activity was 60 nM 4-MU α-L-fucopyranoside (Sigma, Dorset, UK) in 200 mM citric acid/200 mM sodium citrate, 0.1% TritonX-100, pH 5.0 [143,144]. We determined the acid-β-glucosidase activity in the same tissues in a previous publication [20], and further analyses were performed here based on the published activity. Liver homogenates were diluted with the buffer corresponding to each enzymatic determination. Three cycles of freezing (liquid nitrogen) and thawing were performed on the samples. Three biological replicates of the diluted liver extracts were incubated with the corresponding substrate at 37 °C for 30 min (or 1 h for α-neuraminidase, β-D-galactosidase, and chitotriosidase). Cold 0.5 M Na2CO3 (pH 10.7) was added to stop the reaction. Fluorescence intensity in samples was measured in a Synergy HT plate reader (BioTek, Winooski, VT, USA) at 360/460 nm. Protein concentration was measured using a BCA protein assay kit (Thermo Fisher Scientific, New Jersey, NJ USA). Fluorescence values were normalized to protein concentration. A 4-MU standard curve was constructed to calculate specific activity, and the final value was adjusted to one hour of enzymatic reaction. The GSLs were extracted and measured by Normal Phase-High-Performance Liquid Chromatography (NP-HPLC) following published methods [145]. Briefly, the aqueous tissue extract was homogenized in chloroform/methanol (C:M) (1:2 v/v) and kept overnight at 4 °C. Then, the extracts mixture was centrifuged at 3000 rpm for 10 min at room temperature. We added 0.5 mL of PBS and 0.5 mL of chloroform to the supernatant followed by a 3000-rpm centrifugation for 10 min at room temperature. The lower phase was carefully removed and dried under a stream of nitrogen gas (N2) in a heating block (42 °C), resuspended in 40 μL C:M 1:3 v/v and mixed with the upper phase. Afterwards, glycosphingolipids-derived oligosaccharides were purified from the samples using C18 columns (Telos, Kinesis, UK) previously pre-equilibrated with 1.25 mL methanol (four times) and 1.25 mL deionized water (three times). We loaded the mixed phase (lower/upper) onto a column and rinsed the sample tube with 1 × 1 mL of deionized water. Then, the C18 column was washed with 4 × 1.25 mL deionized water and eluted it with 1 × 1 mL (C:M) (98:2 v/v), 2 × 1 mL (C:M) (1:3 v/v), 1 × 1 mL methanol. The eluates were dried under N2 current and digested with a recombinant Endoglycoceramidase I (rEGCaseI) (GenScript, Oxford, UK) in buffer 50 mM sodium acetate, pH 5.0, 0.6% TritonX-100 (4 μL enzyme + 86 μL buffer) at 37 °C for 16 h. The released glycans were labeled with 310 μL of labelling mix (30 mg/mL anthranilic acid (2AA) and 45 mg/mL sodium cyanoborohydride) in 4% sodium acetate, 2% boric acid in methanol, and heated at 80 °C. Then, we cooled the samples and mixed them with 3 × 1 mL acetonitrile: deionized water (97:3) (v/v) and added them to a Discovery DPA-6S-SPE tube (Supelco, PA, USA), pre-equilibrated with 1 × 1 mL acetonitrile, 2 × 1 mL deionized water, and 3 × 1 mL acetonitrile. The columns were cleaned with 3 × 1 mL acetonitrile: deionized water (95:5) (v/v), and the tubes were washed with 2 × 1 mL acetonitrile: deionized water (95:5) (v/v) and eluted in 0.6 mL deionized water. We took 60 μL from 0.6 mL sample eluted, added 140 μL acetonitrile, and injected 50 μL of this mix (deionized water: acetonitrile) (30:70) (v/v) onto NP-HPLC (Waters Alliance 2695 separations module and multi-fluorescent detector set at Ex 360/Em 425 nm). To calculate molar quantities from peaks in the chromatogram, we included a calibration standard containing 2.5 pmol 2AA-labelled chitotriose (Ludger, Oxford, UK) for each NP-HPLC run [145]. The chromatographic data were processed using Waters Empower software 3 (Waters, Milford, MA, USA). Fluorescence values by sample were normalized to protein content using a BCA Assay kit (Merck KGaA, Darmstadt, Germany). We used the genotype of each strain, and the enzymatic activity or substrate as trait, and its kinship matrix to perform the GWAS using The Efficient Mixed Model Association (EMMA) v.1.1.230 in the R package [26,146]. We used PLINK to remove SNVs in linkage disequilibrium to avoid false associations [25], considering an R2 = 0.25, leaving 127,285 independent variants out of the initial four million variants downloaded from the mouse HapMap reference panel (http://mouse.cs.ucla.edu/mousehapmap/full.html, accessed on 28 September 2020) [147]. For gene expression correlations, we obtained inbred mouse hepatic transcript data from the repository GSE16780 UCLA Hybrid MDP Liver Affy HTM430A [24]. The mRNA levels in the repository were expressed as log2 transformed and were calculated from the Affimetrix chip with the robust multiarray average (RMA) method. To plot the heatmaps, we used Morpheus software (https://software.broadinstitute.org/morpheus, accessed on 15 February 2022). The functional impact of genomic variants was assessed using the Sorting Intolerant From Tolerant (SIFT) software (https://sift.bii.a-star.edu.sg/www/SIFT_dbSNP.html, accessed on 12 July 2022) [27]. We used gProfiler [35] with the default settings to perform the pathway enrichment analyses. We consulted the GeneHancer (GH) database, a catalogue of genome-wide enhancer-to-gene and promoter-to-gene associations, through GeneCards® (https://www.genecards.org/Guide/GeneCard, accessed on 6 September 2022) [148]. Only transcription factors with a significative GH Score were considered. We used Student’s t-test, ANOVA with Bonferroni correction, and Pearson correlation. All tests were two-tailed. The significance was considered to be p < 0.05. We used an R package [146] and Prism v9.1.0 (GraphPad software, San Diego, CA, USA) for these analyses.

PMC10002578

Malabendu Jana,Sridevi Dasarathy,Supurna Ghosh,Kalipada Pahan

Upregulation of DJ-1 in Dopaminergic Neurons by a Physically-Modified Saline: Implications for Parkinson’s Disease

28-02-2023

physically modified saline,neurons,DJ-1,CREB

Parkinson’s disease (PD) is the second most common neurodegenerative disorder in human and loss-of-functions DJ-1 mutations are associated with a familial form of early onset PD. Functionally, DJ-1 (PARK7), a neuroprotective protein, is known to support mitochondria and protect cells from oxidative stress. Mechanisms and agents by which the level of DJ-1 could be increased in the CNS are poorly described. RNS60 is a bioactive aqueous solution created by exposing normal saline to Taylor-Couette-Poiseuille flow under high oxygen pressure. Recently we have described neuroprotective, immunomodulatory and promyelinogenic properties of RNS60. Here we delineate that RNS60 is also capable of increasing the level of DJ-1 in mouse MN9D neuronal cells and primary dopaminergic neurons, highlighting another new neuroprotective effect of RNS60. While investigating the mechanism we found the presence of cAMP response element (CRE) in DJ-1 gene promoter and stimulation of CREB activation in neuronal cells by RNS60. Accordingly, RNS60 treatment increased the recruitment of CREB to the DJ-1 gene promoter in neuronal cells. Interestingly, RNS60 treatment also induced the enrollment of CREB-binding protein (CBP), but not the other histone acetyl transferase p300, to the promoter of DJ-1 gene. Moreover, knockdown of CREB by siRNA led to the inhibition of RNS60-mediated DJ-1 upregulation, indicating an important role of CREB in DJ-1 upregulation by RNS60. Together, these results indicate that RNS60 upregulates DJ-1 in neuronal cells via CREB–CBP pathway. It may be of benefit for PD and other neurodegenerative disorders.

Upregulation of DJ-1 in Dopaminergic Neurons by a Physically-Modified Saline: Implications for Parkinson’s Disease Parkinson’s disease (PD) is the second most common neurodegenerative disorder in human and loss-of-functions DJ-1 mutations are associated with a familial form of early onset PD. Functionally, DJ-1 (PARK7), a neuroprotective protein, is known to support mitochondria and protect cells from oxidative stress. Mechanisms and agents by which the level of DJ-1 could be increased in the CNS are poorly described. RNS60 is a bioactive aqueous solution created by exposing normal saline to Taylor-Couette-Poiseuille flow under high oxygen pressure. Recently we have described neuroprotective, immunomodulatory and promyelinogenic properties of RNS60. Here we delineate that RNS60 is also capable of increasing the level of DJ-1 in mouse MN9D neuronal cells and primary dopaminergic neurons, highlighting another new neuroprotective effect of RNS60. While investigating the mechanism we found the presence of cAMP response element (CRE) in DJ-1 gene promoter and stimulation of CREB activation in neuronal cells by RNS60. Accordingly, RNS60 treatment increased the recruitment of CREB to the DJ-1 gene promoter in neuronal cells. Interestingly, RNS60 treatment also induced the enrollment of CREB-binding protein (CBP), but not the other histone acetyl transferase p300, to the promoter of DJ-1 gene. Moreover, knockdown of CREB by siRNA led to the inhibition of RNS60-mediated DJ-1 upregulation, indicating an important role of CREB in DJ-1 upregulation by RNS60. Together, these results indicate that RNS60 upregulates DJ-1 in neuronal cells via CREB–CBP pathway. It may be of benefit for PD and other neurodegenerative disorders. Parkinson disease (PD) is an age-related neurodegenerative disorder of the brain [1] that is clinically identified by resting tremor, bradykinesia, rigidity and postural instability [1,2]. Two pathological hallmarks of PD are progressive deterioration of the dopaminergic neurons in the substantia nigra pars compacta (SNpc) and the manifestation of intracytoplasmic inclusions (Lewy bodies) [1]. Recent studies have also described gliosis in the SNpc of PD patients [1]. Although causes for the disease are not well known, studies have identified association of environmental, genetic, and immunological factors with the onset of the disease [3]. At the same time, the discovery of various genes, such as α-synuclein [4], parkin [5], DJ-1 [6], PTEN-induced kinase 1 (PINK-1) [7], HtrA2/Omi gene [8], and leucine-rich repeat kinase 2 (LRRK2) [9] linked to familial forms of PD provides vital hints for the understanding the pathogenesis of the disease. Oxidative stress is involved in the pathogenesis of many human disorders including PD [10,11,12]. Interestingly, among all the PD-associated genes, DJ-1 is the most important one in providing antioxidant defense [13]. On the one hand, DJ-1 helps a damaged cell to break down superoxide by assisting the transcription of MnSOD [14]. Glutamate cysteine ligase is the rate limiting enzyme of glutathione (GSH) biosynthesis [15]. DJ-1 is known to enrich the level of master antioxidant GSH by upregulating glutamate cysteine ligase [16]. However, mechanisms by which DJ-1 expression could be upregulated is poorly understood and such information is critical for understanding the disease process as well as designing new therapeutic approaches for PD. RNS60 is produced by allowing normal saline to pass through increased oxygen pressure in Taylor-Couette-Poiseuille (TCP) flow [17,18,19]. Therefore, RNS60 is basically electrokinetically modified saline with elevated oxygen, which contains no active pharmaceutical ingredient. Recently, we have shown that RNS60 inhibits the activation of nuclear factor kappa B (NF-κB) to exhibit anti-inflammatory effects in microglia [18]. RNS60 protects nigral dopaminergic neurons in MPTP mouse model of PD [20] and hippocampal neurons in a transgenic mouse model of AD [21]. RNS60 also increased the expression of Nrf1, Tfam, Mcu, and Tom20 (genes associated with the biogenesis of mitochondria) to upregulate mitochondrial biogenesis in dopaminergic neuronal cells [22]. Respirometric assays uncovers that RNS60 increases spare glycolytic capacity of oligodendrocytes (OL) under normal culture conditions, while enhancing OL spare respiratory capacity in the presence of a metabolic stress [23]. Here, we delineated that RNS60 was capable of increasing the expression of DJ-1 in mouse MN9D neuronal cells and primary dopaminergic neurons. Furthermore, we demonstrated that RNS60 induced the activation of CREB and that RNS60 increased the level of DJ-1 in neuronal cells via CREB–CBP pathway. Our studies indicate that this physically-altered saline may be of therapeutic benefit for PD and other neurodegenerative disorders in which oxidative stress and mitochondrial anomaly contribute to disease pathogenesis. DJ-1 is a neuroprotective protein, which have implications in neurodegenerative disorders like Parkinson’s disease (PD), Alzheimer’s disease (AD) and Huntington’s disease [6,24,25,26,27]. To examine whether RNS60 may be used for the upregulation of DJ-1, MN9D dopaminergic neuronal cells were incubated with different doses of RNS60 for 5 h under serum-free conditions. RT-PCR (Figure 1A) and real-time PCR (Figure 1B) results clearly showed that RNS60 dose-dependently increased the mRNA expression of DJ-1 in MN9D cells. Although at a dose of 2% v/v, RNS60 significantly increased the mRNA expression of DJ-1, maximum upregulation was seen at a dose of 10% v/v RNS60 (Figure 1B). Time course study shows that only 1 h of treatment was enough for RNS60 to stimulate the expression of DJ-1 gene (Figure 1C,D). However, a markedly increased level of DJ-1 mRNA expression was observed when MN9D cells were treated with RNS60 for 5 h (Figure 1C,D). Next, we examined whether similar to the increase in DJ-1 mRNA expression, RNS60 could also increase the level of DJ-1 protein at different doses. As evident from Western blot followed by band quantification, RNS60 upregulated the protein level of DJ-1 in mouse MN9D neuronal cells (Figure 2A,B). However, no such increase was found with NS (Figure 2A,B). Time-dependent analysis showed that 10% v/v RNS60 remained unable to increase the protein level of DJ-1 at 2 h of incubation (Figure 2C,D). However, a significant increase in DJ-1 protein was found at 6 h of incubation with RNS60 with maximum increase seen at 24 h of treatment (Figure 2C,D). Therefore, for further experiments while monitoring DJ-1 protein level, cells were treated with RNS60 for 24 h. To further confirm, we also monitored the upregulation of DJ-1 protein by immunofluorescence followed by calculation of DJ-1 MFI. Since MN9D cells are dopaminergic, we employed double-labeling for tyrosine hydroxylase (TH), marker of dopaminergic neurons, and DJ-1. Consistent to Western blot results, 24 h treatment with RNS60, but not NS, also resulted in the augmentation of DJ-1 protein in MN9D neuronal cells (Figure 2E,F). On the other hand, the level of TH did not change after treatment with either RNS60 or NS (Figure 2E), suggesting the specificity of the effect. Since RNS60 upregulated DJ-1 in MN9D dopaminergic neuronal cells, next, we examined whether RNS60 could increase DJ-1 in primary dopaminergic neurons. Mouse dopaminergic neurons were incubated with RNS60 and NS for 24 h followed by double-labeling for TH and DJ-1. Similar to MN9D cells, RNS60 at a dose of 5% v/v also markedly increased the level of DJ-1 in primary dopaminergic neurons (Figure 3A,B). This effect was specific to RNS60 as NS had no effect on the expression of DJ-1 in neurons (Figure 3A,B). Moreover, both RNS60 and NS had no effect on TH (Figure 3A,C), indicating that RNS60 specifically upregulates DJ-1 in dopaminergic neurons without increasing its signature marker TH. Mechanisms by which the transcription of DJ-1 gene occurs are poorly understood. Therefore, to understand the mechanism by which RNS60 increased the expression of DJ-1 in neuronal cells, we analyzed the mouse DJ-1 gene promoter using Mat-Inspector V2.2 search machinery and found the presence of a consensus cAMP response element (CRE) (Figure 4A) that allows the transcription factor CREB to bind. Therefore, we examined if RNS60 alone was capable of inducing the activation of CREB in MN9D neuronal cells by examining the phosphorylation of CREB. RNS60, but not NS, augmented CREB phosphorylation as portrayed by Western blot of phospho-CREB and total CREB (Figure 4B,C). RNS60 treatment upregulated the level of phospho-CREB, but not total CREB, at different minutes of stimulation. Although RNS60 upregulated the level of phospho-CREB in MN9D neurons significantly (p < 0.05 vs. control) at 5 min of stimulation, maximum upregulation (p < 0.01 vs. control) of phospho-CREB was seen at 15 min of RNS60 stimulation (Figure 4B,C). To confirm these results further, we performed double-label immunofluorescence of either phospho-CREB and actin (Figure 4D) or total CREB and actin (Figure 4E) followed by quantification of MFI for either phospho-CREB (Figure 4F) or total CREB (Figure 4G). Consistent to Western blot results, RNS60 increased the level of phospho-CREB (Figure 4D,F), but not total CREB (Figure 4E,G), in MN9D neuronal cells. Upon activation, CREB binds to DNA and therefore, to further reinforce this result, we carried out EMSA in order to check the DNA-binding activity of CREB. As evident from Figure 5, at different minutes (5, 15 and 30) of stimulation, RNS60 treatment markedly led to an increase in DNA-binding activity of CREB. However, RNS60-induced DNA-binding activity of CREB decreased at 60 min of stimulation (Figure 5). Moreover, NS remained unable to stimulate the activation of CREB (Figure 5), signifying the specificity. Next, to understand if CREB was directly engaged in the transcription of DJ-1 gene, we investigated the employment of CREB to DJ-1 gene promoter by ChIP assay. From immunoprecipitates of chromatin fragments of RNS60-treated neuronal cells with antibodies against CREB, we were also able to amplify 306 bp fragments (Figure 6A) from the DJ-1 promoter corresponding to the CRE (Figure 4A). The p300 and CREB-binding protein (CBP), two important histone acetyl transferases, are known to play an important role in gene transcription [28,29]. Therefore, next, we investigated whether p300 and CBP were also involved in RNS60-driven transcription of the DJ-1 gene in neuronal cells. As evident from PCR (Figure 6A) and real-time PCR (Figure 6B), RNS60 treatment markedly induced the staffing of CBP to the DJ-1 gene promoter. In contrast, RNS60 did not induce the recruitment of p300 to the CRE of DJ-1 gene promoter (Figure 6A,B), suggesting that p300 was not involved in RNS60-mediated transcription of DJ-1 gene in neuronal cells. Expectedly, similar to the enrolment of CBP and CREB to the CREs, RNS60 treatment was able to employ RNA polymerase to the DJ-1 gene promoter (Figure 6A,B). These results are specific as we remained unable to detect any amplification product either in control MN9D cells or NS-treated MN9D cells (Figure 6A,B). Furthermore, from the immuno-precipitates obtained with control IgG, we did not observe any amplification product (Figure 6A,B). Together, our results suggest that RNS60-induced transcriptional complex at the DJ-1 promoter contains CREB, CBP and RNA polymerase (Figure 6C). Next, to explore whether CREB is involved in RNS60-induced increase in DJ-1, we employed the siRNA approach. As expected, CREB siRNA, but not control siRNA, decreased the level of CREB in MN9D neuronal cells (Figure 7A,B). To understand whether the expression of DJ-1 depends on CREB in normal cells, we monitored the level of DJ-1 and found that CREB siRNA, but not control siRNA, decreased level of DJ-1 protein in neuronal cells (Figure 7A,C). This result suggest that DJ-1 expression is dependent on CREB in normal cells. Next, we analyzed RNS60-treated cells. Abrogation of RNS60-mediated up-regulation of DJ-1 by siRNA knockdown of CREB (Figure 7D–F) suggests that RNS60 upregulates the expression of DJ-1 via the CREB signaling pathway. Reduced glutathione is an important physiological antioxidant, which also functions as a thiol buffer, and DJ-1 is a key molecule in mediating cell survival by positively regulating the biosynthesis of reduced glutathione [16]. In addition, DJ-1 facilitates Nrf2-dependent detoxification pathways [30,31] and induces Akt activity via the suppression of phosphatase and tensin homolog (PTEN) [32,33]. DJ-1 can also function as peroxiredoxin-like peroxidase in vivo in which it can defend mitochondria against oxidative stress [34]. It has been shown that DJ-1 is capable of upregulating the human tyrosine hydroxylase gene by inhibiting sumoylation of pyrimidine tract-binding protein-associated splicing factor [35]. Therefore, upregulation of DJ-1 is considered to be beneficial for the damaged nigrostriatum in PD. Consequently, genetic inactivation of DJ-1 is found to be associated to early onset PD [6] and down-regulation of DJ-1 is seen in brains of sporadic PD patients [36]. Accordingly, delineating molecules to increase the level of DJ-1 in neurons and characterizing associated intracellular pathways that are required for the transduction of signals from the cell surface to the nucleus required for upregulation of DJ-1 gene are active areas of investigation. RNS60 is a physically reformed saline that is known to contain charge-stabilized nanobubbles [23,37,38,39]. Due to the turmoil caused by TCP, RNS60 is anticipated to contain charge-stabilized nanostructures consisting of an oxygen nanobubble core surrounded by an electrical double-layer at the liquid/gas edge [18]. However, it does not contain any active pharmaceutical ingredients [23,37,38,39]. Here, we establish that RNS60 is capable of increasing DJ-1 in dopaminergic neurons. RNS60, but not NS, increased the level of DJ-1 in mouse MN9D dopaminergic neuronal cells and primary dopaminergic neurons. Recently we have demonstrated protection of dopaminergic neurons and restoration of locomotor functions in MPTP-challenged mice by RNS60 [20]. Since mitochondrial dysfunction and oxidative stress have been shown to contribute to nigrostriatal degeneration in PD patients and in animal models of PD [40,41], our current results indicate that this DJ-1-upregulating efficacy of RNS60 may participate in RNS60-mediated protection of the nigrostriatum in a mouse model of PD. Moreover, due to implications of DJ-1 in the pathogenesis of PD, increase in DJ-1 in neurons by RNS60 may open up an important avenue whereby RNS60 may decrease nigrostriatal injury in PD. The intracellular signal transduction cascades required for the transcription of DJ-1 gene are poorly understood and molecular mechanisms by which RNS60 could upregulate DJ-1 were not known. While analyzing the DJ-1 gene promoter, we found the presence of a consensus cAMP response element (CRE) in the DJ-1 promoter. Recently we have demonstrated that RNS60 is capable of inducing the activation of CREB in microglia and oligodendrocytes [18,42], suggesting that RNS60 might employ the CREB pathway for the upregulation of DJ-1. Several lines of evidence presented in this manuscript also demonstrate that RNS60 increases the expression of DJ-1 gene in neuronal cells via CREB. First, RNS60 treatment increased the amount of phospho-CREB, but not total CREB, in neuronal cells. Second, RNS60 alone induced the DNA-binding activity of CREB in neuronal cells. Third, RNS60 treatment induced the enrollment of CREB to the DJ-1 gene promoter in neuronal cells. Fourth, histone acetyl transferases p300 and CREB-binding protein (CBP) play crucial roles in gene transcription. Interestingly, RNS60 treatment prompted the recruitment of CBP, but not p300, to the DJ-1 gene promoter (Figure 6C). Fifth, siRNA knockdown of CREB abrogated RNS60-mediated upregulation of DJ-1 in neuronal cells. At present, there is no effective therapy for the treatment of PD. Administration of levodopa/carbidopa and/or an agonist of DA has been the standard treatment for PD symptoms all over the world. However, these medications do not affect the disease course of PD. Moreover, these treatments often lead to side effects and disappointing outcomes. As a result, new effective treatments are necessary. Here, we have proven that RNS60 upregulates the expression nigrostriatum-protecting molecule DJ-1 in dopaminergic neuronal cells via activation of the CREB-CBP signaling pathway. Recently we have demonstrated that RNS60 treatment inhibits glial activation and protects dopaminergic neurons in the SNpc of MPTP mouse model of PD [20]. RNS60 is also capable of exhibiting neuroprotection in animal models of multiple sclerosis [19,43], Alzheimer’s disease [44], and traumatic brain injury [45]. In a phase II multicenter, randomized, double-blind, placebo-controlled trial in amyotrophic lateral sclerosis, RNS60 exhibits positive effects on respiratory and bulbar function [37]. Sources of the antibodies are provided in Table 1. Hank’s balanced salt solution (HBSS), fetal bovine serum (FBS), trypsin, and Dulbecco’s modified Eagle’s medium F-12 (DMEM/F-12) were from Mediatech (USA). Antibiotic-antimycotic mixture and poly-D-lysine were obtained from Sigma-Aldrich (St. Louis, MO, USA). RNS60 was generated at Revalesio (Tacoma, WA, USA) using Taylor-Couette-Poiseuille (TCP) flow as described before [18,20,43,45]. Briefly, sodium chloride (0.9%) for irrigation, USP pH 5.6 (4.5-7.0, Hospira, Lake Forest, IL), was processed at 4 °C in the presence of oxygen. Chemically, RNS60 is known to contain water, sodium chloride, 50–60 parts/million oxygen, but no active pharmaceutical ingredients. NS, normal saline from the same manufacturing batch was used as control because NS also contacted the same device surfaces as RNS60 and was bottled in the same way. From careful analysis, it was found that RNS60 and NS were chemically identical [18,19]. No difference between RNS60 and NS was also found by liquid chromatography quadrupole time-of-flight mass spectrometric analysis [18]. On the other hand, by using atomic force microscopy (AFM), we observed that RNS60 has a nanobubble composition different from that of NS [18]. These cells were obtained from Dr. A. Heller (University of Chicago, Chicago, IL, USA). Cells were maintained in DMEM (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% (v/v) heat inactivated FBS, 3.7 g/L NaHCO3, 50 U/mL penicillin, and 50 μg/mL streptomycin in an incubator with an atmosphere of 7% CO2 at 37 °C. These cells express abundant tyrosine hydroxylase (TH), synthesize dopamine (DA) and also quantitatively release DA [46,47,48]. Mouse primary dopaminergic neurons were isolated as described earlier [22,47,49]. Briefly, nigra was dissected as a thin slice of ventral mesencephalon tissue from E12.5 to 14 days old fetus, which was homogenized with 1 mL of trypsin for 5 min at 37 °C followed by neutralization of trypsin as described [22,47,49]. Then single cell suspension of nigral tissue was prepared that was plated in the poly-d-lysine pre-coated 75 mm flask. Cells were allowed to differentiate fully for 9–10 days before treatment. Western blotting was performed as described earlier [28,50] with some alterations. Briefly, cells were scraped in 1X RIPA buffer and protein was measured using Bradford reagent. Sodium dodecyl sulfate (SDS) sample buffer was added to protein samples and equal amount of protein from each group was electrophoresed on NuPAGE® Novex® 4–12% Bis-Tris gels (Thermo Fisher Scientific, Waltham, MA, USA) followed by transferring proteins onto a nitrocellulose membrane (Bio-Rad, Hercules, CA, USA). The membrane was then washed for 15 min in TBS plus Tween 20 (TBST) and blocked for 1 h in TBST containing BSA. Next, membranes were incubated overnight at 4 °C under shaking conditions with primary antibodies listed in Table 1. The next day, membranes were washed in TBST for 1 h, incubated in secondary antibodies (all 1:10,000; Jackson ImmunoResearch, West Grove, PA, USA) for 1 h at room temperature, washed for one more hour and visualized under the Odyssey® Infrared Imaging System (Li-COR, Lincoln, NE, USA). Total RNA was prepared followed by digestion with DNase for the removal of any contaminating genomic DNA. Semi-quantitative RT-PCR was carried out as described earlier [51,52], using a RT-PCR kit from Clontech. Briefly, 1 µg of total RNA was reverse transcribed using oligo(dT) as primer and MMLV reverse transcriptase (Thermo Fisher Scientific, Waltham, MA, USA). The resulting cDNA was appropriately diluted, and diluted cDNA was amplified. Amplified products were electrophoresed on a 1.8% agarose gels to be pictured by ethidium bromide staining. DJ-1: Sense: 5′-CCCCGTGCAGTGTAGCCGTG-3′ Antisense: 5′-CAGGCCGTCCTTCTCCACGC-3′ GAPDH: Sense: 5′-GGTGAAGGTCGGTGTGAACG-3′ Antisense: 5′-TTGGCTCCACCCTTCAAGTG-3′. It was performed using the ABI-Prism7700 sequence detection system (Thermo Fisher Scientific, Waltham, MA, USA) as described earlier [51,52]. The mRNA expressions of respective genes were normalized to the level of GAPDH mRNA. Data were processed by the ABI Sequence Detection System 1.6 software and analyzed by ANOVA. It was performed as described earlier [50,53]. Briefly, cover slips containing 100–200 cells/mm2 were fixed with 4% paraformaldehyde followed by treatment with cold ethanol and two rinses in phosphate-buffered saline (PBS). Samples were then blocked with 3% bovine serum albumin (BSA) in PBS-Tween-20 (PBST) for 30 min followed by incubation in PBST containing 1% BSA and mouse anti-DJ-1 (1:200) or rabbit anti-TH (1:500). After three washes in PBST (15 min each), slides were further incubated with Cy2 (Jackson ImmunoResearch Laboratories, Inc.). For negative controls, a set of culture slides was incubated under similar conditions without the primary antibodies. The samples were mounted and observed under an Olympus BX41 fluorescence microscope. MN9D neuronal cells were treated with RNS60 and NS under serum-free condition. At different time periods of treatment, nuclear extracts were prepared to perform EMSA as described previously [54,55] with some modifications. Briefly, IRDye infrared dye end-labeled oligonucleotides containing the consensus CREB DNA-binding sequence were purchased from Licor Biosciences and nuclear extract was incubated with infrared-labeled probe in binding buffer. Samples were separated on a 6% non-denaturing polyacrylamide gel in 0.25× TBE buffer (Tris borate-EDTA) and analyzed by the Odyssey Infrared Imaging System (LI-COR Biosciences, Lincoln, NE, USA). Recruitment of CREB to DJ-1 gene promoter was determined by ChIP assay as described earlier [56,57]. Briefly, MN9D neuronal cells were stimulated with RNS60 under serum free conditions for 1 h followed by fixation of cells by adding formaldehyde (1% final concentration). Cross-linked adducts were then resuspended and sonicated. ChIP was performed on the cell lysate by overnight incubation at 4 °C with 2 µg of anti-CREB, anti-CBP, anti-p300, or anti-RNA Polymerase II antibodies followed by incubation with protein G agarose (Santa Cruz Biotechnology, Dallas, TX, USA) for 2 h. The beads were washed and incubated with an elution buffer. To reverse the cross-linking and purify the DNA, precipitates were incubated in a 65 °C incubator overnight and digested with proteinase K. DNA samples were then purified, precipitated, and precipitates were washed with 75% ethanol, air-dried, and resuspended in TE buffer. The following primers were used for amplification of chromatin fragments of mouse DJ-1 gene. Sense: 5′-GAGATCTCATTTACCCTGATTTAA-3′ Antisense: 5′-GATCCTGATGCTGCTGCACCCACAG-3′ The “measurement module” of the Microsuite V Olympus software (B3 Biological Suite) was used to measure MFI as described by us in several studies [58,59]. Briefly, images were opened in their green channel and after that, measurement module was opened followed by the selection of two parameters including perimeter and MFI. The perimeter was outlined with rectangular box tool and then associated MFI in that given perimeter was automatically calculated. Statistical analyses were performed with Student’s t test (for two-group comparisons) and one-way ANOVA, followed by Tukey’s multiple-comparison tests, as appropriate (for multiple groups comparison), using GraphPad software (GraphPad Prism Version 9.5.1). Data represented as mean ± SD or mean ± SEM as stated in figure legends. A level of p < 0.05 was considered statistically significant. DJ-1 is a nigral protecting molecule and here, we describe a new neuroprotective property of RNS60, in which this physically-modified saline increases the transcription of DJ-1 in dopaminergic neuronal cells through CREB-CBP signaling pathway. Although the in vitro state of mouse dopaminergic neurons in culture and its dealing with RNS60 may not truly bear a resemblance to the in vivo condition of these cells in the brain of PD patients, our results delineate a new neuroprotective property of RNS60 and indicate that RNS60 may have therapeutic potential in PD and other mitochondria-related disorders.

PMC10002583

Alena I. Krysantieva,Julia K. Voronina,Damir A. Safin

A Novel Ambroxol-Derived Tetrahydroquinazoline with a Potency against SARS-CoV-2 Proteins

28-02-2023

ambroxol,tetrahydroquinazoline,synthesis,crystal structure,NMR,X-ray,molecular docking,molecular dynamics,SARS-CoV-2,COVID-19

We report synthesis of a novel 1,2,3,4-tetrahydroquinazoline derivative, named 2-(6,8-dibromo-3-(4-hydroxycyclohexyl)-1,2,3,4-tetrahydroquinazolin-2-yl)phenol (1), which was obtained from the hydrochloride of 4-((2-amino-3,5-dibromobenzyl)amino)cyclohexan-1-ol (ambroxol hydrochloride) and salicylaldehyde in EtOH. The resulting compound was produced in the form of colorless crystals of the composition 1∙0.5EtOH. The formation of the single product was confirmed by the IR and 1H spectroscopy, single-crystal and powder X-ray diffraction, and elemental analysis. The molecule of 1 contains a chiral tertiary carbon of the 1,2,3,4-tetrahydropyrimidine fragment and the crystal structure of 1∙0.5EtOH is a racemate. Optical properties of 1∙0.5EtOH were revealed by UV-vis spectroscopy in MeOH and it was established that the compound absorbs exclusively in the UV region up to about 350 nm. 1∙0.5EtOH in MeOH exhibits dual emission and the emission spectra contains bands at about 340 and 446 nm upon excitation at 300 and 360 nm, respectively. The DFT calculations were performed to verify the structure as well as electronic and optical properties of 1. ADMET properties of the R-isomer of 1 were evaluated using the SwissADME, BOILED-Egg, and ProTox-II tools. As evidenced from the blue dot position in the BOILED-Egg plot, both human blood–brain barrier penetration and gastrointestinal absorption properties are positive with the positive PGP effect on the molecule. Molecular docking was applied to examine the influence of the structures of both R-isomer and S-isomer of 1 on a series of the SARS-CoV-2 proteins. According to the docking analysis results, both isomers of 1 were found to be active against all the applied SARS-CoV-2 proteins with the best binding affinities with Papain-like protease (PLpro) and nonstructural protein 3 (Nsp3_range 207–379-AMP). Ligand efficiency scores for both isomers of 1 inside the binding sites of the applied proteins were also revealed and compared with the initial ligands. Molecular dynamics simulations were also applied to evaluate the stability of complexes of both isomers with Papain-like protease (PLpro) and nonstructural protein 3 (Nsp3_range 207–379-AMP). The complex of the S-isomer with Papain-like protease (PLpro) was found to be highly unstable, while the other complexes are stable.

A Novel Ambroxol-Derived Tetrahydroquinazoline with a Potency against SARS-CoV-2 Proteins We report synthesis of a novel 1,2,3,4-tetrahydroquinazoline derivative, named 2-(6,8-dibromo-3-(4-hydroxycyclohexyl)-1,2,3,4-tetrahydroquinazolin-2-yl)phenol (1), which was obtained from the hydrochloride of 4-((2-amino-3,5-dibromobenzyl)amino)cyclohexan-1-ol (ambroxol hydrochloride) and salicylaldehyde in EtOH. The resulting compound was produced in the form of colorless crystals of the composition 1∙0.5EtOH. The formation of the single product was confirmed by the IR and 1H spectroscopy, single-crystal and powder X-ray diffraction, and elemental analysis. The molecule of 1 contains a chiral tertiary carbon of the 1,2,3,4-tetrahydropyrimidine fragment and the crystal structure of 1∙0.5EtOH is a racemate. Optical properties of 1∙0.5EtOH were revealed by UV-vis spectroscopy in MeOH and it was established that the compound absorbs exclusively in the UV region up to about 350 nm. 1∙0.5EtOH in MeOH exhibits dual emission and the emission spectra contains bands at about 340 and 446 nm upon excitation at 300 and 360 nm, respectively. The DFT calculations were performed to verify the structure as well as electronic and optical properties of 1. ADMET properties of the R-isomer of 1 were evaluated using the SwissADME, BOILED-Egg, and ProTox-II tools. As evidenced from the blue dot position in the BOILED-Egg plot, both human blood–brain barrier penetration and gastrointestinal absorption properties are positive with the positive PGP effect on the molecule. Molecular docking was applied to examine the influence of the structures of both R-isomer and S-isomer of 1 on a series of the SARS-CoV-2 proteins. According to the docking analysis results, both isomers of 1 were found to be active against all the applied SARS-CoV-2 proteins with the best binding affinities with Papain-like protease (PLpro) and nonstructural protein 3 (Nsp3_range 207–379-AMP). Ligand efficiency scores for both isomers of 1 inside the binding sites of the applied proteins were also revealed and compared with the initial ligands. Molecular dynamics simulations were also applied to evaluate the stability of complexes of both isomers with Papain-like protease (PLpro) and nonstructural protein 3 (Nsp3_range 207–379-AMP). The complex of the S-isomer with Papain-like protease (PLpro) was found to be highly unstable, while the other complexes are stable. The history of mankind is known as a constant fight against problems of health. Of these problems, diseases turned to become pandemics are the most crucial and fatal since they become global and their consequences often cause hard-to-recover human and economic losses. The situation becomes even more crucial upon emergence of new diseases, for which neither efficient drugs nor therapy are known; thus, mankind is constantly on the bloody warpath against diseases. This “war”, obviously, requires the continuous design and synthesis of new molecules with desirable biological and therapeutic properties, as well as the efficient production of novel drugs based on them. During the last three years, mankind has been faced with one of the most dangerous viruses, namely severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2). This virus is a strain of coronavirus that causes coronavirus disease 2019 (COVID-19), which was announced as a pandemic by the World Health Organization (WHO) in March 2020. To date, as of February 2023, about 755 million infections were confirmed with more than 6.8 million deaths [1]. The situation with COVID-19 still remains complicated due new strains, of which variants of concern are alpha, beta, gamma, delta, and omicron; thus, drugs against COVID-19 are of particular value. There are no doubts that heterocyclic compounds are of great importance and play a pivotal role in nature. Suffice it to mention cytosine, guanine, adenine, and thymine, which are nitrogen-containing heterocyclic compounds, all four being nucleobases for deoxyribonucleic acid (DNA). The latter is fundamental for many viruses and all organisms; furthermore, an overwhelming majority of drugs are constructed from heterocyclic compounds. As such, nowadays, a great number of heterocyclic compounds with a pronounced pharmacological activity have been developed and are available on the market [2,3,4,5,6,7,8]. It was also reported that the heterocyclic fragments can serve as valuable and important resources for the development of coronaviruses’ treatment strategies and therapy [9,10]; furthermore, natural products containing heterocycles are also of interest as antiviral agents [11]. Of a great variety of heterocyclic compounds, quinazolines, containing fused benzene and pyrimidine six-membered rings, are a large family with biological properties of particular importance [12,13]. Of a variety of quinazoline derivatives [14,15,16], some important drugs can be highlighted such as: afatinib and gefitinib for treatment of non-small-cell carcinoma [17,18], lapatinib for treatment of advanced-stage or metastatic breast cancer [19], and erlotinib, an antitumor agent [20]. Partial hydrogenation of quinazoline leads to 1,2,3,4-tetrahydroquinazoline, where the pyrimidine fragment turns to 1,2,3,4-tetrahydropyrimidine, which synthesis was first reported by Pietro Biginelly in 1893 [21]. Since that time, 1,2,3,4-tetrahydropyrimidine and its derivatives have attracted attention of scientists from different fields due to both their efficient synthetic approaches [22,23,24,25,26], and biological and chemotherapeutic activities [27,28,29,30]. Furthermore, the novel 1,2,3,4-tetrahydropyrimidine derivative as inhibitor of SARS-CoV-2 was reported recently [31]. Thus, 1,2,3,4-tetrahydroquinazoline derivatives are of potential interest for the therapy of COVID-19. With all this in mind, as well as in continuation of our ongoing interest in the chemistry of nitrogen-containing six-membered rings [32,33,34,35,36,37,38,39] and in in silico studies of bioactive compounds [40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55], we have directed our attention to a novel 1,2,3,4-tetrahydroquinazoline derivative, named 2-(6,8-dibromo-3-(4-hydroxycyclohexyl)-1,2,3,4-tetrahydroquinazolin-2-yl)phenol (1), which was obtained from the hydrochloride of 4-((2-amino-3,5-dibromobenzyl)amino)cyclohexan-1-ol, which is commonly known under its trademark ambroxol, and salicylaldehyde and is constructed from the tetrahydroquinazoline, phenylene, and cyclohexylene rings (Figure 1). Theoretical density-functional-theory (DFT)-based calculations were applied to 1 to reveal its electronic and optical properties. Bioavailability, druggability, as well as absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties of 1 were evaluated using a set of online tools. Using an in silico molecular docking method, we have explored the binding modes and interactions of 1 with binding sites of a series of the SARS-CoV-2 proteins. Recently, it was reported about using ambroxol, which is known for its mucolytic and expectorant properties, as potential pharmacotherapy for SARS-CoV-2 [56]. We have also applied detailed in silico studies to ambroxol and, according to the molecular docking results, it was revealed that the molecule of ambroxol interacts much more efficiently with a series of the studied SARS-CoV-2 proteins in comparison to Favipiravir [48]; furthermore, in 1990 Lai et al. reported on substituted salicylaldehyde Schiff bases as new antiviral agents against coronavirus [57]. All this encouraged us to produce a salicylaldehyde Schiff base from ambroxol with the aim to study possible tautomerism of the resulting product as well as its tautomer-dictated ADMET properties and antiviral activity against a series of the SARS-CoV-2 proteins using computational approaches. As such, we have involved ambroxol hydrochloride into a condensation reaction with salicylaldehyde in ethanol; however, due to the presence of a secondary amine group in the structure of ambroxol, the resulting imine function, being in a close proximity, was further reacted with this amine nitrogen atom, yielding a cyclization product of the novel 1,2,3,4-tetrahydroquinazoline derivative, named 2-(6,8-dibromo-3-(4-hydroxycyclohexyl)-1,2,3,4-tetrahydroquinazolin-2-yl)phenol (1) (Figure 1). The resulting product was formed as colorless crystals, suitable for single-crystal X-ray diffraction, of the 1∙0.5EtOH composition. The IR spectrum of 1∙0.5EtOH, recorded in the KBr pellet, contains broad and sharp bands at 3100–3600 cm−1 (Figure 2), corresponding to OH and NH groups. Low intense bands at about 3050 and 3080 cm−1 are due to CH stretching vibrations of the aromatic rings, while bands at 2790–3000 cm−1 correspond to CH stretching vibrations of the cyclohexane, methylene, and methine fragments. A set of bands at 1530–1675 cm−1 corresponds to C=C bending and N–H stretching. The most intense bands at 1460 and 1485 cm−1 were attributed to bending of the C–H and O–H functionalities. Notably, some contribution to the experimental IR spectrum is also expected from EtOH, which is trapped in the crystal structure of 1∙0.5EtOH (see discussion below). The 1H NMR spectrum of 1∙0.5EtOH recorded in DMSO-d6 contains a single set of peaks, corresponding to both 1 and EtOH (Figure 3). Particularly, peaks of the cyclohexane CH2 protons were observed as a series of multiplets at 1.13–2.27 ppm, while the cyclohexane CH hydrogen atoms were found as two multiplets at about 2.53 and 3.44 ppm, both partially overlapping with the signals from the DMSO and EtOH methylene hydrogens, respectively. The NH and CH protons of the 1,2,3,4-tetrahydropyrimidine fragment were found as two doublets at 4.29 and 5.62 ppm, respectively, while the CH2 protons of the same fragment were shown as two doublets 3.70 and 3.85 ppm, respectively. The cyclohexanol and phenolic OH protons were revealed as a doublet and singlet at 6.43 and 10.57 ppm, respectively. Finally, a set of peaks at 6.60–7.43 ppm was assigned to aromatic protons. According to single-crystal X-ray diffraction, 1∙0.5EtOH crystallizes in triclinic space group P-1, and the asymmetric unit cell contains two molecules, named Molecule A and Molecule B, and a half of the EtOH molecule. Notably, in Molecule B all the cyclohexane atoms, except for the carbon atom attached to the OH fragment, are disordered over two positions with a ratio of 49.5% to 50.5%. The solvent molecule is also disordered over two positions but with ratio of 48.8% to 51.2%. It should also be noted that the molecule of 1 contains a chiral carbon C1 (Figure 4) and the crystal structure of 1∙0.5EtOH is a racemate. Bond lengths, and bond and dihedral angles in Molecule A and Molecule B are within the expected ranges (Table 1). Interestingly, the N2–C2 and O1–C16 bond lengths are about 0.06–0.07 Å shorter in comparison to the related N2–C1 and O2–C12 bonds (Table 1), which is explained by conjugation of the lone pairs of N2 and O1 with the π-system of the corresponding aromatic rings (Figure 4). Both Molecule A and Molecule B in the crystal structure of 1∙0.5EtOH are stabilized by an intramolecular hydrogen bond O1–H1∙∙∙N1, formed between the phenolic OH hydrogen atom and the nitrogen atom of the 1,2,3,4-tetrahydropyrimidine tertiary amine group (Figure 4, Table 2). The bulk sample of 1∙0.5EtOH was examined by means of powder X-ray diffraction analysis (Figure 5). The experimental X-ray powder pattern is in full agreement with the calculated powder pattern obtained from single-crystal X-ray diffraction, showing that the bulk material is free from phase impurities. The absorption spectrum of 1∙0.5EtOH in MeOH contains bands exclusively in the UV region up to about 350 nm with three clearly defined maxima at 212, 258, and 316 nm (Figure 6). The second band is accompanied with a low intense shoulder at about 280 nm (Figure 6). Surprisingly, 1∙0.5EtOH in MeOH exhibits dual emission and the emission spectra contains bands at about 340 and 446 nm upon excitation at 300 and 360 nm, respectively (Figure 7). The high-energy emission band is obviously due to intramolecular charge transfer in 1, which is revealed from comparison of the absorption spectrum (Figure 6) and excitation spectrum at λem = 340 nm (Figure 7). This emission band is remarkably red-shifted (~100 nm) upon excitation at 360 nm. This large shift together with comparison of the absorption spectrum (Figure 6) and excitation spectrum at λem = 450 nm (Figure 7) has allowed to conclude that the low-energy emission is most likely due the origin of a new species upon excitation. Particularly, since the molecule of 1 is characterized by prominent intramolecular hydrogen bonding (Figure 4, Table 2), it might undergo a tautomeric transformation in the excited state, yielding a zwitterion structure 1*, formed upon transition of the phenolic OH hydrogen atom to the tertiary nitrogen atom of the 1,2,3,4-tetrahydropyrimidine fragment (Figure 8). The latter molecule might isomerize with the formation of a zwitterionic form of the corresponding Schiff base 1*′, which is, in turn, structurally related to the corresponding cis-keto form 1*″ (Figure 8) [58,59,60]. We have also applied the DFT calculations to reveal structural and electronic features of 1, which structure was first optimized in gas phase. We have used the crystal structure geometry of Molecule A as a starting model for structural optimization. Notably, both enantiomeric forms of Molecule A yielded the same results of calculations and, for the sake of brevity, we focused on the R-isomer. The calculated geometrical parameters in the optimized structure of the R-isomer of 1 are in good agreement with the experimental ones (Table 1); however, a notable feature in the optimized structure of the R-isomer of 1 can be highlighted. Particularly, in the intramolecular hydrogen bonding O1–H1∙∙∙N1 the O1–H1 bond is 0.16 Å longer, while the H1∙∙∙N1 interaction is about 0.12 Å shorter in comparison to the experimental results (Table 2). Obviously, this discrepancy is explained by the DFT calculations performed in gas phase; thus, the optimized structure of 1 tends to adopt a zwitterionic isomeric form 1* (Figure 8). Analysis of the Mulliken atomic charges in the optimized structure of the R-isomer of 1 revealed that all the hydrogen atoms are positively charged with the highest value corresponding to the phenolic OH hydrogen atom, followed by H3, H4, H5, and H22 hydrogens (Figure 9). Of non-hydrogen atoms, the C7 and C15 are the most positively charged, followed by the N1 atom (Figure 9). The C8 carbon atom carries the most negative charge, followed by the C11, C13, and C9 atoms (Figure 9). The calculated IR and 1H NMR of the optimized structure of the R-isomer of 1 do not contradict the experimental results and some discrepancies are due to optimization of the structure in gas phase (Figure 2 and Figure 3, Table 3). It should be noted that all the frequencies in the calculated IR spectrum were found to be positive, indicating local energy minima for the optimized structure. The calculated UV-vis spectrum of the R-isomer of 1 in gas phase is in good agreement with the experimental one, and also exhibits bands exclusively in the UV region (Figure 6). Main transitions responsible for the bands in the calculated UV-vis spectrum are listed in Table 4. According to the DFT calculations, the energies of the HOMO and LUMO of the R-isomer of 1 in gas phase are −6.02787 and −1.16302 eV, respectively, with the corresponding energy gap of 4.86485 eV (Table 5). The ionization potential (I) and the electron affinity (A) value are large (Table 5) indicating low electron-donating and high electron-accepting properties. Chemical potential (μ) is −3.59545 eV, indicating electron-accepting ability and the low donating ability, which is supported by the corresponding high value of electronegativity, χ (Table 5). The electrophilicity index (ω), which is denoted as the energy of stabilization to accept electrons, is 2.65727 eV, indicating the pronounced electrophilic nature. Finally, the calculated structure can accept about 1.5 electrons, as evidenced from the corresponding ΔNmax value (Table 5). We have also visualized HOMO and LUMO for the R-isomer of 1. It was found that the HOMO is mainly delocalized over the 6,8-dibromo-1,2,3,4-tetrahydroquinazoline fragment, while LUMO is mainly spread over the 6-bromophenylene fragment and the methylene group of the 1,2,3,4-tetrahydropyrimidine fragment (Figure 10). The electrophilic and nucleophilic sites in the discussed optimized structure of the R-isomer of 1 were examined using the molecular electrostatic potential (MEP) analysis. The red and blue colours of the MEP surface correspond to electron-rich (nucleophilic) and electron-deficient (electrophilic) regions, respectively. On the MEP surface the most pronounced nucleophilic centers are located on the both hydroxyl oxygen atoms (Figure 11). As the most electrophilic region the cyclohexanol OH and NH hydrogen atoms, followed by H2, H7 and H14 hydrogens, can be highlighted (Figure 11). According to ProTox-II, a virtual lab for the prediction of toxicities of small molecules [61,62], the R-isomer of 1 belongs to a sixth class of toxicity, and is likely a pronounced inhibitor of kinase, ligand-gated ion channel, enzyme, oxidoreductase, family B G protein-coupled receptor, and phosphodiesterase with the probabilities of 64.0%, 16.0%, 8.0%, 4.0%, 4.0%, and 4.0%, respectively (Figure 12). According to the Toxicity Model Report, the R-isomer of 1 was revealed to be cytotoxic (Figure 12). As evidenced from the SwissADME [63] bioavailability radar, the discussed compound is preferred in all the six parameters, namely lipophilicity, size, polarity, insolubility, insaturation, and flexibility (Figure 13); thus, it is predicted to be suitable for oral bioavailability. The BOILED-Egg method was found to be efficient to predict the human blood–brain barrier (BBB) penetration and gastrointestinal absorption [64]. This approach is based on lipophilicity (WLOGP) and polarity (topological polar surface area, TPSA) (Figure 13). Points located in the yellow region (BOILED-Egg’s yolk) are molecules predicted to passively permeate through the BBB, while points located in the white region (BOILED Egg’s white) are molecules predicted to be passively absorbed by the gastrointestinal tract. Blue (PGP+) and red (PGP−) dots are for molecules predicted to be effluated and not to be effluated from the central nervous system by the P-glycoprotein, respectively. As evidenced from the blue dot position for the R-isomer of 1, both BBB penetration property and gastrointestinal absorption property are positive with the positive PGP effect on the molecule (Figure 13). We have further applied a molecular docking approach for both R-isomer and S-isomer with a series of the SARS-CoV-2 proteins. Docking is the best option to diminish the time and cost of synthesis and to increase the influence of the medicines; in addition, it is considered as a current and advantageous method to have insight information of the possible binding site of the ligand in the protein. The target structures were primarily selected in accordance with the structural features of the virus [65,66] as well as based on biological mechanisms and functions that can be utilized to reduce, prevent, or treat the virus [67] (Table 6). According to the docking analysis results, both isomers of 1 were found to be active against all the applied SARS-CoV-2 proteins with the best binding affinity with Papain-like protease (PLpro) and nonstructural protein 3 (Nsp3_range 207–379-AMP) (Figure 14, Table 6). Interactions responsible for binding of isomers of 1 and 2 with these two proteins are shown in Figure 14 and collected in Table 7. The obtained molecular docking results for both isomers of 1 are comparable with those found for initial redocked ligands [44], Remdesivir [44], Molnupiravir [47], and different tautomers of salen [51] and betulin [53], and superior to those calculated for Favipiravir [44]; furthermore, both isomers of 1, in general, interact with the applied SARS-CoV-2 proteins significantly more efficiently in comparison to the parent ambroxol [48]. Thus, 1 can be considered as a possible agent of further detailed investigation against COVID-19. We have also revealed ligand efficiency scores shed more light on the bioactivity of both isomers of 1 towards the applied SARS-CoV-2 proteins. As such, for all complexes we have calculated inhibition constant (Ki), miLogP, ligand efficiency (LE), ligand efficiency_scale (LE_Scale), fit quality (FQ), and ligand-efficiency-dependent lipophilicity (LELP) [68,69,70,71,72,73] (Table 6); furthermore, for comparison we have also calculated the same ligand efficiency scores for complexes of the studied proteins with initial ligands (Table 6). Notably, the Ki value must be as low as possible for a more efficient inhibition and should fall in the μM range for a compound to be considered as a hit, and >10 nM for a drug [72]. Furthermore, for a compound to be considered as a hit, the LE, FQ, and LELP parameters are recommended as ≥0.3, ≥0.8 and from −10 to 10, respectively [72]. Of all the complexes of the applied proteins, the ligand efficiency scores for complexes with the both isomers of 1 with Papain-like protease (PLpro) as well as for the complex of the S-isomer with nonstructural protein 3 (Nsp3_range 207–379-AMP) are close to be within the recommended ranges for a hit, although the LELP values are somewhat out of the recommended range (Table 6). We have additionally performed molecular dynamics simulations of the 50 ns time to evaluate interactions in complexes PLpro–R/S-isomer and Nsp_range 207–379-AMP–R/S-isomer. Complexes PLpro–R-isomer and Nsp_range 207–379-AMP–R/S-isomer each showed a highly stable root mean square deviation (RMSD) over the whole simulation time with the average values of 0.322, 0.362, and 0.380 nm, reaching the maximum values of 0.467, 0.496, and 0.539 nm, respectively (Figure 15). Contrarily, complex PLpro–S-isomer showed much higher RMSD over the whole simulation time reaching the value of about 1.6 nm with the average value of 1.021 nm (Figure 15), indicating its pronounced instability. The root mean square fluctuation (RMSF) value for complexes PLpro–R-isomer and Nsp_range 207–379-AMP–R/S-isomer was below 0.829, 0.657, and 0.529 nm, respectively (Figure 15). The strongest fluctuations of amino acid residues for complexes PLpro–R-isomer and Nsp_range 207–379-AMP–R/S-isomer are listed in Table 8. The radius of gyration (Rg) values for complexes PLpro–R-isomer and Nsp_range 207–379-AMP–R/S-isomer form relatively stable profiles (Figure 15), with the values varying in the ranges 2.572–2.686, 2.317–2.475, and 2.318–2.488 nm, respectively. The solvent accessible surface area (SASA) profiles were calculated for predicting the interaction between complexes and solvents. It was also established that the binding of the R-isomers to PLpro and Nsp3_range 207–379-AMP, and of the S-isomer to Nsp3_range 207–379-AMP did not impair the proteins’ interaction with the solvent molecule and the stability of the proteins (Figure 15). During the 50 ns simulation time, the average SASA was calculated as 298.41, 155.66, and 162.22 nm2 for complexes PLpro–R-isomer and Nsp_range 207–379-AMP–R/S-isomer, respectively. It was also established that in complex Nsp_range 207–379-AMP–R-isomer mainly 1 intermolecular hydrogen bond is formed during almost the whole simulation time, while in complex PLpro–R-isomer also 1 intermolecular hydrogen bond is formed but at about 18–50 ns (Figure 15). Complex Nsp_range 207–379-AMP–S-isomer is also characterized by 1 intermolecular hydrogen bond at about 8–23, 27–32, and 46–48 ns (Figure 15). The IR spectrum in a KBr pellet was recorded with a FT-IR FSM 1201 spectrometer in the range 400–4000 cm−1. The 1H NMR spectrum in DMSO-d6 were obtained on a Bruker Avance II 400 MHz spectrometer at 25 °C. Chemical shifts are reported with reference to SiMe4. UV–vis and fluorescent spectra from the 10−4 M freshly prepared solutions in freshly distilled MeOH were recorded on an Agilent 8453 instrument and RF-5301PC Shimadzu spectrofluorimeter, respectively. Powder X-ray diffraction was carried out using a Rigaku Ultima IV X-ray powder diffractometer. The parallel beam mode was used to collect the data (λ = 1.54184 Å). Elemental analyses were performed on a Thermo Flash 2000 CHNS analyzer (Waltham, MA, USA). A solution of salicylaldehyde (0.6 mmol, 0.073 g) in ethanol (10 mL) was added to a solution of ambroxol hydrochloride (0.5 mmol, 0.207 g) and KOH (0.5 mmol, 0.028 g) in the same solvent (20 mL). The mixture was heated at reflux for about 2 h. The resulting hot solution was filtered and allowed to cool to room temperature to give colorless crystals 1∙0.5EtOH suitable for single-crystal X-ray diffraction. Yield: 0.217 g (86%). Anal. Calc. for C21H25Br2N2O2.5 (505.25): C 49.92, H 4.99, and N 5.54; found: C 50.04, H 5.07, and N 5.48%. The X-ray diffraction data for 1∙0.5EtOH were collected at 150(2) K on a Bruker Smart Apex-II diffractometer, equipped with a CCD detector (Mo-Kα, λ = 0.71073 Å, graphite monochromator). Semi-empirical absorption correction was applied by the SADABS program [74]. The structure was solved by direct methods and refined by the full-matrix least squares in the anisotropic approximation for non-hydrogen atoms. The calculations were carried out by the SHELX-2014 program package [75] using Olex2 1.2 [76]. CCDC 2235606 contains the supplementary crystallographic data. These data can be obtained free of charge via https://www.ccdc.cam.ac.uk/structures or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44)-1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk. 2(C20H22Br2N2O2), C2H6O; Mr = 1010.50 g mol−1, triclinic, space group P–1, a = 5.7020(5), b = 16.8765(16), c = 21.6839(19) Å, α = 93.483(3), β = 97.142(3), γ = 98.957(3)°, V = 2038.4(3) Å3, Z = 2, ρ = 1.646 g cm−3, μ(Mo-Kα) = 3.999 mm−1, reflections: 18936 collected, 7941 unique, Rint = 0.054, R1(all) = 0.0737, wR2(all) = 0.1105, S = 1.027. The crystal structure geometries of the R-isomer and S-isomer of 1 were used as starting models for structural optimization. The ground state geometries were fully optimized without symmetry restrictions. The calculations were performed by means of the GaussView 6.0 molecular visualization program [77] and Gaussian 09, Revision D.01 program package [78] using the density functional theory (DFT) method with Becke-3-parametre-Lee–Yang–Parr (B3LYP) hybrid functional [79,80] and 6-311++G(d,p) [79,81] basis set. The vibration frequencies were calculated for the optimized structure in gas phase and no imaginary frequencies were obtained. The electronic isosurfaces of the HOMO and LUMO orbitals and MEP surfaces were generated from the fully optimized ground state geometry obtained using the B3LYP/6-311++G(d,p) method. The absorption and 1H NMR spectra of the fully optimized ground state geometry were simulated at the TD-DFT/B3LYP/6-311++G(d,p) and GIAO/B3LYP/6-311++G(2d,p) levels, respectively. Molecular docking of both isomers of 1 with a series of the SARS-CoV-2 proteins were carried using the CB-Dock2 server [82,83], which reveals protein cavities to guide blind docking by the algorithm of AutoDock Vina [84]. The targeted protein structures were subtracted from the RCSB PDB database [85] and were pretreated before the docking, including water removing and inserting hydrogen atoms and missing residues and charges. Docking results were visualized in BIOVIA Discovery Studio 2020 [86]. Molecular dynamics simulations were performed using the WebGro on-line service [87]. Parameters such as root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (Rg), solvent accessible surface area (SASA), and intermolecular hydrogen bonds were assessed. The complex was prepared for molecular dynamics simulations using GROMOS96 54a7 forcefield and was equilibrated using the canonical (NVT) and the isothermal-isobaric (NPT) ensembles. The ligand topology was generated with the PRODRG tool [88]. Simple point charge (SPC) was used as a solvent model (triclinic water box with size 50 × 75 × 70 Å) for protein–ligand complex [89]. This system was neutralized by adding sodium or chlorine ions based on the total charges. For minimization of the system before molecular dynamics simulations the steepest descent algorithm (5000 steps) was applied. The simulations were performed in the presence 0.15 M NaCl using the constant temperature (310 K) and pressure (1.0 bar). Approximate number of frames per simulation was 1000. The simulation time was set to 50 ns. Bioavailability, druggability, as well as absorption, distribution, metabolism, excretion, and toxicity properties were evaluated using the SwissADME [63], BOILED-Egg [64], and ProTox-II [61,62] tools. We have synthesized a novel 1,2,3,4-tetrahydroquinazoline derivative, named 2-(6,8-dibromo-3-(4-hydroxycyclohexyl)-1,2,3,4-tetrahydroquinazolin-2-yl)phenol (1), which was obtained from the hydrochloride of 4-((2-amino-3,5-dibromobenzyl)amino)cyclohexan-1-ol (ambroxol hydrochloride) and salicylaldehyde in EtOH. The resulting compound was produced in the form of colorless crystals of the composition 1∙0.5EtOH. The molecule of 1 contains a chiral tertiary carbon of the 1,2,3,4-tetrahydropyrimidine fragment and the crystal structure of 1∙0.5EtOH is a racemate. It was established that the compound absorbs in MeOH exclusively in the UV region up to about 350 nm; furthermore, 1∙0.5EtOH in the same solvent exhibits dual emission and the spectra contains bands at about 340 and 446 nm. While the high-energy emission band is due to intramolecular charge transfer, the low-energy emission is most likely due the excitation induced origin of a new species of 1, formed upon transition of the phenolic OH hydrogen atom to the tertiary nitrogen atom of the 1,2,3,4-tetrahydropyrimidine fragment. The DFT based calculations allowed to establish values of the global chemical reactivity descriptors, which revealed electron accepting and donating abilities of the reported compound, as well as its molecular electrostatic potential surface, which revealed electrophilic and nucleophilic sites. ADMET properties of the R-isomer of 1 were evaluated using the SwissADME, BOILED-Egg and ProTox-II tools, which predicted its positive human blood–brain barrier penetration and gastrointestinal absorption properties with the positive PGP effect on the molecule. According to the molecular docking analysis, both isomers of 1 were found to be active against all the applied SARS-CoV-2 proteins with the best binding affinity with Papain-like protease (PLpro) and nonstructural protein 3 (Nsp3_range 207–379-AMP). Ligand efficiency scores for both isomers of 1 inside the binding sites of the applied proteins were also revealed and compared with the initial ligands. Of all the complexes, the ligand efficiency scores for complexes of the both isomers of 1 with Papain-like protease (PLpro) as well as for the complex of the S-isomer with nonstructural protein 3 (Nsp3_range 207–379-AMP) are close to be within the recommended ranges for a hit, although the LELP values are somewhat out of the recommended range. Finally, molecular dynamics simulations of the 50 ns time revealed that complexes of the R-isomer with Papain-like protease (PLpro) and nonstructural protein 3 (Nsp3_range 207–379-AMP), and the complex of the S-isomer with nonstructural protein 3 (Nsp3_range 207–379-AMP) are stable, while the complex of the S-isomer with Papain-like protease (PLpro) was found to be highly unstable.

PMC10002590

Monica Terracciano,Simas Račkauskas,Andrea Patrizia Falanga,Sara Martino,Giovanna Chianese,Francesca Greco,Gennaro Piccialli,Guido Viscardi,Luca De Stefano,Giorgia Oliviero,Nicola Borbone,Ilaria Rea

ZnO Tetrapods for Label-Free Optical Biosensing: Physicochemical Characterization and Functionalization Strategies

23-02-2023

surface characterization,quantum yield,label-free detection,zinc oxide nanostructure,surface functionalization

In this study, we fabricated three different ZnO tetrapodal nanostructures (ZnO-Ts) by a combustion process and studied their physicochemical properties by different techniques to evaluate their potentiality for label-free biosensing purposes. Then, we explored the chemical reactivity of ZnO-Ts by quantifying the available functional hydroxyl groups (–OH) on the transducer surface necessary for biosensor development. The best ZnO-T sample was chemically modified and bioconjugated with biotin as a model bioprobe by a multi-step procedure based on silanization and carbodiimide chemistry. The results demonstrated that the ZnO-Ts could be easily and efficiently biomodified, and sensing experiments based on the streptavidin target detection confirmed these structures’ suitability for biosensing applications.

ZnO Tetrapods for Label-Free Optical Biosensing: Physicochemical Characterization and Functionalization Strategies In this study, we fabricated three different ZnO tetrapodal nanostructures (ZnO-Ts) by a combustion process and studied their physicochemical properties by different techniques to evaluate their potentiality for label-free biosensing purposes. Then, we explored the chemical reactivity of ZnO-Ts by quantifying the available functional hydroxyl groups (–OH) on the transducer surface necessary for biosensor development. The best ZnO-T sample was chemically modified and bioconjugated with biotin as a model bioprobe by a multi-step procedure based on silanization and carbodiimide chemistry. The results demonstrated that the ZnO-Ts could be easily and efficiently biomodified, and sensing experiments based on the streptavidin target detection confirmed these structures’ suitability for biosensing applications. The increasing interest in the earliest diagnosis has led to a considerable expansion of the biosensing field, with the continuous search for the most suitable materials for developing high-performance and low-cost detection devices [1,2]. Bulk materials lack important physicochemical properties (e.g., high surface-to-volume ratio, chemical reactivity, optical and electrical properties) needed for the development of biosensors, thus promoting the synthesis of nanostructured materials. Due to their distinct physicochemical properties, nanomaterials are appropriate as building blocks for biosensor development, providing high sensitivity and a valuable platform for analyzing single-molecular activity [3]. Among different materials, zinc oxide (ZnO) has arisen as a versatile, low-cost, and abundant metal oxide semiconductor useful to synthesize new transducer platforms for biosensor development [4,5,6]. Various forms of ZnO nanostructures, such as nanowires, nanotubes, tetrapods (Ts), nanorods, etc., are obtained by simple synthetic processes from low-cost materials [7,8]. The easy tuning of nanostructures’ morphology can significantly alter their physicochemical properties, such as enhancing their surface area and promoting variations in their optical, electrical, and electrochemical responses [9]. In addition, their properties, such as their ultraviolet (UV) light sensitivity, wide bandgap (~3.37 eV), high electron transfer capability (~60 meV), and high isoelectric point (IEP ~ 9.5), make them promising materials for biosensing applications [10]. Thanks to the great progress in nanotechnology and material science, new technological methods have been developed to fabricate nanostructured ZnO templates with high surface area and advanced properties. In a previous work, we demonstrated that the emerging ZnO tetrapodal structures (ZnO-Ts), compared to ZnO nanorods and nanoparticles, have better electrochemical properties as a transducer platform for biosensing applications [10]. ZnO-Ts are characterized by four connective arms protruding from the core center at an average angle of ~110°. Their characteristic three-dimensional structure makes them better than one-dimensional nanostructures for biosensing applications due to the requisite high conductivity, limited agglomeration, and easy fabrication of electrodes and sensing devices, opening the way for the development of a label-free, multiparametric-read-out platform [4,5]. A nanostructured matrix used in biosensor development acts as both a solid support, which is used for the platform for the immobilization of sensing biomolecules (i.e., bioprobes) and as a transducer, which is able to convert target detection into an analytical read-out signal [11]. The bioprobe label can be ascribed to a variety of biomolecules spanning among simple or complex structures: single-stranded (DNA and RNA) oligonucleotides, aptamers, peptide nucleic acids (PNAs), proteins, enzymes, antibodies, peptides, etc. [7,12,13,14,15,16,17]. The fabrication of a new generation of hybrid biosensors in which biological or bio-inspired molecules are fully integrated with transducer platforms strongly depends on the functionalization and bioconjugation strategies of the device’s surface [11,18]. Therefore, not only are the support’s physical and chemical properties fundamental for determining the bioprobe immobilization method but the stability of the transducer surface during the functionalization and detection procedures must also be considered [19]. In this context, we explored ZnO-Ts as potential nanostructures for the development of label-free optical biosensors, paying particular attention to their physicochemical properties. Three different ZnO-T samples (SH0, SH1, and SH2), synthesized by the combustion process and separated by centrifugation method in different size fractions, were characterized by dynamic light scattering (DLS), transmission electron microscopy (TEM), Brunauer-Emmett-Teller (BET) analyses, Fourier-transform infrared spectroscopy (FTIR), UV-vis spectroscopy, spectrofluorimetry, steady-state photoluminescence (PL), and fluorescence microscopy. The use of ZnO-Ts as transducers in biosensor development requires the creation of coupling points for the biomolecules’ immobilization (the so-called bioconjugation process), preserving the specific functionalities of biological receptors through good control of their orientation and organization on the inorganic surface [20]. To this aim, to estimate the number of hydroxyl groups (–OH) exposed on the tetrapods’ surface potentially available for functionalization with the bioprobe, we functionalized the nanostructures’ surface with 5′-DMT-3′-phosphoramidite-thymidine nucleotides via phosphoramidite chemistry and quantified, by colorimetric analysis, the amount of the DMT+ cation released in solution after coupling and acid treatment, which corresponds to the amount of reactive –OH groups exposed on the Ts’ surface [21]. Then, the best T sample in terms of colloidal stability, PL quantum yield, and chemical reactivity was chemically modified and bioconjugated with biotin (used here as the bioprobe model), experiencing chemical protocols able to preserve the physicochemical properties of the matrix. To this end, ZnO nanostructures were chemically modified with a multi-step procedure consisting of a silanization step by a 3-(aminopropyl)triethoxysilane compound (APT) followed by biotin conjugation by carbodiimide chemistry [22,23,24]. The efficacy of the resulting ZnO-biotin biosensor was evaluated by exploiting the real-time detection of streptavidin using optical methods based on steady-state PL and fluorescence microscopy. ZnO-Ts were obtained by a combustion method and then separated by size into three fractions, SH0, SH1, and SH2, as described in Section 3.2. Centrifugation separates ZnO-Ts into fractions by their mass; therefore, the largest ZnO-T structures are obtained in fraction SH1, the smallest are obtained in fraction SH2, and the initial fraction, SH0, has both large SH1 and small SH2 fractions. The morphology of as-synthetized ZnO-Ts was investigated by transmission electron microscopy (TEM). The analysis revealed nanostructures characterized by the typical tetrapodal shape consisting of four connected nanorods (legs) with a diameter of 20–100 nm and a length of 100–1000 nm (Figure 1a). Different tetrapods’ leg morphology could be connected to local fluctuations of synthesis conditions in the combustion chamber. The sample SH2 showed the smallest diameters and lengths; SH1 showed the biggest diameters and lengths, while SH0 had both small and big ZnO-Ts. A BET analysis (Figure S1) revealed the increase in the surface area along the series SH1 < SH0 < SH2, with corresponding values of 7.1 ± 0.1, 8.9 ± 0.1, and 10.8 ± 0.2 m2 g−1. The surface chemistry of bare ZnO-T samples was investigated by attenuated total reflectance Fourier-transform infrared (ATR–FTIR) spectroscopy. The spectra reported in Figure 1b show absorption bands at 3500 cm−1 and 1050 cm−1 due to the O–H stretching and bending vibrations, respectively. The band at 1630 cm−1, attributed to zinc carboxylate due to the synthesis process, is also evident [25,26]. Dynamic light scattering (DLS) analysis was performed to evaluate the stability of colloidal suspensions by measuring the hydrodynamic diameter (size), polydispersity index (PdI), and surface charge (zeta potential) of the three ZnO-T bare samples. To this aim, ZnO-Ts were dispersed in ultra-pure water (pH 7) at a concentration of 0.05 mg mL−1 following the procedure described in Section 3.3. SH0 showed a smaller size, lower PdI, and higher zeta potential than the other samples, thus confirming the good homogeneity and stability of the SH0 ZnO-T colloidal suspension (Table 1). However, we observed that the bare ZnO-Ts had a very high tendency to agglomerate in deionized water, leading to a larger size in the DLS analyses compared to the real size determined by TEM due to the lack of stabilizing surface chemistry [27]. Therefore, it was necessary to chemically stabilize the ZnO-Ts surface for biosensing purposes. The absorbance of ZnO-Ts dispersed in PBS-T was investigated by UV-vis spectroscopy in the wavelength range included between 280 and 800 nm (Figure 2a). In this interval, an absorption peak at 370 nm was observed; the peak can be ascribed to the intrinsic bandgap absorption of ZnO due to the electron transitions from the valence band to the conduction band (O2p → Zn3d) [28]. The fluorescence emission of ZnO-Ts in PBS-T was studied under an excitation light of 370 nm. The emission spectra of samples SH0, SH1, and SH2, reported in Figure 2b, are characterized by a broad band centered at 500 nm due to the defects present in the materials, in particular oxygen vacancies and –OH groups [29,30]. A weak peak at about 420 nm can also be observed. This peak can be attributed to the transitions from Zn interstitials to the valence band [31]. After this preliminary optical characterization, the fluorescence quantum yields (QYs) of SH0, SH1, and SH2 were determined by measuring the absorbance and the fluorescence intensity (λexc = 370 nm) of the samples dispersed in PBS-T at different concentrations ranging from 0.05 to 0.2 mg mL−1 (Figure 3a,b). QYs were estimated relative to Hoechst 33342 (Hoechst) used as a standard dye (the QY of Hoechst in DMF is 35% and calculated using the following equation: where n is the refractive indices of the media (PBS-T or DMF) and α represents the ratio between the integrated fluorescence intensity and the absorbance at λ = 370 nm) [32,33]. The coefficients αSH0 = (3.6 ± 0.4) × 104, αSH1 = (11.0 ± 0.8) × 104, and αSH2 = (4.4 ± 0.9) × 104 were obtained via linear regression from the plots of integrated fluorescence intensity versus absorbance for SH0, SH1, and SH2, respectively (Figure 3c). The coefficient αHoechst = (2.3 ± 0.1) × 105 was calculated for Hoechst in DMF. Using nPBS−T = 1.33 and nDMF = 1.43 as refractive indices. The QY values reported in Table 2 were determined from Equation (1). Proper surface chemical functionalization of the transducer material is paramount for biosensor development. The transducer surface is frequently functionalized to improve its physicochemical properties and enrich its functionalities. Therefore, the precise quantification of available functional groups on the transducer surface is fundamental to adequately control the chemical modification [22]. We determined the number of reactive –OH groups exposed on the bare ZnO-Ts’ surface using a non-conventional methodology based on nucleotide derivatization followed by a colorimetric assay. To this aim, ZnO-Ts (1, Figure 4a) were left to react with tetrazole-activated 5′-DMT-3′-phosphoramidite-thymidine nucleotides (2, Figure 4a) using the well-known phosphoramidite chemistry. This reaction allowed the binding of T nucleotides on the ZnO surface by the fast formation of 3′-phosphite-triester groups with the –OH groups exposed on the ZnO surface (3, Figure 4a). The amount of the bonded nucleotides, which reflects the amount of the reactive –OH groups on the ZnO-Ts’ surface, was assessed by spectrophotometrically quantifying the amount of the 5′-dimethoxytrityl cations (DMT+) released from the ZnO-T-bound 5′-DMT-protected nucleotides (3, Figure 4a) using a solution of dichloroacetic acid in dichloromethane (3% w/w). The release of the DMT+ cations generates a bright red-orange colored solution (Amax = 503 nm) whose color intensity is directly proportional to the content of –OH groups and quantifiable by the Lambert–Beer law (ε = 71,700 M−1 cm−1) as shown in Figure 4b [34]. The amount of DMT+ cation released in solution was 3.9 ± 0.2, 3.5 ± 0.6, and 1.8 ± 0.5 µmol mg−1, respectively, for SH0, SH1, and SH2 (Table 3). The selectivity of the device is another key issue in biosensor development. Selectivity is achievable by using specific bioprobes that can be either directly grown into the PSi matrix (in situ synthesis) or synthetized ex situ and then immobilized on the surface via electrostatic or covalent interactions [12,13,15]. This delicate step might affect the mobility, conformation, and functionalities of the selected bioprobes; therefore, various chemical strategies were developed to preserve the bioprobe’s functionality for biosensing applications [2]. Another important aspect to consider in biosensor development is the pH of aqueous-based solutions used during biofunctionalization procedures. The pH could hugely impact the ZnO-based nanostructures and their properties considering that ZnO is an amphoteric oxide easily dissolvable in both acid and basic conditions. The metal oxide ZnO in water solution undergoes hydrolysis, creating a hydroxide coating on its surface (≡M–OH) [19]. In acid conditions, the H3O+ ions react with the ZnO surface, causing the dissolution of nanostructures with rapid release of Zn2+(aq) in the alkaline condition (pH higher than 8.5) also occurs due to the dissolution of ZnO nanostructures related to their hydroxide which produces soluble species in the form of hydroxyl complexes such as Zn(OH)2(aq). The mechanisms of ZnO-based structures’ dissolution in acid and alkaline conditions are described in detail in the “Supporting Information”. Considering the results reported in the literature, the best working condition to develop a ZnO-based biosensor preserving the physicochemical properties is to use mild alkaline pH solutions [19,35,36]. The SH0 sample was chosen for the experiments of bioconjugation and optical sensing due to two main properties: the superior ability to form a stable colloidal suspension in water-based solutions, as highlighted by the measurements of DLS (Table 1), the reactive surface characterized by the presence of a larger amount of –OH groups, useful for an efficient functionalization, as demonstrated by the studies of –OH quantification (Table 3). The ability of the SH0 sample to act as an optical transducer of biomolecular interactions was investigated by studying selective biotin-streptavidin recognition. Biotin, a member of the water-soluble B-complex group of vitamins, is a molecule involved in a wide range of metabolic processes in humans and other organisms. Due to the strong affinity of biotin to streptavidin, it is commonly used as a model of interaction for the development of diagnostic tools. Once the biotin slides into the tight-fitting pocket of streptavidin, the flexible loop of streptavidin folds over the biotin and acts as a lid for stable binding [23]. Biotin bioprobes were immobilized on the surface of the SH0 ZnO-Ts following the functionalization procedure schematized in Figure 5a. Firstly, the bare ZnO-Ts were silanized by the chemical reaction between the triethoxy groups of APT and the −OH groups on the ZnO-TS surface, generating self-assembled monolayers covalently bonded to the surface via Zn–O–Si bonds able to passivate the surface. This chemical strategy improves the surface stability and introduces coupling points (–NH2 groups) for the immobilization of the bioprobe [11,37]. The surface was then biotinylated by the reaction between the N-Hydroxysuccinimide (NHS) esters of biotin (NHS-biotin) molecules and the amine groups of the ZnO-Ts via carbodiimide chemistry at slightly alkaline conditions (pH 8), yielding stable amide bonds. The chemical modification of the ZnO-T surface was analyzed by ATR-FTIR spectroscopy (Figure 5). As already observed by the graph reported in Figure 1, the spectrum of bare ZnO-Ts is characterized by a broad band at 3500 cm−1 due to the presence of –OH hydroxyl groups on the sample surface [25]. After the silanization process, the ZnO-Ts–APT displayed a decrease in the signals related to –OH groups involved in the covalent bond with the silane compound. The characteristic bands of APT, corresponding to the bending mode of free NH2 at 1520–1330 cm−1 and the rocking CHx vibration of the Si–OCHx bond at 870 cm−1, are well evident in the spectrum of the silanized sample. After the biotin immobilization, no significant changes in the FTIR spectrum were observed. The surface charge of the bare ZnO-Ts passed from −45 ± 10 mV to 5 ± 2 mV after the silanization process and to −27 ± 18 mV after biotinylation, confirming the success of the ZnO-T surface functionalization. The effectiveness of the surface functionalization was also demonstrated by photoluminescence (PL) measurements performed by exposing the sample to UV laser light (λ = 325 nm) and analyzing its emission spectrum after each functionalization step. In these investigations, the sample was deposited on a silicon piece. PL measurements were preferred to standard fluorescence investigations performed in solution because they allow exploring a wider spectral range with higher resolution due to the use of high-performing source and detector devices. Figure 5d shows the PL spectra of ZnO-Ts before and after each functionalization step. Compared to the fluorescence spectrum (λexc = 370 nm) of bare SH0 reported in Figure 2b, its corresponding PL spectrum excited at 325 nm allows observing an intense peak at 372 nm, which can be assigned to the radiative recombination of the electron-hole pairs due to the transition from the conduction band to the valence band (excitonic emission). The visible broad band that peaked at 500 nm (green emission), related to the recombination of electrons with photo-generated holes occupying defect sites, is also well evident [38]. After the silanization process, a slight shift of the peak at 372 nm towards shorter wavelengths (Δλ = −2 nm) was observed; the shift can be attributed to the coordination between the APT molecules and ZnO, which affects the band gap of the material. On the other hand, the decrease in the green emission intensity (about 40%) was due to the binding of the APT molecules with the ZnO-Ts through the –OH groups, reducing the availability of surface holes (i.e., the recombination centers) and, consequently, the intensity of green emission. This result agrees with the FTIR measurements highlighting the decrease in –OH groups after the silanization. After the functionalization with biotin, only a weak increase in the green emission intensity (about 15%) was monitored. The biotinylated ZnO-Ts can interact with streptavidin, a homo-tetrameric (66 kDa) protein from the bacterium Streptomyces avidinii with an extraordinary affinity to biotin with a dissociation constant (Kd) in the order of ≈10−14 mol L−1. The binding of biotin to streptavidin is one of the strongest non-covalent interactions known in nature, forming the basis for many diagnostic assays that require the formation of a specific linkage between biological macromolecules [23]. To verify the interaction between biotinylated ZnO-Ts and streptavidin, functionalized ZnO-Ts were preliminary incubated with 0.4 mM of Cy3-labelled streptavidin solution (PBS, pH 8) for 1 h under stirring (Figure 6a). After the incubation, biotinylated ZnO-Ts were washed, deposited on a silicon piece, left to dry, and analyzed by fluorescence microscopy. Figure 6b reports the fluorescence images of biotinylated ZnO-Ts before and after the incubation with fluorescent streptavidin under two different excitation wavelengths, 365 and 530 nm, respectively. The typical yellow emission of Cy3, well evident in the case of biotinylated ZnO-Ts incubated with Cy3-labeled streptavidin under an excitation of 530 nm, confirms the biomolecular interaction. An analogue investigation was also performed using a label-free streptavidin; in this case, the biomolecular recognition was monitored by PL analysis. The analysis did not reveal any variation in the PL spectrum. We can conclude that the analysis of the photoluminescence emission can be a valid strategy for studying the surface functionalization of ZnO-Ts; indeed, the results are consistent with the preliminary FTIR characterization. On the contrary, the technique did not provide information on the interaction between the biotinylated ZnO-Ts and streptavidin. The weak sensing capability could be attributed to the low specific surface area (up to 10 m2 g−1) that characterizes this nanostructured material compared to other forms of nanostructured ZnO such as ZnO-NWs [39] and porous ZnO [40]. Phosphate-buffered saline tablets (PBS CAS No.: P4417-50), Tween 20 (CAS No.: 9005-64-5), 3-aminopropyltriethoxysilane (APT CAS No.: 919-30-2), DMT-dT Phosphoramidite (CAS No.: 98796-51-1), tetrazole (CAS No.: 919-30-2288-94-8), deblocking solution of trichloroacetic acid in dichloromethane 3% w/w (CAS No.: 8-570-14); tetrahydrofuran anhydrous (THF CAS No.: 109-99-9), acetonitrile (CAS No.: 75-05-8), hydrochloric acid (HCl CAS No.: 7647-01-0), and sodium hydroxide solution (NaOH CAS No.: 1310-73-2) were all purchased from Sigma-Aldrich. Dimethyl sulfoxide (DMSO CAS No.: 67-68-5), sulfo-N-hydroxysuccinimide biotin (biotin-NHS CAS No.: 119616-38-5) water-soluble, streptavidin (SA CAS No.: 9013-20-1) from Streptomyces avidinii, and streptavidin−Cy3 (Cy3-SA CAS No.: S6402) from Streptomyces avidinii were purchased by Merck KGaA (DE). Hoechst 3342 Trihydrochloride Trihydrate-10 mg/mL solution (Hoechst CAS No.: 23491-45-4) in water was purchased from Invitrogen by Thermo Scientific. Absolute ethanol (EtOH CAS No.: A3678) was purchased from PanReac Applichem ITW Reagents. Ultra-pure water (18 Ω·cm) purified from a Milli-Q purification system (Millipore, Bedford, MA, USA) was used to prepare all the aqueous solutions. ZnO-Ts were synthesized in a vertical reactor with heating similar to a combustion method described earlier [41]. Briefly, micron-sized Zn particles, entrained in the air, were supplied to the reactor and combusted to form a nanopowder of ZnO. Differently from the previous method, heating from the combustion of methane gas was used instead of electrical heating, and wet air was additionally supplied to the reactor to control the reaction rate. The nanopowder containing ZnO-Ts and some other forms on nanowires and nanoparticles were collected downstream of the reactor on the cellulose filter and further used in the analysis. The tetrapodal shape of the ZnO nanomaterials obtained was confirmed by in-depth TEM analysis in earlier work [41] using the same synthesis conditions. The combustion synthesis method was used because it delivers the synthesis of chemically pure ZnO-Ts in high yield, which is especially attractive for further practical application [10]. After synthesis, the as-obtained ZnO-Ts mixture was separated into 3 fractions with the help of a centrifuge (Table 4). The initial ZnO-T mixture was marked as SH0; it was suspended in isopropanol (IPA CAS No.: 67-63-0) at a concentration of 1 mg/mL, sonicated in the bath for 1 h, and further separated in the centrifuge at a rotation speed of 1000 rpm. The sediments were collected and marked as SH1. The supernatant was further centrifuged at a rotation speed of 3000 rpm; the sediments were then collected and marked as SH2. All fractions were characterized without further processing. Dynamic light scattering (DLS). Each sample of ZnO-Ts (SH0, SH1, and SH2) was prepared for DLS characterization as described in the following. A stock suspension of ZnO-Ts with a concentration of 0.2 mg mL−1 was obtained by dispersing the powder in PBS 1× + 0.1% Tween 20 (PBS-T). Then, a sample with a concentration of 0.05 mg mL−1 was prepared by dispersing an aliquot of the stock suspension in ultra-pure water (pH = 7). The sample was centrifuged at 3500 rpm for 3 min and resuspended in ultra-pure water to completely remove the PBS-T. Before the DLS analysis, the sample was sonicated for 5 min. Hydrodynamic diameter and surface ζ-potential of ZnO-Ts were measured using a Zetasizer Nano-ZS instrument (Malvern Instrument Ltd., UK) equipped with a He-Ne laser (633 nm, scattering angle of 90°, 25 °C). Size distribution and surface zeta-potential values were obtained by averaging three measurements. UV-Vis spectroscopy. Absorption spectra of ZnO-Ts were acquired using a Jasco V-730 UV-Vis double beam spectrophotometer (Jasco Inc., Easton, PA, USA) in the wavelength range of 280–800 nm. The samples were analyzed at the concentration of 0.2 mg mL−1 in PBS-T using a quartz cell with a path length of 10 mm and a total volume capacity of 0.5 mL. Fluorescence spectroscopy. Fluorescence emission spectra of ZnO-Ts were acquired using a PerkinElmer LS 55 Luminescence spectrometer (PerkinElmer Inc., Waltham, MA, USA) in the wavelength range of 380–700 nm, setting the excitation wavelength at 370 nm, and the excitation and emission bandwidth at 10 nm and 5 nm, respectively. The samples were suspended in PBS-T, and a quartz cell with a path length of 10 mm and a total volume capacity of 0.5 mL was used. Fluorescence microscopy. Fluorescence images of ZnO-Ts were acquired using a Leica AF6000LX-DM6M-Z microscope (Leica Microsystems, Mannheim, Germany), controlled by LAS-X (Leica Application Suite; rel. 3.0.13) software and equipped with a DFC7000T Leica Camera. Fluorescence images were obtained using a 50× objective and an I3 cube constituted by a 365 nm band-pass excitation filter. ZnO-Ts were dispersed in ultra-pure water (250 mg mL−1), and 20 μL of the samples were left to dry on silicon pieces to acquire the images. Brunauer-Emmett-Teller (BET) analysis. The textural parameters of the samples were determined by nitrogen adsorption–desorption isotherms at −200 °C (77 K) using a Quantachrome Autosorb-iQ-KR/MP automated gas sorption analyzer. Before the analysis, the powder samples were outgassed under a vacuum at 200 °C for 3 h. The specific surface area was calculated using the BET (Brunauer–Emmett–Teller) equation. Transmission Electron Microscopy. The morphology of the samples SH0, SH1, and SH2 was investigated using a transmission electron microscope (TEM, Jeol JEM-1400, Jeol Ltd., Akishima, Japan). To this aim, the samples were dispersed in ultra-pure water at a concentration of 0.2 mg mL−1 and dropped on a carbon-coated copper TEM grid before air-drying overnight at room temperature. ATR-Fourier Transform Infrared Spectroscopy. The surface chemical composition of ZnO-Ts was investigated by attenuated total reflectance Fourier transform infrared spectroscopy (ATR–FTIR). The ATR– FTIR spectra of the samples were obtained using a Nicolet iS50 (Thermo Scientific) FTIR spectrometer equipped with a Germanium (Ge) crystal element. The ATR–FTIR spectra were recorded in the wavenumber region 4000–525 cm−1 with a resolution of 4 cm−1. The measurements were carried out on dried ZnO-Ts deposited on the Ge crystal. Photoluminescence. The photoluminescence analysis of ZnO-Ts was performed by depositing 20 μL of the samples dispersed in ultra-pure water (0.2 mg mL−1) on silicon pieces; the samples were left to dry at RT before the investigations. The photoluminescence (PL) spectra of ZnO-Ts were excited by a continuous wave He-Cd laser at 325 nm (KIMMON Laser System). PL was collected at normal incidence to the surface of samples through a fiber, dispersed in a spectrometer (Princeton Instruments, SpectraPro 300i), and detected using a Peltier-cooled charge-coupled device (CCD) camera (PIXIS 100F). A long pass filter with a nominal cut-on wavelength of 350 nm was used to remove the laser line at the monochromator inlet. Quantum yields (QYs) of the samples SH0, SH1, and SH2 dispersed in PBS-T were calculated by measuring their absorbance and the integrated fluorescence intensity at different concentrations (0.05, 0.1, 0.15, and 0.2 mg mL−1). Absorption spectra were obtained using a Jasco V-730 UV-Vis double beam spectrophotometer (Jasco Inc., Easton, PA, USA) in the wavelength range of 280–800 nm. Emission spectra were acquired using a PerkinElmer LS 55 Luminescence spectrometer (PerkinElmer Inc., Waltham, MA, USA) in the wavelength range of 410–600 nm, setting the excitation wavelength at 370 nm. The QYs of ZnO-Ts were estimated relative to Hoechst 33342 (Hoechst) used as a standard dye. To this aim, Hoechst was dispersed in DMF at the concentrations of 0.0075, 0.015, 0.025, and 0.050 mg mL−1 and analyzed. ZnO-T samples SH0, SH1, and SH2 (20 mg) reacted with tetrazole activated (18 mg) 5′-(dimethoxytrityl)-thymidine-phosphoramidite dissolved in dry THF (30 mg/mL) for 1 h at room temperature under mild stirring [34]. Then, the samples were centrifuged for 3 min at 5000 rpm and washed 10 times with acetonitrile to remove adsorbed reagents. The removal of the 5′-dimethoxytrityl protecting group from the supports bound 5′-terminal nucleotide was performed by using the deblocking solution of dichloroacetic acid in dichloromethane (3% w/w). The ZnO-Ts were centrifuged at 5000 rpm, and the amount of DMT in the supernatant was measured by a UV-Vis spectrometer (V-73, Jasco Europe, Italy) at 503 nm (ε = 71,700 M−1 cm−1). Functionalization of the samples was performed in triplicate. To investigate the stability of ZnO-Ts on exposure to acidic and alkaline conditions, ZnO-Ts were dispersed at a concentration of 0.2 mg mL−1 in PBS-T at different pHs (3, 4, 5, 6, 7, and 8) for 0, 2, and 24 h. After the exposure, the samples were investigated by absorbance, fluorescence, and ATR-FTIR analyses. The studies of functionalization were performed on the SH0 sample. ZnO-Ts (1 mg) of SH0 were amino-modified using APT 5% (v/v) in absolute EtOH (1 mL final volume) for 1 h at room temperature (RT) in mild stirring conditions [42]. The sample was centrifuged at 3500 rpm, and the supernatant was replaced twice with EtOH and once with PBS (pH = 8) to remove unreacted APT. Amino-modified ZnO-Ts (1 mg) were resuspended in biotin-NHS solution (1 mM in PBS, pH 8) and left to react for 1 h [23]. ZnO-Ts were characterized after each functionalization step by DLS, photoluminescence, and ATR-FTIR analyses. Two aliquots of biotinylated ZnO-Ts were incubated with 0.4 mM of SA and Cy3-SA and left to interact for 1 h with an agitation of 800 rpm. After the interaction, the samples were washed and resuspended in PBS (pH = 8). The interaction between the biotinylated ZnO-Ts and the streptavidin was monitored by fluorescence microscopy and photoluminescence spectroscopy. The growing advancement of the biosensor field provides a powerful driving force for the constant research and fabrication of advanced nanostructured materials with enhanced properties for developing a new generation of diagnostic devices. Zinc oxide (ZnO) is one of the most interesting metal oxide materials used in biosensing due to its unique and versatile physicochemical properties. It has been observed that the properties of ZnO can be improved through the nanoscale-up architectural process, making ZnO-based structures suitable for biosensing applications and opening the way for the development of multiparametric, label-free transducers. In this work, we discussed the potentiality and performance of various novel ZnO-tetrapod nanostructures for label-free optical biosensing applications. The physical and chemical properties of three different ZnO-T samples (SH0, SH1, and SH2), synthesized by the combustion process and separated by a centrifugation method in different size fractions, were evaluated by DLS, TEM, BET analyses, FTIR, UV-vis spectroscopy, spectrofluorimetry, steady-state PL, and fluorescent microscopy. Then, we explored the chemical reactivity of the samples by surface coupling with monomeric oligonucleotide bases via the phosphoramidite method and quantified the available functional hydroxyl groups (–OH) on the transducer surface necessary for the subsequent functionalization steps. The best ZnO-T sample in terms of colloidal stability, PL quantum yield, and chemical reactivity was chemically modified and bioconjugated with biotin using chemical protocols able to preserve the physicochemical properties of the matrix based on silanization and carbodiimide chemistry. FTIR, zeta potential and PL analyses confirmed the successful obtainment of the biotinylated ZnO-T-based biosensor. The sensing properties of the obtained device were investigated by optical methods based on the steady-state PL and fluorescence microscopy, confirming the biotin-streptavidin interaction. Although further modifications in the synthetic process will have to be carried out to implement the chemical-physical properties of these emerging structures for label-free biosensing applications, the detailed analysis of ZnO nanostructures conducted in this study will contribute to future biosensing applications of these appealing structures.

PMC10002591

Daniel Augustynowicz,Marta Kinga Lemieszek,Jakub Władysław Strawa,Adrian Wiater,Michał Tomczyk

Phytochemical Profiling of Extracts from Rare Potentilla Species and Evaluation of Their Anticancer Potential

02-03-2023

Potentilla,Rosaceae,polyphenols,LC–HRMS,colorectal cancer,LS180 cells,cytotoxicity,CCD841 CoN cells

Despite the common use of Potentilla L. species (Rosaceae) as herbal medicines, a number of species still remain unexplored. Thus, the present study is a continuation of a study evaluating the phytochemical and biological profiles of aqueous acetone extracts from selected Potentilla species. Altogether, 10 aqueous acetone extracts were obtained from the aerial parts of P. aurea (PAU7), P. erecta (PER7), P. hyparctica (PHY7), P. megalantha (PME7), P. nepalensis (PNE7), P. pensylvanica (PPE7), P. pulcherrima (PPU7), P. rigoi (PRI7), and P. thuringiaca (PTH7), leaves of P. fruticosa (PFR7), as well as from the underground parts of P. alba (PAL7r) and P. erecta (PER7r). The phytochemical evaluation consisted of selected colourimetric methods, including total phenolic (TPC), tannin (TTC), proanthocyanidin (TPrC), phenolic acid (TPAC), and flavonoid (TFC) contents, as well as determination of the qualitative secondary metabolite composition by the employment of LC–HRMS (liquid chromatography–high-resolution mass spectrometry) analysis. The biological assessment included an evaluation of the cytotoxicity and antiproliferative properties of the extracts against human colon epithelial cell line CCD841 CoN and human colon adenocarcinoma cell line LS180. The highest TPC, TTC, and TPAC were found in PER7r (326.28 and 269.79 mg gallic acid equivalents (GAE)/g extract and 263.54 mg caffeic acid equivalents (CAE)/g extract, respectively). The highest TPrC was found in PAL7r (72.63 mg catechin equivalents (CE)/g extract), and the highest TFC was found in PHY7 (113.29 mg rutin equivalents (RE)/g extract). The LC–HRMS analysis showed the presence of a total of 198 compounds, including agrimoniin, pedunculagin, astragalin, ellagic acid, and tiliroside. An examination of the anticancer properties revealed the highest decrease in colon cancer cell viability in response to PAL7r (IC50 = 82 µg/mL), while the strongest antiproliferative effect was observed in LS180 treated with PFR7 (IC50 = 50 µg/mL) and PAL7r (IC50 = 52 µg/mL). An LDH (lactate dehydrogenase) assay revealed that most of the extracts were not cytotoxic against colon epithelial cells. At the same time, the tested extracts for the whole range of concentrations damaged the membranes of colon cancer cells. The highest cytotoxicity was observed for PAL7r, which in concentrations from 25 to 250 µg/mL increased LDH levels by 145.7% and 479.0%, respectively. The previously and currently obtained results indicated that some aqueous acetone extracts from Potentilla species have anticancer potential and thus encourage further studies in order to develop a new efficient and safe therapeutic strategy for people who have been threatened by or suffered from colon cancer.

Phytochemical Profiling of Extracts from Rare Potentilla Species and Evaluation of Their Anticancer Potential Despite the common use of Potentilla L. species (Rosaceae) as herbal medicines, a number of species still remain unexplored. Thus, the present study is a continuation of a study evaluating the phytochemical and biological profiles of aqueous acetone extracts from selected Potentilla species. Altogether, 10 aqueous acetone extracts were obtained from the aerial parts of P. aurea (PAU7), P. erecta (PER7), P. hyparctica (PHY7), P. megalantha (PME7), P. nepalensis (PNE7), P. pensylvanica (PPE7), P. pulcherrima (PPU7), P. rigoi (PRI7), and P. thuringiaca (PTH7), leaves of P. fruticosa (PFR7), as well as from the underground parts of P. alba (PAL7r) and P. erecta (PER7r). The phytochemical evaluation consisted of selected colourimetric methods, including total phenolic (TPC), tannin (TTC), proanthocyanidin (TPrC), phenolic acid (TPAC), and flavonoid (TFC) contents, as well as determination of the qualitative secondary metabolite composition by the employment of LC–HRMS (liquid chromatography–high-resolution mass spectrometry) analysis. The biological assessment included an evaluation of the cytotoxicity and antiproliferative properties of the extracts against human colon epithelial cell line CCD841 CoN and human colon adenocarcinoma cell line LS180. The highest TPC, TTC, and TPAC were found in PER7r (326.28 and 269.79 mg gallic acid equivalents (GAE)/g extract and 263.54 mg caffeic acid equivalents (CAE)/g extract, respectively). The highest TPrC was found in PAL7r (72.63 mg catechin equivalents (CE)/g extract), and the highest TFC was found in PHY7 (113.29 mg rutin equivalents (RE)/g extract). The LC–HRMS analysis showed the presence of a total of 198 compounds, including agrimoniin, pedunculagin, astragalin, ellagic acid, and tiliroside. An examination of the anticancer properties revealed the highest decrease in colon cancer cell viability in response to PAL7r (IC50 = 82 µg/mL), while the strongest antiproliferative effect was observed in LS180 treated with PFR7 (IC50 = 50 µg/mL) and PAL7r (IC50 = 52 µg/mL). An LDH (lactate dehydrogenase) assay revealed that most of the extracts were not cytotoxic against colon epithelial cells. At the same time, the tested extracts for the whole range of concentrations damaged the membranes of colon cancer cells. The highest cytotoxicity was observed for PAL7r, which in concentrations from 25 to 250 µg/mL increased LDH levels by 145.7% and 479.0%, respectively. The previously and currently obtained results indicated that some aqueous acetone extracts from Potentilla species have anticancer potential and thus encourage further studies in order to develop a new efficient and safe therapeutic strategy for people who have been threatened by or suffered from colon cancer. Cancer, a non-infectious disease, is one of the most dreadful diagnoses that severely impacts a patient’s life quality. Unfortunately, cancer is a significant and increasing cause of death worldwide. The European Cancer Information System (ECIS) estimated an increase in new cases of cancer in the European Union (EU-27) from 2.68 million in 2020 to 3.24 million in 2040, a 21% increase, while the cancer-related death toll is estimated to increase from 1.26 million to 1.66 million cases, a 31.8% increase. Colorectum cancer is the second-most-diagnosed cancer type in EU-27 countries, with over 0.34 million cases in 2020; however, in 2040, it will overtake breast cancer as the most commonly diagnosed cancer type with over 0.43 million cases [1]. The most frequently used method to treat early-stage colorectal cancer is surgical resection, which effectively relieves the patient’s symptoms. However, approximately 25% to 30% of patients after successful surgery will develop metastases within 5 years [2]. Moreover, in the further stages, unresectable metastatic cancer systemic therapy includes chemotherapy, radiotherapy, immunotherapy, and biological therapy, such as antibodies to cellular growth factors, as well as their combinations [3]. Unfortunately, these treatment methods are inextricably linked with many side effects, such as pain, emotional stress, fatigue, a negative impact on fertility, and subsequent cancers [4]. Biologically active molecules in medicinal plants can be employed to reduce side effects and support the efficacy of the therapy. Notably, Potentilla species are widely used in traditional medicine for the treatment of dysentery, diarrhoea, diabetes mellitus, unspecified forms of cancer, and inflammation of the skin [5,6]. The pharmacological properties of Potentilla species stem from their secondary metabolite composition, which includes a predominant presence of polyphenols, such as hydrolysable and condensed tannins, flavonoids, and phenolic acid, as well as triterpenoids. These substances are associated with antioxidant, anti-inflammatory, and antimicrobial properties [5]. Numerous in vitro experiments on compounds obtained from Potentilla species have shown efficacy against various cancer cell lines, e.g., methanol extract from P. discolor inhibited the proliferation and induced the apoptosis of MC3 and YD-15 (human mucoepidermoid carcinoma) [7], ethyl acetate extracts from P. recta and P. astracanica decreased viability of HEp-2 (human cervix carcinoma) [8], and selected extracts and fractions from aboveground materials of P. alba significantly reduced the viability and proliferation of HT-29 (human colon adenocarcinoma) [9]. In a previous study, we demonstrated that aqueous acetone extracts from the aerial parts of selected Potentilla species showed great chemopreventive potential by decreasing the viability and proliferation of LS180 (human colon adenocarcinoma) cells, simultaneously causing substantial damage to their cell membranes while having a significantly weaker impact on normal colon epithelial cell line CCD841 CoN [10]. The present study is a continuation of that previous investigation conducted by the authors, concerning an assessment of the cytotoxicity and antiproliferative effect of aqueous acetone extracts from selected, rare Potentilla species against human colon cancer cell line LS180 and normal colon epithelial cell line CCD841 CoN. Additionally, identification of the marker metabolites present in extracts using LC–HRMS analysis was conducted to reveal and validate correlations between the qualitative chemical composition of the investigated samples and possible mechanisms of action. Polyphenols are among the major secondary metabolites that are accountable for the pharmacological activities of plant-based preparations. The major group of polyphenols include flavonoids, phenolic acids, hydrolysable and condensed tannins, lignans, and stilbenes [11]. Potentilla species are well-known for their abundance of tannins and flavonoids, which contribute to certain traditional applications aimed at tackling diarrhoea, microbial infections, inflammations of the upper and lower gastrointestinal tract, diabetes mellitus, etc. [5,12]. In our study, extracts from the aerial and underground parts of common and rare Potentilla species were prepared using 70% acetone and were quantitative assessed for the general polyphenolic classes contents using colourimetric methods. The level of phenolic compounds in the extracts from selected Potentilla species are presented in Table 1. Extracts from the underground parts, namely, PAL7r and PER7r, were found to contain the highest total phenolic (TPC) and total tannin (TTC) contents (268.63, 237.56, and 326.28, 269.79 mg gallic acid equivalent (GAE)/g extract, respectively). On the other hand, among extracts from the aerial parts, PFR7 and PPE7 had the highest TPC and TTC values (240.1, 178.65, and 218.85, 195.97 mg GAE/g extract, respectively), while PAU7 and PTH7 revealed the lowest TPC and TTC values (148.38, 129.2, and 149.77, 132.55 mg GAE/g extract, respectively). Moreover, PFR7 was found to contain the highest total proanthocyanidin content (TPrC) (53.59 mg catechin equivalent (CE)/g extract), notably higher than that of other herb extracts. According to our previous study and the results herein, extracts from rhizomes, namely, PAL7r and PER7r, had remarkably higher proanthocyanidin contents than their above-ground counterparts (72.63 and 61.61 vs. 21.28 and 2.05 mg CE/g extract, respectively) [12]. Moreover, PAL7r and PER7r had the highest total phenolic acid content (TPAC), followed by PFR7 (221.08, 263.54, and 197.83 mg caffeic acid equivalent (CAE)/g extract, respectively). On the contrary, PAL7r and PER7r had the lowest total flavonoid content (TFC) values, which were significantly lower than those of all other extracts. PHY7 and PPE7 revealed the highest TFC values (113.29 and 108.2 mg rutin equivalent (RE)/g extract, respectively). All the obtained results were significantly higher than the values available in the literature data reported for various extracts from the aerial parts of P. erecta, P. fruticosa, P. nepalensis, P. pensylvanica, and P. thuringiaca [13,14,15]. Notably, the selection of the solvent in the extraction process is a crucial factor in the explanation of those differences. An aqueous acetone solvent extracts much fewer non-phenol compounds, such as carbohydrates, than methanol and water, which results in higher TPC and TFC values [16]. Moreover, aqueous acetone was reported as an excellent solvent for extracting higher molecular weight flavonoids and proanthocyanidins [17]. The aforementioned solvent prevents the decomposition of hydrolysable tannins during the extraction process, leading to a higher tannin content in the obtained extracts [18]. The identification of the secondary metabolite composition of the aqueous acetone extracts of selected Potentilla species using LC–HRMS (liquid chromatography–high-resolution mass spectrometry) analysis demonstrated the presence of 198 compounds. Among them, three groups of phenolic compounds were dominant in the analysed extracts: tannins, flavonoids, and phenolic acids. Monomeric and dimeric ellagitannins, such as agrimoniin, sanguinis and pedunculagin, are important chemophenetic markers in the Rosaceae family, especially in the Potentilla, Rubus, and Fragaria genera [19]. The chromatographic analysis reported herein led to the identification of a series of hydrolysable tannins that are represented by ellagitannin derivatives, such as laevigatin isomers (84, 109, 114, 124, and 128), laevigatin E isomers (37 and 40), agrimoniin (162) and its structural isomer (151), agrimonic acid A or B (102), galloyl-HHDP-glucose (16, 21, 43, and 48), digalloyl-HHDP-glucose (33 and 60) and trigalloyl-HHDP-glucose (131 and 133), galloyl-bis-HHDP-glucose (108, 118, and 144), ellagic acid (135) and its O-pentosides (132 and 163), O-hexosides (73, 97, and 101), and uronic acid (82, 95, and 130) derivatives. The analysis indicated that the one of the most abundant phytochemicals in all the extracts, except PAL7r, was agrimoniin. Agrimoniin has been frequently described as the major phenolic compound in several Potentilla species, such as P. argentea, P. anserina, P. grandiflora P. kleiniana P. norvegica, P. recta, and P. rupestris [10,20,21,22]. Other present ellagitannins, namely, leavigatins and agrimonic acid, are formed from the partial hydrolysis of agrimoniin (dehydrodigalloyl-di-(bis-HHDP-glucose)) [23]. Furthermore, few degradation products of hydrolysable tannins degradation, such as ellagic acid (135), brevifolincarboxylic acid (46) and its structural isomer (50), and brevifolin (83), were found. Gallotannins were present in a few extracts, which showed the presence of di-, tri-, tetra-, and pentagalloylglucose isomers (35, 36, 80, 86, 103, 137, and 168). However, the analysis revealed the absence of hydrolysable tannins in PAL7r. These findings are in agreement with the previous study, which demonstrated the absence of these metabolites in the aerial parts of P. alba [9]. Moreover, the analysis revealed the presence of condensed tannins, especially in PAL7r, such as catechin (28), epicatechin (61), and their glucosides (11, 22, 23, 41, and 106), as well as products of their polymerisation, such as A-type procyanidins (24, 54, 71, 90, 96, and 110) and dimeric (66, 88, and 107), trimeric (7, 42, 45, and 64) and tetrameric (56 and 93) B-procyanidins, including procyanidin B1 (25), procyanidin B2 (47), procyanidin B3 (27), procyanidin C1 (94), and procyanidin C2 (34). Based on the chromatographic profiles, a number of flavonoids were detected and characterised, including apigenin (92, 119, 161, 166, 184, and 185) as well as isorhamnetin (87, 91, 98, 100, 125, 150, 158, 167, 169, 171–173, 179, 182, 187, 188, 191, 194, and 196), naringenin (180), kaempferol (62, 67, 72, 78, 81, 104, 113, 117, 140, 142, 143, 147, 148, 153, 155, 157, 159, 160, 164, 165, 170, 175, 176, 178, 190, 192, and 193), quercetin (39, 44, 51, 53, 55, 58, 59, 63, 74, 79, 85, 99, 105, 111, 116, 120–122, 126, 127, 129, 134, 136, 138, 139, 141, 145, 149, 152, 154, 156, 177, and 186), acacetin (183), and tricin (189 and 195) derivatives. From a chemophenetic perspective, a few of them may be useful as chemical markers of the Potentilla genus, such as both isomers of tiliroside (190), astragalin (kaempferol 3-O-glucoside) (155), isorhamnetin 3-O-glucoside (169), kaempferol 3-O-glucuronide (157), avicularin (quercetin 3-O-arabinoside) (149), hyperoside (quercetin 3-O-galactoside) (139), isoquercitrin (quercetin 3-O-glucoside) (136), and rutin (quercetin 3-O-rutinoside) (138), which were previously reported to be present in at least one of the Potentilla species investigated to date [5,6,10]. The analysis also revealed the presence of phenolic acids, such as gallic acid (1), caffeic acid (29) and its derivatives (5, 9, 10, 15, 22, 65, and 75), coumaric acid (12, 17, 57, and 197), dihydroxybenzoic acid (13), and syringic acid (89) derivatives. The detailed chromatographic data of the analysed samples are shown in Table 2 and in Supplementary Figures S1–S12. To summarize, the number of compounds shared by all the analysed Potentilla species may typify their chemical profile as homogeneous. In the first step, the extract’s influence on both human colon epithelial cell line CCD841 CoN as well as human colon adenocarcinoma cell line LS180 was examined using an MTT assay. Studies were conducted after 48 h of the cells being exposed to either a culture medium (control) or extracts (25–250 µg/mL). As presented in Figure 1 and Table 3, all the investigated extracts inhibited the metabolic activity of both normal and cancer cells, and the observed effect was dose-dependent. The most significant anticancer effect was presented by extracts PAL7r and PFR7, which, at the highest tested concentration, deceased LS180 cells’ proliferation by 91.3% (IC50 PAL7r LS180 = 82 µg/mL) and 94.8% (IC50 PFR7 LS180 = 89 µg/mL), respectively. On the contrary, the weakest influence on the metabolic activity of colon cancer cells was noted after treatment with PTH7 and PRI7, which, at a concentration of 250 µg/mL, inhibited cell viability by 58.7% (IC50 PTH7 LS180 = 225 µg/mL) and 57.9% (IC50 PRI7 LS180 = 213 µg/mL), respectively. The strongest reduction (by 36.7%) of the viability of colon epithelial cells was caused by both PME7 and PHY7 (IC50 PME7 CCD841 CoN = 380 µg/mL; IC50 PHY7 CCD841 CoN = 489 µg/mL), while the weakest effect, as reflected by the IC50 value, was shown by PRI7 (IC50 PRI7 CCD841 CoN = 2402 µg/mL). Used as a positive control for the experiment, 5-fluorouracil (5-FU) at a concentration of 25 µM decreased the metabolic activity of CCD841 CoN and LS180 by 22.2% and 46.2%, respectively. All the investigated extracts at the highest tested concentrations revealed a stronger anticancer effect than 5-FU. Seven of twelve extracts inhibited LS180 cells’ viability better than 5-FU, when used at lower concentrations; the beneficial effect of PER7r, PHY7, PME7, and PPE7 was observed at concentrations of 150 and 250 µg/mL, while the beneficial effect of PAL7r, PFR7, and PNE7 was observed at concentrations from 100 to 250 µg/mL. In the case of CCD841 CoN cells, only 3 out of 12 of the investigated extracts inhibited the metabolic activity of colon epithelial cells stronger than 25 µM 5-FU: PAL7r (250 µg/mL); PHY7 (250 µg/mL); PME7 (150 and 250 µg/mL). The obtained results for the MTT assay may be strongly associated with high TPrC, especially in the PAL7r, PER7r, and FFR7 extracts. On several occasions, proanthocyanidins were reported to have a strong influence on colon cancer cell viability. Especially oligomeric proanthocyanidins from grape seeds (Vitis vinifera L., Vitaceae), which induce the apoptotic cell death of Caco-2 (human colorectal adenocarcinoma) cells manifested by nuclear condensation, caspase-3 and PARP cleavage, and formation of apoptotic bodies [31]. Additionally, proanthocyanidins from hops (Humulus lupulus L., Cannabaceae) increased the intracellular formation of reactive oxidative species, which was manifested by the augmented accumulation of protein carbonyls and induced cytoskeletal disorganisation of human colon cancer cell line HT-29 [32]. However, in a comparison with a previous study, all extracts exerted a weaker effect on cancer cell viability than extracts obtained from five out of six tested aqueous acetone extracts, namely, P. argentea, P. grandiflora, P. norvegica, P. recta, and P. rupestris [10]. This difference may be associated with the lower TPC and TTC obtained herein. Ellagitannins display great chemopreventive and chemotherapeutic activities. Among them, agrimoniin, the main ellagitannin present in all extracts, except PAL7r, was shown to have prominent anticancer, antioxidant, and anti-inflammatory activities [33]. It is widely recognised that there is a strict correlation between chronic inflammation and colorectal cancerogenesis [34]. Preclinical and clinical studies showed that non-steroidal and anti-inflammatory drugs are effective in preventing the formation of colorectal tumours; however, there are limitations due to severe and fatal side effects, such as gastric bleeding, ulcers, and renal toxicity [35]. Phytochemicals have fewer side effects compared with synthetic drugs, which is advantageous. A study conducted by Shi and co-authors revealed that the use of lyophilised strawberries (Fragaria x ananasa L., Rosaceae) containing agrimoniin as the second-most-abundant phytochemical, in an inflammation-induced colorectal carcinogenesis model, led to downregulating the mRNA expression of the proinflammatory mediators, such as COX-2, iNOS, IL-1β, IL-6, and TNF-α [36]. Moreover, in two consecutive studies, an agrimoniin-enriched fraction from the underground parts of P. erecta showed the dose-dependent inhibition of UVB-induced or TNF-α stimulated IL-6 and PGE2 production as well as reduced NFκB activation in HaCaT cells (human keratinocytes). Further, a UV erythema study in healthy volunteers revealed that an agrimoniin-enriched fraction significantly inhibited the UVB-induced inflammation process [37,38]. In the next step, the antiproliferative activity of Potentilla L. extracts was examined in both the normal and cancer cell lines using a BrdU assay (Figure 2 and Table 1). All the investigated extracts decreased DNA synthesis in the colon cancer cells in a dose-dependent manner. Nevertheless, a significant decrease in LS180 cells’ proliferation in response to the extract, for the whole range of tested concentrations, was only observed for PAL7r, PFR7, and PER7r, which, at concentrations of 100, 150, and 250 µg/mL, reduced DNA synthesis by around 80%. Furthermore, the aforementioned extracts were characterised by the lowest IC50 values (IC50 PAL7r LS180 = 52 µg/mL; IC50 PFR7 LS180 = 50 µg/mL; IC50 PER7r LS180 = 54 µg/mL). On the contrary, the lowest antiproliferative abilities were revealed by PAU7, which, even at the highest tested concentration, decreased DNA synthesis in LS180 cells by only 14.9% (IC50 PAU7 LS180 = 1495 µg/mL). The antiproliferative effect of the examined extracts was also observed in colon epithelial cells; however, the observed effect was weaker than in cancer cells. The only extract that did not affect divisions of CCD841 CoN was PER7, which was characterised by the highest IC50 value of 3705 µg/mL. On the contrary, the most significant changes in normal cells were observed in response to PAL7r, PFR7, and PER7r, which, at the highest tested concentration, reduced the proliferation of epithelial cells by 36.1%, 38.3%, and 43.9%, respectively (IC50 PAL7r CCD841 CoN = 412 µg/mL; IC50 PER7 CCD841 CoN = 282µg/mL; IC50 PER7r CCD841 CoN = 337 µg/mL). As presented in Figure 2, 25 µM 5-fluorouracil (5-FU) decreased DNA synthesis in the investigated cell lines to 90.7% (CCD841 CoN) and 29.7% (LS180). The antiproliferative effect of 5-FU recorded in colon cancer cells was significantly stronger than the changes induced by most of the examined extracts (9 of 12); however, PAL7r, PFR7, and PER7r, in concentrations from 100 to 250 µg/mL, decreased DNA synthesis more than 5-FU. On the contrary, data collected from colon epithelial cells revealed that most of the investigated extracts (PAL7r, PER7r, PFR7, PHY7, PME7, PNE7, PPE7, PPU7, PRI7, and PTH7) at higher concentrations inhibited DNA synthesis stronger than 25 µM 5-FU, while the antiproliferative effect of PFR7 for the whole range of tested concentrations was higher than the changes induced by analysed cytostatic. However, the presented results correspond with data from our previous study, showing that tested Potentilla species possess similar anticancer potential; moreover, for the PAL7r, PER7r, and PFR7 extracts, the results from a BrdU assay were significantly higher than those for all other tested samples [10]. The observed effect may be attributed to the high TPrC values. Kresty and co-authors found that a cranberry (Vaccinium macrocarpon Aiton, Ericaceae) proanthocyanidin-rich fraction significantly inhibited the viability and proliferation of human oesophageal adenocarcinoma SEG-1 cells. The mechanism involved cell cycle arrest in the G1 phase as well as a significant apoptosis induction [39]. Notably, the antiproliferative effect of other extracts may be connected with the presence of ellagitannins and the main product of their decomposition, namely, ellagic acid. The anticancer mechanism of ellagic acid is multidirectional. A study conducted on human colorectal carcinoma HCT-116 and the Caco-2 cell line revealed that ellagic acid induced cell cycle arrest in the G phase, reduced proliferating cell nuclear antigen (PCNA) expression and mitotic activity, and induced apoptosis via increasing the expression of caspase-8 and Bax [40]. Additionally, a further study conducted on HCT-116 cells revealed the involvement of ellagic acid in the decreased gene expression of signalling pathways’ proteins such as mitogen-activated protein (MAPK), p53, PI3K-Akt, and TGF-β [41]. Recently, Han and co-authors found that tiliroside acted as an inhibitor of carbonic anhydrase XII (CAXII), a membrane enzyme that produces a favourable intracellular pH and sustains optimum P-glycoprotein (Pgp) efflux activity in cancer cells. Moreover, tiliroside downregulated E2F1 and E2F3 expression and promoted caspase-3 activity [39]. In addition, the meta-analysis revealed that a high intake of flavonoids, such as quercetin and kaempferol derivatives, in the diet may decrease the risk of colon cancer [42]. Extract cytotoxicity was also examined in both normal and cancer colon cell lines using an LDH (lactate dehydrogenase)-based assay (Figure 3). Most of the examined extracts were not cytotoxic against human colon epithelial cells; however, PME7 was, for the whole range of tested concentrations, while PAL7r, PER7, and PHY7, in concentrations from 100 to 250 µg/mL, damaged the membranes of epithelial cells. The indicated extracts at the highest tested concentration increased the LDH level by an average of 11%. Studies conducted on colon cancer cells showed the cytotoxic effect of all the examined extracts for the whole range of tested concentrations. The strongest damage of cancer cell membranes was caused by PAL7r, which in concentrations from 25 to 250 µg/mL increased the LDH level by 145.7% and 479.0% respectively. The weakest cytotoxic effect was noted in colon cancer cells treated with PRI7, PPU7, and PTH7, which, at the highest tested concentration, caused an increase in the LDH level by 245.1%, 254.7%, and 256.0%, respectively. An LDH assay showed that 5-FU in a concentration of 25 µM was not cytotoxic against colon epithelial, while LDH releases were increased in colon cancer cells by 13.4%. All the investigated extracts damaged the colon cancer cell membranes significantly more than 5-FU. For CCD841 CoN cells, significant differences in the LDH concentration between 25 µM 5-FU and the examined extracts were observed in the case of four extracts (PME7, PAL7r, PER7, and PHY7), for which the cytotoxic impact on colon epithelial cells was reported above. The results of the LDH assay are presumably directly associated with the TTC in the investigated samples. Tannins, due to their specific chemical structure, are known to affect the physical properties of membranes, initiate membrane protein aggregation, increase bilayer adhesion, and regulate cell metabolism [43,44]. The most abundant hydrolysable tannin present in most extracts, agrimoniin, induces the intrinsic pathway of apoptosis, directly influencing the permeability of the mitochondrial membrane via the activation of the mitochondrial permeability transition pore (MPTP) [45]. However, further in vivo studies are required to evaluate the exact mechanism of action. The bioavailability of large ellagitannins is generally low. Therefore, the method of application is limited to topical application. The gut microbiota metabolise ellagitannins and ellagic acid to produce a series of bioavailable metabolites, known as urolithins. Urolithins possess a series of biological activities, such as anti-inflammatory, antioxidant, anticancer, and immunomodulatory activities. The chemopreventive effects of urolithins were extensively studied in several models, including prostate and colorectal cancer models. Urolithins were shown to inhibit colon cancer cell growth in a dose-dependent manner, alter the expression of the genes and proteins modulating the cell cycle, and induce apoptosis [46]. Notably, a clinical study on the aerial parts of P. anserina and the rhizomes of P. erecta confirmed the formation of urolithins in ex vivo conditions [47]. The reference phytochemicals, including isorhamnetin 3-O-glucoside, kaempferol 3-O-glucuronide, and quercetin 3-O-glucuronide were obtained from Extrasynthese (Genay, France). Gallic acid, catechin, and epicatechin were obtained from Carl Roth (Karlsruhe, Germany). Procyanidin B1, procyanidin B2, procyanidin B3, and procyanidin C1 were purchased from Cayman Chemical (Ann Arbor, MI, USA), while agrimoniin, apigenin, 3-O-caffeoylquinic acid, ellagic acid, astragalin (kaempferol 3-O-glucoside), pedunculagin, avicularin (quercetin 3-O-arabinoside), hyperoside (quercetin 3-O-galactoside), isoquercitrin (quercetin 3-O-glucoside), and tiliroside (purity > 96%) were previously isolated in the Department of Pharmacognosy of Medical University of Białystok (Białystok, Poland) [22,48,49,50,51]. All other analytical grade chemicals used in the study were obtained from Sigma-Aldrich (St. Louis, MO, USA). To obtain ultra-pure water, a POLWATER DL3-100 Labopol (Kraków, Poland) system was used. Investigated extracts (100 mg/mL) and 5-fluorouracil (50 mM) were dissolved in dimethyl sulfoxide (DMSO) to prepare stock solutions. Working solutions were prepared by dissolving stock solutions in a culture medium. The final concentration of DMSO in all working solutions used in the studies was 0.25%. Plants used to obtain material for investigations come from the Medicinal Plant Garden at the Medical University of Białystok (Białystok, Poland) and were collected in June-August 2017–2020. Plants were carefully identified by one of the authors (M.T.), and individual voucher specimens were deposed at the Herbarium of the Department of Pharmacognosy, Medical University of Białystok (Białystok, Poland). Plant material was dried at room temperature in the shade and air temperature and subsequently finely ground with an electric grinder. Accurately weighed 2 g of each powdered dry plant material were separately extracted using an ultrasonic bath (Sonic-5, Polsonic, Warszawa, Poland) with 70% acetone at a controlled temperature (40 ± 2 °C) for 45 min in a 1:75 (w:v) solvent ratio to obtain raw extracts. Subsequently. extracts were evaporated to dryness, diluted with water (50 mL). and successively portioned between chloroform (10 × 20 mL). Afterwards. purified extracts were freeze-dried. The list of obtained aqueous acetone extracts from selected Potentilla species detailing plant species, voucher specimen, the parts used and extraction yields are presented in Table 4. The content of total phenolic compounds was measured using standard the Folin-Ciocalteu colourimetric method, with slight modification according to [29]. The content total tannin determination was carried out using the hide powder-binding method and Folin–Ciocalteu assay reported in the corresponding monograph in the European Pharmacopoeia 10th ed. [52]. The absorbance was measured at 760 nm using a EPOCH2 BioTech (Winooski, VT, USA) microplate reader. The obtained results for were expressed as milligrams of gallic acid equivalents per gram of extract (mg GAE/g extract). The determination was repeated at least in triplicate for each sample solution. Total proanthocyanidin content was analysed using the procedure based on the previously published protocol [53]. The analysis was carried out by mixing 50 µL of the sample solution (1 mg/mL) dissolved in methanol and 250 µL of 0.1% methanolic solution of 4-dimethylamino-cinnamaldehyde (DMCA) reagent in 6M HCl. After incubation of the mixture at room temperature for 15 min, the absorbance was measured at 635 nm, and results were expressed as milligrams of catechin equivalents per gram of extract (mg CE/g extract). The determination was repeated at least five times for each sample solution. Total phenolic acid content was estimated with the procedure using Arnov’s reagent (1 g of sodium molybdate and 1 g of sodium nitrate dissolved in 10 mL of distilled water) [54]. Each time the tested solution (30 µL) was mixed with 180 μL of water, 30 μL of 0.5 M HCl, 30 μL of Arnov’s reagent, and 30 μL of 1 M NaOH were sequentially added to the microplate well, and then it was incubated for 10 min at ambient temperature. Afterwards, the absorbance was measured at 490 nm, and results were expressed as milligrams of caffeic acid equivalents per gram of extract (mg CAE/g extract). The determination was repeated at least three times for each sample solution. Total flavonoid content was estimated by the previously described colourimetric method [29]. Each aliquot (100 µL) was mixed with aluminum chloride (AlCl3) solution (100 µL, 2% w:v). After incubation of the mixture at room temperature for 10 min, the absorbance was measured at 415 nm, and results were expressed as milligrams of rutin equivalents per gram of extract (mg RE/g extract). The determination was repeated at least three times for each sample solution. The separation and qualitative evaluation of each extract were conducted using a Kinetex XB-C18 column (150 × 2.1 mm, 1.7 μm, Phenomenex, Torrance, CA, USA) and Agilent 1260 Infinity LC chromatography system coupled to a photo-diode array (PDA) detector and 6230 time-of-flight (TOF) mass spectrometer (Santa Clara, CA, USA). A detailed description of the execution of the above-mentioned assay was presented in the previous study [10]. For the cell culture study, human colon adenocarcinoma cell line LS180 and human colonic epithelial cell line CCD841 CoN were purchased from the European Collection of Cell Cultures (ECACC, Centre for Applied Microbiology and Research, Salisbury, UK) and American Type Culture Collection (ATCC, Menassas, VA, USA), respectively. LS180 cells and CCD841 CoN cells were maintained in Dulbecco′s Modified Eagle′s Medium/Nutrient Mixture F-12 Ham and Dulbecco’s Modified Eagle’s Medium (DMEM), respectively. Then, 10% fetal bovine serum (FBS), penicillin (100 U/mL), and streptomycin (100 g/mL) were added to the cell culture media. Cells were incubated in a humidified atmosphere of 95% air and 5% CO2 at 37 °C. Examination of the anticancer potential of extracts was conducted simultaneously on both cancer (LS180) and normal (CCD841 CoN) colon cells. Cells at a density of 5 × 104 cells/mL were plated on 96-well plates. The next day, the cell growth medium was exchanged for fresh medium supplemented with investigated extracts or 25 μM 5-fluorouracil (5-FU). After 48 h of treatment, compound impacts on cell membrane integrity, metabolic activity, and DNA syntexis were examined using LDH, MTT, and BrdU assays, respectively. The description of the execution of indicated tests was previously presented [10]. The data were presented as the mean ± standard error of mean (SEM). Statistical analyses were carried out using one-way ANOVA with Tukey’s post hoc test and column statistics. Significance was accepted at p < 0.05. The IC50 value (concentration causing proliferation inhibition by 50% compared to control) was calculated according to the method of Litchfield and Wilcoxon [55] using GraphPad Prism 5. In conclusion, the presented study reports, for the first time, an analysis of the LC–HRMS profile of aqueous acetone extracts from rare Potentilla species. The analysis revealed a series of marker metabolites such as agrimoniin, pedunculagin, dimeric and trimeric B-type procyanidins, tiliroside, astragalin (kaempferol 3-O-glucoside), hyperoside (quercetin 3-O-galactoside, ellagic acid, and tri-coumaroyl spermidine. The performed studies revealed that all of the investigated acetone extracts obtained from rare Potentilla species decreased the viability and proliferation of human colon adenocarcinoma LS180 cells. Nevertheless, most of the investigated extracts also decreased metabolic activity and DNA synthesis in human colon epithelial CCD841 CoN cells, and 4 out of 12 of the tested extracts (PAL7r, PER7, PHY7, and PME7) showed cytotoxic effects against normal epithelial cells. Despite the fact that the investigated extracts affected both normal and cancer colon cells, the LS180 cells were more sensitive to tested extracts. Considering the data obtained from all the performed studies, the 2 of the 12 investigated extracts (PFR7 and PER7r) revealed the greatest chemopreventive potential, as manifested by the effective elimination of colon cancer cells, which caused both damage to their cell membranes and inhibition of their proliferation and metabolic activity, with a simultaneous lack of any cytotoxic effect on normal colon epithelial cells and a significantly weaker effect on their metabolism and DNA synthesis compared to cancer cells. The previous [10] and currently obtained results indicated that some acetone extracts from Potentilla species have anticancer potential, however, additional animal and clinical studies, especially including the influence of intestinal flora are required to verify discovered beneficial properties of investigated extracts. Nevertheless, discovered selectivity of the anticancer effects of tested extracts encourages further studies to develop a new efficient and safe therapeutic strategy for people who have been threatened by or suffered from colon cancer.

PMC10002593

Pauline Arbogast,Guillaume Gauchotte,Romane Mougel,Olivier Morel,Ahmed Ziyyat,Mikaël Agopiantz

Neurotensin and Its Involvement in Reproductive Functions: An Exhaustive Review of the Literature

27-02-2023

neurotensin,reproduction,ovulation,acrosome reaction,fertilization,spermatozoa

Neurotensin (NTS) is a peptide discovered in 1973, which has been studied in many fields and mainly in oncology for its action in tumor growth and proliferation. In this review of the literature, we wanted to focus on its involvement in reproductive functions. NTS participates in an autocrine manner in the mechanisms of ovulation via NTS receptor 3 (NTSR3), present in granulosa cells. Spermatozoa express only its receptors, whereas in the female reproductive system (endometrial and tube epithelia and granulosa cells), we find both NTS secretion and the expression of its receptors. It consistently enhances the acrosome reaction of spermatozoa in mammals in a paracrine manner via its interaction with NTSR1 and NTSR2. Furthermore, previous results on embryonic quality and development are discordant. NTS appears to be involved in the key stages of fertilization and could improve the results of in vitro fertilization, especially through its effect on the acrosomal reaction.

Neurotensin and Its Involvement in Reproductive Functions: An Exhaustive Review of the Literature Neurotensin (NTS) is a peptide discovered in 1973, which has been studied in many fields and mainly in oncology for its action in tumor growth and proliferation. In this review of the literature, we wanted to focus on its involvement in reproductive functions. NTS participates in an autocrine manner in the mechanisms of ovulation via NTS receptor 3 (NTSR3), present in granulosa cells. Spermatozoa express only its receptors, whereas in the female reproductive system (endometrial and tube epithelia and granulosa cells), we find both NTS secretion and the expression of its receptors. It consistently enhances the acrosome reaction of spermatozoa in mammals in a paracrine manner via its interaction with NTSR1 and NTSR2. Furthermore, previous results on embryonic quality and development are discordant. NTS appears to be involved in the key stages of fertilization and could improve the results of in vitro fertilization, especially through its effect on the acrosomal reaction. Neurotensin (NTS) is a 13 amino acid peptide initially described in 1973 in the bovine hypothalamus [1]. It is synthesized from a 170 amino acid longform propeptide. The human gene for neurotensin is located on chromosome 12q2.1 [2], and its expression is mainly described in the neurons and N endocrine cells in the small intestine [3]. NTS has a physiological role and is implicated in several functions, mainly in digestive and central nervous systems. In the digestive system, NTS acts on intestinal motility [4] and facilitates fatty acid translocation [5]. In the central nervous system, NTS modulates dopaminergic, serotoninergic, GABAergic, glutamatergic, and cholinergic systems [6]. It plays a major role in decreasing body temperature and sleeping [7]. NTS also presents antinociceptive effects mediated by NTS receptor 2 (NTSR2) [8,9]. NTS effects are mediated by three receptors (NTSR): NTSR1 and NTSR2 are G-protein-coupled receptors with seven transmembrane-spanning domains, and NTSR3 or sortilin (SORT1) is a Vps10p receptor family member, which is characterized by the presence of an extracellular region containing a cysteine-rich domain and only one transmembrane domain [10]. NTSR1 and NTSR3 have a strong affinity with NTS [11]. NTSR1 activation stimulates intracellular second messengers (inositol triphosphate (IP3) and diacylglycerol (DAG)). IP3 increases the concentration of intracellular calcium, and DAG induces the stimulation of protein kinase C (PKC) [12,13], which activates mitogen-activated protein kinases (MAPKs)/Erk pathways [14,15]. The NTS/NTSR1 complex is involved in the progression of many cancers, such as colic adenocarcinoma [16], small-cell lung cancer [17], medullary thyroid cancer [18], hepatocellular carcinoma [19], pancreatic carcinoma [20], breast cancer [21], and non-small-cell lung cancer [22], via cell proliferation, survival, migration, invasion, and neoangiogenesis [23]. Our team has shown that NTSR1 expression levels are correlated with the grade and prognosis of cancer in several tumors, especially in endometrial and ovarian adenocarcinomas, with its cytoplasmic localization increasing with the grade of the tumor [24,25]. Several publications have reported the expressions of NTS and its receptors in reproductive tissues, related to their roles in reproduction functions, particularly in ovulation, sperm capacitation, fertilization, and embryonic development. It has recently been highlighted that NTS receptors are located on male gametes and that NTS is expressed in the uterus. NTS could play a role in reproduction issues by participating in follicular rupture during ovulation [26,27] and improving sperm capacitation and the acrosome reaction [28,29]. Moreover, after fertilization, NTS would improve the rate of embryonic cleavage and embryo quality in cattle and mice [30,31]. The aim of this review was to synthetize the worldwide knowledge about NTS and its potential role in reproductive functions in order to achieve the following: Synthesize knowledge regarding the expression and function of the neurotensinergic complex in the different models in which it has been studied; Propose an integrative mechanistic interpretation; Open perspectives for future research. On the PubMed database, the keywords “fertility”, “fertilization”, “ovulation”, “oocyte”, “embryo”, “endometrium”, “spermatozoa”, “oviduct”, and “reproduction” associated with NTS were tested. After abstract studies, we considered 15 publications of interest [24,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42]. NTS was described in the female tract in the oviduct epithelium using Western Blot, immunohistochemistry, and RT-PCR [28,29,33,39]. Moreover, NTS mRNA levels in female mice treated with equine chorionic gonadotropin (eCG) and human chorionic gonadotropin (hCG) were significantly increased in the ampulla compared to the endometrium and the tubal isthmus epithelial cells [29]. Using a microarray analysis, the expression of NTS mRNA is more than 20 times higher in the isthmus oviduct and 32 times higher in the ampulla during the follicular phase of the cycle than during the luteal phase [43]. NTS expression has also been described in the endometrium in rats [37], mice [29], and cattle [32,38]. Sakumoto et al. studied the expressions of several proteins using immunohistochemistry, including NTS, in the endometrium of cows in summer and fall [32]. The NTS mRNA expression in the endometrium of the cows was more abundant in summer than in autumn. In humans, our team demonstrated using immunohistochemistry and RT-PCR that NTS expression was very weak in the normal endometrium, regardless of the menstrual cycle. NTSR1 was not expressed in the human normal endometrium [24]. Some authors have been interested in the presence of NTS/NTSR1 in the tissues of the human myometrium and leiomyoma and, more particularly, in women having hormonal treatment for in vitro fertilization (IVF) [34]. Rodriguez et al. used in situ hybridization (ISH) to detect NTS mRNA. NTS was present in the connective tissue cells of the normal myometrium. Considering leiomyoma, NTS immunoreactivity has also been detected in smooth muscle cells (up to 14.4%), and its rate is interestingly increased (33.8%) when women have been previously treated for IVF. To test the expression of NTSR1, this team used RT-PCR and detected its expression in connective tissue cells. No significant difference between normal and tumor tissue was found, but NTSR1 levels were also increased in leiomyoma smooth muscle cells after controlled ovarian hyperstimulation (COH). They also showed, using immunohistochemistry, that normal myometrium connective tissue cells only express NTS or NTSR1 but not both, unlike tumor tissue, in which they are both co-expressed [34,44]. Hiradate et al. showed using RT-PCR that NTS mRNA is also expressed in the ovulated cumulus cells of female mice. Its expression was massively increased after stimulation by gonadotropin and hCG, as observed in rats and macaques [26,27]. After in vitro maturation (IVM) with FSH and epidermal growth factor (EGF), the expression of NTS mRNA was higher in mature cumulus cells than in immature cumulus cells [29]. The addition of a MAPK inhibitor (U0126) to the medium completely blocked the release of NTS from cumulus cells, which could suggest that the secretion of NTS is regulated by these two factors via the activation of the MAPK pathway [29,45]. In cattle, it has been shown, using RT-PCR, that cumulus oophorus and corona radiata granulosa cells do not express NTSR1 and NTSR2; NTSR3 was not studied [30]. However, in women and female rats, Al Alem et al. found expressions of the three specific NTS receptors using RT-PCR and immunohistochemistry in granulosa cells, with a major predominance of NTSR3 compared to NTSR1 and 2, which were very weakly expressed [27]. The same result was found in female macaques using RT-PCR, immunohistochemistry, and Western Blot [26]. NTSR3 immunodetection was low to nondetectable in the granulosa cells of ovulatory follicles before hCG and 12 h after hCG administration. Granulosa cell staining for NTSR3 was present 24 h after hCG, with an apparent decline in NTSR3 immunodetection observed by 36 h after hCG. Granulosa cell NTSR3 mRNA showed a similar pattern, with low mRNA levels 0–12 h after hCG, peak NTSR3 mRNA levels 24 h after hCG, and a return to low NTSR3 mRNA levels 36 h after hCG [26]. In female rats, hCG injection did not change the expression levels of NTSR3 in granulosa cells [27]. In mice, NTSR3 expression gradually decreased after eCG injection, whereas NTSR1 expression remained unchanged, and NTSR2 expression was undetectable [42]. Hiradate et al. showed using Western Blot and immunohistochemistry the localization of NTSR1 in mice spermatozoa [29]. Using immunohistochemistry, they demonstrated that NTSR1 is more specifically localized in the neck of bull spermatozoa [28]. NTSR2 is present on their tail [30]. NTSR3 expression has not been studied in the sperm of cattle and mice. In humans, NTSR1 was detected in spermatozoa using Western Blot techniques [46]. In non-human primates, immunohistochemistry detected NTSR1 at several stages of sperm maturation, with the strongest staining in mature spermatozoa in the lumen of the seminiferous tubules [46]. Spermatozoa and their precursors did not express NTSR2, but it was found in the interstitial cells of the seminiferous tubes, unlike NTSR1. NTSR3 was detected using immunohistochemistry in the testis but not in male germ cells at any stage of maturation [46]. NTS secretion was not studied. All these data about the localization of NTS and NTS receptors are summarized in Figure 1 and Table 1. Several studies showed an increase in NTS after ovulation triggering. Wissing et al. showed in 2014 using a microarray analysis and RT-PCR that NTS mRNA expression increased 15,7-fold 36 h after hCG triggering in the granulosa cells (GCs) of women undergoing a long agonist IVF protocol [47]. In order to evaluate the role of NTS in ovulation, Al Alem et al. studied in woman and in female rats the expression levels of NTS in GCs after ovulation triggering via an hCG injection [27]. In woman, the dominant follicle was surgically excised prior to the LH surge (the preovulatory phase), or women were given 250 μg hCG to provoke ovulation triggering, and dominant follicles were collected 12–18 h after hCG (the early ovulatory phase), 18–34 h (the late ovulatory phase), and 44–70 h (the postovulatory phase). NTS mRNA was massively increased during the early and late ovulatory phases in GCs (15,000-fold) and theca cells (700-fold) [27]. Moreover, in cultured granulosa–lutein cells (GLCs) from IVF patients (after 6 or 7 days with the aim of regaining their responsiveness to hCG), NTS expression was induced 6h after hCG treatment [27]. In cultured rat GCs treated with or without 1 IU hCG for 4, 8, 12, or 24 h, the expression of NTS increased after 4 h of hCG treatment by approximately 7-fold and remained elevated after 8 h, after which NTS expression declined to basal levels by 24 h [27]. Likewise, Campbell et al. demonstrated in female macaques that all the granulosa cells of the ovulatory follicle secrete NTS in an increased manner after the LH surge (up to 30 times more). This secretion was not found in theca cells [26]. The NTS regulation of ovulatory mechanisms has been studied in vitro. In both women and rats, the addition of an EGF pathway inhibitor (AG1478) partially blocked the expression of NTS mRNA by granulosa cells after an injection of hCG [27]. In humans, the regulation of NTS expression was blocked by the addition of inhibitors of the PKA and PKC pathways, as well as IP3 kinase and MAPK, suggesting that its regulation occurs via these pathways [27]. In mice, Shrestha et al. confirmed the mediation of NTS expression by PKA, MAPK, and EGF receptor signaling pathways and found several novel genes regulated by NTS (Ell2, Rsad2, Vps37a, and Smtnl2) that could impact the ovulatory process [42]. In order to demonstrate the role of the NTS/NTSR3 complex in ovulation, Campbell et al. inhibited NTS in vivo [26]. Over the monitored cycles, an intrafollicular injection of a rabbit antibody against NTS (ImmunoStar, Hudson) or control IgG was performed during aseptic surgery. Immediately postoperatively, hCG was administered to initiate ovulatory events. They showed that three out of four “anti-NTS-injected” follicles did not break without oocyte release and had an aspect of hemorrhagic cyst, whereas all the control ones showed a clear rupture of the membrane (four/four) [26]. The follicles injected with either the control IgG or the NTS antibody showed evidence of structural luteinization. In addition, both treatments resulted in similar levels of serum progesterone after hCG administration [26]. Finally, NTS stimulated ovarian microvascular endothelial cell migration in a dose-dependent manner but did not alter proliferation [26]. The mechanisms leading to follicular rupture through the interaction of NTS with NTSR3 are not known. However, it has been described in other models, notably in cancer, that the interaction of NTS with NTSR3 activates signaling cascades through focal adhesion kinase (FAK), a key pathway leading to the weakening of cell–cell and cell–extracellular matrix adhesions, a series of events that could be responsible for migration and cancer metastasis. Finally, some future approaches targeting NTSR3 release through the inhibition of matrix metalloproteinases (MMPs) are suggested [48]. These mechanisms leading to cell disjunction could explain the rupture of the follicular membrane during ovulation, showing a major role of NTS/NTSR3 in the mechanisms of ovulation (Figure 2). Sperm viability was studied on human and non-human primates’ spermatozoa using eosin–nigrosin staining [46]. Treatment with 0.1, 1, or 10 μM NTS did not change sperm viability after 5 or 15 min of treatment compared to a control. Treatment for 30 min did result in a small but significant increase in sperm viability when compared to the control [46]. Campbell et al. analyzed the effects of adding NTS to a medium on sperm motility using a computer-assisted sperm analysis (CASA) system. Treatment with NTS did not modify the percentage of motile or progressive sperm or the percentage of sperm scored as rapid, slow, or static, but treatment with NTS at 1 and 10 μM did slightly but significantly increase the percentage of sperm scored as medium speed [46]. The same experiment has been used to study sperm motility in Japanese black cattle. No great effect on sperm motility has been shown, but a small number of sperm cells showed a hyperactivated motility pattern, suggesting that the effect of NTS on sperm motility is small, while cellular responses can be heterogeneous [28]. Capacitation is required to render spermatozoa competent to fertilize oocytes by passing through the female genital tract. It has been shown by Visconti et al. that the tyrosine phosphorylation of many proteins is induced during sperm capacitation [49]. Using a Western Blot analysis, Hiradate et al. showed that NTS can facilitate sperm capacitation in mice by enhancing this tyrosine phosphorylation in a dose-dependent manner in vitro [29]. Using the same technique, these results were confirmed in cattle, where total tyrosine phosphorylation was significantly increased upon the addition of NTS to the medium [28]. In human and non-human primates, Campbell et al. studied the NTS effects on the spermatozoa acrosomal reaction using immunofluorescence [46]. They demonstrated that NTS increased the acrosome reaction in a dose-dependent manner. Sperm incubated for 5 min with 0.1–10 μM NTS showed a dose-dependent decrease in complete acrosomes (p < 0.05), as well as a dose-dependent increase in absent acrosomes (p < 0.05), with no change in partial acrosomes, whereas sperm incubated for 15 min or 30 min showed a decrease in complete acrosomes (p < 0.05) and an increase in both partial and absent acrosomes (p < 0.05) [46]. They also proved that the action of NTS on the acrosome reaction in human spermatozoa was through its interaction with NTSR1. Indeed, the acrosome reaction appeared when NTS was added in the medium and was absent when they used SR48692, a specific antagonist of the NTS-NTSR1 complex [46]. Hiradate et al. had the same results using capacitated sperm from mice, which were incubated with increased concentrations of NTS to study its impact on the acrosome reaction. By using (Alexa Fluor) fluorescein staining, they showed that NTS treatment significantly accelerated the acrosome reaction in a dose-dependent manner [29]. When adding NTSR1 or NTSR2 antagonists (SR48692 and levocabastine, respectively) to the medium, the acrosomal reaction was completely blocked, suggesting the functional expression of not only NTSR1 but also of NTSR2 on sperm cells and the contribution of these receptors to the acrosome reaction [29]. Moreover, when 10 or 50 μM NTS was added, [Ca2+]i levels were increased just after the addition. The calcium intensity when 50 μM of NTS was added was higher than that when 10 μM was added, providing evidence for dose dependency in the [Ca2+]i response. However, the number of sperm cells that immediately increased their calcium levels were low at 50 μM (10 spermatozoa per 29 total examined) and 10 μM (3 spermatozoa per 50 total examined), whereas almost all the sperm experienced increased calcium mobilization when ionomycin was added (24 spermatozoa per 25 total examined) [29]. The same mechanisms appear to exist in bull semen. However, this reaction only concerns a few spermatozoa (only 10 of 29 observed), and this was observed with the highest dose used (50 µM of NTS) [28]. The potential role of NTS in spermatozoa is illustrated in Figure 3. Fertilization was specifically studied in only one publication. The NTS pretreatment of monkey sperm reduced the fertilization rate of monkey MII oocytes in vitro from 72% [46]. Moreover, oocytes fertilized with untreated sperm were generally healthy in appearance, with the development of two pronuclei, whereas oocytes fertilized with NTS-treated sperm did not typically form pronuclei [46]. Two publications studied the role of NTS in embryo development. The addition of NTS to an IVF medium significantly improved the levels of cleavage in the bovine embryo (36.5% to 50% of early cleavage and 49% to 64.4% of cleavage, p < 0.05) [30]. However, NTS had no impact on the number of blastocysts on days 7 and 8 of their experience. A greater number of cells in bovine blastocysts appears to be an advantage in improving the length of gestation and in fetal development in cattle [50]. To assess the quality of embryos, Umezu and his team counted embryo cells using an immunofluorescence technique to distinguish trophectodermal cells and inner mass cells. The blastocysts obtained with the addition of NTS to an IVF medium contained more cells in total, as well as in their inner mass cells, than the ones obtained from the IVF medium without NTS [30]. The effects of NTS were also studied in mice embryos before implantation. Hiradate et al. showed using RT-PCR that mice embryos expressed NTSR 1 and NTSR3 at the zygote stage. Their expressions decreased following the two-cell stage but remained detectable until the blastocyst stage [31]. Embryos were cultured in the presence of different concentrations of NTS for 96 h. There were no significant differences in the two-cell cleavage and four-cell rate between control and NTS treatment (1, 10, 100, and 1000 nM) groups, but a significant increase in blastocyst formation was observed in the 100 nM NTS group (60.6 ± 1.7% vs. 75.6 ± 3.4%, p = 0.03). In this treatment group, the hatching rate was 28.1 ± 3.8% versus that of the control group (13.0 ± 4.1%; p = 0.08) [31]. No data concerning oocyte quality related to embryo development were available. The role of NTS in reproduction was first discussed in 1981 by Clark et al., who analyzed the effect of NTS on blood flow in the uterus of non-pregnant sheep [40]. They compared its effect to that of prostacyclin, which is one of the most potent vasodilator prostaglandins. Their study consisted of injecting neuropeptides directly using a catheter into the main uterine artery. They showed no vasoactive activity for NTS. In 1987, Reinecke et al. showed using an isometric force displacement transducer that NTS enhanced autonomous oviduct contractility up to 55% in vitro and that their amplitude was 9-fold more than that without NTS addition [39]. The effects of NTS on endometrial receptivity appear to be very different depending on the species. In dairy goats, Zhang et al. analyzed the concentrations of NTS mRNA and other molecules in the receptive endometrium and the pre-receptive endometrium [38]. Using inhibitory RNA systems, it was demonstrated that NTS was expressed 195 times more in the receptive endometrium than in the pre-receptive endometrium. To go further in their experiments on the potential functions of NTS in the receptive endometrium, they synthesized siRNA-NTS and a pcDNA3.1 (+) NTS vector to explore the effects of NTS in endometrial epithelial cells (EECs). With this experiment, they showed that the overexpression of NTS increased the proliferation of EECs (p < 0.005), while siRNA-NTS induced the apoptosis of EECs (p < 0.005). In addition, NTS may play a role in endometrial receptivity by increasing the levels of certain biochemical markers of the receptive endometrium, such as leukemia inhibitory factor (LIF), COX2, and HOXA10, in EECs in vitro [38]. However, these results seem contradictory with the results of Sakumoto et al. in cows, where the expression of NTS in the endometrium was higher in the summer when fertility levels were lower [32]. Conflicting results also seem to be observed in rats, where an intra-uterine microinjection of NTS on day 4 or 5 of pregnancy decreases in a significant manner the number of fetuses and the weight and glycogen of the uterus, whereas performing the same experiment on days 8, 9, and 10 or on days 14, 15, and 16 has no consequences on fetuses’ viability [37]. The NTS physiological functions in mammalian reproduction are summarized in Table 2. Data regarding the important role of NTS and its receptors are now well-established. About 15 publications that we reviewed exhaustively show the action of NTS in ovulation and fertilization. The data concerning ovulation are still incomplete. The exponential presence of NTS in granulosa cells, especially after LH, hCG, or eCG triggering, and the co-expression of the specific receptor NTSR3 point us towards an autocrine and/or paracrine mode of action. The limited experimental data support a necessary but not sufficient role in causing ovulation. Mechanistic studies conducting the inhibition testing of signaling pathways confirm a direct action via the MAPK pathway. Likewise, the presence of NTS in the female genital tract, particularly after ovulation, indicates a physiological role. Data concerning the stages of fertilization, as well as the presence of NTSR1 and NTSR2 on the sperm membrane (while they are absent from the epithelial cells of the female tract outside the pathological situation), could suggest a major paracrine effect presiding over capacitation and the acrosomal reaction. However, fertilization rates with NTS-treated sperm were lower than those obtained with untreated sperm. It would be interesting to experiment to determine whether these fertilization rates remain low if NTS is added directly to the fertilization medium rather than to the sperm preparation, as the NTS-induced acrosomal reaction should be performed as close to the oocyte as possible. NTS could potentially play a role in IVF in cases of fertilization failure. The role of NTS in endometrial receptivity is not clear across species, yet implantation failures are a major cause of IVF failure, and its potential mechanisms are still poorly understood. The study of NTS expression at the endometrial level in women could be of interest in patients presenting repeated implantation failures after embryo transfer in this context.

PMC10002594

Shini Kanezawa,Mitsuhiko Moriyama,Tatsuo Kanda,Akiko Fukushima,Ryota Masuzaki,Reina Sasaki-Tanaka,Akiko Tsunemi,Takahiro Ueno,Noboru Fukuda,Hirofumi Kogure

Gut-Microbiota Dysbiosis in Stroke-Prone Spontaneously Hypertensive Rats with Diet-Induced Steatohepatitis

27-02-2023

high-fat- and high-cholesterol-containing diet,Firmicutes/Bacteroidetes ratio,gut-microbiota,metabolic-dysfunction-associated fatty-liver disease,nonalcoholic steatohepatitis,small-intestinal bacterial overgrowth,stroke,hypertension

Metabolic-dysfunction-associated fatty-liver disease (MAFLD) is the principal worldwide cause of liver disease. Individuals with nonalcoholic steatohepatitis (NASH) have a higher prevalence of small-intestinal bacterial overgrowth (SIBO). We examined gut-microbiota isolated from 12-week-old stroke-prone spontaneously hypertensive-5 rats (SHRSP5) fed on a normal diet (ND) or a high-fat- and high-cholesterol-containing diet (HFCD) and clarified the differences between their gut-microbiota. We observed that the Firmicute/Bacteroidetes (F/B) ratio in both the small intestines and the feces of the SHRSP5 rats fed HFCD increased compared to that of the SHRSP5 rats fed ND. Notably, the quantities of the 16S rRNA genes in small intestines of the SHRSP5 rats fed HFCD were significantly lower than those of the SHRSP5 rats fed ND. As in SIBO syndrome, the SHRSP5 rats fed HFCD presented with diarrhea and body-weight loss with abnormal types of bacteria in the small intestine, although the number of bacteria in the small intestine did not increase. The microbiota of the feces in the SHRSP5 rats fed HFCD was different from those in the SHRP5 rats fed ND. In conclusion, there is an association between MAFLD and gut-microbiota alteration. Gut-microbiota alteration may be a therapeutic target for MAFLD.

Gut-Microbiota Dysbiosis in Stroke-Prone Spontaneously Hypertensive Rats with Diet-Induced Steatohepatitis Metabolic-dysfunction-associated fatty-liver disease (MAFLD) is the principal worldwide cause of liver disease. Individuals with nonalcoholic steatohepatitis (NASH) have a higher prevalence of small-intestinal bacterial overgrowth (SIBO). We examined gut-microbiota isolated from 12-week-old stroke-prone spontaneously hypertensive-5 rats (SHRSP5) fed on a normal diet (ND) or a high-fat- and high-cholesterol-containing diet (HFCD) and clarified the differences between their gut-microbiota. We observed that the Firmicute/Bacteroidetes (F/B) ratio in both the small intestines and the feces of the SHRSP5 rats fed HFCD increased compared to that of the SHRSP5 rats fed ND. Notably, the quantities of the 16S rRNA genes in small intestines of the SHRSP5 rats fed HFCD were significantly lower than those of the SHRSP5 rats fed ND. As in SIBO syndrome, the SHRSP5 rats fed HFCD presented with diarrhea and body-weight loss with abnormal types of bacteria in the small intestine, although the number of bacteria in the small intestine did not increase. The microbiota of the feces in the SHRSP5 rats fed HFCD was different from those in the SHRP5 rats fed ND. In conclusion, there is an association between MAFLD and gut-microbiota alteration. Gut-microbiota alteration may be a therapeutic target for MAFLD. The number of patients with nonalcoholic fatty liver disease (NAFLD), including nonalcoholic steatohepatitis (NASH), has increased over the years [1]. As NASH causes cirrhosis and hepatocellular carcinoma (HCC), NASH is one of the important health issues worldwide. However, an unknown mechanism is also present in the pathogenesis of the development of NASH [2,3]. Fatty liver associated with metabolic dysfunction is common [4,5,6]. Metabolic-dysfunction-associated fatty-liver disease, “MAFLD,” may be a more appropriate overarching term [4,5]. Metabolic-dysfunction-associated fatty-liver disease is the principal worldwide cause of liver disease and affects nearly a quarter of the global population [4,5]. Diagnosis of MAFLD is based on the detection of liver steatosis together with the presence of at least one of three criteria that includes overweight or obesity, type 2 diabetes mellitus, or clinical evidence of metabolic dysfunction, such as an increased waist circumference and an abnormal lipid or glycemic profile [5]. Patients with hepatic steatosis and lean/normal weight is diagnosed as MAFLD in the presence of more than two metabolic risk abnormalities of the following criteria: an increased waist circumference, hypertension, an abnormal lipid or glycemic profile [5]. Patients with NAFLD are at a substantially higher risk of fatal and non-fatal cardiovascular events [6]. NAFLD and cardiovascular disease share multiple common conditions, such as obesity, diabetes, dyslipidemia and hypertension. These diseases may also share multiple common mechanisms, such as dietary habits, smoking, lack of exercise, gut-microbial dysbiosis, and genetics [6]. It has been reported that there is an association between NAFLD/NASH and gut-microbiota [5]. Individuals with NASH have a higher prevalence of small-intestinal bacterial overgrowth (SIBO) [7]. Intestinal mucosa-barrier malfunction may also play a role in NASH [8]. Individuals with NASH have a lower percentage of Bacteroidetes (Bacteroidetes total bacteria counts) than those with simple steatosis or healthy controls [9]. Thus, intestinal bacteria and gut-microbiota dysbiosis may play an important role in the development of NAFLD and NASH [3]. It has also been reported that gut-microbiota dysbiosis is linked to hypertension [10]. The gut-microbiota influence stroke pathogenesis and treatment outcomes [11,12]. Spontaneously hypertensive rats (SHR) and stroke-prone spontaneously hypertensive rats (SHRSP) are well-established parallel lines from outbred Wistar–Kyoto (WKY) rats [13,14]. We previously demonstrated a NASH model using arteriolipidosis-prone rats (ALR; SHRSP5), which are sublines obtained by the feeding of high-fat- and high-cholesterol-containing diets (HFCD) to SHRSP rats [15]. SHRSP5 rats fed HFCD possessed NASH, abnormal lipid, lean body, hypertension, and stroke [13,14,15]. In the present study, we examined the gut-microbiota isolated from stroke-prone spontaneously hypertensive-5 rats (SHRSP5) that were fed a normal diet (ND) or HFCD at 12 weeks of age and clarified the difference between their gut-microbiota. We observed differences between the microbiota of the feces in the SHRSP5 rats fed HFCD and those in the SHRP5 rats fed ND, as well as an increase in the Firmicutes/Bacteroidetes (F/B) ratio, which is a signature of gut dysbiosis, in the microbiota from the small intestine in the SHRSP5 rats fed HFCD. Our observation partially supports the concept of “MAFLD” from the point of view of gut-microbiota dysbiosis. Fecal-pellet DNA was isolated from the 12-week-old SHRSP5 rats fed ND or HFCD for 7 weeks [15]. As previously reported [15], in the HFCD group, pathological findings consistent with NASH were observed; however, in the ND group, only diffuse lipid droplets were seen in the hepatocytes at 12 weeks of age. As one rat died in the HFCD group, its fecal DNA could not be analyzed. At the same time, the DNA contents in the small intestines were also isolated from both groups of rats. First, we performed real-time PCR to measure the 16S ribosomal (r)RNA genes of the bacteria in the small intestines and feces in both groups of rats (Table 1). We noticed that the quantities of the 16S rRNA genes in the small intestines of the SHRSP5 rats fed HFCD were significantly lower than those of the SHRSP5 rats fed ND (p < 0.05). However, the DNA from all samples were sufficient for the subsequent analysis. Thus, HFCD reduced the 16S rRNA genes in the small intestines of the SHRSP5 rat, compared with ND. Interestingly, there may be an association between the reduction in bacteria and the fibrosis of the steatosis of the liver in the SHRSP5 rat fed HFCD. The effects of HFCD intake may be more important for the development of hepatic fibrosis in NASH than SIBO. Gut-microbiota dysbiosis is occasionally observed in patients with NASH [16]. The bacterial 16S rRNA gene has been used to define bacterial taxonomy and phylogeny. In order to understand the association between the gut-microbiota and the pathogenesis of NASH, we analyzed the V4–V5 region of the 16S rRNA from the bacteria in the small intestines and feces in the SHRSP5 rats fed ND or HFCD on the Illumina-MiSeq platform. The sequencing-read numbers are shown in Table 2. The sequence-read number ranged from 18,255 to 31,756. In the small intestines, the average sequence-read number of rats fed ND was similar to those of rats fed HFCD (28,819 ± 1944 vs. 29,567 ± 1956; no statistically significant difference). The coverage numbers were in a sequence around ~410 bp. These results indicate successful next-generation sequencing in the present study. Next, we performed weighted UniFrac analyses to calculate the distances between the microbiota populations from the small intestines and feces in the SHRSP5 rats fed with a ND or HFCD [17] (Figure 1A). The microbiota of the small intestines in the SHRSP5 rats fed on ND were more similar to those of the small intestines or feces in the SHRSP5 rats fed on a HFCD than to those of the feces in the SHRSP5 rats fed with ND. The clustering analysis in the ß-diversity analysis of the microbiota populations also supported these results (Figure 1B,C). A clear separation was observed in the principal-components analysis, clustering analysis, and ß-diversity analysis of the microbiota of the feces between the SHRSP5 rats fed on a HFCD and those fed on a ND (Figure 1A–C). Notably, the microbiota of the feces of the SHRSP5 rats fed an HFCD was different from those of the SHRSP5 rats fed an ND. An increase in the Firmicutes/Bacteroidetes (F/B) ratio, caused by an expansion of Firmicutes and/or a contraction of Bacteroidetes, is considered a signature of gut dysbiosis [10]. The F/B ratio in the small intestines of the SHRSP5 rats fed an HFCD increased compared to that of the SHRSP5 rats fed an ND (Figure 2A). The F/B ratio in the feces of the SHRSP5 rats fed with the HFCD tended to increase compared to that of the SHRSP5 rats fed with the ND (Figure 2B). The F/B ratio in the small intestines of the SHRSP5 rats fed with the HFCD was ~4.6-fold higher than that of the SHRSP5 rats fed the ND (Figure 2A). The F/B ratio in the feces of the SHRSP5 rats fed the HFCD tended to be ~1.7-fold higher than that of the SHRSP5 rats fed the ND (Figure 2B). In both the small intestines and the feces of SHRSP5 rats fed on an HFCD, the number of both Firmicutes and Bacteroidetes decreased. In the feces of the SHRSP5 rats fed the HFCD, the number of Proteobacteria increased (Figure 3). In the present study, among the Firmicutes, the Allobaculum decreased in the feces of the SHRSP5 rats fed with a HFCD. The Lactobacillus decreased and the Streptococcus increased in the small intestines of the SHRSP5 rats fed the HFCD. The Clostridium increased in both the small intestines and the feces of the SHRSP5 rats fed the HFCD. Of the Bacteroides, the Porphyromonadaceae decreased in feces of SHRSP5 rats fed the HFCD. Of the Proteobacteria, the Escerichia increased in both the small intestines and the feces of the SHRSP5 rats fed the HFCD. In the present study, we examined the gut-microbiota isolated from 12-week-old SHRSP5 rats fed a ND or a HFCD and clarified the differences between their gut-microbiota. We observed that the F/B ratio in both the small intestines and the feces of SHRSP5 rats fed the HFCD increased compared to that of the SHRSP5 rats fed the ND. Notably, the quantity of 16S rRNA genes in the small intestines of the SHRSP5 rats fed the HFCD were significantly lower than those of the SHRSP5 rats fed the ND. The microbiota of the feces of the SHRSP5 rats fed the HFCD was different from those of the SHRSP5 rats fed the ND. Li et al. reported the ability of Grifola frondosa heteropolysaccharide to ameliorate NAFLD in rats fed a high-fat diet (HFD) and significantly increase the proportions of Allobaculum [18]. Increases in Allobaculum can help infant mice resist the development of obesity, according to an investigation of the intestinal microbiota in mice [19]. These reports partially support our observation that Firmicutes and Allobaculum decreased in the feces of the SHRSP5 rats fed the HFCD. Panasevich et al. reported that soy protein is effective at preventing hepatic steatosis, and an analysis of fecal bacterial 16S rRNA revealed that soy-protein isolate intake elicited increases in Lactobacillus in obese Otsuka Long–Evans Tokushima fatty (OLETF) rats [20]. The rates of Streptococcus belonging to Bacilli were significantly increased in rats fed with a high-fat diet [21]. Compared with healthy subjects, NAFLD patients show an increase in the percentage of bacteria of pathogenic Streptococcus [22]. Previous studies [20,21,22] support our observations that the rates of Lactobacillus decreased and those of Streptococcus increased in the small intestines of the SHRSP5 rats fed the HFCD. Individuals with NAFLD might be at increased risk of the development of Clostridioides difficile colitis [23]. Clostridioides difficile colitis can trigger changes associated with the development of NAFLD [24]. In our study, the Clostridium also increased in both small intestines and the feces of the SHRSP5 rats fed the HFCD. High-fat diets result in quantitative alterations in the aerobes (Escherichia coli) in NASH rats [25]. Of the Proteobacteria, the Escherichia increased in both the small intestines and the feces of the SHRSP5 rats fed the HFCD. In 37.5% (12/32) of the patients with NAFLD, SIBO was present, with Escherichia coli as the predominant bacterium [26]. A previous study also demonstrated an increase in the Escherichia genus among gut-microbiota in the development and progression of NASH [27,28]. The presence of SIBO decreases small-intestinal movement in NASH rats [25]. A high-fat diet did not increase the anaerobics (Lactobacilli) [25]. Bacteroides species are also anaerobic. In the present study, of the Bacteroides, Porphyromonadaceae decreased in the feces of the SHRSP5 rats fed the HFCD. The presence of SIBO and endotoxemia can result in changes in toll-like receptor (TLR)-signaling gene expression, leading to the development of NAFLD [26]. The abundance of Bacteroidetes phylum may be increased, decreased, or unaltered in NASH patients [28]. Thus, SIBO plays a role in the development of NASH pathogenesis [7]. Patients with NASH and those with significant liver fibrosis on liver biopsy had a significantly higher incidence of SIBO than patients without NASH and those without significant liver fibrosis, respectively [29,30]. The onset of NASH in childhood is also a significant health problem [31]. There is an association between NAFLD and SIBO in obese children [32]. SIBO has an effect on the structural and functional characteristics of the liver, resulting in higher insulin and glucose levels, higher neutrophil-to-lymphocyte ratios, and a greater prevalence of NAFLD. A meta-analysis showed a possible association between SIBO and NAFLD in children [33]. The higher the grade of liver steatosis, the higher were the circulating lipopolysaccharide (LPS)-binding protein levels and SIBO rates seen in patients with morbid obesity and NAFLD [34]. The presence of SIBO may enhance intestinal permeability and endotoxemia in NASH patients [35]. Increased endotoxemia may enhance the innate immune response, including TLR-signaling pathways, as well as leading to inflammation and fat deposition in the liver. The symptoms related to SIBO are bloating, diarrhea, malabsorption, body-weight loss, and malnutrition [36]. SIBO is a heterogeneous syndrome characterized by an increased number and/or abnormal type of bacteria in the small intestine [36]. Notably, the SHRSP5 rats fed with the HFCD presented diarrhea and body-weight loss compared to those fed with the ND [15]; these symptoms were consistent with those of SIBO. In the SHRSP5 rats fed the HFCD, abnormal types of bacteria were observed in the small intestines, although the number of these bacteria did not increase (Table 1). We noticed that the HFCD is more important for the development of hepatic fibrosis in NASH than SIBO. High-fat diet (HFD)-dependent differences at the phylum, class, and genus levels appear to lead to dysbiosis, characterized by an increase in the F/B ratio, and Firmicutes was the dominant class in a male Sprague-Dawley (SD) rat (7 weeks old) fed HFD with steatohepatitis [37], supporting our observation (Figure 2B). An eight-week treatment of Gegen Qinlian decoction (GGQLD), a well-known traditional Chinese herbal medicine, improved these HFD-induced change [37]. Hugan Qingzhi tablet (HQT), which is a lipid-lowering and anti-inflammatory medicinal formula, has been used to prevent and treat NAFLD and reduced the abundance of the F/B ratio in HFD-fed rats [21]. Curcumin and metformin, which have a therapeutic effect against NAFLD, reduced the F/B ratio and reverted the composition of the HFD-disrupted gut-microbiota in male Sprague–Dawley rats fed HFD [38]. Gut-microbiota can play a role in the pathogenesis of NAFLD, as dysbiosis is associated with reduced bacterial diversity, altered F/B ratio, a relative abundance of alcohol-producing bacteria, or other specific genera [39]. Major risk factors of MAFLD are overweight/obesity, central obesity, type 2 diabetes mellitus, dyslipidemia, arterial hypertension, metabolic syndrome, insulin resistance, dietary factors, lifestyle, and sarcopenia [5]. It is known that gut-microbiota, hyperuricemia, hypothyroidism, sleep apnea syndrome, polycystic ovary syndrome, polycythemia, hypopituitarism, genetic and epigenetic factors, and family history of metabolic syndrome including high blood pressure are common and uncommon risk factors of MAFLD [5]. An association between hypertension and gut-microbiota alteration has been reported [10], as has an association between stroke and gut-microbiota alteration [11,12]. An association between obese and gut-microbiota alteration has also been reported [40], although fecal microbiota transplantation did not reduce body mass index. Evidence for the role of gut-microbiota in metabolic diseases including type 2 diabetes was provided [41]. Human and animal studies indicate the association between diets and hepatic steatosis [42,43]. The association between MAFLD and gut-microbiota alteration should now be clearer given the results of the present study. Dietary factors, such as high-calorie diets with rich saturated fats and cholesterol, soft drinks high in fructose, and highly processed foods, are known to influence the severity of NAFLD. Changing gut-microbiota also does so, at least in part [44]. In the present study, HFCD had an impact on changing gut-microbiota. We observed an association between NASH and gut-microbiota alteration in the SHRSP5 rats, which originated from the stroke-prone, spontaneously hypertensive rats (SHRSP) fed the HFCD. The recent concept of MAFLD highlights the association between fatty liver disease, hypertension, stroke, and other metabolic diseases. The results from the present study may partially support the association between MAFLD and gut-microbiota alteration. Gut-microbiota alteration may be a therapeutic target for MAFLD. The real interest is how and why the altered microbiota are related to the pathological phenotype. Studies of the associated mechanism should be performed. The 16S rRNA gene is present in multiple copies in the genomes of bacterial pathogens [45,46]. Therefore, amplicon-sequencing of the bacterium-specific 16S rRNA gene is a useful method for investigating a broad range of bacterial species. However, it is unclear whether the amplicon-sequencing-based detection of the 16S rRNA gene is useful for determining the causative pathogen. A major problem is that the 16S rRNA gene can be amplified not only for meaningful bacteria but also for meaningless bacteria, which is one of the limitations of this study. Another limitation of the present study is that the number of rats used was small. This was because the present study was an initial study; we will elucidate the mechanisms further in a future study. For example, further improvement of bioinformatics and their analysis, the use of the QIME2 software, which uses amplicon sequence variant (ASV) instead of operational taxonomic unit (out) [47,48,49,50,51], or a denoising step, which allows for obtaining microbial taxa with a higher confidence [52], will be needed. This investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication no. 85-23, 1996). The Ethics Committee of Nihon University School of Medicine examined all research protocols involving the use of animals and approved this study (no. 11-034). SHRSP5 rats were obtained from Disease Model Cooperative Research Association (Kyoto, Japan) [13,14]. The SHRSP5 rat is a subline obtained by feeding HFCD to SHRSP rats [53,54]. These SHRSP5 rats are characterized by fat deposition in their arteries, as well as fat deposition in and fibrosis of their livers, indicating the development of diet-induced NASH [15]. The ND group was fed only a stroke-prone (SP) diet. SP diet was purchased as MF from Oriental Yeast Co., Ltd., Itabashi-ku, Tokyo, Japan. The HFCD consisted of 68% (w/w) SP diet, 25% (w/w) palm oil, 5% (w/w) cholesterol, and 2% (w/w) cholic acid [15]. In 100 g of ND, there was approximately 7.9 g water, 23.1 g protein, 5.1 g fat, 5.8 g ash, 2.8 g fiber, 55.3 g soluble without asphyxiation, and 359 kcal, according to the information from Oriental Yeast (https://www.oyc.co.jp/bio/LAD-equipment/LAD/ingredient.html (accessed on 13 February 2023)). The quantities of vitamins A, D3, E, K3, B1, B2, C, B6, B12, inositol, biotin, pantothenic acid, niacin, colin, and folic acid were 1283 IU, 137 IU, 9.1 mg, 0.04 mg, 2.05 mg, 1.1 mg, 4 mg, 0.87 mg, 5.5 mg, 439 mg, 27 μg, 2.45 mg, 10.61 mg, 0.18 g, and 0.17 mg, respectively, in 100 g of ND. We expected each rat to eat ~20 g of the diet daily. Experiments were conducted at least twice for consistent observations. Three rats from each group were examined [15]. Their feces were collected for 16S rRNA sequencing analysis. We only gathered the top layers of the feces and performed the isolation under sterile conditions to avoid bacterial contamination. Isoflurane was used as an anesthesia method for sampling the contents of small intestines. Heart blood was collected under general anesthesia; after abdominal median incision, heart blood was collected as described elsewhere [15]. After the incision of perianal, we collected the content of small intestine for further analysis. We performed animal experiments according to the Japanese animal welfare guidelines (https://www.maff.go.jp/j/chikusan/sinko/animal_welfare.html (accessed on 13 February 2023)) at that time. The total bacterial genomic DNA was extracted using the Extrap Soil DNA Kit Plus ver.2 (Nippon Steel Corporation, Tokyo, Japan) and stored at −20 °C prior to further analysis. The DNA was used in equal amounts for further PCR analysis. The total number of bacterial 16S rRNA genes was estimated using a TaqMan-based qPCR approach with primers Bac1055YF, Bac1392R, and Q-probe Bac 1115Probe, which were described previously [55] (Table 3). In general, 16S and/or internal transcribed spacer ribosomal RNA sequencing are performed for the amplicon sequencing methods to identify and compare the flora of bacteria or fungus of collected samples [50]. This method could identify them after concentrating the original materials using the next-generation sequencing. We performed sequencing analysis of 16S rRNA genes in the present study. The PCR with high-fidelity-DNA polymerase was used to amplify the V4–V5 region of the 16S rRNA gene with primers U515F and 926R (Table 3). Agilent 2100 bioanalyzer (Agilent technologies, Santa Clara, CA, USA) and PicoGreen dsDNA Assay Kit (Invitrogen, Carlsbad, CA, USA) were used to purify and quantify the resulting PCR amplicons. The Illumina-MiSeq platform (Illumina, San Diego, CA, USA) was used to pool the amplicons in equal amounts and implement the paired-end 2 × 250-base-pair sequencing. Finally, base-pair sequences of ~410 bp were analyzed. Standard bioinformatics-alignment comparison was utilized for data analysis [56]. The Quantitative Insights Into Microbial Ecology (QIIME) pipeline was employed to process the sequencing data [16]. Paired-end reads were demultiplexed according to a combination of forward and reverse indices. Additional quality filtering included exact match to sequencing primers and an average quality score of 30 or higher on each read. Prior to further analysis, each paired-end read was stitched into one contiguous read using the fast length adjustment of short reads (FLASH) software tool. Reads that could not be joined were excluded from downstream analysis. All sequences passing filters were aligned against a Silva non-redundant 16S reference database (v108) and assigned taxonomic classifications using USEARCH at a 97% identity threshold. Dereplication to unique reference-sequence-based operational taxonomic units (refOTU) was performed using UCLUST at a 97% clustering threshold and summarized in a refOTU table. Additional alpha-diversity measures and normalized-per-level taxonomic abundances were created using custom scripts written in R [10]. Differentially significant features at each level were identified using linear discriminant analysis (LDA), along with effect-size measurements (LEfSe) [57]. Three-dimensional principal-coordinates analysis (PCoA) plots using the tree-based UniFrac distance metric were generated through custom scripts in R and scripts from the QIIME package [16]. The OUT taxonomic classification was conducted by BLAST, searching the representative sequences set against the database using the best hit, as in previous studies [58]. Classification of bacterial taxonomy based on the end product was performed as previously described [59]. Briefly, genera were classified into more than one group if they were defined as producers of multiple metabolites. Genera that were defined as producing equol, histamine, hydrogen, and propionate constituted only a minor portion of the population and were therefore excluded from this analysis. A representative sequence from each OTU was selected according to the default parameters. As in SIBO syndrome, the SHRSP5 rats fed with a HFCD presented diarrhea and body-weight loss with abnormal types of bacteria in their small intestines, although the number of these bacteria did not increase. Our results strongly support the association between MAFLD and gut-microbiota alteration.

PMC10002597

Xiuxiu Li,Jingya Wang,Fali Zhang,Mubin Yu,Ning Zuo,Lan Li,Jinghe Tan,Wei Shen

Cyanidin-3-O-Glucoside Rescues Zearalenone-Induced Apoptosis via the ITGA7-PI3K-AKT Signaling Pathway in Porcine Ovarian Granulosa Cells

23-02-2023

cyanidin-3-O-glucoside,zearalenone,ITGA7-PI3K-AKT signaling pathway,porcine granulosa cells

Zearalenone (ZEN) is an important secondary metabolite of Fusarium fungi, exposure to which can cause reproductive disorders through its effects on ovarian granulosa cells (GCs) in many mammals, especially in pigs. This study aimed to investigate the protective effects of Cyanidin-3-O-glucoside (C3G) on the ZEN-induced negative effects in porcine GCs (pGCs). The pGCs were treated with 30 µM ZEN and/or 20 µM C3G for 24 h; they were divided into a control (Ctrl) group, ZEN group, ZEN+C3G (Z+C) group, and a C3G group. Bioinformatics analysis was used to systematically screen differentially expressed genes (DEGs) in the rescue process. Results showed that C3G could effectively rescue ZEN-induced apoptosis in pGCs, and notably increase cell viability and proliferation. Furthermore, 116 DEGs were identified, and the phosphatidylinositide 3-kinases-protein kinase B (PI3K-AKT) signaling pathway was the center of attention, of which five genes and the PI3K-AKT signaling pathway were confirmed by real-time quantitative PCR (qPCR) and/or Western blot (WB). As analyzed, ZEN inhibited mRNA and protein levels of integrin subunit alpha-7 (ITGA7), and promoted the expression of cell cycle inhibition kinase cyclin-D3 (CCND3) and cyclin-dependent kinase inhibitor 1 (CDKN1A). After the knock-down of ITGA7 by siRNA, the PI3K-AKT signaling pathway was significantly inhibited. Meanwhile, proliferating cell nuclear antigen (PCNA) expression decreased, and apoptosis rates and pro-apoptotic proteins increased. In conclusion, our study demonstrated that C3G exhibited significant protective effects on the ZEN-induced inhibition of proliferation and apoptosis via the ITGA7-PI3K-AKT pathway.

Cyanidin-3-O-Glucoside Rescues Zearalenone-Induced Apoptosis via the ITGA7-PI3K-AKT Signaling Pathway in Porcine Ovarian Granulosa Cells Zearalenone (ZEN) is an important secondary metabolite of Fusarium fungi, exposure to which can cause reproductive disorders through its effects on ovarian granulosa cells (GCs) in many mammals, especially in pigs. This study aimed to investigate the protective effects of Cyanidin-3-O-glucoside (C3G) on the ZEN-induced negative effects in porcine GCs (pGCs). The pGCs were treated with 30 µM ZEN and/or 20 µM C3G for 24 h; they were divided into a control (Ctrl) group, ZEN group, ZEN+C3G (Z+C) group, and a C3G group. Bioinformatics analysis was used to systematically screen differentially expressed genes (DEGs) in the rescue process. Results showed that C3G could effectively rescue ZEN-induced apoptosis in pGCs, and notably increase cell viability and proliferation. Furthermore, 116 DEGs were identified, and the phosphatidylinositide 3-kinases-protein kinase B (PI3K-AKT) signaling pathway was the center of attention, of which five genes and the PI3K-AKT signaling pathway were confirmed by real-time quantitative PCR (qPCR) and/or Western blot (WB). As analyzed, ZEN inhibited mRNA and protein levels of integrin subunit alpha-7 (ITGA7), and promoted the expression of cell cycle inhibition kinase cyclin-D3 (CCND3) and cyclin-dependent kinase inhibitor 1 (CDKN1A). After the knock-down of ITGA7 by siRNA, the PI3K-AKT signaling pathway was significantly inhibited. Meanwhile, proliferating cell nuclear antigen (PCNA) expression decreased, and apoptosis rates and pro-apoptotic proteins increased. In conclusion, our study demonstrated that C3G exhibited significant protective effects on the ZEN-induced inhibition of proliferation and apoptosis via the ITGA7-PI3K-AKT pathway. Folliculogenesis in female mammals is under precise control. Biological events related to follicle development include primordial follicle assembly, follicle recruitment, follicle growth and maturation, ovulation, and follicle atresia. Follicle development is achieved through a series of complex structural and functional changes in ovarian granulosa cells (GCs), and oocyte–GCs interaction can promote follicle maturation and ultimately female reproductive ability [1,2].Alternately, GC apoptosis is the principal inducement of follicle atresia. Although follicle atresia occurs under normal physiological conditions, in excess follicle atresia can lead to reproductive disorders [3]. So, the normal proliferation and functioning of GCs is directly related to oocyte development, follicle maturation, and pregnancy. Zearalenone (ZEN) is a type of estrogen-like biotoxic secondary metabolite mainly produced by Fusarium fungi. It was first isolated from moldy maize by Stob in 1962 [4]. As one of the most common mycotoxins in animal feed, ZEN and its metabolites can accumulate and be detected in animals organs (cardiac tissue, kidney, liver, intestinal tissue, immune organs, reproductive organs, the fetus, etc.) and products (meat, milk, eggs, etc.), with the reproductive system of mammals being most seriously impacted [5,6,7,8]. In recent years, high levels of ZEN pollution have been recorded worldwide [9]. Exposure to ZEN has affected food safety and livestock feed safety and has seriously reduced the economics of intensive animal husbandry. Pigs are highly sensitive to ZEN exposure; it can lead to reduced feed intake, growth inhibition, immunosuppression, reproductive dysfunction, oxidative stress, cellular apoptosis, and even death [10,11]. Studies have revealed that exposure to excessive ZEN can cause reproductive disorders including prolonged estrus cycle, interference with sex hormones, increased area of the vulva, ovarian atrophy, persistent luteal body, false pregnancy, and abortion in pigs [12]. Previous studies have widely reported the damage that ZEN poses to pGCs, including disrupting gene expression and increasing apoptosis [13,14,15]. Therefore, it is crucial to find methods or agents to mitigate ZEN poisoning. Cyanidin-3-O-glucoside (C3G) is one of the main ingredients of mulberry anthocyanins, but is also widely found in blueberries, black rice, black beans, purple potatoes, and other dark colored plants or fruit. Recent research has begun to pay close attention to the benefits of C3G to human health, and has shown that C3G has several beneficial effects: anti-inflammatory, anti-apoptosis, antioxidant, anti-insulin resistance, regulation of blood lipids, and also anti-tumor [16,17,18]. Existing evidence suggests that C3G reduces oxidative stress damage and improves 3-chloro-1,2-propanediol-induced spermatogenesis in mice [19], and reduces apoptosis in rat Leydig cells and interstitial cells induced by cadmium and lead [20,21]. Therefore, this study set out to explore the positive effects of C3G on ZEN-induced damage in pGCs in vitro. Here, it was hypothesized that C3G could protect the porcine reproductive system from ZEN-induced toxicity. Therefore, pGCs, which synthesize ovarian steroids and produce the cytokines and growth factors required for oocyte development, were selected as a model. RNA-sequencing (RNA-seq) and bioinformatics analysis were systematically used to screen key genes in this rescue process and reveal the potential mechanism, to offer a theoretical basis for the production and application of C3G in the future. As seen in the Figure 1A, the brightfield pictures captured by microscope show that the growth status of pGCs in control (Ctrl), ZEN+C3G (Z+C), and C3G groups was good, and cell confluence was about 70~80%, while the confluence of pGCs in the ZEN group was low and cell growth was poor. Next, the cell viability of each group was detected by cell counting kit-8 (CCK-8). As shown in Figure 1B, compared with Ctrl (100 ± 6.52%), the viability of pGCs in ZEN (69.11 ± 3.12%) was significantly reduced; and compared with ZEN, the viability of pGCs in Z+C (103.75 ± 6.82%) was significantly increased. In addition, the results of cell proliferation detected by flow cytometry also showed similar trends, compared with Ctrl (22.83 ± 0.88%); the positive proportion of 5-Ethynyl-2′- deoxyuridine (EdU) in ZEN (15.9 ± 0.31%) was significantly reduced. Compared with ZEN, the positive proportion of EdU in Z+C (20.6 ± 0.31%) was significantly increased (Figure 1C), and the detailed cell proliferation proportion (positive proportion of EdU) statistics are shown in Figure 1D. Furthermore, the expression levels of cell proliferation-related marker protein proliferating cell nuclear antigen (PCNA) detected by Western blot (WB) shows a decrease in ZEN, while the PCNA levels in Z+C were reversed compared to ZEN (Figure 1E). These results suggested that C3G can mitigate the effect of ZEN on pGCs proliferation. As shown in Figure 2A, TUNEL-positive cells in ZEN were more abundant than in C3G, Ctrl, and Z+C. According to the statistical results shown in Figure 2B, in comparison with Ctrl (0.18 ± 0.05%), the proportion of TUNEL-positive cells was significantly increased in ZEN (1.4 ± 0.11%), while the ratio of TUNEL-positive cells in Z+C (0.23 ± 0.06%) was significantly decreased compared with ZEN. Furthermore, the apoptosis rate of pGCs in each group, as detected by flow cytometry, also showed the same trends (Figure 2C). In comparison with Ctrl (10.05 ± 0.13%), apoptosis rates were significantly increased in ZEN (13.49 ± 0.59%). Meanwhile, the apoptosis rates in Z+C (8.75 ± 0.25%) were significantly decreased compared with ZEN (Figure 2D). Moreover, the expression levels of anti-apoptotic-related protein B-cell lymphoma-2 (BCL2), pro-apoptotic-related protein BCL2-associated x (BAX), and apoptotic executioners protein cysteinyl aspartate specific proteinase 9 (caspase9, Casp9) were further verified by WB. After treatment with ZEN, a significant increase in the levels of BAX/BCL2 ratio (Figure 2E) and cleaved-caspase9 (C-Casp9)/Casp9 ratio (Figure 2F) were observed. These results indicated that C3G could reduce apoptosis in ZEN-induced pGCs. RNA-seq was performed to deeply explore the mechanism of C3G rescue of ZEN toxicity in pGCs (Figure 3A). Firstly, the expression profile and function of differentially expressed genes (DEGs) were analyzed. There were 593 DEGs in ZEN vs. Ctrl, among which 306 were up-regulated and 287 were down-regulated. In Z+C vs. ZEN, there were 1595 DEGs, among which 1290 were up-regulated and 669 were down-regulated. In C3G vs. Ctrl, there were 1616 DEGs, among which 580 were up-regulated and 1036 were down-regulated (Figure 3B,C). There were 116 key DEGs in the intersection between the total DEGs in ZEN vs. Ctrl, Z+C vs. ZEN, and C3G vs. Ctrl (Figure 3D). Next, the functional enrichment of these DEGs was analyzed by gene ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG). In ZEN vs. Ctrl, there were a total of 30 biological process (BP) GO terms and the top 11 are shown according to p-values (Figure S2A and Table S4). The top three GO terms were GO:0007167 (enzyme linked protein signaling pathway), GO:0019752 (carboxylic acid metabolic process), and GO:0033993 (response to lipid). In Z+C vs. ZEN, the DEGs were associated with three GO terms, GO:0046777 (protein autophosphorylation), GO: 0051128 (regulation of cellular component biogenesis), and GO:1901888 (regulation of cell junction assembly) according to p-values (Figure S2B and Table S5). In C3G vs. Ctrl, the DEGs were associated with 13 GO terms and the top 15 are shown according to p-values (Figure S2C and Table S6). There were many important GO terms, including GO:0090090 (negative regulation of canonical Wnt signaling pathway), GO:0006119 (oxidative phosphorylation), and GO:0046034 (ATP metabolic process). Furthermore, the phosphatidylinositide 3-kinases-protein kinase B (PI3K-AKT) signaling pathway was the key signaling pathway analyzed by KEGG in ZEN vs. Ctrl, and Z+C vs. ZEN, but not in C3G vs. Ctrl (Figure 3E and Tables S7–S9). ClueGO was also performed for the DEGs, of which the top six pathways were visualized, including the PI3K-AKT signaling pathway, mTOR signaling pathway, JAK-STAT signaling pathway, and focal adhesion (Figure 4A). Next, a network diagram of PI3K-AKT signaling was constructed, which was split into three modules (Figure 4B). Of the expression patterns of the five DEGs, cyclin-dependent kinase inhibitor 1 (CDKN1A), cyclin-D3 (CCND3), phosphoe nolpyruvate carboxykinase 2 (PCK2), activating transcription factor 4 (ATF4) and integrin subunit alpha 7 (ITGA7), shown in Figure 4C, CDKN1A, CCND3, PCK2, and ATF4 were significantly up-regulated and ITGA7 was significantly down-regulated in pGCs after ZEN treatment; however, the opposite trend occurred after C3G treatment. The relative mRNA expression levels confirmed for those genes were in line with their FPKM values (Figure 4D and Table S10). As predicted, compared with Ctrl, the relative mRNA levels of CDKN1A, CCND3, PCK2, and ATF4 increased significantly, and ITGA7 declined significantly in ZEN. Meanwhile, with the addition of C3G, those changes would be reversed dramatically compared with ZEN. WB showed that CCND3 (Figure 4E) increased significantly in ZEN, and decreased significantly in Z+C; ITGA7 (Figure 4F), p-PI3K/PI3K (Figure 4G), and p-AKT/AKT (Figure 4H) decreased significantly in ZEN, and increased significantly in Z+C. The results demonstrate that these genes are indeed regulated by C3G and ZEN, and are related to the PI3K-AKT signaling pathway, which stimulates hypotheses for subsequent research. To better understand the role of ITGA7, RNAi experiments were performed to knock-down ITGA7 to determine its effects on the PI3K-AKT signaling pathway in pGCs (Figure 5A). Real-time quantitative PCR (qPCR), WB, and cell immunofluorescence were used to analyze ITGA7-PI3K-AKT expression after siITGA7 treatment in pGCs. Results showed that the mRNA relative level (Figure 5B), protein level (Figure 5C), and fluorescence in-tensity (Figure 5D,E) of ITGA7 were dropped significantly in ZEN, siITGA7, and Z+C+siITGA7 groups compared with Ctrl and NC. Homoplastically, compared with Ctrl and NC, p-PI3K/PI3K (Figure 5F), and p-AKT/AKT (Figure 5G) protein expression levels were decreased significantly in ZEN, siITGA7, and Z+C+siITGA7. These results demonstrated that C3G plays an important role via the ITGA7-mediated PI3K-AKT signaling pathway in pGCs. The effects of ITGA7 knock-down on pGCs proliferation and apoptosis were further detected. Firstly, as shown in Figure 6A, apoptotic pGCs increased after transfection with si-ITGA7, but this did not take place in Ctrl and NC. Subsequently, the TUNEL-positive proportion increased significantly in ZEN (6.18 ± 0.78%), siITGA7 (5.00 ± 0.69%), and Z+C+siITGA7 (9.69 ± 0.73%) compared with Ctrl (0.83 ± 0.30%) and NC (1.69 ± 0.23%), respectively (Figure 6B,C). Ultimately, the protein expressions of PCNA, BAX, and BCL2 were detected. WB showed that the level of PCNA (Figure 6D) markedly decreased and the level of BAX/BCL2 ratio (Figure 6E) remarkably increased in ZEN, siITGA7, and Z+C+siITGA7 compared with Ctrl and NC, respectively. In conclusion, ZEN inhibited pGCs proliferation by decreasing PCNA expression and promoted apoptosis by increasing the BAX/BCL2 ratio, which was rescued by C3G via the ITGA7-PI3K-AKT signaling pathway. ZEN is widespread and extremely harmful to the reproductive system through its impact on ovarian GCs [22,23,24]. In this study, we demonstrated that C3G reduced ZEN-induced apoptosis in pGCs via the ITGA7-PI3K-AKT signaling pathway. It is well known that apoptosis is a process that is strictly controlled by the BCL2 family, the caspase family, and others, and that polygenes are conserved between species. The Casp9-dependent mitochondrial signaling pathway is a classic apoptotic signaling pathway [25]. Similar to previous studies [13,14,15], we found that ZEN was able to induce pGC apoptosis (Figure 2), which was manifested by an increase in BAX protein content, a decrease in BCL2 protein, and the TUNEL signal was also significantly increased after ZEN exposure. Prior to the current study, little was known about the relationship between C3G and ZEN-induced pGC apoptosis. To gain further insights into how C3G can rescue ZEN’s adverse effects on pGCs in this study, RNA-seq, a technique widely used in the analysis of bioinformatical data, was performed to detail analysis changes in the expression of genes in this process. We found that CDKN1A, CCND3, and ITGA7 showed remarkable changes and were enriched via the PI3K-AKT signaling pathway. PI3K-AKT is an antiapoptotic signaling pathway, and plays an important role in the development and function of ovarian GCs [26,27]. The CCND3 gene, a downstream signal molecule in the PI3K-AKT signaling pathway, is an inducer of inhibitions of the cell cycle and a regulator of apoptosis [28]. The CDKN1A gene, a member of the Clp family, is a cyclin-dependent kinase inhibitor located downstream of the p53 gene and encodes p21 protein [29]. p21 is tightly involved in balancing and coordinating proliferation with cellular processes, and it inhibits cell proliferation directly through binding to cyclin-dependent kinases (CDKs) and PCNA. It is worth noting that PCNA is present in the nuclei of all dividing cells, and plays a vital role in DNA replication machinery and in connecting different DNA metabolic pathways [30,31]. However, p21 is regarded as a modulator of apoptosis. In cancer cells, p21 activates autophagy to expedite cell death; while in normal cells, it inhibits autophagy and induces apoptosis [32]. These reports indicate that C3G may exert an important anti-apoptotic effect and promote cell proliferation, which may be related to the PI3K-AKT signaling pathway. Consistent with previous studies, C3G protects mouse hepatocytes against apoptosis induced by high glucose levels via mitochondria and the PI3K-AKT signaling pathway [33]. Furthermore, C3G also mediates protection against many other physical and chemical substances via endoplasmic reticulum stress- and oxidative stress-induced apoptosis in many animal cells [34,35,36,37]. Although oxidative phosphorylation and the mitogen-activated protein kinase (MAPK) signaling pathway, and others, also varied in our RNA-seq data, they were not addressed in the current work, which was a deficiency and will be further explored in the future. Here, we paid attention to the relationship between ITGA7 and the PI3K-AKT signaling pathway. The ITGA7 gene encodes an integrin subunit alpha 7 protein which is a member of the integrin alpha chain family through which transmembrane heterodimers involved in cell adhesion or cell–extracellular matrix adhesion to regulate cell behavior, and thus play an important role in cell growth, proliferation, differentiation, survival, and migration [38]. Ming et al. report that ITGA7 exerted anti-apoptotic effects via focal adhesion kinase (FAK) to activate the AKT signaling pathway, which subsequently inhibited the release of Cyt c into the cytoplasm and the cleavage of Casp9, Casp3, and poly ADP-ribose polymerase (PARP) [39]. As expected, we confirmed that the PI3K-AKT signaling pathway was inhibited, the level of apoptosis was raised, and proliferation was inhibited (Figure 5 and Figure 6) after knock-down of the ITGA7 gene in pGCs, and the up-regulation effect of C3G on ITGA7 could be offset by siITGA7, even siITGA7 may even boost ZEN’s ability to reduce ITGA7 expression. Therefore, our work highlights that C3G can attenuate apoptosis induced by ZEN via promoting the ITGA7-mediated PI3K-AKT signal pathway and inhibiting the Casp9-dependent mitochondrial signaling pathway in pGCs. Porcine ovaries were collected from the Qingdao Wanfu Pig Breeding Base (Qingdao, Shandong, China). The ovaries were saved in 37 °C saline, with 3% penicillin and streptomycin (Solarbio, P1400, Beijing, China), and transported to the laboratory within 2 h. All animal care and procedures were performed according to the Ethics Committee of Qingdao Agricultural University (approval No. 2019-036). Procedures for primary pGCs isolation in vitro were as previously described, with improvements [40]. Briefly, antral follicles with a diameter between 2 mm and ~4 mm were collected using a syringe (2.5 mL) for the culture of pGCs. Then, the swine follicular fluid was centrifuged at 1500 rpm for 3 min. After the removal of blood by washing with phosphate-buffered saline (PBS) and filtering bulky tissues with 40 µm sieves, the swine follicular fluid was again centrifuged at 1500 rpm for 3 min. Finally, the pGCs were cultured in Dulbecco’s modified Eagle high glucose medium (DMEM, Gibco, C11995500BT, Shanghai, China), with the addition of 10% (v/v) fetal bovine serum (FBS, PAN, ST190318, South America), 1% (v/v) penicillin (100 IU/mL)—streptomycin (0.1 mg/mL)—amphotericin (0.25 µg/mL) (Solarbio, P7630, Beijing, China) and 0.1% (v/v) gentamicin (50 µg/mL) (Solarbio, L1312, Beijing, China) at 37 °C under 5% CO2, 95% air, and saturation humidity. Following 12 h of culture in cell culture flasks, the media were replaced, and cells were cultured until the cell density reached 80~90% within 48 h to ~72 h. Then, the cells were digested by trypsin for subculture in different sized cell culture dishes, which were selected according to different experimental requirements. ZEN (Sigma-Aldrich, Z215, St. Lousis, MO, USA) was dissolved in dimethyl sulfoxide (DMSO) with 100 mg/mL and stored at −20 °C. C3G (Solarbio, IK0070, Beijing, China) was dissolved in DMSO with a stock solution of 10 mM and stored at −20 °C. ZEN and/or C3G was added when the convergence degree reached 40% to ~50% in F1 generation cells. The pGCs were cultured in 4 groups: 30 μM ZEN (ZEN group), 30 μM ZEN with 20 μM C3G (Z+C group), C3G group, and the corresponding concentration of DMSO as a vehicle control (Ctrl group). The concentrations used have been established on the basis of our previous evidence (for ZEN) [13,14,41] and preliminary experiments in the current study (for C3G, Figure S1). For this study, the pGCs were treated with ZEN and/or C3G for 24 h. Cell viability was measured by the CCK-8 (Solarbio, CA1210, Beijing, China). pGCs were seeded into 96-well plates at a density of 2 × 104 cells/well, and divided into 4 groups consistently with grouping in Section 4.3, and 3 repetitions in each group. After culture for 21 h, 10 μL of CCK-8 solution was added to 100 μL medium, and the cells were incubated for an additional 3 h at 37 °C with 5% CO2. Then, the absorbance (A) value was detected at 450 nm by using a microplate reader (Power Wave Xs2, BioTek, Winooski, VT, USA). For the blank group without cells, only medium with 10% (v/v) CCK-8 was added. Cell viability (%) = [A(dosed) − A(blank)]/[A(Ctrl) − A(blank)] × 100%. pGCs were seeded into 6 cm dishes with a density of 1 × 105 cells/dish; when the convergence degree reached 40%~50%, they were treated with ZEN and/or C3G for 20 h, and then EdU was added to the medium (1:2500) according to the cell-light EdU Apollo567 in vitro flow cytometry kit (Ribobio, C10338-1, Guangzhou, China) for another 4 h. Subsequent steps were scheduled for completion under the instruction of the manufacturer. EdU is a thymic nucleotide analogue, which can be a substitute for thymidine incorporation into the replicating DNA during cell proliferation, and fluorescent red in a specific reaction with Apollo fluorescent dye. Hence, positive proportions of EdU as detected by flow cytometry (FACSAria III, BD, San Jose, CA, USA) represented the ability of cells to proliferate. An Annexin V-FITC/PI apoptosis detection kit (Meilunbio, MA0220, Dalian, China) was used to test the apoptosis rates of pGCs, and all protocol steps were completed according to the manufacturer’s instructions. In simple terms, pGCs were collected immediately into 1.5 mL centrifuge tubes after treatment with ZEN and/or C3G for 24 h at 37 °C with 5% CO2. After being washed once in PBS, these cells were resuspended in 1× binding buffer and incubated with Annexin V-FITC and/or PI in the dark for 15 min. The results were analyzed by flow cytometry (FACSAria III, BD, San Jose, CA, USA) using FlowJo (v.10.0) software. TUNEL staining was performed following the instructions of the TUNEL bright green apoptosis detection kit (Vazyme, A113, Nanjing, China). Simply, the pGCs were fixed with 4% paraformaldehyde (PFA) for 25 min at room temperature (RT). After centrifugation, cell deposits were harvested for smear production, which was round with a diameter of 1 cm. After desiccation, the sections were incubated with proteinase K for 5 min. Next, 1 × equilibration buffer was added to the samples for 30 min at RT. After removal of the proteinase K, 50 μL of TUNEL reaction mixture [(ddH2O 34 μL, 5 × equilibration buffer 10 μL, bright green labeling mix (5 μL), and recombinant TdT enzyme (1 μL)) were added, and the recombinant TdT enzyme in the negative group was replaced with ddH2O. PI (red) or hoechst33342 (blue) was used for nuclei staining, and pictures were captured using a microscope (Olympus, BX51, Tokyo, Japan). Positive green fluorescence (TUNEL) was observed at 520 ± 20 nm, red fluorescence (PI) was observed at >620 nm, and blue fluorescence (hoechst33342) was observed at >460 nm. An average proportion of TUNEL-positive cells in 5 fields was photographed for the results, and each group had 3 replicates. The total RNA of pGCs with 107 cells in each sample extracted by RNAiso plus (TaKaRa, SD1412, Kusatsu, Japan) according to the manufacturer’s instructions, were processed through a Novogene (Tianjin, China) Illumina platform for RNA-seq and library construction. The quality and quantity of the library were monitored by a BioAnalyzer 2100 system (Agilent Technologies, Palo Alto, CA, USA). Ten pM libraries were denatured, captured on Illumina flow cells, amplified in situ, and finally sequenced for 150 cycles using the Illumina PE 150 sequencer (Illumina, Davis, CA, USA). Firstly, quality control of raw data was carried out by fastq (version v0.11.8) software [42], from which low quality reads, joints, and poly-N sequences were removed using fastp (version v0.19.5) software [43]. After building the index and aligning clean data to the Sus scrofa Ensembl reference genome (susScr11) utilizing STAR (STAR_2.7.0b), the clean data were mapped and aligned to the Sus scrofa reference genome (susScr11) utilizing STAR software [44]. DEGs were identified using the R/Bioconductor DESeq2 package [45]. To avoid possible bias, an analysis was normalized; adjusted p-values (p-adj) < 0.05 were considered statistically significant. Then, the R/Bioconductor clusterProfiler package was used to analyze the functional profiles of those DEGs, and the ‘org.Ss.eg.db’ database was used to convert the gene symbol to entrezID to perform GO term enrichment analysis. There are 3 components of GO terms: BP, molecular function (MF), and cellular component (CC), of which BP was the core information mining direction related to our research direction. The signal diagrams for KEGG enrichment were produced by R/Bioconductor Pathview package [46]. The results of KEGG pathways examined by ClueGO [47] (a plug-in of Cytospace, v.2.5.9) and the PPI (MCODE, v.1.3, a plug-in of Cytospace) network were visualized by Cytoscape (v.3.9.1, https://cytoscape.org (accessed on 25 May 2022)) [48]. The total RNA of pGCs was extracted by using RNAiso Plus according to the manufacturer’s instructions, and reverse transcription into cDNA was performed with a SPARKscript II RT Plus kit (Sparkjade, AG0304, Jinan, China). The resulting cDNA was then subjected to qPCR following the manufacturer’s protocol of the chamQ universal SYBR qPCR master mix (Vazyme, Q711, Nanjing, China) by using a Bio-Rad CFX 96 real-time PCR System (Bio-Rad, Hercules, CA, USA). Relative mRNA levels were analyzed utilizing the 2(−ΔΔCt) method and normalized against the mRNA expression of the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The primers were designed on the National Center for Biotechnology Information (NCBI, https://www.ncbi.nlm.nih.gov/ (accessed on 9 June 2022)) and purchased from Tsingke Biotechnology Co., Ltd. (Beijing, China). Information on the primer sequences is summarized in Table S1. Total protein was extracted from the pGCs samples using radioimmunoprecipitation assay (RIPA) lysis buffer (Beyotime, P0013C, Shanghai, China) containing protease and phosphatase inhibitors (Beyotime, P1028, Shanghai, China), and the concentrations were measured using the bicinchoninic acid (BCA) method. Subsequently, the protein separated by SDS-PAGE was transferred onto polyvinylidene fluoride (PVDF) membranes by electrophoresis. Samples were blocked with TBST (Tris-buffered saline with Tween-20) that contained 5% BSA at 4 °C overnight; the membranes were then incubated with primary antibodies (Table S2) at different dilutions in 5% BSA at 4 °C overnight. The next day, following washing with TBST, the membranes were incubated with the secondary antibodies for 1.5 h at RT. A BeyoECL plus kit (Beyotime, P0018, Shanghai, China) was used for signal detection, a band containing the target protein was added to the Tanon 5200 system (Tanon, Shanghai, China) and photographed. Finally, Image J software was used to analyze the gray value of target protein bands which represented the relative expression levels of proteins, with β-actin as an internal reference. pGCs were cultured in 6 cm dishes with ZEN and/or C3G for 24 h, and then transfected with 15 µL siRNA of ITGA7 (ITGA7-NC/si1/si2/si3; GenePharma, A10005, Shanghai, China; Table S3 and Figure S3) and 10 μL GP-transfect-Mate (GenePharma, G04008, Shanghai, China), each diluted with 500 μL DMEM, and maintained for 5 min at RT; siRNA and GP-transfect-Mate were then mixed and incubated for a further 20 min at RT. Aliquots of 1 mL of the mixture were added to 4 mL of DMEM and culture was performed at 37 °C for 6 h. After transfection, the medium was exchanged for fresh pGCs medium. After culturing for 48 h, the pGCs were harvested for downstream analysis. After RNAi, pGCs were harvested by centrifugation and fixed by 4% PFA for 30 min at 4 °C. Then, the cells were smeared onto slides as part of “TUNEL staining”. Next, for permeabilization with PBST (PBS with 0.5% Triton-X-100 (Solarbio, P1080, Beijing, China)), the slides were blocked in TBST (10% goat serum in TBS (Boster, AR0031, Wuhan, China)) for 40 min at RT. The slides were then incubated with primary antibodies of ITGA7 (1:100) overnight at 4 °C. The next day, after washing 3 times with PBS, the slides were labeled with secondary antibodies: donkey anti-rabbit lgG HL (Alexa Fluor® 555) (Abcam, ab150074, USA; 1:200) at 37 °C for 1 h. Nuclei were stained with Hoechst33342 (Beyotime, C1022, Shanghai, China) for 5 min. At least 5 representative pictures from each slides were captured under a fluorescence microscope imaging system (Olympus, BX51, Tokyo, Japan). Image J software was used to determine the mean fluorescence intensity per slide. Finally, the data with the Ctrl group as a reference were used to calculate the change in fluorescence intensity in NC, ZEN, siITGA7, and Z+C+siITGA7. All results were expressed as mean ± standard error of the mean (SEM) from 3 repeats and/or 3 independent experiments. Comparisons between multiple groups were analyzed using one-way ANOVA, and further pairwise comparisons were analyzed using LSD tests. All statistical analyses were performed using SPSS 25.0 software (IBM SPSS, Inc., Armonk, NY, USA). GraphPad Prism 8.0 software (GraphPad Software, Inc., San Diego, CA, USA) was used only for generating statistical graphs. * p < 0.05 was considered significantly different. ** p < 0.01 was considered extremely significantly different. Labels of unalike letters indicate significant differences. Capital letters represent p < 0.01. In summary, our study demonstrated that C3G significantly protected pGCs from apoptosis induced by ZEN, and this was attributed to its promotion of the ITGA7-PI3K-AKT pathway. This study reports, for the first time, the specific mechanism of the protective effects of C3G on ZEN-induced pGCs apoptosis in vitro, which offers new theories for protecting porcine ovarian function from ZEN, and a theoretical basis for the application of C3G as a feed additive in the future.

PMC10002601

Marta Majchrzak,Sebastian Sakowski,Jacek Waldmajer,Pawel Parniewski

New Genetic Markers Differentiating IPEC and ExPEC Pathotypes—A New Approach to Genome-Wide Analysis Using a New Bioinformatics Tool

28-02-2023

genetic markers,genome-wide analysis,data analysis,sequence analysis

The increasingly expanding genomic databases generate the need for new tools for their processing and further use. In the paper, a bioinformatics tool, which is a search engine of microsatellite elements—trinucleotide repeat sequences (TRS) in files of FASTA type—is presented. An innovative approach was applied in the tool, which consists of connecting—within one search engine—both mapping of TRS motifs and extracting sequences that are found between the mapped TRS motifs. Accordingly, we present hereby the tool called TRS-omix, which comprises a new engine for searching information on genomes and enables generation of sets of sequences and their number, providing the basis for making comparisons between genomes. In our paper, we showed one of the possibilities of using the software. Using TRS-omix and other IT tools, we showed that we were able to extract sets of DNA sequences that can be assigned only to the genomes of the extraintestinal pathogenic Escherichia coli strains or to the genomes of the intestinal pathogenic Escherichia coli strains, as well as providing the basis for differentiation of the genomes/strains belonging to each of these clinically essential pathotypes.

New Genetic Markers Differentiating IPEC and ExPEC Pathotypes—A New Approach to Genome-Wide Analysis Using a New Bioinformatics Tool The increasingly expanding genomic databases generate the need for new tools for their processing and further use. In the paper, a bioinformatics tool, which is a search engine of microsatellite elements—trinucleotide repeat sequences (TRS) in files of FASTA type—is presented. An innovative approach was applied in the tool, which consists of connecting—within one search engine—both mapping of TRS motifs and extracting sequences that are found between the mapped TRS motifs. Accordingly, we present hereby the tool called TRS-omix, which comprises a new engine for searching information on genomes and enables generation of sets of sequences and their number, providing the basis for making comparisons between genomes. In our paper, we showed one of the possibilities of using the software. Using TRS-omix and other IT tools, we showed that we were able to extract sets of DNA sequences that can be assigned only to the genomes of the extraintestinal pathogenic Escherichia coli strains or to the genomes of the intestinal pathogenic Escherichia coli strains, as well as providing the basis for differentiation of the genomes/strains belonging to each of these clinically essential pathotypes. There is a lack of tools for analyzing the enormity of data obtained through NGS sequencing in today’s world. We propose a new search engine based on the use of microsatellite sequences. Microsatellite sequences, also known as Simple Sequence Repeats (SSRs) or Short Tandem Repeats (STRs) [1], which are sequences of between one and six nucleotides in DNA repeated in a tandem fashion are present in genomes of every organism. They are dispersed throughout the genome and can offer the basis for many molecular tools relevant in medical diagnostic procedures, epidemiology, evolutionary examinations, or criminal studies [2,3]. From the point of view of diagnostics and molecular epidemiology of microorganisms, particularly interesting and promising are the trinucleotide repeat sequences that make one of the most numerous groups among microsatellite sequences in bacteria. Regarding the scope of the abovementioned considerations, a general approach has recently been developed, which has made it possible—with the use of TRS profiling [4] and branching processes—to predict the directions of pathogenicity development in the E. coli population [5]. Recent years have seen the appearance of tools that enable the search for microsatellite sequences [6,7]. However, these tools are focused exclusively on determining the place of occurrence of microsatellite sequences on a genome, a particular case of which can be TRS motifs. There is a shortage of informatic tools that simultaneously map the genome using TRS motifs and find DNA sequences between them. Consequently, in this work, the tool called TRS-omix is presented. The tool represents a new engine to search information in genomes and enables both mapping and extracting sequences between the flanking sequences (each flanking sequence includes only one TRS motif). The analysis of many bacterial genomes shows that the most common microsatellite sequences in genomes represent repetitions of three nucleotides (trinucleotide repeats). Figure 1 shows the distribution of microsatellites in the genomes of Escherichia coli UTI89 and Escherichia coli O157:H7 Sakai, representing two essential and not always genetically distinguishable pathotypes of this species, ExPEC (Extraintestinal Pathogenic Escherichia coli) [8,9,10] and IPEC (Intestinal Pathogenic Escherichia coli) [9]. The analysis was performed with MICdb software 3.0 [11] on the MICAS 3.0 platform using E. coli UTI89 and E. coli O157:H7 strain Sakai genomes available at the database. The number of all possible trinucleotide sequences (TRS) is 64 (nk = 43 = 64, all variations of the k-elements with repetitions of the set of n-elements). For each of these sequences, excluding AAA, CCC, GGG, and TTT, repeating it three times as the co-called TRS motif is considered. The set of all 60 examined TRS motifs arranged in three groups is included in Table 1. There exist various types of scientific software that can be used for whole-genome analyses of microsatellites’ occurrence in the genome. Of these, there are many that allow searching for tandem repeat sequences or microsatellites, e.g., TRF (Tandem Repeats Finder) [12], IMEx (Imperfect Microsatellite Extractor) [7], and many others. However, to the best of our knowledge, no software has yet been created that provides the ability to extract sequences between such trinucleotide repeats. Importantly, each of the genomes tested will contain a different, more or less similar set of such sequences, depending on the overall similarity of the genomes, making the software potentially useful for comparative genomic analyses. In opposition to other well-known software detecting microsatellites, the TRS-omix search engine presented here finds microsatellites—trinucleotide repeats and also extracts sequences between such trinucleotide motifs. Such an approach can find a number of applications in various aspects of genomics. This work focusses on the bacterial model, specifically the genomes of Escherichia coli UTI89 and O157:H7 Sakai. It excludes other examinations, including eukaryotic organisms or other genomes, which is possible with this software. The TRS-omix software was developed with the use of the ANSI C language. This makes it possible, on the one hand, to use it on different bioinformatics IT platforms and, on the other to integrate it with libraries of different languages (including: R or Python) which are dedicated to applications of, among others, DNA computing, statistical analysis, or data mining techniques in big data. When the main TRS-omix search engine computational method was constructed, the structures of data which use dynamic memory allocation were designed and applied. The TRS-omix tool, proposed in this work, includes two types of functionalities. The first one makes it possible to select the examination of genomes with circular structures (e.g., the majority of genomes and bacterial plasmids, mitochondrial DNA and chloroplast DNA) or a linear structure (the majority of eukaryotic genomes). The other functionality enables one to determine the minimal and maximal lengths of sequences searched for, which are found between the flanking sequences (each flanking sequence includes only one TRS motif). Therefore, it is possible to extract DNA sequences of length k (k is a natural number, where k > 0) or of length within a certain range <a, b> (a and b are natural numbers, where a, b > 0 and a < b). The schematic diagram of action and information flow in the TRS-omix tool is shown in Figure 2. The input data are the following: file sequence.fasta and file trs.txt. The first file is one of FASTA type, which contains the examined DNA sequences. The file trs.txt is a textual one that contains searched for TRS motifs grouped in classes. In this file, the searched for patterns (TRS motifs) are defined in successive verses, where motifs belonging to one of the twenty possible classes (see Table 1) are inserted within one verse. Each of the verses constitutes a sequence of TRS motifs, while each of the TRS motifs in the given verse is preceded by the “#” sign. An exemplary class that includes the following sequences of nucleotides (TRS motifs): CCGCCGCCG, CGCCGCCGC, and GCCGCCGCC in file trs.txt are defined by the following verse: #CCGCCGCCG#CGCCGCCGC#GCCGCCGCC. The output data of the engine of TRS-omix are information on sequences existing between the flanking sequences (each flanking sequence includes only one TRS motif) that are found in the textual file interiors.txt. The software was elaborated in such a manner as to enable searching TRS motifs in FASTA files downloaded, for instance, from GenBank—the file called sequence.fasta. After starting the software, the examined linear or circular structure ought to be selected and then the determined value of minimal or maximal length of the sequence searched for. In the case where the minimal value is identical with the maximal one, we want to extract the sequences found between TRS motifs of a concrete length. In the case where these values vary, we define extracting sequences of the existing TRS motifs of the length found within the defined range. On finishing the work of the search engine, the file interiors.txt is returned (see Figure 3). The functioning of the software was tested on FASTA type files containing different nucleotide sequences, e.g., Human chromosome 14—complete sequence, as well as other sequences of bacterial origin (see Table 2), on a computational server. The TRS-omix search engine was tested in both the Linux and Windows operating system environment. The source code and sample data are freely available for download at https://github.com/TRS-omix/software (accessed on 21 February 2023), distributed under the GNU GPLv3 license. The experimental tests were carried out on the basis of genome datasets coming from the NCBI database (https://www.ncbi.nlm.nih.gov). Two genomes of Escherichia coli were analyzed—UTI89 [13], representing the ExPEC pathotype [8], the UPEC sub-pathotype [13], and O157:H7 Sakai [14], representing the IPEC pathotype, the EHEC sub-pathotype [15]. As part of the TRS-omix validation, the general approach was to compare the similarity of the sequences generated by the software in such a way as to eliminate those that are identical/similar in both genomes and leave only those that are unique for each of them. Knowing that TRS-omix only searches for sequences between perfect TRS motifs, additional genomic searches with Vector NTI 11.5 software allowed screening out those sequences that had counterparts in both genomes, but their flanking sequences in one of the genomes were imperfect. Finally, the unique sequences of each genome were verified by BLAST N analysis to leave only those with complementary matches only in the genomes of strains belonging to either the ExPEC or IPEC group of pathogens. These sequences have been proposed as specific markers for these E. coli pathogenic groups. The general approach is schematically presented in Figure 4 and is described in detail in the following (data analysis with the use of TRS-omix and other IT tools). To illustrate the applicability of the TRS-omix software in conjunction with other IT tools, an example analysis of the similarity data of two E. coli genomes for sequences between (TRS) n ≥ 3 motifs will be presented. In this work, we conducted such analyses for two genomes of closely related strains of the Escherichia coli species—uropathogenic UTI89 (Acc. No NC_007946.1) and diarrheal O157:H7 Sakai (Acc. No BA000007.3). Genomic sequences flanked by the TRS motif can be flanked on both sides, on the 3′ direction and on the 5′ direction. It was decided mainly to check whether some of the extracted sequences could be specific for the two genomes studied. In addition, to assess whether it is possible to search for sequences specific for the E. coli ExPEC or IPEC pathotypes represented by these genomes. The analysis was carried out in the following steps: Stage 1. Use of TRS-omix on each of the genomes. All fragments containing DNA sequences between TRS motifs for the studied UTI89 and O157:H7 genomes were 2640 and 2777, respectively (raw text data, Supplementary Figures S1 and S2; and Excel format, Supplementary Figure S3). Interestingly, the algorithm used in the MICdb 3.0 software database [11] showed 2371 and 2491 sequences of this type for these genomes, respectively (see Figure S1). Thus, this algorithm did not find a certain number of TRS repeats, indicating the advantage of our software. Stage 2. Comparison of the similarity of sequences from all studied classes between TRS repeats n ≥ 3 occurring in both genomes, using the Vector NTI 11.5 software, CLUSTAL W algorithm. Only sequences of at least 100 bp length were considered for comparison in these studies. Such a comparison allowed one to determine which sequences are identical or significantly similar in both bacterial genomes tested, leaving only those that remained specific for one or the other genome. The similarity of the sequences extracted by TRS-omix from both genomes was compared for each class separately. For example, in the case of sequences between (CCG) n ≥ 3 and (TCG) n ≥ 3 repeats (Supplementary Figure S4), the analysis showed the presence of a pair of homologous sequences and seven singletons (see Figure 5). Similar analyses were performed for all sequences tested between different combinations of TRS motifs. Stage 3. Checking the actual uniqueness of the sequence. The sequences pre-qualified as unique (seven singletons in the example above) to one genome may still have their counterpart in another genome, which could have occurred due to nucleotide changes (mutations) in the flanking TRS sequence and failure of TRS-omix to find such a sequence (imperfect TRS) or the presence of a different TRS motif within the sequence, which resulted in assigning it to another class. In addition, some sequences were partially homologous to their counterparts. Such a situation can be seen in the UTI-89-1897 1875 bp fragment, which shares homology with 340 bp of the Sakai sequence. Some other genomic rearrangements, such as internal insertions/deletions, may have occurred (SAK-1679 counterpart sequence in the UTI89 genome), leaving homology at 5′- or 3′-parts of sequences, and therefore were not considered truly unique. (Supplementary Figure S4, Table CCG-TCG). Out of seven sequences initially qualified as unique, only two turned out to have no homologous sequences in the second examined genome. There were 7182 bp and 389 bp sequences in the E. coli O157: H7 Sakai genome (SAK-436 and SAK-2393, respectively). (Supplementary Figure S4, Table CCG-TCG). Stage 4. Assess whether the selected DNA sequences, unique for a given pathotype, can be potentially helpful in typing the group of ExPEC and IPEC pathogens. Examination of whether the selected sequences are typical only for the genomes of the ExPEC pathotype, represented by the UTI89 or IPEC genome, represented by the O157: H7 Sakai, was carried out by BLAST N (highly similar sequences, megablast) analysis on the NCBI platform (https://blast.ncbi.nlm.nih.gov/Blast.cgi). The decisive criterion for choosing a given DNA sequence as typical for the ExPEC pathotype was the confirmed lack of homology (1000 maximum target sequences) in the BLAST N database to the genomic sequences of: Enterobacteriaceae other than Escherichia coli, Escherichia other than E. coli, E. coli belonging to the IPEC pathotype, non-pathogenic E. coli. Bacteriophages. Similarly, the decisive criterion for choosing a given DNA sequence as typical for the IPEC pathotype was the confirmed lack of homology (1000 maximum target sequences) in the BLAST N database to the genomic sequences of: Enterobacteriaceae other than Escherichia coli, Escherichia other than E. coli, E. coli belonging to the ExPEC pathotypes, non-pathogenic E. coli. Bacteriophages. In the above example (sequences between CCG n ≥ 3 and TCG n ≥ 3 repeats), only the 389 bp O157: H7 Sakai genomic sequence showed the only similarity to E. coli genomes belonging to the IPEC pathotype. Thus, it can be considered a useful marker in the genetic typing of this group of microorganisms. The other sequence (SAK-436) was excluded as it showed homology to other Escherichia species, Shigella sp., Citrobacter sp., Klebsiella pneumoniae, Enterobacter sp., and Salmonella sp. Taking into account all the investigated sequence classes analyzed and assumed exclusion criteria, 111 sequences from the O:157:H7 Sakai genome and 25 sequences from the UTI89 genome sequences were selected for analysis in terms of their specificity in relation to the ExPEC pathotype or the IPEC pathotype (Supplementary Figure S5). Five sequences typical for IPEC pathotype strain genomes were found (Supplementary Figure S5, tab SAKAI, marked green) and two sequences specific for ExPEC genomes (Supplementary Figure S5, tab UTI89, marked green). We excluded the sequences UTI-98-1111, UTI-98-1112, UTI-98-1113, UTI-98-1114, UTI-98-1115, and UTI-89-1116 (Supplementary Figure S5, tab UTI89, marked yellow) from potential candidates because they showed homology to the E. coli O18ac:H14 genome. This is probably representative of the NMEC [16] sub-pathotype, but we did not find sufficient confirmation for this particular H serotype. These results are summarized in Table 3. The following IPEC genomes, based on unique E. coli O:157:H7 str Sakai sequence fragments, were identified by the BLAST N analysis: E. coli O157:H7 str Sakai (EHEC), E. coli O157:H7 (EHEC), E. coli O157:H7 str F8092B (EHEC), E. coli O157 (EHEC), E. coli O157:H7 str SS52 (EHEC), E. coli O157:H7 str EDL933 (EHEC), E. coli O157:H7 str SS17 (EHEC), E. coli O157:H- (EHEC), E. coli O157:H7 str TW14359 (EHEC), E. coli O157:H7 str EC4115 (EHEC), E. coli Xuzhou21 (EHEC), E. coli O55:H7 (EPEC), E. coli O55:H7 str RM12579 (EPEC), E. coli O55:H7 str CB9615 (EPEC), E. coli O145:H28 (STEC), E. coli O145:NM (STEC), E. coli O145 str RM9872 (STEC), E. coli O145:H28 str RM12581 (STEC), E. coli O145:H28 str RM12761 (STEC), E. coli O145:H28 str RM13514 (STEC), E. coli O145:H28 str RM13516 (STEC), E. coli O145 (STEC), E. coli O26 str RM10386 (EHEC), E. coli O26 str RM8426 (EHEC), E. coli O26:H11 str 11368 (EHEC), E. coli O26:H11 (EHEC), E. coli O39:NM str F8704-2 (ETEC), E. coli O45:H2 (STEC), E. coli O103 str RM8385 (EHEC), E. coli O103:H2 str 12009 (EHEC), E. coli O103:H2 (EHEC), E. coli O111:NM (STEC), E. coli O111:H- str 11128 (STEC), E. coli O111:H- (STEC), E. coli O111 str RM9322 (STEC), E. coli O121 (STEC), E. coli O121:H19 (STEC), E. coli O121 str RM8352 (STEC), E. coli O158:H23 (EPEC), E. coli O169:H41 (ETEC), E. coli E110019 (EPEC), E. coli UMNK88 (ETEC), Escherichia sp E4742 (EHEC, Clade II). The following ExPEC genomes based on unique E. coli UTI89 sequence fragments were identified by BLASTN analysis: E. coli UTI89 (UPEC), E. coli NU14 (UPEC), E. coli O25:H1 (UPEC), E. coli 536 (UPEC), E. coli ABU 83972 (UPEC ABU), E. coli UM146 (AIEC/UPEC) and E. coli RS218 (NMEC). Therefore, these fragments can be used to develop multiplex-PCR kits or dot-blot hybridization systems that distinguish both ExPEC and IPEC and also, to some extent, allow differentiation within the pathotype. TRS-omix extracts all sequences between TRS elements that cover entire genomes (Supplementary Figures S1–S3). The very restrictive exclusion criteria that we applied allowed us to isolate seven specific DNA fragments (Table 3). One may wonder why only some of the unique sequence regions we selected contained virulence factors typical of the analyzed sub-pathotypes. As we show in Table 4, this is due to the fact that such sequences also show partial similarity to elements of the genomes of other microorganisms. Many new genomes of various organisms are sequenced every day, generating vast amounts of data. Growing datasets require new, good, fast, easy-to-use, and widely available genome-wide analysis tools. In our paper, a new informatics tool called TRS-omix is presented. It enables genome mapping and extraction of sequences of nucleotides from a genome using specific types of microsatellites known in all genomes, called TRS motifs (included in the flanking sequence). This type of microsatellite was dictated by the fact that they occur more frequently in the studied genomes and are mainly grouped in coding regions [11]. To our knowledge, no software has yet been developed to analyze genomic sequences in this way. The software allows for genomic analysis of linear and circular structures (with a given range or any length of sequence fragments), which offers a wide range of applications and universal use, e.g., prokaryotic and eukaryotic genomes, human chromosomes, etc. In our research, we have shown that conducting a further study using TRS-omix can lead to the elaboration of a new method of differentiating organisms based on complex systems of TRS motifs. The tool mentioned above can take a fresh look at the genome in the category of a unique arrangement of TRS motifs characteristic of different organisms belonging to an evolutionary group. The approach mentioned above is considered to search for similarities and differences within the sequences contained between TRS motifs for different genomes, which can be significant, while determining specific genetic features of living organisms. In our comparative analysis of two E. coli genomes, we have shown that fragments of genomic sequences that are either unique to the ExPEC pathotype or unique to the IPEC pathotype can be found. We also see the potential of using such sequence sets to differentiate genomes/strains within these pathotypes. We propose strict criteria to distinguish the two selected genomes in our new genetic approach to differentiating IPEC and ExPEC pathotypes (Section Results, Stage 3), as well as the criteria for the elimination of homologous sequences in genomes available in databases and not being sequences belonging only to the group of ExPEC or IPEC pathogens (Section Results, Stage 4). We excluded, among others, a pool of sequences that was specific for the ExPEC pathotype, but also for the AIEC sub-pathotype, which, despite genetic similarity to the UPEC sub-pathotype, nevertheless belongs to the IPEC pathotype [17,18]. This resulted in leaving only three regions characteristic of the IPEC pathotype and two such regions characteristic of the ExPEC pathotype. For E. coli O157: H7 str Sakai genome, the unique regions identified for IPEC, as derived by automated computational analysis using the gene prediction method: the homology of proteins (Vector NTI 11,5), are characterized as follows: the SAK-687 region (1,390,007–1,390,725 bp) partially overlaps the sequence of the CDS 340 gene (hypothetical protein; ncbi: protein/BAB34742.1) and the ureD gene (encoding urease accessory protein D; ncbi: protein/BAB34744.2), the region SAK-937-939 (1,937,664–1,938,325 bp) lies within CDS 626 (encoding hypothetical protein; ncbi: protein/BAB35378.1), and the SAK-1032 region (2,147,832–2,147,944 bp) lies inside the ftrA gene (encoding transcriptional regulator; ncbi: protein/BAB35566.2. We have not identified any genes that are known markers of virulence for this group of pathogens. As we showed in Table 4, the sequences of these genes are similar to some extent to those of many E. coli strains. However, the identified regions are, in our opinion, good markers of the IPEC sub-pathotype since they allowed the identification of 19 genomes of the EHEC sub-pathotype, 16 genomes of the STEC sub-pathotype, five genomes belonging to the EPEC sub-pathotype, and three genomes belonging to the ETEC sub-pathotype. Similarly, in the case of the E. coli UTI89 genome, the unique regions identified for the ExPEC/UPEC genomes, as derived by automated computational analysis using gene the prediction method, Protein homology (Vector NTI 11,5), are characterized as follows: the UTI89-2514 region (4,810,579–4,811,161 bp) partially overlaps the CDS sequence 2305 (encoding hypothetical protein; ncbi: protein/WP_001304629.1) and contains the gene encoding urease accessory protein; ncbi: protein/WP_001335216.1 and the UTI89-2517 region (4,818,430–4,818,864 bp), which partially overlaps the cnf-1 gene (protein/WP_000528123.1), genes associated with the pathogenicity of the UPEC sub-pathotype. These sequences allowed the identification of five UPEC genomes (including one genome of ABU strain), one AIEC genome, but with the confirmed high similarity to the uropathogenic UPEC strain (E. coli UM146) [19], and one genome belonging to the NMEC sub-pathotype. We also did not identify most of the typical virulence factors of the ExPEC pathotype, which indicates partial similarity of such genomic sequences to elements of genomes of other E. coli strains. In this work, we did not conduct this research in more detail. We focused only on using comparative genomic analyses and finding such elements that differentiate the genomes of the two groups of very similar pathogens that can be used as markers of a specific E. coli pathotype. Based on the results shown in Table 3, it is possible to design appropriate genetic markers that will enable the differentiation described above by the multiplex-PCR or dot blot hybridization reaction. The main effect of our interdisciplinary work is, on the one hand, the development of software that allows scientists to automate their research. On the other hand, it enabled the search for new markers differentiating closely related microorganisms. One of the most important functionalities of this software is to support the extraction of sequences between the TRS motifs, determining the position on the genome and some other information necessary for a new genetic approach to the differentiation of IPEC and ExPEC pathotypes. In further studies of the new approach to the deep differentiation of microorganisms, we will try to generalize our method for analyzing entire families of organisms using machine learning or various statistical procedures. This line of research is very promising for potential clinical applications but requires further research in this area from the point of view of biology and computer science. At the present stage, TRS-omix software is not adapted to the parallel study of thousands of genomes simultaneously, and its scaling is an interesting line of research in this area. In the presented form, TRS-omix allows searching for DNA sequences that may be specific to a particular group of microorganisms; in our case, for the ExPEC or IPEC, two approaches can be used. The first is to subject many genomes from each of these groups to subsequent analyses using TRS-omix and then to check by BLAST N which fragments occur only in one of them and additionally meet the other defined criteria. The second way is to use two genomes for analyses using TRS-omix, one representative for the ExPEC group and one representative for the IPEC group, and then use BLAST N to check which fragments occur only in the genomes of ExPEC strains or only in the genomes of IPEC strains and meet the other defined conditions. In our research, we chose the second approach, as illustrated in Figure 4 and described in Section 2.3. A general approach for differentiation of organisms with the use of TRS-omix. At present, TRS-omix is a tool supporting the work of researchers, and its mode of operation is sufficient for genetics. Analysis using TRS-omix makes it possible to perform calculations on a personal computer. It seems to us that this is the advantage of TRS-omix because it may be available to many geneticists, for whom this software will enable genome analysis and cheap research in the field of comparing different genomes. We can imagine scalable TRS-omix software computing in big data environments like Apache Hadoop or Apache Spark. However, this approach requires designing the architecture of calculations on individual computing nodes and owning or renting servers with such infrastructure. Using big data technology would enable horizontal scaling (scale-out/scale-in), i.e., adding/removing subsequent machines, and vertical scaling (scale-up/scale-down), i.e., increasing/decreasing resources within one machine. The TRS-omix is a powerful tool that allows a specific way to compare genomes, particularly those sequences that lie between the trinucleotide repeats. Therefore, it may be an interesting and faster alternative to comparing whole genomes, indicating that these classes of sequences are typical for certain groups of microorganisms. The TRS-omix search engine was implemented with the use of a GNU compiler collection called gcc, compliant with ISO 2019 of language C in the programmers’ environment, bearing the name Code::Blocks (release 20.03) under license from GNU. We note that it was developed on the basis of the imperative programming paradigm with the use of the language C, which is dedicated to low level programming [20]. It is worth underlining that the use of language C enables integration of the search engine with other software tools made in popular programming languages such as language R, or Python. In the code architecture of the TRS-omix search engine, data structures corresponding to single-dimensional arrays (linked list) and two-dimensional arrays, which were elaborated based on dynamic memory allocation [21]—the memory was dynamically allocated by the malloc() function of language C. In the case of Linux and Windows operating systems, in order to execute the search of the TRS motifs, it needs inserting into the following in one directory: software file (file name: TRS-omix.exe), defining a file of the class of TRS motifs (file name: trs.txt), and a file for analysis (sequence.fasta). In the case of the Linux operating system, we used only the computational server with limited memory access to 8 GB of the Faculty of Mathematics and Computer Science University of Lodz with the following parameters: CPU: 2× AMD EPYC 7302—3 GHz (3.3 GHz Turbo), Cores/Threads: 32C/64T RAM: 128 GB (4× 32 GB) Hard drive: 2× 2.4 TB (Seagate)—RAID1, Operating system: Debian GNU/Linux 10 (buster). The experimental running tests of the TRS-omix software were carried out on the computer with the following parameters: CPU: Intel Core i7 CPU 860 @ 2.80 GHz, RAM: 8.0 GB RAM, Hard drive: 1 TB., operating system: Windows 10 Professional. In particular, the following genomes were subjected to analysis: Escherichia coli UTI89 strain (Acc. no NC_007946.1), a representative of the pathotype ExpEC (Extraintestinal Pathogenic E. coli), sub-pathotype UPEC (Uropathogenic/Extraintestinal Pathogenic E. coli), and O157:H7 Str. Sakai (Acc. no BA000007.3), a representative of pathotype IPEC (Intestinal Pathogenic E. coli), sub-pathotype EHEC (Enterohemorrhagic E. coli). The Vector NTI 11.5 software using the CLUSTAL W algorithm was used to compare the level of similarity of the sequences extracted from the two investigated genomes by TRS-omix and determine which of these sequences could be pre-qualified unique to a given genome. This software was also used to search for sequences present in a given genome that TRS-omix could not extract because of a mutation within the TRS (imperfect TRS) or because a given complementary sequence contained an internal TRS motif and was therefore assigned to other classes. The NCBI platform https://blast.ncbi.nlm.nih.gov/Blast.cgi was used to evaluate the available microbial genomes to determine whether a given genomic DNA fragment was specific only for the genome group of the ExPEC or IPEC pathotype. The default settings were used for the analysis, with the exception of the maximum number of hits set to 1000. (1) TRS means a sequence of three nucleotides, e.g., CCG. (2) TRS motif means a sequence of nucleotides, in which there occurs a triple repetition (directly one after another) of the same TRS, e.g., CCGCCGCCG. (3) Class of TRS motifs means TRS motif occurring in one line in the file trs.txt, each of which is preceded by the sign „#”, e.g.,: #CCGCCGCCG#CGCCGCCGC#GCCGCCGCC. (4) Number of class of TRS motifs means a natural number (greater than 0), which corresponds to the number of the line in the file trs.txt. (5) Flanking sequence means a sequence of nucleotides in which there occurs at least a triple repetition (directly one after another) of the same TRS. (6) Extracted sequence means a sequence of nucleotides (SEQ) that is found between two flanking sequences that consists of at least one nucleotide and is not a flanking sequence. (7) Left flanking sequence (LSF) means a flanking sequence which is found at the site 5′ of an extracted sequence. (8) Right flanking sequence (RSF) means a flanking sequence which is found at the site 3′ of an extracted sequence.4.4. Using TRS-omix on the example of selected genomes The TRS-omix software works with the use of files formed in the FASTA format, which is applied in bioinformatics to record nucleotide sequences representing information on the genome of living organisms and also amino acids in proteins. This offers the possibility of performing analyses with the use of an open-access genome database, e.g., GenBank. Further in this Chapter, we will characterize input files, output files, and software options of the TRS-omix software. The file called TRS-omix.exe is a workable one of the TRS-omix search engine. It accepts two input files (sequence.fasta, trs.txt) and creates one output file (interiors.txt). The input files should be placed in the same directory as the TRS-omix software. A similar case offers when it comes to the output file—it is formed in the same directory, in which the TRS-omix software finds itself. The file called sequence.fasta contains a genome of the examined organism, with the use of which the TRS-omix search engine searches for microsatellites of the trinucleotide repeats type, and also extracts sequences between such trinucleotide repeat and executes initial analyses of the genome. Let us note that in a file that is downloaded, in an exemplary fashion, from GenBank, it needs merely to change the name of the file into that of “sequence”, but we do not alter the type of file (the type of file is still FASTA). The file trs.txt contains classified TRS motifs—each line includes TRS motifs preceded by the sign “#”. One such line is identified as one considered class of TRS motifs. In a similar sense, the file called trs.txt is treated as a file with a set of search rules in files of the FASTA type. The file interiors.txt contains information on the positions of flanking sequences and also about those and the very extracted sequences themselves. The first line of the file includes headings of 14 columns, while the following lines contain relevant data. The line including the headlines of the columns was formatted in the following way: L-NoClass;L-No;LFS;Len(LFS);L-POS(LFS);R-POS(LFS);R-NoClass;R-No;RFS;Len(RFS); L-POS(RFS); R-POS(RFS);>SEQ;Len(SEQ) where: L-NoClass—denotes the number of the class of TRS motifs from the file trs.txt for the left flanking sequence, L-No—denotes the number of TRS motifs from the file trs.txt for the left flanking sequence, LFS—denotes the left flanking sequence; Len(LFS)—denotes the number of nucleotides of the left flanking sequence, L-POS(LFS)—denotes the position from which the left flanking sequence begins in the genome, R-POS(LFS)—denotes the position at which the left flanking sequence ends in the genome, R-NoClass—denotes the number of the class of TRS motifs from the file trs.txt for the right flanking sequence, R-No—denotes the number of the TRS motif from the file trs.txt for the right flanking sequence, RSF—denotes the right flanking sequence, Len(RFS)—denotes the number of nucleotides of the right flanking sequence, L-POS(RFS)—denotes the position from which right flanking sequence starts in the genome, R-POS(RFS)—denotes the position at which the right flanking sequence ends in the genome, >—denotes the place from which the extracted sequence begins. SEQ—denotes the extracted sequence, Len(SEQ)—denotes the number of nucleotides of the extracted sequence. Between the elements indicated above there are found semicolons that constitute separators with the omission of a semicolon between the sign “>” and an extracted sequence. Figure 3 illustrates the recording of data in the file interiors.txt. The results of the flanking sequence searched in the file interiors.txt are recorded in ascending order. When starting the executable file of TRS-omix, there appear on the computer screen two options which are possible to select: Analysis of the linear case with conditions: use of this option enables to search TRS motifs in linear genomes with additional search conditions: giving the minimal (Min) and maximal (Max) length of the searched sequence found between the flanking sequence. Analysis of the circular case with conditions: use of this option enables searching TRS motifs in circle genomes with analogous additional search conditions like in the case of analysis of the linear case with conditions. The results of our studies have shown that it is a new approach to genome-wide analysis using a new bioinformatics tool. Furthermore, it has been discovered that it is possible to use the proposed TRS-omix search engine to find new genetic markers that differentiate IPEC and ExPEC pathotypes. Using this IT tool, we are able to analyze virtually any group of genomes, from viral, through pro- and eukaryotic, to human chromosomes. The paper presents an efficient search engine for TRS motifs and shows the use of TRS-omix to find new genetic markers according to a novel approach to differentiating pathotypes. The new approach to genome analysis presented in the paper allows one to look at the nucleotide sequences in FASTA files from the point of view of the mapped TRS motif sequences, which enables comparison of the genomes. In addition, the work includes detailed tests of the presented search engine, and also shows an example of an analysis using this bioinformatics tool. The presented informatics tool—TRS-omix—provides a reliable basis for new research on the phylogeny, diagnostics, and epidemiology of organisms. It offers a new look at the genome through the prism of analyses of TRS profiling, which can find application in many areas of biology and medicine in the future. The tool allows—on the one hand—to carry out analyses using files from open-access bioinformatics databases, such as GenBank, which presently offers a massive amount of data (big data). On the other hand, the tool was created using a low-level language, making its integration with other data science software and tools possible.

PMC10002603

Christine Swinka,Eva Hellmann,Paul Zwack,Ramya Banda,Aaron M. Rashotte,Alexander Heyl

Cytokinin Response Factor 9 Represses Cytokinin Responses in Flower Development

22-02-2023

cytokinin,Cytokinin Response Factor,transcription,repressor,flower development

A multi-step phosphorelay system is the main conduit of cytokinin signal transduction. However, several groups of additional factors that also play a role in this signaling pathway have been found—among them the Cytokinin Response Factors (CRFs). In a genetic screen, CRF9 was identified as a regulator of the transcriptional cytokinin response. It is mainly expressed in flowers. Mutational analysis indicates that CRF9 plays a role in the transition from vegetative to reproductive growth and silique development. The CRF9 protein is localized in the nucleus and functions as a transcriptional repressor of Arabidopsis Response Regulator 6 (ARR6)—a primary response gene for cytokinin signaling. The experimental data suggest that CRF9 functions as a repressor of cytokinin during reproductive development.

Cytokinin Response Factor 9 Represses Cytokinin Responses in Flower Development A multi-step phosphorelay system is the main conduit of cytokinin signal transduction. However, several groups of additional factors that also play a role in this signaling pathway have been found—among them the Cytokinin Response Factors (CRFs). In a genetic screen, CRF9 was identified as a regulator of the transcriptional cytokinin response. It is mainly expressed in flowers. Mutational analysis indicates that CRF9 plays a role in the transition from vegetative to reproductive growth and silique development. The CRF9 protein is localized in the nucleus and functions as a transcriptional repressor of Arabidopsis Response Regulator 6 (ARR6)—a primary response gene for cytokinin signaling. The experimental data suggest that CRF9 functions as a repressor of cytokinin during reproductive development. The plant hormone cytokinin regulates numerous developmental processes throughout the plant, from early events in embryogenesis to reproductive development and leaf senescence [1,2,3,4,5]. The Two-Component Signaling pathway (TCS) is a modified version of the bacterial two-component system consisting of receptor histidine kinases, phosphotransfer proteins, and two classes of Response Regulators (RRs) [5,6,7,8,9]. Type-B RRs (RRBs) are transcription factors that positively regulate responses to cytokinin, while type-A RRs lack DNA binding activity and act as negative regulators [6,10]. Although these key components of the TCS pathway have been well characterized in Arabidopsis and other species, many of the downstream regulatory mechanisms of the cytokinin responses remain unclear [3,8,10]. A small subgroup of AP2/ERF family transcription factors, known as Cytokinin Response Factors or CRFs, has been shown to be involved in the regulation of cytokinin signaling and response [11,12,13,14,15]. CRFs have been shown to act in concert with RRBs in the transcriptional regulation of cytokinin-responsive genes and are involved in protein–protein interactions with components of the primary signal transduction pathway [11,16,17]. Functional characterizations of CRFs have implicated these transcription factors in the regulation of multiple processes related to shoot growth and development including leaf and cotyledon expansion, leaf vascular patterning, and senescence [11,17,18]. There have also been connections between CRFs and an array of abiotic stress responses in addition to development, which was reviewed in [12,13,14,19]. These are in line with similar findings that have generally connected cytokinin to different stress responses [20,21]. The CRFs were first identified in a screen for cytokinin-responsive genes in Arabidopsis as related AP2/ERF transcription factor genes [13]. These were later, named and specifically described as CRFs, including three CRFs (CRFs 2, 5, and 6) that are members of a group of genes that show strong transcriptional induction by cytokinin treatment [11,12,22,23]. Additional members of the CRF group (Clade VI, VI-L—ERFs) were identified later, yielding a total of 12 CRFs in Arabidopsis (CRF1-12) [11,12,17,24,25]. Examinations of CRF protein and gene sequences across Angiosperm species have revealed that the CRF group can be generally spilt into five distinct phylogenetic clades (I to V) [12,17]. In nearly every diploid Angiosperm species examined, there are either 1 or 2 genes found per CRF-Clade, with expected increases in polyploid plants (i.e., 2–4 in a tetraploid). However, Clade V CRF groups routinely have 3–4 genes in diploid species examined. Primary examples of species where there have been full examinations of CRFs include tomato (3 Clade V genes: SlCRF9, 10, 11) and Arabidopsis (Four Clade V genes: AtCRF9, 10, 11, 12). Further evidence from the gathering of additional Angiosperm CRF sequence phylogenetic examinations, suggests that Clade V should likely be split into two distinct clades or at least subclade Clade Va and Vb, with AtCRF9, SlCRF9, and their orthologs in one group Va and the other Clade V members in a distinct group Vb, AtCRF10, 11, 12, and SlCRF10, 11 [18,26]. In a genetic screen to identify novel transcriptional regulators of the cytokinin response in Arabidopsis, CRF9 was found to regulate the transcription of the reporter gene ARR6 [27]. This study aimed to functionally characterize CRF9. It was demonstrated that this gene acts to repress cytokinin responses. Additionally, we show that CRF9 is a direct regulator of the ARR6 expression. While the phylogeny of CRFs is well established in modern land plants [25], the origin of this protein family is not clear. To gain some insights, a phylogenetic analysis targeting early land plants and algae was conducted. The results show that all known CRFs from the Arabidopsis clade in one branch are clearly different from sequences of most proteins (Figure 1). Within this branch of the tree, there is also one sequence from M. polymorpha and four from P. patens, the two most early diverging land plants in this analysis. Interestingly, the only sequence of Charophyceae algae, C. braunii, is also found in this branch. All other sequences are distinct and probably represent members of the AP2/ERF family, which are not CRFs. No members of the CRF family were found in Chlorophyceae algae or the other Charophyceae algae included in this analysis. To learn about the biological function of CRF9, the expression of the gene was analyzed using a promoter–GUS construct (Figure 2A). This construct used the whole CRF9 upstream region and the 3′UTR of the next gene upstream AT1G49110 (in a total length of 1180 bp). The results showed GUS expression in the flower, primarily the anthers with a small amount of expression in the leaf mid-vein at 27 DAG (Figure 2B). Publicity available RNA-seq data confirmed that the expression pattern of CRF9 expression was primarily localized to RNA samples collected from flowers and pollen (Figure 2C,D). In summary, the expression analysis using GUS and RNA seq data demonstrates CRF9 expression mainly in the reproductive tissues. The bioinformatics analysis places CRF9 in the AP/ERF transcription factor group [17]. Thus, we wanted to test if the protein indeed localizes to the nucleus. CRF9 N- and C-terminal GFP fusion constructs, driven by the 35S promoter, were made and used to transiently transform tobacco leaf cells. After a 48 h incubation period, the samples were analyzed by confocal microscopy. For both CRF9 GFP fusion proteins, strong signals exclusively in the nucleus were detected (Figure 3). The nuclear localization was later confirmed by a DAPI staining of the nucleus. To investigate the function of a gene in a given organism, the gene of interest can be either mutated or overexpressed. We chose both approaches and analyzed the phenotypes of a T-DNA insertion mutant, and four independent overexpressor lines (# 1.17; 17.5; 24.3; 28.8), using the 35S promoter to drive CRF9 expression. The T-DNA insert of the knock-out mutant (crf9-2) was located at the C-terminal end of the gene downstream of the functional domains but within the Clade V C-terminal specific region. Surprisingly, qRT PCR revealed even higher levels of CRF9 in this mutant line compared to the wild type. The same analysis demonstrated very high levels of CRF9 transcripts in the overexpressor lines tested (Figure 4A). The crf9-2 line showed a similar growth pattern as seen for wild-type plants. However, plants of the CRF9 ox lines were to a varying degree smaller than the wild type (Figure 4B,C). Two of the overexpressor lines, 17.5 and 24.3, also had fewer and smaller leaves. The other two lines investigated showed a similar leaf size, but higher leaf number compared to the wild-type plants (Figure 4D). Next, the effect of CRF9 on root growth was evaluated. Again, the crf9-2 line did not significantly differ in root elongation from the wild type. In contrast, the roots of the CRF9 overexpressor lines were significantly shorter than the wild-type, seemingly independent of the level of CRF9 overexpression (Figure 5). CRF9 overexpression led to smaller plants compared to the wild-type, while the crf9-2 line was very similar to the wild-type. After analyzing the effects of CRF9 on vegetative development, we looked at the reproductive traits. The time to floral transition was determined in all plant lines and was defined as the age that the stem reached a height of longer than 1 cm. In both the wild-type and the crf9-2 lines, the floral transition started 24 days after germination (DAG), and all plants had transitioned from vegetative to reproductive growth at 30 DAG. In contrast, the CRF9-overexpressing lines also started flowering or bolting at 24–26 DAG, but the transition took a longer time, as only at 36 DAG all plants had started their reproductive growth (Figure 6). After fertilization, the resulting siliques of the CRF9-overexpressing lines were significantly shorter than those of wild-type and the crf9-2 plants—with the strongest phenotype overserved again in lines 17.5 and 24.3 (Figure 7A,B). These reduced silique lengths resulted in lower seed density (seeds/mm of silique) as well as fewer seeds in total. The greatest reduction was detected in lines 17.5 and 24.3 with a gradual decrease in seed number in the other two overexpressing lines and no detectable difference between crf9-2 and the wild-type plants (Figure 7C,D). However, no difference was detected concerning the size of the seeds themselves. In summary, CRF9-overexpressing lines showed a later transition from vegetative to reproductive growth and their siliques were smaller and had a lower seed density. Only some CRFs have been shown to be transcriptionally regulated by cytokinin and affect cytokinin response in plants [12]. Thus, the effect of CRF9 on the response to cytokinin was examined using standard bioassays. First, the effect of cytokinin on root elongation was tested. Cytokinin treatment did not lead to a detectable difference between the wild-type and crf9-2 plants. In contrast, all the CRF9-overexpressing lines showed a significant reduction in root elongation after cytokinin treatment (Figure 8A). A similar picture emerged in a second, well-established cytokinin bioassay, the chlorophyll retention assay. The overexpressing lines showed a significant difference in chlorophyll content compared to the wild type (Figure 8B). In summary, the CRF9 overexpression led to detectable differences in long-term cytokinin response assays in both leaves and roots. After detecting the effect of CRF9 overexpression on the cytokinin response on the whole plant level, the effects of the hormone on the molecular level were also investigated. Since CRF9 is a member of the CRF transcription factor family, we looked at its effect on the expression of the well-characterized cytokinin response gene ARR6, a member of the RRA family using two different assays. In a protoplast transactivation assay (PTA) the effect of the expression of CRF9 on the ARR6 promoter was tested via a coupled Luc reporter gene [27]. The co-expression of CRF9 led to a decrease in the ARR6 promoter activity compared to the vector control, which would be consistent with CRF9 functioning as a negative regulator of the molecular cytokinin response (Figure 9A). A similar picture emerged when the amount of ARR6 transcripts were examined in the different plant lines. The crf9-2 mutant lines and lines 1.17 and 28.8 displayed only weak or no reduction of the transcript of the primary response gene. In contrast, the strong CRF9 overexpressor lines 17.5 and 24.3 displayed a dramatic reduction of ARR6 mRNA (Figure 9B). Thus, on the molecular level, CRF9 acts as a negative regulator of at least one of the primary cytokinin response genes. It has long been known that the cytokinin response is mediated to a large extent by changes in transcriptional patterns [6,11,22,23,28,29]. RRBs and CRFs are the main regulators facilitating these changes in the transcriptome [12,16]. Since their original identification in Arabidopsis more than a decade ago, there have been several studies of some CRF genes as to their roles in development as well as cytokinin response, and environmental response in several different species [12,30,31]. Despite these efforts, there is still much that is not known about this group, and some phylogenetic clades of CRFs are basically unexamined. One such CRF, CRF9 or At1g49120 (a Clade V CRF), which had not been studied in detail before, was identified in a genetic screen for transcriptional regulators [27]. This study aimed to functionally characterize this transcription factor. One of the best ways to study the function of any gene is to look for mutations. These can either be loss-of-function or gain-of-function mutations. In this study, both types of mutations have been utilized. In the different experiments shown here, the crf9-2 mutant phenotype was not different from that of the wild type. This might be due to either redundancy among the CRFs or due to the allele that we used not leading to a loss of CRF9 function. Only one prior report examined any CRF9 mutant phenotype [17]. While that study did find some minor changes in leaf vascular patterning, no other phenotypes were reported, and a different T-DNA insertion mutant (crf9-1) was utilized. Loss-of-function mutations are often seen as more effective in characterizing the function of a gene. However, most modern land plants have undergone several rounds of whole genome duplications and consequentially show a high-level gene redundancy [32]. This high level of redundancy is also true for cytokinin-regulated transcription factors [33]. This could be a reason for the similarity of the phenotypes of crf9-2 and wild-type plants. Another reason might be due to the position of the T-DNA insertion, downstream of the functional domains, or due to the level of CRF9 transcript not being reduced but rather increased by the T-DNA insert (Figure 4A). In contrast, the overexpressor lines investigated displayed a distinctive phenotype (Figure 4). They had smaller or more leaves than the wild type, which would be consistent with the role of transcriptional repressor as a similar phenotype has been reported for ARR1-SDRX—a dominant repressor of the RRBs [34,35]. In addition, the later transition from vegetative to reproductive growth, which was detected in the overexpressor lines, has been reported before for ARR1-SRDX plants [34]. While those results are informative about the general function, the expression analysis showed the strongest expression for CRF9 in the inflorescences and, thus, a biological function for the proteins is most likely to be found there. Interestingly, CRF9 overexpressor lines showed significantly smaller siliques and a lower density (seeds/length of the silique) than the wild type (Figure 7), indicating a function for CRF9 as a negative regulator for inflorescences development. This would be comparable to what has been reported for ARR1-SRDX plants [34]. The opposite phenotype was reported when the cytokinin content was increased, e.g., by mutating cytokinin metabolizing enzymes (CKX), as the siliques of those plants become longer [35,36]. So how does CRF9 function on the molecular level? It was identified in a genetic screen to be a repressor of the primary cytokinin response gene ARR6 [27]. The detected nuclear localization would be consistent with a function as a transcriptional regulator (Figure 3). The protoplast transactivation assay (PTA) has long been used to investigate the function of putative transcription factors involved in cytokinin signaling [6,27,34]. Our PTA experiment validates that CRF9 regulates ARR6 as a repressor (Figure 9A). Furthermore, the expression analysis of ARR6 in the different CRF9 overexpressor lines confirmed a role for CRF9 as a negative regulator of this cytokinin primary response gene. Interestingly, other CRFs (CRF2, 3, and 6) have been described as directly targeting and inducing the auxin transport protein PIN1, which also has been linked to female reproductive organ developmental processes [37,38]. The phylogenetic analysis conducted in this study focused on the early diverging land species and green algae (Figure 1). The CRFs of Arabidopsis can be roughly separated into two groups (CRF1-CRF8 and CRF9-CRF12) as noted before [25]. The second group (B-clade) was further defined as CRF Clade V in a more detailed examination of CRFs [17]. Clade V seems to be specific for modern land plants as the CRFs from early diverging land plants and algae only clade with CRF1-CRF8. Given that CRF9 seems to play a role in flower development and flowers are an evolutionary novelty of modern land plants, it is tempting to speculate that the CRF9 branch is a result of an early whole genome duplication event and was neo-functionalized to function in the regulation of reproductive growth. However, more research is needed to support this hypothesis. The protein CRF9 was used as a query in HMMR in Plant Ensembl (plants.ensembl.org) [39] to identify homologs in different plant species (Ostreococcucs taurii, Chlamydomonas rheinhardtii, Klebsormodium nitans, Chara braunii, Mesotigma viride, Marchantia polymorpha, Physcomitrium patens, Selaginella moelendorfii). All sequences that had an E-value of >1−10 and did not start with a Met were eliminated. All remaining sequences were checked for the presence of an AP2/ERF domain using Interpro (https://www.ebi.ac.uk/interpro, accessed on 21 December 2022) [40]. An alignment of the sequences was performed using MAFFT (www.ebi.ac.uk/Tools/msa/mafft, accessed on 21 December 2022) in the default setting [41]. The resulting alignment was imported into MEGA XI (www.megasoftware.net, accessed on 21 December 2022) [42]. A Maximum Likelihood tree was calculated using the JTT + G + I + F substitution setting. The resulting tree was evaluated using 200 bootstrap repetitions. Plants of the Columbia (Col-0) ecotype of Arabidopsis thaliana and Nicotiana benthamiana were used. For the analysis of the CRF9 (At1g49120) function the mutant line GABI-Kat GK-351-H05, crf9-2 was used [43]. Plants for the experiments were grown under long-day conditions in the greenhouse or a climate chamber at 21 °C. To generate transgenic lines for different purposes, the desired sequences were shuttled into vectors indicated using the GATEWAY system (InvitrogenTM). Wild-type plants from Col-0 were transfected by floral dip [44]. All used plant lines were homozygous for the T-DNA insertion. For overexpression studies, CRF9 was cloned into the vector pB2GW7 and for reporter studies, the promoter (1180 bp) of CRF9 was shuttled into the pBGWFS7 vector. For subcellular localization studies, CRF9 was shuttled into the vector pB7WGF2 and pB7FWG2 to generate N- and C-terminal fusion proteins with GFP using the GATEWAYTM system. Plants from N. benthamiana were infiltrated with an agrobacteria solution containing the constructs for transient protein expression. Plants were incubated in the greenhouse for 2–3 days. For DAPI staining [45] leaf disks were punched out, put in a DAPI staining solution and vacuumed, followed by a 15 min incubation at 37 °C. The fluorescent signal from both DAPI and GFP was detected using a confocal microscope. Arabidopsis seeds were surface sterilized with sodium hypochlorite solution and vertically grown on 0.5× MS medium on square plates. For cytokinin treatment, a medium containing 100 nM BA was used. Plates with an equal DMSO supplement served as a control. The position of the root tip was marked on day three and day ten after germination. The root elongation was determined using the program Image J (https://imagej.nih.gov/ij/index.html, accessed on 21 December 2022). The respective plant tissues were stained with GUS staining solution as published before [46] and incubated overnight at 37 °C. After discarding the staining solution, the tissues were destained with 70% ethanol. The ethanol step was repeated several times until the tissue was cleared. Leaf three and leaf four of 21 days old plants were analyzed for their chlorophyll content as described previously [47]. For lines #17.5 and #24.3, leaves five to eight were analyzed because leaves three and four already showed the early beginning of senescence. The initial chlorophyll content and the content after dark incubation were measured. The leaves were incubated in the dark for six days in an MES buffer containing 1 µM BA or DMSO as control. Three repetitions were performed. Arabidopsis plants from the ecotype Col-0 were grown in a growth chamber under short-day conditions (8 h light and 16 h dark) and low light at 21 °C. Protoplasts were isolated from leaves five to ten from six-week-old plants by using the Tape-Arabidopsis Sandwich method [48]. The transfection of the protoplasts was performed following the description from Yoo [49] with some modifications. Protoplasts were transfected with three plasmids expressing the specific reporter pARR6 2,4 kb::LUC [6], an effector and the transfection control reporter pUbi::GUS [50]. The plasmids were used at the ratio 5:4:1. The double amount of each component of the transfection step was used to obtain higher values at the end. For cytokinin treatment, the transfected protoplasts were incubated overnight in the dark at 23 °C for around 16 h in 1 mL WI solution containing 500 nM trans-Zeatin. After that, LUC activity was analyzed following the instructions of the manufacturer (Promega Kit E1500) with the following modifications. For measuring the luminescence 50 µL of the protoplast solution was mixed with 50 µL Luciferase-Agent in a black 96-well plate. To analyze the transfection efficiency of each sample, 10 µL protoplast solution was assayed with 100 µL MUG solution (1 mM MUG, 10 mM Tris ph 8.0, 2 mM MgCl2). A kinetic read over 15 min was performed and the slope of the values was calculated. The activities were calculated as relative LUC/GUS values. The protoplast transactivation assay was performed as described before [27]. For the analysis of the time point of transition from vegetative growth to reproduction growth, about 50 plants for each line were analyzed. The days as well as the number of leaves produced were counted once the stem reached 1 cm high. In summary, the results of our experiments indicate that CRF9 is a repressor of the cytokinin response in inflorescences. This would make this transcriptional regulator one of the few examples of negative regulators of the transcriptional response to cytokinin known today. However, further experiments are needed to determine to which extent CRF9 controls other primary cytokinin response genes using gene editing to create a CRF9 functional knock-out and RNA-sequencing to see the effect on the floral transcriptome.

PMC10002607

Luca Mignani,Jessica Guerra,Marzia Corli,Davide Capoferri,Marco Presta

Zebra-Sphinx: Modeling Sphingolipidoses in Zebrafish

01-03-2023

gene knockout,hereditary disease,lysosome,morpholino,sphingolipid,zebrafish

Sphingolipidoses are inborn errors of metabolism due to the pathogenic mutation of genes that encode for lysosomal enzymes, transporters, or enzyme cofactors that participate in the sphingolipid catabolism. They represent a subgroup of lysosomal storage diseases characterized by the gradual lysosomal accumulation of the substrate(s) of the defective proteins. The clinical presentation of patients affected by sphingolipid storage disorders ranges from a mild progression for some juvenile- or adult-onset forms to severe/fatal infantile forms. Despite significant therapeutic achievements, novel strategies are required at basic, clinical, and translational levels to improve patient outcomes. On these bases, the development of in vivo models is crucial for a better understanding of the pathogenesis of sphingolipidoses and for the development of efficacious therapeutic strategies. The teleost zebrafish (Danio rerio) has emerged as a useful platform to model several human genetic diseases owing to the high grade of genome conservation between human and zebrafish, combined with precise genome editing and the ease of manipulation. In addition, lipidomic studies have allowed the identification in zebrafish of all of the main classes of lipids present in mammals, supporting the possibility to model diseases of the lipidic metabolism in this animal species with the advantage of using mammalian lipid databases for data processing. This review highlights the use of zebrafish as an innovative model system to gain novel insights into the pathogenesis of sphingolipidoses, with possible implications for the identification of more efficacious therapeutic approaches.

Zebra-Sphinx: Modeling Sphingolipidoses in Zebrafish Sphingolipidoses are inborn errors of metabolism due to the pathogenic mutation of genes that encode for lysosomal enzymes, transporters, or enzyme cofactors that participate in the sphingolipid catabolism. They represent a subgroup of lysosomal storage diseases characterized by the gradual lysosomal accumulation of the substrate(s) of the defective proteins. The clinical presentation of patients affected by sphingolipid storage disorders ranges from a mild progression for some juvenile- or adult-onset forms to severe/fatal infantile forms. Despite significant therapeutic achievements, novel strategies are required at basic, clinical, and translational levels to improve patient outcomes. On these bases, the development of in vivo models is crucial for a better understanding of the pathogenesis of sphingolipidoses and for the development of efficacious therapeutic strategies. The teleost zebrafish (Danio rerio) has emerged as a useful platform to model several human genetic diseases owing to the high grade of genome conservation between human and zebrafish, combined with precise genome editing and the ease of manipulation. In addition, lipidomic studies have allowed the identification in zebrafish of all of the main classes of lipids present in mammals, supporting the possibility to model diseases of the lipidic metabolism in this animal species with the advantage of using mammalian lipid databases for data processing. This review highlights the use of zebrafish as an innovative model system to gain novel insights into the pathogenesis of sphingolipidoses, with possible implications for the identification of more efficacious therapeutic approaches. Sphingolipids were first described during the second half of the nineteenth century [1]. The term “sphingolipid” was coined based on the complexity and sphinxlike nature of this class of lipids characterized by a core long chain aliphatic amino alcohol (sphingoid base). The most common member is represented by sphingosine, which can be functionalized by a fatty acid condensed at its aminic moiety and by polar molecules at its hydroxyl terminus, including small organic molecules, amino acids, or carbohydrates [2] (Figure 1). De novo synthesis of sphingolipids begins in the endoplasmic reticulum (ER) and may move towards the Golgi apparatus. Finally, their mature forms are delivered to cell membranes [2,3,4]. Sphingolipids play a key structural role in cellular membranes and/or act as signaling molecules. Owing to their molecular structure, sphingolipids can organize within plasma membranes into ordered focal regions named lipid rafts, crucial for the arrangement of raftophilic molecules or transmembrane protein domains [5]. During recycling or after signaling events, sphingolipids may reach the lysosomes, where specific enzymes catabolize them to less-complex molecules, which can enter different metabolic pathways or act as novel signaling molecules [6]. Educated reviews have described the anabolic and metabolic pathways that characterize the sphingolipid metabolism and the involvement of the different sphingolipid species in physiological and pathological processes [7,8]. Lysosomal storage diseases are a group of genetic disorders characterized by the gradual lysosomal accumulation of metabolites due to a defective lysosomal hydrolytic activity [9]. Among them, alterations of the lysosomal catabolic pathways responsible for the progressive breakdown of complex sphingolipids may translate into the accumulation of their corresponding undegraded substrates in lysosomes, leading to inherited sphingolipid storage diseases gathered under the name of sphingolipidoses [10]. In this review, we will focus our attention on zebrafish (Danio rerio) as an animal model for the study of this sub-class of lysosomal storage diseases and as a “zebra-sphinx” platform for a better understanding of the complex and, at least in part, still sphinxlike biology of sphingolipid metabolism (Figure 2). Sphingolipidoses affect approximately 1 in 20,000 newborns [11]. The clinical presentation of patients affected by sphingolipid storage disorders is quite diverse, ranging from a mild progression for some juvenile- or adult-onset forms to severe and fatal infantile forms. To date, approved and investigational therapies for the treatment of lysosomal storage diseases, including sphingolipidoses, comprise hematopoietic stem cell transplantation (HSCT), in vivo and ex vivo gene therapy, enzyme replacement therapy (ERT), substrate reduction therapy (SRT), and pharmacologic chaperone therapy [12,13,14] (Figure 3). These strategies have improved the life of many affected patients by preventing progression or ameliorating various signs and symptoms. However, given the complexities resulting from the alterations of sphingolipid metabolism in different systemic organs, much is still needed at the basic, clinical, and translational levels to improve patient outcomes. For the purposes of the present paper, we will briefly describe the major types of human sphingolipidoses (Table 1). Diseases associated with deficiency of the sphingolipid activator proteins saposins A-D generated by proteolytic processing of the common precursor prosaposin will not be described here. Gaucher disease (GD) is one of the most common sphingolipidoses with an incidence ranging from 1:40,000 to 1:60,000 live births in the general population, with 1:850 in the Ashkenazi Jewish population [15]. GD is caused by recessive mutations in the GBA1 gene that encodes for acid β-glucocerebrosidase, also known as β-glucosidase, a lysosomal enzyme responsible for the degradation of glucosylceramide. The deficiency of acid β-glucocerebrosidase activity leads to the accumulation of its substrate primarily in the lysosomes of macrophages (Gaucher cells) found in the spleen, liver, bone marrow, lungs, and lymph nodes of affected patients [16]. Marked enlarged liver and splenomegaly are clear signs of the disease in children and teenagers that give rise to defects in the blood circulation with anemia and bleeding tendency [17]. Gene expression analysis of cultured skin fibroblasts from GD patients demonstrated that glucosylceramide accumulation triggers the activation of inflammatory responses via the upregulation of genes involved in cytokine and JAK-STAT signaling pathways, the downregulation of genes involved in cell-to-cell and cell-to-matrix interaction, and the inhibition of PI3K-Akt and survival signaling pathways [18]. Several factors may contribute to the severity of GD depending on the type of GBA1 mutation, including the levels of ER stress and proteasomal degradation. In particular, ER stress responses may entail the accumulation of α-synuclein aggregates, causative of neuronal injury and degeneration, as in Parkinson’s disease [19,20]. According to the degree of severity and impairment, GD is classified into three main groups (GD type I–III) based on clinical presentation. The most frequent and less aggressive form of GD is type I, also known as nonneuropathic GD. The onset of the disease varies from childhood to adulthood, and is characterized by bone pain and fractures, splenomegaly, hepatomegaly, anemia, leukopenia, and thrombocytopenia [17]. Although it is considered nonneuropathic, a continuum of clinical forms between GD types may exist, with some neuropathic defects also observed in GD type I patients [21]. GD type II and GD type III are historically classified as primary neurologic diseases. GD type II represents the most severe form as it affects children eliciting rapid degeneration that leads to death before 4 years of age. GD type III usually has a later onset with slower progression [22]. Nowadays, macrophage-directed ERT is the standard of care for symptomatic GD type I and type III patients. It is efficacious in reducing splenomegaly and hematological signs, favoring the growth of GD children, whereas, at variance with ERT, SRT based on the administration of glucosylceramide synthase inhibitors has been shown to be effective in also reducing the skeletal complications [23]. At present, no approved treatment exists for neuropathic GD, but recent studies suggest that the use of ambroxol, an over-the-counter drug that can cross the blood–brain barrier, might be effective [24]. The Fabry disease, also known as the Anderson–Fabry disease, was first described by W. Anderson and J. Fabry in 1898 as a systemic vascular disorder [25]. The Fabry disease is a X-linked disorder caused by mutations in the GLA gene encoding for α-galactosidase A that catalyzes the hydrolysis of terminal non-reducing α-d-galactose residues in α-D-galactosides [26]. Pathogenic variants in GLA result in absent or non-functional α-galactosidase A, leading to the accumulation of its substrate globotriaosylceramide (Gb3) and the deacylated derivative globotriaosylsphingosine in the lysosomes of endothelial cells, myocytes, renal cells, and neurons [27,28]. At the molecular level, the pathogenesis of Fabry disease is still unclear [29]. Gb3 accumulation results in the deregulation of the mitochondrial function and of mTOR and autophagy/lysosome pathways in peripheral blood mononuclear cells from Fabry patients. Of note, similar lysosomal, autophagy, and mitochondrial alterations were also observed in Faber cells, suggesting that a common pathogenic mechanism may exist for both sphingolipidoses [30]. Further confirmation that autophagy and mitochondrial dysfunctions may occur in Fabry disease comes from studies performed on cardiovascular endothelial cells derived from Fabry-induced pluripotent stem cells in which the GLA mutation was corrected by clustered regularly interspersed short palindromic repeats/CRISPR-associated 9 (CRISPR/Cas9) technology [31]. The Fabry disease is typically divided into the major classical or infantile phenotype and the late-onset phenotype. The classical form of Fabry disease affects males that have little or no residual α-galactosidase A activity. It is characterized by clinical heterogeneity and symptoms arise around 1 to 3 years of age. Children with classical Fabry disease usually present acroparesthesia (“Fabry crisis”), angiokeratoma, hypohidrosis, and heat intolerance. The initial symptoms are followed by gastrointestinal disorders, ocular abnormalities, and Gb3 accumulation, causing renal, cardiac, and neurological complications. The milder late-onset Fabry disease involves only a single organ system, typically the heart or the kidneys. Female Fabry patients have a mosaic expression for GLA as a result of X chromosome inactivation and they usually show less severe symptoms [32]. Increasing evidence suggests that cardiovascular morbidity is the main cause of death in Fabry patients, mainly due to increased risk of sudden cardiac death and heart failure [33]. The identification of serum biomarkers derived from collagen type I metabolism has been proposed to predict early fibrotic damage in Fabry patients to be followed by a prompt ERT procedure [34,35,36]. A second currently approved medication is based on chaperone therapy to correct the misfolded enzyme, but an increase in enzymatic activity and a decrease in Gb3/lyso-Gb3 accumulation does not occur in all patients. Currently, SRT and mRNA-based therapy are under evaluation [37]. Niemann–Pick disease (NPD) is an autosomal recessive inherited disorder due to hydrolase deficiency or impaired intracellular cholesterol trafficking. Mutations in acid sphingomyelinase (aSMase), encoded by SMPD1, are causative of the NPD type A and B forms, whereas NPD type C, a lysosomal storage disease distinct from sphingolipidoses, is a cholesterol trafficking defect due to mutations in NPC1 or NPC2 genes [38]. aSMase catalyzes the breakdown of sphingomyelin in ceramide and phosphocholine. The degree of severity of NPD type A and B depends on the aSMase residual activity owing to the type of SMPD1 mutation [39]. When aSMase is mutated, its primary substrate accumulates in the monocytes and macrophages (foam cells) of the liver, spleen, lymph nodes, adrenal cortex, and bone marrow [40]. In children with NPD type A, foam cells infiltrate the brain, causing structural changes, gliosis, demyelination, and neuronal cell loss. Thus, NPD type A is the most severe form of NPD and death occurs within the second or third year of age. NPD type A has a high incidence in the Ashkenazi Jewish population, with a carrier frequency of 1 in 90, whereas NPD type B is a pan-ethnic disease characterized by a later onset and milder symptoms [40,41]. Currently, there is no efficacious treatment for NPD type A and B. Recombinant human aSMase selectively reduces sphingomyelin accumulation in NPD type B fibroblasts in vitro [42] and ERT is now under clinical trial [43]. NPD type C is due to mutations in NPC1, which encodes for a transmembrane protein of the lysosomal membrane, or NPC2, which encodes for an intracellular cholesterol transporter. Both deficiencies lead to intracellular accumulation of unesterified cholesterol and glycosphingolipids [44]. Its incidence is about 1 in 100,000 live births and can be divided into neonatal, late infantile, and juvenile [45]. Neonatal presentation is rare and characterized by progressive liver disease, which represents the most common cause of death among neonatal-onset NPD type C patients [46]. Late infantile and juvenile forms are the most common, characterized by the outbreak of neurological disorders; in contrast to the infantile form, there is no liver or spleen enlargement. NPD type C is usually treated with anti-hypercholesterolemic drugs, but this does not ease the symptoms [47,48]. Also known as globoid cell leukodystrophy, Krabbe disease is an autosomal recessive disorder characterized by the deficiency of the acid hydrolase β-galactosylceramidase (GALC) encoded by the GALC gene. GALC catalyzes the removal of β-galactose from β-galactosylceramide (a major component of myelin) and other terminal β-galactose-containing sphingolipids, including the neurotoxic metabolite β-galactosylsphingosine (psychosine) [49]. By acting at different cellular levels, GALC deficiency causes psychosine accumulation paralleled by neuroinflammation, degeneration of oligodendroglia, and progressive demyelination [50]. Psychosine has been shown to inhibit protein kinase C signaling, activate the caspase cascade, disrupt the trans Golgi network and endosomal vesicles, and impair mitochondria and peroxisome function [51]. In addition, the detergent-like action of psychosine may disturb the membrane microdomain organization of lipid rafts, causing demyelination [51,52,53]. Moreover, deregulation of brain neovascularization occurs in Krabbe patients and in twitcher mice, an authentic model of the disease [54], whereas neuroinflammation leads to increased levels of long pentraxin 3, an innate immune response mediator that acts at the site of inflammation [55]. The early infantile form (onset at birth to 5 months of age) represents the most common and severe type of Krabbe disease. It is characterized by fast progression and the symptoms include regression of psychomotor development followed by seizures, loss of vision and hearing, and early death [56]. The late-infantile onset occurs between 13 and 36 months and is characterized by motor regression, ataxia, and progressive blindness [57]. Adult forms of Krabbe disease are rare; they display progressive spastic paraplegia and sometimes neuropathy [58]. ERT is not the most effective treatment because of its poor ability to pass the blood brain barrier and the immune response against the recombinant GALC protein [51]. Currently, the standard of care is HSCT, which significantly improves the lifespan of Krabbe patients when performed before the outbreak of symptoms [57]. Farber disease is a rare autosomal inherited metabolic disorder caused by inactivating mutations in the ASAH1 gene that encodes for the lysosomal acid ceramidase. Acid ceramidase promotes the breakdown of ceramide in sphingosine and fatty acid, and its deficiency leads to the progressive accumulation of ceramide in bone, cartilage, immune system, central nervous system, lungs, and other organs [59]. Farber lipogranulomatosis has a wide range of age onset and clinical features, even though subcutaneous nodules, made of ceramide engorged macrophages, arthritis, and dysphonia are the three major signs of the disease [60]. As for other sphingolipidosis, the infantile form is the most severe, characterized by progressive neurologic regression and lung disorders. Milder forms present only modest or no alterations of the central nervous system [61]. Unfortunately, no effective therapies are currently available for this disease [59]. Gangliosides are glycosphingolipids that account for up to 10% of brain lipid content and were isolated from the human brain for the first time in 1939 by E. Klenk [62]. They are composed of sialic-acid-containing oligosaccharide chains linked via a β-glycosidic bond to ceramide, which is responsible for their insertion into cell membranes. Deficiencies in enzymes involved in their metabolism cause an accumulation of unmetabolized gangliosides in lysosomes, mainly in neurons where ectopic neurite outgrowth may occur [63]. β-Galactosidase is a lysosomal hydrolase that cleaves β-linked galactose residues from the non-reducing end of glycan moieties found in various glycoconjugates [62]. Deficiency in the β-galactosidase encoding gene GBL1 leads to the accumulation of the GM1 ganglioside and its derivative GA1 mainly in lysosomes. Like all of the other lysosomal disorders, GM1 gangliosidosis is an inherited metabolic disease with an estimated incidence of 1 in 100,000–200,000 newborns [64]. The most severe form of the disease is the infantile type I GM1 gangliosidosis, characterized by hydrops fetalis developmental psychomotor regression and, as the child grows, hepatosplenomegaly and skeletal abnormalities. Type II GM1 gangliosidosis is named late infantile or juvenile, depending on the age at which the first symptoms arise: between 12 and 24 months for the late infantile form and 3–5 years for the juvenile form. Children quickly lose their ambulatory capacity and need a gastrostomy placement. In the juvenile form, ataxia and dysarthria follow the psychomotor decline [65]. The adult-onset type III GM1 gangliosidosis is characterized by milder and more varied symptoms, with a longer life expectancy [66]. Currently, no specific treatment exists for GM1 gangliosidosis; the therapy aims to relieve symptoms and is mostly palliative [67]. Recently, miglustat, a glucosylceramide synthase inhibitor, used for SRT in GD and NPD type C diseases [68,69], has also been proposed for the treatment of children affected by type II GM1 gangliosidosis [70]. The disease is due to the lysosomal accumulation of the GM2 ganglioside [71], which represents about 5% of all brain gangliosides [72]. The hydrolysis of GM2 to GM3 ganglioside is performed by β-hexosaminidase A (HEXA), a heterodimer whose α and β subunits are encoded by HEXA and HEXB genes, respectively, and requires the GM2 activator protein (GM2AP) as a cofactor [73]. In an ERT prospective, an enzymatically active recombinant protein homodimer HexM has been developed that is able to interact with the GM2AP–GM2 complex in vivo [73]. Currently, the use of HexM as ERT has not been transferred to the clinics and works are in progress to optimize an AAV vector for gene therapy [74,75] with reduced immune response reactions [76]. Three forms of GM2 gangliosidoses have been described: the AB variant, Tay–Sachs disease, and Sandhoff disease. They are characterized by neurological disorders that vary from hypotonia regression to cerebellar ataxia according to the age of onset [71]. The AB variant is the rarest form of GM2 gangliosidosis, with about 30 cases reported in the scientific literature. It is caused by inherited mutations of the GM2A gene that disrupt the activity of the GM2AP cofactor. The AB variant is characterized by severe cerebellar atrophy that causes dysphagia, muscle atrophy, psychotic episodes, and manic depression [72]. More than 130 mutations of the HEXA gene have been reported for Tay–Sachs disease, which has an incidence of 1 in 100,000 live births [77]. HEXA encodes for the α-subunit of the enzyme and the disease presents an ample heterogeneity of clinical symptoms based on hexosaminidase residual activity [69]. Tay–Sachs disease can be divided according to the age of onset. The infantile form represents the most aggressive form and is associated with very low hexosaminidase activity. Developmental delay arises around the sixth month of age and is followed by blindness, cognitive impairment, seizures, and paralysis, resulting in death before 5 years of age [78]. The juvenile form is characterized by ataxia, dysarthria, and developmental delay; the survival time is usually around 14 years [79]. The adult form is less severe and has 5–20% of hexosaminidase residual activity. With the progression of the disease, patients complain of leg weakness, ataxia, tremor, and psychiatric disorders [80]. Current treatments for Tay–Sachs patients involve SRT, bone marrow transplantation, hematopoietic or neural stem cell transplantation, and the use of anti-inflammatory drugs. However, most of the treatments have failed to relieve neurological symptoms owing to the difficulty in restoring hexosaminidase activity in the brain [77]. Sandhoff disease accounts for approximately 7% of GM2 gangliosidoses. In this type of GM2 gangliosidoses, HEXB variants prevent the correct catabolism of GM2 ganglioside with its lysosomal accumulation in the central nervous system and somatic cells [62]. As for other sphingolipidoses, Sandhoff disease has been classified into infantile, juvenile, and adult forms according to the severity of the disease and the age of onset. The cardinal clinical features of infantile Sandhoff disease are seizure, muscle weakness, developmental delay, and regression; death occurs before 3 years of age [81]. Late onset forms are less common and characterized by lower motor neuron disease and neurological degeneration [82,83]. Clinical manifestations, mainly in juvenile and adult Sandhoff patients, are heterogeneous and based on residual hexosaminidase activity. A case report of two siblings with compound heterozygous HEXB mutations further confirmed the clinical heterogeneity of Sandhoff disease [84]. As in Tay–Sachs disease, efficacious therapy for Sandhoff patients is still lacking owing to poor diffusion of the drugs into the nervous system [83]. Metachromatic leukodystrophy (MLD) is an autosomal-recessive inherited sphingolipidoses caused by deficiency of the enzyme arylsulfatase-A encoded by the ARSA gene. The enzyme cleaves sulfatides in galactosylceramide and its deficiency leads to the formation of sulfatide-engulfed metachromatic granules in oligodendrocytes, microglia, Schwann cell, neurons, and macrophages, causing myelin degradation and inflammation [85]. Motor neurons derived from induced pluripotent MLD stem cells are characterized by lysosomal accumulation of sulfatides, mitochondrial fragmentation, and impaired autophagy, leading to premature cell death [86]. The worldwide incidence of MLD is around 1.5 in 100,000 live births, being much higher in Habbanite Jews (1:75) and Navajo Indians (1:2500) [85]. Different mutations in ARSA are associated with two groups with different residual arylsulfatase-A activity: the allele 459+1G>A is the most frequent mutation in Europe and belongs to group 0 with no residual activity, whereas the alleles 1277C>T and 536T>G represent the R group with minimal residual activity [87]. The disease can be also divided into four groups according to the age at onset: late infantile, early, and late juvenile, and adult forms. Late infantile and early juvenile MLD are the most frequent forms with severe and rapid progression; they arise during the second and fourth year of life, respectively, and the symptoms affect both the central and the peripheral nervous system [87]. Adult MLD is often misdiagnosed as early-onset dementia or schizophrenia because of its slow progression [85]. The most promising treatment is bone marrow transplantation or HSCT when performed before the onset of symptoms [85]. Moreover, HSCT leads to stabilization or reduced decline in motor and cognitive functions and the positive effects were particularly meaningful in the peripheral nervous system in patients with late-infantile MLD, refractory to other therapeutic interventions [88]. Beginning with use as a vertebrate animal model during the 1980s [89], the teleost zebrafish has emerged as a useful platform for studies in diverse fields of research. The high grade of genome conservation between human and zebrafish (around 70%, and the percentage increases to 84% when focusing on genes associated with human diseases) [90], combined with precise genome editing and the ease of manipulation, enable to model several human diseases in zebrafish, such as cancer [91], neurodegenerative [92], cardiovascular [93], behavioral [94,95], and inherited [96,97] disorders, including sphingolipidoses. Indeed, lipidomic analysis has revealed the presence in zebrafish of all the principal lipid classes present in mammals (see [98,99] and Figure 4), supporting the possibility to model lipidic metabolism diseases in the fish with the advantage of using existing mammalian lipid databases for data processing [100]. In addition, zebrafish is useful to study lipidic changes after exposure to industrial pollutants [101], drugs [102], toxic compounds [103], or a high-cholesterol/high-fat diet [99]. Moreover, zebrafish larvae can be fed with fluorescent BODIPY-lipids to serve as metabolic tracers when incorporated in vivo into more complex lipid products [99]. The dynamic composition of lipids in the body and yolk sac of zebrafish embryos was investigated during the first 5 days of development via liquid chromatography/mass spectrometry (LC/MS), demonstrating significant differences between the two embryonic compartments [98]. The results have shown that cholesterol, phosphatidylcholine, and triglycerides are the most abundant lipids in zebrafish embryos. Of note, the yolk not only represents simple storage of lipids to provide energy for the growing embryo, but also an active organ where lipids are remodeled before reaching the embryo body. Desorption electrospray ionization MS imaging followed by nanoelectrospray MS and tandem MS (MS/MS) were used to detect phosphatidylglycerols, phosphatidylcholines, phosphatidylinositols, free fatty acids, triacylglycerols, ubiquinone, squalene, and other lipids during zebrafish embryonic development from 0 to 96 h post fertilization (hpf) [105]. In addition, high-spatial-resolution matrix-assisted laser desorption/ionization (MALDI) MS imaging was applied to map and visualize the 3D spatial distribution of phosphatidylcholine, phosphatidylethanolamines, and phosphatidylinositol molecular species in zebrafish embryos at the one-cell stage, whereas high-spatial-resolution 2D MALDI MS imaging was used to analyze zebrafish embryos at the 1- to 16-cell stages [106]. These studies have allowed to investigate the composition and distribution of lipids in zebrafish, with insights about lipidic dynamics during embryonic development. Given the growing interest in the study of the zebrafish lipidome, attempts have been performed to improve the quality of lipid analysis. For instance, conventional one-dimensional LC (1D-LC) was compared to comprehensive two-dimensional LC (2D-LC) coupled to a high-resolution time-of-flight mass spectrometer for a full-scale lipid characterization of lipid extracts from zebrafish embryos. The results demonstrate that 2D-LC is 2.5 times more efficient than 1D-LC, allowing the annotation of more than 1700 lipid species [107]. Recently, a direct infusion MS/MS approach using multiple reaction monitoring was applied to precisely quantify membrane lipid composition both in the yolk and in the zebrafish embryo body during the gastrula stage [108]. Around 700 membrane lipids were annotated, divided into two main lipid classes: sphingolipids and phospholipids, with the latter including phosphatidylcholine, phosphatidylinositol, phosphatidylserine, and phosphatidylethanolamine. The composition of the embryo body and yolk was quite similar, with phosphatidylcholine representing the most abundant species. However, major differences were found in the content of phosphatidylserine, dihydrosphingolipids, and sphingomyelin with short-chain fatty acids (significantly higher in the embryo body than in the yolk). Notably, the fine tuning of the sphingolipid synthesis appears to be related to the wnt pathway and is fundamental for proper orientation during cell division. Lipidomic analysis can also be applied to specific organs from adult zebrafish. For instance, livers from 6-month-old animals were analyzed with different MS techniques, identifying 712 unique lipid species from four categories (fatty acyls, glycerolipids, glycerophospholipids, and sphingolipids) [109]. Moreover, adult zebrafish brains have been analyzed for changes in the lipid profile after exposure to different xenobiotics [102,110]. The central hub of the sphingolipid pathways ceramide and its derivatives play a pivotal role in different biological processes [3]. The ceramide profiles of adult zebrafish brain, 7-day-old zebrafish larvae, and human cells were compared using a parallel reaction monitoring approach in which a targeted quantification method was associated with high-resolution hybrid MS [111]. The results highlighted a significant overlapping in ceramide distribution, even though a scarcity of sphingadiene-containing ceramides was observed in zebrafish specimens, despite their biological importance in mammals. These results raised the hypothesis about possible alternative unexplored lipidic pathways in zebrafish that might pave the way for novel discoveries in human sphingolipid disorders. Targeted sphingolipidomics performed at various stages of embryonic development and in adult animals under different physiological and pathological conditions are required for a better understanding of sphingolipid metabolism and function in zebrafish. Zebrafish and human genomes share a high homology [90], thus several lipid-metabolizing enzymes involved in human diseases have a zebrafish counterpart (Figure 2). Ceramide synthases (CERS), the enzymes responsible for ceramide production, play a central role in the sphingolipid metabolism. Highly conserved through evolution, the CERS gene family encompasses six isoforms (CERS1–6) with diverse spatial/temporal expression in mammals. All CERSs except for CERS1 show an N-terminal homeobox-like domain whose functions remain elusive [112]. In zebrafish, nine genes encoding for the six cers subtypes have been identified with a sequence homology with human and mouse counterparts ranging from 46% to 79% identity. Owing to the genome duplication typical of zebrafish, cers2, cers3, and cers4 are present as double copy genes (a and b), while cers1, cers5, and cers6 are present as single copy genes. As in mammals, all zebrafish orthologs display the Hox domain, except for the Cers1 protein. The tissue-/stage-dependent expression of the cers genes has been analyzed during zebrafish embryo development by whole mount in situ hybridization (WISH) [113]. The results suggest that these enzymes are involved in diverse biological processes and that the production of ceramides may dynamically vary in different tissues. For instance, only cers2a and cers3b are expressed in the embryonic zebrafish pronephros, congruent with the high expression of Cers2 in murine kidney, while all cers are expressed in the nervous tissue, possibly pointing to the requirement for various ceramide species in the developing brain. Notably, the expression of cers can be modulated in zebrafish embryo when a perturbation in the lipidic composition occurs. Indeed, zebrafish embryo mutants for the sphingosine kinase gene sphk2, in which a potentially dangerous accumulation of the metabolite sphingosine occurs, upregulate the expression of cers2b to activate the sphingolipid salvage pathway and turn the excess of sphingosine in ceramide [114]. In this frame, the ortholog of the human peroxisome proliferator-activated receptor γ-responsive transmembrane gene FAM57B, involved in the regulation of ceramide metabolism, was found to maintain the homeostasis of sphingolipids and glycerol lipids during brain development in zebrafish [115]. Indeed, untargeted lipidomic analysis performed in the brain tissue of 7-day-old fam57b null and heterozygous zebrafish lines has revealed remarkable differences in the lipid profile with consequences on membrane composition and permeability when compared with wild type animals. As described above, ceramide catabolism is catalyzed by the lysosomal acid ceramidase encoded by the ASAH1 gene. In silico analysis has revealed the presence of two ASAH1 co-orthologs in zebrafish (asah1a and asah1b) [116]. Genome editing techniques have revealed the importance of lysosomal acid ceramidase to maintain the physiological levels of ceramide in zebrafish. Indeed, asah1a and asah1b enzymes are both able to hydrolyze ceramide and the presence of either asah1a or asah1b prevents substrate accumulation, with ceramide being increased only in double asah1a−/−/asah1b−/− mutants [117]. Ceramide represents the substrate for the generation of more complex sphingolipids, such as sphingomyelin, a central component of myelin. The enzymes responsible for the production of sphingomyelin from ceramide are named sphingomyelin synthases (SMSs). As reported in the ZFIN database [118], two duplicated genes are predicted for the two human SMS genes in zebrafish (sgms1a, sgms1b, sgms2a, and sgms2b). Moreover, a gene ortholog for the human enzymatically inactive SMS-related protein (SMSr) has been reported in zebrafish, named zgc:175139 [119]. SMSr represents a key regulator of ceramide homeostasis that may operate as a sensor rather than a converter of ceramides in the ER [120]. However, its role in zebrafish remains unexplored. SMases, also named sphingomyelin phosphodiesterases, catalyze the production of ceramide and phosphocholine from sphingomyelin, representing one of the three ceramide production pathways together with the de novo synthesis and the salvage pathways [121]. In zebrafish, smpd1 has been identified as a single ortholog of the aSMase-encoding gene SMPD1, which shares 59% identity with the human counterpart. A mutant line of smpd1 was created in zebrafish via the CRISPR/Cas9 technique; the SMase enzymatic activity was abolished by 93% at 5 days post fertilization (dpf) with a consequent increase in various sphingolipid metabolites [122]. SMPD2 encodes for the membrane-bound Mg2+-dependent neutral SMase1. Its smpd2 ortholog has been cloned in zebrafish and it has been shown to mediate ceramide production and activation of apoptosis following heat stress in zebrafish embryonic cells [123]. Of note, neutral SMase1 is activated by phosphorylation at Ser-270 downstream of the c-Jun N-terminal kinase pathway in both zebrafish and human cells [124], and thalidomide exerts an antiangiogenic effect on zebrafish embryos due to the upregulation of neutral SMase activity and the consequent production of ceramides [125]. In addition, a mitochondrial neutral SMase (mtSMase) has been characterized in zebrafish. mtSMase was purified from zebrafish cells and tested for its enzymatic activity, showing an optimum working pH of 7.5 and sphingomyelin as the main substrate. Cell fractionation and immunofluorescence analysis demonstrated the mitochondrial localization of this novel SMase. Another neutral SMase has been identified in zebrafish that represents the ortholog of the human gene SMPD3 with a conserved identity of 55% [126]. The sphingoid base sphingosine is an important component of sphingolipid metabolism. Its biologically active metabolite sphingosine-1-phosphate (S1P) is involved in a variety of physiological and pathological processes by binding specific G-coupled receptors (S1PRs) [127]. The study of sphingosine and related metabolites in zebrafish can provide novel information about the human counterparts, favoring a better understanding of the biological mechanisms involved in sphingolipidoses and other pathologies. Two S1P phosphatase (spp1 and spp2), two sphingosine kinase (sphk1 and sphk2), and one sphingosine lyase (spl) encoding genes have been identified in zebrafish, together with seven conserved s1pr orthologs corresponding to the five human S1PRs (s1pr3 and s1pr5 being duplicated in zebrafish). s1pr1 is highly expressed in the brain, while s1pr4 is expressed mainly in the kidney, which represents the zebrafish hematopoiesis site, thus reflecting the mammalian S1PR expression in lymphoid and hematopoietic tissues [128]. Knockdown (KD) and knockout (KO) approaches have shown that the S1P/S1PR pathway plays a pivotal role in vascular and cardiac development in the zebrafish embryo [129,130]. A single ortholog of the human GLA gene is present in zebrafish (a-gal) with significant similarities between human and zebrafish proteins (>70%). [131]. Enzymatic and immunohistochemical analyses have shown that the zebrafish protein retains significant α-galactosidase activity and a distribution in zebrafish kidney like in humans, suggesting that it may retain the same biological functions. Finally, the zebrafish gene arsa is reported in the ZFIN database [118] as an ortholog of the human gene ARSA with a predicted arylsulfatase activity; however, its role in zebrafish remains to be investigated. In the last decades, the use of zebrafish to study gene function has increased exponentially thanks to the multiple advantages offered by this model, such as a high number of offspring generated, embryo transparency, and quick genetic manipulation. The necessity to target distinct genes to study their function led to the development of different techniques to block gene function either in a transient manner or permanently. One of the most used commonly techniques involves the injection of antisense oligonucleotides complementary to specific genetic loci, named morpholinos (MOs), which temporarily KD protein production [132]. There are two different strategies by which MOs can interfere with protein expression. The first strategy is based on the block of the translation of the targeted gene (ATG-MO). The second one is aimed at interfering with the splicing process that occurs during mRNA maturation (splicing-MO) [96]. Genome editing techniques have been extensively applied in the zebrafish field. To date, the most used techniques are the Zinc-Finger Nuclease, the Transcription Activator Like Effector Nuclease (TALEN) [133,134], and a more recent approach based on the CRISPR/Cas9 system [135,136,137]. Briefly, these systems use different approaches to drive proteins with nuclease activity to a specific DNA sequence. Once bound to the locus, the nucleases cut the double-stranded DNA, forcing the cell to activate double-strand break repair mechanisms. The repair process mediated by the non-homologous end-joining repair system can introduce deletions or insertions (indels) into the break-point region, which can lead to alterations of the reading frame and hence to an altered protein sequence and loss of gene function. Using these procedures, several models of inherited human diseases have been generated in zebrafish. In line with this review, we will discuss the phenotypes of the main models of sphingolipidoses in zebrafish (Table 2). Injection of a splicing-MOs directed against gba1, the zebrafish ortholog of the human GBA1 gene, caused the appearance of specific alterations in 5 dpf embryos, including curvature of the trunk, defects of primary bone ossification with a decrease in col10a1 and runx2b gene expression associated with a dysfunction of the osteoblast population, and severe erythropenia and thrombocytopenia caused by early hematopoietic defects. Microarray analysis demonstrated alterations in the expression of genes involved in different biological processes, including mitochondrial activity and intracellular vesicle trafficking. These defects were paralleled by increased oxidative stress and reduced signaling of the Wnt/β-catenin pathway [138]. In a different study, gba1 KD led to an increased number of vacuolated macrophages, characterised by migratory defects and enlarged lysosomes, pointing to an impairment in the macrophage function due to the low levels of acid β-glucocerebrosidase activity [148]. The first KO model for GD was derived from a forward genetic screening that led to the identification of a gba1sa1621/sa1621 mutant zebrafish line. The characterization of this mutant revealed a decrease in the body length and curvature of the spine at 7 dpf [138]. Like gba1 morphants, gba1sa1621/sa1621 mutants show a reduction in col10a1 and runx2b expression related to osteoblast function, as well as erythropenia. In addition, splenomegaly and hepatomegaly can be observed in 3-month-old mutants. To date, different zebrafish KO models of GD have been generated taking advantage of genome editing techniques. The first engineered gba1 null mutant was obtained by TALEN approaches [139]. Mutants did not show any significant defect during the early stages of development, with the first alterations in the swimming behaviour occurring at 8 weeks of age and curvature of the spine at 12 weeks. Histopathological analysis revealed the presence of Gaucher-like cells in the brain, liver, thymus, and pancreas of adult KO animals. Furthermore, dopaminergic neuron degeneration, the presence of cytoplasmic inclusions resembling Lewy bodies, an increased number of autophagosomes, and microglia activation were observed in KO brains. MS demonstrated the accumulation of sphingolipid metabolites in gba1−/− larvae at 5 dpf. They included hexosylsphingosine, glucosylceramide, lactosylceramide, and galactosylceramide, and their levels were further increased in juvenile brains. A second zebrafish model of GD was generated by the CRISPR/Cas9 technique [140]. As observed for the gba1 null zebrafish generated by TALEN, adult gba1 KO mutants showed a curved back, and swimming and feeding impairment starting from 10 weeks of age. In addition, adults were characterized by the presence of Gaucher-like cells in the liver, spleen, and pancreas, together with an increase in glucosylsphingosine and glucosylceramide levels in the brain and liver. However, in this study, gba1 KO larvae appeared to accumulate solely glucosylsphingosine at 5 dpf, and no altered levels of other glycosphingolipids were observed at this stage. Expression analysis of specific mRNAs in the brain of adult gba1−/− zebrafish mutants [117] revealed the upregulation of macrophage (gpnmb, chia.6), microglia (apoeb), and complement system (c1qa, c3.1, c5, c5aR1) markers, as well as the upregulation of proinflammatory cytokines (il1-b, tnf-a2), whereas downregulation of the dopaminergic neuron marker (th1) and of the myelin-related gene (mbp) were observed. Autophagy was also increased in these brains. More recently, a further KO model was generated in zebrafish using TALEN [151]. Animals showed a reduction in dopaminergic and noradrenergic neurons at 3 months of age, confirming the importance of gba1 function for neuronal survival. Cytosol-facing GBA2 metabolizes cytosolic glucosylceramide. Genetic ablation of the Gba2 gene exerts beneficial effects in murine models of GD and NPD type C [152,153]. In order to investigate the potential role of GBA2 in compensatory glucosylceramide metabolism during inadequate GBA1 activity, double gba1 and gba2 null animals were generated by CRISPR/Cas9 in zebrafish [140]. Lipid analysis performed on double mutants at 5 dpf showed increased glucosylceramide levels when compared with single gba1−/− larvae, but similar to those detected in single gba2−/− animals. Moreover, glucosylcholesterol was significantly decreased in the double KO animals and in single gba2−/− mutants. In addition, a significant accumulation of glucosylsphingosine occurs in double gba1/gba2 null animals when compared with controls. Notably, in keeping with an SRT approach for the treatment of GD, the administration of the potent specific glucosylceramide synthase inhibitor eliglustat elicited a significant decrease in hexoxylceramide and in the derived lipids glucosylsphingosine and hexoxylcholesterol in gba1−/− larvae. Together, these data indicate that zebrafish larvae offer an attractive model to study glucosidase actions on glycosphingolipid metabolism in vivo. To study the role of excessive glucosylsphingosine formation during acid β-glucocerebrosidase deficiency, KO zebrafish lines were generated for the two ASAH1 orthologs asah1a and asah2b [117]. Of note, double asah1a/asah1b null larvae, but not single asah1a or asah1b mutants, accumulate the primary substrate ceramide. Nevertheless, only asah1b appears to be involved in the formation of glucosylsphingosine in a gba1-deficient background. Accumulation of glucosylsphingosine in gba1−/−/asah1b−/− zebrafish did not prevent the formation of Gaucher-like cells, glucosylceramide accumulation, or neuroinflammation. However, these double mutants show an ameliorated course of disease reflected by a delay in the appearance of locomotor abnormalities and curvature of the back, reduced loss of dopaminergic neurons, and increased lifespan, suggesting that the accumulation of glucosylsphingosine may play a role in the pathogenesis of GD. A similar approach was used to investigate the impact of acid SMase activity on a glucocerebrosidase-deficient background by generating double gba1−/−/smpd1−/− zebrafish mutants [122]. Unexpectedly, double gba1−/−/smpd1−/− mutants showed a markedly prolonged survival, rescue of neuronal and mitochondrial damages, and normalization of the motor phenotype when compared with gba1 KO animals. This occurred in the presence of an additive increase in the levels of various sphingolipid metabolites. Both GSC1 and SMPD1 variants represent inherited risk factors for Parkinson’s disease [154]. In keeping with the data obtained in double KO animals, human cells with combined glucocerebrosidase and aSMase deficiency showed an unpredicted reduction in intracellular α-synuclein levels. Together, these observations indicate that a better understanding of the crosstalk among sphingolipid metabolizing enzymes is required to dissect the pathogenesis of sphingolipid-related pathologies and for the development of efficacious therapeutic approaches. A model of Fabry disease was obtained in zebrafish by the generation of a α-gal KO fish line using the CRISPR/Cas9 technique [131]. This led to the decrease in α-gal protein expression in the kidney associated with a marked reduction in α-galactosidase activity. Even though KO mutants did not show significant differences in body size when compared with wild type animals, they were characterized by an increased mortality during the early embryonic stages. A more in-depth analysis unveiled an increase in creatinine levels and the leak of high molecular weight proteins, suggesting that an impairment of glomerular filtration may occur in these mutants. Accordingly, microscopic analysis of the kidney revealed an increased glomerular size, dilated capillary loop, and thinner Bowman’s space. In contrast with the results obtained in other animal models of Fabry disease that do not show renal abnormalities [155], these data are in keeping with the nephropathy that occurs in Fabry patients [156]. Notably, the absence of a Gb3 synthase encoding gene ortholog in zebrafish provides the unique opportunity to identify pathogenic processes that may work in concert with Gb3 accumulation in Fabry disease [131]. A KO model of NPD type A was generated in zebrafish using the CRISPR/Cas9 system [122]. Smpd1 KO animals showed a 93% reduction in the enzymatic activity at 5 dpf, with a consequent increase in sphingolipid metabolites, including sphingomyelin, ceramide, lactosylceramide, and sphinganine. However, despite the absence of enzyme activity and the significant increase in key glycolipid substrates, no obvious phenotype was observed in embryo and adult KO animals. At variance with the paucity of zebrafish models for the aSMase-deficient forms of NPD, various attempts have been made to model the type C form of NPD, a lysosomal storage disease distinct from sphingolipidoses that depends on cholesterol trafficking defects due to mutations in NPC1 or NPC2 genes. For instance, injection of specific MOs for npc1, the orthologous of NPC1, induces an accumulation of unesterified cholesterol at early embryonic stages [141]. Morphological evaluation of the zebrafish KD morphants injected at one cell stage or in the yolk syncytial layer revealed a disorganization of the actin cytoskeleton and a delay in the development during epiboly, unveiling a role for npc1 in cell movement at this embryonic stage. Interestingly, a lower dose of MO was associated with a milder phenotype characterized by neuronal death, like in the human pathology. KD of npc1 in zebrafish has also been associated with thrombocytopenia, as observed in some NPD patients [157]. Various KO models for NPD type C have been developed using the CRISPR/Cas9 technology to inactivate the npc1 [142,143] or the npc2 [144] gene. KO of npc1 caused premature death of half of the animals during the embryonic and juvenile stages, with a significant reduction in body length, together with hepatomegaly, splenomegaly, neurological defects, and ataxia—features that resemble those observed in patients and other animal models of NPD type C. Moreover, analysis of hepatocytes unveils a massive accumulation of cholesterol and changes in the levels of different types of lipids, including ceramide, diacylglycerol, and lysophosphatidic acid [144,145]. At variance with npc1 null animals, npc2 KO fishes were able to reach adulthood, even though they showed a reduction in body size and impairment in the locomotor system starting from 2 months and 4 months of age, respectively [144]. Histopathological analysis of npc2 KO adults revealed the presence of foam cells in liver and kidney, defects in axonal myelination, and alterations of cerebellar Purkinje cells. Notably, significant alterations have also been observed in npc2 null zebrafish at early stages of development [145]. They include the accumulation of unesterified cholesterol, upregulation of markers of inflammation and activated microglia, mitochondrial dysfunction, defects in the myelination process, and an anxiety-like behaviour. Like what was observed in npc1 maternal mutants, npc2 KO derived from homozygote females show an aberrant phenotype already at 30 hpf, such as a curved tail, absence or abnormalities of the otoliths, defects in the brain structures, and lack of circulating cells—defects that may arise from an impairment of the Notch3 signaling pathway [145]. As for NPD type C, zebrafish has been used as a platform to model lysosomal storage diseases other than sphingolipidoses, including mucolipidosis type II (MLII) and mucopolysaccharidosis type II (MPSII), providing novel information about the pathogenesis of these disorders. Briefly, MLII is due to the mutation in the GNPTAB gene, encoding for the catalytic subunit of N-acetylglucosamine-1-phosphotransferase that catalyses the first step of the formation of mannose 6-phosphate (M6P)-tagged lysosomal soluble hydrolases. As a consequence of GNPTAB mutation, the lack of the M6P tag causes the missorting and secretion of such hydrolases, with lysosomal accumulation of their substrates. A first zebrafish model of MLII was obtained by MO injection. Morphant embryos showed craniofacial defects, impaired motility, and abnormal otolith and pectoral fin development. This model allowed to undercover alterations in the spatial-temporal expression of type II collagen and Sox9 [158]. Stable mutant lines for the gnptab gene were obtained by TALEN and site-directed mutagenesis technologies. Zebrafish mutants showed a craniofacial phenotype and elevated levels of cathepsin K activity associated with abnormal cartilage development and heart and valve malformations [159,160]. MPSII is caused by mutation in the IDS gene, encoding for the lysosomal enzyme iduronate 2-sulfatase, leading to the toxic accumulation of glycosaminoglycans into lysosomes (mainly dermatan and heparan sulphates) and multi-organ damage. An MO approach in zebrafish targeting ids, the single ortholog for human IDS, caused early defects in embryonic development. In particular, the abnormal migration and differentiation of neural crest cells into chondroblasts were responsible for craniofacial cartilage defects, while sonic hedgehog pathway disruption led to congenital heart defects [161,162]. In addition, KO of ids in zebrafish has provided novel information about the role of early deregulation of the fibroblast growth factor signaling pathway in the occurrence of irreversible skeletal defects before glycosaminoglycans’ accumulation [163]. With a different approach, human-mutated IDS mRNAs have been injected into zebrafish embryos for a rapid preliminary study about novel IDS point mutations associated with MPSII [164]. Two GALC co-orthologs have been identified in zebrafish (GALCa and GALCb) that share a high identity with their human counterpart [148]. Further analysis confirmed that both isoforms are endowed with enzymatic activity and are localised in the lysosome. Moreover, WISH analysis revealed their co-expression in the central nervous system during embryonic development. Injection of single GALCa or GALCb specific MOs in zebrafish resulted in the partial reduction in enzymatic β-galactosylceramidase activity, which was completely abolished by the simultaneous injection of both MOs. Nevertheless, no evident morphological alterations were observed in both single- and double-injected morphants during embryonic development. Notably, no alterations in psychosine levels were detected in double GALCa/GALCb morphants, suggesting that the transient abrogation of GALC activity is not sufficient to accumulate this metabolite [146]. Relevant to this point, myelination in zebrafish starts in the hindbrain at day 4 of development and is not completed at day 10 [165], making the study of the effect of β-galactosylceramidase deficiency on myelination in zebrafish morphants unfeasible. However, analysis of the expression pattern of a set of neuronal marker genes unveiled a significant reduction and partial disorganization in neurod1 expression and neuronal death in double GALCa/GALCb morphants, in keeping with the neurodegenerative features of Krabbe disease. These data suggest that GALC loss-of-function may have pathological consequences independent of psychosine accumulation, thus providing new insights into the pathogenesis of Krabbe disease. This possibility is supported by the observation that psychosine levels do not correlate with nervous system regions exhibiting demyelination and axonopathy in twi-5J mice harboring a spontaneous missense GALC mutation [166]. Thus, double GALCa/GALCb zebrafish morphants may represent an interesting option for addressing previously unrecognized psychosine-independent key aspects of the pathogenesis of Krabbe disease. Transient downregulation of asah1b using an ATG-MO approach led to a 74% decrease in acid-ceramidase activity in zebrafish embryos [147]. Embryo morphants develop macroscopic phenotypic alterations by 48 hpf. Further analysis has disclosed increased neuronal apoptosis localised only in the spinal cord, leading to a reduction in the number of motor neuron branches. This defect does not affect peripheral projections, indicating a specific susceptibility of motor neurons to the reduced levels of lysosomal acid-ceramidase [147]. A more recent zebrafish model of Farber disease was generated using the CRISPR/Cas9 technique [111]. Three different mutant lines were generated: single KOs for each of the two co-orthologs (asah1a or asah1b) and a double asah1a/asah1b KO. At variance with the MO model, the abrogation of only one gene (asah1a or asah1b) did not lead to the appearance of an evident phenotype until adulthood, whereas double KO animals display a progressive reduction in body size when compared with wild type and single KO siblings. These differences became more evident 3 months after birth and double KO animals died within 4 months [111]. Accordingly, sphingolipid analysis performed on the brain of 3.5-month-old fishes revealed a significant accumulation of ceramide only in double KO animals. These results indicate that the activity of a single asah1 ortholog is sufficient to maintain physiological levels of ceramide and to guarantee a normal phenotype in zebrafish. At variance, reminiscent of the joint deformations observed in Farber patients [61], the complete abrogation of acid-ceramidase activity impairs the normal growth of the skeletal system in zebrafish and induces a premature death, probably due to heart failure or seizure related to progressive ceramide accumulation. During a multi-gene analysis to understand the role of macrophages in tuberculosis progression, a zebrafish model for Tay–Sachs disease was generated by injecting an MO targeting hexa, the ortholog of human HEXA. This model was characterized by augmented macrophages that show migratory defects and enlarged lysosomes [148]. Analysis of different MOs in a wide range screening for angiogenesis inhibitors in zebrafish revealed that downregulation of hexb, the HEXB ortholog, induces defects in the vascular system at 48–56 hpf, as shown by FITC-dextran microangiography and by WISH analysis of the expression of the endothelial cadherin-5 encoding gene cdh5 in the intersegmental vessels [149]. More recently, a KO model of Sandhoff disease was generated using a CRISPR/Cas9 approach targeting hexb [150]. The enzymatic activity of hexa+/+/hexb−/− animals was reduced by 99% compared with controls, indicating that the hexa does not contribute significantly to the total β-hexosaminidase activity in zebrafish. Despite the lack of enzymatic activity, hexb null adult fishes are viable and show a normal morphological phenotype. However, mutant fishes showed an accumulation of different oligosaccharides in the brain and various internal organs. A more in-depth analysis performed on hexb KO embryos at 5 dpf evidenced abnormality in the lysosome morphology of the microglia and radial glia, probably associated with defects in the lysosome fusing process. A behavioural analysis of 4.5 and 6 dpf embryos showed a reduced locomotor activity of hexb KO animals compared with controls, an alteration that resembles the impaired locomotor function observed in Sandhoff patients [167]. Interestingly, the manifestation of this locomotor alteration is simultaneous with the appearance of lysosomal abnormalities in the radial glia, suggesting a correlation between glial function and locomotor activity. Moreover, hexb KO animals exhibit a slight increase in neuronal loss at 5 dpf that partially mimics the neurodegeneration observed in humans. The only zebrafish model established so far for MLD was obtained by injection of a splicing-MO specific for the arsa gene [148]. An initial characterization of this KD model showed an increased number of vacuolated macrophages presenting enlarged lysosomes compared with control embryos. Moreover, as observed in gba1 and hexa null animals, abnormal macrophages showed an impairment of movement associated with migratory defects. These results indicate that diverse lysosomal storage disorders may impair macrophage function with an impact on their anti-microbial function [148]. In this review, we have highlighted the use of zebrafish to develop new animal models of sphingolipidoses. Starting from a gene knockdown approach via MO injection at the one–two cell stage of embryonic development, the more recent use of the TALEN and CRISPR/Cas9 gene editing techniques has allowed to knock out enzymes involved in sphingolipid metabolism whose deficiency is responsible for various human hereditary sphingolipid disorders. Notably, many of these models recapitulate, at least in part, the phenotypic defects observed in patients (Figure 5). In addition, lipidomic analysis has allowed the study of the impact of enzymatic deficiencies on the sphingolipid metabolism in zebrafish, providing useful insights into the pathogenesis of these diseases. It must be pointed out that these studies can be performed not only in adult animals, but also in zebrafish embryos, thus providing invaluable information about the early biochemical alterations that may occur in patients before birth. At present, different therapeutic approaches, including HSCT, ERT, SRT, pharmacological chaperones, and in vivo and ex vivo gene therapy, are envisaged for patients affected by sphingolipidoses. However, given the complexities resulting from the alterations of sphingolipid metabolism in different systemic organs and the early appearance of serious pathological alterations in the infantile forms of various sphingolipidoses, more efficacious therapeutic strategies are required to improve patient outcomes. As described in numerous reviews [96,97], the zebrafish system includes several advantages that make this organism a powerful platform for the study of the pathogenesis of human hereditary diseases and for the development of novel drug-based therapeutic strategies. Indeed, as also pointed out in this review, most of the pathogenic processes of genetic diseases are conserved between humans and zebrafish, with high similarity among possible drug targets. Compared with cell-based and biochemical screening of putative drug candidates, zebrafish models offer the great advantage of providing a whole organism response to the delivery of drug candidates, thus also allowing the evaluation of side effects such as teratogenicity, toxicity, and metabolic alterations, as well as the study of drug pharmacokinetics and pharmacodynamics [168]. In addition, the zebrafish embryo offers multiple advantages that make this model attractive for a cost-effective drug screening, including external fertilisation, high fecundity, and ease of use; furthermore, embryo transparency enables imaging at cellular resolution and internal organ visualization. Given these features, zebrafish has been used as a tool for high-throughput screening of different drug candidates relevant to a broad range of human diseases [169]. Many of these molecules have reached the clinical trial phase, confirming the possibility of using zebrafish as a platform for the development of new potential therapeutic strategies. In this frame, even though some of the approaches envisioned for the therapy of sphingolipidoses cannot be modelled in zebrafish (like HSCT and gene therapy), various studies indicate the possibility to assess, in KO zebrafish mutants, the therapeutic potential of novel drugs to be used in an SRT approach. In addition, double KO zebrafish mutants harbouring the pathogenic mutation together with the genetic deficit of an upstream enzyme involved in the synthesis of the accumulating substrate may provide useful information about the possible efficacy of drug-driven SRT strategies. Moreover, zebrafish models of sphingolipidoses might be useful for the screening of pharmacological chaperones in zebrafish lines obtained by CRISPR/Cas9-based point mutation gene editing that harbour the identical pathogenic mutation detected in human patients. Clearly, given the obvious differences with humans, zebrafish models may not fully reflect the pathophysiology of the human disease. In addition, the duplication of the zebrafish genome may result in the presence of two co-orthologs of the human pathogenic gene. This may require the understanding of the biological role of both proteins encoded by the two orthologs during zebrafish development and in adults to evaluate how and to what extent the single or double KO mutants may mimic the human disease. Despite these and other drawbacks, the “zebra-sphinx” system represents an innovative and informative tool to gain insights into the biology of sphingolipid metabolism for a better comprehension of the pathological processes contributing to sphingolipid disorders, thus enabling the development of novel potential therapies and their translation to patients.

PMC10002608

Celine Pöhlking,Sebastian Beier,Jan Patrick Formanski,Michael Friese,Michael Schreiber,Birco Schwalbe

Isolation of Cells from Glioblastoma Multiforme Grade 4 Tumors for Infection with Zika Virus prME and ME Pseudotyped HIV-1

24-02-2023

glioblastoma multiforme,human cerebrospinal fluid,oncolytic viruses,pseudotyped virus,Zika virus,HIV-1

This study aimed to isolate cells from grade 4 glioblastoma multiforme tumors for infection experiments with Zika virus (ZIKV) prME or ME enveloped HIV-1 pseudotypes. The cells obtained from tumor tissue were successfully cultured in human cerebrospinal fluid (hCSF) or a mixture of hCSF/DMEM in cell culture flasks with polar and hydrophilic surfaces. The isolated tumor cells as well as the U87, U138, and U343 cells tested positive for ZIKV receptors Axl and Integrin αvβ5. Pseudotype entry was detected by the expression of firefly luciferase or green fluorescent protein (gfp). In prME and ME pseudotype infections, luciferase expression in U-cell lines was 2.5 to 3.5 logarithms above the background, but still two logarithms lower than in the VSV-G pseudotype control. Infection of single cells was successfully detected in U-cell lines and isolated tumor cells by gfp detection. Even though prME and ME pseudotypes had low infection rates, pseudotypes with ZIKV envelopes are promising candidates for the treatment of glioblastoma.

Isolation of Cells from Glioblastoma Multiforme Grade 4 Tumors for Infection with Zika Virus prME and ME Pseudotyped HIV-1 This study aimed to isolate cells from grade 4 glioblastoma multiforme tumors for infection experiments with Zika virus (ZIKV) prME or ME enveloped HIV-1 pseudotypes. The cells obtained from tumor tissue were successfully cultured in human cerebrospinal fluid (hCSF) or a mixture of hCSF/DMEM in cell culture flasks with polar and hydrophilic surfaces. The isolated tumor cells as well as the U87, U138, and U343 cells tested positive for ZIKV receptors Axl and Integrin αvβ5. Pseudotype entry was detected by the expression of firefly luciferase or green fluorescent protein (gfp). In prME and ME pseudotype infections, luciferase expression in U-cell lines was 2.5 to 3.5 logarithms above the background, but still two logarithms lower than in the VSV-G pseudotype control. Infection of single cells was successfully detected in U-cell lines and isolated tumor cells by gfp detection. Even though prME and ME pseudotypes had low infection rates, pseudotypes with ZIKV envelopes are promising candidates for the treatment of glioblastoma. Glioblastoma multiforme (GBM) is the most common and aggressive tumor of the central nervous system. The average survival time of patients with GBM is between 12–15 months [1,2]. The poor prognosis after surgical excision results from tumor recurrence, which is mainly caused by a subpopulation of highly tumorigenic cells, called GBM stem cells (GSCs) [3]. Without a cure, treatment is complicated and primarily consists of surgical removal followed by radiation and chemotherapy [4]. The GSC subpopulation, the so-called “tumor-initiating cells” [3], is responsible for tumor recurrence as they are resistant to therapeutic interventions [5,6,7] and therefore lead to the failure of conventional therapy [8,9,10] and the development of treatments [11,12]. Since GBM is extremely heterogeneous, it is impossible to identify cells that have moved away from the original tumor, even with the best staining and imaging techniques available. This makes safe and complete tumor resection impossible [13]. Successful therapy must target the remaining tumor-initiating cells that cannot be removed by surgery. One strategy is to administer drugs directly at the site of the tumor. Lower-grade gliomas, for example, are characterized by the presence of the mutated enzyme isocitrate-dehydrogenase-1 (IDH-1). The absence of IDH-1 activity makes the cells dependent on a secondary pathway to produce nicotinamide adenine dinucleotide (NAD) by the enzyme nicotinamide phosphoribosyltransferase (NAMPT). Shankar et al. (2018) have developed a system based on a mixed polymer of lactic acid and glycolic acid loaded with an inhibitor for NAMPT [14]. The ultimate aim is to administer the polymer in situ after surgery to release the drug over days or weeks. Such a strategy would allow the administration of NAMP-inhibiting drugs with an overall increased toxic potential. Normally, these drugs would cause severe side effects if administered intravenously. Shankar et al. reported that they had not observed such negative side effects when applying the NAMPT inhibitor to tumor sites in a murine animal model [14]. Besides NAMPT, enzymes of the metabolic pathway are perfect targets to inhibit tumor growth. These strategies are based on drug toxicity, with or without genotyping tumor cells [14,15]. One aim of genotyping strategies is the identification of mutated genes in GBM tumor cells that play a role in positive selection. Parallel mutations are DNA nucleotide substitutions that occur at the same gene location in tumor cells from different GBM patients. Such parallel mutations occur in low-grade gliomas for IDH1 as previously described, but can also be found in other genes for proteins like EGFR (epidermal growth factor receptor) [16], TP53 (human tumor protein 53) [17], PTEN (Phosphatase and tensin homologue deleted on chromosome 10) [16], and RB1 (retinoblastoma tumor suppressor) [16]. In addition, any mutation, such as silent mutations or mutations outside of genes, can be used to identify parallel mutations. Such mutations may not be relevant to the change in cellular phenotype, but they can be used as targets in a genotype-targeted approach to inhibit or kill tumor cells. Since these mutations distinguish GBM tumor cells from normal brain cells, a CRISPR/Cas9 (Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR associated protein 9) guiding RNA (gRNA), designed for one of these mutations, could inactivate these tumor cells. CRISPR/Cas9 “gene scissors” can bind to short sequence motifs on genomic DNA and cut double-stranded DNA at a site defined by the target sequence present in the gRNA molecule. This system can be used to inactivate genes because the cellular DNA repair mechanism will introduce mismatches, rendering them inactive. For this purpose, a special vector is needed to transfer the CRISPR/Cas9 system into the target cells. Human immunodeficiency virus type-1 (HIV-1) genome-based lentiviral vectors can be used to transfer the whole CRISPR/Cas9 system into eukaryotic cells by using artificial HIV-1 particles, called HIV-1 pseudotype. Pseudotyping is a technique for producing virus particles with a viral envelope that is not derived from the virus used to produce the virus core. An envelope commonly used for pseudotyping HIV-1 is the glycoprotein G from the vesicular stomatitis virus (VSV). VSVg pseudotyped HIV-1 is often used because almost all cells, including brain cells, are permissive to VSV or VSVg pseudotypes [18]. Interestingly, GBM stem-like cancer cells express higher levels of the Zika virus (ZIKV) receptor molecules Axl [19,20,21] and integrin αvβ5 [22,23]. HIV-1 pseudotypes, loaded with the two ZIKV envelope proteins, prM and E, showed infectivity for two glioma-derived cell lines, U87 and 86Hg39 [24]. In particular, the infection experiments using a lentiviral plasmid containing the firefly luciferase reporter gene have shown that HIV-1 genomes cleared of HIV-1 viral proteins (gag, pol, and env) can in principle be used for pseudotyping and gene transfer into glioma cells. However, only two cell lines were used in these studies. Therefore, it is an important task to show that fresh tumor cells isolated directly from tumors can, in principle, be infected with the ZIKV-HIV-1 pseudotype. For the pseudotype infection experiments, we collected tumor samples from GBM patients with IDH-1 wild-type and isolated cells from these tumors. There are currently no standardized methods for in vitro cultivation of tumor cells [25]. Therefore, we have developed a cell culture method for pseudotype infection studies using only human cerebrospinal fluid (hCSF) or hCSF as a supplement to the standard cell culture medium. Cerebrospinal fluid provides the brain with nutrients for proper neural functions and growth [26,27], and the use of standard or adapted cell culture media could compromise the functionality of isolated tumor cells [28]. For fast and efficient isolation of tumor cells, we also tested different surfaces to which the cells could adhere. Pseudotypes were generated by transfection of COS-1 cells with plasmids for the ZIKV envelope prME, HIV-1 gag, and pol genes, and for the HIV-1 viral genome. To monitor infection of the prME (Z3) pseudotype, firefly luciferase (Luc) or green fluorescence protein (GFP) was used as a reporter expressed from the viral genome. Additionally, we constructed a new ZIKV ME (Δpr) envelope lacking the pr part of the prME envelope. The corresponding Z3- and Δpr-HIVluc and HIVgfp pseudotypes successfully infected standard glioma cell lines and the freshly isolated tumor cells. Tumor cells were isolated from tissue samples and transferred directly into cell cultures after the surgical removal of the tumor. All patients were diagnosed with isocitrate-dehydrogenase-1 (IDH-1) wild type and diagnosed positive for Glioblastoma multiforme. Inhibition of the MGMT gene by methylation of the promotor leads to reduced O-6-methylguanine-DNA methyltransferase expression and reduced DNA repair activity. This may be relevant since such tumors have an increased sensitivity to therapeutic interventions. The MIB-1 labeling index indicates how many cells from a biopsy were mitotically active before surgery. Glial fibrillary acidic protein (GFAP) is a protein found specifically in brain cells and is not found outside the central nervous system (CNS). This protein is only released outside the CNS after cell death or brain injury and is therefore an additional diagnostic marker for glioblastoma multiforme. Most patients have received therapy according to the Stupp protocol [4] before surgery. Four tumor samples were collected from each of the twelve patients P01-P12 between June and September 2021. The tumor samples were found to be heterogeneous, as differences in growth and appearance of adherent cells were observed between patients and between the four tissue subsamples taken from each tumor. From each of the tumor samples, we obtained a mixed cell suspension using a 70 µm cell strainer. The cell suspensions were each divided into three portions and cultured in DMEM, 1:1 hCSF+DMEM, and hCSF. The first appearance of adherent cells was monitored between days one and four, and non-adherent cells were removed when the density of adherent cells reached 40–50%. Replication increased as cells began to connect with each other through newly developed long filopodia. The 20-day growth rates of our tumor cell cultures are shown in Table 1. A representative example of the isolation of cells from tumor tissue samples is shown for the tissue sample P-09, designated AKH-09. The tumor cells in this example were cultured in a 1:1 mixture of hCSF and DMEM with 10% FBS (FBS final concentration of 5%). On the first day, small, needle-shaped cells appeared (Figure 1A), which developed into a much longer shape within the next few days, surrounded by small non-adherent cells (Figure 1B). Adherent cells developed long filopodia, and lamellipodia began to grow (Figure 1C). From the first subculture, cells were seeded on 96-well plates for infection assays (Figure 1D). The visual phenotype of the isolated cells remains stable in the third subculture (Figure 1E) and cultures started from frozen stocks (Figure 1F). The cell shape and growth pattern varied significantly between the individual tissue samples. A common feature was the formation of distinct filopodia, which were typically 50–300 µm long (Figure A2). When cells began to make cell-to-cell contact, the filopodia became thicker, while others regressed, and cell replication increased significantly. Tumor cells with cell-to-cell contacts usually form only two or three filopodia, which contact cells in the immediate vicinity and form a meshed, netlike structure, or they form large, triangular clusters of adherent cells, leaving the center of the cluster empty. The isolated cells used for pseudotype infection experiments (AKH-01, -05, -09, -10, -12) were grown on a larger scale, and the corresponding cell pellets were embedded in paraffin wax for standard p53 and S100 tumor marker diagnostics (Figure A3). P53 and S100 detection was <10% in the cultured samples, with two exceptions: AKH-09 and -10, where 80% and 50% of the cells were positive for p53, respectively. ZIKV receptors were detected using Axl- (mAb clone C4A8) and integrin-αvβ5-(mAb clone P1F6) specific monoclonal antibodies, as shown in Figure 2. All cells tested were positive for Axl. In addition to these cells, we also tested our standard laboratory cell lines COS-1, VeroB4, and HEK293T for Axl expression. Axl was positive in all these cells, but they were clearly negative for integrin αvβ5. The overall expression of integrin αvβ5 was also very low in AKH-01 and AKH-05, and only a subset of the cells was positive. A very low level of integrin αvβ5 was detected in U87 and U138. AKH-09, -10, -12, and U343 showed similar levels of integrin αvβ5 as Axl. As mentioned above, we detected low overall integrin expression in both AKH-01 and AKH-05 cell cultures. Figure 3 shows that the low expression of integrin αvβ5 can be seen in the silent cells. Cells in their late dividing state show spots of high integrin αvβ5 expression. Three HIV-1 pseudotypes with the envelopes Z3 (prME), Δpr (ME), and as a control, VSVg, were used for the infection experiments. The amino acid sequences of the ZIKV capsid-to-pr and capsid-to-M transitions are shown in Figure 4. The complete prME sequence is shown in the Appendix A in Figure A1. The amino acid recognition site for proteolytic cleavage AMAAEI of the capsid-to-pr transition was retained in the prME expression vector. For the preparation of the capsid-to-M construct, the AMAAE sequence was coupled to the VTLPSHS start of the M domain, creating an AMAAEV recognition site. Using these two ZIKV envelope constructs, HIV-1 pseudotypes with firefly luciferase as entry reporters were produced using vectors pNLlucAM. For infection of glioma cell lines U87, U138, and U343, cells were seeded in 96-well plates and infected using cell culture supernatants containing the pseudotype particles VSVg-HIVluc, Z3-HIVluc, and Δpr-HIVluc. After pseudotype infection, firefly luciferase activity in the culture supernatants was low 24 and 72 h post-infection. Thus, even shortly before the lysis of the cells, no firefly luciferase was measured in the cell culture supernatants. The 24-h values represent the luciferase activity derived from the initial COS-1 transfection supernatants. The 72-h values represent the luciferase activity released during HIV genome integration and subsequent expression (Figure 5A,B). Only the VSVg-HIVluc infected cultures had a higher luciferase background in their supernatants, which was most likely due to the weak syncytia-inducing and cell lysis effects observed in these experiments. In all infection experiments, firefly luciferase activity in the cell extracts was 2–3 log10 higher than the so-called background activity detected in the supernatants tested at 24- and 72-h post-infection (Figure 5C). We also produced HIV-1 pseudotypes with the pNLgfpAM vector to show infections of single cells by their green fluorescence (Figure 5D). Compared to VSVg-HIVgfp infection rates, which were estimated at 5–9%, infection rates for Z3- or Δpr-HIVgfp were basically lower, ranging between 0.1 and 0.5%. The best results were obtained with U138 cells, showing an estimated infection rate of about 0.5% for Δpr-HIVgfp. Both infection experiments using luc or gfp as entry reporters showed that the ME envelope lacking the pr domain can promote entry into glioma cell lines. For pseudotype infection of AKH tumor cells, these were seeded into Cell+™ 96-well plates and infected using respective cell culture supernatants of COS-1 transfected cell cultures containing VSVg-HIVgfp, Z3-HIVgfp, and Δpr-HIVgfp. Pseudotype entry was monitored by green fluorescence, as shown in Figure 6. COS-1 cells transfected by pNLgfpAM and pCMV-VSVg developed gfp-positive syncytia, indicating that both plasmids within the transfected cells were productive. In Figure 6A, infections are shown for AKH-01, -05, and -10, and in Figure 6B, infections are shown for AKH-09 cells. Compared to VSVg-HIVgfp infections, cells in Z3-HIVgfp and ∆pr-HIVgfp infection experiments showed partial detachment (Figure 6B). As expected from the Z3-HIVluc experiments shown in Figure 4A–C, the number of Z3- and ∆pr-HIVgfp-related gfp-positive cells were significantly lower compared to VSVg-HIVgfp infections in AKH-01 cells. In Figure 6B, data from AKH-09 infections showed that the infection rate for Z3-HIVgfp was about one-third that of VSVg-HIVgfp. In Figure 6C, more examples of Z3-HIVgfp and ∆pr-HIVgfp single-cell infections of AKH tumor cells were exemplarily shown. For the pseudotype infection studies cells were isolated from Glioblastoma multiforme tumor samples and cultivated to establish cell cultures of adherent cells, designated AKH cells. Usually, DMEM and Neurobasal media are suitable for the in vitro cultivation of tumor cells [29]. Considering that DMEM with FBS addition is used in cell culture for rapid cell division but is not developed for post-mitotic cells such as neurons, the lack of proteins and ions required for optimal neuronal growth is one reason [25]. Furthermore, the presence of serum can induce neural stem cell differentiation [27]. Therefore, neurobasal media with various additives have been developed to ensure long-term survival as well as low cellular differentiation during in vitro cultivation [30]. However, several studies have shown that DMEM is often used as a basic medium with or without selected additives for the cultivation of neuronal cells. One possible reason is that DMEM/FBS is much cheaper than commercially available neurobasal media with expensive additives. Due to the great heterogeneity of GBM tumor cells, the choice of media primarily depends on the scientific task [25]. The main methodological objective of the present study was to obtain a high yield of adherent tumor cells that could be used for infection experiments as soon as possible. A recent study investigated the use of human cerebrospinal fluid (hCSF) as a sole medium or as a medium additive [28]. In the first weeks after the isolation of tumor cells, an increase in cell growth was observed compared to the use of DMEM (10% FBS) medium. Tumor cell proliferation was significantly increased in the presence of hCSF. Thus, the average density of the tumor cell population was reached more quickly. Bardy et al. (2015) also reported that a medium enriched with hCSF components supports basic cellular functions and the overall activity of human neurons [31]. This supports the assumption that hCSF provides the brain with a variety of important nutrients, enabling proper neuronal functions [26,27]. According to a report by Reiber et al. (2001), important proteins needed for cell growth are provided by hCSF [30,32], suggesting that hCSF would be an appropriate supplement to DMEM or Neurobasal media or could be used as a sole medium for the first phase of cell isolation. This is contrary to the often-stated necessity for a full-fledged and thus expensive culture medium, to establish a universal method for the isolation and long-term cultivation of GBM tumor cells [29]. In the present study, cultivation in DMEM with hCSF added showed an overall improvement in cell growth and tumor cell isolation during the first weeks of cultivation. However, when complete cell confluence was achieved, and cells were passaged 2–3 times, the cells adapted to the Cell+™ surface. As a result, after successful isolation, the tumor cells could be grown at the same rates in sole DMEM (10% FBS without hCSF). We, therefore, recommend the use of hCSF as a supplement or sole medium as the preferred starting medium for the isolation of GBM tumor cells for infection experiments with oncolytic viruses or viral pseudotypes. Another important factor for the isolation of tumor cells is the specific surface characteristics of the cell culture flasks [33]. To determine the best surface for adherent GBM cell growth, Sarstedt Cell™ (red cap), Cell+™ (yellow cap), Eppendorf CCC-FN1-coated surfaces, and surfaces coated with an extracellular matrix (ECM) created by 86Hg39 cells were used. The standard Cell™ polystyrene surface (Sarstedt Red Cap) was optimal for the growth of Glioma laboratory cell lines U87, U343, and U138. For these cell lines, no differences were observed between the different surfaces tested. In contrast, when cells were isolated from tumor tissue, a major difference in cell growth was observed. No adherent cells could be detected on the Standard Cell™ surface for more than 4 weeks. Therefore, the standard Cell™ polystyrene surface was an inappropriate surface for the selection of adherent cells. We then tested the Cell+™ surface (Sarstedt, yellow cap). The Cell+™ culture flasks are coated with polar groups in addition to the standard hydrophilic surface to mimic an in vitro environment that allows the adhesion of so-called fastidious primary cells [34]. Although Sarstedt does not provide information about the exact coating, it states that surface irradiation generates polar amino groups, which provide a closer resemblance to the in vivo microenvironment. For this study, we compared cell growth on different surfaces like Cell+™, ECM-coated, and fibronectin-coated. As mentioned earlier, in this study, cultivation on the Cell+™ surface was the most efficient and economical method for isolating tumor cells. To create a surface that mimics the microenvironment of Glioma cells, the standard surface of the cell culture flask can be ECM-coated [35]. We used the glioma cell line 86HG39 for the coating. AKH-01 through AKH-05 tumor cells were additionally cultivated on ECM-coated flasks and 24-well plates. The cultivation experiments showed neither recognizable differences nor any benefit compared to cellular growth on Cell+™ flask or 24-well plates. An important argument against this method was that tumor cell cultures can be accidentally contaminated by the 86HG39 cells used for the ECM-coating. Such contamination must be carefully avoided in rare infection events in the isolated tumor cells to be monitored. However, due to the more complicated procedure to produce the ECM-coating, Cell+™ flasks were still used for cultivation. The growth of AKH-05, -09, and -12 was also tested on Eppendorf CCC-FN1 24-well plates. Tumor cells grew particularly well on these surfaces, but the price difference and small cultivation area (available only in 24-well format) were negative aspects. In general, therefore, no advantage over the Cell+™ surface was observed by using the fibronectin-FN1-coated plates. FN1 plates mimic the cell attachment site of a native extracellular matrix and ensure passage over twenty-five passages without surface-induced cell differentiation [36]. In our experience, FN1 plates were well-suited for the isolation of tumor cells. Unfortunately, FN1 plates are no longer available, and since such coated plates are expensive, we did not search for an equivalent FN1 replacement. In summary, cell culture plasticware with the Cell+™ surface in combination with DMEM/FBS supplemented with hCSF, or hCSF alone was the method of choice to isolate and culture adherent cells from GBM tumor samples for our pseudotype infection studies. We previously demonstrated that VeroB4 and two cell lines derived from brain tumors, U87 and 86HG39, can be infected with four different prME-HIV-1 pseudotypes [24]. Based on these studies, we decided to use the prME pME-Z3 expression vector. The Z3 prME envelopes showed high infection rates for U87 and 86HG39 cell lines compared to Z2 and Z4. Since firefly luciferase, which was used as a reporter to detect pseudotype infection in our previous study, is not a suitable reporter to study infections of single cells, we now performed experiments expressing intracellular gfp. The present study provides further evidence that ZIKV-HIV pseudotypes could be a promising candidate for virotherapy targeting gliomas. In addition to the infection of cell lines, single-cell infections of isolated cells from tumor samples are another important proof of concept. Another important point is that the study clearly focused on cells from GBM tumors, which are highly malignant and resistant to treatment, even after the main tumor has been surgically removed. A therapeutic approach by Shankar et al. (2018) targets the remaining cells, aiming not to harm healthy tissue [14]. This approach will apply only to brain tumors showing the IDH-mutated phenotype and is therefore not suitable for tumors with wild-type, functional IDH-1. This is important because grade 4 GBM tumors all express functional IDH-1, and remaining GBM tumor cells can therefore not be inhibited by the Shankar approach. However, tumor samples from GBM are generally heterogeneous [33,37]. Due to the diversity of tumor cells, we collected four different tissue samples from each tumor, suggesting that at least one sample includes enough tumor cells and implying that contaminations that cannot be avoided most likely do not occur in all samples. Since GBM tumor cells do not necessarily show rapid cell growth or some samples did not contain enough tumor cells, it was found that some tissue samples did not provide enough adherent cells (AKH-04, 06, 07, 08) and therefore were not included in the infection studies. Owing to the large differences between samples, cultivation and passage, must be adapted in such a way that, once established, cell-to-cell contact should be maintained to avoid any collapse in cellular growth. Of the forty-eight tissue samples collected, fifteen cell cultures, not cell lines, were established with sufficient growth rates to allow the preparation of test plates. All these cultures were infected by the Z3-HIVgfp pseudotype. Consistent with the present knowledge, Axl was detected in all the cells [21], while integrin αvβ5 expression varied significantly. Some evidence suggests that Axl may mediate ZIKV entry into astrocytes [38]. However, Axl is not a universal major receptor for ZIKV since Axl is not required for ZIKV infection of neuronal cells [39] or infections in mouse models [40]. In studies using freshly isolated primary human GBM slices, integrin αvβ5 is an important molecular feature mediating infection. Blocking integrin αvβ5 by antibody attenuated ZIKV replication in these slices [41]. Thus, integrin αvβ5 was identified as an internalization factor that increases the ZIKV permissiveness of glioma cells [23,42]. To date, the precise molecular interactions that determine the individual steps of ZIKV entry into glioma cells have not been fully elucidated. Therefore, it is important to study pseudotype infections in fresh tumor cells. To our knowledge, the studies using Δpr-HIVluc or Δpr-HIVgfp are the first examples of successful tumor cell infections with a partially truncated ZIKV envelope complex. The Δpr, ME envelope is mimicking the ZIKV envelope as it appears after proteolytic maturation during virus budding. At this point, we argue that the pseudotype infection experiments with prME and ME serve overall as proof-of-principle rather than definitive therapeutic applications. However, the pseudotype model is a very well-suited method to optimize the ZIKV envelope in terms of its receptor affinity and especially its particle packing efficiency. The development of the Δpr pseudotype is another step toward the optimization of the pseudotype envelope. Further work will show whether the prM envelope protein can be completely omitted and whether only the E protein is required for an infectious pseudotype. In general, enveloped viruses acquire their envelope through a budding process in which ZIKV and HIV-1 are completely different. For viral particle budding, the envelope proteins must accumulate at the appropriate membrane before the final budding step. Zika virus envelope proteins contain transmembrane (TM) localization signals specific for integration into the membrane of the endoplasmic reticulum (ER), and budding takes place in the ER. In contrast, HIV-1 envelopes and the gag and gag/pol precursors are transported to the cellular membrane, where budding is initiated by linking viral membrane-associated proteins to a process called endosomal sorting complexes required for transport (ESCRT). These two different strategies may explain the low efficiency of ZIKV-HIV pseudotype production, which remains a challenge for achieving high pseudotype yields for flavivirus-HIV in general. An alternative approach was followed by Liu et al., in which stem and ancor regions of E were exchanged against the TM and cytoplasmic domain (CD) of the VSVg envelope protein. This ZIKV-VSV protein chimera had a kidney-specific binding affinity. Together with a lentiviral vector, efficient gene transfer was observed through the corresponding pseudotype. This suggests that functional HIV pseudotypes for glioma cells can likely be produced using a ZIKV E protein in combination with TM and CD sequences from cell membrane-integrated proteins [43]. Although infection rates are relatively low, ZIKV HIV pseudotypes represent a promising method for the treatment of glioblastoma and targeted gene transfer. Since infection rates are low, quantification is difficult. In positive infection experiments, we have identified between 1–5 gfp-positive tumor cells in about 1000 cells. This is in agreement with the observed differences in the measurements with luciferase as a reporter compared to the VSVg-HIV values. In isolated GBM tumor cells, it is even more complicated to quantify precisely, as the cell cultures are very heterogeneous by nature. Therefore, infection rates cannot be directly compared with rates in cell lines. As shown in Figure 6B, the VSV infection rates for the AKH-09 tumor cells are also very low. In comparison, the rates for ZIKV-HIV are one-third of the VSV-HIV rates. This shows that the different tumor cell cultures differ greatly from each other. However, accurate quantification of infection rates with, e.g., FACS requires much more efficient pseudotypes. As a result, greatly improving pseudotype efficiency is an important future goal. The experiments of Liu et al. [43], who describe a 100-fold higher efficacy of their E-TM-CD construct compared to VSVg, give hope for the development of a more efficient pseudotype. VSV is also formed at the outer cell membrane like HIV. However, VSVg does not contain the PTAP amino acid motif, unlike the HIV p6 protein. This reveals linking the viral budding complex to the TSG101 protein of the ESCRT machinery. Therefore, many more steps need to be taken to create a ZIKV envelope that, like the HIV-1 proteins, finds its way to the cell surface, is efficiently assembled and is finally released by the cellular ESCRT machinery. Historically, our experience in developing a working pseudotype protocol is in line with reports that we are currently unable to successfully produce infectious ZIKV-HIV-pseudotypes [44]. In line with these findings, we also failed when using the viral pNL4.3R-E- vector (nef−). ZIKV-HIV, when using identical vector concentrations, as described by Ruiz-Jimenez et al. (2021), we were again unable to detect infectious particles. Successful pseudotype generation established by Kretschmer et al. (2020) relies preferentially on the use of a nef+ viral background [45,46], a high vector concentration for prME expression using a modified version of pcDNA3.1 [47], the additional use of a gag/pol packaging vector, and an appropriate transfection protocol [24]. Since the ZIKV-HIV-pseudotypes infect glioma cell lines as well as freshly isolated cells from GBM tumors, it is a future task to enhance pseudotype efficiency by (i) optimizing the codon usage to enhance expression of the envelope [48], (ii) changing ER localization signals present in the envelope sequence [49], modifying signal sequences for outer membrane localization [48,50], and (iii) creating a link to the ESCRT machinery [51]. Since all these steps seem to take their time, the next achievable aim is to generate more efficient prME-, ME- and probably E-HIV-pseudotype particles targeting genes in freshly isolated GBM tumor cells by the CRISPR/Cas9 system [52,53,54,55,56] using the methodology described in this study. Twelve patients diagnosed with Glioblastoma multiforme (Glioblastoma, IDH-wild type) according to the WHO classification [57] were included in the study. The study design was approved by PV6041 by the Ethical Commission of the Hamburg Medical Chamber (Ethik-Kommission der Ärztekammer, Hamburg, Germany). Tumor operations were performed at the Asklepios Klinik Nord-Heidberg (Hamburg, Germany). For cell differentiation during surgery, tumor cells were stained with 5-aminolevulinic acid. After surgery, tissue samples were placed in sterile screw cap tubes with a standard cap (2 mL, Type H, Sarstedt, Nümbrecht, Germany) previously filled with 1.5 mL DMEM supplemented with 10% fetal bovine serum (FBS) (PAN-Biotech, Aidenbach, Germany). Tissue samples were collected from four different tumor regions, and each was placed separately in one of the reaction tubes. The tubes were placed in a 50 mL screw cap tube (Sarstedt, Nümbrecht, Germany) to avoid any contamination during transport. The tissue samples were immediately transported to the cell culture laboratory at the Bernhard Nocht Institute for Tropical Medicine (Hamburg, Germany) to start the cell cultivation procedure immediately. A sterile cell strainer (70 µm mesh size; Fisherbrand, Schwerte, Germany) was placed on top of a 50 mL sterile screw cap tube (Sarstedt, Nümbrecht, Germany). The tissue sample was pressed through the mesh in a circular motion using a sterile plunger flange from a 2 mL syringe (B. Braun, Melsungen, Germany). During the cell separation procedure, the strainer was rinsed multiple times with 2 mL of DMEM/10% FBS. The tube was filled up to 50 mL with DMEM/10% FBS, centrifuged for 10 min (1500 rpm, RT, Megafuge 3.0R, Thermo Scientific Heraeus, Schwerte, Germany), and cells were resuspended in medium. The prepared cell suspension was cultured in 25 cm2 filter cap cell culture flasks (T-25, Cell+™, Sarstedt, Nümbrecht, Germany), each containing 10 mL of (i) human cerebrospinal fluid (hCSF), (ii) a 1:1 mixture of hCSF and DMEM/10% FBS, and (iii) DMEM/10% FBS. Cells were grown in a 5% CO2 atmosphere at 37 °C. The cell suspension was monitored daily for the appearance of adherent cells. To change the culture medium from the mixed cell culture (adherent and non-adherent cells), non-adherent cells were transferred from the cell culture flask into a sterile 50 mL centrifuge tube (Sarstedt, Nümbrecht, Germany) and centrifuged for 10 min (1500 rpm, RT, Megafuge 3.0R, Thermo Scientific Heraeus, Schwerte, Germany). The cell pellet was suspended in a new medium and placed back in the cell culture flask containing the adherent cells. When the density of adherent cells reached 40–50%, the cell cultures were washed several times with DMEM/10% FBS to remove non-adherent cells. As cell growth increased, the medium was changed every seven days, or the cells were split into two cultures or transferred into larger 75 cm2 cell culture flasks (T-75, Cell+, Sarstedt, Nümbrecht, Germany). To split adherent cells, the medium was removed, and the cells were washed with phosphate-buffered saline (PBS) and treated with 1 mL of trypsin 0.05%/ethylenediaminetetraacetic acid (EDTA) 0.02% in PBS (PAN-Biotech, Aidenbach, Germany). After a 2–3 min incubation at room temperature, the cells were resuspended by up-and-down pipetting using 4 mL of cell culture medium and were finally transferred to a new cell culture flask. Images of adherent cells were taken with a fluorescence microscope (EVOS FL Auto, Thermos Fisher Scientific, Schwerte, Germany). Various cell culture flasks and 24-well plates with different surfaces were used for the growth of tumor cells. One method for the cultivation of neuronal cells is to use surfaces coated with extracellular matrix (ECM) [17]. Therefore, 86Hg39 cells were cultured to high density in DMEM/10% FBS. The cell lawns were treated with 0.5% triton X-100 (2 mL per 25 cm2 flask, 0.2 mL per 24-well) for 30 min and washed with PBS. Cell culture flasks and 24-well plates were further incubated with 0.25 M ammonium hydroxide (2 mL per 25 cm2 flask, 0.2 mL per 24-well) for ten minutes. After ammonium hydroxide treatment, the surfaces were washed four times with PBS. The flasks or 24-well plates were stored until use at 4 °C. Another method describes the cultivation of various types of stem cells using fibronectin (FN) [58] or RGD-peptide-coated surfaces [59,60]. A commercially available surface was used (Eppendorf CCC-FN1, Hamburg, Germany) with a surface coated with synthetic RGD-based motifs. The Cell+™ surface provided by Sarstedt (Nümbrecht, Germany) is promoted as a surface for “sophisticated adherent cells”. The Cell+™ surface is loaded with polar amino groups in addition to the hydrophilic polystyrene surface. Cell growth was measured on day 20 by counting adherent cells and expressed as cells/cm2 at + = <40, ++ = 40–160, +++ = 160–280, and ++++ = >280 cells/cm2. Teflon-coated 12-well microscope slides (Thermo Scientific, Schwerte, Germany) were washed with warm, soapy water followed by normal tap water. The slides were then washed with distilled water, isopropanol, and ethanol. They were then autoclaved. For cell culture, the slides were placed in 92-mm Petri dishes, and approximately 1–2 mL of a fresh cell suspension was added on top of the slides. Then, 10 mL of medium was carefully added, and the Petri dishes were stored at 37 °C and 5% CO2. After the appearance of adherent cells, slides were washed three times in PBS buffer to remove DMEM residues. After air drying, slides were placed in a 3.7% formaldehyde solution (30 min, RT) and washed three times with PBS. Fixed cells were then treated with Triton-X100 (0.1% in PBS) for 15 min at RT, and blocking was performed with PBS/5% BSA for one hour in the dark, followed by three washing steps with PBS 0.05% Tween 20 (PBST). Cells were first stained with phalloidin (Phalloidin-iFluor 555 conjugate, AAT Bioquest, Biomol, Hamburg, Germany). 25 µL of a phalloidin working dilution (1 µL/mL PBS/1% BSA) was used per well. Staining was performed for one hour in the dark. For receptor staining, the primary antibody against Axl (mouse mAb clone C4A8, Invitrogen, Thermofisher, Schwerte, Germany) or integrin αvß5 (mouse mAb clone P1F6, Abcam, Berlin, Germany) was diluted in PBST/1% BSA at 4 µL/mL or 10 µL/mL, respectively. For staining, 25 µL of the antibody working solution was added to each well, and the slides were incubated overnight at 8 °C in the dark. The slides were washed three times with PBST and incubated with 25 µL of the secondary antibody solution (2 µL goat anti-Mouse IgG H&L-Alexa 488, [ab150117, Abcam, Berlin, Germany]/mL PBST/1% BSA) for one hour at RT in a dark chamber. After washing the cells three times with PBST, the last step was to color them with a commercially available DAPI solution (ROTI Mount FluorCare DAPI, Carl Roth, Karlsruhe, Germany). Images of green, red, and blue fluorescent cells were acquired using a fluorescence microscope (EVOS FL auto imaging system, Thermofisher Scientific, Schwerte, Germany). Transfection of cells was carried out as described before in 6-well and 24-well formats [24]. COS-1 cells were transfected with plasmids for the HIV-1 virus core pNLlucAM or pNLgfpAM, provided by Nikolas Friedrich (Institute for Medical Virology, University of Zürich, Switzerland), a packaging plasmid psPAX2 (Addgene, #12260, Teddington, UK), a Zika prME and ME envelope plasmid (pME-Z3, -Δpr, Figure A1), and a VSVg envelope plasmid pCMV-VSV-G (Addgene, #8454) [19]. For transfection in 25 cm2 cell culture flasks (T-75, Cell+™, Sarstedt, Nümbrecht, Germany), COS-1 cells were seeded one or two days before transfection, to reach about 70–80% confluence. For transfection, 480 µL of ScreenFect™ Dilution Buffer and 480 µL of ScreenFectA (SFA) (Screenfect GmbH, Eggenstein-Leopoldshafen, Germany) were mixed in a 15 mL sterile screw cap tube (Sarstedt, Nümbrecht, Germany). In a 1.5 mL reaction tube, 480 µL of dilution buffer and the respective plasmid DNA—(i) 150 µg of the pME-Z3 envelope expression vector, (ii) 60 µg of the psPAX2 packaging vector, and (iii) 30 µg of the pNLgfpAM vector—were mixed. The DNA mixture was added to the SFA solution and rapidly mixed with at least ten pipette strokes. After 20 min of incubation, OptiPro serum-free medium (Gibco, Schwerte, Germany) was added to a final volume of 6 mL. The culture flask was prepared by discarding the medium and washing the cell layer once with PBS to remove the medium and FBS. The DNA/OptiPro SFM mixture was carefully added to the cell layer, and the flask was incubated at 37 °C and 5% CO2. After three hours, the transfection mixture was discarded. The cell layer was washed with PBS, and 12 mL of DMEM containing 10% heat-inactivated FBS (30 min, 56 °C) was added. The cells were incubated at 37 °C with 5% CO2. Transfection efficiency was monitored by fluorescence microscopy (EVOS FL Auto, Thermofisher Scientific, Schwerte, Germany). Pseudotypes with firefly luciferase as a reporter were prepared using pNLlucAM instead of pNLgfpAM according to this protocol and as described previously [24]. Cell culture supernatant was harvested 72 h post-transfection, and centrifuged at 10,000× g for 60 s (Eppendorf, Centrifuge 5415C, Hamburg, Germany). Supernatants were directly used for infection experiments or transferred into 1.5 mL high-speed centrifuge tubes (1.5 mL, microcentrifuge tube, Beckman Coulter, Krefeld, Germany). Centrifugation was carried out at 125,000 g for 4 h at 4 °C (TLA-55 rotor, Optima TL, Beckman Coulter, Krefeld, Germany). Pseudotype particles were directly used or stored for up to 7 days at 8 °C. The fetal bovine serum used for infection studies was heat-inactivated at 56 °C, under sterile conditions, for 30 min, shaking the tube every 10 min [33]. After heat-inactivation, the FBS was added to the DMEM medium at a final concentration of 10% (PAN-Biotech, Aidenbach, Germany). Two days before infection, tumor cells were seeded in 96-well plates (Cell+™, Sarstedt, Nümbrecht, Germany) in a total volume of 200 µL of the respective medium (hCSF; 1:1 HCSF/DMEM 10% FBS; DMEM 10% FBS) to reach about 70% confluence on the day of infection. Empty microplate wells were filled with 200 µL of PBS to prevent medium evaporation during cultivation. On the day of infection, the medium was discarded and 100 µL of pseudotype solution was added per 96-well plate. For infection, cells were incubated for 3 h at 37 °C with 5% CO2, and then 100 µL of cell culture medium was added. Cells were incubated for 7 days while infection events were observed at daily intervals by an automated multichannel fluorescence life cell imaging system (EVOS FL Auto Imaging System, Life Technologies, Fisher Scientific, Schwerte, Germany). GBM is a highly malignant brain tumor. Treatment usually consists of surgery, but despite maximum treatment with chemotherapy and radiation, the tumor recurs. A variety of potential gene targets have been identified to inhibit GBM tumor cells on a DNA level. The challenge remains to develop the appropriate vectors for successful delivery of genetic tools to hit these targets. Our study describes the development of a vector for gene transfer into GBM cells. Pseudotyping is a useful technology to study viral entry into cells by an envelope-specific process. The pseudotype transfer of a reporter gene such as gfp into the target cells allows a simple but also precise study of pseudotype entry and gene delivery. We constructed HIV particles that contain the prME or ME envelope of ZIKV. To study infection by these pseudotypes, a culture method was established using hCSF as a supplement or sole medium to isolate cells from GBM tumors. The isolated tumor cells were successfully infected by pseudotypes, as shown by the expression of the entry reporters gfp and firefly luciferase. Although the infection rates were low, these results provide further insight into the construction and use of ZIKV-HIV pseudotypes as a model for the development of an efficient oncolytic pseudotype or virus with a distinct GBM tropism.

PMC10002675

Shihui Liu,Toshihiko Matsuo,Takumi Abe

Revisiting Cryptocyanine Dye, NK-4, as an Old and New Drug: Review and Future Perspectives

23-02-2023

NK-4,anti-allergic,anti-cancer,anti-inflammatory,antiviral,dilated cardiomyopathy,anti-oxidative,neuroprotective effects,cryptocyanine dye,heterocycles

NK-4 plays a key role in the treatment of various diseases, such as in hay fever to expect anti-allergic effects, in bacterial infections and gum abscesses to expect anti-inflammatory effects, in scratches, cuts, and mouth sores from bites inside the mouth for enhanced wound healing, in herpes simplex virus (HSV)-1 infections for antiviral effects, and in peripheral nerve disease that causes tingling pain and numbness in hands and feet, while NK-4 is used also to expect antioxidative and neuroprotective effects. We review all therapeutic directions for the cyanine dye NK-4, as well as the pharmacological mechanism of NK-4 in animal models of related diseases. Currently, NK-4, which is sold as an over-the-counter drug in drugstores, is approved for treating allergic diseases, loss of appetite, sleepiness, anemia, peripheral neuropathy, acute suppurative diseases, wounds, heat injuries, frostbite, and tinea pedis in Japan. The therapeutic effects of NK-4’s antioxidative and neuroprotective properties in animal models are now under development, and we hope to apply these pharmacological effects of NK-4 to the treatment of more diseases. All experimental data suggest that different kinds of utility of NK-4 in the treatment of diseases can be developed based on the various pharmacological properties of NK-4. It is expected that NK-4 could be developed in more therapeutic strategies to treat many types of diseases, such as neurodegenerative and retinal degenerative diseases.

Revisiting Cryptocyanine Dye, NK-4, as an Old and New Drug: Review and Future Perspectives NK-4 plays a key role in the treatment of various diseases, such as in hay fever to expect anti-allergic effects, in bacterial infections and gum abscesses to expect anti-inflammatory effects, in scratches, cuts, and mouth sores from bites inside the mouth for enhanced wound healing, in herpes simplex virus (HSV)-1 infections for antiviral effects, and in peripheral nerve disease that causes tingling pain and numbness in hands and feet, while NK-4 is used also to expect antioxidative and neuroprotective effects. We review all therapeutic directions for the cyanine dye NK-4, as well as the pharmacological mechanism of NK-4 in animal models of related diseases. Currently, NK-4, which is sold as an over-the-counter drug in drugstores, is approved for treating allergic diseases, loss of appetite, sleepiness, anemia, peripheral neuropathy, acute suppurative diseases, wounds, heat injuries, frostbite, and tinea pedis in Japan. The therapeutic effects of NK-4’s antioxidative and neuroprotective properties in animal models are now under development, and we hope to apply these pharmacological effects of NK-4 to the treatment of more diseases. All experimental data suggest that different kinds of utility of NK-4 in the treatment of diseases can be developed based on the various pharmacological properties of NK-4. It is expected that NK-4 could be developed in more therapeutic strategies to treat many types of diseases, such as neurodegenerative and retinal degenerative diseases. NK-4 (1-ethyl-4-[(1Z,3E,5E)-1-(1-ethylquinolin-1-ium-4-yl)-5-(1-ethylquinolin-4-ylidene)penta-1,3-dien-3-yl]quinolin-1-ium;iodide, IUPAC name) (Figure 1) is a divalent, cationic pentamethine trinuclear cyanine dye that consists of three quinolinium rings, short N-alkyl side chains (C2), and two iodine anions [1]. It has been studied in Japan for over 100 years and has been popularly used as an over-the-counter drug since 1951. NK-4 exhibits a variety of biological activities, such as anti-allergy, anti-cancer (inhibition of cancer cell proliferation), anti-inflammation, antiviral infection, anti-oxidative, and neuroprotective effects [2]. Additionally, it has a potential to treat dilated cardiomyopathy and muscular dystrophy [2]. In this review, we reviewed all relevant literature on NK-4 drugs. Gell and Coombs’s classification divides allergies into four pathophysiological types: type I: anaphylaxis; type II: antibody-mediated cytotoxic reactions; type III: immune complex-mediated reactions; and type IV: delayed type hypersensitivity. A type I hypersensitivity is a hypersensitivity reaction that occurs within minutes after the sensitized body is exposed to the same antigen again. A type II hypersensitivity reaction is a pathological immune reaction in which IgM or IgG antibodies are combined to the corresponding antigen on the surface of target cells together with the participation of phagocytes, complement, and NK cells, leading to cell lysis or tissue damage. A type III hypersensitivity is an inflammation and tissue damage caused by the deposition of soluble immune complexes in the tissues, such as kidney, blood vessel wall, and skin, by activating the complement system; furthermore, it includes the participation of effector cells such as neutrophils and platelets, leading to cellular infiltration and localized necrosis. A type IV hypersensitivity reaction is an inflammatory reaction in which T cells are bound to corresponding antigens, leading to mononuclear cell infiltration and tissue cell damage [3]. A representative disease of the type I hypersensitivity reaction is allergic rhinitis (hay fever). Treatments of allergic rhinitis are, for instance, intranasal corticosteroids, oral and intranasal antihistamines, decongestants, intranasal anticholinergics, intranasal cromolyn, leukotriene receptor antagonists, combination therapy, immunotherapy, etc. [4]. Representative diseases of the type II hypersensitivity are Hashimoto’s disease and autoimmune hemolytic anemia (AIHA). Treatment for Hashimoto’s disease involves observation and medication, and the main therapy for Hashimoto’s disease is to control hypothyroidism, including the oral synthetic hormone levothyroxine 4 (L-T4) [5]. Prednisolone is recommended as the initial first-line treatment for primary warm AIHA. In the treatment of pathogenic B cell clones, rituximab monotherapy has become the most commonly used first-line therapy for cold AIHA. Nonpharmacological management includes thermal protection to limit hemolysis and relieve any ischemic symptoms [6]. The representative disease of the type III hypersensitivity is systemic lupus erythematosus (SLE). Treatments for SLE include use of immunomodulators (i.e., vitamin D and hydroxychloroquine), targeted therapy, and immunosuppressants [7]. The representative disease of the type IV hypersensitivity is allergic contact dermatitis. The primary treatment for allergic contact dermatitis is allergen avoidance. Databases such as the Exposure Allergen Management Program help patients choose allergen-free products. Treatment of acute exacerbations uses topical corticosteroids which are, however, not recommended to be used as a long-term treatment [8]. The studies investigated the immunopharmacological effects of NK-4 on type I and type IV hypersensitivity. The results demonstrated that NK-4 has a mild inhibitory effect on IgE antibody production, which is induced by heterologous passive cutaneous anaphylaxis (PCA) for 3 hours; NK-4 was shown to have a mild inhibitory effect on the homologous PCA response for 48 hours and was also demonstrated to have an inhibitory effect on the histamine release reaction by the in vitro antigen–antibody reaction in male Wistar rats (Charles River Laboratories Japan, Inc., Kanagawa, Japan). On the other hand, NK-4 significantly inhibited the cyclophosphamide (CY)-induced response from a type IV hypersensitivity reaction (delayed-type hypersensitivity (DTH)) model [1]. New Zealand white (NZB/W) F1 mice have been used as a model for autoimmune disease, such as SLE. NZB/W F1 mice produce autoantibodies such as natural thymocytotoxic autoantibodies (NTA) and anti-single DNA antibodies, and then develop into immune complex nephritis. The experiments show that NK-4 can significantly inhibit the level of NTA in the blood of NZB/W F1 mice while promoting the induction of suppressor T cells. On the other hand, NK-4 restored the anti-sheep red blood cells (SRBC) antibody response and anti-TNP-LPS PFC response in NZB/W Fl mice [9]. NK-4 exerts immunomodulatory effects by preventing T-cell damage and by directly activating dysfunctional B cells. Moreover, previous studies have investigated whether NK-4 plays a regulatory role in Th2 cell activation and effector function. The results showed that NK-4 appears to selectively eliminate IL-4 and IL-5 production by Th2 cells that have been activated by antigen or anti-CD3ε monoclonal antibody. These phenomena have been accomplished by means of inhibiting the mRNA expression of the Th2-related transcription factors GATA-3 and NFATc1. On the regulation of Th2 cell effector function, NK-4 inhibits the secretion of eotaxin and TARC from IL-4/TNF-α-activated fibroblasts by inhibiting the STAT6 signaling pathway [10]. These results provide evidence for NK-4 as a therapeutic agent for Th2-mediated allergic inflammation. Risk factors for cancer include chemicals, radiation, tobacco, excess alcohol, infections, stress, obesity, and more [11]. Cancer occurs as a series of consecutive genetic mutations that alter cellular function [12]. Cancer-related genes can promote cancer development when they are mutated, affect the cell cycle, and lead to abnormal proliferation. A tumor contains mutations in two to eight genes that promote tumorigenesis (driver genes). Driver genes can be divided into twelve signaling pathways that regulate three key cellular processes: cell fate, cell survival, and genome maintenance. Lines of evidence suggest that mutations in about 140 driver genes, such as ABL1, BRCA1, and CDKN2A, lead to cancer [13]. Additionally, a lack of tumor suppressor genes can lead to uncontrolled cell division [14]. From an epigenetic point of view, cancer cells are characterized by aberrant DNA methylation, which primarily targets CpG islands in regulatory elements of gene expression [15]. Past experiments have demonstrated that the expression of six genes (CLDN3, DECR2, EVA1B, NTSR1, NME4, and XPNPEP2) were highly significantly changed by alterations in DNA methylation, which can be detected by reduced representation bisulfite sequencing and RNA-seq techniques [16]. Near-infrared photoimmunotherapy (NIR-PIT) is a newly developed, molecularly targeted phototherapy based on the injection of a near-infrared conjugate IRDye700DX (IR700), which targets antigens expressed on the surface of cancer cells. NIR-PIT selectively destroys cancer cells, leading to immunogenic cell death, which elicits local immune responses as well as the reactivation of polyclonal CD8+ T cells against various released cancer antigens [17,18,19]. NIR-PIT not only induces immediate and highly selective cancer cell killing, but also stimulates highly effective anti-tumor immunity, thereby reducing side effects and helping patients avoid side effects associated with surgery, chemotherapy, and radiotherapy [17,18,19]. Many methods and drugs are available to treat cancer, and many more are being researched. Cancer treatment is mainly classified into local treatment, systemic treatment, and palliative care. Local treatments are used to treat cancer in specific body parts, such as surgery and radiation therapy. Systemic treatments can affect the entire body, such as chemotherapy, immunotherapy, or targeted therapy [20]. Palliative care is about improving the quality of life of patients by relieving pain and symptoms and by providing mental and psychological support [21]. High doses of NK-4 under light are destructive to cancer tissues, and low doses of NK-4 can effectively activate macrophages. Inflammatory lesions in photodynamic therapy with damaged and dead cancer cells which have still remained generate specific immunity, and dead cells and debris are cleared by macrophages. Studies have found that very small doses of NK-4 through photodynamic activation stimulate lymphocytes and activate macrophages, resulting in beneficial immune effects on the host organism. The activation of macrophage functions by low-dose NK-4 involves the mechanism of singlet oxygen [22]. Other studies have shown that reactive oxygen species did not participate in photodynamic cytocidal activity but that the activation of macrophages resulted from electron transfer between cationic dyes and cellular components [23]. NK-4 with an ethyl group on each quinoline structure is the most effective derivative, which can maximize the capability for the activation of macrophages. The conditions for the activation of macrophages include a 660 nm red laser or 780 nm near-infrared laser light [22,24]; a 670 nm red laser or 780 nm near-infrared laser are needed for cancer treatment by NK-4 [24,25]. Human lung cancers were transplanted into nude mice, and NK-4 was injected into the cancers six times after the cancers became hypertrophic. Cancers were exposed to a near-infrared laser (2 mW, 1 min) every other day for two weeks. After the first photoimmunotherapy treatment, the mice were free to drink water containing NK-4; then, the mice were allowed to drink water containing NK-4 every day. The results showed that the use of a low dose of NK-4 and laser treatment significantly enhanced the activity of macrophages, thereby increasing the immune effect. At the same time, the cancer is scarred by collagens, which have been produced from fibroblasts in the stroma [24]. Photoimmunotherapy is effective for the treatment of local deep cancer by the developed needle-type system in the presence of NK-4. At the same time, NK-4 can promote the differentiation of macrophages and lymphocytes, help healing, and improve immune function [25]. Inflammation based on time course is mainly divided into acute inflammation, subacute inflammation, and chronic inflammation. Acute inflammation occurs immediately after injury and persists for several days. Subacute inflammation is the transition from acute to chronic, lasting from 2 to 6 weeks. Chronic inflammation may persist for months or even years [26,27,28]. Acute inflammation is characterized by vasodilation, neutrophil infiltration, and fluid exudation [29]. Molecular mechanisms of inflammation are primarily initiated by the identification of characteristic molecular patterns associated with tissue damage or infection. After a generation of inflammatory response, natural innate immunity cells, such as neutrophils, macrophages, CD8+ T lymphocytes, and natural killer cells, provide an early response to noxious factors to eliminate noxious stimuli [30]. The pathological mechanisms of chronic inflammation are mainly related to stress response, adaptive immunity, and damage-associated molecular patterns [31,32,33]. At present, anti-inflammatory drugs mainly include non-steroidal anti-inflammatory drugs such as aspirin, indomethacin, ibuprofen, naproxen, diclofenac, celecoxib, etoricoxib, and mefenamic acid [34,35], which exert anti-inflammatory effects by inhibiting the synthesis of prostaglandins, inhibiting the aggregation of leukocytes and reducing the formation of bradykinin [36,37]. Wound healing includes granulation tissue proliferation, scar tissue formation, and the regeneration of various tissues. The basic process of wound healing is as follows: acute inflammation stage → cell proliferation stage → scarring stage → epidermis and tissue regeneration. There are three main types of wound healing: primary healing, secondary healing, and tertiary healing. Inflammation is part of the physiological phase of wound healing; its purpose is to attract different immune cells to remove debris and pathogens from the wound and to create an ideal environment for the differentiation of keratinocytes and fibroblasts, which finally leads to their migration to close the wound. The main topical medicines used for wound healing include medical device dressings and hyperbaric or negative pressure oxygen therapy [38]. In recent medical research, there is still a shortage of oral medications that directly improve wound healing. Among the available oral medications, most of them play an adjunctive role, such as infection relief, nutrition, and pain management [39]. In Japan, NK-4 has been used as an oral therapeutic agent to promote wound healing. Previous studies have shown that interferon-gamma (IFN-γ) production by splenocytes can be enhanced by the oral administration of NK-4 to male BALB/c mice in which the splenocytes have been stimulated with lipopolysaccharide (LPS). This phenomenon may be related to the activation of T cells by IL-12 produced by macrophages [40]. Another study elucidates the underlying mechanisms of NK-4 for wound healing. This study demonstrates that NK-4 drives macrophage polarization toward an inflammatory M1-like phenotype to increase macrophage phagocytic activity in the tests using the human monocytic cell line THP-1. This study also shows that NK-4 has the potential for treating persistent inflammation in chronic wounds [41]. Viruses are non-cellular organisms composed of a nucleic acid molecule (DNA or RNA) and proteins. It can only synthesize its own nucleic acid and protein components by using the metabolic system in the host’s living cells and can only reproduce itself in large quantities. Under the condition of leaving the host cell, it can exist in the state of inanimate biological macromolecules and maintain its invasive and infectious viability for a long time. The nucleic acid of some viruses can also integrate into the genome of the host cell, then inducing latent infection [42]. According to the classification of strategies, antiviral drugs are mainly concentrated in two directions: targeting the virus itself or the host cytokines [43]. Based on the mechanism of antiviral drugs, the current antiviral drugs can be divided into the following categories: (1) preventing viruses from penetrating into the host cell and from inhibiting the uncoating (amantadine, rimantadine) [44]; (2) DNA polymerase inhibitors (acyclovir, ganciclovir) [45]; (3) nucleoside reverse transcriptase inhibitors (lamivudine, emtricitabine), and non-nucleoside reverse transcriptase inhibitors (efavirenz, nevirapine) [46,47]; (4) neuraminidase inhibitors (oseltamivir, zanamivir) [48]; (5) protein inhibitors saquinavir [49]; and (6) broad-spectrum antiviral drugs (interferon, ribavirin) [50]. Ushio et al. validated the antiviral effect of NK-4 and used herpes simplex virus (HSV)-1 to produce pathological effects on human amniotic fluid cells as a model. The experimental results showed that NK-4 had no direct inhibitory effect on HSV-1, but mainly due to an indirect effect mediated by fluid cells to reduce HSV-1 replication by a dose-dependent manner. Furthermore, NK-4 itself significantly induced the alkalinization of intracellular organelles, leading to the inhibition of viral entry into cells. NK-4 also enhances the antiviral effects of interferon (IFN)-α [51]. Dilated cardiomyopathy (DCM) is a primary cardiomyopathy of unknown cause. It is characterized by the progressive enlargement and exacerbation of the left, right, or bilateral ventricles, leading to myocardial contractile dysfunction with or without congestive heart failure [52]. Death from dilated cardiomyopathy can occur at any stage of DCM. Common causes of DCM include viral infection of cardiomyocytes, genetic inheritance, and autoimmune disease. Pathological factors for left ventricular expansion are mainly related to remodeling and fibrosis. The pathological mechanisms of DCM mainly include: (1) genetics: the most common genes responsible for DCM are TTN, BAG3, TNNT2, MYH7, RBM20, LMNA44, PRDM16, etc. [53]; (2) autoimmunity: immune cell infiltration, aberrant expression of adhesion molecules, or HLA II in the heart was found in 50% of biopsy samples. According to the Rose–Witebsky criteria, DCM may be caused by an autoimmune condition [54]; (3) infection: infectious factors (mainly myocarditis) account for approximately 30% of the pathophysiology of DCM. Common groups of viruses associated with DCM include parvovirus B19, enteroviruses, herpesviruses, and adenoviruses [55]; (4) inflammation: inflammation associated with autoimmunity and viral infection is involved in the pathogenesis of DCM. Myocardial biopsy specimens from patients with DCM showed that high expression of tenascin-C resulted in poor patient survival [56]; and (5) exposure to toxins and chemicals, such as metals (mercury, lithium, antimony, cobalt), scorpion venom, antidiabetic drugs, anticancer drugs, antiretroviral agents, cocaine, ethanol, methamphetamines, and carbon monoxide. The mechanisms of toxic cardiomyopathy include the production of reactive oxygen species (ROS), intracellular calcium handling, interference with mitochondrial respiration in cardiomyocytes, neurohormonal stress, genetic susceptibility, and apoptosis [57]. The main treatment methods for dilated cardiomyopathy include: (1) controlling blood pressure, improving blood flow, and reducing the burden on the heart. Antihypertensive drugs include angiotensin II receptor blockers (ARBs), angiotensin-converting enzyme (ACE) inhibitors, beta-blockers, and sacubitril; (2) controlling the heart rhythm, strengthening the contraction of the heart muscle, slowing down the heartbeat, and reducing the symptoms of heart failure, such as digoxin; (3) preventing blood clots and anticoagulants such as warfarin and direct oral anticoagulants; (4) reducing fluid in the body and improving dyspnea caused by pulmonary hypertension, such as from diuretics; and (5) surgery and heart transplantation [58]. Transient receptor potential vanillin 2 (TRPV2) is a prime candidate for aberrant Ca2+ entry pathways and a potential target for the treatment of DCM [59]. NK-4 is one of the TRPV2 inhibitor candidates. Experiments demonstrate that low-dose NK-4 inhibits TRPV2 channel activity, which in turn inhibits abnormally increased Ca2+ influx, prevents the progression of DCM in dystrophic hamsters (J2N-k), and improves cardiac function [2]. It is noteworthy that the structure of NK-4 (Figure 2A) has a 1,4-dihydropyridine moiety-like structure which is similar to a representative Ca channel inhibitor, nifedipine (Figure 2B). Thus, the TRPV2 channel inhibitory activity of NK-4 might be derived from the 1,4-dihydropyridine moiety. Oxidative stress is considered to be an imbalance between the oxidative stress and antioxidative systems of cells, leading to inflammatory infiltration of neutrophils, increased secretion of proteases, and the production of large amounts of reactive oxygen species (ROS) and reactive nitrogen species (RNS). In particular, oxygen free radicals can cause damage to the phospholipids, proteins, enzymes, and nucleic acids of cells, resulting in cell dysfunction. There are many markers of oxidative stress, including lipid hydrogen peroxide, isoprostanes, and more [60,61,62]. According to the solubility, antioxidants are divided into oil-soluble (butylated hydroxyanisole (BHA) and butylated hydroxytoluene (BHT)) [63] and water-soluble (tea polyphenols, ascorbic acid, ascorbyl palmitate) [64,65,66]. Antioxidants can also be divided into synthetic antioxidants (butylated hydroxyanisole (BHA); butylated hydroxytoluene (BHT); propyl gallate (PG); tertiary butylhydroquinone (TBHQ)) [67] and natural antioxidants (caffeic, rosmarinic acids, carnosol, quercetin, eugenol, anthocyanin) [68,69]. On the basis of the mechanism of action of antioxidants, antioxidants are classified into free radical scavengers [70], hydrogen peroxide scavengers [71], metal chelating agents [72], enzymatic and non-enzymatic antioxidants [73], and singlet oxygen quenchers [74]. Previous studies have shown that NK-4 has been tested to have significant hydroxyl radical scavenging activity, whereas NK-4 has also been demonstrated to be effective at scavenging peroxyl radicals in vitro by electron spin resonance (ESR) techniques [75]. In vivo, NK-4 was intravenously injected twice in an animal model of ischemic stroke (the middle cerebral artery occlusion (MCAO) model rat), which was induced by the temporary ligation of middle cerebral artery, followed by the reperfusion. NK-4 reduced the infarct volume by 57.0%. The results demonstrate that NK-4 can prevent cerebral ischemic injury and reduce cerebral ischemic damage by reducing reactive oxygen species (ROS), including superoxide (•O2−), hydroxyl radicals (•OH), and hydrogen peroxide (H2O2) (Figure 3A–C). NK-4 also reduces ischemic swelling of the brain hemispheres [75,76]. Another study evaluated the protective effects of NK-4 on oxidatively damaged nerves in vitro and in vivo. In vitro, NK-4 has free radical-scavenging activity by means of clearing hydroxyl radical, peroxy radical, and superoxide. The studies which compared NK-4 with other neuroprotectants showed that NK-4 has significantly higher hydroxyl radical scavenging activity than ascorbic acid and edaravone (Figure 3D). In vivo, the activation of PI3K and its downstream signaling effector Akt may be designated as a key mediator system that is beneficial to neuronal survival by NK-4 injection, and the antioxidant properties of NK-4 may also be associated with neuronal survival and functional maintenance [77]. In a recent study, researchers injected NK-4 into the eyes of Royal College of Surgeons rats (a model rat of retinitis pigmentosa) that exhibit inherited retinal dystrophy. The results show that NK-4 delays photoreceptor apoptosis through anti-oxidation, the maintenance of intracellular ion homeostasis, and other mechanisms [78]. Neurodegenerative diseases are caused by the loss of neurons, myelin sheaths, and synapses. Neurodegenerative diseases can be caused by aging and genetic mutations, and the condition of the diseases worsens over time, leading to functional impairment [79]. Common pathogenic mechanisms of neurodegenerative diseases include: (1) abnormal protein dynamics (protein misfolding and aggregation); (2) oxidative stress (formation of reactive oxygen species and free radicals); (3) dysfunction of neurotrophic factors; (4) mitochondrial dysfunction; (5) neuroimmune inflammation; (6) neuronal Golgi breakdown; (7) disruption of cell/axon transport; and (8) altered cell signaling. Altogether, the diversity of multiple pathogenic factors leads to multifaceted neuronal death [80]. The main research areas of neurodegenerative diseases include: (1) tau protein disease—Alzheimer’s disease (AD); (2) extrapyramidal disorder: Parkinson’s disease (PD), Huntington’s disease (HD); (3) spinocerebellar degeneration: multiple system atrophy (MSA); (4) autonomic disorders: Shy-Drager syndrome (SDS); and (5) motor neuron disorders: amyotrophic lateral sclerosis (ALS), Werdnig–Hoffmann disease. Ophthalmological neurodegenerative diseases mainly include retinitis pigmentosa (RP). The main drugs for neurodegenerative diseases include: galantamine, rivastigmine, and donepezil for Alzheimer’s disease [81,82,83]; levodopa, monoamine oxidase-B inhibitors, and dopamine agonists for Parkinson’s disease [84]; tetrabenazine (Xenazine) and deutetrabenazine (Austedo) for Huntington’s disease [85]; fingolimod (Gilenya), dimethyl fumarate (Tecfidera), and teriflunomide (Aubagio) for multiple sclerosis (MS) [86]; and Radicava, rilutek, exservan, nuedexta, and tiglutik for amyotrophic lateral sclerosis [87]. As a therapeutic drug for retinitis pigmentosa, Luxturna® (voretigene neparvovec) is the only Food and Drug Administration (FDA)-approved retinitis pigmentosa therapy, designated for a small subset of patients with RPE65 mutations [88]. On 23 June 2022, the FDA published a 5-year action plan for drugs of neurodegenerative diseases, focusing on ALS [89,90]. Therefore, with the deepening of neurodegenerative disease research, multi-pathway and multi-target therapeutic drugs urgently need to be developed. In a report, besides the neurotrophic and neurogenesis activity of NK-4 observed in a transgenic mouse model of Alzheimer’s disease (Tg 2576), the effect of NK-4, which was better than acetylcholinesterase inhibitors (AChEIs), was also observed in the early stages of mouse dementia (6 months old). NK-4 may be a new drug for the treatment of early- to late-stage Alzheimer’s disease [91]. Another study showed that NK-4 had neurotrophin-like activity and exhibited neuroprotective effects in vitro and in vivo. In vitro, NK-4 significantly enhanced nerve growth factor (NGF)-induced neurite outgrowth in PC12HS cells. In vivo, NK-4 effectively prevented injury in a rat stroke model (middle cerebral artery occlusion (MCAO) Rats) through neurotrophin-like activity and antioxidative activity [75]. In vitro, NK-4 was shown to dose-dependently protect PC12 cells from oxidative stress-induced toxicity by 6-hydroxydopamine (6-OHDA) or hydrogen peroxide (H2O2). In an ataxia animal model (Syrian hamster marked by Purkinje cell degeneration, PCD model) of neurodegeneration, the studies showed that the neuroprotective effects of NK-4 are mediated by the PI3K-Akt signaling pathway [92]. NK-4 can also reduce the accumulation of Aβ in the brain, inhibit Aβ aggregation, scavenge free radicals, and produce neuroprotective effects by its intraperitoneal injection in Alzheimer’s disease model AβPP transgenic mice (Tg2576). It is thus suggested that NK-4 can also be used to treat Alzheimer’s disease [77,93]. In a recent study, researchers administered NK-4 into the eyes of RCS rats via intravitreal injection; the researchers found that NK-4 could inhibit the apoptosis of photoreceptor cells. Hmox1, Mt1, Atf5, Slc7a11, and Bdh2 genes were up-regulated by the RNA-seq analysis and confirmed by the RT-PCR analysis. Functional and pathway enrichment analyses of up-regulated genes in that study suggest that the neuroprotective effect of NK-4 in RCS rat retina might be related to the retinal pigment epithelial metabolic process, transition metal ion homeostasis, and negative regulation of neurons’ apoptosis by Metascape analysis. They also uploaded five genes (Hmox1, Mt1, Slc7a11, Bdh2, and Atf5) to the DAVID database for the functional annotation clustering of bioinformatics resources. Based on the gene function distributed by DAVID, it was divided into the following categories: response to oxidative stress, negative regulation of neuron apoptotic process, and iron ion homeostasis [78]. All of these results revealed the molecular mechanism by which NK-4 inhibits the apoptosis of photoreceptor cells, indicating that NK-4 upregulates genes involved in anti-oxidative stress and anti-apoptotic pathways. The synthetic route for NK-4 is shown in Figure 4 [94,95]. 1-Ethyl-4-methylquinolin-1-ium iodide was treated with Vilsmeier reagent generated in situ from P(O)Cl3 and N,N-dimethylformamide, affording the desired NK-4 at a 41% yield (Figure 4A). A different C1 unit protocol was also reported (Figure 4B). However, these routes could be used for the same three-quinoline moiety, but not for hetero-quinoline moieties. In the future, the development of a new synthetic methodology is highly required to supply various derivatives which have different quinoline cores (Figure 4C). In nearly 70 years of research, NK-4 has been developed for various pharmacological effects, including anti-inflammatory, anti-allergic, anti-cancer, wound healing, antiviral, antioxidative, and neuroprotective effects. NK-4 is a good candidate for treating various diseases, and it is expected that the pharmacological properties of NK-4 can be applied to the treatment of many more diseases, such as neurodegenerative and retinal degenerative diseases. With respect to the antioxidative effect, NK-4 has a higher hydroxyl radical scavenging activity compared with other antioxidants. To plan experiments for assessing the antioxidative and neuroprotective effects of NK-4, more animal models are needed to verify these effects and the pharmacological mechanisms of NK-4, and to proceed towards the goal of successfully entering clinical trials. Overall, this review provides a summary of the various functions of NK-4 as insights for the development of potential therapeutic agents.

PMC10002748

Andrea R. Daamen,Prathyusha Bachali,Amrie C. Grammer,Peter E. Lipsky

Classification of COVID-19 Patients into Clinically Relevant Subsets by a Novel Machine Learning Pipeline Using Transcriptomic Features

03-03-2023

COVID-19,severity,classification,transcriptomics,bioinformatics,machine learning

The persistent impact of the COVID-19 pandemic and heterogeneity in disease manifestations point to a need for innovative approaches to identify drivers of immune pathology and predict whether infected patients will present with mild/moderate or severe disease. We have developed a novel iterative machine learning pipeline that utilizes gene enrichment profiles from blood transcriptome data to stratify COVID-19 patients based on disease severity and differentiate severe COVID cases from other patients with acute hypoxic respiratory failure. The pattern of gene module enrichment in COVID-19 patients overall reflected broad cellular expansion and metabolic dysfunction, whereas increased neutrophils, activated B cells, T-cell lymphopenia, and proinflammatory cytokine production were specific to severe COVID patients. Using this pipeline, we also identified small blood gene signatures indicative of COVID-19 diagnosis and severity that could be used as biomarker panels in the clinical setting.

Classification of COVID-19 Patients into Clinically Relevant Subsets by a Novel Machine Learning Pipeline Using Transcriptomic Features The persistent impact of the COVID-19 pandemic and heterogeneity in disease manifestations point to a need for innovative approaches to identify drivers of immune pathology and predict whether infected patients will present with mild/moderate or severe disease. We have developed a novel iterative machine learning pipeline that utilizes gene enrichment profiles from blood transcriptome data to stratify COVID-19 patients based on disease severity and differentiate severe COVID cases from other patients with acute hypoxic respiratory failure. The pattern of gene module enrichment in COVID-19 patients overall reflected broad cellular expansion and metabolic dysfunction, whereas increased neutrophils, activated B cells, T-cell lymphopenia, and proinflammatory cytokine production were specific to severe COVID patients. Using this pipeline, we also identified small blood gene signatures indicative of COVID-19 diagnosis and severity that could be used as biomarker panels in the clinical setting. Since its emergence in December 2019, infection by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of coronavirus disease 2019 (COVID-19), has led to the deaths of over 6.5 million individuals worldwide [1,2,3]. COVID-19 is characterized by a wide range of disease presentations from mild with flu-like symptoms to severe with acute hypoxic respiratory failure (AHRF) requiring hospitalization and admittance to the intensive care unit (ICU) [4]. In addition, symptoms may persist for months after patients present with acute illness [5]. A number of pre-existing clinical risk factors (age, gender, obesity, respiratory conditions, diabetes, immunodeficiency) [6,7] and post-infection multi-organ disease complications, including cardiac injury, thrombosis, renal disease, and liver injury [8,9,10], have been associated with severe COVID. Despite intense study, however, the major drivers of severe and potentially fatal COVID-19 have not been fully elucidated. Considerable resources have been applied to study the immune response to SARS-CoV-2 infection as a means to understand COVID-19 pathogenesis in greater detail [11,12]. Among these efforts, our group and others have employed a bioinformatics-based approach using multi-omic, and in particular, transcriptomic data to construct immune profiles of COVID-19 patients at various stages of disease onset and severity [13,14,15,16,17,18,19,20,21,22]. These studies have identified numerous abnormalities in the strength and nature of the adaptive and innate immune responses of severe COVID patients, including neutrophilia, T-cell lymphopenia, plasmablast expansion, antibody or autoantibody production, and IFN levels. With this accumulated knowledge base, the focus has evolved to attempt to apply the information to identify subsets of individuals with COVID-19 with prognostic relevance. Recent advances in computer science and the use of artificial intelligence (AI) and machine learning (ML) for clinical applications offer a promising approach to identify biomarkers and predict the risk of SARS-CoV-2 infected individuals to develop severe and possibly fatal disease [23,24]. Several recent studies have employed ML or deep learning approaches to patient demographic data [25], lung CT scan images [26,27,28], and transcriptomic data [29,30,31,32,33] to stratify COVID-19 patients with varying degrees of success. However, more work is required to improve model performance and reproducibility before biomarkers of severe COVID-19 identified by ML can be translated into a clinical setting. This is dependent on: (1) developing a standardized approach to dealing with high dimensional data; (2) selecting the best ML algorithm for each application; and (3) selecting the most informative features as input for ML applications. To that end, we have developed an iterative ML pipeline based on curated gene signatures previously used to characterize immune profiles of COVID-19 patients from whole blood gene expression data [13,14]. As a result, we have identified immunological signatures and individual genes with strong predictive capacity for classifying COVID-19 patients most at risk of severe disease that could be employed as a diagnostic and prognostic tool in the clinic. We employed three publicly available whole blood transcriptome datasets (GSE161731, GSE172114, and PRJNA777938) that had previously been analyzed for immune profiling of COVID-19 patients based on gene signature enrichment by gene set variation analysis (GSVA) [14]. The combined datasets consisted of individuals with COVID-19, individuals with non-COVID-induced AHRF, individuals admitted to the ICU without AHRF, and healthy controls. COVID-19 patients were further categorized as “non-critical” if they exhibited symptoms but were not hospitalized or were admitted to a non-critical care ward and as “critical” if they exhibited more severe symptoms requiring hospitalization and admittance to the ICU. To determine the top immune signatures and genes differentiating subsets of COVID-19 patients from healthy individuals and other ICU patients, combined GSVA enrichment scores for 40 curated immune cell and pathway gene signatures [13,14] were used as features to train 9 ML algorithms (Figure 1, Table S1). Each ML algorithm was used for 4 binary classifications: COVID patients versus healthy individuals, non-critical COVID patients versus healthy individuals, critical versus non-critical COVID patients, and COVID ICU patients versus non-COVID ICU patients. Then, the top 5 performing ML algorithms were employed in an iterative approach to identify the GSVA modules contributing most to each classification (Figure 1). After each iteration, feature importance was calculated for each ML algorithm and the top 50% of features were used for the next round of ML. Then, the log2 expression values of genes composing the final 10 gene modules were used as features for ML algorithms to identify the top 20 individual genes that could effectively classify COVID-19 patients in each comparison. Expression values used as input for the final ML iteration were normalized to account for batch effects between separate RNA-seq datasets (Figures S1–S4A). To ensure reproducibility of the results, each ML iteration was repeated 10 times and the most represented modules or gene features were used moving forward. Overall, the ML algorithms successfully performed each binary classification and the top performing algorithms for each comparison achieved accuracies of >0.7 for all iterations of ML. In addition, accuracies improved between the first and last ML iteration as the gene modules were refined and distilled into individual gene features, and final accuracies reached >0.9 for the top 5 algorithms in each classification (Table S2). Notably, the final gene lists derived from this iterative ML approach reflected distinct immune cell and inflammatory pathways providing the greatest contribution to each binary classification of COVID-19 patients (Table S3). The iterative ML pipeline was first applied to determine the top modules and individual genes for classification of COVID-19 patients from healthy individuals. For the initial iteration, support vector machine (SVM) was consistently the top performing algorithm with average areas under the receiver operating characteristic and precision/recall curves (AU-ROC and AU-PR) of 1.0 (Figure 2A) as well as an average overall accuracy of 1.0, indicative of perfect model performance across 10 repetitions (Table S2). The 10 modules with the highest feature importance were divided between pathways related to cell proliferation/metabolism (Cell Cycle, Pentose Phosphate Signature, Glycolysis, Oxidative Phosphorylation (OxPhos), Fatty Acid Alpha Oxidation (FAAO)) and inflammatory modules (Anti inflammation, Alternative Complement Pathway, Granulocyte, MHC II, Proinflammatory IL-1 Family) (Figure 2B, Table S3). In the final ML iteration, the perfect classification metrics observed in the first iteration were maintained, but linear regression (LR) and random forest (RF) were the top performing algorithms (Figure 2C, Table S2). This result highlights the importance of comparing multiple algorithms when inputting different sets of features to the ML pipeline. The final 20 genes with greatest performance in classifying COVID-19 patients from healthy individuals were dominated by genes from modules of the Cell Cycle (CCNB1, CCNB2, CDC20, CEP55, GINS2, MCM10, MKI67, PTTG1) and OxPhos (COX6C, COX7B, COX7C, NDUFAF5, NDUFAF7, NDUFB3, TMEM126B, UQCRB) but also included genes in the Glycolysis (SLC2A3), and Granulocyte (CD177, CLC, OSM) modules (Figure 2D and Figure S1B, Table S3). To determine the exact relationship between expression of these 20 genes and COVID-19 classification, we used SHapley Additive exPlanations (SHAP) values, which provide information on the magnitude and directionality of the contribution of each gene to classification of an individual as a COVID-19 patient or a healthy control. SHAP values are displayed in a summary plot (Figure 2E) in which each dot represents the expression of each gene in an individual patient and higher gene expression values in conjunction with increased SHAP values are indicative of a positive association with COVID-19 classification, and lower gene expression values in conjunction with increased SHAP values indicate a negative correlation with COVID-19 classification. Overall, we found that increased expression of cell cycle and glycolysis genes and decreased expression of OxPhos genes were associated with COVID-19 classification (Figure 2E). Genes in the Granulocyte module were split with increased expression of CD177 and decreased expression of CLC and OSM contributing to COVID classification (Figure 2E). The COVID-19 patient cohort consisted of individuals with mild, non-critical disease as well as individuals with severe disease requiring ICU admission. Therefore, we sought to determine whether an iterative ML approach could also effectively identify subtle differences between COVID-19 patients with mild disease and healthy individuals (Figure 3). Interestingly, the non-critical COVID vs. healthy classifier performed similarly to classification including COVID-19 patients with severe disease. In the first ML iteration, SVM was the top performing algorithm over 10 repetitions with an average AU-ROC of 0.994, AU-PR of 0.992, and accuracy of 0.976 (Figure 3A, Table S2). The top 10 gene modules for the non-critical COVID-19 vs. healthy classifier shared common proliferation/metabolism pathways with results when including all COVID-19 patients but, interestingly, had more representation of immune cell gene modules including Plasma Cells (PCs), CD40 Activated B Cells, and Inflammatory Neutrophils (Figure 3B, Table S3). Algorithm performance improved by the final ML iteration and LR and RF algorithms performed the best with perfect classification metrics in 9 out of 10 repetitions (Figure 3C, Table S2). The additional immune modules selected as top features identifying non-critical COVID-19 patients from healthy controls were reflected in the final list of genes as in addition to Cell Cycle (CCNB1, CCNB2, CCNE1, CDC20, CEP55, E2F3) and OxPhos (COX7B, COX7C, DNAJC15, NDUFA5, NDUFAF7, TMEM126B, UQCRB), genes in the Inflammatory Neutrophil (CARD16, CFL1, CLEC4D, HBC21), CD40 Activated B Cell (DUSP4), Alternative Complement Pathway (CFD), and Apoptosis (FAS) modules were also included (Figure 3D and Figure S2B, Table S3). SHAP values for the top 20 genes indicated that increased expression of genes in the Cell Cycle, Alternative Complement Pathway, CD40 Activated B Cell, and Inflammatory Neutrophil modules and decreased expression of genes in the OxPhos and apoptosis modules were associated with classification of non-critical COVID-19 patients as compared to healthy individuals (Figure 3E). Thus, an iterative ML approach effectively differentiated COVID-19 patients with varying levels of disease severity from healthy individuals based solely on expression of gene modules and individual genes in each patient. To extend the clinical applications of gene expression-based classification of COVID-19 patients, we employed the iterative ML pipeline to the classification of COVID-19 patients with critical, severe disease requiring hospitalization from patients with a mild, non-critical form of the disease (Figure 4). For the critical vs. non-critical COVID-19 classifier, the gradient boosting (GB) algorithm performed best in the first iteration with an average AUC-ROC and AUC-PR of 0.82 and an average accuracy of 0.83 across 10 repetitions (Figure 4A, Table S2). The top 10 gene modules distinguishing critical from non-critical COVID-19 largely consisted of inflammatory cell/pathway modules and, in particular, T cells (T Cell, Inflammatory Neutrophil, IFN, Tactivated, Treg, Cytotoxic, Activated T Cell) as well as metabolic pathways (Amino Acid (AA) Metabolism, Pentose Phosphate Signature, FAAO), and the stress response (Unfolded Protein) (Figure 4B, Table S3). As with the COVID-19 vs. healthy classifier, performance across all algorithms improved by the final iteration and GB remained the top performer with an average AUC-ROC of 0.98, AUC-PR of 0.98, and overall accuracy of 0.93 (Figure 4C, Table S2). As a result, the majority of the top 20 individual genes identifying critical COVID-19 patients with more severe disease were derived from the Inflammatory Neutrophil module (CARD16, CAST, CD177, FKBP5, H2BC21, HIF1A, MCEMP1, NT5C3A) as well as genes from the IFN (EIF2AK2, SP100), Unfolded Protein (DERL1, ERAP1, SSR1), AA Metabolism (OXCT1), and T-cell-related modules (CCR2, CD28, CREM, EOMES, SGK1) (Figure 4D and Figure S3B, Table S3). SHAP analysis of the directionality of relationships between relative expression of the top 20 genes and classification of critical COVID-19 revealed that, as a whole, critical classification was associated with increased expression of inflammatory neutrophil and IFN genes and decreased expression of T-cell, metabolism, and stress response genes (Figure 4E). These results emphasize the differences in the nature of inflammation present in critical COVID-19 patients as compared to those with mild disease. Our previous analysis of patients admitted to the ICU with or without COVID-19-induced AHRF found few differences in gene-expression-based immune profiles, predominantly centered on the nature and magnitude of the PC antibody response [14]. To explore these differences in greater depth, we utilized the iterative ML pipeline for classification of COVID-19 vs. non-COVID-19 ICU patients (Figure 5). The first iteration of this classifier had the weakest initial algorithm performance, and SVM performed best in 5 out of 10 repetitions with an average AUC-ROC of 0.77, AUC-PR of 0.81, and accuracy of 0.76 (Figure 5A, Table S2). Of the top 10 gene modules for the COVID-19 vs. non-COVID-19 ICU classifier, the PC module had the greatest overall feature importance, as expected from our previous work (Figure 5B, Table S3). The remaining modules in the top 10 included those related to the inflammatory response (IFN, pDC, CD40 Activated B Cell, LDG, Erythrocytes, TNF) as well as cell proliferation and metabolism (Cell Cycle, TCA Cycle, Pentose Phosphate Signature). Algorithm performance improved dramatically by the final ML iteration in the pipeline, and linear regression (LR) was the best overall algorithm with average AUC-ROC of 0.91, AUC-PR of 0.99, and accuracy of 0.97 (Figure 5C, Table S2). The top 20 genes predominantly originated from the TNF module (AMPD3, ASAP1, CDKN3, EGR1, FBXL2, FCER2, GMIP, GP1BA) but also included genes from the PC (SDC1), IFN (GBP4, IFITM1, IFITM3), Cell Cycle (MCM10, PTTG1), TCA cycle (CS, SDHC), Pentose Phosphate Signature (G6PD, H6PD), LDG (CAMP), and Erythrocyte (GYPA) modules (Figure 5D and Figure S4B, Table S3). Interestingly, the SHAP summary plot indicated that decreased expression of most of these genes including those from the TNF and metabolism-related modules and increased expression of genes from the PC, IFN, and LDG modules was associated with COVID-19 classification from all patients admitted to the ICU (Figure 5E). Altogether, these results demonstrate that gene expression data can effectively classify subsets of COVID-19 patients and provide insight into inflammatory cells and pathways with the greatest contribution to disease severity. Next, we determined whether the small gene signature output from our iterative ML pipeline could distinguish COVID-19 patients with mild/moderate and severe disease in an independent dataset (GSE157103) [21]. Blood RNA-seq data from 100 hospitalized COVID-19 patients (50 admitted to the ICU and 50 non-ICU) was used to validate the 20 gene signature from the critical vs. non-critical COVID-19 classifier (Figure 4D, Table S3). Scaled log2 expression values for the 20 genes were used as input for 9 ML algorithms and combined ROC and PR curves were generated (Figure 6A). SVM was the top performing algorithm with excellent AUC-ROC and AUC-PRs of 0.97. Classification performance metrics were high for all algorithms with sensitivities ranging from 0.67 to 0.87, specificities from 0.7 to 1, and accuracies from 0.72 to 0.88 (Figure 6B). Therefore, gene signatures derived from the iterative ML pipeline can be successfully applied to classification of COVID-19 severity in an independent patient cohort and achieve a comparable level of model performance. The heterogeneity of COVID-19 and the wide range of individuals affected by acute and/or persistent disease manifestations necessitates an accurate and reproducible approach to identify drivers of pathology and disease trajectory over time. We have developed a novel, iterative ML approach to identify key molecular signatures stratifying COVID-19 patients based on disease severity and to differentiate them from healthy individuals and those with other infectious or non-infectious causes of AHRF. This information could be leveraged to predict whether an individual that tests positive for COVID-19 will develop mild/moderate disease warranting standard of care treatment or severe disease requiring more careful monitoring and deployment of targeted therapeutics to prevent multi-organ damage and death. Thus, our work is directly applicable to optimizing the deployment of medical resources for COVID-19 patient care. As a result, we identified four separate gene biomarker panels distinguishing COVID-19 patients from healthy individuals, critical from non-critical COVID-19 patients, and critical COVID-19 patients from those admitted to the ICU for non-COVID-19 AHRF. Furthermore, each classifier achieved excellent ML performance metrics across 9 different ML algorithms and over 10 repetitions of each binary classification and effectively translated to an independent validation cohort of COVID-19 patients. In developing a novel ML pipeline that would yield the most reliable and reproducible results, we considered a number of critical factors including (1) the type of input data to use, (2) the ML algorithm(s) that would be most appropriate, and (3) the method to select the most informative, non-redundant features. First, we used gene expression data as input for the ML pipeline as it produces thousands of data points and, thus, provides a comprehensive overview of COVID-19 disease profiles that is easily obtainable from a single blood draw [34,35,36]. In addition, our previous studies demonstrated the utility of bulk transcriptome data to characterize COVID-19 immunopathology across different tissues and levels of disease activity [13,14], results that have been validated by other multi-omics studies [15,16,17,18,19,20,21,22]. Second, to account for differences and potential biases in ML algorithms used for patient stratification, we utilized 9 different supervised classifiers spanning a wide range of approaches to model construction. These included if/then (DTREE), regression (LR), ensemble (RF, ADB, GBM), Bayesian (NB), dimensionality reduction (LDA), and instance-based algorithms (SVM, KNN) [37,38,39]. In addition, multiple repetitions of each step in the ML pipeline further increased confidence in the reliability and reproducibility of the results. Finally, in working with high-throughput gene expression data, we faced the issue of feature selection and how to reduce transcriptome-wide datasets into a manageable number of gene features that would optimize ML algorithm performance [40,41]. We accounted for this by using curated gene modules representative of immune cells and pathways previously associated with COVID-19 pathogenesis. Then, using an iterative approach in which the most informative features were selected moving forward allowed us to reduce the number of genes used for the final iteration and select the top 20 overall genes able to distinguish COVID-19 patients in each classifier. This iterative ML pipeline was used for classification of COVID-19 patients from healthy individuals, critical from non-critical COVID-19 patients, and COVID-19 from non-COVID-19 patients admitted to the ICU with AHRF. Overall, the pathologic cell types and processes represented in the resulting modules and genes reflected previous characterizations of COVID-19 patients, providing validation for these results. Comparing differences in the output of each classifier provided additional insights into the distinguishing features of severe manifestations of COVID-19 as compared to other conditions. Classification of all COVID-19 patients from healthy individuals was dependent on increased expression of genes involved in the cell cycle and decreased expression of genes involved in mitochondrial function through OxPhos. Dysregulated expression of cell cycle genes in COVID-19 patients has been reported in several studies in conjunction with outgrowth of peripheral blood leukocyte populations and thus may serve as a marker of generalized inflammation in infected individuals [42,43,44]. The decreased expression of OxPhos genes is indicative of mitochondrial dysfunction and oxidative stress, both of which are a general feature of the response to viral infection and have also been implicated specifically in COVID-19 patients [45,46,47]. Interestingly, isolating non-critical COVID-19 patients in comparison to healthy individuals yielded a final gene list that, in addition to cell cycle and OxPhos genes, incorporated genes linked with pro-inflammatory cell populations, including inflammatory neutrophils and activated B cells. Our group and others have previously associated neutrophil-driven inflammation and rapid outgrowth of antibody secreting cells (ASCs) with early SARS-CoV-2 infection [14,48], suggesting that these signatures are early signs of mild/moderate disease and that continued expansion of these populations could result in critical illness. The final 20 gene list from the critical vs. non-critical COVID-19 classifier indicated that increased expression of inflammatory neutrophil, IFN, and stress response pathway genes in conjunction with decreased T-cell genes were hallmarks of severe COVID-19. Most of the 20 gene signature was composed of inflammatory neutrophil genes, which were originally identified in the blood of severe COVID-19 patients [49,50]. Of these, CARD16 and H2BC21 were shared in the signature differentiating non-critical COVID-19 patients from healthy individuals. However, their increased expression in critical patients emphasizes the greater prevalence of pathologic neutrophils in promoting severe disease complications. An elevated IFN response has also been associated with severe COVID-19 in numerous studies in line with a dysregulated systemic inflammatory response and increased expression of stress response genes as tissue damage occurs [16,51,52]. As pro-inflammatory response signatures were increased in severe COVID-19 patients, this was accompanied by a decrease in immunoregulatory population genes and, in particular, T cells indicative of T-cell lymphopenia, which has been frequently described in severe COVID-19 [15,53,54]. Notably, the specific inclusion of cytotoxic-T-cell-specific genes (EOMES and SGK1) in the 20 gene signature differentiating critical patients further supports a role for inability to clear viral infection in COVID-19 severity. We also used our iterative ML pipeline to differentiate ICU patients with COVID-19 from those admitted with non-COVID-19 conditions. The most prevalent module for genes resulting from this classifier were indicative of a high TNF response in COVID-19 ICU patients and high expression of TNF family members has been found in damaged tissue of COVID-19 patients indicative of severe disease [55]. This result suggests that while numerous inflammatory cytokines have been implicated in COVID-19 pathogenesis, a TNF response is unique to COVID-19-associated AHRF compared with other conditions requiring admittance to a critical care ward and provides evidence that TNF inhibitors may be a viable option for treatment specifically of COVID-19 [56]. In addition, inclusion of the PC gene SDC1 in the final gene list distinguishing COVID-19 from non-COVID-19 ICU patients is in agreement with our previous work in which PC enrichment was the major difference in these cohorts [14]. Our innovative ML pipeline extends and improves upon previous efforts at ML-based stratification of COVID-19 patients [25,26,27,28,29,30,31,32,33]. A main difference in our approach is the use of whole blood gene expression data, further reduced into groups of gene modules indicative of pathologic cell populations and processes associated with SARS-CoV2 infection and COVID-19 severity. This is in contrast to ML studies utilizing patient demographic/medical record data or lung CT scan images which are limited by the quality and quantity of input data for ML algorithms [25,26,27,28]. Among other ML classification studies using transcriptomic data, our approach is the first to identify the most informative genes from within gene modules previously identified to be abnormally enriched in whole blood of COVID-19 patients. The use of whole blood RNA-seq samples rather than gene expression from sorted single-cell populations [32], peripheral blood mononuclear cells (PBMCs) [32], or throat swabs [31] provided the basis of a clinically useful blood test in which sample handling was minimal. Moreover, this work represents the first study to build transcriptome-based ML models predictive of variations in COVID-19 severity as well as differences from other individuals in the ICU, which were previously assessed separately [29,33]. Overall, our iterative ML pipeline for COVID-19 patient stratification offers a robust method of identifying critical inflammatory processes driving progression to severe disease. In addition, the excellent performance of the resulting algorithms instills confidence that they can be applied to predict disease trajectories of newly diagnosed individuals as demonstrated by the effectiveness in distinguishing severe COVID-19 patients admitted to the ICU from non-ICU COVID-19 patients. We propose that this approach be used for development of rapidly deployable blood-based PCR tests employing a small number of highly informative genes to provide clinical support for COVID-19 patient risk assessment in the clinic. Three publicly available whole blood bulk RNA-seq datasets previously analyzed for characterization of COVID-19 patients with varying degrees of severity were used as input for machine learning classification algorithms and an additional dataset was analyzed for validation. Raw SRA files were downloaded from Gene Expression Omnibus (GEO) accessions: GSE161731 [19], GSE172114 [22], PRJNA777938 [14], and GSE157103 [21], converted into FASTQ files, and processed into log2 gene expression values as previously described [14]. The R/Bioconductor package GSVA [57] (v1.36.3) was used as a non-parametric, unsupervised method for determining enrichment of pre-defined gene modules in individual gene expression samples as previously described [14]. GSVA enrichment scores for 40 gene modules (Table S1) previously used for characterization of immune profiles in blood from COVID-19 patients [13,14] were calculated for each patient and used as input for the iterative machine learning pipeline. The iterative ML pipeline is outlined in Figure 1. Briefly, GSVA enrichment scores for 40 gene modules (Table S1) from publicly available whole blood bulk RNA-seq datasets were used as input for 9 ML algorithms in each of 4 binary classifications. The pipeline consisted of 3 ML iterations each of which was repeated 10 times. After the first 2 ML iterations, the top 50% of features from the top 5 performing algorithms were used as input for the next iteration. Then, the top 10 gene modules were filtered to remove redundant genes with correlation coefficients >0.8 and log2 gene expression values from the remaining genes were normalized across datasets. Normalized z-scores of gene expression values were used as input for the final ML iteration to select the top 20 gene features for each classifier. Detailed methods for each step in the pipeline are described below and available through the following link: https://github.com/adaamen/iterative-ML-COVID-pipeline (accessed on 2 February 2023). Binary classifications were carried out in Python (v3.8.2) using scikit-learn (v0.24.1) [58]. Nine ML algorithms were implemented to distinguish COVID-19 patients from healthy individuals, critical from non-critical COVID-19 patients, and COVID-19 patients admitted to the ICU from non-COVID ICU patients (logistic regression (LR), random forest (RF), support vector machine (SVM), decision tree (DTREE), AdaBoost (ADB), Gaussian naïve Bayes (NB), linear discriminant analysis (LDA), k-nearest neighbors (KNN), and gradient boosting classifier (GB)). Synthetic minority oversampling technique (SMOTE) [59] was used to account for class imbalances between patient groups used for each classification by expanding the number of samples in the minority class to ensure equal numbers between classes used for ML. For each classification, datasets were split into 70% training and 30% validation and SMOTE was applied to the training data. Classification models for each ML algorithm were then built on the training set using default parameters from the scikit-learn library and evaluated on the validation set. Average algorithm performance was calculated based on the following metrics: sensitivity, specificity, accuracy, and areas under the receiver operating characteristic (ROC) and precision-recall (PR) curves. ROC and PR curves were plotted using the Python library matplotlib (v3.3.4) [60]. Binary classification of ICU and non-ICU COVID patients from an independent dataset were used as validation for results of the iterative ML pipeline. For this classification, SMOTE was not applied. To optimize model performance on the validation dataset, a 75% training, 25% testing split was used and the class_weight = ‘balanced’ parameter was added to the SVM algorithm. Feature importance was calculated for the top 5 performing algorithms in each ML iteration. For each input feature a higher feature importance score indicates how relevant that feature is towards the output of each classifier. The method of calculating feature importance was determined based the appropriate metric for each algorithm and implemented through the scikit-learn Python library. Gini importance, or average decrease in impurity, was used for RF, DTREE, ADB, GB, LR, and LDA, while permutation importance was used for SVM, KNN, and NB. Feature redundancy was calculated for genes in the top 10 modules of each classifier using Pearson correlations between each feature and every other feature. When two genes shared a correlation coefficient >0.8, the gene with the highest degree of correlation with all other genes in the module was removed. Pearson correlations were computed using the cor() function in R and plotted using the corplot library. Log2 transformed gene expression values were normalized before the final ML iteration to account for batch effects across separate datasets. To do this, z-scores were calculated for each gene across samples in each dataset using the scale() function in R and then normalized datasets were combined into a dataframe used as input for the 9 ML classification algorithms. The pre- and post-normalization expression values for each dataset combination are summarized as boxplots (Figures S1–S4A). The precise contribution (magnitude and directionality) of the top 20 gene features output from each classifier was determined using Shapley additive explanations (SHAP) [61]. SHAP values were calculated based on the RF algorithm for the final iteration of each classifier. SHAP summary plots were visualized in Python using the shap library (v0.39.0). Normalized gene expression values of the top 20 genes were compared between the two classes of each binary classification using multiple unpaired t-tests with Welch’s correction in GraphPad Prism (v9.1.0; www.graphpad.com, (accessed on 20 December 2022)). A two-tailed p-value < 0.05 was considered statistically significant.

PMC10002756

Smaranda Maier,Laura Barcutean,Sebastian Andone,Doina Manu,Emanuela Sarmasan,Zoltan Bajko,Rodica Balasa

Recent Progress in the Identification of Early Transition Biomarkers from Relapsing-Remitting to Progressive Multiple Sclerosis

22-02-2023

multiple sclerosis,neurodegeneration,relapsing-remitting,secondary progressive,progression,biomarker

Despite extensive research into the pathophysiology of multiple sclerosis (MS) and recent developments in potent disease-modifying therapies (DMTs), two-thirds of relapsing-remitting MS patients transition to progressive MS (PMS). The main pathogenic mechanism in PMS is represented not by inflammation but by neurodegeneration, which leads to irreversible neurological disability. For this reason, this transition represents a critical factor for the long-term prognosis. Currently, the diagnosis of PMS can only be established retrospectively based on the progressive worsening of the disability over a period of at least 6 months. In some cases, the diagnosis of PMS is delayed for up to 3 years. With the approval of highly effective DMTs, some with proven effects on neurodegeneration, there is an urgent need for reliable biomarkers to identify this transition phase early and to select patients at a high risk of conversion to PMS. The purpose of this review is to discuss the progress made in the last decade in an attempt to find such a biomarker in the molecular field (serum and cerebrospinal fluid) between the magnetic resonance imaging parameters and optical coherence tomography measures.

Recent Progress in the Identification of Early Transition Biomarkers from Relapsing-Remitting to Progressive Multiple Sclerosis Despite extensive research into the pathophysiology of multiple sclerosis (MS) and recent developments in potent disease-modifying therapies (DMTs), two-thirds of relapsing-remitting MS patients transition to progressive MS (PMS). The main pathogenic mechanism in PMS is represented not by inflammation but by neurodegeneration, which leads to irreversible neurological disability. For this reason, this transition represents a critical factor for the long-term prognosis. Currently, the diagnosis of PMS can only be established retrospectively based on the progressive worsening of the disability over a period of at least 6 months. In some cases, the diagnosis of PMS is delayed for up to 3 years. With the approval of highly effective DMTs, some with proven effects on neurodegeneration, there is an urgent need for reliable biomarkers to identify this transition phase early and to select patients at a high risk of conversion to PMS. The purpose of this review is to discuss the progress made in the last decade in an attempt to find such a biomarker in the molecular field (serum and cerebrospinal fluid) between the magnetic resonance imaging parameters and optical coherence tomography measures. Multiple sclerosis (MS) is a chronic, inflammatory neurodegenerative disorder of the central nervous system (CNS) that represents the leading cause of non-traumatic disability in young adults. Despite extensive research into the pathophysiology of the disease in recent years, the underlying mechanisms have yet to be unravelled. The process of neurodegeneration was considered to occur secondarily to inflammation and demyelination, a consequence of an evolutive progression. However, recent data have confirmed that neurodegeneration is present in all clinical phenotypes, including early disease. Therefore, it was hypothesised that neurodegeneration and axonal loss are driven by independent pathological processes occurring before, or concomitant to, demyelination [1,2,3,4,5,6]. In 85–90% of cases, MS presents a relapsing-remitting course and is defined as relapsing-remitting MS (RRMS). RRMS is characterised by episodic occurrence of symptoms with complete or partial recovery followed by periods of clinical stability and remission. According to the natural history data of MS, 25% of RRMS patients eventually convert to progressive MS (PMS) after one decade of evolution. This progression is characterised by a gradual decline in neurological function with or without superimposed relapses. Due to recent developments in potent disease-modifying therapies (DMTs) and high-accuracy diagnostic protocols, the percentage of patients who transition to a progressive phase has decreased. Nonetheless, two-thirds of RRMS patients will progress to PMS after 20 years of disease evolution. The conceptual delineation between RRMS and PMS is not clearly defined; the evolution is gradual, with an intermediate phase of ‘transitional MS’. Compensatory mechanisms may initially protect against neuronal injury; however, in time, their continuous decline progresses to exhaustion and irreversible neurological deficits, marking the transition to PMS. However, the presence of these compensatory mechanisms represents one of the main obstacles in early PMS diagnosis. MS patients have a substantial heterogeneity concerning the brain reserve, modifiable cognitive reserve and neuroplasticity [7,8,9,10,11,12,13,14,15,16,17,18]. In 2014, a new definition by the MS Phenotype Group classified MS as relapsing-remitting and progressive disease. The early identification of the MS phenotype in which the predominant pathophysiological process is either inflammation (in RRMS) or neurodegeneration (in PMS) has a substantial impact on the therapeutic decision and the prognosis [18,19]. Neurodegeneration, independent of the inflammatory response, although present even in the early, asymptomatic stages of MS, represents the primary pathogenic mechanism in PMS and is responsible for the irreversible progression of disability, independent of relapses. A PMS diagnosis is exclusively retrospective and based on disability accumulation over 6 months to 1 year in the case of a patient who initially had a relapsing-remitting evolution of the disease. Currently, it is considered that the best definition of PMS is an increase by at least one point in the Expanded Disability Status Scale (EDSS) for patients who initially had an EDSS score ≤5.5 points or 0.5 points if the initial score was ≥6.0 points. The other conditions that must be fulfilled in order to meet the diagnostic criteria for PMS are the presence of an EDSS score of at least four points and involvement of the pyramidal system (functional score ≥2). Identifying the transition between RRMS and PMS according to the moment between the onset of disease progression and PMS diagnosis is difficult, and studies report an approximate 3-year delay in PMS diagnosis in 70% of cases. A delay in PMS diagnosis is associated with disability progression, alteration of the neurological status and a decrease in quality of life. Additionally, the financial burden is represented by the high reimbursement required for highly efficient DMTs in RRMS that hold no therapeutic value in PMS. Until recently, there was no incentive to select patients who were already in the transition phase due to the lack of an effective treatment for this form of MS. With the approval of siponimod in March 2019 as the first efficient treatment for active PMS, it became imperative to identify reliable and objective biomarkers of transition that could facilitate early diagnosis of PMS in the preclinical stage. This would enable early therapy adaptation (including timely escalation) to prevent PMS conversion and the irreversible progression of disability. In addition to the immunomodulator effect on the peripheral immune cells, siponimod is able to cross the blood–brain barrier (BBB) and interact with sphingosine 1-phosphate receptors inside the central nervous system, exerting a neuroprotective effect in PMS [7,8,13,20,21,22,23,24,25]. In the last 2 years, considerable progress has been made in identifying a feasible biomarker for early transition to PMS. According to Lublin et al. [18], the diagnosis of PMS can be established only retrospectively, and the authors underline the lack of clinical, imaging, immunological and pathological biomarkers for the transition point from RRMS to PMS, and the pressing need to identify biological biomarkers that can enhance the MS phenotype definition. The authors suggested using optical coherence tomography (OCT) as an indirect biomarker for whole-brain tissue loss and the importance of additional studies aiming to extend the indications for OCT as a potential biomarker of MS phenotype. They suggest prospective follow-up for MS patients and their clinical and paraclinical parameters that should be prioritised to understand the disease’s subtypes better and to accurately predict the transition moment between RRMS and PMS. Recently, a Delphi panel in Portugal recognised that the identification of progression is a clinical challenge that would be completed by the discovery of biomarkers and diagnostic tools that point to the transition phase. The authors suggested that neurofilament light chains and OCT parameters hold value as biomarkers for disease progression [23]. This review aims to discuss the progress made in identifying biomarkers (molecular and imaging) that have prognostic value in the evolution of disability and the conversion of RRMS to PMS (Figure 1). Identifying an early serum biomarker for transition in MS patients that has value in current clinical practice is challenging. Most experimental studies were proven invalid due to a lack of specificity, sensibility and ease of sample harvest. Neurofilaments are major components of the neuronal cytoskeleton and are released into the extracellular space, cerebrospinal fluid (CSF) and bloodstream as a result of neuro–axonal lesions that appear in various pathologies of the central and peripheral nervous systems [26,27]. Three subunits have been described: NfL, neurofilament medium and neurofilament heavy chain (NfH). Since their discovery in 1989, the assessment of NfLs has proven effective for quantifying central neurodegeneration. Rosengren et al., demonstrated that Alzheimer’s disease and amyotrophic lateral sclerosis patients have significantly higher CSF NfL levels than healthy subjects, suggesting that CSF NfL titre can be used as a biomarker of neurodegeneration [28]. Further studies demonstrated that CSF NfL levels increase during MS relapses and are positively correlated with disability index and radiological activity. This evidence supports the hypothesis that CSF NfL can be used as biomarkers for treatment response [29,30,31,32]. Even though NfL concentrations are higher in the CSF than in the serum, the invasiveness of lumbar puncture greatly limits the applicability of a CSF examination in current practice. However, this limitation was overcome with the introduction of the state-of-the-art single-molecule array (SiMoA) technique, which allowed for more precise assessment of low serum NfL concentrations. The feasibility of serum assessment using harvested peripheral blood allowed for the routine use of NfL assessment in clinical practice. The potential of NfL as a biomarker of disease progression emerged as a result of broad studies on large cohorts of MS patients. One of the most important studies was conducted by Disanto et al. The authors included 142 MS patients diagnosed with clinically isolated syndrome (CIS), RRMS, secondary-progressive (SPMS) and primary progressive MS (PPMS). In the cohort, serum NfL levels correlated with the duration of the disease and were significantly higher in SPMS and PPMS patients compared with CIS and RRMS patients. Univariate analysis revealed a significant association between NfL serum levels and EDSS (p < 0.001), recent decline in EDSS (p = 0.003) and relapse within the 2 months before harvest (p < 0.001). Furthermore, the authors evaluated the potential predictive role of NfL in predictive future clinical disease activity. They noted that in the first 2 years after NfL assessment, the incidence of clinical relapses was twofold higher in patients with NfL serum levels above the 97.5th percentile (IFF = 1.94, 95% CI = 1.21–3.10, p = 0.006 for the first year and IRR = 1.96, 95% CI = 1.22–3.15, p = 0.005 for the second year). Serum NfL levels correlated with disease worsening (calculated by the EDSS) at 12 months; 6.7% of the patients who initially had NFL levels below the 80th percentile presented with disease worsening, compared with 15% of the patients who had baseline NfL levels above the 97.5th percentile. These complex results support the role of NfL as an early biomarker for disease progression [26]. Siller et al., evaluated the role of NfL in neuronal destruction. They included 74 recently diagnosed CIS and RRMS patients, with a mean time from disease onset to serum sampling of 1 month; 63 of the patients were naïve to DMTs. The patients were assessed for NfL serum level at baseline and underwent 3T cerebral magnetic resonance imaging (MRI). Furthermore, 42 of the patients underwent 1–3 additional cerebral MRI assessments over the next 6–37 months. They demonstrated that baseline NfL serum levels significantly correlated with T2 lesion volume (T2-LV) and T2-LV progression throughout the study. Patients who presented with higher baseline NfL level had a higher T2-LV at 2-year follow-up, with an increase in cerebral volume loss in the 6–37 months following the initial MRI assessment. The accuracy of the cerebral atrophy prediction was higher for the 2-year follow-up cerebral MRI than the 1-year follow-up [33]. The results of this study further support the use of NfL as an early biomarker for progression in MS, before apparent imaging structural lesions appear. The role of NfL in the pathophysiology of MS has been extensively studied. Clinical studies have demonstrated the utility of NFLs as a predictive biomarker for the progression of disability, brain and spinal cord atrophy, and cognitive decline [34,35,36,37]. A study by Barro et al., included 259 MS patients (189 with RRMS, 70 with PMS) and 259 healthy controls (HCs). The patients and HCs were assessed based on brain and spinal MRI, clinical data, and serum NfL over a mean period of 6.5 years. The results favoured NfL level as an independent predictive factor for disability progression in the next year (p < 0.001), and increased EDSS was directly correlated with NfL serum level. In addition, higher NfL levels were associated with more severe subsequent cerebral and spinal cord atrophy; at NfL values above the 97.5th percentile, an additional loss of 1.5% of the brain parenchyma and 2.5% of the spinal volume was observed over a period of 5 years [38]. Other longitudinal studies that followed patients with MS for one decade from the baseline assessment of serum NfL level, such as those performed by Thebault et al., and Chitnis et al., demonstrated the predictive role of NfL for the progression of disability, the risk of evolving towards a progressive form of MS, cerebral atrophy and increasing volume of T2 lesions [39,40]. Based on the reported evidence, NfL level was used in clinical studies to monitor treatment response. The results of a post-hoc analysis of the data compiled from two phase 3, placebo-controlled clinical trials published in 2022 by Leppert et al., supported the use of NfL as a biomarker of disease progression. A total of 1452 SPMS and 378 PPMS patients were included. A high baseline NfL level was associated with a significant increase in the risk of confirmed disability progression at 3 and 6 months, early wheelchair restriction and cognitive decline [41]. Serum NfL assessment using high-quality laboratory techniques, such as the SiMoA, represents a simple quantification method for neuronal damage. NfL assessment can be routinely used in clinical practice as a prognostic biomarker for disease progression. Numerous studies demonstrated the utility of NfL as a prognostic biomarker in MS but, in contrast, NfH and phosphorylated NfH have been understudied. NfH and pNfH are cytoskeleton proteins expressed by the neurons in reaction to injury. They are of interest in neurodegenerative disorders such as amyotrophic lateral sclerosis, Alzheimer’s disease and frontotemporal dementia. Still, they have been studied to a lesser extent in MS. A recent study evaluated the role of serum/CSF NfH and OCT parameters in MS patients. The authors demonstrated that SPMS patients had significantly higher NfH CSF levels and lower thickness of the peripapillary retinal nerve fibre layer (pRNFL) compared with RRMS patients and controls (p < 0.0001). The pRNFL thickness positively correlated with serum and CSF NfH for both groups (p < 0.01). The EDSS scores positively correlated with the CSF NfH levels both in RRMS and SPMS groups (p < 0.001, p = 0.002) [42]. The importance of NfH levels in MS patients is disputed. While some authors demonstrated that the serum and CSF NfH levels are higher in PMS patients compared with RRMS patients, others disproved any association of this biomarker with a disability, progression risk and imaging parameters [43,44,45,46,47,48,49,50,51,52,53]. Due to the lack of consistency in the research results, the utility of NfH as a biomarker of transition in MS has yet to be established and further validation is needed. MicroRNAs (miRNAs) are short, non-coding, endogenous RNA molecules consisting of 21–25 nucleotides that are mainly involved in regulating gene expression, especially at the posttranscriptional level, by binding to target miRNAs, which leads to either degradation or translational repression [54,55]. MiRNAs have an essential role in tissue development and homeostasis, immune system up-regulation and maturation, and were shown to be involved in the pathophysiology of autoimmune disorders, such as MS [56,57,58,59]. Over 1500 human miRNAs regulating one-third of human genes have been identified [56]. Because they are susceptible to underlying pathological processes, as important molecules involved in cellular function regulation, and because of their stability, resistance to endogenous RNases, and easy assessment, circulating miRNAs have been proposed as biomarkers for early transition to PMS in RRMS patients [56]. Clinical studies have demonstrated the variable expression of certain miRNAs between RRMS and PMS (Table 1), suggesting that these miRNAs can be used as early biomarkers of conversion from RRMS to PMS. Haghiki et al., in a study that enrolled 53 MS patients (RRMS, SPMS, and PPMS) and 39 patients with other neurological pathologies, identified 50 miRNAs in the CSF. They compared the expression of five selected miRNAs (miR-132, miR-181c, miR-494, miR-633, miR-922) between the two groups of patients and between MS phenotypes. They reported that miR-633 and miR-181c exhibited significantly lower expression in the CSF of patients with SPMS compared with patients with RRMS. Following statistical analysis of the combination of the two miRNAs, a high sensitivity and specificity (69% and 82%, respectively) for differentiating between the two major types of MS was found [60]. The same researchers aimed to validate their results in a larger cohort of patients with MS (n = 218) and various other neurological pathologies (n = 211). They demonstrated that the same two miRNAs (miR-633 and miR-181c) were expressed differently in MS patients; however, the differences were not sufficient to differentiate RRMS from PMS patients [61]. Ghandi et al., aimed to identify the specific miRNAs present in different MS evolutive phenotypes. They included RRMS, SPMS and PPMS patients in the study and evaluated the expression of 368 miRNAs. The expression of eight miRNAs was significantly different between RRMS and SPMS patients: hsa-miR-92a-1, hsa-miR-135a, hsa-miR-454, hsa-miR-500, hsa-miR-574-3p, hsa-let-7c, hsa-let-7d and hsa-miR-145. Of these miRNAs, miR-92a-1 had the strongest association with RRMS (adjusted OR: 1.35, p = 0.022). MiR-92a-1 expression also correlated with the EDSS and the duration of the disease; furthermore, miR-92a-1 is a part of the miR-17-92 miRNA cluster that targets genes involved in cell cycle regulation, signalling, modulation and activation of naïve CD4+ T lymphocytes [62,63,64]. Recent imaging studies have demonstrated that cortical lesions are specific to PMS; however, conventional MRI sequences have technical challenges associated with their assessment. For this reason, Tripathy et al., aimed to identify miRNAs that could be used as predictive biomarkers of cortical involvement in MS. They evaluated 27 miRNAs that were differentially expressed in grey matter (GM) lesions (GMLs) compared with normal-appearing grey matter (NAGM), of which 10 were up-regulated and 17 were down-regulated [65]. By comparing their results with those of Fritsche et al. [66], who identified 31 miRNAs that were differentially expressed in cortical demyelinating lesions compared with HCs, the authors identified 4 miRNAs with specific expression in GMLs in both studies: miR-1180-3p, miR-219-a-2-3p, miR-328-3p and miR-432-5p. The results of this study were compared with the analysis performed by Regev et al. [67], who identified 87 serum miRNAs that correlated with the degree of cortical GM atrophy. Four of these miRNAs also had modified expression in GMLs in a study by Tripathy et al.: has-miR-149, has-miR-20a, has-miR-25 and has-miR-29c [65]. Furthermore, in Keller et al.,’s study [68], these four miRNAs were found in the peripheral circulation of MS patients and were specifically expressed by immune cells. These results support the use of these four miRNAs detected in MS patients’ serum as biomarkers of PMS, because of their specific expression in cortical demyelinating lesions and their association with cerebral atrophy. In a study of 120 MS patients, Regev et al. [67] aimed to identify the correlations between different selected miRNAs and MRI parameters (i.e., T2-LV, T1/T2 ratio lesion volume, global brain atrophy, GM atrophy, cervical spinal lesions and cervical spinal cord atrophy). Statistical analysis identified two miRNAs (has-miR-375 and has-miR-629-5p) that significantly correlated in both groups with cerebral atrophy, and two miRNAs (miR-486-5p and miR-92a-5p) that significantly correlated with the ratio of T1/T2 lesion volumes. The miRNAs associated with T2-LV (inflammatory and demyelinating processes) differed from those associated with cerebral atrophy (neurodegenerative processes). This observation supports the hypothesis that different pathological processes drive inflammation and neurodegeneration in MS. In another study, Regev et al., proposed identifying circulating miRNAs that could be used to differentiate between RRMS and PMS phenotypes. The authors analysed 652 miRNAs to determine which were associated with inflammation and which were associated with neurodegeneration. In the discovery phase, by comparing the expression of miRNAs between RRMS and SPMS patients, 27 miRNAs with distinct expression between the 2 groups were identified. However, in the validation phase, of the 27 proposed miRNAs, only the expression of miR-27a-3p was significantly different (area under the curve value of 0.78) [69]. Additional studies demonstrated that miR-27a-3p inhibits the differentiation of Th1 and Th17 lymphocytes, is involved in the regulation of the neurotrophin pathway and is highly expressed in RRMS compared with SPMS patients [69,70]. These studies suggest that miR-27a-3p could function as a biomarker of transition from RRMS to PMS. In the described study, the expression of 10 miRNAs correlated with the degree of disability, and the highest correlation was for miR-199a-5p [69]. The same authors identified another miRNA as a potential biomarker of disease progression in MS patients: miR-337-3p. The expression of this miRNA was negatively correlated with disability (calculated by EDSS) in both the discovery and validation phases. Furthermore, they noticed that miR-337-3p expression was significantly lower in patients with SPMS compared with patients with RRMS (p = 0.01) [71]. Decreased miR-337-3p expression was associated with increased T1/T2 lesion load [67]. miR-337-3p targets RAP1A, a key component of chemokine-induced integrin activation and migration, which suggests that it can be used as a biomarker for natalizumab treatment response due to the drug’s mechanism of action, which involves inhibition of α4β1-integrin [67,71]. Sharaf-Eldin et al., aimed to identify biomarkers that facilitate the differential diagnosis of immune-mediated neuro-inflammatory pathologies. They included patients with neuromyelitis optica spectrum disorder, systemic lupus erythematosus with neuropsychiatric manifestations and MS. Even though the purpose of the study was to identify the differences between miRNAs in the selected disorders, the researchers determined the serum expression of three miRNAs (miR-145-5p, miR-223-3p and miR-326-5p) and noted that miR-326-5p expression was increased in RRMS compared with SPMS (p = 0.018). Furthermore, different combinations of the three miRNAs allowed for more accurate differentiation between the two clinical MS phenotypes: miR-145-5p + miR-326-5p (p = 0.014), miR-223-3p + miR-326-5p (p = 0.027) and miR-145-5p + miR-223-3p + miR-326-5p (p = 0.005). However, because of the small cohort, the data require validation in a larger population [72]. Exosomes are cell-derived extracellular vesicles that contain various molecules, including miRNAs, and can cross the BBB. Ebrahimkhani et al., hypothesised that the pathophysiological processes associated with different stages of MS evolution should reflect different expression of serum exosomal miRNAs between patients with RRMS and SPMS. They identified nine miRNAs (miR-15b-5p, miR-23a-3p, miR-223-3p, miR-374a-5p, miR-30b-5p, miR-433-3p, miR-485-3p, miR-342-3p, miR-432-5p) that could be used to differentiate patients with relapsing MS from patients with progressive forms of MS; furthermore, they demonstrated that the combination of miR-223-3p + miR-485-3p + miR-30b-5p differentiated between the subtypes with an accuracy of 95%. The results of this study support the use of exosomal miRNAs for early determination of the conversion from RRMS to PMS [73]. Glial fibrillary acid protein (GFAP), which is predominantly expressed by astrocytes, is another promising biomarker for the transition from RRMS to PMS, even if the study results are controversial. GFAP is an intermediate astrocytic cytoskeletal protein and a well-established marker of astrogliosis that plays a vital role in astrocytic-neuronal cross-talk. Abdelhak et al., performed a study on 42 RRMS, 13 SPMS and 25 PPMS patients and 20 patients with non-inflammatory neurological disorders. They assessed GFAP and NfL levels in the CSF and serum. The CSF and serum levels of GFAP were significantly increased in progressive MS patients compared with RRMS patients and patients with non-inflammatory neurological disorders; however, after performing age-adjusted corrections, the statistical significance was lost. In the PPMS patients, serum GFAP strongly correlated with EDSS after age-adjusted corrections. Furthermore, CSF and serum GFAP levels positively correlated with NfL levels in the PMS group [77]. A meta-analysis by Sun et al., identified four studies in which CSF GFAP levels were assessed in RRMS and PMS patients. Three of these studies demonstrated that CSF GFAP levels were increased in PMS compared with RRMS [77,78,79,80,81,82], indicating GFAP as a potential biomarker of disease progression in RRMS patients. Ayrignac et al., analysed serum GFAP levels in a cohort of 11 RRMS and 18 PPMS patients. Serum GFAP levels were significantly higher in PPMS patients than in RRMS patients, and the results remained statistically significant after multivariate analysis that included patient age and duration of the disease. Furthermore, serum GFAP levels were negatively correlated with white matter (WM) and GM volume [83]. Tryptophan is an essential amino acid and 95% of it is metabolised by the kynurenine pathway (KP). Kynurenines are molecules resulting from tryptophan metabolism by the KP. Kynurenic acid (KYNA), the most studied molecule produced by astrocytes, has neuroprotective effects. In contrast, quinolinic acid (QUIN), produced by microglia, exhibits neurotoxic effects. The duality of the two kynurenines resides in their effect on NMDA, AMPA and kainite receptors (KYNA is an antagonist, while QUIN is an agonist). In addition, KYNA inhibits the influx of calcium ions into the intracellular space and prevents the formation of reactive oxygen species, while QUIN has the opposite effect [84,85,86]. Lim et al., analysed the metabolomic KP profile in MS patients, including RRMS, SPMS and PPMS patients. The QUIN/KYNA ratio was significantly higher in PPMS and SPMS patients compared with RRMS patients, and was highly correlated with the EDSS (r = 0.62, p < 0.0001). RRMS patients had higher KYNA levels compared with SPMS and PPMS patients, and high QUIN levels were reported in PPMS patients. Of the 37 potential biomarkers assessed by the authors, 6 could be used as predictive biomarkers of MS subtype, with an accuracy of 83%. In order of relevance, these biomarkers were KYNA, QUIN, tryptophan, picolinic acid, fibroblast growth factor-basic and tumour necrosis factor α [86]. In a study by Aeinehband et al., SPMS patients had lower KYNA and tryptophan levels compared with RRMS patients [87]. The results of these studies confirm the critical role of KP in neurodegenerative processes and the potential of KYNA and QUIN as biomarkers for the prediction of conversion to PMS. These studies demonstrate that kynurenine metabolites have different expression patterns depending on the MS clinical phenotype, and they may be used to indicate the transition from RRMS to PMS. The exact role of chitinase-3-like-1 (CHI3L1) in MS pathophysiology has yet to be completely understood. Inside the central compartment, activated microglia and reactive astrocytes are the main producers of CHI3L1. Chemokine C-X-C motif ligand 13 (CXCL13) is a B lymphocyte chemoattractant involved in forming ectopic lymphoid follicles adjacent to the meninges. Recent clinical studies demonstrated a significant association between the serum and CSF levels of CHI3L1 with disability accumulation, cognitive decline and disease progression in MS patients [88,89,90,91,92]. In a recent systematic review and meta-analysis, Floro et al. [93] evaluated the potential to use serum and CSF CHI3L1 levels as a biomarker for distinguishing different MS phenotypes. They identified 20 studies published over a period of 10 years (2010–2020). Five studies compared the serum/CSF CHI3L1 levels in RRMS, SPMS and PPMS. They noted that PPMS patients are significantly higher compared with RRMS patients, but no differences were noted between RRMS and SPMS patients [93]. Lamancova et al. [94] evaluated serum levels of sNfL, CXCL13 and CHI3L1, among other molecular biomarkers, and noted that they were significantly higher in PPMS and SPMS patients compared with RRMS. Furthermore, the CXCL13 levels correlated with the EDSS score. In the SPMS group, the authors noted a positive correlation between CHI3L1 and sNfL. Similar results are reported by Gil-Perontin et al. [95]. The authors analysed CSF levels for NfL and CHI3L1 in MS patients. The CSF CHI3L1 levels were lower in RRMS patients compared with SPMS and PPMS patients, but reached statistical significance only between RRMS and PPMS patients. In the same study, the CSF CHI3L1 positively correlated with the EDSS score and proved to be an independent factor of prediction for EDSS worsening by one point (HR = 2.99, 95% CI = 1.27–7.07). The patients were followed for 10 years and 14 patients with RRMS converted to SPMS. The statistical analysis identified the CSF CHI3L1 levels as a predictive biomarker for conversion. The results of these studies support the hypothesis that NfL, CHI3L1 and CXCL13 can be used to identify different MS phenotypes and have the potential to become predictive biomarkers of transition from RRMS to SPMS. Two types of parallel inflammation are specific to MS. The first type is typical for the relapsing-remitting forms and is characterised by B and T lymphocyte passage through the vulnerable BBB, mainly with the formation of white matter demyelinating lesions. The second type of inflammation, although present from the early stages of the disease, defines compartmentalised inflammation in the presence of an impenetrable BBB. This results in the formation of tertiary ectopic lymphatic follicles adjacent to the meninge and the perivascular Virchow Robin spaces, and is associated with cerebral and cerebellar subpial demyelinating lesions [96]. The results of the abovementioned studies suggest that an increase in CHI3L1 production is associated with the second type of inflammation. This hypothesis was confirmed by the study of Cubas-Núñez et al. [97], who analysed the CSF CHI3L1 levels in different MS patients and evaluated the CHI3L1 expression and the degree of inflammation and neurodegeneration found in the autopsy tissue of 22 MS patients. The CSF CHI3L1 level was significantly lower in PMS patients compared with RRMS and, furthermore, correlated with the EDSS score. At the level of chronic active lesions, CHI3L1 is mainly expressed by astrocytes and is associated with neurodegeneration [97]. The results of these studies suggest that CHI3L1 and CXCL13 can be used to identify different MS phenotypes, but can also be used as predictive biomarkers for the moment of transition to a PMS. Still, this hypothesis needs to be confirmed in a larger cohort. Tau protein is a key component of the cytoskeleton of both neurons and oligodendrocytes and is essential for axonal transport. Abnormally phosphorylated P-tau leads to cell deterioration and is a characteristic feature of neurodegenerative diseases [98,99]. Anderson et al. [99,100] reported the presence of these abnormally phosphorylated tau proteins in PMS and patients with early aggressive MS. Other studies compared CSF tau protein levels between patients with RRMS and PMS; however, no significant differences were observed between these two groups [101]. The studies that analysed S100B in patients with RRMS and SPMS did not report significant differences [101]. Contactin-1 is a member of the contactin family and is mainly expressed in the paranodal axonal domain. Contactin-1 is involved in myelin formation and its loss is associated with neuronal dysfunction and axonal loss. Contactin-2 is found in the juxtaparanodal domain and plays a key role in axonal growth and guidance [102]. Chatterjee et al. [103] reported that CSF contactin-1 and -2 levels were lower in RRMS and SPMS patients compared with controls; however, no significant difference was observed between RRMS and SPMS patients. Impairment of the visual system in MS is frequent. One-third of patients present at the onset with optic neuritis (ON), and most MS patients develop visual dysfunction throughout the course of the disease [104,105]. Therefore, the visual system has been the focus of numerous studies aiming to unravel the pathophysiological mechanisms that govern MS because the techniques used to assess the optic pathways are easily accessible, non-invasive and can be used in routine clinical practice. These include imaging investigations, such as MRI and OCT, and electrophysiological investigations, such as visual evoked potentials (VEPs). The current literature supports the hypothesis that the visual system is a mirror of the pathophysiological processes at the CNS level, both in terms of the acute focal lesions that would have as a model at the level of the visual system the acute episodes of NO, but also in terms of the neurodegeneration process that would correspond to chronic retinopathy, optic neuropathy and transsynaptic degeneration [34,106,107]. In recent years, clinical studies have demonstrated that OCT can be used as a predictive biomarker for disease progression as it facilitates analysis of the neurodegenerative processes in MS [108,109]. OCT is a modern imaging technique that has significantly improved over the past two decades. It is a non-invasive, fast, easy-to-use, high-resolution technique that uses the same technology as ultrasound. It allows for the assessment of the optic nerve axon via a retinal nerve fibre layer (RNFL) measurement and neuronal assessment via a ganglion cell layer (GCL) measurement, and can detect subclinical neurodegeneration [108]. In 2009, Blumenthal et al. [110] compared RNFL measurements obtained by OCT and histopathological examination and concluded that they were similar. The potential use of OCT parameters as biomarkers of transition to PMS was demonstrated in numerous clinical studies. RNFL and GCL thickness can be used to differentiate RRMS patients from those with progressive forms of MS who present with apparent axonal and neuronal loss [111,112,113,114,115]. In a study on 414 MS patients, Oberwahrenbrock et al. [115] demonstrated that SPMS patients without a history of ON had a significantly thinner RNFL (p = 0.007) than patients with RRMS. RNFL thickness can be used as a biomarker of an early transition from RRMS to SPMS. A selection of the most relevant studies that demonstrated the usefulness of OCT parameters in differentiating RRMS from PMS patients is presented in Table 2. RNFL and GCL thickness is correlated with consecrated biomarkers of neurodegeneration, such as global irreversible disability (quantified by EDSS; Table 3), global and specific brain atrophy [116,117], cortical lesion volumes [118] and fluid biomarkers, such as NfL. Given that the GCL thickness is difficult to measure accurately, in most studies the thickness of the composite GCL and inner plexiform layer (GCIPL) was evaluated [111,118]. Longitudinal studies demonstrated that RNFL and GCL thickness were independent predictive factors for disability progression [34,119,120,121]. Martinez-Lapiscina et al. [122] performed a multicentric research that included 74 CIS, 664 RRMS and 141 PMS patients. EDSS was calculated for all patients at baseline followed by OCT assessment. The patients were observed between 6 months and 5 years, with a median of 2 years. Patients that had at baseline RNFL thickness <88 µm or <87 µm (depending on the OCT), with no history of ON, had a higher risk of disability progression from year 1 to year 3 of follow-up (90% increase in adjusted risk), and the risk increased 4-fold after year 3 of follow-up. The results of these studies suggest that RNFL and GCIPL are promising OCT biomarkers of axonal loss and neurodegeneration and can be used to monitor disease progression and hold value as predictive biomarkers of transition to PMS. The recent advances in neuroimaging, especially in MRI techniques, have drastically challenged the understating of basic pathophysiological processes in MS and have facilitated early diagnosis of neurodegeneration. Conventional MRI sequences allow for the assessment of WM demyelination and the identification of active lesions. Today, the leading cause of irreversible disability accumulation and cognitive decline is neurodegeneration, not inflammation. Due to the availability of new DMTs, it is imperative to identify imaging biomarkers with prognostic value for disability progression and conversion to PMS [149,150,151,152]. For many years, the accumulation and expansion of new T2 lesions and contrast-enhancing T1 lesions (both biomarkers of inflammation) were the only parameters used for radiological follow-up and treatment-response monitoring in MS patients. However, these parameters do not correlate with the clinical evolution of the disease. Irreversible progression of disability is associated with neurodegeneration rather than inflammation, and it is highly associated with the degree of cortical atrophy [153,154,155,156]. Cortical atrophy is the most studied biomarker of neurodegeneration and represents one of the final sequelae of neurodegenerative processes. Novel MRI techniques allow for global and regional brain atrophy assessment from the early stages of the disease. However, they are unfeasible in current clinical practice due to technical limitations and physiological variations in cerebral volume (e.g., hydration, menstrual cycle) that require extensive knowledge of, and experience with, these imaging techniques. Other factors that influence the accuracy of the results are genetics (expression of apolipoprotein E), lifestyle (e.g., alcohol, smoking), concomitant disorders (diabetes and cardiovascular diseases) and medications [149]. Numerous clinical studies have demonstrated that the quantification of global brain atrophy can be a predictive biomarker for CIS conversion to RRMS [151,157]. Clinical progression of disability independent of relapses in RRMS patients, referred to as ‘silent progression’, is associated with a higher cerebral atrophy rate than in clinically stable RMS patients [158]. The role of cerebral atrophy as a biomarker for MS phenotype stratification, its association with disability severity and the presence and severity of cognitive decline were demonstrated in clinical studies. Therefore, cerebral atrophy assessment was added to the ‘no evidence of disease activity (NEDA)’ criteria 4 (NEDA-4) for the evaluation of DMT efficiency alongside other radiological (new T2 lesions/increasing T2 lesions, gadolinium-enhancing lesions) and clinical parameters (clinical relapses, confirmed disability progression) [149,159,160,161,162,163]. Global cerebral atrophy is present in all stages of MS. A study that followed 206 patients (180 RRMS, 14 SPMS, 12 PPMS) and 35 HCs demonstrated that the annualised percent brain volume change (PBVC/y) was significantly higher in MS patients (−0.51 ± 0.27%) than in HCs (−0.27 ± 0.15%). The annual brain volume loss rate exceeded the physiological limits in all disease phenotypes, with a PBVC/y of −0.52 ± 0.29% in RRMS patients and −0.45 ± 0.18% in PMS patients. ROC analysis established a PBVC/y cut-off of −0.37%, with a sensitivity of 67% and a specificity of 80% in distinguishing MS patients from HCs. When the cut-off was modified to −0.52%, the specificity reached 95% (i.e., the rate of false positives decreased to 5%). Even a cut-off of −0.4% had clinical significance; patients with a brain volume loss rate higher than −0.4% had a higher rate of EDSS decline [164]. Evaluating the segmental volume of different cerebral regions has been of interest in recent years. The development of new imaging techniques has allowed for a more detailed assessment of cortical GM and deep grey matter (DGM). Global brain atrophy appears secondary to neurodegeneration and neuroaxonal loss at the level of the GM. Patients with SPMS have reduced cortical GM and DGM volume compared with RRMS patients. Loss of GM volume in the temporal pole and posterior insula is more accelerated in SPMS patients (−1.21%) than in RRMS patients (−0.77%). DGM volume has a predictive value for disability accumulation in the time interval until progression if confirmed with EDSS. Eshaghi et al. [161] reported the highest predictive value of EDSS for progression in RRMS patients in thalamic volume (baseline thalamic atrophy increased the disability risk by 37%), hippocampal volume and angular gyrus volume. DGM atrophy appears early in MS evolution. A significant decrease in DGM volume was noted in CIS patients compared with HCs; however, no differences were reported regarding global brain volume and WM volume between the two groups. DGM atrophy evolves more rapidly than WM atrophy [149,161,163]. In addition to regional brain atrophy, isolated thalamic atrophy has the highest potential to be a prognostic biomarker for disease progression and disease progression. In a recent study by Hänninen et al. [163], the authors included 24 newly diagnosed RRMS patients and 36 SPMS patients. They demonstrated that isolated thalamic atrophy appeared before global brain atrophy (1/60 patients had global brain atrophy in the absence of thalamic atrophy and 16/60 patients had thalamic atrophy in the absence of global brain atrophy). SPMS patients had a significantly lower thalamic volume at baseline and at the end of the study than RRMS patients. Isolated thalamic atrophy at the onset of the study, even when not associated with global brain atrophy, represented a risk factor for EDSS progression and not reaching NEDA-3 at 2-year follow-up. The evaluation of global brain atrophy and atrophy of specific regions can be used to identify patients at risk of disease progression and transition from RRMS to PMS. Recent pathological studies have demonstrated that the progressive neurodegeneration found in more advanced stages of MS occurs secondary to chronic inflammation that is compartmentalised into the CNS. When the BBB is intact, inflammatory infiltrates are primarily found in the meningeal and perivascular spaces [165,166,167]. Some demyelinating lesions undergo early remyelination after the initial inflammatory stage, which prevents axonal degeneration, while others remain chronically demyelinated, without the presence of the inflammatory infiltrate, but with axonal loss and gliosis, or ‘chronic inactive lesions’ [168]. A third category of WM lesions is represented by chronic active lesions, or so-called ‘smouldering lesions’ or ‘slowly expanding lesions’ (SELs). In terms of pathophysiology, SELs present a central inactive demyelinated core surrounded by a slow, continuous inflammatory demyelinating process at the periphery, overlapped with incomplete remyelination. The result is irreversible myelin loss and axonal degeneration. Ongoing myelin degradation is followed by phagocytosis, which occurs at the periphery of the demyelinating lesions. Iron accumulates in activated microglia, which induces a paramagnetic rim at the lesion’s edge consisting of iron-loaded microglia on susceptibility-weighted images (SWI), and is the reason why this type is also called an ‘iron rim lesion’ (IRL). One-half of WM lesions have persistent inflammation and function as active chronic lesions in relapsing MS patients and at least in 60% of progressive MS patients [168,169]. Frequently, these lesions slowly expand and progress rapidly and with increased severity; their microstructural abnormalities are reflected by a T1 hypo signal, and they represent promising biomarkers for the transition of RRMS to PMS [165,170,171,172]. Cortical lesions (CLs) are another promising biomarker of disease progression. CLs are present in all evolutive phenotypes of MS, but their burden is highest in SPMS patients. CL number and volume correlate with disability progression, cognitive decline and the degree of cortical and global brain atrophy. A 7T MRI is required to detect CLs; conventionally, most clinical sites have 3T MRIs. Given the current radiological techniques, their applicability as progression biomarkers in current clinical practice is minimal [173,174]. Calabrese et al. [175] performed a longitudinal prospective study over a period of 5 years that included RMS and PMS patients. The authors demonstrated that a high CL burden at baseline was associated with clinical progression of the disease throughout the study. Therefore, the number of baseline CLs is an independent predictive factor for the progression of disability. Treaba et al. [170] performed a study on 111 MS patients: 74 with RRMS and 37 with SPMS. They used a 7T MRI scanner for radiological assessment. The number and volume of CLs in SPMS patients was significantly higher than in RRMS patients (p < 0.001). The model for disease-stage prediction reached a mean ± standard deviation (SD) for the area under the curve of 0.82 ± 0.08. The six most important radiological parameters that facilitated RRMS and SPMS distinction were global WM volume, thalamic volume, number and volume of WM lesions with no rim, leukocortical lesion volume and intraventricular CSF volume. In this model, the number of leukocortical lesions and IRLs was ranked low, as the 12th and 14th predictors of MS subtype; however, they were the main predictors of neurological disability progression (measured using the EDSS) over a mean period of 3.2 years. Therefore, IRLs and leukocortical volume have value as predictive biomarkers for disability progression, with salient implications on therapeutic strategies. Evaluating the hypothesis that the number and characteristics of SELs are predictive biomarkers for disability progression and conversion to PMS, Preziosa et al., published a study of 52 RRMS patients who were prospectively followed for 9.1 years and underwent imaging tests at baseline and after 6, 12 and 24 months. A significant association was noted between EDSS progression at 9.1 years and the median proportion of SELs (p = 0.045), the presence of at least four SEL lesions (p = 0.018), and the mean baseline magnetisation transfer ratio of the SELs (p = 0.026). Regarding the conversion risk to SPMS at 9.1 years from the baseline, the data revealed a significant association only with a decreased SEL magnetisation transfer ratio at onset (p = 0.037) and a reduction in T1 SEL signal on the 2-year MRI compared with the baseline MRI (p = 0.041). The results of this study suggest that the proportion of SEL-type lesions and their microstructural alterations are independent risk factors for disability progression (measured using the EDSS) and conversion to SPMS [165]. Various other clinical studies demonstrated the association between chronic active lesions and the clinical evolution of MS. Absinta et al., demonstrated that patients with ≥4 IRLs at baseline had a 1.6-fold increased risk of developing clinical disease progression compared with patients with no IRLs at baseline. When patients over 50 years old were excluded from the statistical analysis, the risk of developing clinical progression increased to 3.2-fold. Moreover, patients with ≥4 IRLs had a higher degree of global and regional brain atrophy (thalamus, putamen, caudate nucleus) than patients with no IRLs [169]. In most clinical studies, SEL and IRL assessments were performed using a 7T MRI apparatus; however, in recent years, published studies have demonstrated that these lesions can be easily visualised and assessed using a 3T MRI with SWI. This gradient echo imaging technique facilitates the identification of different MS lesions based on their paramagnetic properties. Blinderbacher et al., followed a group of patients with CIS (n = 32) and RRMS (n = 34) over 2.9 years. Using the SWI sequence, the researchers noted that IRLs were present in early MS stages. In 13 (19%) patients, these lesions were found at baseline, and the subjects had significantly lower Symbol Digit Modalities Test scores compared with the non-IRL subjects. A high IRL index correlated with higher brain volume loss throughout the study. Moreover, baseline IRLs correlated with a higher EDSS during the study, and statistical analysis revealed that baseline IRLs were independent risk factors for future disability progression [176]. After Hemond et al., demonstrated that paramagnetic rim lesions can also be visualized on susceptibility-weighted sequences using 1.5 T MRI, chronic active lesions have become an important imaging biomarker with the potential to be used in current clinical practice to evaluate the evolution of MS patients and the risk of disease progression [177]. As research advances on the identification of radiological biomarkers for the conversion from RRMS to PMS, in addition to conventional cerebral parameters, it has become apparent that spinal cord imaging has value as a predictor of progression. The annual rate of spinal cord atrophy is approximately −1.78% when both relapsing and progressive forms of MS are analysed and −2.08% when only progressive cases are analysed. The mean annual rate of cerebral atrophy is only −0.4% [178,179]. Spinal cord atrophy is less studied as a radiological parameter than cerebral atrophy in MS, mainly due to anatomical (smaller dimensions, higher flexibility) and radiological (lower tissue contrast, possible artefacts due to breathing patterns) limitations. Recent studies have demonstrated that three-dimensional sagittal T1-weighted sequences, which are part of routine cerebral MRI assessment in MS patients, include the upper part of the spinal cord, which is less affected by breathing patterns and pulmonary artefacts. Assessment of the spinal cord area is usually made at the level of C1–C2, C1–C3 and C2–C3 [149,180,181,182]. In a study published by Rocca et al., that followed 326 patients with RRMS, 41 with SPMS and 179 HCs for about 5 years, one-third of the patients presented with an increase in the EDSS, and 14% of the RRMS patients converted to SPMS. The low surface area of the spine and high baseline spinal cord lesions were independent predictive factors for disability in the next 5 years. Moreover, statistical analysis revealed that baseline spinal lesion burden was an independent risk factor for conversion to SPMS [183]. Bischof et al., followed an MS cohort (360 with RRMS and 47 with SPMS) and 80 HCs for a period of 12 years. They reported that 54 patients with RRMS converted to SPMS, and silent progression of disability was noted in 159 cases. The patients who had RRMS at baseline and converted to SPMS presented a high rate of cervical spinal cord atrophy, assessed at C1 vertebral level (−2.19%/year) at least 4 years before conversion, while the patients who remained diagnosed with RRMS had a mean rate of cervical spinal cord atrophy of −0.88%/year. After progressing to an SPMS phenotype, the spinal cord atrophy rate slowed to 1.63%/year. Statistical analysis revealed that the annual spinal cord atrophy rate was a predictive factor for disability progression and conversion to SPMS, as even a 1% decrease was associated with a reduced time interval (59%) to reaching the progressive phase of the disease. The patients who experienced silent progression of disability had an annual spinal cord atrophy rate of −1.1%/year, while the patients with clinical stability had a significantly lower rate of spinal area loss (−0.72%/year) (p = 0.004) [184]. Spinal cord atrophy is a valuable biomarker for predicting the progression of disability and progression to PMS. Early clinical progression of disability in RRMS patients towards PMS can be quantified by the atrophied T2-LV, which represents either the lesional tissue volume being replaced by CSF secondary to atrophy or the direct destruction of the lesion. In a study that included 176 RRMS patients who were followed clinically and radiologically for 10 years, the atrophied T2-LV in the first 6 months from baseline was used to differentiate between the patients who remained clinically stable throughout the study (n = 76) and the patients who presented with confirmed disability progression (n = 100). This is clinically important, as PBVC assessment could only be used to differentiate between the two groups after 2 years of radiological follow-up [149,153,185]. These results were confirmed in a much larger MS cohort (1089 RRMS, 217 SPMS, 124 CIS patients). Genovese et al., demonstrated that SPMS patients had a higher annualised atrophied T2-LV rate compared with RRMS patients (p = 0.001). In addition, the patients who were clinically stable throughout the entire study period had a significantly lower annualised atrophied T2-LV than the patients with confirmed disability progression (p < 0.001), and logistic regression analysis revealed that an annual increase of 1 mL of atrophied T2-LV was associated with a five-fold increase in the risk of disability progression. The patients who converted to SPMS (n = 67) had a higher annualised atrophied T2-LV compared with clinically stable patients (p = 0.002), and an increase of 1 mL per year of annualised atrophied T2-LV was associated with a 4.7-fold increased risk of developing SPMS [186]. Therefore, atrophied T2-LV is a potential radiological biomarker for early prediction of disability progression and disease conversion from RRMS to PMS. This study aimed to review potential biomarkers of transition that could facilitate early identification of RRMS patients who are at risk of progressing to PMS. This knowledge would allow for prompt DMT adjustment or escalation to constrain and delay the conversion towards PMS. The clinical implications of the latter objective would shorten the time interval until a diagnosis of PMS is established, and an appropriate and prompt therapeutic decision could be made before irreversible disability occurs. Early identification of transition biomarkers to PMS could represent a key step in paving the way towards personalised therapy. The molecular, OCT and MRI biomarkers presented in this review have promising value as predictive biomarkers of early transition from RRMS to PMS. The most promising paraclinical biomarkers that can feasibly be translated to current clinical practice are serum NfL, thalamic atrophy, chronic active lesions, spinal cord atrophy, GCIPL and RNFL; however, they need to be further validated in longitudinal, long-term studies with larger cohorts of MS patients.

PMC10002761

Yuanfen Gao,Xuewu He,Huayang Lv,Hanmei Liu,Yangping Li,Yufeng Hu,Yinghong Liu,Yubi Huang,Junjie Zhang

Epi-Brassinolide Regulates ZmC4 NADP-ME Expression through the Transcription Factors ZmbHLH157 and ZmNF-YC2

27-02-2023

bHLH157,C4 NADP-ME,epi-brassinolide,NF-YC2,photosynthesis

Maize is a main food and feed crop with great production potential and high economic benefits. Improving its photosynthesis efficiency is crucial for increasing yield. Maize photosynthesis occurs mainly through the C4 pathway, and NADP-ME (NADP-malic enzyme) is a key enzyme in the photosynthetic carbon assimilation pathway of C4 plants. ZmC4-NADP-ME catalyzes the release of CO2 from oxaloacetate into the Calvin cycle in the maize bundle sheath. Brassinosteroid (BL) can improve photosynthesis; however, its molecular mechanism of action remains unclear. In this study, transcriptome sequencing of maize seedlings treated with epi-brassinolide (EBL) showed that differentially expressed genes (DEGs) were significantly enriched in photosynthetic antenna proteins, porphyrin and chlorophyll metabolism, and photosynthesis pathways. The DEGs of C4-NADP-ME and pyruvate phosphate dikinase in the C4 pathway were significantly enriched in EBL treatment. Co-expression analysis showed that the transcription level of ZmNF-YC2 and ZmbHLH157 transcription factors was increased under EBL treatment and moderately positively correlated with ZmC4-NADP-ME. Transient overexpression of protoplasts revealed that ZmNF-YC2 and ZmbHLH157 activate C4-NADP-ME promoters. Further experiments showed ZmNF-YC2 and ZmbHLH157 transcription factor binding sites on the −1616 bp and −1118 bp ZmC4 NADP-ME promoter. ZmNF-YC2 and ZmbHLH157 were screened as candidate transcription factors mediating brassinosteroid hormone regulation of the ZmC4 NADP-ME gene. The results provide a theoretical basis for improving maize yield using BR hormones.

Epi-Brassinolide Regulates ZmC4 NADP-ME Expression through the Transcription Factors ZmbHLH157 and ZmNF-YC2 Maize is a main food and feed crop with great production potential and high economic benefits. Improving its photosynthesis efficiency is crucial for increasing yield. Maize photosynthesis occurs mainly through the C4 pathway, and NADP-ME (NADP-malic enzyme) is a key enzyme in the photosynthetic carbon assimilation pathway of C4 plants. ZmC4-NADP-ME catalyzes the release of CO2 from oxaloacetate into the Calvin cycle in the maize bundle sheath. Brassinosteroid (BL) can improve photosynthesis; however, its molecular mechanism of action remains unclear. In this study, transcriptome sequencing of maize seedlings treated with epi-brassinolide (EBL) showed that differentially expressed genes (DEGs) were significantly enriched in photosynthetic antenna proteins, porphyrin and chlorophyll metabolism, and photosynthesis pathways. The DEGs of C4-NADP-ME and pyruvate phosphate dikinase in the C4 pathway were significantly enriched in EBL treatment. Co-expression analysis showed that the transcription level of ZmNF-YC2 and ZmbHLH157 transcription factors was increased under EBL treatment and moderately positively correlated with ZmC4-NADP-ME. Transient overexpression of protoplasts revealed that ZmNF-YC2 and ZmbHLH157 activate C4-NADP-ME promoters. Further experiments showed ZmNF-YC2 and ZmbHLH157 transcription factor binding sites on the −1616 bp and −1118 bp ZmC4 NADP-ME promoter. ZmNF-YC2 and ZmbHLH157 were screened as candidate transcription factors mediating brassinosteroid hormone regulation of the ZmC4 NADP-ME gene. The results provide a theoretical basis for improving maize yield using BR hormones. As one of the important crops in the world, the yield of maize is only lower than wheat and rice. Enhancing photosynthetic capacity is one of the ways to increase maize yield [1]. Photosynthesis is a crucial process in green plants; therefore, improving the efficiency of photosynthesis is one of the ways of increasing yield [2]. Maize photosynthesis occurs mainly through the C4 pathway. Unlike C3 plants, C4 plants first form malic acid by phosphoenolpyruvate (PEP) and CO2 in mesophyll cells [3]. Malic acid is transferred to sheath cells and releases CO2 by decarboxylase, which enters the photosynthetic carbon cycle via ribulose diphosphate (RuBP) carboxylase in the chloroplasts of sheath cells [4]. Therefore, decarboxylase is crucial for CO2 fixation in maize, and it is of great significance to study the regulation of the expression of its encoding genes to improve photosynthetic efficiency. Three types of decarboxylases catalyze deacidification of malic acid, namely NADP-dependent malic enzyme (NADP-ME), NAD-dependent malic enzyme (NAD-ME), and phosphoenolpyruvate carboxykinase (PCK) [5], and the decarboxylase in maize is mainly NADP-ME [6]. Three NADP-ME isoforms have been identified in maize: one is the plastid type (C4 NADP-ME) involved in C4 photosynthesis, and it is subcellularly localized in the chloroplast [7]; the other is the plastid non-phototype (C4 non-NADP-ME) that responds to plant defense inducers [8]; and the third is the cytoplasmic subtype (CYT-NADP-ME), which is enriched in embryos and roots and participates in the control of cytosolic acid levels [7]. Although the three NADP-MEs are similar, the expression level of ZmC4 NADP-ME in the leaves is 20 times more than that of cytoplasmic NADP-ME [9]. Overexpression of maize C4-NADP-ME in tobacco increases the efficiency of net CO2 fixation, and this transgenic tobacco exhibited a higher phenotype [10], and overexpression of SbC4-NADP-ME in sweet sorghum, which is a C4 plant, has a positive effect on its photosynthetic capacity. Brassinosteriods (BRs), a sterol hormone type, have the ability to affect plant photosynthesis. More than 70 BRs have been identified to date, with 24-epibrassinolide (24-EBL) and 28-homobrassinolide (28-homobrassinolide, 28-HBL) being the most active and commercialized [11]. BRs can promote CO2 fixation and dark reactions in plant photosynthesis [12], and studies have demonstrated that exogenous spraying or overexpression of endogenous brassinolide (BL) on crops, such as soybean [13], Brassica juncea [14], Triticum aestivum [15], rice [16], cucumber [17], and tomato [18], has improved the transcription level and activities of dark reaction enzymes, and promoted CO2 assimilation, resulting in an increase in photosynthetic products. In fact, EBL regulates photosynthesis through different pathways. EBL affects stomatal development by regulating the bHLH family transcription factor SPEECHLESS, and further has an impact on plant photosynthesis [19]. The BES1(BRINSENSITIVE 1-EMS-SUPPRESSOR1) mediated the regulation of EBL on chloroplast development [20]. However, the regulation of transcription factors mediated by EBL on carbon fixation is not clear. It has been reported that the transcription levels of RCA (rubisco activase) and fructose-1,6-bisphosphatase (FBPase) are promoted by BR treatment [21,22]. However, whether BL has an effect on the expression of the C4-NADP-ME gene in C4 plants and the possible mechanisms is still poorly understood. In this experiment, the transcriptome analysis of maize leaves treated with EBL was performed, and it was confirmed that EBL played a role in the differential expression of maize leaf photosynthesis genes. We studied the effect of EBL on the expression of the ZmC4-NADP-ME gene in maize, and screened and identified two transcription factors that might mediate BL regulation of the expression of ZmC4-NADP-ME. To determine the genes regulated by EBL in maize seedlings, DEGs were analyzed following EBL treatment. The results showed that 2041 DEGs were observed in 0.5 μM EBL treatment, of which 1051 DEGs were upregulated and 990 DEGs were downregulated (Figure 1, Supplementary Table S1). The KEGG pathway analysis was performed in DEGs to identify the enriched pathways. The five main KEGG pathways were photosynthetic antenna protein, ribosome, porphyrin and chlorophyll metabolism, photosynthesis, and synthesis of secondary metabolites (Figure 1), indicating that EBL treatment affects plant chlorophyll and photosynthetic electron transport [20]. The carbon fixation pathway was not the main enriched pathway in KEGG, but the results also showed that DEGs in the EBL treatment were significantly enriched in the carbon fixation pathway (p < 0.05) (Supplementary Table S1). Previous experiments have shown that exogenous spraying of EBL promotes the transcription of dark reaction enzymes [23]. We found that the expression of ZmC4 NADP-ME following EBL treatment was upregulated in RNA-seq (Figure 2A). To further study how EBL regulates ZmC4 NADP-ME expression, different concentrations of EBL were used to spray maize seedlings or whole plants in the field. The results showed that the expression of the ZmC4 NADP-ME gene significantly increased by 1.8 times on the 5th day after 150 nmol/L EBL in the field. On the 10th day after treatment with 100 nmol/L EBL, the expression of the gene increased by approximately seven times. The expression of ZmC4 NADP-ME in the leaves of maize seedlings increased significantly by 8.6 times after 24 h EBL treatment (Figure 2B). These results show that EBL can promote the expression of C4-NADP-ME in maize leaves. We screened for transcription factors involved in the regulation of ZmC4 NADP-ME expression mediated by EBL. The upregulated transcription factors in the transcriptome data were selected, and Pearson correlation coefficients (PCCs, R) were calculated with the reference transcriptome in leaves at different stages provided by Stelpflug [24] (Supplementary Table S1). Fluorescence quantification was performed to verify the transcription levels of transcription factors after EBL treatment. The bHLH and NF-Y family transcription factors that showed enhanced expression after EBL treatment and may be involved in carbon fixation were selected (Figure 2C) [25]. The R coefficients between ZmC4 NADP-ME and bHLH157 or NF-YC2 were 0.66 and 0.75, respectively (Supplementary Table S2). This indicated that ZmC4 NADP-ME was moderately positively (0.5 ≤ |r| < 0.8) correlated with the two transcription factors, and the expression levels of bHLH157 and NF-YC2 were significantly increased by 2.7 and 9.3 times, respectively, after exogenous EBL treatment (Figure 2C). Moreover, the dual-luciferase system showed that bHLH157 and NF-YC2 promoted the expression of genes downstream of the C4-NADP-ME promoter (Figure 3A). The yeast one-hybrid assay showed that bHLH157 and NF-YC2 in yeast combined with the C4-NADP-ME promoter to make the experimental group of yeast grow better on the SD/-Leu/-Trp/-His medium with 150 mmol/L 3-AT than the control (Figure 3C,D). We also verified the synergy between bHLH157 and NF-YC2; when bHLH157 and NF-YC2 were transformed together, the promoter activity was further improved compared to when they were transformed separately (Figure 3B). However, yeast two-hybrid experiments showed that bHLH157 and NF-YC2 did not interact in yeast cells (Figure 3E). The subcellular localization and self-activation activity were used to determine the transcription factor characteristics of bHLH157 and NF-YC2. NF-YC2 and bHLH157 were located in the nuclei of maize leaf cells (Figure 4A). The results of yeast self-activation indicated that ZmbHLH157 and ZmNF-YC2 have transcriptional activation activities (Figure 4B). Additionally, the expression patterns of ZmC4 NADP-ME, ZmbHLH157 and ZmNF-YC2 in different parts of maize were analyzed to explore the temporal and spatial specificity of gene expression. The results showed that ZmbHLH157 and ZmNF-YC2 have higher expression levels in leaves (Figure 4C). To further study the binding sites between the transcription factors, bHLH157 and NF-YC2 and the promoter of ZmC4 NADP-ME, Plant Care (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/, accessed on 22 March 2019) was used to analyze the cis-acting elements in the promoter region [26]. To preliminarily explore the possible binding regions, a dual luciferase system was used to verify the transactivation of bHLH157 and NF-YC2 on ZmC4 NADP-ME promoters of different lengths. The cis-acting element analysis results indicated that the binding site of cis-acting element E-box that bHLH157 may bind to is distributed in the ZmC4 NADP-ME promoter at −1152 bp, −669 bp and −435 bp (Table 1). The cis-acting element CCAAT-box that NF-YC2 may bind to was mainly distributed at the −1495 bp promoter (Table 1). When the length of the ZmC4 NADP-ME promoter was 1118 bp, the effect of bHLH157 and NF-YC2 on the relative activity of LUC was significantly reduced in comparison to a length of 1616 bp, indicating that the region from −1616 bp to −1118 bp was the binding region of bHLH157 and NF-YC2 (Figure 5A–C). We speculated that NF-YC2 may bind to the CCAAT-box or CAAT-box in this region, while bHLH157 may bind to the G-box in this region, and the remaining sites are presumably related to the constitutive activity of the promoter. To determine whether the effect of EBL on the expression of bHLH157 and NF-YC2 further activated the ZmC4-NADP-ME promoter, EBL was added to the protoplasts. The results showed that the transactivation effect of bHLH157 and NF-YC2 on the ZmC4-NADP-ME promoter was significantly increased by 1.55 and 0.79 times, respectively, compared with the control after adding 1 nmol/L EBL for 12 h (Figure 5D). In contrast, EBL treatment had no effect on the expression of the LUC reporter under the Ubi promoter, indicating that the promotion effect of EBL on the expression of ZmC4-NADP-ME in maize protoplasts was related to the transcription factor bHLH157 or NF-YC2. BR has previously been found to indirectly affect plant photosynthesis by controlling stomatal size and chloroplast development [11,27]. We found that the DEGs of EBL treatment were enriched in the photosynthetic antenna protein, and porphyrin and chlorophyll metabolism pathway of maize leaves. BR affects chlorophyll content in Brassica juncea [11]. And we have previously found that chlorophyll content in maize leaves is affected by EBL treatment [23]. The light-harvesting complex in chloroplast mainly consists of photosynthetic antenna protein and chlorophyll. It has been found that BRs play a role in the regulation of chloroplast development [20]. We speculate that EBL differentially regulates genes in the photosynthetic antenna protein and chlorophyll metabolism pathway in leaves, which is an important factor affecting photosynthesis in chloroplast. Many studies have shown that bHLH family transcription factors can participate in the expression of carbon fixation-related enzyme genes to regulate photosynthesis [28]. Transcription factors of the bHLH family, such as BES1 (BRASSINOSTEROIDROIDISTTIVE 1-EMS-SUPPRESSOR 1) and BZR1 (BRASSINAZOLE RESISTANT 1), are important transcription factors in BRs signal transduction, and E-box (CANNTG), a cis-acting element response to BRs, is a putative binding element of the bHLH family [29]. This experiment showed that the bHLH157 transcription factor activates the promoter of ZmC4 NADP-ME. The transcription factor was highly expressed in maize leaves and responded to exogenous spraying. Moreover, bHLH157 transcription factors may combine with the G-box in the C4-NADP-ME promoter region, which is a common binding region of the bHLH family transcription factors [30]. Moreover, it has been found that certain transcription factors in the bHLH family can bind to the G-box of the C4-NADP-ME promoter [28]. The bHLH157 transcription factor transferred into protoplasts were treated with EBL, which had no affect on the relative LUC activity; however, when the bHLH157 and C4-NADP-ME promoter were co-transferred and treated with EBL, the relative value of LUC increased by 1.55 times. Presumably, a low concentration of EBL further promoted the Ubi::bHLH157 and C4-NADP-ME promoter interaction, and this may be related to the regulation of protein synthesis or modification by EBL. The F-Y family transcription factors are also involved in the regulation of dark reaction enzymes [31]. The three different subunits in the NF-Y family (NF-YA, NF-YB, and NF-YC) generally form a complex that binds to the CCAAT cis-acting element in the promoter and can also interact with transcription factors of other families to regulate gene expression [32], but only a few active NF-Y complexes are known in plants [33]. In this study, ZmNF-YC2 was found to be a nuclear localization protein. The self-activation verification of NF-YC2 showed that the NF-YC2 subunit alone was self-activated, but only the overexpression of the NF-YC2 subunit in protoplasts still increased the activity of the ZmC4-NADP-ME promoter. We attribute this phenomenon to the correlation between gene expression levels. A correlation between the expressions of related proteins was also found in rice. In transgenic rice plants with antisense or RNAi of the chloroplast biogenesis regulatory gene OsHAP3A, the expression of OsHAP3A was reduced, and OsHAP3B and OsHAP3C formed a complex with it [31]. Transcription factors of the NF-Y family have been found to be involved in BL signal transduction, such as LEAFY COTYLEDON1 (LEC1) [33]. Overexpression of LEC1 would inhibit the negative regulator BRH1 of BR signaling, and the gene DOGT1 related to BR catabolism [34]. Notably, Su et al. discovered ZmNF-YC2 is a positive regulator of maize flowering time under long-day conditions [35]. We speculate that NF-YC2 may be involved in multiple light-related physiological processes with powerful functions. Previous studies have shown that NF-Y family proteins can interact with other families; for example, OsbZIP76 in rice interacts with OsNF-YBs to regulate endosperm development [36]; and the NF-YB1-YC12-bHLH144 ternary complex regulates the key granule-bound starch synthase gene Wx (granule-bound starch synthase) [33]. In this study, we found that bHLH157 and NF-YC2 synergistically promoted the activity of the ZmC4-NADP-ME promoter compared with that of each single promoter (Figure 3B). However, bHLH157 and NF-YC2 did not interact in yeast (Figure 3E). We speculate that there is a third transcription factor involved in this binding [37]. Future studies should explore the function of the bHLH157 and NF-YC2 transcription factors in EBL signaling, which could help to understand in detail the mechanism of EBL in enhanced plant photosynthetic capacity. In this experiment, the maize inbred line Mo17 was used as the experimental material, and the seeds were provided by Sichuan Agricultural University, Sichuan China. Mo17 seeds were sown in peat soil (peat soil: vermiculite = 1:1) and the day and night culture conditions were: 25 °C, 3000 lx light for 16 h, and 23 °C, dark, for 8 h. Maize in the field was planted in Ya’an City, Sichuan Province. In the field, maize plants at the 8-leaf stage were sprayed with EBL all over the plant at concentrations of 50, 100, 150, and 500 nmol/L once every five days, with each group comprising five plants. The sixth leaf of maize in the field at the 8-leaf stage was selected for quantitative real-time PCR (qRT-PCR) with two biological replicates [38]. Six-leaf stage roots (roots, R), stems (St), leaves (leaves, L), filaments (F), anthers (A), pollen (pollen, P), post-pollination seeds (S), embryos (Em), and endosperm (En) in maize were used for semi-quantitative analysis of the relative expression of transcription factors [39]. Cellulase and pectinase were used in the enzymatic cell wall to produce protoplasts. These cells were then treated with 0.02 nmol/L and 0.1 nmol/L EBL, including three biological replicates [40]. Primers were designed using Primer Premier 5.0; the primer sequences are presented in the supplementary file. The isolated maize leaves (the two leaves of a one-leaf seedling are lightly divided by a sharp blade into segments of about 2 cm) used for RNA-seq were treated with 1 mL of 0.5 μmol/L EBL on 1/2MS medium (Solarbio, Beijing, China) for 6 h, and distilled water was used as the control. Leaves treated with 0, 0.2, and 0.5 μmol/L EBL, respectively, were cultured for 24 h. The content of EBL was chosen by a previous study [23]. According to the manufacturer’s protocol, the Macherey-Nagel NucleoSpin miRNA isolation kit (Solarbio, Beijing, China) was used to extract RNA from different treated maize leaves, and these RNA (>200 nt) components were used for RNA-seq (novogene, Beijing, China). The original data were filtered using HISAT software and compared with the reference genome (ftp://ftp.ensemblgenomes.org/pub/plants/release-41/fasta/zea_mays/dna/, accessed on 28 December 2018). The expected number of fragments per kilobase of transcript sequence per million base pairs sequenced (FPKM) was evaluated, and the gene expression levels of each sample were analyzed using the union model in the HTSeq software [38]. The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway of differentially expressed genes (DEGs) was significantly enriched [41]. The ClonExpress®II One-Step Cloning Kit (Vazyme, Nanjing, China) was used to recombine bHLH157 or NF-YC2 with the pGBKT7 vector. The recombinant plasmid was transformed into AH109 yeast cells and cultured at an OD600 of 0.5. Transformed yeasts were diluted 10−1/10−2/10−3/10−4 times, then spotted on the surface of an SD/-Trp or SD/-Trp/-His/X-α-gal solid medium, and cultured at 28 °C for 3 days [39]. The pCAMBIA2300-35S-EGFP vector was chosen to carry the bHLH157or NF-YC2 genes with the stop codon removed. The construct concentration for protoplast transformation was measured using the NanoDrop2000 (Thermo Scientific, Waltham, MA, USA). The construct was transformed into protoplasts of maize leaf cells by PEG-mediated transformation and cultured for 12 h at 25 °C under 2000 lx light intensity [42]. A laser confocal microscope (FV1200, OLYMPUS, Shinjuku City, Japan) was used to observe the location of bHLH157 or NF-YC2. The pGADT7-Rec2 vector carrying the CDS sequence of bHLH157 or NF-YC2 and the pHIS2 vector carrying the promoter of C4 NADP-ME were co-transferred to the Y187 yeast competent and then expanded to an optical density at 600 nm (OD600) of 0.5. The yeast concentration was diluted with sterile water by 1, 10−1, 10−2, 10−3, 10−4 times. Then, 5 μL of the sample was added to SD/-Leu/-Trp/-His solid medium containing 0, 50, 100, and 150 mmol/L 3-AT. PEG-Ga solution (40% PEG 4000, 0.2 M Mannitol, 0.1 M GaCl2) was used by protoplasts transformation. The single sample needed 110 μL PEG-Ga solution, and this transformation reaction lasted six minutes before dilution with 440 μL MMG (15 mM MgCl2, 0.4 M Mannitol, 4 mM MES). Protoplasts after transformation reaction were washed by WI (0.5 M Mannitol, 4 mM MES, 20 mM KCl) for dual-Luciferase assay. The C4-NADP-ME (Zm00001eb121470) promoter sequence was recombined with the PBI221 vector, in which the GUS (β-glucuronidase) reporter gene was replaced with the LUC (fireflyluciferase) reporter gene [43]. The coding sequences of bHLH157 and NF-YC2 were cloned into the PBI221 vector, and the two recombinant vectors were co-transformed into maize leaf protoplasts to evaluate the relative activity of LUC. Dual-luciferase activity (Promega) was measured to verify the interactions between bHLH157 and/or NF-YC2 and C4 NADP-ME. GUS activity was used as an internal reference to normalize the transformation efficiency of protoplasts, and the relative ratio of LUC/GUS (4 h–0 h) between the experimental group and control was used to represent the relative activity of the promoter [44]. SPSS 22.0 was used to process the data, and Tukey’s test was performed to analyze significant differences among groups. In this study, transcriptome of maize seedlings treated with EBL indicated the role of EBL in metabolic pathways related to maize leaf photosynthesis. Genes in photosynthetic antenna proteins, porphyrin and chlorophyll metabolism are regulated by EBL. We found ZmC4-NADP-ME was upregulated by EBL, and the ZmNF-YC2 and ZmbHLH157 transcription factors were moderately positively correlated with ZmC4-NADP-ME by co-expression analysis. Transcription factors ZmNF-YC2 and ZmbHLH157 can promote ZmC4 NADP-ME promoter activity, and have a certain degree of superposition effects. Further research showed that both ZmNF-YC2 and ZmbHLH157 transcription factor binding sites on the promoter of ZmC4 NADP-ME were located between −1616 bp and −1118 bp of the target promoter. EBL is involved in regulating multiple metabolic pathways related to photosynthesis. The results provide a theoretical basis for improving maize yield using BL hormones.

PMC10002778

Roxana Resnik,Fabiana Lopez Mingorance,Francisco Rivera,Florencia Mitchell,Claudio D. Gonzalez,Maria I. Vaccaro

Autophagy in Inflammatory Response against SARS-CoV-2

03-03-2023

COVID-19,macroautophagy,ATG proteins,mitophagy,SIRS,MOF,pyroptosis,NETosis

The coronavirus disease pandemic, which profoundly reshaped the world in 2019 (COVID-19), and is currently ongoing, has affected over 200 countries, caused over 500 million cumulative cases, and claimed the lives of over 6.4 million people worldwide as of August 2022. The causative agent is severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Depicting this virus’ life cycle and pathogenic mechanisms, as well as the cellular host factors and pathways involved during infection, has great relevance for the development of therapeutic strategies. Autophagy is a catabolic process that sequesters damaged cell organelles, proteins, and external invading microbes, and delivers them to the lysosomes for degradation. Autophagy would be involved in the entry, endo, and release, as well as the transcription and translation, of the viral particles in the host cell. Secretory autophagy would also be involved in developing the thrombotic immune-inflammatory syndrome seen in a significant number of COVID-19 patients that can lead to severe illness and even death. This review aims to review the main aspects that characterize the complex and not yet fully elucidated relationship between SARS-CoV-2 infection and autophagy. It briefly describes the key concepts regarding autophagy and mentions its pro- and antiviral roles, while also noting the reciprocal effect of viral infection in autophagic pathways and their clinical aspects.

Autophagy in Inflammatory Response against SARS-CoV-2 The coronavirus disease pandemic, which profoundly reshaped the world in 2019 (COVID-19), and is currently ongoing, has affected over 200 countries, caused over 500 million cumulative cases, and claimed the lives of over 6.4 million people worldwide as of August 2022. The causative agent is severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Depicting this virus’ life cycle and pathogenic mechanisms, as well as the cellular host factors and pathways involved during infection, has great relevance for the development of therapeutic strategies. Autophagy is a catabolic process that sequesters damaged cell organelles, proteins, and external invading microbes, and delivers them to the lysosomes for degradation. Autophagy would be involved in the entry, endo, and release, as well as the transcription and translation, of the viral particles in the host cell. Secretory autophagy would also be involved in developing the thrombotic immune-inflammatory syndrome seen in a significant number of COVID-19 patients that can lead to severe illness and even death. This review aims to review the main aspects that characterize the complex and not yet fully elucidated relationship between SARS-CoV-2 infection and autophagy. It briefly describes the key concepts regarding autophagy and mentions its pro- and antiviral roles, while also noting the reciprocal effect of viral infection in autophagic pathways and their clinical aspects. The coronavirus disease pandemic, which profoundly reshaped the world in 2019 (COVID-19), and is currently ongoing, has affected over 200 countries, caused over 500 million cumulative cases, and claimed the lives of over 6.4 million people worldwide as of August 2022 [1]. The causative agent, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is a member of the Coronaviridae family of viruses, a group of positive-sense, single-stranded RNA genome viruses of approximately 30 kb in length [2]. Its genome consists of 11 genes with 11 reading frames that produce 16 non-structural proteins (NSP1 to NSP16) and 4 structural proteins, including the fusion trimeric spike (S), the envelope protein (E), the nucleocapsid protein (N), and the membrane glycoprotein (M) [3]. These reading frames generate eight accessory proteins: ORF3a, ORF3b, ORF6, ORF7a, ORF7b, ORF8a, ORF8b, and ORF9b [3,4]. NSPs are key for viral RNA replication and immune avoidance, and the accessory proteins play diverse roles, aiding in viral infection, replication, and transmission [5]. Figure 1 contains a diagram of the SARS-CoV-2 virus genome, highlighting the two open reading frames: ORF1a and ORF1b [4]. Autophagy is a catabolic process that sequesters damaged cell organelles, proteins, and external invading microbes and delivers them to the lysosomes for degradation [6,7,8,9]. It is a fundamental and evolutionarily conserved eukaryotic cellular process that has multiple effects on cell survival, homeostasis, and immunity [10]. Non-canonical autophagy describes other processes that use the autophagy molecular machinery, such as phagocytosis, inflammatory signaling, and secretion [11]. Autophagy plays an essential role in promoting RNA virus replication by inhibiting innate antivirus immune responses [12], or promoting infectivity via the autophagy-related secretion of vesicles loaded with virus [13,14,15,16]. Coronaviruses can cause respiratory infections of varying severity. Seasonal human CoVs (HCoVs) can cause mild to moderate upper respiratory tract infections with cold-like symptoms in humans [17]. On the other hand, highly pathogenic beta-CoVs have been responsible for multiple deadly outbreaks in the 21st century, including SARS-CoV (2003), the Middle East respiratory syndrome coronavirus (MERS-CoV) in 2012, and the current SARS-CoV-2 (2019) [17,18]. For the ongoing pandemic, self-isolation quarantine and vaccination are still the mainstream strategies in responding to the virus, and therapeutic strategies remain challenging. For knowledge and therapeutic purposes, depicting the virus’ life cycle and pathogenic mechanisms, as well as the cellular host factors and pathways involved during SARS-CoV-2 infection, has great relevance [19]. The viral replication process comprises mainly six steps: binding and attachment, where the virus and the host cell receptors interact—this interaction is performed by the viral protein spike, (S) and the cell host receptors, angiotensin-converting enzyme-2 (ACE2), TMPRSS2 and integrins [19]; viral entry, using the two entry pathways of membrane fusion and endocytosis—where the virus fuses with the host cell membrane via the cleavage of protein S by the serine proteases of the host cell endocytic pathway; transcription and translation, where viral proteins are translated from viral RNA; viral replication; nucleocapsid packaging, where new virion packaging at the endoplasmic reticulum (ER) and Golgi apparatus occurs; budding and egress, i.e., the release of viral particles by exocytosis, as these RNA viruses traverse the Golgi apparatus and trans-Golgi network (TGN), where envelope proteins receive additional post-translational modifications and exit the cell via lysosomal exocytosis [20]. The assembled viral components further undergo maturation in the Golgi vesicles to form the mature virion, ready to be released to the extracellular environment [21,22,23]. Gosh and colleagues found that β-coronaviruses utilize lysosomal trafficking to exit host cells, rather than other conventional secretory pathways [24]. This review aims to describe the main aspects that characterize the complex and not yet fully elucidated relationship between SARS-CoV-2 infection and autophagy. It briefly describes the key concepts regarding autophagy, including its pro- and antiviral roles, while also mentioning the reciprocal effect of viral infection in autophagic pathways and their clinical consequences. Thus, this review is relevant to the goals of developing new therapeutic targets and opportunities to treat COVID-19 and establishing better prognostic markers by describing molecules that correlate with disease severity, in order to gain a better understanding of a disease that has modified the world in such an impactful way. Canonical autophagy includes three different major types: (1) microautophagy, which implies the direct sequestering of small molecules by the invagination or protrusion of the membranes of lysosomes [25]; (2) chaperone-mediated autophagy (CMA), which transfers proteins or other targeted molecules through a chaperone to the lysosome, specifically to its LAMP2A receptor; regarding this, Cuervo and colleagues led to the identification of the lysosome-associated membrane protein type 2A (LAMP-2A) as a CMA receptor [26]; LAMP2A is a single-span membrane protein with a very heavily glycosylated luminal region and a short (12 amino acids) C-terminus tail exposed on the surfaces of the lysosomes, where substrate proteins bind [27]; (3) macroautophagy, which is the third and most prevalent form, in which double-membrane vesicles called auto-phagosomes form and engulf the substrates to be transported to the lysosome, in which they then fuse and release their cargo for degradation [28,29]. The molecular machinery involved in macroautophagy is composed of evolutionarily conserved autophagy-related proteins (ATG) [7,30,31]. ATG protein complexes are sequentially recruited for autophagosome biogenesis, and, after successive steps, the autophagosome fuses with a lysosome, forming the degradation compartment called the autolysosome (Figure 2). Autophagy is induced in mammalian cells by mTORC1 inhibition and AMPK activation, which activate autophagy when a lack of nutrients or energy is detected. The UKL1 complex, ULK1-ATG13-FIP200-ATG101, drives the pre-autophagosomal structure. Then, the class III PI3K-kinase complex, BCN1-ATG14-VPS34-VPS15, mediates membrane phosphorylation and two ubiquitin-like systems involving ATG7, ATG12, ATG10, ATG5, ATG16, and ATG3, which allows the recruitment of Atg8 family proteins, promoting membrane expansion and cargo sequestration [7,32,33]. Two transmembrane proteins are involved in this sophisticated membrane trafficking process, ATG9 and (VMP1) [34,35,36,37]. VMP1 is a well-described transmembrane protein essential for mammalian cell autophagy [7,34]. TMEM41B is a transmembrane protein recently linked to autophagy that shares a conserved VTT domain with VMP1, and, together, they belong to the conserved DedA family of half transporters [14,38]. Finally, soluble N-ethylmaleimide sensitive factor attachment protein receptors (SNAREs), such as STX17, facilitate the fusion between autophagosomes and lysosomes, forming the autolysosome structure, where cargoes are degraded and recycled into the cytosol [39]. The autophagosomal fusion is regulated by the homotypic fusion and protein sorting (HOPS) complex. Then, mTOR phosphorylates UVRAG and activates the PI3KC3-C2 complex to produce a lysosomal pool of PI3P. In this way, mTOR controls this process and regulates tubular initiation and maintenance. Macroautophagy can be non-selective or selective. In selective autophagy, one specific cellular component is recognized and sequestered in autophagosomes [40]. Examples of this mechanism are mitophagy, lipophagy, zymophagy, and xenophagy, the latter being a pathway by which the cell sequesters and degrades external invading microorganisms, including viruses, as a protective defense mechanism [41] (Figure 2). Non-canonical autophagy describes other processes that use the autophagy molecular machinery, such as phagocytosis, inflammatory signaling, and secretion. One of the most relevant non-canonical autophagy pathways is the recently described secretory autophagy [42]. Secretory autophagy has been related to unconventional secretion, which links autophagy with an anabolic condition where cytosolic proteins lack an N-terminal signal peptide, which is the main peptide needed to undergo the conventional secretion pathway through the ER and Golgi apparatus. These are cytosolic proteins secreted from the cells to perform their biological functions. This process is associated with the canonical autophagic pathway as it shares regulation factors (ATGs), which also lead to the formation of autophagic membranes. These autophagic regulating factors include ULKs, BCN1, LC3s, and GABARAPs (analogs of the yeast Atg8) [42]. There is considerable evidence suggesting that autophagy could play a role in assisting with the viral replication cycle at different stages. As is well known, the SARS-CoV-2 virus enters the host by binding its receptor-binding domain (RBD), located in the spike protein, to the ACE2 host cell receptor. Moreover, a bioinformatic analysis showed that two integrin-binding regions are present in the cytoplasmic C-terminal tail of the host ACE2 protein. Additionally, the viral S protein has an arginine–glycine–aspartic acid (RGD) motif-binding domain that links to integrins, suggesting that integrins could be used as co-receptors for viral entry [43]. Furthermore, one specific type of short linear motif (SLiM) has been found to be present in the tails of integrin β3 and ACE2. Thus, these structures from the autophagic machinery could be used to facilitate viral attachment, entry, and replication (as highlighted in the first step of Figure 3). Additionally, LC3-interacting region motifs (LIRs) have been identified in the tails of integrin β3 and ACE2. These LIRs are involved in the autophagic pathway through interaction with LC3, a marker of autophagosomes [19,44]. These LIR motifs also facilitate viral attachment, entry, and replication. Several viruses, including coronaviruses (CoVs), take advantage of cellular autophagy to facilitate their own replication. SARS-CoV-2 mediates its replication through a dependent or independent ATG5 pathway using specific double-membrane vesicles that can be considered similar to autophagosomes [45]. As previously described, the virus enters the cell via the endosomal pathway. The spike protein is composed of two subunits: one binding subunit (S1) and one fusion subunit (S2). First, the S1 subunit binds to the host cell receptor ACE2. This union induces conformational changes in protein S. Second, the S2 subunit is activated. This activation occurs through two consecutive cleavage steps: a cleavage between the S1 and S2 domains of protein S by Furin is produced, and it then undergoes further cleavage at the S2’ site, which promotes the unmasking and activation of the fusion peptide [46,47]. To activate the spike protein’s fusion potential, a second cleavage performed by the host’s proteases is required. The cleavage can occur at different stages of the virus infection cycle by different host proteases, such as Furin (convertase), TMPRSS2 (cell surface protease), and Cathepsin L (lysosomal protease) [48,49]. Cathepsin L, which links the virus cycle to the autophagy process, acts at a low pH, degrades cargo, and maintains autolysosome homeostasis and autophagic flux [50]. By cleaving the spike protein at S2’, it mediates virus membrane and autolysosome fusion, thus facilitating the release of viral RNA into the host cell. Smieszek et al., demonstrated that the inhibition of Cathepsin L could significantly reduce the entry of viruses into host cells [51]. It has also been shown that TMPRSS2, Furin, and Cathepsin L proteases have cumulative effects to activate virus entry and increase the pathogenicity of SARS-CoV-2 [52,53] (Figure 3, pathway 4). Once the virus has entered the cell, it uses the autophagy machinery for its own benefit through viral proteins, NSPs. Recently, as shown in Figure 3, pathway 3, transmembrane proteins related to autophagy, i.e., VMP1 and TEMEM41B, have been reported as critical host factors at the early stages of viral infection [52]. VMP1 and TMEM41B both contribute to different stages of DMV formation. Ji et al., have revealed that DMV biogenesis is impaired in VMP1 and TMEM41B knockout cells. Analysis using transmission electron microscopy revealed that the formation of DMVs was substantially inhibited in cells lacking these autophagy proteins. Hence, by inhibiting VMP1 and TMEM41B expression, the virus is unable to hide from the immune sensors in DMVs, thus reducing its protection from the immune system [54,55]. Shneider et al., showed that TMEM41B participates in the transport of lipids to the membrane and that, together with VMP1, they are involved in the remodeling of the ER to form double-membrane vesicles (DMV) [38,52]. Scudellari compared the double-membrane spheres to bubbles being blown by the endoplasmic reticulum [20]. These DMVs may act as replication organelles (RO), which might provide a safe place for viral RNA to be replicated and translated, protecting it from innate immune sensors in the cell, similar to other β coronaviruses [56]. Therefore, these structures play a central role in infection, and, consequently, the loss of RO integrity due to the lack of VMP1 or TMEM41B could lead simultaneously to altered viral replication and enhanced antiviral signaling, as viral RNA is a very potent inducer of innate antiviral signaling [54,57]. However, the mechanism by which ER is transformed into these vesicles is still not fully elucidated. SARS-CoV-2 mediates its replication through a dependent ATG5 pathway using specific DMVs that can be considered similar to autophagosomes. Mutations in the NSP6 protein with a positive influence on autophagosome production suggest a potential link with autophagy [45]. Thus, we hypothesize that some of these DMVs could be related to autophagy structures, and, more specifically, to autophagosomes. We are certain that, in the near future, it will be found that well-described autophagy markers colocalize with these DMVs in SARS-CoV-2-infected cells. Another connecting pathway between the viral replication cycle and autophagy is represented by SNX27, one of the sorting nexin (SNX) family members, which down-regulates autophagy by increasing the level of mTORC1 signaling [58]. Figure 3, pathway 2 shows how SNX27 regulates the traffic of endosomal receptors towards recycling endosomes. Kim et al., found that mTORC1 acts as a signal integrator at the lysosome and can act as an inhibitor of later stages of autophagy, suppressing phosphorylation on UVRAG, which is a component of VPS 34 complex II. In this way, it avoids autophagosome and endosome maturation [59]. These events are relevant for the viral cycle, given that the virus enters the cell via directly fusing to the membranes in the cell surface pathway or via the endocytic pathway through endosome/lysosome-mediated endocytosis. Interestingly, it has recently been found that the ACE2 receptor possesses a type I PDZ binding motif (PBM) and can therefore interact with a PDZ domain-containing protein such as SNX27. A recent study has shown SNX27 to be critical for ACE2 cell surface regulation, and SNX27 prevents ACE2-bonded viral particles from entering the lysosome, down-regulating the endocytic viral entry pathway and therefore serving as a viral trafficking regulator [60]. Regarding autophagy machinery, a class III PI3-kinase that produces PI3P has a role in cellular trafficking and in the nucleation step in both canonical and non-canonical autophagy. In fact, inhibition of VPS34 kinase activity by VPS34-IN1, a well-known inhibitor for this kinase, reduced PI3P production and suppressed SARS-CoV-2 infection and replication in ex vivo human lung tissues [61] (Figure 3, pathway 5). The features mentioned above provide consistent evidence that the autophagy machinery is actively involved in the viral entry and replication process of SARS-CoV-2 infection and therefore could be used as potential therapeutic targets to battle infection and prevent viral entry and replication at different steps. In Figure 3, we show the stage modeling of how SARS-CoV-2 is related to autophagy structures and molecules (Figure 3). This section aims to describe the interplay between autophagy machinery proteins and newly described viral proteins and how this interaction affects viral replication and pathogenicity. Although strong evidence points toward the SARS-CoV-2 virus having an inhibiting role in some stages of autophagy, paradoxically, it has been suggested that the virus enhances autophagy in other steps of this process. Li et al., explored the regulatory role of the SARS-CoV-2 spike protein in infected cells and attempted to elucidate the molecular mechanism of SARS-CoV-2-induced inflammation. They found that SARS-CoV-2 inhibits the PI3K/AKT/mTOR pathway by upregulating intracellular reactive oxygen species (ROS) levels, and, in this way, promotes the autophagic response. Subsequently, SARS-CoV-2-induced autophagy triggers inflammatory responses and apoptosis in infected cells [62]. Recent studies suggest that SARS-CoV-2 inhibits autophagy at different stages, limiting the autophagic flux to suppress viral clearance by selective autophagy, known as virophagy, a process mediated by autophagy receptors that recognize and sequester viral components inside autophagosomes. To avoid virus inactivation, SARS-CoV-2 uses the autophagy machinery for its benefit [14]. Autophagy also regulates adaptive immunity through antigen presentation. Gassen and colleagues found that SARS-CoV-2 modulates cellular metabolism and reduces autophagy; therefore, the induction of autophagy limits SARS-CoV-2 propagation [16]. Autophagy machinery proteins and viral proteins interact, leading SARS-CoV-2 to successfully survive and complete the replication cycle in the infected host cells. Regarding SARS-CoV proteins, Mohamud and colleagues have shown that NSP3, one of the 16 nonstructural proteins, also known as papain-like protease (PLpro), can cleave the serine/threonine unc-51-like kinase (ULK1) and prevent the formation of the autophagy initiation complex in the absence of nutrients. In addition, PLpro showed deubiquitinase activity, which allows the virus to interrupt selective autophagy, preventing its proteins from being ubiquitinated [15]. Another recent study showed that different viral proteins targeted and inhibited autophagy to avoid viral clearance and to block the antiviral functions of autophagy [63]. SARS-CoV-2 uses autophagy to its benefit, hijacking the autophagy mechanism in the host cell to improve viral replication and to avoid the immune response and extracellular release. Viral proteins ORF3a and ORF7a were shown to cause the accumulation of autophagosomes [64]. Specifically, ORF3a interacts with autophagy-related protein UVRAG, suppressing autophagosome maturation and therefore the autophagy flux [53]. In two other studies, it was demonstrated that ORF3a interacts with VPS39, colocalizing with lysosomes. In this way, it impairs the binding of HOPS with RAB7, avoiding the regulation of the fusion of autophagosomes with lysosomes [65,66]. Another effect of viral protein ORF3a is its ability to promote lysosomal exocytosis, blocking autophagy flux and facilitating the lysosomal targeting of the BORC-ARL8b complex. Additionally, BORC-ARL8b is involved in lysosomal trafficking and modulates the exocytosis-related SNARE complex (VAMP7, STX4, and SNAP23). Following this pathway, the complex is oriented towards the plasma membrane area. This entire process is Ca 2+-dependent [67]. ORF7a generates a dysfunctional deacidified lysosome; therefore, autophagosomal degradation is interrupted and the virus can exit the host cell [64]. On the other hand, Koepke et al., used an mCherry-GFP-LC3B reporter system to show that lysosomal acidity, implicated in lysosomal degradation, was reduced in the presence of SARS-CoV-2. Regarding other viral protein effects, Nsp15 modulates autophagy regulation hypothetically by interfering with the mTOR pathway, in this way facilitating SARS-CoV-2 replication [64]. Non-structural protein NSP6 interacts with autophagy in different ways. It can join ER membranes, stimulating the rearrangement of its membranes and facilitating phagophore formation. This is another example of how SARS-CoV-2 uses the autophagy machinery to form DMVs to hide from the immune system and to replicate RNA. Moreover, it was found that viral protein NSP6 impairs autophagic flux, inhibiting autophagy at a late stage and impairing lysosomal acidification by targeting ATP6AP1, a vacuolar ATPase proton pump component. Consequently, the inflammasome is activated through NLRP3. To confirm that NSP6 elicits pyroptosis, Sun and colleagues experimentally overexpressed NSP6 and found that NLRP3/ASC Caspase-1-dependent activation released IL1β and IL18 in lung epithelial cells, thus being a crucial factor in viral pathogenicity [68]. It has been demonstrated that SARS-CoV-2 uses the mechanism of mitophagy, a well-known type of selective autophagy, as a strategy to regulate the host cell immune response. Hui and colleagues found that viral structural protein M joins with translation elongation factor (TUMF) M located in the mitochondrial external membrane and interacts with the LC3 II LIR domain; thus, the SARS-CoV-2 M protein breaks mitochondria networks by inducing mitophagy and then breaks downstream innate immunity signaling through inhibiting the type I IFN response [69]. Simultaneously, Li and colleagues demonstrated a similar effect of the ORF10 viral protein, which translocases to mitochondria and interacts with NIX—a protein very similar to the conforming protein of mitophagy receptor Nip3—and joins LC3 II. The activation of mitophagy leads to the degradation of mitochondrial antiviral signaling protein (MAVS), disrupting the activation of type I INF. This suppresses cellular pyroptosis and cytokine release, hijacking the immune response in favor of SARS-CoV-2 survival [70]. Additionally, SARS-CoV-2-infected cells are much less sensitive to lysis by cytotoxic T lymphocytes. This could be due to non-structural viral protein ORF8, which impairs antigen presentation with major histocompatibility complex I (MHCI) and leads MHCI to lysosomal degradation, mediated via autophagy. This mechanism also helps to evade the immune response [66]. ORF8 also mediates the escape from the immune system by degrading major histocompatibility complex (MHC) [71]. SARS-CoV-2 is a cytopathic virus, which means that it produces cell and tissue death as part of its replication cycle [72]. In order to restrain the infection, infected cells can undergo a type of inflammatory programmed cell death called pyroptosis [73,74]. Pyroptosis produces an inflammatory response led by the secretion of interleukin 1 beta (IL1β) and interleukin 18 (IL18) and the recruitment of immune cells such as neutrophils, eosinophils, and macrophages, causing an excessive immune response and potentially massive multiorgan failure [75]. There are two different types of pyroptosis: canonical pyroptosis, which involves inflammasomes triggered by pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs), and non-canonical pyroptosis, which is triggered by LPS and activates Caspase-1 [76]. Non-canonical pyroptosis results in Caspase-11 activation in mice, or Caspase-4 and Caspase-5 activation in humans, in response to lipopolysaccharide (LPS), a component of the Gram-negative bacterial cell wall, but does not cleave pro-IL-1β or pro-IL-18 [77,78]. This pathway results in cell swelling and lysis, preventing the replication of pathogens inside the infected cells. Canonical pyroptosis is induced by the formation of inflammasomes, which are large cytosolic multiprotein complexes assembled in response to infection and cellular stress. Inflammasomes are crucial for the activation of inflammatory caspases and the subsequent processing and release of pro-inflammatory mediators, such as interleukin-1β (IL-1β) and IL-18 [79]. Inflammasomes are assembled by pattern recognition receptors (PRR) such as NOD-like receptor protein 3 (NLRP3), which are sensors of exogenous and endogenous “danger” signals from PAMPs and DAMPs. Hyperinflammation caused by unrestrained inflammasome activation is linked with multiple inflammatory diseases, including inflammatory bowel disease, Alzheimer’s disease, and multiple sclerosis. It has been shown that patients with SARS-CoV-2 infection have high levels of IL1β because of the imbalance of autophagy and inflammasome formation caused by the virus. Autophagy responds to PRR from PAMPs and DAMPs to control homeostasis. Almost all PRR can induce autophagy, either directly or indirectly. Autophagy presents antigens to PRR and contributes to pathogen clearing. In addition, the induced autophagy forms a negative feedback regulation of PRR-mediated inflammation in a cell-/disease-specific manner to maintain homeostasis and prevent excessive inflammation. Understanding the interaction between PRR and autophagy in a specific disease can promote drug development for immunotherapy [80]. It has been proposed that autophagy modulates IL-1β production by means of inflammasome ubiquitination and the recruitment of p62 and LC3 [81,82]. Autophagy stimulates the clearance of inflammasomes, and inflammatory proteins related to pyroptosis [83]. In addition, some microorganisms, such as Streptococcus pneumoniae, increase ROS levels in the mitochondria of infected cells, causing mitochondrial damage and lower ATP production. In response to this damage, mitophagy is stimulated and consequently inactivates the NLRP3 inflammasome [84,85]. It was recently reported that protein E of SARS-CoV-2 activates the NLRP3 inflammasome [75,81,86]. Zhong et al., suggested that the parkin-dependent clearance of p62-bound mitochondria can reduce NLRP3 activation and IL-1β release in macrophages, providing another mechanism for the autophagy-mediated inhibition of inflammasome activation [87]. Later studies indicated that the regulation of inflammasome activation by autophagy can occur in multiple ways, through either the removal of endogenous inflammasome activators or removal of inflammasomes and their downstream cytokines directly. Saitoh et al., demonstrated that autophagy can negatively regulate pyroptosis. They found that when Atg16L1 is not present, it causes the hyperactivation of the inflammasome, with the excessive release of IL1β and IL18, and susceptibility to the development of intestinal inflammation was observed [88]. Autophagy also regulates non-canonical pyroptosis through the interaction of ATG8 with LPS and Caspase-11. The stimulation of autophagy increases ATG proteins and causes the inhibition of non-canonical pyroptosis [89]. Sun et al., reported pyroptosis induction in patients with COVID-19. They found that NSP6 overexpression in SARS-CoV-2 infection may affect lysosome acidification in lung epithelial cells by interacting with vacuolar ATPase, leading to the blockade of autophagic flux and inducing NLRP3 inflammasome activation and pyroptosis. When autophagy flux is pharmacologically restored, NSP6-induced pyroptosis is suppressed [90]. In contrast, there is some evidence that autophagy could promote pyroptosis. Dupont and colleagues have demonstrated that mice macrophages experience increased release of interleukins under starvation [91]. Finally, autophagy and pyroptosis could have a synergic effect in clearing microorganisms such as SARS-CoV-2. Xenophagy, the selective type of autophagy that degrades pathogens, may also be involved in SARS-CoV-2 infection. Moreover, when xenophagy is surpassed by pathogens, Caspase-1 might be cleaved and pyroptosis triggered. As described above, SARS-CoV-2 infection in the lung recruits immune cells and causes the overproduction of pro-inflammatory cytokines such as IL1β, IL2, IL10, and TNF. The imbalance in cytokine production causes a cytokine storm, leading to multiorgan damage, affecting the liver, heart, and kidneys [92]. Huang, in his prospective study, was the first to include an analysis of cytokine levels in severe and mild COVID-19, showing the presence of a cytokine storm analogous to that found for SARS-CoV-2 infection [93]. Excessive cytokine production could lead to multiorgan failure and worsen a patient’s prognosis. Moreover, the damage to epithelial and endothelial lung cells in SARS-CoV-2 infection, plus the cytokine storm, initiates the coagulation cascade via the tissular factor pathway [80]. Specifically, the excessive immune response characterized by an increase in IL1β, IL2, IL6, GSF, TNFα, and IFNɤ generates a local and systemic inflammatory response [94]. It is believed that these two inflammatory responses can cause hypercoagulability by increasing procoagulant molecules such as Von Willebrand factor, tissular factor, fibrinogen, and thrombin, leading to a reduction in blood flow and macro- and micro-thrombosis. In addition, there is an endotheliopathy caused by the vasoconstriction that also enhances the prothrombotic state. Masi and colleagues found that acute respiratory distress syndrome in COVID-19 was associated with procoagulants, specifically with plasminogen activator factor. Furthermore, they suggested that there is a potential role of endothelial dysfunction in the imbalance between procoagulant and anticoagulant agents. This procoagulant state can cause pulmonary embolism, acute respiratory distress syndrome (ARDS), and multiorgan failure, as is observed in patients with severe SARS-CoV-2 infection [95]. Additionally, the accumulation and infiltration of lymphocytes in the lungs leads to lymphopenia and neutrophilia, which leads to the formation of neutrophil extracellular traps (NETs). This process is called “NETosis”, which is a newly described type of programmed cell death that involves neutrophils as the key players in this mechanism, generating NETs by the extrusion of DNA, histones, and antimicrobial proteins, which are important for preventing pathogen infection. However, if NETs are in excess, a series of negative effects, such as autoimmune inflammation and tissue damage, could occur [96,97]. Interestingly, chronic metabolic diseases such as type 2 diabetes are also associated with high levels of NET production. Diabetes is characterized by inflammation, endothelial dysfunction, a risk of infection, and cardiovascular disease. NETosis is observed in patients with type 2 diabetes [98] and in SARS-CoV-2-infected individuals. Although it is still unclear whether this increased NETosis in type 2 diabetes patients is associated with the elevated incidence of thromboembolic events seen under SARS-CoV-2 infection, it has been postulated that NET overproduction may explain part of this increased risk. Similarly, non-diabetic obese patients have an increased incidence of thromboembolic events, as well as increased NETosis, and obesity is a risk factor for poor outcomes in COVID-19 patients. Moreover, it has been demonstrated that the increased level of angiotensin II seen in patients with hypertension could trigger NETosis, increasing the cardiovascular risk in these patients [99,100]. It has been found that the presence of plasma NETs correlates with high sequential organ failure assessment (SOFA) scores in COVID-19. Interestingly, there is an increase in plasma NETs in patients with SARS-CoV-2 and this correlates with the severity of the clinical presentation [101]. NETs mediate harmful effects caused by neutrophils, leading to unfavorable coagulopathy and immune thrombosis, and are a major element of micro- and macrovascular thrombi [102,103]. There is an intricate and complex relationship among SARS-CoV-2 infection, NETs, and cytokine storms, which is not entirely understood at present. When SARS-CoV-2 enters the cells, pyroptosis is activated, causing an increase in cytokines released by endothelial and epithelial lung cells, which recruit and activate neutrophils, triggering NETosis [104], which increases the severe effects of SARS-CoV-2 infection [105]. It is important to highlight that autophagy participates in NET formation and regulation. An increased concentration of IL8 due to SARS-CoV-2 infection induces autophagy and then autophagy triggers NETosis in neutrophils [106]. In addition, secretory autophagy plays an important role in the release of IL1β, enhancing the activation and recruitment of neutrophils [91]. Nevertheless, autophagy regulates the externalization of membrane-bound and cytosolic proteins, modulating the NET vacuolation process. Autophagy may also have a role in limiting the respiratory burst, preventing cytoskeletal dynamics and chromatin decondensation, and generating histone citrullination [68]. Additionally, Remijsen et al., detected defective intracellular chromatin decondensation when they suppressed neutrophil autophagy, elucidating that autophagy is essential for the first stage of NETosis (DNA decondensation), rather than the cell killing process itself [107]. It has been suggested that the PI3K-AKT-mTOR axis links autophagy with NET formation and has a significant impact on both processes. The serine/threonine kinase mTOR controls cellular stress, proliferation, and autophagy. On the other hand, mTOR down-regulates autophagy when activated by dephosphorylation. However, Rapamycin and WYE-354 are autophagy inducers that inhibit mTOR and enhance NET formation in human neutrophils via autophagy downstream of formyl peptide receptor (FPR) signaling [68]. In contrast with the previous evidence, it has also been found that autophagy modulates NETosis to prevent an excessive immune response. For instance, as was discussed above, the viral proteins M and ORF10 could interfere in the development of NET formation by stimulating mitophagy and inhibiting the release of IL1β. Kim and colleagues have discovered an increase in autophagy activity and NET formation in hemodialysis patients. Interestingly, when autophagy was blocked, neutrophil activity and NET release rose significantly, demonstrating that autophagy may also play a role in limiting excess NET production [59]. To summarize, NETosis is a beneficial immune process triggered by inflammation and infection to protect the host organism and cells against pathogens such as SARS-CoV-2. However, the overproduction of NETs could cause an immune thrombotic state that exacerbates the systemic inflammation caused by severe COVID-19. Considering that autophagy can prevent or limit NET production, the activation of autophagy may play an essential role in modulating systemic inflammation. Taken together, it is important to highlight the relationship between the immune system, inflammation, and autophagy, to treat and improve the outcomes of patients with SARS-CoV-2 infection. The interplay between local and systemic inflammatory responses, SARS-CoV-2, and autophagy is summarized in Figure 4. Post-COVID-19 syndrome is described as symptoms that are persistent for many weeks or months after the acute disease has resolved [108,109,110]. The World Health Organization (WHO) defined it in October 2021 as “Illness that occurs in people who have a history of probable or confirmed SARS-CoV-2 infection; usually within three months from the onset of COVID-19, with symptoms and effects that last for at least two months” [67]. The pathogenetic mechanism of post-COVID-19 pulmonary fibrosis is currently a topic of intense research interest but is still largely unexplored (recently reviewed by [111]). Although they are still poorly understood, the immunological dysregulations are probably associated with post-COVID-19 syndrome. In particular, patients with a prolonged symptom duration maintained antigen-specific T-cell response magnitudes to the virus in CD4+ and increased T follicular helper cell (Tfh) populations throughout late convalescence, while those experiencing a full recovery demonstrated a decline in these cellular populations [112,113,114,115]. The symptoms can vary among people experiencing post-COVID-19 syndrome, but there are general symptoms that are the most prevalent. The five most common symptoms are fatigue (58%), headache (44%), attention disorder (27%), hair loss (25%), and dyspnea (24%) [116]. Respiratory and cardiovascular symptoms include cough, dyspnea, and chest pain, among others, and neuropsychiatric symptoms range from difficulty thinking, sleep problems, and changes in smell or taste, to symptoms of clinical depression. There are certain specific patient populations that present a higher risk of developing this condition. These include patients with underlying autoimmune diseases, patients who have had severe COVID-19, people who needed intensive care, and unvaccinated patients, among others [116,117,118]. The pathophysiological mechanisms that underlie post-COVID-19 syndrome include endothelial damage, direct viral toxicity, a pro-inflammatory and prothrombotic state, and immune system dysregulation. Apparently, severe COVID-19 infection survivors have an increased risk of presenting bacterial, fungal, and viral infections thereafter [111,112,114,115]. As described above, autophagy regulates the immune response and NETosis in SARS-CoV-2 infection, and its dysregulation by the virus itself can worsen the clinical outcomes of patients. It is believed that long COVID syndrome is more common in patients who have had severe symptoms and who have required long-term hospitalization because they have had a prolonged inflammatory response [117]. Taken together, it could be hypothesized that autophagic dysregulation could increase the risk of developing long COVID or post-COVID-19 syndrome. Thus, maintaining autophagy within normal functioning could aid in preventing long-term sequelae by controlling the inflammatory and immune response, and therefore diminishing post-COVID syndrome. This needs to be elucidated in future investigations. Autophagy would be involved in the entry, as well as transcription and translation, of viral particles in a host cell [119]. Additionally, autophagy could be involved in the systemic inflammatory response and post-COVID-19 syndrome. Selective autophagy, especially mitophagy, was reported to be induced by SARS-CoV-2 proteins, modulating the inflammatory response. Secretory autophagy may also be involved in the development of the thrombotic immune-inflammatory syndrome seen in a significant number of COVID-19 patients that leads to severe illness and even death. However, autophagy does not only interact with SARS-CoV-2 infection by participating in its viral replication cycle. Studies suggest that, surprisingly, reciprocal dysregulation of autophagy by the viral infection itself could be one of the mechanisms of viral survival and tissue damage, given the antimicrobial functions of autophagy, its ability to aid with viral clearance through xenophagy, and its immunological role in battling infection and regulating excessive inflammation [120]. Autophagy’s role as a balancer of the beneficial and detrimental effects of immunity and inflammation becomes disrupted by viral effects on autophagy’s complex machinery [120,121]. Dysregulation of autophagy could imply the disinhibition of pyroptosis, excess NETosis, and other molecular processes, stimulating the release of pro-inflammatory cytokines and interleukins that favor an exaggerated inflammatory response overall, leading to a thrombotic immunoinflammatory state that correlates with more severe clinical illness in COVID-19. In fact, one of the described cargoes of unconventional autophagy-associated secretion pathways is the export of cytosolic protein IL1β, a pro-inflammatory cytokine, which has a central role in inducing pro-inflammatory signaling [122]. A possible link between this process and the cytokine storm that characterizes the immunoinflammatory state seen in patients with severe COVID-19 could be further explored. Consequently, identifying different autophagic biomarkers could help to correlate with the severity of illness, and thus serve as a biological marker for the prognosis of the disease. Many viruses with induced direct and indirect mechanisms explaining most of the short-term complications of the disease have correlations with alterations in autophagy. Long-term post-COVID-19 syndrome may also be related to dysfunctional autophagy. Associations between interindividual markers of short- and long-term prognosis and dysfunctional autophagy offer many gaps for further investigation. More research is needed to clarify the involvement of these abnormalities in disease infection and clinical evolution.

PMC10002779

Michio Yamashita,Junko Shibato,Randeep Rakwal,Naoko Nonaka,Takahiro Hirabayashi,Brian J. Harvey,Seiji Shioda,Fumiko Takenoya

Molecular and Physiological Functions of PACAP in Sweat Secretion

26-02-2023

sweat gland,eccrine gland,PACAP,aquaporin5 (AQP5),DNA microarray

Sweat plays a critical role in human body, including thermoregulation and the maintenance of the skin environment and health. Hyperhidrosis and anhidrosis are caused by abnormalities in sweat secretion, resulting in severe skin conditions (pruritus and erythema). Bioactive peptide and pituitary adenylate cyclase-activating polypeptide (PACAP) was isolated and identified to activate adenylate cyclase in pituitary cells. Recently, it was reported that PACAP increases sweat secretion via PAC1R in mice and promotes the translocation of AQP5 to the cell membrane through increasing intracellular [Ca2+] via PAC1R in NCL-SG3 cells. However, intracellular signaling mechanisms by PACAP are poorly clarified. Here, we used PAC1R knockout (KO) mice and wild-type (WT) mice to observe changes in AQP5 localization and gene expression in sweat glands by PACAP treatment. Immunohistochemistry revealed that PACAP promoted the translocation of AQP5 to the lumen side in the eccrine gland via PAC1R. Furthermore, PACAP up-regulated the expression of genes (Ptgs2, Kcnn2, Cacna1s) involved in sweat secretion in WT mice. Moreover, PACAP treatment was found to down-regulate the Chrna1 gene expression in PAC1R KO mice. These genes were found to be involved in multiple pathways related to sweating. Our data provide a solid basis for future research initiatives in order to develop new therapies to treat sweating disorders.

Molecular and Physiological Functions of PACAP in Sweat Secretion Sweat plays a critical role in human body, including thermoregulation and the maintenance of the skin environment and health. Hyperhidrosis and anhidrosis are caused by abnormalities in sweat secretion, resulting in severe skin conditions (pruritus and erythema). Bioactive peptide and pituitary adenylate cyclase-activating polypeptide (PACAP) was isolated and identified to activate adenylate cyclase in pituitary cells. Recently, it was reported that PACAP increases sweat secretion via PAC1R in mice and promotes the translocation of AQP5 to the cell membrane through increasing intracellular [Ca2+] via PAC1R in NCL-SG3 cells. However, intracellular signaling mechanisms by PACAP are poorly clarified. Here, we used PAC1R knockout (KO) mice and wild-type (WT) mice to observe changes in AQP5 localization and gene expression in sweat glands by PACAP treatment. Immunohistochemistry revealed that PACAP promoted the translocation of AQP5 to the lumen side in the eccrine gland via PAC1R. Furthermore, PACAP up-regulated the expression of genes (Ptgs2, Kcnn2, Cacna1s) involved in sweat secretion in WT mice. Moreover, PACAP treatment was found to down-regulate the Chrna1 gene expression in PAC1R KO mice. These genes were found to be involved in multiple pathways related to sweating. Our data provide a solid basis for future research initiatives in order to develop new therapies to treat sweating disorders. Sweat is an exocrine fluid secreted by the skin and is an essential body fluid for thermoregulation and skin water regulation [1]. It is secreted from the sweat gland of the eccrine gland and apocrine gland, and 90% of the sweat secretory function is performed by the eccrine gland, of which 99% is water and the rest is inorganic. Sweat secretion is promoted by an increased body temperature and stress, etc., and contributes to the maintenance of homeostasis in the organism. It is known that abnormalities in sweat secretion can lead to dyshidrosis, such as hyperhidrosis, which causes excessive sweating, and anhidrosis, which prevents sweat secretion. It is known that dyshidrosis can result in not only a high fever and heat stroke but also causes serious skin conditions such as pruritus and erythema, which markedly reduce the quality of life of patients [2,3]. The cause of these disorders is thought to be one of the causes of nervous system diseases and other underlying diseases, such as autonomic dysrhythmia, but its basis is yet to be clarified [1]. Therapies such as using an anticholinergic drug and surgical treatment are available for the treatment of hyperhidrosis; however, no well-defined therapy exists for anhidrosis [4,5,6,7,8]. Previous studies have reported the involvement of neurotransmitters and peptides, such as acetyl choline (ACh), noradrenaline (NA) and non-noradrenergic transmitter (NANC), and atrial natriuretic peptide, in the regulation of sweat secretion in sweat glands [9]. It is known that ACh and NA, which are the typical neurotransmitters involved in sweat secretory action, act on the G protein-coupled receptor of the Gs and Gq system expressed in sweat gland cells and increase Ca2+ and cyclic adenosine monophosphate (cAMP) in cells. Increased levels of intracellular Ca2+ and cAMP are thought to promote water secretion by translocating aquaporin5 (AQP5), the main water channels in sweat glands, from the cytosol to the plasma membrane [10]. Pituitary adenylate cyclase-activating polypeptide (PACAP) is a bioactive peptide isolated and identified to potently activate the adenylate cyclase in pituitary cells from the hypothalamus of the sheep brain [11,12]. PACAP is composed of 27 or 38 amino acid residues, with PACAP27 and PACAP38 almost showing the same action to increase cytosolic cAMP levels. PACAP belongs to the vasoactive intestinal peptide (VIP)/secretin/glucagon family. PACAP and VIP share three GPCR receptors: the PAC1 receptor (PAC1R) and VPAC1R and VPAC2R. In addition, various physiological roles of PACAP are becoming clear from the analysis of the tissue distribution of PACAP and its receptors and in vitro and in vivo physiological and pharmacological studies. PACAP is considered to act as a neurotransmitter/modulator and in addition to its hormonal action, PACAP has been reported to have many physiological effects, including the promotion of glucose-dependent insulin secretion in pancreatic islet B cells, the regulation of pain inhibition, immune suppression, protection against ischemic neuronal cell death, and nerve regeneration [13,14,15,16,17,18]. In recent years, research has also shown that PACAP can stimulate the secretion of exocrine glands in tissues such as the lacrimal gland, sweat gland, and salivary gland [19]. It was reported that PACAP gene knockout (KO) mice developed corneal damage similar to the dry eye [20]. In addition, the function of PACAP in promoting tear secretion in the lacrimal gland has been revealed, and its mechanism of action has been shown to involve the activation of the cAMP/protein kinase A (PKA) pathway by PACAP through its action on PAC1R in the lacrimal gland, leading to the phosphorylation of AQP5 and promoting the translocation of AQP5 from the cytosol to the cell membrane [20]. In sweat glands, immunohistochemical analysis using human and mouse skin tissue has shown that PACAP is present in nerve endings near eccrine glands and that PAC1R is expressed in the acinar cells of eccrine gland secretory cells [21]. The addition of PACAP to the plantar surface of the mice’s forepaws has been reported to increase sweat secretion in mice, and this effect is inhibited by an inhibitor of PAC1R [21]. In addition, it is reported that PACAP has an ability to increase the cytosolic Ca2+ concentration through PAC1R and to promote the transition of AQP5 to the cell membrane in NCL-SG3 cells, which are immortalized human eccrine gland cells [22]. However, the intracellular signaling mechanisms by PACAP are poorly defined [9]. Therefore, in this study, we used PAC1R gene KO mice (PAC1R KO) and wild-type (WT) mice to observe changes in AQP5 localization and gene expressions (DNA microarray, GSE223124) in sweat gland tissues by PACAP in order to identify the cytosolic cell signaling mechanism. The distribution of PAC1R immunoreactivity was observed by using fluorescent light microscopy. In the WT mouse, PAC1R was localized in the acinus of the eccrine gland (Figure 1A,B). In the PAC1R KO mouse, PAC1R was not observed in the eccrine gland (Figure 1C,D). We used confocal laser microscopy to construct a 3D model of mouse eccrine gland tissue and observed the intracellular localization of AQP5-like immunoreactivity (LI) vesicles using immunostaining. As a result, the presence of AQP5 positive vesicles, about 50 nm in diameter, both on the cell membrane and in the cytoplasm of mouse exocrine gland acinar cells, were found (Figure 2). In previous studies, it was reported that AQP5-containing vesicles transiently migrate to the cell membrane in the eccrine gland following PACAP treatment [21]. Therefore, in this study, the intracellular localization of AQP5-LI was observed using immunostaining at 30 and 60 min after an intradermal injection of PACAP to the WT mice and PAC1R KO mice. In the WT mice group 30 min after PACAP treatment, the localization of AQP5-LI in the acinus was strongly observed in the apical cell membrane over the vehicle-treated WT mice (Figure 3A,B). However, after 60 min of PACAP treatment, the localization of AQP5-LI was not observed as intensely in the acinus (Figure 3C,D). In addition, in the PAC1R KO mice group, the transition of AQP5 to the lumen of the acinus was weakly observed by PACAP administration (Figure 3E,H). Based on previous results, it has been suggested that the activation of PAC1R by PACAP leads to the translocation of AQP5 from the cytoplasm to the apical side of the secretory cells of the mouse eccrine gland [22]. Therefore, in order to clarify the mechanism by which PACAP promotes sweating, we observed gene expression changes in WT mice and PAC1R KO mice after an intradermal injection of PACAP in the skin of the forefeet for 30 and 60 min using DNA microarray analysis. The results revealed changes in the expression of multiple genes at 30 min after PACAP treatment, of which 70 genes were up-regulated and 45 genes were down-regulated in the WT mice, while 190 up-regulated genes and 43 down-regulated genes were found in the PAC1R KO mice (Figure 4). Additionally, 197 genes were up-regulated and 76 genes were down-regulated in the WT mice, while 170 genes were up-regulated and 39 genes were down-regulated in the PAC1R KO mice at 60 min after PACAP treatment (Figure 4). Several genes involved in sweat secretion showed changes in their expression levels: in WT mice, three genes involved in sweat secretion, Ptgs2, Kcnn2, and Cacna1s, were upregulated at 30 min after PACAP treatment; in PAC1R KO mice, the gene whose expression was decreased was Chrna1s at 30 min after PACAP treatment (Table 1). Figure 5 shows the top 20 genes with expression changes in WT and PAC1R KO mice at 30 and 60 min after PACAP administration, showing those with expression changes of 2.0-fold or more and 0.5-fold or less, respectively. Comparing WT and KO, genes involved in antimicrobial activity (yellow) increased more in KO, and 13 of the top 20 genes were related to antimicrobial activity, especially at 60 min post-treatment. Genes related to muscle contraction (gray) were increased in WT, whereas they were decreased in KO. Trub2, which was decreased at 30 min in the KO, is an oxidative phosphorylation-related gene (green) that is responsible for ATP biosynthesis in the mitochondrial inner membrane electron transport system in a phosphorylation reaction that is coupled to the oxidation of NADH. Many Trub2 gene probes have been detected, and the change in the decreased Trub2 expression seems significant. Other cytokines such as Il6, Cxcl9, and Cxcl5 were also frequently expressed but were confirmed in both the WT and KO. Previous in vivo studies reported that local sweat secretion promotion through PAC1R is elicited by administering PACAP into the subcutaneous tissue of the mouse footpad under anesthesia [21]. In addition, it has been reported that PACAP-related peptide VIP promotes sweat secretion through the cAMP pathway [23]. In the in vitro studies using human eccrine gland immortalized cells (NCL-SG3 cells), we have reported that PACAP plays important roles in the translocation of AQP5 from cytoplasm to the cell membrane by increasing the intracellular Ca2+ concentration by PAC1R [22]. However, there are less data showing genome-wide changes in sweat glands after the administration of PACAP. In this study, we first observed changes in the intracellular localization of AQP5 followed by the gene expression in the sweat glands after an intradermal injection of PACAP. At first, we confirmed that the location of PAC1R was in the eccrine gland of the WT mouse. Simultaneously, in the PAC1R KO mouse, PAC1Rs were not observed in the eccrine gland. Using confocal laser scanning microscopy, we identified many AQP5-positive vesicles in the cytoplasm of eccrine gland secretory cells by analyzing the mouse eccrine gland in three dimensions. AQP5-LI in the eccrine gland was observed in small vesicles in the cytoplasm of secretory cells [24]. AQP5 has been reported to promote water secretion by transitioning to the cell membrane in response to external stimuli [25]. After an intradermal injection of PACAP, the WT mice had more AQP5 localized on the luminal side of the eccrine gland compared to the control group. However, after 60 min of PACAP administration, the localization of AQP5-LI vesicles on the luminal side decreased and its immunoreactivity increased in the cytoplasm. In the group of PAC1R KO mice administered with PACAP, a weak AQP5-LI was found on the cell membrane in the acinus. Therefore, it is suggested that in the mouse eccrine gland, PACAP promotes the transition of AQP5 from cytoplasm to the cell membrane acting through PAC1R. Figure 6 describes the gene categories for their involvement in the PACAP-induced membrane localization of AQP5. The expression levels of four genes (Ptgs2, Kcnn2, Cacna1s, and Chrna1) known to be involved in sweat secretion were found to be elevated in WT mice at 30 min. One of these genes, prostaglandin endoperoxide synthase 2 (Ptgs2/COX-2), has been suggested to be involved in sweat secretion rather than vasodilation through studies in which COX inhibitors and selective COX2 inhibitors were used in human subjects subjected to exercise. The mechanism of sweating mediated by COX is thought to work though the interaction with nitric oxide synthase (NOS), and it is suggested that these interactions may activate Cl- channels and/or the Na+/K+-ATPase enzyme [26,27,28,29]. Kcnn2 encodes the small conductance calcium-activated potassium channel 2 (SK); the SK channel activation is voltage independent and depends on intracellular calcium levels [30]. Kcnn4, which belongs to the same potassium–calcium-activated channel family, is one of the main regulatory factors of eccrine glands, and its expression has been reported to co-localize with Foxa1 and eccrine gland secretory cells, and Kcnk5, a potassium voltage-gated channel, double knockout mice were found to have significantly reduced sweating compared to WT mice [27]. In particular, Cacna1s is a calcium voltage-gated channel and previous studies have reported that the inhibition of this channel leads to the suppression of sweat secretion [31]. Changes in the intracellular calcium concentrations within the eccrine glands have been suggested to be involved in a calcium influx from both intracellular and extracellular sources. Furthermore, an increase in the intracellular calcium concentration has been shown to promote sweating through the action of calcium-dependent chloride channels (CaCCs) and other factors [32,33]. Previous research has also demonstrated that PACAP can increase intracellular calcium concentrations in eccrine gland cells and promote the translocation of AQP5 to the cell membrane [22,34]. The decreased expression of the cholinergic receptor nicotinic α1 subunit (Chrna1), which plays an important role in sweating in eccrine glands, was observed in PAC1R KO mice in this study. CHRNA1 has been reported as a causative factor in primary focal hyperhidrosis and, similarly to CACNA1, it is strongly involved in the mechanism of eccrine sweat production in mice injected with the siRNA Chrna1 gene [31]. Aquaporins function as inflammatory mediators in some lesions. The dysfunction of aquaporins is involved in the development of inflammatory skin diseases characterized by the disruption of the skin barrier [35,36,37]. Genes involved in inflammation, such as neutrophil degranulation and antibacterial activity, were increased in KO mice compared to WT mice. In particular, S100a8 and S100a9 have been shown to be increased in inflammatory skin diseases [38,39,40], suggesting that the PAC1R KO mice have skin inflammation due to defective aquaporin action. This is likely because the genes KRT17 (2.1-fold change), KRT16, and KRT6A (2.3-fold change), whose expression increased at 60 min in KO mice, are upregulated during skin damage and inflammation. Recent studies have recognized that keratinocytes (KCs) located at the skin surface exposed to external stimuli, such as pathogens and/or injury, are activates that in turn secrete an array of alarmin molecules, and which can be considered a rapid and specific innate immune response [41]. Myosin light chain phosphorylation in the lung plays an important role in cell contraction, extracellular permeability, and lung water homeostasis, and the inhibition of myosin light chain kinase leads to a decreased AQP5 expression. Contraction induced by myosin light chain phosphorylation can lead to tight junction opening because the cell junction complex is bound to actin and non-muscle myosin in the cytoskeleton [42]. Furthermore, the cytosolic Ca2+ concentration has been shown to upregulate the AQP5 expression by regulating it via motor proteins such as myosin and dynein, resulting in AQP5 plasma membrane redistribution. Tmod1, whose expression was decreased in KO mice, has been reported to be involved in the regulation of the body water balance [43]. In the kidney, actin and actin-related proteins are involved in the regulation of water and salt homeostasis. Several channels interact with actin, including aquaporin 2 (AQP2), the cystic fibrosis transmembrane regulator (CFTR), and the epithelial sodium channel (ENaC). Increases in intracellular calcium and ATP are involved in the translocation of AQP5 [44,45]. Panx3 is a member of the pannexin family found in skin and is present at the plasma membrane and endoplasmic reticulum membrane and functions as a channel to move ATP and Ca2+ into and out of the cell. Pannexin 3 channels are also involved in skin homeostasis and wound healing [46,47]. Trub2, whose expression was decreased in KO mice, is involved in oxidative phosphorylation, and mitochondrial oxidative phosphorylation is a major pathway for ATP generation [48]. Disseminated intravascular coagulation syndrome (DIC) is an acquired syndrome characterized by the extensive activation of coagulation, leading to the dysfunction of multiple organs in the body [49]. Coagulation reactions have also been identified in Sjögren’s syndrome [50]. Although there are no reports of an association with AQP5, TBC1D1 (2.5-fold change), which is involved in an enhanced glucose uptake via glucose transporter 4 (GLUT4) translocation, was identified at 30 min in the WT mice. TBC1D1 is involved in the regulation of glucose processing and the substrate metabolism within the skeletal muscle and is essential for the insulin secretory function. An important role for TBC1D1 in the translocation of GLUT4 to the skeletal muscle protoplasm membrane via exercise, and contraction has been reported [51,52]. Recently, it has been reported that TBC1D1 is potentially associated with the etiology of atopic dermatitis in dogs, a chronic inflammatory and pruritic skin disease [53]. Therefore, we are interested in whether TBC1D1 is involved in the translocation of AQP5 by PACAP. Overall, from this study, it is suggested that the sweat-promoting effects of PACAP mediated by PAC1R are due to its ability to promote the translocation of AQP5 to the luminal side of the eccrine gland. Additionally, various pathways via PAC1R were identified in mouse skin, including those involved in sweat secretion that was affected by an intradermal injection of PACAP. In the future, it will be necessary to deeply investigate how these pathways contribute to sweat secretion by using the inhibitors of these pathways and the eccrine cells obtained from the study. All experimental procedures involving animals were approved by the Institutional Animal Care and Use Committee of Hoshi University. The male Adcyap1r1 -/- mice (PAC1R KO) and the male C57BL/6 (WT) were bred and maintained under specific pathogen-free conditions in the animal facility of Hoshi University. Animals were housed with a 12 h/12 h light/dark cycle and were provided free access to water and standard rodent chow. Nine- to ten-week-old mice were used for the experiments. Adult C57BL/6 mice (Tokyo Laboratory Animals Science Co., Ltd., Tokyo, Japan) and the PAC1R KO mice were anaesthetized with three types of mixed anesthetic agents (5 mL kg−1, i.p.). These anesthetic agents were midazolam (0.3 mg kg−1, Maruishi Seiyaku, Osaka, Japan), medetomidine (4 mg kg−1, Kyoritsu Seiyaku, Tokyo, Japan), and butorphanol (5 mg kg−1, Meiji Seika, Tokyo, Japan). In total, 5 µL of the vehicle (saline containing 0.1% BSA) containing 10−7 mol/L of PACAP and 5 µL of the vehicle alone were intradermal injected into the center of the footpad with a 26G needle attached to a 10 µL glass syringe (HAMILTON, Reno, NV, USA) [54]. The mice were euthanized 30 and 60 min after the administration of PACAP by spinal dislocation and the skin from the forepaws was harvested. The harvested samples were fixed overnight at 4 °C in a 4% paraformaldehyde solution in 0.1 M of phosphate buffer (PB). The fixed tissues were processed in 20% sucrose solution overnight and 30% sucrose solution for 48 h, followed by embedding. Cryosections (10 μm thickness) were cut from the frozen tissue using a cryostat and used for immunostaining. The cryosections were washed with phosphate-buffered saline (PBS) and then blocked with 5% horse serum in PBS for 1 h and incubated overnight at 4 °C in a solution of the following primary antibody: rabbit anti-PAC1R antibody (1:200, developed by our laboratory) [55]. It was then washed with PBS and reacted with a secondary antibody: Alexa Flour 594 Donkey anti-rabbit IgG (1:800, Invitrogen, MA, USA) for 60 min at room temperature. It was re-washed with PBS, followed by nuclear staining with DAPI (1:10,000, Invitrogen) for 3 min at room temperature; then, it was washed with PBS and encapsulated using a VECTASHIELD Vibrance Antifade Mounting Medium (Vector Laboratories, Newark, CA, USA). After drying, the samples were visualized and imaged on a fluorescent microscope (BZ-X710; Keyence, Osaka, Japan). The cryosections were washed with PBS and then blocked with 5% horse serum in PBS for 1 h and incubated overnight at 4 °C in a solution of the following primary antibody: rabbit anti-AQP5 antibody (1:200, Merck Millipore, Billerica, MA, USA). It was then washed with PBS and reacted with a secondary antibody, Alexa Flour 488 Donkey anti-rabbit IgG (1:800, Invitrogen), for 60 min at room temperature. It was re-washed with PBS followed by nuclear staining with DAPI (1:10,000, Invitrogen) for 3 min at room temperature; then, it was washed with PBS and encapsulated using a VECTASHIELD Vibrance Antifade Mounting Medium (Vector Laboratories). After drying, the stained sections were photographed cross-sectionally using a confocal laser microscope A1R/Ti-2E (Nikon, Tokyo, Japan). After an intradermal injection of PACAP to six WT mice and PAC1R KO mice each, the skin of the forepaws was sampled at 30 and 60 min later, which were immersed immediately post-dissection in liquid nitrogen and thereafter transferred to an −80 °C deep freezer. The samples were individually ground in liquid nitrogen to prepare very fine powders for subsequent downstream gene expression analyses [56,57]. Prior to each DNA microarray analysis, the total RNA was extracted using an optimized protocol using the QIAGEN RNeasy Mini Kit (QIAGEN, Germantown, MD, USA). Optimization was done based on our expertise in extracting total RNA from multiple samples each requiring an individual trial experiment/s to optimize the protocol for each of the organisms, tissues, and cells in aspects of the amount of tissue and time of grinding, the amount of phenol (including preparation of the reagents), the use of a single or double chloroform extraction, ethanol washes, etc., and finally confirming both the quality and quantity of total RNA, and, as also mentioned below, using RT-PCR to check the gene expressions as a positive control. The quantity and quality were measured spectrophotometrically with DeNovix DS-11 (DeNovix, Wilmington, DE, USA) and re-confirmed using formaldehyde-agarose gel electrophoresis [56,57]. Briefly, the obtained total RNA quality measurements revealed a good quality, as demonstrated by the A260/280 values > 1.8, A260/230 > 1.8; in our experiments, we always aim for a higher value of more than 2.0 to 2.3. Furthermore, the total RNA was obtained in an optimum quantity of >300 ng/μL. Following the visualization for confirming the pre-experimental integrity of the total RNA subunits by gel electrophoresis, an additional step comprised of cDNA synthesis was used for examining the quality of the synthesized cDNA by determining the expression of a commonly used housekeeping gene, glyceraldehyde 3-phosphate dehydrogenase (GAPDH). This was done by RT-PCR (AffinityScript QPCR cDNA Synthesis Kit; Strategene, Agilent Technologies, Santa Clara, CA, USA, and an Emerald Amp PCR Master Mix; TaKaRa, Shiga, Japan) on a Thermal Cycler (Applied Biosystems, Tokyo, Japan). Moreover, only gene-specific primers were used, whose sequences are shown in the articles. Post-electrophoresis, the PCR products were visualized and quantified using a ChemiDoc XRS + imaging system (Bio-Rad Laboratories, Hercules, CA, USA). All the steps followed in the sequence provided confidence that the best quality total RNA (sample) was used in the DNA microarray chip. As a last step, the Agilent mouse, whole genome was 4 × 44 K (G4122F) DNA slide (composed of 4 chips on one slide), was used for the microarray analysis, which was performed according to the manufacturer’s instructions (Agilent Technologies, Santa Clara, CA, USA) and detailed in our publications [56,57]. Briefly, the total RNA (400 ng) was labeled with either Cy3 or Cy5 dye using an Agilent Low RNA Input Fluorescent Linear Amplification Kit. Fluorescently labeled targets of the control as well as the treated samples were hybridized to the same microarray slide with 60 mer probes. A flip labeling (dye-swap or reverse labeling with Cy3 and Cy5 dyes) procedure was followed to nullify the dye bias associated with an unequal incorporation of the two Cy dyes into cDNA. Hybridization and wash processes were performed according to the manufacturer’s instructions, and hybridized microarrays were scanned using an Agilent Microarray scanner G2505C. For the detection of significantly differentially expressed genes between the control and treated samples, each slide image was processed by Agilent Feature Extraction software (version 9.5.3.1). This program measures the Cy3 and Cy5 signal intensities of whole probes. Dye-bias tends to be signal intensity-dependent; therefore, the software selected probes using a set by a rank consistency filter for dye normalization. Said normalization was performed by LOWESS (locally weighted linear regression) which calculates the log ratio of dye-normalized Cy3 and Cy5 signals, as well as the final error of the log ratio. The significance (P) value was based on the propagate error and universal error models. In this analysis, the threshold of significant differentially expressed genes was <0.01 (for the confidence that the feature was not differentially expressed). In addition, erroneous data generated due to artifacts were eliminated before the data analysis using the software. The whole-genome DNA microarray data of treated skin (forepaws) of C57BL/6 WT and PAC1R KO mice have been submitted to NCBI’s GeneExpression Omnibus under the GEO series accession number GSE223124 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE223124; accessed on 19 February 2023).

PMC10002780

Xiaomei Dong,Haishan Luo,Jiabin Yao,Qingfeng Guo,Shuai Yu,Xiaoyu Zhang,Xipeng Cheng,Dexuan Meng

Characterization of Genes That Exhibit Genotype-Dependent Allele-Specific Expression and Its Implications for the Development of Maize Kernel

01-03-2023

maize,heterosis,kernel development,allelic expression,epigenetic regulation

Heterosis or hybrid vigor refers to the superior phenotypic traits of hybrids relative to their parental inbred lines. An imbalance between the expression levels of two parental alleles in the F1 hybrid has been suggested as a mechanism of heterosis. Here, based on genome-wide allele-specific expression analysis using RNA sequencing technology, 1689 genes exhibiting genotype-dependent allele-specific expression (genotype-dependent ASEGs) were identified in the embryos, and 1390 genotype-dependent ASEGs in the endosperm, of three maize F1 hybrids. Of these ASEGs, most were consistent in different tissues from one hybrid cross, but nearly 50% showed allele-specific expression from some genotypes but not others. These genotype-dependent ASEGs were mostly enriched in metabolic pathways of substances and energy, including the tricarboxylic acid cycle, aerobic respiration, and energy derivation by oxidation of organic compounds and ADP binding. Mutation and overexpression of one ASEG affected kernel size, which indicates that these genotype-dependent ASEGs may make important contributions to kernel development. Finally, the allele-specific methylation pattern on genotype-dependent ASEGs indicated that DNA methylation plays a potential role in the regulation of allelic expression for some ASEGs. In this study, a detailed analysis of genotype-dependent ASEGs in the embryo and endosperm of three different maize F1 hybrids will provide an index of genes for future research on the genetic and molecular mechanism of heterosis.

Characterization of Genes That Exhibit Genotype-Dependent Allele-Specific Expression and Its Implications for the Development of Maize Kernel Heterosis or hybrid vigor refers to the superior phenotypic traits of hybrids relative to their parental inbred lines. An imbalance between the expression levels of two parental alleles in the F1 hybrid has been suggested as a mechanism of heterosis. Here, based on genome-wide allele-specific expression analysis using RNA sequencing technology, 1689 genes exhibiting genotype-dependent allele-specific expression (genotype-dependent ASEGs) were identified in the embryos, and 1390 genotype-dependent ASEGs in the endosperm, of three maize F1 hybrids. Of these ASEGs, most were consistent in different tissues from one hybrid cross, but nearly 50% showed allele-specific expression from some genotypes but not others. These genotype-dependent ASEGs were mostly enriched in metabolic pathways of substances and energy, including the tricarboxylic acid cycle, aerobic respiration, and energy derivation by oxidation of organic compounds and ADP binding. Mutation and overexpression of one ASEG affected kernel size, which indicates that these genotype-dependent ASEGs may make important contributions to kernel development. Finally, the allele-specific methylation pattern on genotype-dependent ASEGs indicated that DNA methylation plays a potential role in the regulation of allelic expression for some ASEGs. In this study, a detailed analysis of genotype-dependent ASEGs in the embryo and endosperm of three different maize F1 hybrids will provide an index of genes for future research on the genetic and molecular mechanism of heterosis. Heterosis or hybrid vigor refers to the superior phenotypic traits of hybrids relative to their parental inbred lines [1]. Phenotypic traits include plant height, development rate, male and female fertility, nutrient quality, grain yield, and tolerance to stress. This phenomenon was first described by Charles Darwin and was later independently rediscovered by George H. Shull and Edward M. East in 1908. In the last few hundred years, heterosis has been widely exploited to increase crop yield and improve agricultural production [2,3,4,5,6]. Although not well understood at the molecular level, heterosis has been exploited over the past half-century in plants and animals [7,8]. Extensive studies on heterosis using RNA sequencing (RNA-seq) technologies have identified differentially expressed genes (DEGs) between F1 hybrids and their parental inbred lines in plants [9,10,11,12,13,14,15,16]. For example, in maize reciprocal F1 hybrids, a total of 1510 and 647 genes showed additive expression in shoots and roots, respectively [17]. Allele-specific expression (ASE) refers to the specific or preferential expression of one parental allele in the hybrid due to variations in regulatory sequences between the maternal and paternal genomes. The detection of single nucleotide polymorphisms (SNPs) in parent lines can be used to distinguish parental alleles and identify genes showing ASE in heterozygotes. To date, ASE has also been analyzed in several plants, including Arabidopsis, rice, maize, and barley [17,18,19,20,21]. For example, in rice, 23.8% of genes showed a preferential allele expression that was genotype-dependent in leaves of reciprocal crosses [16]. ASE accounted for 79.8% of the genes that showed more than a 10-fold expression level difference between an F1 and its parents. The expression difference caused by ASE may lead to phenotypic variation depending on the function of the genes. Several studies have suggested that ASE plays a role in heterosis because genetic variations often cause differences in gene expression, which may lead to phenotypic variations [22]. Genes showing allele-specific expression can lead to heterosis-relevant phenotype variation [23]. In addition to genetic variations, epigenetic variations have been suggested to play a role in the regulation of differential gene expression in plant hybrids, leading to the hybrid phenotype [24,25,26,27]. Genome activity and chromatin states can be regulated by epigenetic modifications in eukaryotes, mainly DNA methylation and histone modifications [28]. DNA cytosine methylation, as an important epigenetic modification, occurs in the context of CG, CHG, and CHH (where H is A, C, or T) in plants. The major role of DNA cytosine methylation is to silence transposable elements (TE) and repetitive sequences and suppress gene promoter activity [29,30,31]. Genome-wide allele-specific DNA methylation has been investigated in plants, including Arabidopsis, rice, and maize [32,33,34,35]. In endosperm, differential levels of DNA methylation have been observed around imprinted genes and are essential for allele-specific expression of imprinted genes [36,37]. Recently, in rice, DNA methylation differences between two inbred lines, ZS97 and MH63, and parental methylation interactions in reciprocal hybrids were investigated [38]. The results revealed a specific role for the divergence of parental CHG methylation in ASEGs, which is associated with phenotype variation and hybrid vigor in several plant species. Maize is an ideal model system for the study of ASEGs in hybrids due to its significant heterotic performance and well-known complex genome [39,40]. In this study, using RNA sequencing technology, we systematically identified genes exhibiting genotype-dependent ASEGs in embryos and endosperm from three maize F1 hybrids. Comparison of the allelic expression of these ASEGs in different tissues and hybrid crosses suggests that these ASEG patterns may have distinct implications for the genetic and molecular basis of heterosis. Further functional analysis indicated that these genotype-dependent ASEGs may make important contributions to kernel development. Finally, the potential relationship between DNA methylation and genotype-dependent ASEGs was also investigated. To explore global ASEGs in hybrid maize and reveal the mechanism of differential expression in the embryo and endosperm of F1 hybrids, three maize inbred lines (B73, Mo17, and CAU5) were chosen to generate three reciprocal crosses, B73 × Mo17 (BM) and Mo17 × B73 (MB), B73 × CAU5 (BC) and CAU5 × B73 (CB), and Mo17 × CAU5 (MC) and CAU5 × Mo17 (CM). RNA sequencing (RNA-seq) of the immature embryo and endosperm at 11 days after pollination (DAP) of three reciprocal crosses was performed. Here, a combination of proportion filters and statistical significance was applied to identify and classify genes that exhibit genotype-dependent allele-specific expression in the study (see Section 4). As illustrated in Figure 1A–F, most genes exhibited the expected maternal-to-paternal ratio of 1:1 or 2:1 in the embryo or endosperm (q > 0.05, x2 test). Based on statistically significant deviation (q < 0.05, x2 test), read counts from one parental allele being at least two-fold, five-fold, or nine-fold higher than read counts from another parental allele were used to identify ASEGs (Figure 1G–I). Under the criteria of a nine-fold difference between reads from two parents, a total of 740, 497, and 777 genes showed ASE in the embryos from BC/CB, MC/CM, and BM/MB, respectively (Figure 1G; Tables S1–S3). A total of 599, 347, and 657 genes showed ASE in endosperm from BC/CB, MC/CM, and BM/MB, respectively (Figure 1H; Tables S1–S3). These ASEGs were further classified according to which parent they preferred to express (Figure 1I). For example, according to criteria with a nine-fold difference, 740 ASEGs in the BC/CB embryos included 414 genes that preferred to express the CAU5 allele and 326 genes that preferred to express the B73 allele, and 599 ASEGs in endosperm from BC/CB included 327 genes that preferred to express the CAU5 allele and 272 genes that preferred to express the B73 allele (Figure 1I). In the BM/MB hybrid cross, the number of ASEGs was the largest in both the embryo and the endosperm (Figure 1G,H). The differences in the number of ASEGs for the three hybrid crosses were largely due to differences in the number of genes with polymorphisms. Using the circus program, the chromosomal locations of the ASEGs were detected in three hybrid crosses, and these ASEGs were evenly distributed in all chromosomes without obvious location preference (Figure 1J). The average distance between ASEGs was 2.23 Mb in BC/CB, 3.33 Mb in MC/CM, and 2.23 Mb in BM/MB. Then, the genome was searched for clusters containing at least two ASEGs within a 1 Mb region. A total of 35, 11, and 51 clusters of ASEGs were identified in BC/CB, MC/CM, and BM/MB crosses (Tables S1–S3), which is significantly higher than the numbers expected by chance (Fisher test; p-value < 0.001). For example, five genes (Zm00001d006941, Zm00001d006942, Zm00001d006943, Zm00001d006944, and Zm00001d006945) were located within ~17 kb on chromosome 2 from 219,959,370 to 219,976,357 bp. Interestingly, all five genes preferred to express the CAU5 allele in the BC/CB hybrid (Table S1). To increase the precision of the subsequent analysis, only ASEGs that reached a nine-fold difference between the reads of the two parents were used. First, we examined the genes that show consistent ASE across tissues in one hybrid cross. As visualized in the Venn diagram, approximately half of the ASE was consistent in both the embryo and endosperm of a hybrid cross (Figure 2A). For example, in BC/CB crosses, a comparison of ASEGs revealed 380 genes (51% in embryo and 63.4% in endosperm) that showed a consistent direction of expression bias in two tissues (embryo and endosperm), which included 193 genes that showed B73-biased expression and 187 genes that showed CAU5-biased expression (Figure 2A). Furthermore, ASEGs were found in a single tissue that usually did not have informative SNPs or had insufficient reads to identify whether they were ASEGs in other tissues (Figure 2B). For example, among the 414 B73-biased ASEGs identified in the BC/CB embryo, 223 genes (53.8%) exhibited B73-biased expression, only 29 genes (7.0%) were biallelically expressed, and 162 genes (39.1%) were not expressed or analyzed in BC/CB endosperm. Indeed, we found that the expression levels of ASEGs exhibited differences in the embryo and endosperm (Figure 2C and Figure S1). Therefore, for ASEGs analyzed and expressed in different tissues, most of them had consistent ASE across different tissues from one hybrid cross. To further analyze whether some of the ASEGs showed consistency in different hybrid crosses, a Venn diagram of three hybrid crosses was generated. As illustrated in Figure 2D–F, only ~15% of ASEGs exhibited consistent ASE in the same tissue from all three hybrid crosses. Further analysis found that for ASEGs analyzed or expressed in different hybrid crosses, approximately half exhibited ASE from some genotypes but not others (Figure S2). For example, among the 414 B73-biased ASEGs identified in BC/CB embryos, 136 genes (32.8%) exhibited B73-biased expression, 99 genes (23.9%) were biallelically expressed, and 179 genes (43.2%) were not expressed or not analyzed in the BM/MB embryo. The subcellular locations of ASEG encoded proteins that exhibited consistent ASE in the same tissue from the three hybrid crosses were then analyzed on the website of GenScript-PSORT II (https://www.genscript.com/psort.html?src=leftbar, accessed on 20 November 2022, Figure 2G–I). These ASEGs were separated into various subcellular locations, and nearly 40% of the ASEGs were located in the nucleus. For example, 414 B73-biased ASEGs identified in BC/CB embryos were mainly located in the nucleus (35.8%), followed by mitochondria (20.1%) and cytoplasm (19.4%), and the rest were distributed in several other organelles. To explore the function of these ASEGs in the development of embryos and endosperm of maize, we performed gene ontology (GO) analysis by distinguishing ASEGs in different genotypes and tissues (Figure S3). Three enriched GO terms for molecular functions, including the tricarboxylic acid cycle, aerobic respiration, and energy derivation by oxidation of organic compounds, were detected in B73-biased ASEGs identified in the embryos from BC/CB. A GO term for molecular functions, ADP binding, was detected in Mo17-biased ASEGs identified in embryos from BM/MB. CAU5-biased and Mo17-biased ASEGs in endosperm from MC/CM were enriched in the nuclear envelope and cytosol, respectively. Hence, GO analysis indicated that ASEGs have important roles in biosynthesis, development, and regulation. However, limited GO terms were common to ASEGs identified in different genotypes and tissues, suggesting that allele-specific genes have different roles in different genomic backgrounds. Then, an ASEG, Zm00001d046765 (Zm765), was selected for further phenotype analysis. Zm765 is a B73-biased ASEG detected in the BC/CB embryo and is highly expressed in the early period of the kernel, which encodes a glycosyl hydrolase of unknown function. Therefore, we focused on comparing the kernel phenotypes of the overexpression lines and mutant lines with the transgenic receptor line to determine whether abnormal kernel phenotypes occurred during development. Both the area of the immature 15 DAP kernels and the mature 30 DAP kernels in the mutant lines showed a significant decrease compared to those in the transgenic receptor line (p-value < 0.01, Student’s test) (Figure 3). Then, we used transgene technology to create an overexpression line (Zm765-OE) to further analyze the function of Zm765. As shown in Figure 3C–E, at 15 and 30 days after self-pollination, the kernel areas of the Zm765 overexpression lines were significantly larger than those of the transgenic receptor line at the corresponding period, which indicated that Zm765 may participate in the development of the kernel. Based on the MethylC-seq performed for MC/CM endosperm, we scanned the genome to identify genotype-dependent differentially methylated regions (gDMRs) using a sliding window strategy in MC/CM endosperm (see Section 4). As a result, 1225 gDMRs were identified in the CG context (CG_gDMRs) (Table S4), including 649 CG_gDMRs showing hypermethylation in the CAU5 allele (CG_gDMRs_HC) and 576 CG_gDMRs showing hypermethylation in the Mo17 allele (CG_gDMRs_HM). In the CHG context, 307 CHG_gDMRs (gDMRs in the CHG context) were identified in MC/CM endosperm (Table S5), including 169 CHG_gDMRs hypermethylated in the CAU5 allele (CHG_gDMRs_HC) and 138 CHG_gDMRs hypermethylated in the Mo17 allele (CHG_gDMRs_HM). In Figure 4A–D, the allelic methylation pattern in the gDMRs region identified in MC/CM endosperm is shown. The availability of ASEG and DNA methylome data in MC/CM allowed us to investigate the relationship between epigenetic modification and genotype-dependent allelic expression in maize. First, the patterns of allele DNA methylation at ASEGs were determined in the MC/CM endosperm (Figure 4E–J). As a result, in the context of CG, the methylation levels of the activated allele were slightly lower than those of the silenced alleles in the 5′ portion of the gene body regions (Figure 4E,F). However, the methylation levels of the activated allele were slightly higher than those of the silenced alleles in the 3′ portion of the gene body regions (Figure 4E,F). The levels of DNA methylation between two alleles of all genes were similar (Figure 4G). In the CHG context, similar results were observed. Then, we analyzed the association of gDMRs with ASEGs. Approximately 3% of ASEGs overlapped with CG_gDMRs or CHG_gDMRs (Figure 4K,L and Figure S4). For example, among 65 Mo17-biased ASEGs in the MC/CM endosperm that overlapped with the analyzed methylation region in the CHG context, 7 ASEGs overlapped with regions exhibiting hypermethylation in the CAU5 allele, and 2 ASEGs overlapped with CHG_gDMR showing hypermethylation in the CAU5 allele (Figure S4). To explore global ASEGs in hybrid maize and reveal the mechanism of differential expression in embryo and endosperm from F1 hybrids, three maize inbred lines (B73, Mo17, and CAU5) were chosen to generate three reciprocal crosses, BC/CB, MC/CM, and BM/MB. The proportion of genotype-dependent ASEGs identified in the three F1 hybrids did not differ significantly, which is similar to a previous report in rice and maize hybrids [16,41]. When ASEGs are compared in two tissues and three hybrids, ASEGs can be classified into two major patterns: consistent ASEGs and inconsistent ASEGs. In embryo and endosperm, most ASEGs were consistent in the same hybrid. Such a consistent biased expression of the genes would result in partially to fully dominant effects on the traits regulated by the genes [22]. However, in the same tissue from different hybrids, half of the ASEGs were inconsistent. Therefore, these results implied that the regulatory mechanism for the allele-specific expression of genotype-dependent ASEGs was mainly influenced by genetic variations [40,42]. In addition, ASEGs identified in three hybrid crosses also tend to be clustered in the genome. Moreover, the expressed directions of ASEGs located in one cluster were the same (Tables S1–S3), which indicated that the cause and regulatory mechanism of one ASEGs clusters might be the same. For an inbred line to cross with different inbred lines, the favorable allele of the genes can be variable, and the hybrid can make use of the favorable allele of the genes and express them at high levels. The observed genotype-dependent ASEGs in the embryo and endosperm tissue of hybrid maize could represent a common mechanism of complementary allelic effects in hybrids and show the importance of the parental genotype in both cross-breeding and hybrid breeding [41,43]. The function of genotype-dependent ASEGs was involved in plant development and resistance to stress [44,45]. For example, Ghd7 (a major QTL for grain number, plant height, and heading date) is present in rice-inbred MH63 but absent in inbred ZS97 and exerts a large pleiotropic dominance effect on all traits [22]. In our study, Zm765, expressed in B73 allele but silenced in CAU5 allele in BC/CB embryo, contributed to the development of maize kernel. In further work, whether Zm765 exerts a pleiotropic dominance effect will be investigated. In addition, the GO annotation of genotype-dependent ASEGs was mainly enriched in metabolic pathways of substances and energy, such as the tricarboxylic acid cycle (TCA). TCA is the final oxidation pathway for glucose, fats, and amino acids and is the most important source of ATP production in cells [46]. Previous work also suggested that the TCA cycle can regulate plant reproductive development [47]. Therefore, our results indicated the importance of the parental genotype in the superior performance of the hybrid. In recent work, ASE was negatively associated with allele-specific methylation (ASM) in CHG [38], indicating a specific pattern of DNA methylation reprogramming in hybrid rice and pointing to the role of parental CHG methylation divergence in ASE, which is associated with variation in phenotypes and hybrid vigor in several species of plants. In our study, although hypermethylation at CG and CHG repressed allele expression from one parent line, the relationship between ASEG and CHG_gDMR, but not CG_gDMR, was significantly higher than that of all genes. This is consistent with the finding that the silent maternal allele of paternally expressed genes in the endosperm of Arabidopsis lyrata is marked by hyper CHG methylation [48]. The present results indicate that CHG methylation of the allele-specific gene body is likely to be inherited from the parental epigenomes and is maintained or reinforced in an allele-specific manner in the hybrid and during development. Of course, only 10% of ASEGs overlapped with gDMRs. Except for DNA methylation, extensive allele-level histone modification was correlated with genome-wide changes in the allelic expression of genes [25,49]. Hence, the regulation for allele-specific expression of ASEGs was complex and should be explored with more datasets in the future. The hybrid lines B73(♀) × Mo17(♂), Mo17(♀) × B73(♂), B73(♀) × CAU5(♂), CAU5(♀) × B73(♂), CAU5(♀) × Mo17(♂), Mo17(♀) × CAU5(♂) were obtained from the inbred lines B73, Mo17, and CAU5 in the summer of 2021 at the experimental station of Shenyang Agriculture University in Shenyang, Liaoning. The ears and tassels of the three lines were bagged with kraft paper one day prior to pollination. The next day, each paper bag was patted to collect pollen from one parent, which was used to pollinate the ear of the other parent. After 11 days, the ears of six reciprocal crosses (BM, MB, BC, CB, MC, and CM) were collected. In this study, BM /MB represents the crosses of B73 × Mo17 and Mo17 × B73, BC /CB represents the crosses of B73 × CAU5 and CAU5 × B73, and MC /CM represents the crosses of Mo17 × CAU5 and CAU5 × Mo17. The embryo and endosperm samples were isolated using a Quick RNA Isolation Kit (Huayueyang Biotechnology of Beijing, China). mRNA library construction and sequencing were conducted following the Illumina manufacturer instructions. Total RNA was extracted as input material for the RNA sample preparations. The NEB Next® Ultra TM RNA Library Prep Kit from Illumina® (NEB, San Diego, CA, USA) was used to generate mRNA libraries. High-throughput mRNA sequencing was performed using the Illumina NovaSeq 6000 platform, and 150 bp paired-end reads were generated for each library. An average of 3 Gb data for each replicate was obtained and used for the following analyses, providing sufficient sequencing depth for the imprinting analysis. Genomic DNA was extracted by the DNeasy plant mini kit (QIAGEN, Hilden, Germany). The purity was detected by a Nano Photometer® spectrophotometer (IMPLEN, Westlake Village, CA, USA). Then, DNA fragments were treated with bisulfite (Accel-NGS Methyl-Seq DNA Library Kit for Illumina, Swift). Finally, the library quality of MethylC-seq was checked by the Agilent Bioanalyzer 2100 system. Pair-end sequencing was performed on the Illumina platform (Illumina, San Diego, CA, USA). First, clean reads were mapped to the B73 reference genome (Version 4) using HISAT2 software with default parameters [50]. Cufflinks software (V2.2.1) was used to estimate the normalized gene expression values (FPKM) [51]. The calculated log2 (FPKM + 1) values were used to analyze the correlation coefficient between replicates. Hierarchical clustering analysis was performed on the relative expression value by setting the parameters’ average linkage and Euclidean distance using MeV (http://www.tm4.org/mev.html, accessed on 15 April 2022). Resequencing reads of B73, Mo17, and CAU5 inbred lines were downloaded from NCBI (SRR12415217, SRR12415218, and SRR3124079). Reads were mapped using BWA with default parameters [52]. Samtools were used to exclude reads that were not uniquely mapped with the -q 20 parameter [52]. SNPs between B73, Mo17, and CAU5 inbred lines were called using Bcftools with default parameters [53]. First, clean reads were mapped to the B73 reference genome (Version 4) using HISAT2 software (accessed on 12 March 2022) with default parameters [50]. To avoid bias, SNP sites were converted to CAU5 nucleotides to obtain the SNP-substituted CAU5 genome. All clean reads from three biological replicates of each sample were mapped to the B73 (Version 4) and SNP-substituted CAU5 genomes using HISAT2 with default parameters. Samtools was used to exclude reads that were not uniquely mapped with the -q 20 parameter [53]. Three replicates from each sample were merged for further identification of the imprinted genes. According to the SNP information, the reads aligned at the SNP site were split into maternal or paternal alleles using Samtools mpileup. The maternal and paternal read counts of each gene were summed. If the sum of the read counts of the annotated genes at all SNP sites was ≥20, the gene’s imprinting status could be analyzed. The maternal-to-paternal allele ratio of the genes analyzed was determined using the χ2 test to detect the deviation of the maternal: paternal ratio from the theoretically suggested 1:1 ratio in the embryo and the 2:1 ratio in the endosperm. Finally, read counts from one parental allele were used to identify ASEGs at least two-fold, five-fold, or nine-fold higher than read counts from another parental allele. GO analysis of ASEG was performed using Agri GO v2.0 (accessed on 20 July 2022) [54]. Only GO terms are displayed among cell components, molecular functions, and biological processes with significant (p-value < 0.05) enrichment compared to all genes. MethylC-seq reads were generated using the same workflow as in previous work. First, low-quality reads were filtered using SolexaQA [55]. The remaining reads were mapped to the B73 genome using Bismark [56]. The bulk methylation of endosperm was calculated by the ratio of Cs to all Cs and Ts from all CG, CHG, or CHH sites. Then, SNPs were used to separate allele-specific MethylC sequence reads from the hybrid endosperm. Only sites with at least five reads were used in subsequent analyzes. The same criteria were used to identify CG_gDMR and CHG_gDMR as in previous work. First, a sliding-window approach with a 200-bp window and 20-bp step was adopted throughout the genome. Only windows containing more than five CG/CHG sites supported with at least five reads were kept as CG/CHG analyzed regions. Second, the statistical significance of the allelic methylation bias in each window was assessed by the p-value using Fisher’s exact test. The resulting p values were converted to Q values. Finally, the gDMRs were identified according to the following criteria: FDR < 0.01; the methylation level between two alleles differed by >30%; and the hypermethylated alleles had methylation levels > 40% in the context of CG. The candidate gDMRs were then further filtered using a smaller window size of 50 bp, and gDMRs within 200 bp were merged. We prepared overexpression constructs for the genetic transformation of one ASEG, Zm00001d046765 (Zm765). Full-length CDS (without stop codon) of Zm765 was amplified from Zm765 cDNA and cloned into the binary vector pBCXUN-MYC to generate the pOE Zm765-MYC construct driven by the ubiquitin promoter (using the primers Zm765-CDS-F/R listed in Table S6). Transformations using the overexpression construct were introduced into the maize receptor line KN5585 via Agrobacterium-mediated transformation, and we verified the transgenic positive line with the primer of Bar-F/R (Table S6) [57]. For the CRISPR/Cas9 gene-editing construct, a 19-bp sequence from the first exon of Zm765 was selected as a guide RNA (gRNA) and introduced into the pBUE411 vector as previously described [58]. For transformations using the CRISPR/Cas9 construct, two homozygous knockout lines of this gene with insertions or deletions at the target sites were identified from the independent positive transgenic lines (T0) by PCR amplification and sequencing analysis (using the primer of Zm765-CDS-F/R listed in Table S6). Independent positive transgenic lines were obtained and self-pollinated to generate homozygous progenies for kernel phenotype analysis. One-third of the kernels in the middle of the ear of the six crossed ears (BC, CB, MC, CM, BM, and MB) at 15 DAP and 30 DAP were separated and imaged under a light microscope (Olympus, Tokyo, Japan) one by one. Image J software was used to measure the area of each kernel. All primers used in this study are listed in Table S6. Allelic expression profiles in hybrid maize determined by RNA-sequencing technology demonstrated a type of genotype-dependent monoallelic expression gene in plants. The association analysis of DNA methylation and ASEGs indicated that epigenetic modifications have potential effects on the expression of ASEGs. In the future, we will pay more attention to the detailed functional analysis of the ASEGs detected in our study. Nonetheless, the ASEGs have provided an index of the genes for future studies, especially with respect to the genetic and molecular mechanism of heterosis, which would be helpful for hybrid breeding.

PMC10002785

Kota Takemoto,Luga Santo Lomude,Sachio Takeno,Tomohiro Kawasumi,Yukako Okamoto,Takao Hamamoto,Takashi Ishino,Yuki Ando,Chie Ishikawa,Tsutomu Ueda

Functional Alteration and Differential Expression of the Bitter Taste Receptor T2R38 in Human Paranasal Sinus in Patients with Chronic Rhinosinusitis

24-02-2023

bitter taste receptor (T2R),T2R14,T2R38,chronic rhinosinusitis (CRS),eosinophils,ciliated cells,nitric oxide (NO),single nucleotide polymorphism (SNP)

The bitter taste receptors (T2Rs) expressed in human sinonasal mucosae are known to elicit innate immune responses involving the release of nitric oxide (NO). We investigated the expression and distribution of two T2Rs, T2R14 and T2R38, in patients with chronic rhinosinusitis (CRS) and correlated the results with fractional exhaled NO (FeNO) levels and genotype of the T2R38 gene (TAS2R38). Using the Japanese Epidemiological Survey of Refractory Eosinophilic Chronic Rhinosinusitis (JESREC) phenotypic criteria, we identified CRS patients as either eosinophilic (ECRS, n = 36) or non-eosinophilic (non-ECRS, n = 56) patients and compared these groups with 51 non-CRS subjects. Mucosal specimens from the ethmoid sinus, nasal polyps, and inferior turbinate were collected from all subjects, together with blood samples, for RT-PCR analysis, immunostaining, and single nucleotide polymorphism (SNP) typing. We observed significant downregulation of T2R38 mRNA levels in the ethmoid mucosa of non-ECRS patients and in the nasal polyps of ECRS patients. No significant differences in T2R14 or T2R38 mRNA levels were found among the inferior turbinate mucosae of the three groups. Positive T2R38 immunoreactivity was localized mainly in epithelial ciliated cells, whereas secretary goblet cells generally showed lack of staining. The patients in the non-ECRS group showed significantly lower oral and nasal FeNO levels compared with the control group. There was a trend towards higher CRS prevalence in the PAV/AVI and AVI/AVI genotype groups as compared to the PAV/PAV group. Our findings reveal complex but important roles of T2R38 function in ciliated cells associated with specific CRS phenotypes, suggesting the T2R38 pathway as a potential therapeutic target for promotion of endogenous defense mechanisms.

Functional Alteration and Differential Expression of the Bitter Taste Receptor T2R38 in Human Paranasal Sinus in Patients with Chronic Rhinosinusitis The bitter taste receptors (T2Rs) expressed in human sinonasal mucosae are known to elicit innate immune responses involving the release of nitric oxide (NO). We investigated the expression and distribution of two T2Rs, T2R14 and T2R38, in patients with chronic rhinosinusitis (CRS) and correlated the results with fractional exhaled NO (FeNO) levels and genotype of the T2R38 gene (TAS2R38). Using the Japanese Epidemiological Survey of Refractory Eosinophilic Chronic Rhinosinusitis (JESREC) phenotypic criteria, we identified CRS patients as either eosinophilic (ECRS, n = 36) or non-eosinophilic (non-ECRS, n = 56) patients and compared these groups with 51 non-CRS subjects. Mucosal specimens from the ethmoid sinus, nasal polyps, and inferior turbinate were collected from all subjects, together with blood samples, for RT-PCR analysis, immunostaining, and single nucleotide polymorphism (SNP) typing. We observed significant downregulation of T2R38 mRNA levels in the ethmoid mucosa of non-ECRS patients and in the nasal polyps of ECRS patients. No significant differences in T2R14 or T2R38 mRNA levels were found among the inferior turbinate mucosae of the three groups. Positive T2R38 immunoreactivity was localized mainly in epithelial ciliated cells, whereas secretary goblet cells generally showed lack of staining. The patients in the non-ECRS group showed significantly lower oral and nasal FeNO levels compared with the control group. There was a trend towards higher CRS prevalence in the PAV/AVI and AVI/AVI genotype groups as compared to the PAV/PAV group. Our findings reveal complex but important roles of T2R38 function in ciliated cells associated with specific CRS phenotypes, suggesting the T2R38 pathway as a potential therapeutic target for promotion of endogenous defense mechanisms. Bitter taste receptors (T2Rs) are chemosensory proteins belonging to the G protein-coupled receptor (GPCR) superfamily [1]. Humans are known to express 25 different T2Rs, which are variously activated by more than one thousand compounds [2]. The bitter molecules include plant extracts (alkaloids, terpenoids, flavonoids, etc.), bacterial products, and synthetic chemical compounds [3,4]. Numerous studies conducted in recent years have shown expression of bitter receptors in a wide variety of extraoral tissues. These receptors are considered to perform diverse and important physiological functions apart from their better-understood role in taste [5,6,7]. The bitter taste receptors in the human airways are known to elicit innate immune responses and eradicate airborne pathogens [8,9,10]. Recently, T2Rs have been studied for their roles in sinonasal immunity and their involvement in chronic rhinosinusitis (CRS) pathophysiology [11,12,13,14]. Epithelial ciliated cells express at least one bitter receptor (T2R38, encoded by the TAS2R38 gene) that recognizes foreign bacteria, viruses, or fungi. The downstream pathways trigger immune defenses leading to the release of nitric oxide (NO), which plays crucial roles in the maintenance of physiological homeostasis with bactericidal effects [15]. T2R14 is one of the highly expressed T2Rs in the upper respiratory epithelial cells of humans [16] and is activated by plant flavones, Pseudomonas aeruginosa quinolones [17], and many pharmaceutical drugs [18,19]. Chronic rhinosinusitis (CRS) is a common disease with a considerable social burden. It is defined as the symptomatic inflammation of the sinonasal mucosa with evidence of inflammation on exam and/or imaging lasting more than 12 weeks [20]. CRS patients are categorized into two subtypes based on their phenotypic features: those with nasal polyps (CRSwNP) and those without (“sans”, CRSsNP) [21]. Eosinophilic chronic rhinosinusitis (ECRS) is a subgroup of CRSwNP that is associated with severe eosinophilic infiltration and intractable social burden according to the Japanese Epidemiological Survey of Refractory Eosinophilic Chronic Rhinosinusitis (JESREC) [22]. ECRS is a predominantly type 2-mediated inflammatory disease characterized by the presence of increased levels of cytokines including interleukins (IL) -4, -5, -13, -25, and -33, as well as thymic stromal lymphopoietin (TSLP) [23]. Persistent symptoms include nasal blockage, nasal discharge, and olfactory dysfunction, with patients frequently reporting an altered sense of taste [24,25,26]. As a CRS symptom, a loss of taste has been drawing increasing attention and is believed to be associated with a loss of smell [27,28]. However, studies have yet to elucidate the cellular mechanisms related to taste receptor function and its downstream immunologic effects, and to associate these features with the different CRS subsets. In the present study, we hypothesized the possible roles of T2Rs in CRS in relation to underlying pathophysiologic mechanisms that translate into clinical phenotypes, i.e., non-ECRS and ECRS. For this purpose, we investigated the gene expression level and protein localization of T2R14 and T2R38 in the inferior turbinate and paranasal sinus mucosae of CRS patients and controls and assessed their relationship with the levels of fractional exhaled NO (FeNO). In addition, we examined whether differences in TAS2R38 alleles at the pertinent chromosomal location influenced CRS prevalence in the study population. The T2R38 isoform is the most well-known and well-characterized T2R, the haplotypes of which have common polymorphisms, one encoding a functional receptor and the other encoding a nonfunctional one [29,30,31]. The functional allele of “high tasters” contains the proline-alanine-valine (PAV) sequence, whereas “low tasters” have the alanine-valine-isoleucine sequence (AVI). Our results herein suggest the potential value of T2R38 as a therapeutic target for the promotion of endogenous immune responses in CRS patients involving increased NO production. The background and clinical characteristics of the study population are summarized in Table 1. We divided the 92 CRS patients into two groups based on the JESREC criteria for diagnosing ECRS [22]: namely, a non-ECRS group (n = 56) and an ECRS group (n = 36). No significant difference was found among the three groups in the baseline data of gender, allergic rhinitis (AR) co-morbidity, or body mass index (BMI). Mean age was significantly higher in the non-ECRS group (p < 0.001) and the ECRS group (p < 0.01) than in the control group. Significant differences were observed between the non-ECRS and ECRS groups in the proportion of asthma co-morbidity, the degree of blood and tissue eosinophils, and the severity of CT scores. Further, the non-ECRS patients showed a significantly lower level of blood eosinophils than the control subjects. We examined single nucleotide polymorphisms (SNPs) of the TAS2R38 gene in the study population. The proportion of the PAV/PAV allele and other alleles (PAV/AVI and AVI/AVI) in the controls and CRS (ECRS + non-ECRS) groups are shown in Figure 1a. The PAV/PAV proportion tended to be higher in the control group than in the CRS group (41.1% vs. 27.1%), although the difference was not statistically significant (p = 0.085). The CRS prevalence classified by TAS2R38 gene polymorphisms is shown in Figure 1b. There was a trend towards higher CRS prevalence in the non-PAV/PAV genotype group as compared to the PAV/PAV group (69% vs. 54.3%). However, the difference was not significant (p = 0.085). As for gender difference, the PAV/PAV proportion in females tended to be higher than in males in the control group, although the difference was not statistically significant (p = 0.265). No significant differences in gender proportion for each SNP allele was found in the CRS prevalence. Previous studies have reported that human bitter taste receptors are expressed in human upper airway tissues and that the density of these receptor proteins is higher in the ethmoid sinus than in the nasal cavity [16]. We therefore sought to evaluate T2R mRNA expression in different areas of the sinonasal mucosae in both CRS and control groups. The mRNA levels of T2R14 and T2R38 in the ethmoid sinus mucosa, nasal polyps, and inferior turbinate mucosa were assessed by quantitative RT-PCR (Figure 2). Nasal polyp specimens in ECRS patients showed a significant downregulation of T2R14 mRNA expression compared to the ethmoid mucosa samples across all groups. We also observed significant downregulation of T2R38 mRNA levels in the ethmoid mucosa of non-ECRS patients and in the nasal polyps of ECRS patients. There was no significant difference in T2R14 and T2R38 mRNA levels among the inferior turbinate mucosae of the three groups. We also plotted the ratio of mRNA expression levels of ethmoid sinus mucosa to inferior turbinate mucosa (Eth/IT ratio) for each subject (Figure 3). There was no significant difference in the Eth/IT ratios for T2R14 among the groups. In contrast, the non-ECRS patients showed a significantly lower Eth/IT ratio for T2R38, reflecting the lower mRNA levels in the ethmoid sinus area. Since transcriptional changes in T2R38 were associated with CRS pathology and clinical manifestations, we examined the sinus tissue distribution of T2R14 and T2R38 proteins in representative cases. Figure 4 provides immunohistological images of the distributions of T2R38- and T2R14-positive cells in the ethmoid sinus and nasal polyp mucosae. In the non-ECRS group, intense inflammatory cell infiltration with neutrophils and lymphocytes dominated the ethmoid mucosa on conventional histological examination. By contrast, dense eosinophil infiltration was observed in the ECRS group (Figure 4i,j). Positive T2R38 immunoreactivity was localized mainly in epithelial ciliated cells along the luminal surface. The ethmoid specimens from the patients in the ECRS group generally showed higher rates of T2R38-positive cells throughout the epithelial area as compared to those from the non-ECRS group, with ciliated cells being predominant (Figure 4a,b). In contrast, secretary goblet cells that were widely scattered in the epithelial layer of nasal polyp specimens showed a lack of staining (Figure 4c). The degree of T2R14 staining appeared to be identical among the specimens of the three groups. Epithelial ciliated cells covering a large area of the human paranasal sinuses produce large amounts of NO, which plays a major role in airway physical defense by inducing increased ciliary beating [15]. T2R activation in response to T2R38-specific agonists results in the release of NO by means of calcium-dependent activation of constitutive nitric oxide synthase (NOS), contributing substantially to antibacterial defense mechanisms [32]. We therefore performed measurements of oral and nasal FeNO values in each group as a possible surrogate marker of T2R functional activities. As shown in Figure 5, the median oral FeNO levels were 22.5 (interquartile range; 15.7–35) ppb in control subjects, 16.0 (12–21.5) ppb in non-ECRS patients, and 24 (16.8–44) ppb in ECRS patients. The median nasal FeNO levels were 32.5 (25.3–52) ppb in control subjects, 24 (17–32) ppb in non-ECRS patients, and 28.3 (18.9–49.4) ppb in ECRS patients. Compared with the control group, the non-ECRS patients showed significantly lower both oral and nasal FeNO levels. The nasal cavity and paranasal sinuses are frontline defense systems of the human respiratory tract, where immunologic responses occur against various aerosolized pathogens contaminating inhaled air [10,33]. CRS is considered a category of heterogenous syndromes resulting from dysfunctional interactions between various environmental factors and the host immune system [20]. The complexity of CRS pathology has been evidenced by numerous research efforts aimed at recognizing more detailed endotypes, i.e., those defined by the presence of particular patterns of immune cells or biomarkers, and at developing optimal treatment modalities for each subset of patients [23,34,35,36,37]. As a CRS symptom, loss or changes of taste have drawn attention and are reported to be associated with a loss of smell [25,26,27,28]. Although extraoral T2R receptor expression in sinonasal epithelium would not affect the ability of taste and flavor perception in CRS patients, the association between improvements in taste and smell is intriguing given that the latter has been shown to correlate with CRS disease severity [20,24,38]. Studies have yet to elucidate the cellular mechanisms related to taste receptor function and immunologic responses underlying the different CRS subsets. In this sense, bitter taste receptors have gained attention for their roles in sinonasal immunity and contributions to CRS pathophysiology [11,12,13,14]. So far, two different cell types in the sinonasal mucosa have been reported to express T2Rs: namely, epithelial ciliated cells and solitary chemosensory cells (SCCs) [39]. In the present study, we tried to assess the possible relationship between the expressions and distributions of representative T2Rs and CRS pathology, with emphasis on the CRS phenotypic classification, as a measure for innate immunity function. To date, there has been no study concerning the interplay of T2R38 functional expression and development of CRS with different phenotypes in the Japanese population. We also examined whether TAS2R38 gene polymorphisms caused subjects to be more susceptible to sinonasal infection through attenuated T2R38 function, because the receptor’s function is dictated by specific genetic polymorphisms [13,40]. Further, no previous report has directly examined the relationship between T2R expression levels and nasal FeNO concentrations, as well as the frequency of genetic variants of T2R38 in the Japanese population. In this study, we found significantly decreased T2R38 mRNA levels in the ethmoid mucosae of non-ECRS patients and in the nasal polyps of ECRS patients. Nasal polyp specimens in ECRS patients also showed a significant downregulation of T2R14 mRNA expression. There was no significant difference in the T2R14 and T2R38 mRNA levels of the inferior turbinate mucosae among the groups. The results indicate differences in anatomical structure and physiological function between the inferior turbinate and the ethmoid sinus mucosa, with the latter being intimately involved in the development of CRS pathology. Chen et al. reported that expression of T2Rs was higher in the ethmoid sinus than in other locations of the nasal cavity (inferior turbinate, middle turbinate, and nasal septum) [16]. In contrast with rhinitis, whose lesions are restricted to the nasal cavity with the inferior turbinate of particular importance [41], the paranasal sinuses and especially the ethmoid sinus are the major areas affected in CRS. Biopsies obtained from the ethmoid sinus may serve as the best location for the functional study of upper airway taste receptors in humans. Interestingly, nasal polyps showed significantly lower mRNA levels of T2R14 and T2R38 than ethmoid sinus mucosae obtained from the same patients. Our immunohistochemical studies indicate that positive T2R immunoreactivity was localized mainly in epithelial ciliated cells along the luminal surface, whereas secretary goblet cells predominantly observed as epithelial non-ciliated components of nasal polyp specimens showed a lack of staining [20,42]. So far, there are no data available to directly compare T2R gene expression between the ethmoid sinus and nasal polyps in human CRS. Our results are supported by previous reports that T2R proteins are found exclusively within ciliated cells [43]. Recent immunohistochemical analysis reported a significant difference in T2R38 protein levels between both CRSsNP and CRSwNP patients and healthy controls [44]. We consider that the decreased expression of T2R38 in nasal polyps is attributable to the fact that nasal polyps contain higher proportions of secretory cells with fewer ciliated cells. T2R activation results in the release of cellular NO and the increase in ciliary beat frequency (CBF), both of which contribute to the maintenance of airway physiological homeostasis [10,31]. T2R38 appears to be an essential mediator of sinonasal epithelial defense against respiratory bacterial infections [9,13,45]. When ciliated cells were stimulated with known T2R38-specific agonists, such as the bitter-tasting synthetic compound phenylthiocarbamide (PTC), they exhibited calcium-dependent activation of nitric oxide synthase (NOS) [13]. The robust NO production leads to cellular protein phosphorylation by protein kinase G (PKG) and increases ciliary beat frequency to facilitate the mucous movement out of the airway [19]. T2R38 also detects bacterial products such as acyl-homoserine lactones (AHLs) produced by P. aeruginosa [8,9,45] and activates NO production specifically via the NOS3 isoform localized on the cilia [42]. The NO that is generated diffuses into the airway surface liquid (ASL) and plays roles in anti-bacterial defense mechanisms [17,46]. We found that both oral and nasal FeNO levels in non-ECRS patients were significantly lower than those of control subjects and were associated with decreased T2R38 mRNA expression levels in the ethmoid mucosae of such patients. The non-ECRS patients also showed a significant decrease in the Eth/IT ratio for T2R38, which might primarily reflect lower mRNA levels in the ethmoid sinus area. Further, immunostaining revealed that T2R38 expression in the ethmoid sinus mucosa in the non-ECRS patients generally showed lower degrees of T2R38-positive cells throughout the epithelial area compared to ECRS patients. The results support our hypothesis that the onset and persistence of CRS morbidity might be partly triggered by decreased NO production due to attenuated expression of T2R38 in the middle meatus area, leading to the vulnerability of defense systems against airborne pathogens. The role of nasal FeNO in CRS patients has been a matter of debate due to its multiple origin, with contributions from both the paranasal sinus cavities and the inflamed sinonasal mucosa [15,47]. Non-ECRS patients generally showed lower FeNO levels as a result of occluded paranasal sinus ventilation and damaged ciliated epithelia [48]. On the other hand, several attempts have been made to measure nasal NO levels as a marker for assessing the severity of type 2 inflammation with tissue eosinophils of ECRS patients [48,49]. The present results of FeNO levels are compatible with our previous studies from different patient populations [48]. The treatment of CRS may restore both the NOS expression of the sinus ciliated cells and the ability of NO to pass through the paranasal sinus ostia. This is particularly important in cases of non-ECRS patients with a limited area of sinus disease [50]. Anyway, further research is required to elucidate how the post-surgery recovery process of sinus ciliary epithelial cells is functionally related to recovery of T2R38 expression levels that leads to NO production with morphological integrity. The present study also suggests that development of T2R38 stimulatory pharmaceutical components combined with delivery devices might provide an attractive therapeutic option to augment natural host responses in the treatment of CRS. The use of secreted bacteria-derived products for treatment in adult CRS patients is empirically recommended and is also supported by the EPOS 2020 with level 1b evidence [20,51]. On the other hand, no significant difference was observed in T2R14 expression levels among the groups. T2R14 is one of the highly expressed T2R isoforms in nasal and lung epithelial cells [52]. T2R14 also responds to AHL quorum-sensing molecules produced by Gram-negative bacteria [17]. Further study is required to understand how different T2R isoforms’ responses are modulated within the inflammatory milieu of CRS pathology. The most well-known and well-characterized example is the T2R38 isoform [7,29]. The TAS2R38 gene has two common polymorphisms, one encoding a functional receptor and the other encoding a nonfunctional receptor [45]. The differences in the resulting proteins are at amino acid positions 49, 262, and 296. The functional T2R38 receptor contains proline (P), alanine (A), and valine (V) residues, while nonfunctional T2R38 contains alanine (A), valine (V), and isoleucine (I). Loss of the valine in the AVI variant is responsible for the impairment of receptor activation [30,53]. The resulting haplotypes influence the perception of bitter taste: PAV–PAV as “supertasters,” PAV–AVI as variable or intermediate-level tasters, and AVI–AVI as “nontasters”. In the study of TAS2R38 genotypes, we found a tendency for a higher prevalence of CRS in subjects with PAV/AVI and AVI/AVI SNP alleles as compared to subjects with the PAV/PAV allele. The proportion of PAV/PAV T2R38 genotype also tended to be lower in CRS patients than in controls. Genetic polymorphisms are common within the taste receptors [9]. T2Rs are genetically diverse, which helps to explain the wide variety of taste preferences both within and among cultures [30]. In addition, several recent linkage studies have demonstrated associations of T2R isoforms genetics with CRS morbidity, including TAS2R13, TAS2R19, TAS2R38, and TAS2R49 [11,12]. We examined associations between the TAS2R38 polymorphisms and phenotypic CRS prevalences in the Japanese population due to the broad clinical implications of extraoral expression of T2R38. These polymorphisms are reported to be distributed in a nearly Mendelian ratio in Caucasian populations [54]. Adappa et al. reported that distributions of AVI/AVI, AVI/PAV, and PAV/PAV alleles were 37%, 54%, and 8.5%, respectively, in CRS patients in Caucasian populations compared with 29%, 51%, and 20%, respectively, in the general regional population in America [12]. Their results were comparable with ours: namely, in our study, AVI/AVI, AVI/PAV, and PAV/PAV alleles were 20.6%, 52.1%, and 27.1%, respectively, in CRS patients as compared to 19.6%, 39.2%, and 41.1% in the control subjects [9,52]. A series of these studies have highlighted the potential relevance of T2R38 in CRS. Individuals who express the fully functional PAV/PAV genotype are less likely to contract upper airway infection by gram-negative bacteria [11,40] or to require surgical intervention [12,55]. This study has several limitations. First, the study included a relatively small number of Japanese patients; hence, caution should be taken when extrapolating our results to other ethnic groups. Further prospective studies on a larger scale are requisite to elucidate the mucociliary clearance function via T2Rs activation. Second, we failed to examine the second sinonasal cell type that expresses bitter receptors, i.e., the solitary chemosensory cell (SCC), because these cells also modulate the epithelial innate immune system [32,39,56]. Third, the degree of functional expression of T2R38 and nasal FeNO levels is a partial element of recalcitrant CRS, which should be viewed in the context of other genetic and environmental influences yet to be clarified. For example, our previous study demonstrated that ECRS patients with genetically longer (CCTTT)n repeat polymorphisms had higher expressions of NOS2 mRNA in ethmoid sinus mucosae with higher FeNO levels in certain clinical manifestations [57]. In conclusion, we have demonstrated that changes in T2R38 expression levels and localization in the sinonasal pathway are associated with specific CRS phenotype (non-ECRS and ECRS) subsets, suggesting the T2R38 pathway as a potential therapeutic target to promote endogenous immune responses in CRS patients. This discovery is likely to have significant clinical impact immediately and to prompt further studies to define the T2R38 signaling pathways as well as to identify other T2Rs that similarly activate immune responses. We conducted a case–control study of 56 patients with non-ECRS and 36 patients with ECRS, all of whom underwent endoscopic sinus surgery. The diagnosis of sinus disease was based on the patient’s history, clinical symptoms, endoscopic findings, and computed tomography (CT) imaging. Patients with a previous sinus surgery were excluded. None of the patients had received topical or systemic steroids for ≥4 weeks prior to the surgery. The CT images were subjected to radiological grading using the Lund-Mackay system [58]. The diagnosis of AR was based on clinical history, presence of nasal symptoms together with positive nasal eosinophils, and positive allergen-specific IgE antibodies. JESREC scoring was used to differentiate ECRS from non-ECRS. The scores include 4 items: bilateral sinus disease, the presence of nasal polyps, the degree of eosinophilia in peripheral blood, and mucosal eosinophil count ≥70/high-power field (HPF) [22]. Fifty-one patients without sinus infection who underwent endonasal surgery served as controls. All controls had paranasal sinus mucosa of normal appearance and normal radiological findings. Oral and nasal FeNO levels were measured before surgery using a handheld electrochemical analyzer (NObreath®, Bedfont Scientific Ltd., Rochester, UK) according to ATS/ERS guidelines [47]. For oral FeNO measurements, subjects were advised to exhale at a flow rate of 50 mL/s through a mouthpiece. For nasal FeNO measurements, subjects were instructed to exhale transnasally with their mouth closed into a nose adaptor as described elsewhere [59]. Each measurement was performed in triplicate, and the mean value was used for analysis. Mucosal specimens were obtained from the ethmoid sinus, nasal polyps (if any), and the inferior turbinate at the time of surgery. When CRS was present bilaterally, specimens were taken from both sides. The specimens were divided and either immersed in RNAlater® solution (Ambion, Austin, TX, USA) for RT-PCR or fixed in 10% neutral buffered formaldehyde for immunohistochemistry. Quantitative PCR analysis was performed on an ABI Prisms 7300 system (Applied Biosystems, Foster City, CA, USA). Cellular RNA was isolated using RNeasy mini kits (Qiagen, Valencia, CA, USA). Total RNA was then reverse-transcribed to cDNA using a high-capacity RNA-to-cDNA kit (Applied Biosystems) according to the manufacturer’s instructions. Gene expression was measured on a real-time PCR system using TaqMan Gene Expression Assays (Thermo Fisher Scientific, Waltham, MA, USA). PCR primers specific for TAS2R14 (Hs00256800_s1) and TAS2R38 (Hs00604294_s1) were used (Thermo Fisher Scientific). Primers for GAPDH (Hs03929097_g1) were used as a reference. Amplifications of the PCR products were quantified by the number of cycles, and the results were analyzed using the comparative cycle threshold (Ct) method (2−ΔΔCt). The quantities of target gene expression are presented as relative rates compared to the expression of the reference gene (ratio: target gene/GAPDH expression). The primary antibodies used were anti-human TAS2R14 rabbit polyclonal antibody (#PA020246; Cusabio, Houston, TX, USA) and anti-human TAS2R38 rabbit polyclonal antibody (#PA023155LA01HU; Cusabio). Surgical tissue specimens embedded in paraffin were sliced into 5 µm thick sections for immunostaining. For antigen retrieval, sections were immersed in Histo VT One (Nacalai Tesque, Kyoto, Japan) at 70 °C for 40 min. The slides were then incubated overnight at 4 °C with the primary antibodies. Color development was achieved using the streptavidin-biotin amplification technique (ChemMate EnVision kit; Dako, Glostrup, Denmark). Peroxidase activity was visualized by diaminobenzidine solution. Sections were counterstained with hematoxylin. Control specimens with IgG1 isotype control were used to verify that the nonspecific binding was not detectable. Consecutive sections were routinely stained with hematoxylin-eosin (HE) for the assessment of mucosal pathology and the degree of eosinophil infiltration. Peripheral whole blood was collected from all patients for genotyping of the single nucleotide variation at rs10246939 in the TAS2R38 gene in chromosome 7. Genomic deoxyribonucleic acid (DNA) was extracted from blood using the PAXgene® Blood DNA kit (Qiagen, Hilden, Germany). The genotypes were assessed using the TaqMan SNP genotyping assay (C___9506826_10, Thermo Fisher Scientific). Then, haplotypes (C/T) and diplotypes (C/C and T/T) were identified and recorded using Applied BiosystemTM StepOnePlus® Real-Time PCR systems (Applied Biosystems). We categorized subjects with two copies of the PAV allele (PAV/PAV) as high tasters and those with only one copy of the PAV allele (PAV/AVI) or no PAV alleles (AVI/AVI) as low tasters [11]. Power and sample size calculations for the study design were performed based on data from previous studies of sinonasal T2R and NOS expression. The G*power program, version 3.1.9.6, was used for estimation (https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower.html (accessed on 2 January 2023)). For multiple comparisons, a screening of the data for differences was first carried out using the Kruskal-Wallis test. If the analysis gave a significant result, a further comparison was done by the Mann-Whitney U-test for the between-group analysis. Fisher’s exact test was used to compare qualitative data. p-values < 0.05 were considered significant. All procedures contributing to this work complied with the ethical standards expressed in the Helsinki Declaration. The study protocol was approved by the Institutional Review Board at the Hiroshima University School of Medicine (Approval No. Hi-136-2). Written informed consent was obtained from all patients prior to their participation.

PMC10002786

Pablo Cores Ziskoven,Andressa V. B. Nogueira,Lorena S. Gutierrez,Jens Weusmann,Sigrun Eick,Nurcan Buduneli,James Deschner

Apelin Enhances the Effects of Fusobacterium nucleatum on Periodontal Ligament Cells In Vitro

01-03-2023

apelin,APJ,periodontal ligament cells,Fusobacterium nucleatum,periodontitis,obesity

This study aimed to explore effects of Fusobacterium nucleatum with or without apelin on periodontal ligament (PDL) cells to better understand pathomechanistic links between periodontitis and obesity. First, the actions of F. nucleatum on COX2, CCL2, and MMP1 expressions were assessed. Subsequently, PDL cells were incubated with F. nucleatum in the presence and absence of apelin to study the modulatory effects of this adipokine on molecules related to inflammation and hard and soft tissue turnover. Regulation of apelin and its receptor (APJ) by F. nucleatum was also studied. F. nucleatum resulted in elevated COX2, CCL2, and MMP1 expressions in a dose- and time-dependent manner. Combination of F. nucleatum and apelin led to the highest (p < 0.05) expression levels of COX2, CCL2, CXCL8, TNF-α, and MMP1 at 48 h. The effects of F. nucleatum and/or apelin on CCL2 and MMP1 were MEK1/2- and partially NF-κB-dependent. The combined effects of F. nucleatum and apelin on CCL2 and MMP1 were also observed at protein level. Moreover, F. nucleatum downregulated (p < 0.05) the apelin and APJ expressions. In conclusion, obesity could contribute to periodontitis through apelin. The local production of apelin/APJ in PDL cells also suggests a role of these molecules in the pathogenesis of periodontitis.

Apelin Enhances the Effects of Fusobacterium nucleatum on Periodontal Ligament Cells In Vitro This study aimed to explore effects of Fusobacterium nucleatum with or without apelin on periodontal ligament (PDL) cells to better understand pathomechanistic links between periodontitis and obesity. First, the actions of F. nucleatum on COX2, CCL2, and MMP1 expressions were assessed. Subsequently, PDL cells were incubated with F. nucleatum in the presence and absence of apelin to study the modulatory effects of this adipokine on molecules related to inflammation and hard and soft tissue turnover. Regulation of apelin and its receptor (APJ) by F. nucleatum was also studied. F. nucleatum resulted in elevated COX2, CCL2, and MMP1 expressions in a dose- and time-dependent manner. Combination of F. nucleatum and apelin led to the highest (p < 0.05) expression levels of COX2, CCL2, CXCL8, TNF-α, and MMP1 at 48 h. The effects of F. nucleatum and/or apelin on CCL2 and MMP1 were MEK1/2- and partially NF-κB-dependent. The combined effects of F. nucleatum and apelin on CCL2 and MMP1 were also observed at protein level. Moreover, F. nucleatum downregulated (p < 0.05) the apelin and APJ expressions. In conclusion, obesity could contribute to periodontitis through apelin. The local production of apelin/APJ in PDL cells also suggests a role of these molecules in the pathogenesis of periodontitis. Periodontitis is a chronic inflammatory disease mainly caused by a subgingival dysbiotic microbiota whose balance is shifted by several factors [1]. Additionally, there is also a dysbiotic status between the host and the subgingival microbiota in periodontitis. Hyperinflammatory immune responses of the host to this microbiota can lead to alveolar bone resorption and eventually tooth loss [1]. Risk factors such as smoking or genetic predisposition can contribute to the initiation and progression of periodontitis [2]. There is strong evidence that periodontitis is associated with systemic diseases and conditions, such as diabetes mellitus, cardiovascular disease, hypertension, obesity, and metabolic syndrome. It is thought that the oral microorganisms, their components, or metabolites as well as inflammatory mediators get into the systemic circulation and therefore to other parts of the human body [3,4,5,6,7,8]. Obesity is defined as abnormal or excessive fat accumulation that presents a risk to health [9]. Because adipose tissue is not only an energy reservoir but also a metabolic organ, dysregulation of cytokines, hormones, and metabolites occurs when this tissue increases [10]. There is evidence that obese individuals have systemically high levels of CRP, TNF-α, and IL-6 in comparison to normal-weight subjects and, therefore, are in a chronic subclinical inflammatory state [11]. A lot of possible pathomechanisms have been suggested to be responsible for the link between periodontitis and obesity, such as adipokines [12]. Adipokines are cytokines produced by adipocytes, but also by other cell types, such as periodontal cells [13,14,15,16,17,18]. Various adipokines such as leptin, visfatin, adiponectin, and resistin have been identified and studied in regard to systemic diseases. It is suggested that these adipokines have a wide range of functions, which include regulation of insulin metabolism, thirst and hunger sensation, angiogenesis, energy balance, bone metabolism, coagulation, and hematopoiesis, as well as inflammation and its resolution [13,19]. Adiponectin has mainly anti-inflammatory effects, whereas resistin, visfatin, and leptin are more pro-inflammatory [20,21]. Another adipokine, which has been rather less studied so far, is apelin. Apelin was first isolated and described in 1998 [22]. As early as 1993, the apelin receptor (angiotensin II protein J receptor, APJ) had been discovered in humans as a G protein-coupled receptor whose gene locus is located on chromosome 11 [23]. Apelin has a wide range of effects, which differ depending on cell types and tissues. Originally, apelin was isolated from tissues of the central nervous system. Accordingly, the molecule was found to be important in central signal transduction [24]. As research progressed, the apelin-APJ system was discovered in other tissues as well. For example, the molecule interferes with the regulation of bone turnover by modulating apoptosis, proliferation, and differentiation of osteoblasts [25,26]. It has been shown that apelin levels are increased in systemic diseases and conditions such as obesity and diabetes [27,28]. A recent study looked at serum levels of apelin in diabetes and/or periodontitis patients [29]. Those patients who suffered from both diabetes and periodontitis exhibited the highest serum levels of apelin as compared to healthy individuals. Another study could show that the salivary apelin levels of diabetic patients with periodontitis were increased as compared to healthy individuals [30]. This adipokine also has modulatory properties regarding inflammation. For example, apelin can increase the expression of TNF-α and IL-1β in glial cells, but at the same time downregulate inflammatory mediators in lung and heart cells [31,32]. Therefore, apelin could be a critical molecule, which may mediate the harmful effects of obesity on periodontal tissues. The aim of this in vitro study was to explore the regulatory effects of Fusobacterium nucleatum in the presence or absence of apelin on periodontal ligament (PDL) cells in order to test the hypothesis that apelin might be one of the pathomechanistic links between periodontal disease and obesity. First, we wanted to verify whether F. nucleatum would regulate the expression of COX2, CCL2, and MMP1 in PDL cells. F. nucleatum caused a significant (p < 0.05) and dose-dependent (O.D.660: 0.000, 0.025, 0.050, and 0.100) upregulation of the pro-inflammatory and proteolytic molecules COX2, CCL2, and MMP1 with the highest expression for the highest bacterial concentration (O.D.660 = 0.100) at 24 h (Figure 1a). In addition, the stimulatory effect of F. nucleatum (O.D.660 = 0.025) on these molecules was also time-dependent (p < 0.05), as shown in Figure 1b. Next, we studied whether apelin (1 ng/mL) could modulate the stimulatory actions of F. nucleatum (O.D.660 = 0.025) on the expression of pro-inflammatory markers in PDL cells. Apelin was used at a concentration corresponding to physiological plasma levels and consistent with previous in vitro studies. For F. nucleatum, O.D.660 = 0.025 was chosen because even this minimal dose had a proinflammatory effect on PDL cells, as evidenced by a significant increase in the expression of COX2, CCL2, and MMP1. As shown by real-time PCR analysis, apelin significantly (p < 0.05) increased the F. nucleatum-stimulated expression of CCL2 at 24 h (Figure 2a). For COX2, CXCL-8, and TNF-α, no significant modulatory effect of apelin on the F. nucleatum-triggered expression was observed at this time point (Figure 2a). Moreover, apelin caused a further significant (p < 0.05) elevation of the F. nucleatum-induced expressions of COX2, CCL2, CXCL-8, and TNF-α at 48 h (Figure 2b). This shows that the stimulatory influence of apelin on the effects of F. nucleatum was stronger at 48 h as compared to 24 h. We then examined the effect of apelin (1 ng/mL) on the regulation of MMP1, TGF-β1, and RUNX2 by F. nucleatum (O.D.660 = 0.025) in PDL cells (Figure 3). F. nucleatum increased the expression of MMP1 at 24 h (Figure 3a) and 48 h (Figure 3b), and this upregulation was significantly (p < 0.05) enhanced by apelin at both time points. No upregulation by F. nucleatum was observed for TGF-β1 and RUNX2 at 24 h (Figure 3a) and 48 h (Figure 3b). Apelin had no significant effect on the actions of F. nucleatum on TGF-β1 at 24 h (Figure 3a) and 48 h (Figure 3b) and RUNX2 at 48 h (Figure 3b). Interestingly, apelin significantly (p < 0.05) counteracted the inhibitory effect of F. nucleatum on RUNX2 expression at 24 h (Figure 3a). We next sought to identify intracellular signaling pathways potentially involved in the actions of F. nucleatum on CCL2 and MMP1 in PDL cells. For this purpose, cells were pre-incubated with specific inhibitors for NF-κB or MEK1/2 signaling and subsequently stimulated with F. nucleatum (O.D.660 = 0.025) and/or apelin (1 ng/mL). Pre-incubation of cells with an NF-κB inhibitor resulted in a significant (p < 0.05) downregulation of the CCL2 expression in cells treated with either F. nucleatum alone or in combination with apelin at 24 h (Figure 4a). In contrast, the expressions of CCL2 and MMP1 induced by F. nucleatum and/or apelin were always significantly (p < 0.05) inhibited by the MEK1/2 inhibitor after 24 h (Figure 4a,b). We also investigated whether apelin is expressed in PDL cells and, if so, whether this adipokine as well as its receptor are regulated by F. nucleatum (O.D.660 = 0.025). The periodontopathogen downregulated (p < 0.05) the expression of apelin and APJ over a variety of doses (Figure 5a). A slight time dependence was observed (Figure 5b). Finally, we investigated whether apelin (1 ng/mL) can modulate the stimulatory effect of F. nucleatum (O.D.660 = 0.025) on pro-inflammatory markers also at protein level in PDL cells. As detected by ELISA, F. nucleatum resulted in increased protein levels of CCL2 and MMP1 in cell supernatants at 24 h and 48 h (Figure 6a,b). Incubation of F. nucleatum-stimulated cells with apelin resulted in a further significant (p < 0.05) increase in protein levels of CCL2 at 48 h (Figure 6a) and of MMP1 at 24 h and 48 h (Figure 6b). This study aimed to investigate the modulatory effect of the adipokine apelin on the action of the periodontopathogen F. nucleatum on PDL cells to better understand the relationship between periodontitis and obesity. Interestingly, apelin was able to modify bacterial regulation of molecules related to inflammation and hard and soft tissue turnover. The combination of F. nucleatum and apelin resulted in the highest expression levels of pro-inflammatory and proteolytic molecules, suggesting that apelin may be a pathomechanistic link mediating deleterious effects of obesity on periodontal tissues. In addition, F. nucleatum caused downregulation of the expression of apelin and its receptor, suggesting a role of these molecules in the pathogenesis of periodontitis. There is strong evidence for an association between periodontitis and obesity [12,33]. It has been shown in several studies of our research group that adipokines represent a possible pathomechanistic link underlying the association between periodontitis and obesity [15,16,17,18,34,35,36,37]. Leptin, visfatin, and resistin exert pro-inflammatory effects on periodontal cells and tissues, whereas adiponectin has rather protective effects on periodontal cells [33]. However, with respect to the periodontium, almost nothing is known about the production, regulation, and action of apelin, another adipokine whose serum levels are altered in obesity [28]. Recently, Hirani et al. investigated the serum level of apelin in periodontally and systemically healthy individuals and periodontitis patients with and without type 2 diabetes [29]. The study showed that apelin levels were higher in the periodontitis group compared with the healthy control. When patients had concomitant periodontitis and obesity, apelin levels were highest. The authors concluded that the increased expression of apelin in patients with periodontitis and type 2 diabetes might indicate a possible role of this adipokine in inflammation and glucose regulation. Sarhat et al. examined the salivary apelin levels of periodontally diseased diabetic patients and of periodontally and systemically healthy individuals [30]. They also found the highest apelin levels in periodontitis patients with diabetes. In our study, the periodontopathogen F. nucleatum led to a dose- and time-dependent upregulation of pro-inflammatory and proteolytic molecules. Interestingly, apelin caused an increase in the F. nucleatum-stimulated expression of these pro-inflammatory and proteolytic molecules. In this respect, our in vitro data confirm that apelin may be associated with inflammation. Lee et al. also investigated the relationship between apelin and periodontitis and found a decrease in apelin expression in gingival tissues from periodontitis patients, which is in contrast to the aforementioned studies [38]. Moreover, overexpression of apelin or treatment with exogenous apelin suppressed TNF-α-stimulated gene expressions of MMP1, IL-6, and COX2 in PDL cells [38]. Further studies are needed to clarify whether apelin levels in gingiva, sulcus fluid, saliva, and serum are increased or decreased in gingivitis and periodontitis, and whether apelin exerts pro- or anti-inflammatory effects. In addition, it should be investigated whether periodontal therapy results in a change in these apelin levels. Furthermore, we were interested in whether apelin and its receptor are expressed in periodontal cells, and if so, whether this expression can be regulated by F. nucleatum. Our in vitro experiments with PDL cells showed that both apelin and its receptor are constitutively produced in these cells. Moreover, our experiments revealed that the periodontopathogen F. nucleatum inhibited the expression of apelin and its receptor. In the study by Lee et al., incubation of PDL cells and gingival fibroblasts with the inflammatory mediator TNF-α also resulted in downregulation of apelin [38]. Therefore, this and our study suggest that the apelin-APJ system is downregulated during periodontal infection and inflammation, at least initially. Because our results suggest that apelin exerts rather pro-inflammatory effects, the initial downregulation of apelin and its receptor may represent the host tissues’ attempt to limit inflammation and associated tissue destruction. However, our experiments also showed that this possibly tissue-protective downregulation of apelin and its receptor was no longer observed after 48 h, which may suggest that in persistent periodontal infection, the apelin-APJ system may be of critical importance in the pathogenesis of periodontitis. F. nucleatum is an obligate anaerobic gram-negative bacterium very prevalent in the subgingival biofilm and associated with the etiopathogenesis of periodontitis [39,40]. Infection with F. nucleatum alone has shown to cause alveolar bone loss in a murine experimental periodontitis [41]. When in combination with T. forsythia or P. gingivalis, F. nucleatum synergistically stimulated the host immune response and induced alveolar bone loss in this experimental periodontitis model [42,43]. F. nucleatum, such as other red complex bacteria, is associated with periodontitis [44]. As expected according to our previous studies [45,46,47], F. nucleatum led to increased expressions of pro-inflammatory and proteolytic molecules, underlining the special role of this bacterium in periodontal inflammation and destruction. As in our previous studies, F. nucleatum was used as lysate, so several factors may have been responsible for the observed stimulatory effects of F. nucleatum. Studies using live F. nucleatum or even biofilms consisting of a variety of different bacteria should be performed in the future to confirm the results of this study. Our study clearly demonstrates that apelin can exert pro-inflammatory effects and thus enhance periodontal inflammatory processes. Although there are numerous publications on anti-inflammatory and thus protective effects of apelin [48,49], there are also studies that have demonstrated pro-inflammatory effects of apelin [32,50]. Our analyses regarding intracellular signal transductions suggest that pro-inflammatory effects of F. nucleatum and/or apelin are realized at least partially through MAPK and NF-kB. Further studies should clarify which other intracellular signaling pathways apelin uses for its modulatory effects. Our results are thus in agreement with other studies that have also shown that apelin uses the MAPK and NF-kB signaling pathways, among others, for its effects [27,48,51,52]. In the present study, apelin and APJ were also shown to be produced in periodontal cells and regulated by periodontal pathogenic bacteria, suggesting that apelin and APJ may play an important role in the pathogenesis of periodontitis. Interestingly, F. nucleatum led to downregulation of apelin and its receptor in PDL cells, which would imply an anti-inflammatory effect in accordance with the other results of this study. However, because the inhibitory effect of F. nucleatum was lost with increasing duration of bacterial incubation, this protective effect might also be lacking in persistent periodontal infection. Future studies should also address the apelin-APJ system in other cells of the periodontium, e.g., gingival epithelial cells, and fibroblasts. The increased production of apelin by periodontal cells after bacterial stimulation suggests that this adipokine is increased in saliva, sulcus fluid, gingiva, and serum during periodontal inflammation. Clinical studies of experimental gingivitis and periodontitis as well as periodontal therapy, i.e., intervention, should further clarify the role of apelin locally in the periodontium but also systemically for the whole organism. In summary, within its limitations, our in vitro study demonstrated that the adipokine apelin is able to modulate the effects of F. nucleatum on molecules associated with inflammation and hard and soft tissue turnover. Apelin was able to further increase the expression of pro-inflammatory and proteolytic molecules induced by F. nucleatum, which may suggest that apelin may be a pathomechanistic link mediating the deleterious effects of obesity on periodontal tissues. In addition, our study revealed that PDL cells express apelin and APJ and that these expressions are inhibited by F. nucleatum, suggesting a possible role for this adipokine and its receptor in the pathogenesis of periodontitis. A human PDL cell line PDL26 was used for cell culture. As described previously, this cell line was obtained from a third molar tooth of a healthy, 26-year-old non-smoking patient [47]. Cells were first cultured in cell culture flasks provided with nutrient medium. The culture medium was Dulbecco’s Modified Eagle Medium (DMEM) GlutaMAX (Invitrogen, Karlsruhe, Germany) supplemented with 10% fetal bovine serum (FBS, Invitrogen). Furthermore, 100 units of penicillin and 100 μg/mL streptomycin (Invitrogen) were added to the medium. Cells were maintained in the incubator at 37 °C and with a humidified atmosphere of 5% CO2. Cells were cultured (1 × 105 cells/well) on 6-well culture plates and grown until 70–80% confluence. The medium was changed every other day and 24 h before stimulation; the FBS concentration was reduced to 1%. The periodontopathogenic bacterium F. nucleatum ATCC 25586 was used at different concentrations (optical density, O.D.660 = 0.025, 0.050, and 0.100) to simulate microbial infection in vitro. The bacterial strain was pre-cultivated on Schaedler agar plates (Oxoid, Basingstoke, UK) in an anaerobic atmosphere for 48 h. Successively, bacteria were suspended in phosphate-buffered saline (O.D.660 = 1, corresponding to 1.2 × 109 bacterial cells/mL) and submitted twice to ultrasonication (160 W for 15 min) leading to total killing. Furthermore, apelin (recombinant human apelin protein, Abcam, Cambridge, United Kingdom) was used for in vitro stimulation at a concentration corresponding to physiological plasma levels (1 ng/mL) and consistent with previous in vitro studies [53,54,55]. In addition, cells were pre-incubated with PDTC (10 µM, Cell Signaling Technology, Danvers, MA, USA), a specific inhibitor of NF-κB, and U0126 (10 µM, Calbiochem, San Diego, CA, USA), a specific inhibitor of MEK1/2 signaling. Untreated cells served as control. RNA isolation was performed using RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. To determine the RNA concentration, the spectrophotometer NanoDrop ND-2000 (Thermo Fischer Scientific, Waltham, MA, USA) was used. Five hundred ng of total RNA was reverse transcribed using iScrip Select cDNA Synthesis Kit (Bio-Rad Laboratories, Munich, Germany) according to manufacturer’s protocol. Gene expression analysis of apelin and its receptor (APJ), C-C motif chemokine ligand 2 (CCL2), cyclooxygenase-2 (COX-2), C-X-C motif chemokine ligand 8 (CXCL8), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), matrix metalloproteinase 1 (MMP1), runt-related transcription factor 2 (RUNX2), transforming growth factor-beta 1 (TGF-β1), and tumor necrosis factor alpha (TNF-α), was performed by real-time PCR using the PCR thermal cycler CFX96 (Bio-Rad Laboratories), SYBR green PCR master mix (QuantiFast SYBR Green PCR Kit, Qiagen), and specific primers (QuantiTect Primer Assay, Qiagen). One µL of cDNA was mixed with 12.5 µL master mix, 2.5 µL primer, and 9 µL nuclease-free water. The mix was heated at 95 °C for 5 min, followed by 40 cycles of denaturation at 95 °C for 10 s, and a combined annealing/extension step at 60 °C for 30 s. Data were analyzed by comparative threshold cycle method. The protein levels of CCL2 and MMP1 in the cell supernatants were measured using commercially available ELISA kits (DuoSet, R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions. The optical density was determined using a microplate reader (BioTek Instruments, Winooski, VT, USA) set to 450 nm. The readings at 450 nm were subtracted from the readings at 540 nm for optical correction as per manufacturer’s recommendation. Cell numbers were checked and there was no significant difference between groups. The statistical analysis was performed using the software GraphPad Prism (version 9.2.0, GraphPad Software, San Diego, CA, USA). For data analysis, mean values and standard errors of the mean (SEM) were calculated. Data were checked for normal distribution and, subsequently, analyzed with the t-test (parametric) or Mann–Whitney-U test (non-parametric). For multiple comparisons, ANOVA or the Kruskall–Wallis test was applied, depending on normal distribution. The Dunnet’s (parametric) or Dunn’s test (non-parametric) served as post hoc tests. The significance level was set at p < 0.05 for all experiments.

PMC10002853

Giulia Semenzato,Sara Del Duca,Alberto Vassallo,Angela Bechini,Carmela Calonico,Vania Delfino,Fabiola Berti,Francesco Vitali,Stefano Mocali,Angela Frascella,Giovanni Emiliani,Renato Fani

Genomic, Molecular, and Phenotypic Characterization of Arthrobacter sp. OVS8, an Endophytic Bacterium Isolated from and Contributing to the Bioactive Compound Content of the Essential Oil of the Medicinal Plant Origanum vulgare L.

02-03-2023

plant microbiota,endophytes,volatile organic compounds,genome,essential oil

Medicinal plants play an important role in the discovery of new bioactive compounds with antimicrobial activity, thanks to their pharmacological properties. However, members of their microbiota can also synthesize bioactive molecules. Among these, strains belonging to the genera Arthrobacter are commonly found associated with the plant’s microenvironments, showing plant growth-promoting (PGP) activity and bioremediation properties. However, their role as antimicrobial secondary metabolite producers has not been fully explored. The aim of this work was to characterize the Arthrobacter sp. OVS8 endophytic strain, isolated from the medicinal plant Origanum vulgare L., from molecular and phenotypic viewpoints to evaluate its adaptation and influence on the plant internal microenvironments and its potential as a producer of antibacterial volatile molecules (VOCs). Results obtained from the phenotypic and genomic characterization highlight its ability to produce volatile antimicrobials effective against multidrug-resistant (MDR) human pathogens and its putative PGP role as a producer of siderophores and degrader of organic and inorganic pollutants. The outcomes presented in this work identify Arthrobacter sp. OVS8 as an excellent starting point toward the exploitation of bacterial endophytes as antibiotics sources.

Genomic, Molecular, and Phenotypic Characterization of Arthrobacter sp. OVS8, an Endophytic Bacterium Isolated from and Contributing to the Bioactive Compound Content of the Essential Oil of the Medicinal Plant Origanum vulgare L. Medicinal plants play an important role in the discovery of new bioactive compounds with antimicrobial activity, thanks to their pharmacological properties. However, members of their microbiota can also synthesize bioactive molecules. Among these, strains belonging to the genera Arthrobacter are commonly found associated with the plant’s microenvironments, showing plant growth-promoting (PGP) activity and bioremediation properties. However, their role as antimicrobial secondary metabolite producers has not been fully explored. The aim of this work was to characterize the Arthrobacter sp. OVS8 endophytic strain, isolated from the medicinal plant Origanum vulgare L., from molecular and phenotypic viewpoints to evaluate its adaptation and influence on the plant internal microenvironments and its potential as a producer of antibacterial volatile molecules (VOCs). Results obtained from the phenotypic and genomic characterization highlight its ability to produce volatile antimicrobials effective against multidrug-resistant (MDR) human pathogens and its putative PGP role as a producer of siderophores and degrader of organic and inorganic pollutants. The outcomes presented in this work identify Arthrobacter sp. OVS8 as an excellent starting point toward the exploitation of bacterial endophytes as antibiotics sources. Antimicrobial resistance is a worldwide issue associated with high morbidity and mortality. Conventional antimicrobials are no longer able to overcome bacterial pathogens with a multidrug-resistant profile, resulting in difficult-to-treat or even untreatable infections [1]. The insufficiency of effective drugs and inefficient prevention measures have engaged prescribers and researchers in the development of novel treatment possibilities and alternative antimicrobial therapies [1]. In this context, medicinal plants can play an important role in the discovery of new natural bioactive compounds with antimicrobial activity. For thousands of years, plants have been the primary source of compounds with biological activities. Medicinal and aromatic plants have been widely used for the treatment and the prevention of various human diseases, such as cardiovascular, neurodegenerative, and hepatic diseases, skin disorders, and bacterial and viral infections [2,3,4,5,6]. In the last few decades, the research on natural products of plant origin has been almost fully replaced by drug design and combinatorial chemistry due to the time- and resource-consuming procedures that are required for the isolation, identification, and development of a product with pharmaceutical applications [7]. However, while chemical synthesis creates random molecules, plant secondary metabolites are produced as a result of an adaptation process and might be more efficient as therapeutic molecules in the battle against multidrug-resistant pathogens [7]. These bioactive compounds can also strongly affect the plant-associated microbiome and its physiological traits. It has been shown that the composition and structure of the microbial communities of the roots and the rhizosphere microenvironments change in response to the production of specific root exudates secreted by the plant as a result of microorganisms exposure [8,9,10]. Moreover, there is evidence of the different distribution of bacterial communities inside the internal tissues of the same medicinal plants, suggesting the existence of selective forces able to compartmentalize the microbes into specific microenvironments, which have been attributed to antagonistic interactions between microorganisms and their specific adaptation to the plant’s secondary metabolites [11,12,13,14,15]. Large-scale comparative genomic analyses also revealed that plant-associated bacteria have gained specific metabolic attributes which allow their adaptation to plant anatomical parts [16]. However, how secondary metabolites influence the relative abundance of a particular bacterial genus/species/strain and model the microbial community remains unclear. A large number of medicinal plants have been characterized for their pharmacological properties and applications, but members of their microbiota are also able to synthesize bioactive molecules, both soluble and volatile ones [11,14,17,18,19]. Plants rely on their microbiome for specific traits, including nutrient acquisition, growth promotion, induced systemic resistance, and tolerance to abiotic stress factors [20]. Endophytes are considered an almost untapped source of natural compounds with potential therapeutical properties. Egamberdieva et al. [21] showed that plants whose extracts exhibited the highest antibacterial properties also hosted bacterial endophytes able to exert a comparable action toward bacterial pathogens. This confirms the evolutionary adaptation of endophytes to the plant microenvironments and identifies medicinal plants as an excellent starting point to isolate microorganisms able to produce biologically active molecules. Among these bacteria, strains belonging to the genera Arthrobacter, Gram-positive bacteria of the family Micrococcaceae, are commonly found associated with the plant’s microenvironments: they are regular inhabitants of the rhizosphere [22,23], but they have also been isolated from the surface [24] and the inner tissues [25] of plants. Bacteria of the Arthrobacter genus often establish a beneficial relationship with their host plants through various mechanisms of action. Many Arthrobacter strains fulfill a plant growth-promoting activity, mostly due to the production of indole-3-acetic acid and siderophores, which is able to increase the solubility of various minerals and enhance nutrient acquisition [22,23,26]. In addition, many works focused on the interesting ability of Arthrobacter species to degrade organic and inorganic compounds which they utilize as substrates for their metabolism, thus making Arthrobacter genus a promising tool for plant-based bioremediation [24,27]. The Arthrobacter genus can be defined as a metabolically versatile group of bacteria, but their role as antimicrobial secondary metabolite producers has not been fully explored. Some plant-associated Arthrobacter strains, however, were able to antagonize both plant and human pathogens [28,29,30]. Taken together, these features highlight the importance of Arthrobacter strains from an ecological perspective and call attention to the possible biotechnological and therapeutical application of this bacterial genus. In a recent work on Origanum vulgare L., the culturable endophytic microbiota isolated from different anatomical parts of the plant was characterized through molecular and phenotypic analysis, revealing a high degree of biodiversity at the strain level, different antibiotic resistance phenotypes, and the ability, for some strains, to antagonize the growth of bacteria isolated from the same or different O. vulgare microenvironments and some human pathogens [14]. Among others, Arthrobacter sp. OVS8, isolated from the stem of O. vulgare L., was able to inhibit the growth of various endophytes isolated from different compartments of the plant. Moreover, it efficiently inhibited the growth of ten members of the Burkholderia cepacia complex [14]. The antibacterial activity was attributed to the production of volatile organic compounds (VOCs): it has been demonstrated that some O. vulgare endophytes, including Arthrobacter sp. OVS8, synthesize a plethora of VOCs, some of which belong to the content of O. vulgare essential oil (EO), opening the possibility that (at least) some of the bioactive compounds of the plant EO might be synthesized by the microbiota itself [19]. A better understanding of the pathways and genes involved in the biosynthesis of endophytic secondary metabolites and of the role of such compounds in the interplay between the plant and its associated microbial communities should be strongly encouraged to unravel the mechanisms implied in the plant–microbiota interactions and so enable the prediction of endophytes’ capacity to synthesize novel bioactive secondary metabolites and potential antibacterial drug candidates [31]. Thus, this work aims to delineate molecular and phenotypic features of Arthrobacter sp. OVS8 from different viewpoints in order to evaluate its adaptation and influence on the plant internal microenvironments and its biotechnological potential as a producer of antibacterial volatile molecules. To examine its adaptation to the plant essential oil, Arthrobacter sp. OVS8 was tested for its ability to grow in the presence of diesel fuel and β-caryophyllene as the only carbon and energy sources. The endophyte growth in the presence of diesel fuel was evaluated as the variation of the density of the bacterial streak, measured from 4 (complete growth, as the positive control) to 0 (absence of growth). Its growth was maximal (4) on the Tryptic Soy Agar (TSA) control plate and on Minimum Davis medium (MMD) supplemented with glucose (4), and it was strong (3) on MMD supplemented with diesel fuel. We then assessed the ability of Arthrobacter sp. OVS8 to resist in the presence of β-caryophyllene and/or use it as a carbon and energy source. This sesquiterpene hydrocarbon is one of the main components of O. vulgare essential oil [14]. The strain was plated in the presence of a 6 mm paper disk soaked with different concentrations of the volatile compound, as described in the Materials and Methods section. If the strain can degrade/utilize β-caryophyllene, its growth on minimum medium would only be possible in the proximity of the paper disc soaked with the compound. The experiment was repeated twice, and results are reported in Table 1. Arthrobacter sp. OVS8 was sensible to β-caryophyllene, as its growth was inhibited on Mueller Hinton (MH) agar and on MMD supplemented with glucose, while it did not grow in the presence of β-caryophyllene as the only source of carbon and energy. Since we previously demonstrated that Arthrobacter sp. OVS8 synthesize VOCs able to completely inhibit the growth of bacteria belonging to the Burkholderia cepacia complex (Bcc) [19], it was tested for its ability to produce VOCs able to antagonize the growth of 36 multidrug-resistant human pathogenic strains belonging to the Staphylococcus aureus, Pseudomonas aeruginosa, Klebsiella pneumoniae and Coagulase-Negative Staphylococci (CoNS) groups. The analysis was carried out through the cross-streaking method, using Petri dishes with a septum physically separating the culture plate into two compartments, as described in Materials and Methods. The antibacterial activity of Arthrobacter sp. OVS8 was evaluated qualitatively as the reduction in the density of the target strain streak, measured from 0 (where the strain grew as the positive control) to 4 (absence of growth). The obtained data are shown in Table 2. Of the 36 target strains, Arthrobacter sp. OVS8 VOCs were able to slightly reduce the growth of 7 strains, belonging to the P. aeruginosa and S. aureus groups, while it exhibited a moderate antagonistic effect toward 4 K. pneumoniae strains and 1 of the CoNS strains, namely, S. epidermidis 5321. Given the results obtained through the qualitative cross-streaking, we decided to focus on the K. pneumoniae group. The inhibitory activity of the VOCs emitted by Arthrobacter sp. OVS8 was quantified through a quantitative cross-streaking method, which allows the determination of the number of viable target strain cells at the beginning (t0, soon after their streaking onto plates) and at the end of the experiment (t1, after 48 h incubation in the presence or absence of the tester strain) [19]. The experiment was performed in triplicate according to the protocol detailed in the Materials and Methods section. The obtained results are schematically represented in Figure 1 and Figure 2. Arthrobacter sp. OVS8 was able to significantly reduce the growth of all the K. pneumoniae strains (Figure 1A and Figure 2); indeed, the t-test between the average CFU obtained at t1 (normalized for the initial titer) for treated and non-treated K. pneumoniae strains highlighted a significant difference between the two conditions (p-value < 0.005). Among all the strains, K. pneumoniae ATCC 700603 resulted to be the least sensitive target for the VOCs emitted by the endophyte. The strongest antibacterial activity was reported against K. pneumoniae 4409 and 4420, with a CFU reduction higher than 50% compared with the growth controls (Figure 1B). Data regarding the genome assembly of the sequencing reads performed through the Canu assembler plugin is shown in Table 3. Arthrobacter sp. OVS8 embeds a single contig with an overall length of 4,175,013 bp and a GC content of 67.14%. The genome was annotated using the NCBI GenBank annotation pipeline, revealing the presence of 4273 genes, of which 2598 are coding genes, 15 are rRNAs, 50 are tRNAs, and 3 are ncRNAs. The complete genome sequence is available in GenBank under the accession number CP116670. We also performed an Average Nucleotide Identity (ANI) and a digital DNA–DNA hybridization (dDDH) analyses, as described in Materials and Methods. Data obtained are reported in Supplementary Material (Tables S1–S3 and Figures S1 and S2) and revealed that this strain might potentially be affiliated to a new species. Secondary metabolite biosynthetic gene clusters (BGCs) analysis predicted five types of BGCs (Table 4); however, only one BGC had a percentage of similarity higher than 50% with clusters available in the database. This BGC accounts for the production of desferrioxamine, a non-ribosomal peptide-synthetase (NRPS)-independent siderophore; a 66% of similarity was found with the desferrioxamine BGC from Streptomyces argillaceus. The DuctApe software suite is able to handle annotated genomic data, enabling metabolism reconstruction according to the KEGG database: each protein that has a KEGG annotation in the genome is mapped to its KEGG reaction ID and the corresponding pathway [32]. Given the previous data showing that Arthrobacter OVS8-emitted VOCs are able to induce a bactericidal effect against some human opportunistic pathogens belonging to the Bcc, we focused on the metabolic pathways involved in the production of such VOCs; in particular, to those which are also part of O. vulgare essential oil composition, i.e., α-pinene, p-cymene, and γ-terpinene [19]. The biosynthesis of these volatile compounds, all belonging to the monoterpenes family, contemplates the catabolic activity of various terpene synthases and begins with the ionization of the geranyl diphosphate [33]. As depicted in Figure 3, Arthrobacter sp. OVS8 shows the metabolic potential to produce geranyl diphosphate. However, no enzyme belonging to the subsequent monoterpenoid biosynthetic pathway was identified in the annotated genome of Arthrobacter sp. OVS8 (Figure 4). The phenome of Arthrobacter sp. OVS8 was investigated using the GEN III aerobe microbial identification plates. Data were analyzed with the DuctApe software suite [32] to obtain activity values (AVs) for each well. The AV is a synthetic parameter able to recapture the whole shape of the respiration curve, accounting for five parameters of the curve, namely, the length of the lag phase, the slope, the average height, the maximum cell respiration, and the area under the curve. The GEN III microplate contains 23 chemical sensitivity assays. Based on the AV, we may classify Arthrobacter sp. OVS8 as being able or not to grow under different stresses. For exploration, samples with an AV lower than seven were grouped as having low or absence of growth (i.e., no resistance to the harmful molecule in the sensitivity assay), while those with AVs equal to or higher than seven were considered high growth (i.e., showing resistance to the harmful molecule in the sensitivity assay). Regarding chemo-physical stress assays, Arthrobacter sp. OVS8 was able to grow in the presence of NaCl concentrations up to 4% but showed no growth at 8% NaCl, and it was able to grow at a pH of 6 but showed no growth at pH 5 (Figure 5). Antibiotic resistance of Arthrobacter sp. OVS8, based on the antibiotics included in the GEN III plate, was fairly low. It showed resistance to nalidixic acid and aztreonam, but sensitivity to troleandomycin, rifamycin SV, minocycline, lincomycin, and vancomycin. For other classes of chemical compounds, Arthrobacter sp. OVS8 displayed resistance to 1% sodium lactate, d-serine, lithium chloride, potassium tellurite, and sodium bromate, while it showed sensitivity to fusidic acid, guanidine hydrochloride, Niaproof 4, tetrazolium violet, and tetrazolium blue (Figure 5 and Table 5). The same approach used for the evaluation of results on the sensitivity wells was also used for the C- and N-utilization wells. Samples with AV lower or equal to three were considered as showing absence of growth (i.e., inefficient utilization of the C or N source), while those with AV equal to or higher than seven were considered as high growth (i.e., efficient utilization of the C or N source). Overall, Arthrobacter sp. OVS8 was able to efficiently utilize 23 out of 68 C sources included in the GEN III microplate, while showing low or no growth in the presence of 27 out of 68 C sources included in the GEN III microplate (Figure 5 and Table 6). In addition, Arthrobacter sp. OVS8 showed modest utilization of 18 C sources included in the GEN III microplate, showing AVs between 3 and 7. As far as nitrogen utilization is concerned, the GEN III microplates include a total of four wells with different N sources. Arthrobacter sp. OVS8 efficiently utilized the g-amino-N-butyric acid, but inefficiently utilized D-aspartic acid, glycil-L-proline, and L-aspartic acid. The lack of new compounds able to combat the current issues of antibiotic resistance, along with the emergence of multidrug-resistant pathogens and the continued presence of untreatable diseases, currently represent a challenge for the scientific community in the study of new and effective molecules with antibacterial activity. In fact, only a small fraction of the antibiotics approved over the past 40 years represent new compound classes, while the majority are derived from already known chemical structures, with the most recent new class of antibiotics being discovered during the 1980s [34]. Over the last few decades, and more recently over the SARS-COVID19 pandemic, there has been a remarkable resurgence of interest in natural product research to both prevent the toxicity effects of some medical products and to discover new antibiotic-producing strains [35]. When speaking about natural products, traditional medicine, also known as alternative or complementary medicine, cannot be ignored. The herbs, plants or formulas used in traditional medicine contain a plethora of phytochemicals that function alone or in combination with one another to produce a pharmacological effect. Indeed, many plant-originated drugs were discovered from traditional medicine knowledge, and it has been demonstrated that many of them were discovered thanks to their application in such practices [36]. The renewed interest in medicinal plants and their bioactive secondary metabolites has also put the spotlight on the complex and multifaced world of bacterial endophytes and their intimate and intricate relationship with their hosts. There is evidence of endophytes adapting to the plant microenvironments and their direct or indirect contribution to plant secondary metabolites [37,38,39], so one may ask (i) if the phytochemicals obtained from medicinal plants could be a result of the bacterial metabolisms, and (ii) if such bioactive compounds could be directly obtained from endophytes. In this work, we performed a genomic, molecular, and phenotypic characterization of Arthrobacter sp. OVS8, a promising bacterial endophyte isolated from the medicinal plant O. vulgare L. [14]. In particular, we focused on specific features of the strain: its adaptation to the plant microenvironment and essential oil and its ability to inhibit the growth of some multidrug-resistant bacterial pathogens through the emission of VOCs. O. vulgare essential oil is widely known for its antimicrobial potential [40]; thus, it can be imagined that bacterial endophytes inhabiting the inner microenvironment of such plants might be adapted to the bioactive volatile molecules of which the plant tissues are rich in [12], becoming resistant to them or using them as a carbon and energy source. Shimasaki et al. revealed that tobacco roots, a microenvironment abounding in nicotine and santhopine, were enriched in Arthrobacter strains with the catabolic capacity to detoxify or utilize them as nutrients, further supporting the relevance of plant secondary metabolites in the shaping of the plant’s microbial community with specific metabolic competences [41]. The diesel fuel-degrading potential test revealed the ability of Arthrobacter sp. OVS8 to grow in the presence of diesel fuel as the only carbon and energy source. The GC/MS analysis of the essential oil hydro-distilled from the same O. vulgare plant from which the endophyte was isolated revealed that the major chemical groups consisted of sesquiterpene hydrocarbons (73.5%), monoterpene hydrocarbons (17.6%), oxygenated sesquiterpenes (4.8%), and 3.7% oxygenated monoterpenes (4.8%) [14]. The ability of the endophyte to grow in the presence of diesel fuel could suggest its adaptation to the hydrocarbon components of O. vulgare essential oil, leading to the hypothesis that its composition might represent one of the factors involved in the plant–endophytes symbiosis [13]. For this reason, we set up another experiment using β-caryophyllene, a bicyclic sesquiterpene representing 19.2% of O. vulgare essential oil, as the only carbon and energy source. Sesquiterpenes are a class of terpenes that consist of three isoprene units. Arthrobacter strains isolated from the under-canopy soil of the isoprene-emitting Tectona grandis and Madhuca latifolia trees had a high isoprene tolerance and a great degrading potential toward the compound [42]. Our results suggest that β-caryophyllene does not represent a carbon or energy source for Arthrobacter sp. OVS8. Moreover, according to Ponce et al., the strain can be classified as extremely sensitive (inhibition zone diameter ≥ 20 mm) to β-caryophyllene (100%) [43]. As the strain was isolated from the stem of O. vulgare, it can be hypothesized that the content and composition of essential oil from flowers, leaves and stems differ among anatomical parts [44] or that the endophyte is not intimately associated with plant organs and cells responsible for the production of the essential oil [45]. The hydrocarbon-degrading potential of the endophytes could be then associated with an adaptive mechanism through which the endophyte takes advantage of the abundant long chain aliphatic compounds that constitute the plant tissues, as hypothesized for epiphytic bacteria [46]. Arthrobacter sp. OVS8 was selected amongst the collection of endophytes isolated from O. vulgare because of its antibacterial activity against 10 members of the Burkholderia cepacia complex (Bcc), opportunistic pathogens able to cause severe infections in immunocompromised patients, such as cystic fibrosis patients. This activity was attributed also to the production of VOCs, which induced a bactericidal effect toward 7 out of 10 Bcc target strains, most of which were of clinical origin [19]. The endophyte’s emitted VOCs were tested against a panel of 36 multidrug-resistant human pathogens through the cross-streaking method. Data obtained revealed that target strains used in this work were much more resistant than Bcc strains, and only K. pneumoniae strains’ growth seemed to be affected by Arthrobacter sp. OVS8 VOCs. There is still little evidence on the antibacterial metabolites produced by Arthrobacter species, and most of them refer to strains isolated from the Antarctic or marine environments [47,48]. Further strategies for the identification of new antibiotics produced by unexplored targets (such as bacterial endophytes associated with medicinal plants) are required to ensure the availability of effective drugs and overcome the issue of antimicrobial resistance. The genome of Arthrobacter sp. OVS8 has a total length of 4,175,013 bp and a GC content of 67.14%, which reflects the characteristic high GC content of the genus. Secondary metabolite biosynthetic gene clusters (BGCs) analysis predicted a desferrioxamine BGC, a NRPS-independent siderophore, with a 66% similarity to the desferrioxamine BGC from Streptomyces argillaceus [49]. The production of desferrioxamine is quite common for species belonging to the Streptomyces genus, and it benefits the plant–microorganism interaction by being able to promote plant growth, alleviate oxidative stress, and promote the solubilization of iron and other metals [50,51,52]. Siderophore-producing bacteria play a crucial role in plants’ survival and growth, enhancing metal bioavailability in the rhizosphere [53]. Siderophore production has also been reported for the genus Arthrobacter. A high siderophore-yielding Arthrobacter strain isolated from the wild grass Dichanthium annulatum colonizing an abandoned mine efficiently solubilized iron and increased iron-stress resilience in iron-deficiency-sensitive maize [26]. Moreover, the naturally occurring rhizosphere bacteria Arthrobacter oxydans releases a variety of desferrioxamine-like compounds that induce direct plant growth-promoting effects and increase the mobility and solubility of a great variety of metals and minerals through the formation of soluble element-organic complexes that can move toward the plant roots [23]. Hence, the presence of the desferrioxamine BGC in the Arthrobacter sp. OVS8 genome could suggest a putative plant-growth promoting role of the endophyte in facilitating O. vulgare nutrient acquisition. Given the previous data regarding the ability of Arthrobacter sp. OVS8-emitted VOCs to induce a bactericidal or bacteriostatic effect against some clinical and environmental Bcc isolates, we focused on the metabolic pathways involved in the production of such VOCs; in particular, to three monoterpenes also found in O. vulgare essential oil composition, i.e., α-pinene, p-cymene, and γ-terpinene [19]. In nature, we can find different terpene carbon skeletons. Such abundance is attributed to the large number of the enzyme class of terpene synthases and to their ability to convert the same substrate to multiple products [33]. Even though there is a solid degree of amino acid sequence similarity among plant and fungi monoterpene synthases, this similarity is based more on taxonomic affinities of the plant species rather than the type of compound formed. The best recognized structural motif of the terpene synthase family is an aspartate-rich region, [D/N)DXX(D/E) or DDXXXE], located within 80–120 (bacteria and fungi) or 230–270 aa (plants) of the N terminus [33,54]. However, the discovery and biochemical characterization of bacterial terpene synthases represent a great challenge because, unlike plant and fungal enzymes, bacterial terpene synthases do not exhibit an overall amino acid sequence similarity to those from plants and fungi and display a relatively low degree of sequence similarity to other known bacterial terpene synthases [54]. Through the use of hidden Markov models and protein family database searching methods, various structurally diverse bacterial terpene synthases were identified; most of these were sesquiterpene synthases widely distributed amongst Gram-negative bacteria and the Actinomycetales order, as the geosmin producer Streptomyces genus [54,55]. The genome analysis of Arthrobacter sp. OVS8 revealed the presence of the metabolic pathway leading to the synthesis of geranyl diphosphate, the common precursor of most of the cyclic monoterpenes; however, no monoterpene biosynthetic pathways were highlighted. Given the fact that the endophyte actually produces monoterpenes [19], the absence of the identification of the latter pathway could be attributed to the high diversity existing between the amino acid sequence of plant and bacteria terpene synthases and the lack of bacterial terpene synthase sequences from the Arthrobacter genus. GEN III microplates dissect and analyze the ability of strains to metabolize all major classes of compounds and determine other important physiological properties such as pH, salt, and lactic acid tolerance, reducing power, and chemical sensitivity. It could be assumed that strains metabolizing a wide range of substrates and showing tolerance to high salinity concentrations and acidic pH could be more adapted to changing environmental conditions. Arthrobacter sp. OVS8 was able to grow in a medium containing up to 4% NaCl and a pH of 6, suggesting its ability to tolerate mild changes in abiotic stress factors. The strain showed a high resistance toward 10 out of 23 available chemical sensitivity assay compounds such as D-serine, which exerts its antimicrobial activity by replacing D-alanine residues of the peptidoglycan rigid envelope surrounding the cytoplasmic membrane of bacterial species [56]. Resistance to sodium lactate, a salt of lactic acid widely used as a food preservative able to lower water activity [57], and to lithium chloride, reported to induce hyperosmotic stress [58], could indicate the endophyte’s potential to adapt to changes in water availability and high saline concentration. Potassium tellurite toxicity is due to its ability to act as a strong oxidizing agent over a variety of cell components [59]. Arthrobacter sp. OVS8 growth in the presence of such compounds could hint at the endophyte’s ability to remove toxic tellurite from polluted environments [60]. Lastly, Arthrobacter sp. OVS8 resistance to sodium bromate, which mainly originates when ozonation is adopted to treat bromide (Br−)-containing water, could suggest its role as a potential candidate for bioremediation processes [61]. On the contrary, the strain did not grow in the presence of the disinfectant polyhexamethylene guanidine hydrochloride, which disrupts the cellular envelope causing leakage of the cytoplasmic content [62], and the surfactant Niaproof 4. Arthrobacter sp. OVS8 was sensitive to most of the antibiotics present in GEN III plates: the protein synthesis inhibitors fusidic acid, troleandomycin, rifamycin, minocycline, and lincomicyn, as well as vancomycin, which hampers proper cell wall synthesis in Gram-positive bacteria, did not permit the growth of the endophytic strain. As a Gram-positive bacterium, Arthrobacter sp. OVS8 showed resistance in the presence of nalidixic acid and aztreonam, antimicrobial molecules more effective toward Gram-negative strains [63,64]. Arthrobacter sp. OVS8 was isolated from the stem of the medicinal plant O. vulgare L. as described in Castronovo et al. [14]. This strain belongs to a collection of isolates obtained from a pool of O. vulgare plants cultivated in a common garden at the “Giardino delle Erbe”, Casola Valsenio (Italy). Arthrobacter sp. OVS8 was maintained on Tryptic Soy Agar (TSA, Biolife, Milan, Italy) plates for 48 h at 30 °C. Arthrobacter sp. OVS8 was tested against 36 pathogenic strains: 10 CoNS, 10 P. aeruginosa, 10 S. aureus, and 6 K. pneumoniae strains, all characterized by their resistance to multiple antibiotics (as reported in Table 7) [65]. The strains were grown on TSA plates at 37 °C for 24 h. The endophytic strain was tested for its ability to grow in the presence of diesel fuel as the sole carbon and energy source. A single colony of Arthrobacter sp. OVS8 was suspended in 100 μL of 0.9% w/v NaCl solution and then streaked on minimal medium Davis [66] (MMD, 1 g (NH4)2SO4, 7 g K2HPO4, 2 g KH2PO4, 0.5 g Na3-citrate·2H2O, 0.1 g MgSO4·7H2O, pH 7.2, per liter of deionized water) containing 0.4% v/v diesel fuel or 1% w/v glucose as the sole carbon and energy source. Diesel fuel (Esso Italiana, Roma, Italy) was previously filtered through a 0.2 μm-pore-size filter (Sartorius) for sterilization and particle removal. TSA plates were used as growth control. Once streaked, the endophytes were incubated at 30 °C for 3 days. Positivity to this assay was assessed as the presence or the absence of visible growth, expressed as a range from 4 (complete growth) to 0 (absence of growth). The strain was also tested for its ability to grow in the presence of β-caryophyllene as the sole carbon and energy source. A few colonies of the strain were inoculated in 10 mL of Tryptic Soy Broth (TSB, Biolife) and incubated at 30 °C overnight. A dilution of the inoculum was prepared (OD600 = 1) and 100 μL of such suspensions was spread onto different types of agar media: Muller–Hinton Agar (MH, 17.0 g/L agar, 2 g/L beef heart infusion, 17.5 g/L casein acid hydrolysate, 1.5 g/L starch), MMD supplemented with 1% of glucose, and MMD. Immediately after the spreading, sterile filter paper disks (Oxoid S.p.A. Milan, Italy) of 6 mm diameter were placed on the surface of the dishes and soaked with 10 µL of β-caryophyllene and with either a 1:10 or 1:100 dilution of the compound in dimethyl sulfoxide (DMSO) 0.5% v/v. In addition, positive and negative controls were prepared using the antibiotic chloramphenicol (1 mg/mL) (Oxoid S.p.A.) and a solution of DMSO 0.5% in sterile deionized water, respectively. The plates were checked after a 48 h incubation at 30 °C and the diameter of the inhibition zones, including the paper disc diameter, was measured in mm. Arthrobacter sp. OVS8 antibacterial activity was evaluated through cross-streaking, using Petri dishes with a septum separating two compartments to allow the growth of the tester and the target strains without any physical contact. A few colonies of Arthrobacter sp. OVS8 were suspended in 2 mL of a 0.9% NaCl w/v solution, reaching a McFarland turbidity standard of 0.5, corresponding to 1 × 107 CFU/mL. Using a cotton swab, the obtained suspension was streaked on one of the two halves of the Petri dishes. Plates were then incubated at 30 °C for 48 h to allow growth of the strain and the production of volatile organic compounds. A single colony of each target strain was then streaked perpendicularly to the septum using an inoculation needle, and plates were incubated at 37 °C for a further 48 h. Additionally, target strains were streaked on half of a Petri plate in the absence of the tester as a growth control. The antagonistic effect was evaluated at 24 and 48 h and measured as the reduction of the target strains’ growth in the presence of the tester compared to control plates. The different inhibition levels were indicated as follows: complete (4), strong (3), moderate (2), weak (1), and absent (0) inhibition. The antibacterial activity of VOCs synthesized by Arthrobacter sp. OVS8 was quantified through the quantitative cross-streaking method described in Polito et al., 2022 [19], with some modifications. Briefly, a few Arthrobacter sp. OVS8 colonies were suspended in 2 mL of a 0.9% NaCl w/v solution, reaching a McFarland turbidity standard of 0.5. Using a cotton swab, the obtained suspension was streaked on one of the two halves of the Petri dishes. Plates were incubated at 30 °C for 48 h. After two days, a few colonies of each target strain (previously grown at 37 °C for 24 h on TSA plates) were suspended in 2 mL of a 0.9% NaCl w/v solution in order to obtain turbidity corresponding to a 0.5 McFarland standard; then, a 1:100 dilution was prepared. With the aid of a micropipette, 30 μL of the obtained suspensions were applied on the second half of the Petri plate, near the septum. The six drops were then streaked perpendicularly to the septum with an inoculation needle. For the growth control, the same procedure was repeated on Petri plates in the absence of the tester strain. The target suspensions were spread on a TSA plate, using appropriate dilutions to obtain the number of viable cells streaked at the beginning of the experiment (t0). All these plates were incubated at 37 °C for 48 h days. To determine the number of target strains’ viable cells grown in the presence and in the absence of Arthrobacter sp. OVS8 (t1), target cells were recovered in 2 mL of saline solution with a spatula. The suspensions obtained were appropriately diluted, spread onto TSA plates, and incubated at 37 °C for 24 h to determine the viable titer. The number of CFUs obtained at t1 in the presence of Arthrobacter sp. OVS8 was then compared with those obtained for the growth control in the absence of the tester strain and with the number of viable cells streaked at the beginning of the experiment (t0). Phenotype microarray analysis was performed using the Biolog GEN III MicroPlateTM (Catalog No.1030, Biolog). The test provides a phenotypic fingerprint of the microorganism as it contains 71 carbon sources and 23 chemical sensitivity assays. All the reagents used were provided by Biolog, Inc. (Hayward, CA, USA). Arthrobacter sp. OVS8 was grown on TSA at 30 °C for 48 h. The bacterial suspension was prepared by picking bacterial colonies with a sterile inoculation loop and resuspending them in IF-B (Catalog No.72402, Biolog), until 98% transmittance was reached, as assessed with the Biolog turbidimeter (Catalog No.3587). Then, 100 µL of the obtained suspension was dispensed into each well of a Biolog GEN III MicroPlateTM. The plate was incubated at 30 °C in an Omnilog reader (Biolog) for 4 days. The data obtained was visualized using the Biolog Data Analysis software v.1.7. Phenomic data were analyzed using the DuctApe software suite (version 0.18.2) [32] to obtain the activity values (AVs) of each well, and plotted using the ggplate R package (https://github.com/jpquast/ggplate, https://jpquast.github.io/ggplate/, accessed on 30 January 2023). A single colony of Arthrobacter sp. OVS8 was inoculated in 10 mL of fresh Tryptic Soy Broth (Biolife) in a 50 mL tube and incubated at 30 °C overnight under shaking (130 rpm). The following day, bacterial cells were collected by centrifugation (15,500× g for 4 min) and the PowerLyzer PowerSoil DNA Isolation Kit (MO BIO Laboratories, Inc., Carlsbad, CA, USA) was used to extract genomic DNA, following the protocol provided by the manufacturer, with some modifications [31]. Specifically, the cell pellet was resuspended in the PowerSoil Bead Solution in the presence of 1 mg/mL of lysozyme and incubated for 1 h at 37 °C. PowerSoil Solution C1 and 0.5 mg/mL proteinase K were added to the sample, which was then incubated at 55 °C for 2 h before proceeding with the next DNA purification steps. Nanopore sequencing was performed with a PCR-free approach following the native barcoding genomic DNA protocol provided by Oxford Nanopore Technologies (ONT) (version NBE_9065_v109_revY_14Aug2019), as described in Semenzato et al. (2022). The gDNA of Arthrobacter sp. OVS8 was sequenced as follows. Briefly, 1 µg of each input gDNA was repaired and end-prepped using the NEBNext Companion Module for Oxford Nanopore Technologies Ligation Sequencing (E7180S, New England Biolabs, MA, USA). Purification with Agencourt AMPure XP beads (Beckman Coulter, CA, USA) on a magnetic separator followed, and the concentrations of DNA samples were determined using a Qubit dsDNA HS Assay Kit and a Qubit 4 Fluorometer (ThermoFisher Scientific, MA, USA). Then, 500 ng of each end-prepped DNA sample was barcoded using NEB Blunt/TA Ligase Master Mix (M0367, New England Biolabs) and the Native Barcoding Expansion 13–24 (EXP-NBD114, ONT). After purification, equimolar amounts of barcoded DNA samples were pooled to obtain a total of 700 ng and subjected to the adapter ligation. During the subsequent clean-up step, the Long Fragment Buffer included in the Ligation Sequencing Kit (SQK-LSK109, ONT) DNA library was used to enrich the DNA library with >3 kb-long fragments, and it was immediately sequenced. A R9.4.1 Flow Cell (FLO-MIN106D, ONT) was primed with a Flow Cell Priming Kit (EXP-FLP002, ONT). The library was loaded following the instructions provided and sequencing was performed with a MinION MK1B (ONT) and the MinKNOW software v.21.10.4 for 72 h. Base calling and demultiplexing were performed using Guppy v.4.3.4. De novo assembly was achieved using Canu assembler software v.2.1.1 [67] and the quality of contigs was assessed by QUAST v.5.0.2 [68]. These functions were performed in a Galaxy environment (https://usegalaxy.eu, accessed on 3 September 2021). The assembled genome sequence was annotated using the NCBI Prokaryotic Genome Annotation Pipeline (PGAP) v.6.4 (https://www.ncbi.nlm.nih.gov/genome/annotation_prok/, accessed on 20 January 2023). The Average Nucleotide Identity (ANI) analysis was performed using FastANI v.1.3, with default options [69]. The genomic sequences of the genera closely related to the bacterial strain under investigation (Arthrobacter and Pseudarthrobacter) were downloaded from the NCBI “assembly” database and used as reference input for the ANI analysis. The genome sequence of Arthrobacter sp. OVS8 was then uploaded to the Type (Strain) Genome Server (TYGS), a free bioinformatics platform available under https://tygs.dsmz.de, accessed on 30 January 2023, for a whole genome-based taxonomic analysis [70,71]. The results were provided by the TYGS on 22 February 2023. The antiSMASH v.6.0.1 webserver was used for the identification of gene clusters involved in the biosynthesis of secondary metabolites [72]. The query genome was uploaded in a FASTA format; to identify only well-defined clusters containing genes with a significant alignment, the analysis was performed using the strict method of detection. In conclusion, this work aimed at characterizing the newly isolated endophytic strain Arthrobacter sp. OVS8, isolated from the stem of the medicinal plant O. vulgare, to test its ability to synthesize antimicrobial compounds effective against bacterial human pathogens and its adaptation to plant microenvironments. For this purpose, a set of phenotypic and genetic parameters was tested. Cross-streaking experiments revealed that Arthrobacter sp. OVS8-emitted VOCs are able to antagonize the growth of different bacterial pathogens, especially K. pneumoniae strains, which exhibit a multi-drug-resistance phenotype. This is particularly intriguing, considering the worldwide spreading of pathogenic bacteria with multidrug-resistance profiles. Arthrobacter sp. OVS8 association with the plant internal tissues does not seem related to its ability to degrade and/or utilize the essential oil compound β-caryophyllene, but its hydrocarbon degrading potential might suggest an intimate relationship with the plant internal tissues. The phenotype microarray analysis revealed the capacity of Arthrobacter sp. OVS8 to grow in the presence of various chemicals, confirming its ecological role in the degradation of organic and inorganic compounds, thus representing a tool for bioremediation. Finally, genome analysis pointed out the production of siderophores, which might suggest its role as a plant-growth-promoter bacterium, while its genetic features regarding VOCs biosynthesis could not be fully elucidated. Arthrobacter sp. OVS8 represents an excellent starting point toward the exploitation of bacterial endophytes as antibiotics sources.

PMC10002856

Hady Shahin,Sallam Abdallah,Jyotirmoy Das,Weihai He,Ibrahim El-Serafi,Ingrid Steinvall,Folke Sjöberg,Moustafa Elmasry,Ahmed T. El-Serafi

miRNome and Proteome Profiling of Human Keratinocytes and Adipose Derived Stem Cells Proposed miRNA-Mediated Regulations of Epidermal Growth Factor and Interleukin 1-Alpha

04-03-2023

keratinocytes,adipose-derived stem cells,direct co-culture,miRNA,proteome,epidermal growth factor,interleukin 1 alpha,stem cell differentiation

Wound healing is regulated by complex crosstalk between keratinocytes and other cell types, including stem cells. In this study, a 7-day direct co-culture model of human keratinocytes and adipose-derived stem cells (ADSCs) was proposed to study the interaction between the two cell types, in order to identify regulators of ADSCs differentiation toward the epidermal lineage. As major mediators of cell communication, miRNome and proteome profiles in cell lysates of cultured human keratinocytes and ADSCs were explored through experimental and computational analyses. GeneChip® miRNA microarray, identified 378 differentially expressed miRNAs; of these, 114 miRNAs were upregulated and 264 miRNAs were downregulated in keratinocytes. According to miRNA target prediction databases and the Expression Atlas database, 109 skin-related genes were obtained. Pathway enrichment analysis revealed 14 pathways including vesicle-mediated transport, signaling by interleukin, and others. Proteome profiling showed a significant upregulation of the epidermal growth factor (EGF) and Interleukin 1-alpha (IL-1α) compared to ADSCs. Integrated analysis through cross-matching the differentially expressed miRNA and proteins suggested two potential pathways for regulations of epidermal differentiation; the first is EGF-based through the downregulation of miR-485-5p and miR-6765-5p and/or the upregulation of miR-4459. The second is mediated by IL-1α overexpression through four isomers of miR-30-5p and miR-181a-5p.

miRNome and Proteome Profiling of Human Keratinocytes and Adipose Derived Stem Cells Proposed miRNA-Mediated Regulations of Epidermal Growth Factor and Interleukin 1-Alpha Wound healing is regulated by complex crosstalk between keratinocytes and other cell types, including stem cells. In this study, a 7-day direct co-culture model of human keratinocytes and adipose-derived stem cells (ADSCs) was proposed to study the interaction between the two cell types, in order to identify regulators of ADSCs differentiation toward the epidermal lineage. As major mediators of cell communication, miRNome and proteome profiles in cell lysates of cultured human keratinocytes and ADSCs were explored through experimental and computational analyses. GeneChip® miRNA microarray, identified 378 differentially expressed miRNAs; of these, 114 miRNAs were upregulated and 264 miRNAs were downregulated in keratinocytes. According to miRNA target prediction databases and the Expression Atlas database, 109 skin-related genes were obtained. Pathway enrichment analysis revealed 14 pathways including vesicle-mediated transport, signaling by interleukin, and others. Proteome profiling showed a significant upregulation of the epidermal growth factor (EGF) and Interleukin 1-alpha (IL-1α) compared to ADSCs. Integrated analysis through cross-matching the differentially expressed miRNA and proteins suggested two potential pathways for regulations of epidermal differentiation; the first is EGF-based through the downregulation of miR-485-5p and miR-6765-5p and/or the upregulation of miR-4459. The second is mediated by IL-1α overexpression through four isomers of miR-30-5p and miR-181a-5p. Skin is a self-renewing organ that covers the entire surface area of the body. It forms an anatomical guard that works as a barrier from the outer environment. Skin broadly consists of 3 layers: epidermis, dermis, and subcutaneous adipose tissue [1,2]. Because of the complex organization, wound healing involves orchestrated synergy between different skin cells with interplay between a plethora of signaling chemokines, growth factors, and cytokines. The healing process starts with the formation of a fibrin clot, followed by the recruitment of inflammatory cells. Granulation tissue then starts to form alongside angiogenesis. Re-epithelialization takes place with the recruitment of proliferating fibroblasts and migrating keratinocytes causing the dermis to contract [3,4]. Epithelialization is an integral component of wound healing used as a decisive factor of its success. Impaired epithelialization is a characterizing feature of chronic wounds [5]. Keratinocytes, the predominant cellular constituent of the epidermis, play a central role in restoring the epidermis after injury as epithelization is largely mediated by local keratinocytes at the wound edges and by epithelial stem cells in the hair follicles or sweat glands [3,4]. Adipose-derived stem cells (ADSCs), known for their plasticity and low immunogenicity, can be isolated by minimally invasive procedures from subcutaneous adipose tissue with a remarkable yield of ADSCs [6]. These cells have shown potential to differentiate into cell types from different lineages, including osteogenic, chondrogenic, neural, cardiomyogenic, hepatic, endocrinal, as well as epithelial lineages [7,8,9,10]. Additionally, the plasticity and immunomodulatory capacity of ADSCs could increase the chances for the success of cellular transplantation [11]. Establishing sustainable cultures of epidermal cells while eliminating the progressive loss of epidermal stem cell population can be a challenging process due to rapid clonal conversion [12]. As a result, in vitro differentiation protocols were considered to convert ADSCs into epidermal-like cells [13]. Several reports tried to optimize epidermal differentiation protocol with organic and non-organic culture media additives [14,15]. Unfortunately, the effectiveness, reproducibility, and compatibility with the regulations for clinical use are always challenging. Cross-talk within the epidermal niche between keratinocytes, fibroblasts, stem cells, endothelium, immune cells, and other cell types is vital for successful cutaneous wound repair [5]. Keratinocytes release signaling molecules that act in an autocrine and paracrine manner to stimulate the epithelialization through modulating the proliferative and migratory pattern of the surrounding cells. Multiple signaling molecules, including cytokines, chemokines, growth factors, integrins, and coding and non-coding RNAs function as active mediators of cell–cell communication and are crucial for its effectiveness [5,16]. Furthermore, dermal adipocytes and ADSCs play important roles during skin repair via endocrine secretions and signaling [17,18,19,20,21]. Extracellular vesicles (EVs) can promote early stages of healing through pleiotropic effects, including enhancing fibroblasts migratory and proliferative capacity, as well as inducing collagen deposition. EVs derived from ADSCs have been reported to accelerate murine cutaneous wound healing when applied through local and intravenous injections [20]. MicroRNAs (miRNAs) are small non-coding RNAs that are a major constituent of the EVs cargo; miRNAs can regulate gene expression in neighboring cells when released extracellularly, either freely or within Evs. Additionally, miRNAs play a role in all major cellular processes, including metabolism, cell proliferation, differentiation, and apoptosis [2,22]. A single miRNA can regulate several mRNAs at the post-transcriptional level, but several miRNAs can conversely bind to a single gene and cooperatively fine-tune its expression, which attests to the complexity of miRNA-mediated gene regulation [22,23]. Epigenetically, miRNAs can activate/repress gene expression by controlling the rate of transcription or translation [24,25]. Additionally, miRNAs play a central role in epidermal development where various miRNAs have been detected throughout skin cell lineages during embryonic skin morphogenesis [26]. Studies involving conditional depletion of either of the major regulators of miRNAs biogenesis, Dicer and Drosha, in murine epidermis showed the loss of barrier function, aberrations in hair follicle growth, and impaired epidermal differentiation [26,27]. Various studies identified a multitude of miRNAs and their role in regulating key epidermal developmental processes by creating feedback loops that modulate the proliferation, differentiation, and the migration of epidermal cells in both normal and disease conditions [27,28,29,30]. Additionally, miRNAs are involved in the overlapping phases of cutaneous wound healing, including the inflammatory phase, and as regulators of angiogenesis during re-epithelialization [2,31]. In the context of stem cells, miRNAs are involved in the regulatory pathways modulating the differentiation process of ADSCs into various cell lineages including osteogenic, chondrogenic, neuronal and adipogenic differentiation [32,33,34]. In this study, the effect of a direct co-culture model on stem cell differentiation is reported. The model consists of human keratinocytes and ADSCs co-cultivated at equal numbers. Additionally, this study aims to characterize the differential repertoire of miRNAs and proteins in the lysate of cultured human keratinocytes and ADSCs. A combined approach of computational and experimental analyses was employed (Figure 1) in order to establish a miRNome-proteome axis with a focus on potential regulations that can alter the fate of ADSCs to differentiate into keratinocyte-like cells. HaCaT, the epidermal cell line, was considered a positive control (100% group) and ASC52, the immortalized adipose-derived stem cell line, was considered a negative control (0% group). The study group consisted of a mixture of an equal number of both cell lines (50% group) as a direct co-culture system, as shown in Figure 2a–c. After 7 days in culture, gene expression analysis indicated that direct co-cultures of HaCaT with ASC52 expressed higher levels of the transcription factor p63 and the early epithelial marker KRT18 (0.9 folds) while both markers could not be detected in the 0% group (p-value = 0.0004 and 0.01 respectively). The expression of both markers in the co-culture (50% group) was similar to f HaCaT cells (100% group) with a p-value = 0.51 and 0.67, respectively (Figure 2d,e). Similar up-regulation pattern was shown for the basal-specific epidermal marker KRT5 (p-value = 0.0004; Figure 2f). Interestingly the co-culture group showed a trend of upregulation of the basal-specific epidermal marker KRT14 with 0.4 folds over the 100% group (p-value = 0.051; Figure 2g) while it kept the same upregulation pattern in comparison to the 0% group (p-value = 0.0005; Figure 2g). On the contrary, cells at day 7 in a co-culture did not express the late epidermal differentiation markers KRT1 and 10 compared with that of the ASC52 cells in the 0% group (p-value = 0.1883 and 0.1841, respectively; Figure 2h,i). To confirm our initial findings, the expression of the selected epidermal-specific differentiation markers was evaluated with immunocytochemistry (ICC) after 7 days in culture. Monocultures of HaCaT (100% group) and ASC52 (0% group) were used as positive and negative controls, respectively. ASC52 cells in the 0% group exhibited no expression for KRT18 or 5 (Figure 2j,n) and faint expression of KRT14 (Figure 2r). Interestingly, in the 50% group, some of the ASC52 cells, with their distinct “spindle-like” morphology, were clearly expressing the three studied markers, especially the cells in close proximity to HaCaT colonies (Figure 2k,o,s). The relative expression of the immunohistochemical marker (Figure 2m,q,u) showed that cells in the 50% group collectively expressed higher levels of KRT18, 5, and 14 with 0.7, 0.8, and 0.6 folds respectively compared to the 100% group (p-value = 1.73 × 10−9, 0.004 and 1.76 × 10−9 respectively). The 0% group comprising solely of an adipose-derived stem cell line showed the lowest expression levels of KRT18, 5, and 14 with fold changes of 0.1, 0.01, and 0.4, respectively (p-value = 2.95 × 10−16; 1.46221 × 10−12; and 1.21 × 10−3; respectively) compared to the 50% group. miRNAs represent an important vehicle for communication between cells. To evaluate the differentially expressed miRNA, GeneChip® miRNA arrays were used to profile the expression in primary human keratinocytes and ADSCs. A total of 378 miRNAs were identified as differentially expressed miRNAs (DEmiRNAs) between primary keratinocytes and ADSCs (Figure 3a,b; Supplementary Figure S1 and Table S5). Of these, 114 miRNAs (30.16%) were upregulated in keratinocytes while 264 miRNAs (69.84%) were downregulated. Microarray data was validated with qPCR analysis for the selected miRNA targets using individual miRNA assays. The following miRNAs, miR-30b-5p (3.1 fold, p-value = 0.002), miR-30c-5p (2.5 fold, p-value = 0.002), and miR-203a (5018 fold, p-value = 0.03), followed the same pattern of significant differential expression and showed upregulation in keratinocytes. The expression of miR-34a-3p (0.1 fold, p-value = 0.008) showed a significant downregulation in keratinocytes in a similar pattern to that in the microarray analysis. Furthermore, miR-29b-3p (0.6 fold, p-value = 0.12), miR-195-5p (1.1 fold, p-value = 0.48) and miR-374a-5p (0.9 fold, p-value = 0.24) were non-significant in both assays. Our validation of selected individual miRNA revealed overall agreement with the microarray data. The target genes of the identified significant differentially expressed miRNAs (DEmiRNAs) were investigated separately for upregulated and downregulated miRNAs. A total of 659 unique target mRNAs related to 33 upregulated miRNAs in keratinocytes were identified by combining the three target regions, UTRs (3′ and 5′) and CDS (Supplementary Table S6). A similar analysis was performed on the downregulated miRNAs in keratinocytes and a total number of 555 unique target mRNAs for 58 downregulated miRNAs were observed (Supplementary Table S7). To investigate the relation between the upregulated miRNAs and their target genes expression in skin, we explored the Expression Atlas database “https://www.ebi.ac.uk/gxa/home (accessed on 2 December 2022)” and collected the tissue related gene expression data from the FANTOM5 [35] dataset with a cut-off score. The cut-off score (CS) calculated as follows: where : is the expression value of the gene and : is the mean value of total 13,476 genes expressed in Skin, . In this case, a list of 109 unique target genes related to 26 upregulated miRNAs were identified, including AGO2, CDKN1A, MAPK1, MCL1, SEPTIN2, SMAD5, TP53, and TSC1 (Figure 4b; Supplementary Table S8). Additionally, 14 pathways were found to be significantly enriched with this list of 109 genes, including signaling by interleukins, RUNX3 regulated CDKN1A transcription, membrane trafficking, vesicle-mediated transport, and others (Figure 4b; Supplementary Table S9). The protein content in the lysate of primary keratinocytes and ADSCs was experimentally explored, using proteome profiler arrays (Supplementary Table S10). Interestingly, only two proteins were significantly upregulated in keratinocytes, which were Epidermal growth factor (EGF) and Interleukin 1-alpha (IL-1α), as shown in Figure 5a. Integrated analysis between DEmiRNAs and differentially expressed proteins (DEproteins) in the lysate of the studied cell types was conducted. The fisher’s exact test suggested a strong association between a number of DEmiRNAs and the upregulated proteins (p-value = 0.04; Supplementary Table S11). This integrated analysis presented two predictions for miRNA-mediated gene regulations and their protein products. EGF is likely to be directly controlled by the downregulated miRNAs miR-485-5p and miR-6765-5p or the upregulated miR-4459 (Figure 5b). On the other hand, IL-1α can be controlled by 5 of the upregulated miRNAs, 4 isomers of miR-30 (b, c, d, and e)-5p, and miR-181a-5p (Figure 5c). The interaction between keratinocytes and stem cells is not only crucial to understand the wound healing process, but also to identify the key regulators of stem cell differentiation into the epidermal lineage. Culturing keratinocytes alongside ADSCs in monolayer allow for cell–cell contact and communication. In our study, a co-culture of keratinocyte and stem cell lines showed genotypic and phenotypic changes allowed by the cross-communication between cells from the two sources. The expected result was the activation of intracellular signaling cascades that enhanced ADSCs differentiation. The effect of direct co-culture on ADSCs differentiation toward other target cell lineages has been previously reported. For example, ADSCs’ potential for osteogenic differentiation was enhanced when co-cultured at a 50:50 ratio with dental pulp stem cells while the effect was diminished with the use of EVs release inhibitor. In support of our findings, this study confirmed the positive effect of co-culture on enhancing cell differentiation, through cell signaling mediators exchanged between the co-cultured cells [36]. Similarly, direct co-culture has also promoted adipogenesis in ADSCs when seeded at a ratio of 70:30 with umbilical vein endothelial cells and cultured in adipogenic differentiation media [37]. Furthermore, the capacity of ADSCs to differentiate into keratinocyte-like cells was described in another co-culture system where keratinocytes were cultured in a transwell above a monolayer of ADSCs. In this model, secreted molecules and growth factors travel through the pores by gravity and induced the differentiation of ADSCs monolayer. In addition, the authors investigated the effect of keratinocyte conditioned medium on ADSCs. Starting at day 7, ADSCs demonstrated gene and protein expression of epidermal markers KRT5, 14, involucrin, filaggrin, and stratifin, comparable to those of keratinocytes. Furthermore, it was not until day 10, when 20–30% of ADSCs changed their morphology from the typical spindle-like appearance of stem cells into a polygonal morphology resembling that of keratinocytes [38]. In our direct co-culture model, upregulation of p63 expression was detected after 7 days to a level resembling that of HaCaT. This marker can be considered as a key transcription factor for the epidermal lineage, being the first gene product distinguishing epidermal progenitor cells, as well as a prevalently expressed marker in proliferating keratinocytes [39,40,41]. The upregulation of p63 suggests that the ASC52 in the co-culture started the commitment of differentiation into epidermal-like cells. In agreement, a previous study showed nuclear p63 expression at day 7 when cultivating MSCs derived from umbilical cord in keratinocyte-specific media with EGF and calcium [42]. KRT18 is a major component of intermediate filaments that acts as an early epithelial differentiation marker, expressed exclusively in simple epithelium prior to stratification [40,43]. Our gene and protein expression data showed the upregulation of KRT18 in the co-cultured cells in a similar trend to p63 gene expression, which supported the evidence for early epidermal differentiation. KRT18 has been shown in differentiating MSCs induced by a cocktail of growth factors, including KGF, EGF, HGF, and IGF-2 for 14 days and altered their fibroblastic morphology to epithelial-like [44]. As none of these differentiation inducers were added to the culture, the intracellular communication was expected to trigger the same effect. Furthermore, KRT5 and 14 were upregulated in our direct co-culture on both the gene expression and protein levels. Dos Santos et al. (2019) reported that the expression of KRT14 in umbilical cord MSCs cultured in keratinocyte media reached its peak at the first day of cultivation followed by abrupt descend at day 4 and maintained a steady state until day 14 [42]. This could be attributed to the role of KRT14 in sustaining proliferation in mitotically active basal keratinocytes followed by downregulation when cells become committed to differentiation [45]. Undifferentiated ADSCs in monoculture showed expression of the epithelial basal marker KRT14 at the protein level, in agreement with previous studies [46,47]. KRT1 and KRT10 are epidermal stratification markers expressed by differentiated keratinocytes in the suprabasal cutaneous layers, including the stratum corneum [5,41]. Longer differentiation protocol in literature showed upregulation of KRT10 starting at day 11, which could explain the absence of upregulation in our 7-day culture [42]. A temporal expression analysis of various epidermal differentiation markers in our system could be considered as an interesting future analysis. Cell-to-cell communication can occur through several approaches, including receptor-mediated events, direct cell–cell contact, and cell–cell synapses. Often released within EVs, miRNAs are one important mediator in communication, and they disseminate through the extracellular fluid to act as signaling molecules by altering gene expression and protein production in the recipient cell. miRNA-mediated cell-cell communication can be achieved through direct exchange of exosomes between adjacent cells, as well as by shuttling exosomes through the systemic circulation [48,49,50]. Additionally, miRNAs have been shown to regulate various aspects of wound healing including cell proliferation, migration, collagen biosynthesis, and vascularization. Moreover, the field of miRNA-based therapeutics is emerging with vast potential to improve wound healing through targeting of antagomir treatments [51]. In keratinocytes, differential miRNA expression showed that miR-203a was among the most highly expressed. The miRNA miR-203a is one of the most abundant miRNA species in the skin and plays a major role in keratinocytes proliferation and differentiation, alongside the miR-30 family [52,53,54]. On the other hand, miR-34a was downregulated. This result was expected as the cells involved in this study were normal healthy keratinocytes, as the overexpression of miR-34a is known to inhibit keratinocyte proliferation and promote apoptosis [55]. Other major mediators of cell–cell communication are cytokines, which are responsible for a wide range of functions across non-immune cells, including a trophic role in the cell repair and regeneration [56,57]. In the context of stem cell differentiation, chemokines mediate vital cellular processes by establishing the cell communication between proliferating and migrating cells [56]. In this study, both cell types are known for their natural ability to produce cytokines. The secretion can aim at modulating the surrounding tissues in physiological conditions or in response to stimuli, such as cellular stresses imposed by infection, inflammation, tissue damage, or specific culture conditions [58,59]. The expression patterns of a group of cytokines and growth factors as a relevant part of the proteome was explored, with a focus on ADSCs differentiation into epidermal-like cells in response to keratinocytes signaling. Out of the 105 studied proteins, EGF and IL-1α were upregulated in keratinocytes. Keratinocytes are known to both produce and respond to EGF, as it is considered as a major regulator for epidermal homeostasis. The production of EGF by keratinocytes was in agreement with our findings [60,61]. Endogenous growth factors play a major role in orchestrating the proliferative phase in epithelization and are essential for effective wound healing [5,62]. In the epidermis, EGF regulates the barrier function, terminal cell differentiation, cell adhesion, protease secretion, and wound healing [63]. Nevertheless, EGF has long been considered as a crucial additive in epidermal cell culture systems as it stimulates keratinocyte migration and proliferation, as well as ADSCs differentiation into the epidermal lineage [5,42]. Adding the cell culture supernatant of HaCaT cells that are induced to overexpress EGF to ADSCs was associated with enhanced proliferation, migration, and invasion of ADSCs. These findings were abolished when HaCaT were transfected with the EGF inhibitor small interfering RNA (siEGF) [64]. Clinically, the topical application of EGF accelerated healing of split-thickness cutaneous wounds through the stimulation of keratinocytes migration across the wound bed [65]. Epidermal keratinocytes, similar to all epithelial cells with a barrier function, are rich in IL-1α in the physiological state. Upon skin injury, trauma, or infection, IL-1α is released promptly, inducing the local inflammation necessary to initiate wound healing [3,66]. Interestingly, IL-1α is amongst the most frequently reported keratinocyte secretion in culture supernatant, which supports our finding. In skin wounds, IL-1 can mobilize locally located stem cells, as well as enhance the keratinocyte migration [58]. The cluster related to miR-30 is known to be functionally involved in cell fate determination and lineage differentiation of mesenchymal stem cells into adipogenic, chondrogenic, and osteogenic lineages [67]. However, to the best of our knowledge, there are no reports of the direct association of this family to epidermal homeostasis or ADSCs differentiation into epidermal-like cells. The miRNA mir-30a has been shown, in systems other than skin, to block the release of inflammatory cytokines, including IL-1α. However, our results showed the upregulation of miR-30b, c, d, and e and not miR-30a. Different members of miR-30 clusters share a common seed sequence near the 5′ end, but they differ in compensatory sequences near the 3′ end, targeting different genes, and pathways [68,69,70]. The positive regulation of miRNA on proteins can be explained by: (1) miRNA-mediated post-transcriptional upregulation, (2) translation upregulation, or (3) competing with repressive proteins, preventing them from binding to their target sites, leading to increased mRNA stability, thus promoting the expression of the target protein [71]. The upregulation of miR-181a was found to deaccelerate keratinocyte proliferation and promote keratinocyte differentiation when induced by high calcium or UVA irradiation [72,73]. On the other hand, miR-4459 plays a role in decreasing the stemness of human embryonic stem cells through inhibiting its target proteins Cell Division Cycle Protein 20 Homolog B (CDC20B) and Autophagy-Related Protein 13 (ATG13) [74]. To the best of our knowledge, there was no evidence pointing in the direction of miR-4459 mediated EGF regulation in the literature. On the other hand, miR-485 downregulation in human skin has been previously reported, specifically in terminally differentiated keratinocytes [53]; however, the miR-485 mediated EGF interaction in the context of epidermal development or stem cell differentiation has not been reported before, to the best of our knowledge. KRT17 is known to be a direct target of miR-485, while the signaling cascade involving miR-485/KRT17 may result in suppressing EGFR in oral squamous cell carcinoma cell lines [75]. The accumulated evidence, including the miR485-EGF inhibitory regulation shown here, postulate that miR-485 may constitute an appealing target to be investigated in MSCs differentiation into epidermal-like cells. The inhibitory regulation between downregulated miR-6765 and upregulated EGF is another predicted miRNA-target interaction, which should be explored. To the best of our knowledge, this interaction has not been reported in the context of epidermal development, skin repair, or stem cell differentiation. In summary, this study reported a direct co-culture model that can be used to study the cell-to-cell interaction in monolayer, including stem cell differentiation to the epidermal lineage. The integrated analysis of miRNA–protein characterization predicted novel pathways for the regulation of EGF and IL-1α in keratinocytes. The investigation of these pathways may help in providing new concepts for stem cell differentiation into epidermal cells as well as for wound repair. Based on our findings and the known role of IL-1α in regulation of cell differentiation, this cytokine should be studied as a media additive for stem cells differentiation into keratinocytes. The limitations of this study included the use of cell lines in the co-culture experiments rather than primary cells. Obtaining enough primary cell numbers to conduct these experiments would be challenging, which is the reason beyond modelling with cell lines instead. The co-culture duration was limited to only 7 days in order to prove the efficiency of the system and to detect early differentiation. Extension of the culture could help in showing more positive cells for the studied markers as well as the expression of late keratinocyte markers. Another limitation for analyzing the study finding was the vast possibilities of intermediate effectors, potential feedback loops, and the post-translational changes affecting the predicted miRNA-mediated gene regulations. Future studies should focus on the experimental validation of the newly proposed miRNA-mediated gene and protein regulations to provide a better understanding of their associated signaling cascades. Loss/gain function studies for the effect of suggest-miRNA to EGF and IL-1α can provide biological evidence for our computational model. Additionally, miRNA-based approach can be explored for in vitro differentiation of stem cells into epidermal cell lineage, as well as in non-healing in vivo wound models. The application of EGF, IL-1α, or a combination of them may be investigated for therapeutic potential. HaCaT cell line (accession: CVCL_U602, Cellosaurus database) of spontaneously transformed, non-tumorigenic keratinocytes isolated from histologically normal skin (Elabscience Biotechnology Inc., Houston, TX, USA) and ASC52 (accession: CVCL_0038, Cellosaurus database) and hTERT immortalized adipose-derived Mesenchymal stem cells (ATCC, Manassas, VA, USA) were used in this study. Both cell lines were cultured in DMEM (Gibco, Billings, MT, USA) with 10% fetal bovine serum (FBS; Life Technologies, São Paulo, Brazil) and 10,000 units penicillin and 10 mg streptomycin/mL (Sigma-Aldrich, St. Louis, MO, USA). Upon reaching 90% confluence, the cells were dissociated with Trypsin-EDTA (Sigma Aldrich, St. Louis, MO, USA), stained with 0.4% trypan blue (1:1) and counted using a TC20 automated cell counter (Bio-Rad Inc., Singapore). Cells from both cell lines were mixed with a concentration of 1:1 in a direct co-culture system. Control groups were either HaCaT alone as a positive control or ASC52 alone as a negative control. The groups were designated as 0, 50, & 100% representing the percentage of HaCaT in the mixture. Cells were seeded at a density of 4 × 103 cell/cm2 and kept at 37 °C and 5% CO2 for 7 days, with media change every second day. On day 7, cells were either fixed for immunocytochemistry or harvested for RNA extraction and gene expression analysis. Full thickness skin biopsies were obtained from healthy donors during abdominoplasty and/or breast reduction procedures under the ethical approval no. 2015/177-31 by the Swedish Ethical Review Authority. The fat portion was carefully separated. Skin cut into 2–3 mm2 then incubated in 1:1 volume of 10 mg/mL Dispase II solution (Gibco, Tokyo, Japan) overnight at 4 °C. The epidermis was then gently peeled from the dermis and incubated with Trypsin-EDTA (Sigma-Aldrich, St. Louis, MO, USA) on a tube rotator at 37 °C for 30 min. Trypsin was deactivated with media with 10% FBS. The cell suspension was allowed to pass through a 70 µm cell strainer (Corning, New York, NY, USA), and then, the keratinocyte suspension was centrifuged at 700 RCF for 4 min. The cell pellets were washed twice with phosphate buffered saline (PBS) (Life Technologies, Grand Island, NY, USA) before cells were resuspended in keratinocyte serum free media (Life Technologies, Grand Island, NY, USA) supplemented with bovine pituitary extract and epidermal growth factor (Life Technologies, Grand Island, NY, USA), and the media was changed every other day. The fat tissue was cut into 0.5–1 cm2 slices. Collagenase I (1mg/mL) (Life Technologies, Grand Island, NY, USA) was then added at a 3:1 ratio of the tissue volume and incubated on a tube rotator at 37 °C for 90 min. The tissue solution was then centrifuged at 700 RCF for 4 min, the oil phase was removed, and pre-warmed DMEM with 10% FBS was added to stop the enzymatic digestion. Digested tissue was then passed through a 70 µm cell strainer and washed twice with serum-free DMEM. ASDCs were re-suspended in DMEM with 10% FBS and 1% penicillin-streptomycin and seeded in monolayer culture. Total RNA, including small RNAs, were extracted from primary keratinocytes and ADSCs using miRNeasy kit (Qiagen, Hilden, Germany). Briefly, cell pellets were lysed using QIAzol lysis reagent with mechanical agitation. Cell lysates were dissolved in chloroform and the aqueous phase supernatant was loaded into RNeasy mini column, washed and eluted in RNase free water. RNA yield was measured using NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Low molecular weight RNA molecules were labelled with the FlashTag Biotin RNA Labeling Kit (Affymetrix, Santa Clara, CA, USA). Briefly, 500 ng RNA from each sample was ligated to a poly (A) tail followed by binding to a biotinylated signal molecule. miRNA microarray hybridization was then performed with Affymetrix GeneChip miRNA Array 3.0 (Affymetrix, Santa Clara, CA, USA), according to manufacturer’s instructions. Briefly, biotin-labeled samples were incubated with hybridization master mix at 99 °C for 5 min, followed by 45 °C for another 5 min. Hybridization was performed in a rotating hybridization oven (60 rpm) at 48 °C for 18 h. The arrays were washed and stained on GeneChip automated fluidics station and scanned with an Affymetrix GCS 3000 7G-plus scanner (Affymetrix, Santa Clara, CA, USA). The microarray data were analysed using the Transcriptome Analysis Console (TAC)® (Thermo Fisher Scientific, Waltham, MA, USA). Following quality check (Supplementary Table S4), differential expression (DE) analysis was performed between the two study groups for homo sapiens specific miRNA with ID contain “hsa-miR” and log2 fold change ≥2 with p-value < 0.05 (Figure 1a,b). DE result was annotated using the Affymetrix GeneChip® miRNA 4.0 array annotation, version HG38. Quantitative real-time PCR was performed to detect the expression of early and late epidermal markers from 5 independent cell line replicates of direct co-culture (n = 5). Total cellular RNA was extracted using RNeasy mini kit (Qiagen, Hilden, Germany). RNA was reverse transcribed into cDNA using Maxima First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA) following the elimination of double-stranded DNA as recommended by the manufacturer. Gene expression was determined by the PowerUp© SYBR green master mix (Applied biosystem, Waltham, MA, USA). Sequences for the oligonucleotide primers of target genes are listed in Supplementary Table S1. For miRNA quantification in keratinocytes and ADSCs, total RNA, including small RNAs, as described above were reverse transcribed with the miScript II or miRCURY LNA Reverse-Transcription kits (Qiagen, Hilden, Germany) with RNA input of 250 ng or 10 ng respectively, according to the manufacturer’s instructions. The miScript SYBR® Green or the miRCURY® LNA SYBR® Green were used according to the availability of the marker of interest in either workflow (Supplementary Table S2) following the manufacturer’s protocols. Gene and miRNA expression levels were quantified in a 7500 Fast Real-Time PCR System (Applied Biosystem, Thermo Fisher) and the assays were performed in a minimum of three technical replicates. Gene and miRNA expression were normalized against the endogenous controls Glyceraldehyde 3-phosphate Dehydrogenase (GAPDH) or U6 snRNA, respectively and the fold change was calculated with the 2−ΔΔCT method [76]. To identify the target genes of the significant DEmiRNAs, the list of up and downregulated miRNAs were uploaded separately to the miRwalk database v3 “http://mirwalk.umm.uni-heidelberg.de (accessed on 1 December 2022)” [77]. The targeted genes were determined according to the following criteria, (1) the targeted mRNAs should be present on the three databases—TargetScan [78], miRDB [79] and miRTarBase [80], (2) the miRwalk score should be ≥0.95. In order to obtain the “Skin” related expression from the upregulated DEmiRNA-targeted mRNAs in keratinocytes, an expression dataset from the FANTOM5 project in the Expression Atlas [35] was collected. A mean cut-off score was applied on the dataset to draw genes which were highly expressed in skin tissues. The resulting DEmiRNAs-mRNA interactome diagram was generated using the Cytoscape software (v3.8.2) [81]. The Reactome database v83 “https://reactome.org (accessed on 2 December 2022)” [82] was considered to identify the enriched pathways from the up- and downregulated DEmiRNA-targeted mRNAs. Immunocytochemistry (ICC) was used to detect the expression of the key epidermal differentiation markers (KRT5, KRT14, and KRT18). Following cell fixation in ice-cold methanol 99% for 15 min at −20 °C, endogenous peroxidase activity was quenched with H2O2 (10 min). Cells were then permeabilized by incubation for 15 min at room temperature with 0.05% Triton X-100 (Thermo Scientifc, IL, USA). Non-specific binding was blocked using 3% bovine serum albumin in PBS, for 1 h. Primary antibodies, listed in Supplementary Table S3, were incubated overnight with the fixed cells at 4 °C followed by incubation with the biotinylated mouse and rabbit specific secondary antibody for an hour at room temperature. The immune complex was visualized using the streptavidin-biotin immunoenzymatic antigen detection system where the streptavidin-enzyme conjugate binds to the biotin present on the secondary antibody. Positive cells were stained with chromogen 3-Amino-9-Ethylcarbazole (AEC) using detection immunohistochemistry kit (Abcam, Cambridge, UK) following the manufacturer’s instruction. Then the cells were rinsed with Tris-buffered saline (TBS) and counterstained with 8GX alcian blue solution (Sigma-Aldrich, St. Louis, MO, USA) for 2 min before a final wash and mounting with anhydrous mounting medium. For negative controls, PBS was applied instead of the primary antibody. ICC were run in triplicate for each individual antibody stain (n = 3). Stained cells photographed under inverted microscope (CKX53, Olympus Corp., Tokyo, Japan) using the Imageview software version X64 (Olympus Corp., Japan). To quantify stain intensity, colour deconvolution was performed using ImageJ 1.53c with the necessary plugins (Wayne Rasband National institutes of health, Bethesda, MD, USA) then setting a unified threshold for integrated pixel density. Keratinocytes and ADSCs from 3 donors were harvested and solubilized in lysis buffer 17, according to the manufacture instructions (R&D Systems Inc., Minneapolis, MN, USA). Total protein concentration was determined using the DC protein assay (Bio-Rad Inc., Hercules, CA, USA), following the manufacturer’s instructions. Proteome Profiler™ Human XL Cytokine Array (R&D Systems, Inc., Minneapolis, MN, USA) was used to simultaneously assess soluble human proteins and their differential expression between the two cell types. Following the manufacturer’s instructions, the array membranes were blocked for 1 h on a rocking platform, and 150 µg of the cell lysates were incubated with array membranes overnight at 4 °C. The arrays were then incubated with a cocktail of biotinylated detection antibodies for 1 h followed by chemiluminescent detection with Streptavidin-HRP. Membranes were imaged using the ChemiDoc™ MP imaging system (Bio-Rad Inc., Hercules, CA, USA). The pixel densities at each capture spot were quantified and normalized to the reference spots of each blot. Images were analyzed using ImageJ 1.53c (Wayne Rasband National institutes of health, MD, USA) where the mean intensity corresponded to the relative expression of each blotted protein in the cell lysate. Integrated analysis was performed between DEmiRNAs and DEproteins in keratinocytes and ADSCs with the assumption that the resulting protein products were directly affected by miRNA-target interaction. To maintain a reliable resource for miRNA target genes, only experimentally validated targets from the miRwalk database were curated and cross-checked with the miRTarBase as well as the miRNA-gene interactions annotated in the Affymetrix GeneChip® miRNA 4.0 array annotation, version HG38. All possible alternative nomenclature was considered for each gene and protein and were used to search for possible miRNA-gene interactions on the aforementioned databases. For PCR-based assays and ICC, the data were analysed using the Data Analysis ToolPak (Microsoft® Excel, Microsoft® Office 365, Redmond, DC, USA), and the graphs were created using GraphPad Prism Version 9 (GraphPad Software Inc., San Diego, CA, USA). Statistical significance was evaluated using Student’s t-test for unequal variance. Bar charts showed the mathematical mean and the standard error of mean. To estimate the DEmiRNAs from the microarray analysis, p-value < 0.05 was considered as the level of significance. To identify the enriched pathways in the analysis, Benjamin-Hochberg (BH) corrected p-value < 0.05 was applied. Fisher’s exact test was used to test a null-hypothesis, assuming there was no regulation between DEmiRNAs and the genes associated with DEproteins.

PMC10002857

Bonita H. Powell,Andrey Turchinovich,Yongchun Wang,Olesia Gololobova,Dominik Buschmann,Martha A. Zeiger,Christopher B. Umbricht,Kenneth W. Witwer

miR-210 Expression Is Strongly Hypoxia-Induced in Anaplastic Thyroid Cancer Cell Lines and Is Associated with Extracellular Vesicles and Argonaute-2

24-02-2023

microRNA,miRNA,miR-210,hypoxia,HIF1-alpha,extracellular vesicles,anaplastic thyroid cancer

Hypoxia, or low oxygen tension, is frequently found in highly proliferative solid tumors such as anaplastic thyroid carcinoma (ATC) and is believed to promote resistance to chemotherapy and radiation. Identifying hypoxic cells for targeted therapy may thus be an effective approach to treating aggressive cancers. Here, we explore the potential of the well-known hypoxia-responsive microRNA (miRNA) miR-210-3p as a cellular and extracellular biological marker of hypoxia. We compare miRNA expression across several ATC and papillary thyroid cancer (PTC) cell lines. In the ATC cell line SW1736, miR-210-3p expression levels indicate hypoxia during exposure to low oxygen conditions (2% O2). Furthermore, when released by SW1736 cells into the extracellular space, miR-210-3p is associated with RNA carriers such as extracellular vesicles (EVs) and Argonaute-2 (AGO2), making it a potential extracellular marker for hypoxia.

miR-210 Expression Is Strongly Hypoxia-Induced in Anaplastic Thyroid Cancer Cell Lines and Is Associated with Extracellular Vesicles and Argonaute-2 Hypoxia, or low oxygen tension, is frequently found in highly proliferative solid tumors such as anaplastic thyroid carcinoma (ATC) and is believed to promote resistance to chemotherapy and radiation. Identifying hypoxic cells for targeted therapy may thus be an effective approach to treating aggressive cancers. Here, we explore the potential of the well-known hypoxia-responsive microRNA (miRNA) miR-210-3p as a cellular and extracellular biological marker of hypoxia. We compare miRNA expression across several ATC and papillary thyroid cancer (PTC) cell lines. In the ATC cell line SW1736, miR-210-3p expression levels indicate hypoxia during exposure to low oxygen conditions (2% O2). Furthermore, when released by SW1736 cells into the extracellular space, miR-210-3p is associated with RNA carriers such as extracellular vesicles (EVs) and Argonaute-2 (AGO2), making it a potential extracellular marker for hypoxia. Anaplastic thyroid carcinoma (ATC) is a rare malignancy, yet it accounts for the majority of all thyroid tumor-related deaths [1,2,3,4]. In contrast with thyroid cancers such as papillary thyroid cancer (PTC), ATC cases are characteristically aggressive and almost always fatal, as these tumors acquire numerous deleterious genetic aberrations during de-differentiation [1,5]. Anaplastic thyroid carcinomas, as with many other types of solid tumors, are highly proliferative and often hypoxic (oxygen-deprived) [5,6,7,8]. Tumors adapt to hypoxic stress by activating various cell signaling pathways through the action of hypoxia-inducible factors (HIFs). HIF-1 is a heterodimer composed of an alpha and a beta subunit [9,10]. In normoxic conditions, HIF-1α is produced constitutively but is rapidly degraded by the proteasome upon hydroxylation. This occurs at specific proline residues by prolyl hydroxylase (PHD), followed by ubiquitination by the von Hippel–Lindau (VHL) protein [11,12,13]. In hypoxic conditions, HIF-1α proteins remain stable and translocate to the nucleus to activate genes by associating with the HIF-1ß subunit and binding to HIF Response Elements (HREs) within the proximal promoters of target genes [14,15,16]. Several HIF targets are involved in cell cycle regulation, metabolism and angiogenesis [14,17,18,19,20,21,22]. Furthermore, hypoxic tumor cells with elevated HIF-1α protein levels are often resistant to chemotherapy and ionizing radiation [8,23,24]. Hence, developing approaches to identify and target hypoxic cells within ATC tumors may improve patient outcomes. miRNAs are short, ~22 nucleotide, post-transcriptional gene regulators that bind to complementary regions within the 3′UTR of mRNAs, thereby inhibiting translation or mediating degradation [25,26,27,28]. Cells release miRNAs into the extracellular milieu, where they are protected from RNase digestion in carriers including extracellular vesicles (EVs) and extracellular proteins such as Argonaute-2 (AGO2) [29,30,31,32,33,34,35]. Thus, miRNAs may be particularly attractive for the purpose of “liquid biopsies”: using biological fluids to assess the presence and state of cancer cells in the body. We and several others have reported dysregulated microRNAs (miRNAs) in thyroid cancers. The most commonly aberrantly expressed miRNAs in ATC include miR-222, miR-221, miR-146b and miR-21 [36,37,38,39,40,41]. However, there is a paucity of information on miRNA expression differences in hypoxic vs. non-hypoxic regions within thyroid tumors [30,31,32,33,34,35]. In this study, we examined miR-210-3p (“miR-210”) as a potential marker of hypoxia. miR-210 is a well-known hypoxia-responsive miRNA and a direct HIF-1 target [42,43,44,45]. Its expression is frequently up-regulated in hypoxic tumor cells. As with HIF-1α protein levels, elevated miR-210 expression is associated with a poor prognosis [46,47]. We report herein on miR-210 expression in both PTC and ATC cell lines, examine the influence of hypoxia on cellular and extracellular miR-210-3p in the ATC line SW1736 and assess the extracellular association of miR-210 with EVs and AGO2. To confirm the inducible expression of miR-210-3p in response to HIF-1α stabilization in hypoxia, SW1736 (ATC) cells were cultured at 21% O2 (normoxia) or 2% O2 (hypoxia) for 24, 48, 72, 96 and 120 h. Increased HIF-1α protein levels in hypoxia were observed at each time point; the greatest increases compared with normoxic controls were at 24 and 48 h (Figure 1B). Concurrent with increased HIF-1α protein levels, miR-210-3p expression increased at each hypoxic time point, with the highest expression levels also at 24 h (~9-fold) and 48 h (~16-fold) relative to normoxia (Figure 1C). Both HIF-1α and miR-210-3p levels slightly declined by 72 h and thereafter (Figure 1B,C). This is consistent with our observation and previous reports of decreased levels of HIF-1α protein due to mRNA instability in prolonged hypoxia (Figure 1D) [48,49,50] as well as the negative regulation of HIF-1α by miR-210 in a study of human T-cells [51]. Additionally, cell morphology, viability and glucose consumption were compared in hypoxia vs. normoxia. Cells from each condition were visibly comparable (Figure 1E,F) and shared a similar viability, but cell counts were lower in hypoxia (Figure 1G). Additionally, glucose consumption significantly increased in hypoxia (Figure 1H). Taken together, our results confirm the positive correlation between miR-210-3p expression and HIF1-α protein levels in SW1736. Small RNA sequencing was performed on ATC cell lines SW1736 and C643 and PTC cell lines BCPAP and TPC-1, all grown under normoxic culture conditions. We identified 28 miRNAs that were differentially expressed (fold-change > 2, p-value < 0.05) between normoxic ATC and PTC lines (Figure 2A). Under normoxia, miR-210 was ~5-fold less abundant in ATC lines versus PTC lines (Figure 2A). qPCR validation confirmed this result for normoxia (Figure 2B). However, in hypoxia, ATC miR-210 expression reached levels comparable to or greater than those in PTC cells (Figure 2B). A time course assay examining points from 0 to 48 h of hypoxia revealed that ATC cell lines showed a higher fold change, >2-fold, of miR-210-3p expression compared with the PTC cell lines (Figure 2C). Therefore, although basal (normoxic) levels of miR-210 are lower in ATC than PTC lines, the miRNA is more strongly up-regulated by hypoxia in ATC lines. To examine how hypoxia might impact the expression of additional miRNAs in SW1736, small RNA-sequencing was performed to assess miRNA differences in normoxia vs. hypoxia at 72 h. Although miR-210-3p was up-regulated in hypoxia (~6.5-fold, p < 0.05), no other miRNAs were significantly differentially expressed greater than 2-fold, p < 0.05, except for the precursor premiRNA-210 (~2.5-fold, p < 0.05) (Figure 3A). Interestingly, most of the reads aligned to 3-p termini of pre-miR-210 were 1-2 nucleotides longer than the canonical mature miR-210 form and frequently contained mismatched bases at +1 location (T → G, C, A) and +2 location (C → T, G, A) (Supplementary Figure S1). The latter observation may hint on the occurrence of certain miRNA editing events during processing. We also inspected the differential expression of protein-coding mRNA as well as long non-coding RNA in response to hypoxia in SW1736 by small RNA-seq analysis. Specifically, we found 12 protein-coding and 2 lncRNA transcripts to be up-regulated more than 2-fold p < 0.05 (Figure 3A; Supplementary Table S1). Surprisingly, no genes were significantly down-regulated within the given range (LFC < −1). However, hierarchical clustering strongly emphasized coordinated responses to hypoxia with at least 60 differentially expressed genes (−0.5 < LFC < 0.5) between the two groups (Figure 3B). Finally, gene ontology, KEGG and pathway analysis confirmed the enrichment of the deregulated transcripts in processes related to hypoxia response including the HIF-1 signaling pathway and glucose metabolism (Figure 3C,D). Cells and conditioned cell culture media (CCM) were collected after 72 h of normoxic and hypoxic culture, and CCM was fractionated by size-exclusion chromatography (SEC) to obtain fractions enriched in extracellular vesicles (EVs), intermediate mixed EVs and proteins and extracellular proteins. Fractions were characterized per MISEV2018 guidelines [52]. The presence of EVs was verified by Western blot (WB) using EV-rich markers CD63, CD9, CD81, TSG101 and Syntenin, which were detected in EV-rich fractions but not in the later protein fractions (Figure 4B). The depletion of cellular markers such as GM130 and calnexin in extracellular fractions was confirmed relative to the cell source (Figure 4B). Single-particle interferometric reflectance imaging sensing (SP-IRIS) further confirmed EV-rich markers CD63, CD9 and CD81 relative to MIgG control (Figure 4C). Fractions were imaged by transmission electron microscopy (TEM), revealing that EVs ranged in diameter from ~50 to 300 nm in both normoxia and hypoxia in pooled EV-rich SEC fractions; EVs were not observed in protein-rich fractions (Figure 4D). We also assessed particle concentration and size by nano-flow cytometry (NFCM) (Figure 4E), finding no significant differences. qPCR for miR-210-3p showed that miR-210-3p was released from cells in all fractions during normoxia, and that this release increased, also across all extracellular fractions, by >2-fold in hypoxia (Figure 4F). Finally, immunoprecipitation confirmed the presence of miRNA carrier AGO2 in both mixed and protein fractions (Figure 4G). Highly aggressive tumors, which are often hypoxic, are frequently resistant to oxygen-dependent treatments such as chemotherapy and radiation. Thus, several direct and indirect methods to assess tumor hypoxia for targeted therapy, including miRNA expression, have been explored [53,54,55,56,57]. However, to our knowledge, miRNAs have not been extensively studied in hypoxic vs. non-hypoxic regions within ATC tumors. In this work, we examined miR-210, a direct HIF-1 target, as a potential marker of hypoxia in ATC [58,59]. Our results demonstrate that precursor miR-210 is down-regulated in ATC cell lines compared with PTC cells, but it is more robustly up-regulated in ATC in response to hypoxia. Additionally, we show that miR-210-3p can be detected extracellularly and is associated with EVs and extracellular AGO2. miR-210 has paradoxically been described as both oncogenic and as a tumor suppressor in many studies [42,60,61,62]. It is believed to be involved in cell cycle regulation, metabolism and angiogenesis [63,64]. However, the exact role of miR-210-3p in tumorigenesis remains elusive. In this study, we also examined the differential expression of mRNA, but no down-regulated transcripts were identified. Polysome profiling by the proteomic analysis of ATC cell lines in hypoxia may be required to identify putative miR-210-3p targets in this cell type, as the use of a miR-210-3p mimic may result in off-target gene down-regulation. Hypoxia can impact the differentiation state of cells by altering the expression of genes involved in various cellular pathways including cell cycle regulation and anaerobic metabolism [17]. Depending on the cell type, hypoxia can drive cells to a stem-like or differentiated state [65]. Furthermore, miR-210-3p may have a functional role in altering the differentiation state of cells in hypoxia [66]. It was interesting to observe lower basal levels of miR-210-3p in ATC cell lines SW1736 and C643, which are de-differentiated, compared with poorly and well-differentiated PTC cell lines BCPAP and TPC-1, respectively. This finding suggests that miR-210 might also be linked to tumor differentiation. It is believed that ATC can originate from PTC [67,68,69]; therefore, it is possible that miR-210 expression is impacted during de-differentiation. However, because miR-210 down-regulation was observed at both the precursor and mature level, it is unclear if the basal level expression differences are directly related to HIF-1 or miR-210 genetic or epigenetic alterations due to the differentiation levels of the thyroid cell lines. In either case, lower basal expression levels make miR-210 induction more pronounced and rapid in hypoxia; hence, miR-210-3p may be a better marker of hypoxia in ATC compared with PTC. Furthermore, miR-210-3p might also be a predictive marker of the de-differentiation of thyroid cancer cells. We hypothesized that miR-210-3p might also be detected extracellularly in hypoxia, as miRNAs are remarkably stable in biological fluids such as blood plasma and are promising non-invasive biomarkers [29,32,34,35]. This is ascribed to their association with RNA carriers such as EVs and RNA-binding proteins, which offer protection from degradation [34,70]. In our study, we observed increased extracellular miR-210-3p in conditioned cell culture medium in hypoxia after at least 48 h, which further increased after 72 h. This suggests that, although cellular levels of miR-210 increase almost immediately in response to hypoxia, additional time is needed for the newly produced miRNAs to pass into the extracellular space and be detected there. Apart from miR-210-3p, we documented the upregulation of various small RNAs aligning to more than 60 different protein-coding and lncRNA transcripts. Accordingly, gene ontology analysis revealed that multiple upregulated RNA fragments derived from protein-coding genes strongly related to hypoxia response including the HIF-1 signaling pathway and glucose metabolism. The mechanism of formation and the biological role of these mRNA-aligned small RNAs remains unknown; however, we suggest that they could be used as promising biomarkers for hypoxia in addition to miR-210-3p. The presence of those mRNA- and lncRNA-aligned short RNAs (mlnca-sRNA) could be explained by the uneven degradation of the parental transcripts that are retained in the small RNA fractions and the extracellular space. In summary, our observations, albeit limited to in vitro studies, suggest that miR-210-3p might be a suitable miRNA marker of hypoxia in ATC. SW1736, C643, BCPAP and TPC-1 thyroid cell lines [71,72,73,74] were authenticated by the Johns Hopkins Genetic Resources Core Facility by Short Tandem Repeat (STR) profiling. Cells were cultured in RPMI-1640 medium (Gibco cat#11875-119, Grand Island, NY, USA) supplemented with: 10% Exosome-Depleted fetal bovine serum (Gibco A27208-01, Lot#2165597), 2mM L-glutamine (Thermo Fisher 25030-081, Waltham, MA, USA), 1% Non-Essential Amino Acids (Thermo Fisher cat#1140050), 10mM HEPES (Life Tech cat#5630-080, Newmarket, Australia), 100 U/mL Penicillin Streptomycin (Thermo Fisher cat#15140-122). Cells were cultured in a humidified 37 °C, 5% CO2 incubator under 21% O2 (normoxic) or in a 2% O2 (hypoxic) chamber, balanced with N2 using a compact O2/CO2 Controller ProOx c21 (Biospherix RS485, Parish, NY, USA) for 2, 4, 8, 24, 48, 72 or 96 h. Cell counts and viability were obtained using the Muse cell analyzer (EMD Millipore cat#1000-5175 serial#1846277, Burlington, MA, USA) and Muse Viability reagent (Millipore cat#MCH600103). Glucose concentrations from conditioned media were measured using a glucose monitoring system (bioreactor sciences GM-100 serial#15100005, Lawrenceville, GA, USA) and test strips (GMTS-50 bioreactor science). Cells were seeded into 6-well plates and cultured until 100% confluent in normoxic conditions. Media was removed from wells and cells were washed twice gently with PBS. A 200 µL pipette was used to create a scratch. Cells were washed again, and a sharpie was used to mark a location for imaging. Images were taken at 0 h and 24 h during hypoxia. Cell culture conditioned medium was collected from cells and centrifuged at 1000× g for 5 min to pellet dead cells. The supernatant was centrifuged at 2800× g for 10 min to clear cell debris and apoptotic bodies. The 2800× g supernatant was then concentrated down to 500 µL at 3500× g using Centricon Plus-70 10kD cut-off concentrators (Millipore cat#UFC701008) for 1 h at 4 °C. A total of 500 µL of concentrated medium was fractionated over a qEV original 70 nm size-exclusion chromatography column according to the manufacturer’s instructions (IZON Science SP1). Pooled EV fractions (7–9) were concentrated down to ~100 µL at 4000× g using Amicon Ultra-2 Centrifugal Filter Units10kD cut-off concentrators (Millipore cat#UFC201024) at 4 °C. The EV and protein fractions were negatively stained and analyzed by transmission electron microscopy (TEM) at the Johns Hopkins Microscope Core Facility as previously described [75]. Samples were adsorbed (10 µL) to glow-discharged (EMS GloQube) carbon-coated 400 mesh copper grids (EMS), by flotation for 2 min. Grids were quickly blotted then rinsed in 3 drops (1 min each) of tris-buffered-saline (TBS). Grids were negatively stained in 2 consecutive drops of 1% uranyl acetate with tylose (1% UAT, double filtered, 0.22 µm filter), blotted then quickly aspirated to get a thin layer of stain covering the sample. Grids were imaged on a Phillips CM-120 TEM operating at 80 kV with an AMT XR80 CCD (8 megapixel). The NanoFCM Flow NanoAnalyzer was used to measure concentration and particle size following the manufacturer’s instructions and as described previously [75,76]. The instrument was calibrated separately for concentration and size using 250 nm Quality Control Nanospheres (NanoFCM) and a Silica Nanosphere Cocktail (NanoFCM cat#S16M-Exo) for detection of side scatter (SSC) of individual particles. Events were recorded for 1 min. Using the calibration curve, the flow rate and side scattering intensity were converted into corresponding particle concentrations and size. Measurements were performed as described previously [75]. A total of 50μL of EVs was diluted 1:1 in 1xIncubation Solution II and incubated at room temperature on ExoView Human Tetraspanin chips (Unchained Labs, Pleasanton, CA, Cat # 251-1000) printed with anti-human CD81 (JS-81), anti-human CD63 (H5C6), anti-human CD9 (HI9a) and anti-mouse IgG1 (MOPC-21). After incubation for 16 h, chips were washed with 1X Solution A I 4 times for 3 min each under gentle horizontal agitation at 460 rpm. Chips were then incubated for 1 h at room temperature with a fluorescent antibody cocktail of anti-human CD81 (JS-81, CF555), anti-human CD63 (H5C6, CF647) and anti-human CD9 (HI9a, CF488A) at a dilution of 1:1200 (v:v) in a 1:1 (v:v) mixture of 1X Solution A I and Blocking Solution II. The buffer was then exchanged to 1X Solution A I only, followed by 1 wash with 1X Solution A I, 3 washes with 1X Solution B I and 1 wash with water (3 min each at 460 rpm). Chips were immersed in water for approximately 10 s each and removed at a 45-degree angle to allow the liquid to vacate the chip. All reagents and antibodies were supplied by Unchained Labs (Pleasanton, CA, USA). Samples were diluted in 1X Incubation Solution II to load 50 μL of 4.0 × 108 particles/mL, nominally, per chip. All chips were imaged in the ExoView R100 (Unchained Labs, Pleasanton, CA, USA) by interferometric reflectance imaging and fluorescent detection. Data were analyzed using ExoView Analyzer 3.1 Software (Unchained Labs). Fluorescent cutoffs were as follows: CD63 channel 200, CD81 channel 400, CD9 channel 400. Cells were removed from the incubator and immediately placed on ice. The medium was removed and immediately processed for EV purification. Cells were washed with ice-cold PBS and lysed in ice-cold 1X RIPA buffer (Cell Signaling 9806S, Danvers, MA, USA) with 1X Protease inhibitors (Santa Cruz sc-29131, Santa Cruz, CA, USA). Cells were incubated on ice for 10 min then transferred to tubes using cell scrapers. Lysates were cleared at 20,000× g for 10 min at 4 °C. The pellet was discarded. Total protein was measured using Pierce BCA Protein Assay Kit according to the manufacturer’s microplate protocol (Thermo Fisher 23225). A final concentration of 1X Laemmli sample buffer (Bio-Rad 161-0747) with 10% beta-mercaptoethanol (BME) (Bio-Rad 161-0710, Hercules, CA, USA) was added to 15 µg of total protein. The EV and mixed samples were vortexed for 30s and incubated for 10 min at room temperature in 1X RIPA buffer (Cell Signaling 9806S) and 1X Protease inhibitors (Santa Cruz sc-29131) to lyse EVs. Total protein was measured using Pierce BCA Protein Assay Kit according to the manufacture’s microplate protocol (Thermo Fisher 23225). 1X Laemmli sample buffer (Bio-Rad 161-0747) was added to 1 µg of total protein. Mixed and protein fractions were pre-cleared of IgG from FBS with 100 µL of protein G coated magnetic Dynabeads (Thermo Fisher 10003D Lot# 00715594) overnight at 4 °C with rotation. A total of 50µL of protein G beads was coated with 2ug of anti-AGO2 antibody (Sigma SAB4200085 Lot# 0000089486, St. Louis, MO, USA) according to the manufacturer’s instructions. The clearing protein G beads were removed from the sample and replaced with the anti-AGO2 coated beads overnight at 4 °C with rotation. The beads were washed three times with 1X PBST and resuspended in 1X Laemmli sample buffer 10% BME (Bio-Rad 161-0747). Samples were heated to 95 °C for 5 min then separated alongside a spectra multi-color Ladder (Thermo Fisher 26634) through a 4–15% Tris-Glycine extended Stain-Free gel (Bio-Rad 5678085), at 100 V for 1.5 h using 1X Tris-Glycine SDS buffer (Bio-Rad 161-074). Gels were imaged with an EZ DocGel Stain-free imaging system (Bio-Rad 170-8274). Proteins were then transferred to a methanol activated (10s) PVDF membrane (Bio-Rad 1620177) at 100 V for 1 h in 1X Tris-Glycine buffer (Bio-Rad 161-0734) at 4 °C. The PVDF membrane was blocked for 1 h in blocking buffer (1XPBS (Gibco 14190-144), 0.05% Tween20 (Sigma-Aldrich 274348500), 5% blotting-Grade Blocker (Bio-Rad 170-6404)). The membrane was incubated with primary antibodies anti-CD63 (BD 556019 Lot#7341913) diluted 1:1000 and anti-CD81(sc-7637 Lot#C2318) diluted 1:500, anti-CD9 (BioLegend 312102 Lot#B351275, San Diego, CA, USA), ant-TSG101 (abcam ab228013 Lot#GR3306738-8, Cambridge, UK) diluted 1:500, anti-Syntenin (abcam ab133267 Lot# GR3375272-1) diluted 1:500, anti-GM130 (ab52649 Lot#GR3427322-2, anti-Calnexin (ab22595)) diluted 1:1000, anti-Albumin (abcam ab28405 Lot#GR3367930-2) diluted 1:1000, or anti Hif-1a (Cayman 10006421) diluted 1:200, anti-AGO2 (Sigma SAB4200085 Lot# 0000089486) diluted 1:1000 and anti-beta-actin (Sigma A1978) diluted 1:10,000 in blocking buffer overnight at 4 °C with rotation. The membrane was washed 3x for 5 min with rotation in blocking buffer. Secondary anti-mouse-HRP (Santacruz sc-516102 Lot#c1419) was diluted 1:10,000 or anti-rabbit-HRP (Dako P0448, Glostrup, Denmark) 1:1000 in blocking buffer and incubated for 1h at room temperature. Membranes were washed 3X in blocking buffer then 2X in 1XPBS 0.05% Tween20. Membranes were then incubated with Super Signal Chemiluminescent Substrate (Thermo 34580) for 5 min with gentle rotation and imaged by iBright FL1000 (Invitrogen, Waltham, MA, USA). For HIF-1a, band intensity in normoxia was normalized to hypoxia using Image J software. Total RNA was extracted from cells using mirVana miRNA isolation kit (Ambion cat#AM1560, Austin, TX, USA) following the manufacturer’s protocol for adherent cells. Total RNA was extracted from size-exclusion chromatography (SEC) EV and protein fractions using miRNeasy serum/plasma kit (Qiagen 1071073 Lot#160020206, Hilden, Germany) after adding 1.0 × 106 copies/µL of cel-miR-39 exogenous spike-in control (Qiagen 219610 Lot#157036035), according to the manufacturer’s instructions. The RNA concentration and purity were measured by NanoDrop (Thermo Fisher). The ligation-independent Capture and Amplification by Tailing and Switching (CATS) small RNA-seq method was used to profile cellular small RNA originally described in [77]. Libraries were constructed with CATS RNA-seq Kit (Diagenode C05010041 Lot#4) following the manufacturer’s instructions. The remaining NGS library primers and other products shorter than 100 bp were removed by AMPureXP beads (Beckman, Brea, CA, USA). Quality control was assessed by Agilent bioanalyzer high sensitivity assay (Agilent, Santa Clara, CA, USA). Libraries were then additionally size-selected between 160 and 180 bp by BluePippin (Sage Science, Beverly, MA, USA). Prior to sequencing, libraries were spiked with 20% PhiX Control v3 (Illumina 1501766, San Diego, CA, USA) then run using the NovaSeq Illumina sequencing platform by the Johns Hopkins Microarray and Deep Sequencing Core. Sequencing quality control was performed using FASTQC (Babraham Bioinformatics, Cambridge, UK) followed by adapter and PolyA trimming with Cutadapt 1.17. The reads were first aligned to various housekeeping small non-coding RNA references including rRNA, tRNA, RN7S, snRNA, snoRNA, scaRNA, VT-RNA and Y-RNA (custom-curated from NCBI RefSeq and GENCODE). All reads that did not map to the aforementioned RNAs were sequentially aligned to mature miRNA (miRBase 22 release), pre-miRNA (miRBase 22 release), protein-coding mRNA transcripts and long non-coding RNAs (custom-curated references from GENCODE v28). The numbers of reads mapped to each RNA transcript type were extracted using eXpress software based on a previous publication [78]. All reads mapped to minor transcripts, pseudogenes and non-protein-coding parts of mRNAs were not included in the final analysis. The intronic and intergenic small RNA reads (combined) were extracted by mapping the remaining unaligned reads to hg38 genome reference. The differential expression analysis was performed by edgeR and limma packages as described in [54] using raw count tables as an input. Reads distribution over certain transcripts was visualized using Integrative Genomics Viewer (IGV). The GO-terms enrichment was assessed using the Enrichr knowledgebases as described previously [79] and visualized by the ggpot2 package. Two step RT-qPCR was performed for miRNA analysis starting with 2 µL of RNA input from cells, EVs and Ex-protein fractions using TaqMan microRNA Reverse transcription kit (ABI 4366597 Lot#00636931), TaqMan miRNA stem-loop RT primers/qPCR primer probe set (ABI 4427975): (cel-miR-39 ID# 000200 Lot#P180110-003B10, miR-16 ID# 000391 Lot#P171018-000H05, miR-210-3p ID#000512) and TaqMan Universal Master Mix II, no UNG (ABI 4440040 Lot#1802074) as described in manufacturer’s protocol. For mRNA analysis, high-capacity cDNA synthesis kit (ABI cat#4368813), TaqMan qPCR primer probe set ABI: (HIF-1α ID#Hs00153153_m1, GAPDH ID# Hs02786624_g1, Beta-Actin ID# Hs01060665_g1) and TaqMan Universal Master Mix II, no UNG were used following kit protocol. qPCR was run on a CFX96 Real-time System (Bio-Rad). miR-210 was normalized to miR-16 and cel-miR-39 using the 2−ΔΔCT method. HiF-1α was normalized to the average of GAPDH and Beta-actin Pooled by 2−ΔΔCT.

PMC10002862

Yingchao Shi,Wenhao Liu,Yang Yang,Yali Ci,Lei Shi

Exploration of the Shared Molecular Mechanisms between COVID-19 and Neurodegenerative Diseases through Bioinformatic Analysis

02-03-2023

COVID-19,Alzheimer’s disease,Parkinson’s disease,bioinformatics,differentially expressed genes,gene ontology,protein–protein interaction,hub genes,drugs

The COVID-19 pandemic has caused millions of deaths and remains a major public health burden worldwide. Previous studies found that a large number of COVID-19 patients and survivors developed neurological symptoms and might be at high risk of neurodegenerative diseases, such as Alzheimer’s disease (AD) and Parkinson’s disease (PD). We aimed to explore the shared pathways between COVID-19, AD, and PD by using bioinformatic analysis to reveal potential mechanisms, which may explain the neurological symptoms and degeneration of brain that occur in COVID-19 patients, and to provide early intervention. In this study, gene expression datasets of the frontal cortex were employed to detect common differentially expressed genes (DEGs) of COVID-19, AD, and PD. A total of 52 common DEGs were then examined using functional annotation, protein–protein interaction (PPI) construction, candidate drug identification, and regulatory network analysis. We found that the involvement of the synaptic vesicle cycle and down-regulation of synapses were shared by these three diseases, suggesting that synaptic dysfunction might contribute to the onset and progress of neurodegenerative diseases caused by COVID-19. Five hub genes and one key module were obtained from the PPI network. Moreover, 5 drugs and 42 transcription factors (TFs) were also identified on the datasets. In conclusion, the results of our study provide new insights and directions for follow-up studies of the relationship between COVID-19 and neurodegenerative diseases. The hub genes and potential drugs we identified may provide promising treatment strategies to prevent COVID-19 patients from developing these disorders.

Exploration of the Shared Molecular Mechanisms between COVID-19 and Neurodegenerative Diseases through Bioinformatic Analysis The COVID-19 pandemic has caused millions of deaths and remains a major public health burden worldwide. Previous studies found that a large number of COVID-19 patients and survivors developed neurological symptoms and might be at high risk of neurodegenerative diseases, such as Alzheimer’s disease (AD) and Parkinson’s disease (PD). We aimed to explore the shared pathways between COVID-19, AD, and PD by using bioinformatic analysis to reveal potential mechanisms, which may explain the neurological symptoms and degeneration of brain that occur in COVID-19 patients, and to provide early intervention. In this study, gene expression datasets of the frontal cortex were employed to detect common differentially expressed genes (DEGs) of COVID-19, AD, and PD. A total of 52 common DEGs were then examined using functional annotation, protein–protein interaction (PPI) construction, candidate drug identification, and regulatory network analysis. We found that the involvement of the synaptic vesicle cycle and down-regulation of synapses were shared by these three diseases, suggesting that synaptic dysfunction might contribute to the onset and progress of neurodegenerative diseases caused by COVID-19. Five hub genes and one key module were obtained from the PPI network. Moreover, 5 drugs and 42 transcription factors (TFs) were also identified on the datasets. In conclusion, the results of our study provide new insights and directions for follow-up studies of the relationship between COVID-19 and neurodegenerative diseases. The hub genes and potential drugs we identified may provide promising treatment strategies to prevent COVID-19 patients from developing these disorders. With the liberalization of epidemic prevention and control in various countries, more than 600 million people worldwide have been diagnosed with coronavirus disease 2019 (COVID-19), which is caused by the novel Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2). It is well known that SARS-CoV-2 mainly attacks the human respiratory system and causes typical symptoms, including fever, sore throat, cough, shortness of breath, and fatigue. Moreover, current evidence supports that SARS-CoV-2 is capable of targeting and invading the central nervous system (CNS) [1]. Neurological symptoms have been observed during and after the acute COVID-19 phase, including both CNS symptoms and vegetative/peripheral manifestations [2]. In particular, recent studies have suggested that COVID-19 may trigger clinical manifestations of neurodegenerative disorders, such as cognitive decline [3], dementia [4], and parkinsonism [5], bringing the potential role of COVID-19 in the future development of neurodegenerative diseases into the spotlight. Some studies reported increased risk of these disorders among COVID-19 positive patients [6,7]. The changes in COVID-19 patients’ brain structure also reinforced this hypothesis [8]. In addition, COVID-19-induced impairment of the frontal cortex, a critical area for cognitive function, was described in complementary studies that combined neuro-imaging and cognitive screening [9]. Neurodegenerative diseases are characterized by progressive dysfunction and loss of neurons [10], and they can affect an individual’s movement, speech, memory, cognition, intelligence, and much more [11,12]. These diseases include Parkinson’s disease (PD), Alzheimer’s disease (AD), Huntington’s disease (HD), multiple sclerosis (MS), amyotrophic lateral sclerosis (ALS), epilepsy, and others [13]. AD and PD are the two most common human neurodegenerative diseases, and AD is the leading cause of dementia. Although PD is traditionally considered a movement disorder, dementia is becoming more widely accepted as part of the clinical spectrum of PD [14]. A previous study found that mild cognitive impairment (MCI) was one of the most common non-motor symptoms of early-stage PD patients, and dementia was presented in 83% of 20-year PD survivors [15]. There is growing evidence suggesting an association between AD and PD at the molecular level, such as failure in redox homeostasis, improperly folded modified proteins, and neuroinflammation [16]. The damage to the frontal cortex has been implicated in both AD and PD [17,18]. Clinical studies have reported that patients with a previous neurodegenerative disease have an increased risk for COVID-19, as well as COVID-19-related hospitalization and mortality [19,20,21]. Progress in deciphering the common pathogenesis of COVID-19, AD, and PD is conducive to developing effective strategies to treat the neurological symptoms of infected individuals and to prevent these patients from developing neurodegenerative diseases. To explore the molecular mechanisms of COVID-19-related neurodegenerative symptoms, we estimated transcriptome alterations in the frontal cortex of patients with COVID-19, AD, and PD using two datasets. Further analyses, including Gene Ontology and pathway enrichment, protein–protein interaction (PPI) and key module extraction, identification of hub genes and potential drugs, and transcription factor (TF) regulatory network construction, were performed based on the common DEGs among COVID-19, AD, and PD. The sequential workflow of our research is presented in Figure 1. To determine shared genetic interrelations among COVID-19, AD, and PD, we initially accessed the transcriptomic data from the frontal cortex of each disease in the GEO database. Before the procedure of differential analysis, we performed normalization and removal of batch effects to standardize the expression matrices, and the results of the processing are shown in a density plot (Supplementary Figure S1A,B). After standardization, the normality test (Supplementary Figure S1C) and the PCA plot of each dataset (Figure 2A) indicate that the source of samples is reliable. Next, differential analysis of gene expression was performed by controlling age and sex, which was significantly different between the patients and the healthy controls (Table 1, Table 2 and Table 3). Finally, 1344 genes were identified as DEGs for COVID-19, including 927 up-regulated and 417 down-regulated genes. In the same way, 2655 DEGs (651 up-regulated and 2004 down-regulated) in the AD dataset and 2589 DEGs (882 up-regulated and 1707 down-regulated) in the PD dataset were obtained. The results are shown in the volcano plots (Figure 2B). Using a cross-comparative analysis, we identified 52 common DEGs, including 9 common up-regulated genes and 43 common down-regulated genes, after excluding genes with opposite expression trends among COVID-19, AD, and PD (Figure 2C). This common gene set was submitted to further experiments. The connectivity of common DEGs may indicate crucial information about similar biological roles. To further understand the underlying common biological characteristics among COVID-19, AD, and PD, we implemented four canonical and widely used databases to analyze the common DEGs, including GO, KEGG pathway, Reactome, and GSEA. Typically, GO enrichment analysis is performed to identify the most important molecular features associated with genes, which can be categorized into three subsections, including biological process (BP), molecular function (MF), and cellular component (CC), for the annotation of gene products. For the biological process, the top GO terms that we enriched were associated with synaptic signaling and its regulation, such as neurotransmitter transport, modulation of chemical synaptic transmission, catecholamine transport, and regulation of trans-synaptic signaling. According to the cellular component, synaptic vesicle, transport vesicle, exocytic vesicle, and distal axon were the top terms. In the molecular function, voltage-gated ion channel activity was the main enriched GO term. The top 10 GO terms of each subsection are illustrated in a dot graph (Figure 3A) and summarized in Table 4. For the pathway enrichment analysis, the KEGG analysis revealed that these common DEGs were significantly enriched in the synaptic vesicle cycle pathway, GABAergic synapse, MAPK signaling pathway, cAMP signaling pathway, and nicotine addiction (Figure 3B). The Reactome analysis showed that these genes were most related to transmission across chemical synapses, neuronal system, presynaptic depolarization, calcium channel opening, LGI-ADAM interactions, transcriptional regulation by MECP2, and regulation of insulin secretion (Figure 3C). Further independent analysis for the common up-regulated and down-regulated DEGs revealed that the GO terms and pathways mentioned above were mostly down-regulated, suggesting that a dysfunction of the synaptic vesicle cycle might be the common pathogenesis of COVID-19, AD, and PD. More information for the pathway enrichment results is presented in Table 5. In addition, we used the GSEA to analyze common up-regulated and down-regulated GO terms and KEGG pathways in the COVID-19, AD, and PD expression datasets. The results demonstrated that cytokine–cytokine receptor interaction and humoral immune response were up-regulated. The down-regulated terms were mainly linked to synapses, synaptic membrane, and synaptic properties (Figure 3D). Based on these findings, we supposed that SARS-CoV-2 infection might cause a general down-regulation of genes associated with the synaptic vesicle cycle and synaptic signal transmission in the patients’ frontal cortex. The DO (Disease Ontology) enrichment analysis was conducted to identify the diseases associated with the common DEGs, thereby providing novel perspectives on our intended diseases. Through the DO analysis, we found that the common DEGs were mainly related to a loss of cognitive function or mental diseases, such as different types of epilepsy, dementia, and autism disorder, supporting that these common DEGs might be involved in the neurological symptoms of COVID-19, AD, and PD (Figure 4). PPI networks have been used to discover novel protein functions, as well as identify functional modules and conserved interaction patterns [22]. Thus, the construction of a PPI network is regarded as the crucial step of cellular biology study and works as a precondition for system biology [23]. Here, the PPI network of the common DEGs is depicted in Figure 5, containing 52 nodes and 320 edges. Based on the PPI network, two closely connected modules were obtained through the MCODE plugin, and module 1 is shown in Figure 6A, which has the highest score (18.222) with 19 nodes. The GO and KEGG pathway analyses of module 1 were performed using ClueGO. The results of the GO analysis indicated that it was primarily related to synaptic vesicles (Figure 6B), and the KEGG pathway analysis also showed that module 1 was significantly correlated with the synaptic vesicle cycle (Figure 6C). According to the PPI network, we intended to explore the hub genes that play indispensable roles in the shared biological mechanisms of COVID-19, AD, and PD. Based on three widely used methods, MCC, Degree, and Betweenness Centrality, we listed each algorithm’s top 10 hub genes (Figure 6D). After taking the intersection of these genes, we identified five common genes as the hub genes, including TAGLN3, GAD2, SST, SYP, and KCNJ4. Furthermore, we searched the drug targets of the common DEGs to identify potential therapeutic targets. Here, we identified 6 drug targets and 18 related drugs based on DrugBank (Figure 7). Among them, five drugs, including Ibutilide, Azelnidipine, Dotarizine, Copper, and Artenimol, were considered to have potential therapeutic effects. The detailed information of these drugs and their targets is summarized in Table 6. Ibutilide is a class III antiarrhythmic agent used to correct atrial fibrillation and atrial flutter [24], and it can be considered as an alternative to cardioversion. Azelnidipine is a dihydropyridine calcium channel blocker [25]. It has a gradual onset of action and produces a long-lasting decrease in blood pressure, with only a small increase in heart rate. It is currently being studied for post-ischemic stroke management [26]. Dotarizine is a calcium antagonist used to treat and prevent migraines [27]. Copper is an essential element in the body and is incorporated into many oxidase enzymes as a cofactor [28]. The precise mechanisms of the effects of copper deficiency are vague due to the wide range of enzymes which use its ion as a cofactor. Artenimol is an artemisinin derivative and an antimalarial agent used in the treatment of uncomplicated Plasmodium falciparum infections [29]. The mapping and characterization of TFs regulating the expression of common DEGs can provide insights into the comprehensive biological processes [30]. In this study, 10 possible TFs, including SP2, SIN3A, REST, ATF3, MYF6, TBX5, RFX1, RPL6, NR3C1, and HDAC2, were discovered to be related to 42 common DEGs (Figure 8). Among these DEGs, the five hub genes were all involved. In this study, we used three datasets of COVID-19, AD, and PD patients’ frontal cortex from the GEO database to discover the underlying mechanisms and potential therapeutic strategies for neurodegenerative disorders caused by SARS-CoV-2 infection. Through the intersectional analysis, we identified 52 common DEGs, and most of them were down-regulated, indicating that COVID-19, AD, and PD might cause a suppression of common cellular functions in patients’ frontal cortex. We next performed a pathway-based analysis to identify shared biological pathways of COVID-19, AD, and PD. The pathway analysis revealed that these common DEGs were significantly enriched in the synaptic vesicle cycle pathway. Synaptic vesicles undergo a complex trafficking cycle, which could be divided into sequential steps: the formation of synaptic vesicles; the docking of synaptic vesicles in the active zone of the presynaptic membrane; the priming of synaptic vesicles; the fusion of synaptic vesicles with the presynaptic membrane; the release of neurotransmitters by exocytosis; and the endocytosis of vesicles [31]. An independent GO analysis of the common up- and down-regulated DEGs showed that the top terms, which were mainly associated with the synaptic vesicle cycle, were all down-regulated (Supplementary Figure S2A,B). Furthermore, the GSEA of the three datasets also demonstrated that synapses, components of synapses, and synaptic function were down-regulated in these three diseases. Our results indicated that the loss and damage of synapses and synaptic dysfunction might be the cause of neurodegenerative disease-related symptoms in COVID-19 patients or survivors. Consistent with our results, some studies have shown that SARS-CoV-2 infection may cause synaptic disorder based on high-throughput sequencing and systematic bioinformatic analyses. Andrew et al. found that the synaptic signaling of upper-layer excitatory neurons, which are linked to cognitive function, is preferentially affected in COVID-19 patients through profiling large single-nucleus transcriptomes from the frontal cortex and choroid plexus samples across control individuals and patients with COVID-19 [32]. Cheng et al. also identified that SARS-CoV-2-infected neurons undergo degeneration, including shortened neurite length and reduced synapses [33]. AD may be primarily a disorder of synaptic failure. Synapse loss and synaptic dysfunction are the best-known pathological correlates of cognitive deficits found in AD patients [34,35]. Recently, studies have shown that synaptic pathology occurs in the early stage of AD progression, mainly manifested by a loss of synaptic proteins [36]. It has been reported that synaptophysin, a presynaptic vesicle protein, is decreased by around 25% in MCI patients, and this change occurs before Aβ plaque formation [37]. The dysfunction of synapses in the frontal cortex is considered a marker of AD progress and a very promising therapeutic target. Moreover, researchers have started to develop synaptic therapies to restore and prevent synaptic dysfunction in AD. These treatment strategies aim to avoid synaptic loss, strengthen synaptic connections, and improve synaptic signal transmission function. Moreover, recent studies have shown that a disorder of synaptic vesicle trafficking plays a vital role in the pathogenesis of PD [38]. Among the reported PD-related genes, alpha-synuclein (αS) [39,40], LRRK2 [41,42], Parkin [43,44], PINK1 [45,46], and DJ-1 [47,48] have also been found to regulate the release of neurotransmitters from presynaptic vesicles and the circulation of synaptic vesicles. Obviously, synaptic dysfunction is closely related to the progression of neurodegenerative diseases. Collectively, previous studies have shown that the dysfunction caused by synaptic vesicle circulation is strongly related to neurodegenerative diseases. Since the pathological changes of synapses generally occur in the early stage of neurodegenerative diseases [49,50], we reasonably speculate that patients may have synaptic dysfunction in the cerebral cortex after SARS-CoV-2 infection, including a loss of synapses and an inhibition of synaptic vesicle transport. We identified two key modules and five hub genes based on the topological measures of the PPI network analysis. The GO pathway analysis of the dominant module was consistent with our previous results, which also highlighted that the synaptic vesicle cycle was the potential pathogenesis shared by COVID-19 and neurodegenerative diseases. Five hub genes, including GAD2, SST, TAGLN3, SYP, and KCNJ4, were further verified using the test_datasets of COVID-19, AD, and PD (Figure 6E). GAD2 is expressed in both pancreatic islets and the brain at later ages [51], particularly in the hypothalamus. It encodes one of the isoforms of glutamic acid decarboxylase enzyme, which is responsible for catalyzing the production of γ-aminobutyric acid (GABA) neurotransmitter. GABA is the primary inhibitory neurotransmitter in the central nervous system, and dysfunction of GABAergic mechanisms is associated with different neurological conditions. Previous studies have shown that stimulation of GABAergic signaling protects neurons against the neurotoxicity of amyloid β-protein. Therefore, GAD2 might be a potential therapeutic target for AD treatment [52,53]. Somatostatin (SST), encoded by SST, is a well-known neuropeptide that is expressed throughout the brain. In the cortex, SST is expressed in a subset of GABAergic neurons and is known as a protein marker of inhibitory interneurons. Recent studies have identified the critical functions of SST in modulating cortical circuits in the brain and cognitive functions [54]. Furthermore, reduced expression of SST is a hallmark of various neurodegenerative and neuropsychiatric disorders, such as AD [55], PD [56], HD [57], major depressive disorder (MDD) [58], bipolar disorder, and schizophrenia (SCZ) [59]. TAGLN3, which is preferentially expressed in the CNS, is homologic to transgelin and calponin, two cytoskeleton-interacting proteins. TAGLN3 is a member of the calponin family and co-localizes with actin and tubulin, which indicates that TAGLN3 has a part in neuronal plasticity. Recently, Laurie et al. confirmed that TAGLN3 was significantly down-regulated in the brain of patients with AD, and they considered it to be a molecular target to modulate neuroinflammation and a potential biomarker for AD [60]. SYP is a synaptic vesicle membrane protein that accounts for approximately 7–10% of the total vesicle proteins [61], and it is also used as a marker for synaptogenesis and synaptic density [62]. It has been reported that SYP could affect the efficiency of the synaptic vesicle cycle [63], which would then undermine cognitive ability. Schmitt et al. found SYP knock-out mice showed a significant dysfunction in learning and memory compared to wild-type mice, confirming the role of SYP in modulating cognitive functions [64]. KCNJ4 encodes potassium voltage-gated channel subfamily J member 4, which is an inward rectifier potassium channel family member. Previous studies have shown that KCNJ4 is associated with the progression and poor prognosis of lung adenocarcinoma [65], dilated cardiomyopathy [66], and prostate cancer [67]. Recent bioinformatic research revealed the ion channel-related gene features in COVID-19, of which the up-regulated gene, KCNJ4, was identified as the hub gene. This study indicated a correlation between KCNJ4 and SARS-CoV-2 infection [68]. Moreover, Wang et al. showed that an overexpression of KCNJ4 can protect against rotenone-induced apoptosis in cell models during the neurodegenerative process, suggesting the protective effect of KCNJ4 on neurodegeneration [69]. Since the beginning of the pandemic, extensive global research studies have been underway to find appropriate drug agents to treat COVID-19. However, most of these drugs and therapies aim to reduce COVID-19-related hospitalization rates and deaths, without considering the improvement of neurological complications and ‘Long COVID’ [70,71]. Current treatments for post-COVID conditions are based on symptom relief and rehabilitation as there is no documented specific medical treatment [72]. Therefore, there is an urgent need for drugs to treat neurological symptoms related to COVID-19. Since developing a novel drug is a lengthy, expensive, and risky process, drug repurposing is the best approach to identify therapeutic options for COVID-19-related neurodegenerative diseases in a limited time [73]. Here, we identified candidate drugs from the DrugBank database, which contains very comprehensive information about approved drugs. Ibutilide fumarate is the first ‘pure’ class III intravenous antiarrhythmic agent indicated for the acute termination of atrial fibrillation and flutter [74]. Its predominant action is prolongation of the myocardial action potential duration through a unique ionic mechanism of action [75]. Dotarizine, a novel piperazine derivative, belongs to wide-spectrum Ca2+ channel antagonists. Compared to other Ca2+ channel blockers, Dotarizine was found to have a lower oral toxicity [76]. To date, there are no reports of these two drugs in the treatment of neurological diseases or COVID-19. They still need further experiments to explore their therapeutic potential in neurology. Azelnidipine, a long-acting calcium channel blocker, is highly lipid soluble and selective for the vascular wall [26]. Clinical studies have demonstrated that Azelnidipine markedly reduces heart rate and proteinuria in hypertensive patients by inhibiting sympathetic nerve activity. Azelnidipine has also been confirmed to have cardio-protective, neuroprotective, and anti-atherosclerotic properties, and it has also been found to prevent insulin resistance [77]. Many studies have reported its neuroprotective effects in ischemic stroke (IS) [26,78,79]. Since IS is considered an important contributing factor for the development of vascular dementia (VaD) and AD [80], Azelnidipine may also have neuroprotective effects on neurodegenerative diseases. Copper, a trace element, is present throughout the brain and is most prominent in the basal ganglia, hippocampus, cerebellum, numerous synaptic membranes, and in the cell bodies of cortical pyramidal and cerebellar granular neurons [81]. As a coenzyme factor, copper plays an important role in central nervous system development, and copper deficiency may result in neurological disorders [82,83,84]. Previous studies have found that copper is implicated directly or indirectly in the pathogenesis of numerous neurodegenerative diseases, such as AD, PD, ALS, and HD [83]. Artenimol is an artemisinin derivative and an antimalarial agent used in the treatment of uncomplicated Plasmodium falciparum infections. Recent evidence has demonstrated the potential effect of artemisinin against SARS-CoV-2 [85]. Nair et al. found that artemisia annua L. extracts inhibited the in vitro replication of SARS-CoV-2 and two of its variants [86]. Ruiz-Nuño et al. revealed that artemisinin and its derivatives portrayed more potent binding to Lys353 and Lys31-binding hotspots of SARS-CoV-2 spike protein than hydroxychloroquine, suggesting the potential repurposing of Artenimol for the treatment of COVID-19 [87]. The transcriptome profiling used in this study was obtained from the GEO database (http://www.ncbi.nlm.nih.gov/geo/ accessed on 20 May 2022). The inclusion criteria of GSE188847 included the following: (a) severe COVID-19 patients with pre- or peri-mortem positive test results for SARS-CoV-2 by nasopharyngeal swab qPCR and history of hospitalization, and (b) age- (±2 years) and sex-matched uninfected controls without a history of neurological disorders or psychiatric diseases. The inclusion criteria of GSE150696 included the following: (a) all AD patients were diagnosed according to the Consortium to Establish a Registry for Alzheimer’s disease (CERAD); (b) all PD patients were selected on the basis of the Movement Disorders Society criteria; and (c) age- and sex-matched healthy people without a history of COVID-19. In this study, one RNA-seq dataset of COVID-19 (GSE18847) and one array dataset containing AD and PD (GSE150696) patients were acquired as the training sets. To subsequently validate hub genes, we downloaded the GSE164332 dataset as a validation set for COVID-19, the GSE104704 dataset for AD, and two datasets, GSE20168 and GSE8397, for PD, which were merged, and their batch effects were corrected. Table 7 summarizes the detailed information of included datasets in this study. Firstly, the “ComBat” function in the SVA package (version: 3.38.0) was applied to the merged datasets to correct batch effects. Next, we normalized the datasets and adjusted for covariates using the “Normalizebetweenarrays” and “removeBatchEffect” function in the limma package (version: 3.46.0) [88]. Principal component analysis (PCA), a classic dimension reduction approach, was conducted to verify intra-group data repeatability in each group using the FactoMineR package. A DEG is characterized as being expressed differently at the transcription level when there is a statistically significant difference between diverse conditions [89]. Herein, we performed differential expression analysis using the limma package to identify DEGs in R programming language (version: 4.1.3). The cutoff criteria (p-value < 0.05 and |logFC (fold change)| > 1) were applied to screen significant DEGs for all datasets. A Venn diagram analysis was performed to determine the shared and unique DEGs among COVID-19, AD, and PD. The Gene Ontology (GO) database provides a comprehensive and computational source to annotate gene product-based functions, comprising classes for molecular functions, the biological processes these contribute to, and the cellular locations where these occur [90]. Typically, the canonical pathway databases, Kyoto Encyclopedia of Genes and Genomes (KEGG) and Reactome, are considered as they are well-known databases to grasp the signaling and metabolic pathways [91]. Gene set enrichment analysis (GSEA) is another powerful analytical method for interpreting gene expression data [92]. The Disease Ontology (DO), a comprehensive and standardized knowledge base for inherited, developmental, and acquired human diseases, is utilized for disease annotation by major biomedical databases (e.g., Array Express, NIF, and IEDB) [93]. In this study, we performed enrichment analysis of GO, KEGG pathway, and GSEA for the common DEGs utilizing the clusterprofiler package (version: 3.18.1). The ReactomePA package was applied to the Reactome pathway analysis, and the DOSE package was employed for the Disease Ontology (DO) analysis. A p-value < 0.05 was considered statistically significantly different. The Search Tool for the Retrieval of Interacting Genes (STRING) (https://string-db.org/ accessed on 1 July 2022), which supplies experimental and predicted interaction-based information [94], was used to predict potential interactions between the identified common DEGs at the protein level with a medium confidence score. Additionally, Cytoscape software (version: 3.9) was used to construct and visualize the PPI network. Then, we used an important plugin of Cytoscape, Molecular Complex Detection (MCODE), to extract profound functional modules of genes in the PPI network with default parameters (K-core = 2, degree cutoff = 2, max. depth = 100, and node score cutoff = 0.2) [95]. The MCODE method is generally used to find densely connected regions in a PPI network that may represent molecular complexes or parts of pathways based on graph-theoretic clustering algorithms. Hub genes are identified as having high intramodular connectivity (or module membership), and previous research has revealed critical biological functions by assessing hub genes. To extract hub genes from the PPI network, we applied CytoHubba, which is a plugin of Cytoscape, to identify essential nodes and sub-networks from the complex interactome. It provides several topological algorithms that researchers can select (e.g., MCC, Degree, DMNC, MNC, EPC, and Bottleneck). The study of drug–target interaction is of great importance for drug discovery and design. Based on the common DEGs, candidate drugs and drug–target interactions were predicted using the DrugBank database (https://go.drugbank.com/ accessed on 10 July 2022), which is the world’s most widely used reference drug resource comprising detailed drug, drug–target, drug action, and drug interaction information about FDA-approved drugs, as well as experimental drugs going through the FDA approval process [96]. The intersection of the common DEGs and drug target genes (DTGs) downloaded from DrugBank was then used to identify related drugs. Finally, we excluded drugs that have an opposite effect on their target genes and acquired candidate drugs that might contribute to phenotypes. The statistical significance was set at p-value < 0.05. Precise regulation of gene expression is imperative for all biological processes. In this study, to identify substantial changes happening at the transcriptional level and obtain insights into the hub proteins’ regulatory molecules, we employed the RcisTarget package [97,98] to decode the regulatory transcription factors (TFs), and a p-value < 0.05 was considered significant. RcisTarget is an R-package to identify TF-binding motifs that are over-represented on a gene list. Previous studies have shown that COVID-19 survivors are at high risk of neurodegenerative diseases [6], and degeneration of brain regions related to cognitive functions has been detected in milder cases [8]. It has also become evident that SARS-CoV-2 infection has a negative effect on the outcome of patients with neurodegenerative diseases. In the future, with an increasing number of infections, it is imperative to prevent or treat these neurological symptoms. Our study explored the relations among these three diseases in the context of transcriptomic analysis on AD, PD, and COVID-19 using bioinformatic analyses. We identified the five most significant hub genes from the common DEGs of these three diseases, and other transcriptome data can validate them. Most importantly, we found that the synaptic vesicle cycle was the common pathway shared by COVID-19, AD, and PD. Further analysis indicated that SARS-CoV-2 infection might lead to synaptic dysfunction and extensive synaptic down-regulation in the cortex of patients, thus triggering or aggravating neurodegenerative diseases. Our research contributes to a deeper understanding of the linkage of SARS-CoV-2 to neurodegenerative diseases, and it proposes potential therapeutic targets and related drugs, which may be promising therapeutic strategies for further clinical research studies. Our study also had some limitations. Firstly, this research was performed based on bioinformatic and transcriptomic analyses; the differences in microarray platforms, tissue collection, RNA extraction methods, and statistical methods could produce potential bias in the results. In addition, our study was limited by the amount of available transcriptome expression data derived from the frontal cortex; thus, the size of the datasets used in this study needs to be larger to generate more compelling results. The inclusion of more large cohorts of COVID-19, AD, and PD patients should be better, and future cellular or animal experiments can also be conducted to provide convincing evidence to support our results. Therefore, the above findings should be taken with caution. Nevertheless, our study sheds light on the shared pathogenesis and molecular mechanism behind COVID-19, AD, and PD. Our results suggest the critical role of synaptic signaling and provide several promising genes for the potential correlation between SARS-CoV-2 infection and neurodegenerative diseases.

PMC10002864

Cristina Fantini,Clarissa Corinaldesi,Andrea Lenzi,Silvia Migliaccio,Clara Crescioli

Vitamin D as a Shield against Aging

25-02-2023

vitamin D,immunosenescence,inflammaging,molecular mechanisms,immunocytes,cardiomyocytes,skeletal muscle cells

Aging can be seen as a physiological progression of biomolecular damage and the accumulation of defective cellular components, which trigger and amplify the process, toward whole-body function weakening. Senescence initiates at the cellular level and consists in an inability to maintain homeostasis, characterized by the overexpression/aberrant expression of inflammatory/immune/stress responses. Aging is associated with significant modifications in immune system cells, toward a decline in immunosurveillance, which, in turn, leads to chronic elevation of inflammation/oxidative stress, increasing the risk of (co)morbidities. Albeit aging is a natural and unavoidable process, it can be regulated by some factors, like lifestyle and diet. Nutrition, indeed, tackles the mechanisms underlying molecular/cellular aging. Many micronutrients, i.e., vitamins and elements, can impact cell function. This review focuses on the role exerted by vitamin D in geroprotection, based on its ability to shape cellular/intracellular processes and drive the immune response toward immune protection against infections and age-related diseases. To this aim, the main biomolecular paths underlying immunosenescence and inflammaging are identified as biotargets of vitamin D. Topics such as heart and skeletal muscle cell function/dysfunction, depending on vitamin D status, are addressed, with comments on hypovitaminosis D correction by food and supplementation. Albeit research has progressed, still limitations exist in translating knowledge into clinical practice, making it necessary to focus attention on the role of vitamin D in aging, especially considering the growing number of older individuals.

Vitamin D as a Shield against Aging Aging can be seen as a physiological progression of biomolecular damage and the accumulation of defective cellular components, which trigger and amplify the process, toward whole-body function weakening. Senescence initiates at the cellular level and consists in an inability to maintain homeostasis, characterized by the overexpression/aberrant expression of inflammatory/immune/stress responses. Aging is associated with significant modifications in immune system cells, toward a decline in immunosurveillance, which, in turn, leads to chronic elevation of inflammation/oxidative stress, increasing the risk of (co)morbidities. Albeit aging is a natural and unavoidable process, it can be regulated by some factors, like lifestyle and diet. Nutrition, indeed, tackles the mechanisms underlying molecular/cellular aging. Many micronutrients, i.e., vitamins and elements, can impact cell function. This review focuses on the role exerted by vitamin D in geroprotection, based on its ability to shape cellular/intracellular processes and drive the immune response toward immune protection against infections and age-related diseases. To this aim, the main biomolecular paths underlying immunosenescence and inflammaging are identified as biotargets of vitamin D. Topics such as heart and skeletal muscle cell function/dysfunction, depending on vitamin D status, are addressed, with comments on hypovitaminosis D correction by food and supplementation. Albeit research has progressed, still limitations exist in translating knowledge into clinical practice, making it necessary to focus attention on the role of vitamin D in aging, especially considering the growing number of older individuals. The undeniable increase observed in human life expectancy and longevity in developed countries is too often accompanied by a decrease in the quality of life (QoL) [1]. This means that living longer is not living better, thus, the current challenge is how to maintain and improve a good QoL during a longer lifespan. The aging process is a gradual functional decline, that starts early in adulthood, continues throughout the human lifespan, and might end in chronic diseases, e.g., sarcopenia, metabolic disorders, cardiovascular and neurodegenerative diseases, or cancer [2]. These macroscopic age-related alterations reflect the changes occurring at the cellular and molecular levels: mitochondrial dysfunction, damaged protein accumulation, epigenetic alterations, telomere shortening, aberrant intracellular signaling, and altered nutrient sensing are recognized to be the biomolecular pillars of aging [3,4]. Altogether, these molecular alterations converge in frailty, a multidimensional syndrome, clinically marked by decreased physiological reserve and resistance to stressors, leading to a higher rate of disability and mortality [5]. Frailty is undeniably a multifaceted condition, due to different determinants, however, the reduced capacity in immune defense, known as “immunosenescence”, and the consequent increase in chronic low-grade inflammation, named “inflammaging”, emerge as the main triggers. Whereas aging is a natural and unstoppable process, several adaptable lifestyle factors can favorably impact QoL, by strengthening the immune system and promoting anti-inflammatory processes. It is well-known that diet can drive human health status toward wellbeing or disease. Indeed, the nutritional pattern per se plays a pivotal role in promoting healthy aging, by modulating several intracellular processes. Among many factors, vitamin D plays an important role as a nutrient capable of affecting the aging process at a cellular/molecular level, with a wide range of actions. Indeed, beside bone defects, insufficient vitamin D is associated with the increased risk of developing a wide range of pathologies, from neurological diseases to cancer, obesity, diabetes and metabolic disturbance, cardiovascular diseases, autoimmune diseases, infections, and menopause-related diseases [6,7,8,9,10,11,12,13]. A clear cause–effect relationship still lacks in many cases, nevertheless hypovitaminosis D is a marker for poor health, especially in the elderly, often characterized by co-morbid conditions [14]. Most of the actions of vitamin D are undoubtedly ascribable to its fine-tuned immunomodulatory properties, impacting immune system function and inflammation. This review aims to provide an overview of some of the cellular biomechanisms and biomolecules engaged in immunosenescence and inflammaging, as biotargets for vitamin D interventions during adult life, or even earlier, in order to maintain a level of health as high as possible while aging. Particular attention will be given to the ability of vitamin D to shape the human immune status, acting on the immune system or immune-related activity of some tissues, such as skeletal muscle or heart, whose function highly affects QoL during life, including in old age. Human aging is a complex process influenced by several factors, including sex, genetics, socioeconomic status, and lifestyle. Age-related changes are associated with higher susceptibility to diseases, depending on macroscopic modifications in the whole-body physiology and organ function, which, in turn, mirror the changes occurring at the cellular and intracellular levels. Accumulating data suggest that dietary pattern, among other factors, can prevent some molecular/cellular changes underlying human aging-related diseases, and maintain the functional abilities associated to wellbeing [World Health Organization Ageing: Healthy Ageing and Functional Ability 2020; https://www.who.int/news-room/questions-and-answers/item/healthy-ageing-and-functional-ability, (accessed on 26 October 2020)]. Some of the main processes triggering and mediating senescence and inflammation are discussed below. Understanding the molecular drivers of age-dependent multimorbidity can help to develop new interventions and strategies to delay this condition [15]. The multimodal process of aging includes different biological system remodeling leading to the loss of homeostasis and, consequently, deterioration of several organs and tissues. In this scenario, age-related biomolecular changes in the immune response and immune cells are acknowledged to play a causal role in driving whole-body aging [16]. In humans, the natural decline in immune system function usually starts in the sixth decade, and tends to continuously progress toward immunosenescence and decreased protection against pathogens; at the same time, the inflammatory response increases, in terms of duration/intensity [17]. Age-dependent “re-shaping” of both the innate and adaptive arms of the immune landscape can increase vulnerability to illness, and promote and allow multimorbidity, frailty, and adverse health outcomes [18]. Indeed, whereas the adaptive immunity decreases and leads to weakened antigen-specific response and impaired memory formation, the nonspecific innate immunity overreacts, leaving older individuals unprotected from chronic inflammation. The concurrence of these conditions is defined as “inflammaging” [19,20]. In turn, inflammaging is an additional risk factor in the elderly, since it amplifies other age-associated modifications related to morbidities. The macroscopic changes reflect microscopic remodeling at the cellular and subcellular levels, involving significant modifications in intracellular paths, especially in the T cell repertoire [21,22,23]. Cellular senescence combines morphological and molecular characteristics, makes cells modify cyclin-dependent kinases (CDKs) regulation, and undergo permanent cell cycle arrest, in a steady status different from terminal differentiation or quiescence [24,25]. The diluted cytoplasmic domain, the enlarged cell size, and increased β-galactosidase lysosomal activity—mostly used as a senescence cell marker—are the major features of senescent cells, along with an altered DNA and chromatin landscape, a damaged protein and lipid profile, and dysfunctional mitochondria and lysosomes [24,26,27]. Remarkably, senescent cells display a “senescence-associated secretory phenotype” (SASP) due to a specific secretory profile or “secretome”, which is characterized by an excessive secretion of proinflammatory molecules, such as cytokines and chemokines, matrix metalloproteinases, and growth and angiogenic modulators [28,29,30]. Figure 1 shows a schematic of the cellular changes associated with senescence. In this scenario, the production and release of proinflammatory molecules increase, including interleukin (IL)-6, IL-1, tumor necrosis factor(TNF)α, interferon (IFN)γ, acute-phase proteins, reactive oxygen species (ROS), and autoantibodies constantly increase and accumulate. These factors function as mediators of signal communication between cells resident within organs/tissues, and immune cells, and thus amplify inflammaging [31,32]. To date, several in vitro and in vivo studies show that the senescence-related signature is highly variable, but a robust approach to establishing a senescent secretome profile is still lacking [33]. Nevertheless, senescence of the immunocytes is undeniably acknowledged as a critical step in triggering sustained chronic-persistent inflammation, therefore, representing a potential target for interventions. The central defect in immunosenescence is the decline in T cell function, as shown by animal and human studies, leading to overactivity, inflammation, and autoantibody production, with quite predictable consequences on health [34,35,36,37]. The main age-dependent changes include a significant decrease in naïve T cells, along with a simultaneous increase in memory cells, impairments in the T cell receptor (TCR) repertoire, defects in NK cells and neutrophil chemotaxis, myeloid skewing, and monocyte dysregulation. Albeit immunosenescence involves most of the immunity-related components, T cells emerge as playing a triggering role in aging, as shown by investigations on rheumatoid arthritis (RA), an autoimmune disease in which the faster progression of this process allowed the characterization of some underlying molecular paths [38,39,40]. Age-related thymus involution is the first event that triggers impairments and defects in T cell differentiation and maturation. The number of new naïve T cells decreases, and circulating naïve T cells live longer, accumulate defects, and cause a decreased TCR diversity [34]. T cell-TCR signaling activation is required for IL-2 release and T cell expansion, which are known to be critical for mounting an effective immune response [41]. In aging, the downtuning of T regulatory (Treg) cells leads to an unbalanced Th17/Treg ratio, with Th17 upregulation and Treg decrease, ending in polarization of the immune response towards inflammation [42]. Furthermore, age-related thymic involution, naïve T cell dysfunction, and aberrant expression of age-associated gene profiles seem to be linked to IL-33/ST2 signaling activation, suggesting that targeting IL-33 or ST2 could be a promising strategy to rejuvenate T cell immunity [43]. IL-33 has been shown to orchestrate the signals within the skeletal muscle/immune system/nervous system in response to injury. Remarkably, the IL-33/ST2 axis is suggested as a promising potential target in managing age-related sarcopenia and muscle repair due to injury or atrophy, both major problems impacting mobility and QoL in the elderly [44]. The dysregulation of T cell signaling includes defects in calcium mobilization, phosphorylation of tyrosine and serine/threonine, mitogen-activated protein kinases (MAPK) activity, and activation of transcription factors such as nuclear factor kB (NF-kB), nuclear factor of activated T cells (NFAT), or activator protein 1 (AP-1) [45,46,47,48,49,50,51,52]. Other age-dependent damage, i.e., the loss of co-stimulatory molecule CD28—which plays a pivotal role in cell activation/proliferation/survival—impacts on the function of B cells (proliferation and Ig production) and the antigenspecific cytotoxic CD8 T cell subset [34]. In addition to T cell deregulation, aberrant increases in NF-kB activation and cyclooxygenase (COX)-2 expression, with a higher production of prostaglandin E2 (PGE2) from macrophages, have been reported in the elderly [53,54,55]. Remarkably, NF-kB dysregulation/hyperactivation represents a potential molecular target for intervention, with particular significance in the elderly, as addressed later in this review. The number of natural killer (NK) cells, which are known to have cytotoxic/lytic activity against cancers and viruses, seems to not vary with aging, whereas cytokine and chemokine production declines [56,57,58]. Like NK, neutrophil number does not change with age, but neutrophil activity, including oxidative burst, phagocytosis, and chemotaxis (all defense mechanisms) is significantly compromised [57]. In the complex network of aging, oxidative stress is highly determinant, as suggested by the coined term “oxi-inflamm-aging” [59]. Indeed, the high percentage of polyunsaturated fatty acids present in immune cell plasma makes the cells inclined to lipid peroxidation; any damage in the oxidative process can alter signal transmission within/between several types of immunocytes, ending in a defective immune response [60,61]. Thus far, age-related changes in the immune system expose the elderly to a higher risk of disease, first of all, cancer, and infections. The significant impact of genetic and environmental factors on immune system function is undeniable, and nutrition emerges as an acknowledged tool to regulate the immune status. Indeed, age-related immune alterations seem to be associated with a suboptimal nutrient status, so that nutritional interventions with macronutrients (i.e., polyunsatured fatty acids or PUFA) or micronutrients (i.e., vitamins, elements) are highly recommended for ameliorating QoL in the elderly. Whereas macronutrients provide a substrate for the biosynthesis of molecules engaged in immune response (acute-phase proteins, cytokines, new receptors, amino acids for immunoglobulins), and are fuel for immune cell energy, micronutrients, such as zinc, iron, and vitamins are exquisite regulators of the immune response at the cellular and molecular level [62]. Overall, this review focuses on vitamin D as a fine-tuned regulator of both the innate and adaptive immune responses, playing a pivotal role in the health of individuals of any age, particularly, in the elderly. Vitamin D has historically been known as the necessary nutrient to guarantee a correct bone metabolism and health. This molecule is classically defined as a steroid hormone, since it shares with steroids the common progenitor molecule (cyclopentanoperhydrophenanthrene). It is synthesized as the precursor molecule in skin exposed to sufficient UV rays, and is transformed into the biologically active compound by two enzymatic hydroxylations in the liver and kidneys (25- and 1-α-hydroxylase, respectively). Therefore, while “vitamin” is considered an imprecise term [63], this molecule is present in food such as fatty fish, i.e., salmon and trout, beef liver, cheese, egg yolk, albeit the latter ones provide small amounts, mushrooms and some vegetables, and is fully identified as an essential micronutrient with critical regulatory functions [64]. Vitamin D deficiency is a worldwide problem, affecting people of all ages, due to several impacting factors, from ethnicity and skin color, to latitude, habits and lifestyle, sex, and age. It has been reported that the elderly in Europe, the USA, and Australia suffer from insufficient levels of vitamin D, mainly due to less solar exposure, associated with low levels of outdoor activity or with clothing, and also due to reduced synthesis, due to atrophic skin modifications and a lower amount of precursor or reduced renal function [65,66]. Remarkably, a less varied diet, with a lower vitamin D content, is a typical age-associated habit, contributing to vitamin D deficiency in aging populations [67,68]. Thus far, vitamin D intake to a sufficient level is recommended not only for bone homeostasis but to maintain a general status of good heath as well, especially during aging, considering its pleiotropic effects. Vitamin D intake from dietary sources, in particular foods enriched or fortified with this nutrient, is considered an excellent strategy to counteract vitamin D deficiency, a condition currently defined as being a “world problem” [69]. It is acknowledged that this nutrient exerts such an important effect on the immune response that it can be considered a tool to tackle immunosenescence and oxi-inflamm-aging. In general, vitamin D deficiency is associated with a higher risk of infections and autoimmune diseases, involving dysfunctional biological activity of the specific vitamin D receptor (VDR), which is expressed in the majority of immunocytes [70]. Hypovitaminosis D is acknowledged as a pandemic condition, affecting all stages of life, with detrimental consequences on health; but in the elderly, vitamin D deficiency retains even more clinical significance, since this condition often converges with other age-related deficiencies (i.e., hormonal) or diseases, worsening the outcome. Most of the immune cells, including T and B cells, dendritic cells (DC), macrophages and monocytes, express VDR and respond to vitamin D with fine-tuned modulations in cell signaling, path activation, and molecule production, with significant consequences on immune response [71,72,73,74,75,76,77,78]. Furthermore, many immunocytes express 1-α-hydroxylase and can themselves produce the active metabolite and, thus, control the local cell microenvironment [79,80]. In the elderly, adequate vitamin D levels help to counteract the natural decline in immune surveillance by a fine-tuned orchestration of several effects. Vitamin D strengthens the first line of host defense, that is particularly relevant during aging, when the risk of infection is higher. Indeed, this molecule can maintain the barrier integrity and induce a set of genes encoding antimicrobial proteins (AMPs), such as cathelicidin, defensins, hepcidin, and neutrophil peptides, which behave as antibiotics against various types of pathogens [81,82]. The vitamin D–cathelicidin axis is the most studied and best characterized among the VDR-dependent signaling engaged in infections and autophagosome formation—the latter one playing a critical role in microorganism clearance and infection resolution. Clinical studies on human tuberculosis, sepsis, viral infection, peritonitis, and pneumonia, whose incidence rise in the elderly [83], document that, after vitamin D supplementation, serum cathelicidin—human cathelicidin LL-37 or human cationic AMP 18 (hCAP-18)—increases, and correlates with improved clinical outcomes [84,85,86]. Evidence in human monocytes/macrophages shows that the activation of vitamin D signaling, triggered by toll-like receptor (TLR)2/1 or TLR8, leads to an increase in antimicrobial response, autophagy, antimicrobial peptide expression, and phagosome–lysosome fusion, in association with an increase in IFNγ, which is likely required to strengthen antibacterial activity, together with IL-12 and IL-18 [87,88,89,90]. Furthermore, IFNγ expression, in combination with CD40–CD40 ligand signaling, increases the activity of the hydroxylase converting 25-hydroxyvitamin D (25D) to the active metabolite, in human monocytes [91]. In Mycobacterium tuberculosis, the most studied infection, vitamin D-induced cathelicidin acts as a second messenger to activate autophagy genes, such as autophagy related 5 (ATG5) and beclin-1 (BECN1), and triggers a downstream wide signaling cascade, including intracellular calcium (Ca2+) release, Ca2+-dependent kinases, extracellular ATP-gated ion channel, purinergic receptor (P2X) 7, mammalian target of rapamycin (mTOR)/AMP-activated protein kinase (AMPK)/phosphoinositide 3-kinase(PI3K) pathway, and reactive ROS signaling [92,93,94]. During Mycobacterium infection, vitamin D boosts cytokines/chemokines production through the induction of IL-1β, that regulates defensin beta 4 gene (DEFB4), encoding for human beta-defensin-2 (HBD2) in macrophages [95]. Thus far, vitamin D/VDR signaling acts as a fine regulator of AMP-dependent autophagy, cytokines/chemokines production, IFN-dependent signaling, and ROS generation [71,96,97]. Since ROS mediates TLR2-induced cathelicidin expression in human monocytes/macrophages [98], and ROS-autophagy events are mutually regulated [99], further understanding the function of the vitamin D–cathelicidin axis in redox homeostasis and autophagy activity seems necessary. The antimicrobial action(s) of host defense proteins induced by vitamin D might be particularly relevant, since infections from drug-resistant pathogens are emerging worldwide [100]. Remarkably, vitamin D-mediated LL-37 induction is specific in human defense system, as this vitamin D function has been reported to fail in a murine system [101]. This observation is in line with the concept that, results from animals are often not translatable to humans, especially concerning studies on immunity [96]. However, the specific role of the vitamin D–cathelicidin axis in the different infections caused by bacteria, viruses, and parasites goes beyond the aim of this paper, and was exhaustively reported in another recent paper [88]. To date, whereas inflammatory cytokines such as TNF-α and IL-1β are central within vitamin D-mediated antibacterial activity during host defense, alterations in the vitamin D–cathelicidin axis, leading to lower vitamin D and excessive cathelicidin, result in overinflammation, and take part in the pathological setting of chronic inflammatory diseases, such as rosacea, a disease whose prevalence increases with age [102,103]. The paradoxical effect of vitamin D emerges; while this molecule works on the one hand to keep active the macrophage phenotype (M1), associated with proinflammatory/anticancer activity, to preserve physiological integrity, on the other hand, it helps to maintain the pro-tolerogenic/anti-inflammatory phenotype (M2), to counteract inflammaging, that is the route opening to a variety of diseases. The main anti-inflammaging effects related to vitamin D are addressed in the following paragraph. Figure 2 depicts the effects of vitamin D against inflammation, as well as some other molecular effects, reported in the following paragraphs. Following the original concept, inflammaging is a consequence of immunosenescence, even if this model has been re-interpreted in favor of a mutual interplay [104]. The immunomodulatory role of vitamin D in inflammation is widely recognized. Substantially, an adequate level of vitamin D counteracts inflammation with multilevel targeting effects, i.e., inhibiting the expression and signaling of TLR2, 4, and 9, reducing the production of cytokines such as TNF-α, IL-6, IL-23, and IL-1, and repressing the activity of T cells recruiting chemokines [105]. The main inhibitory effects are on CD4+ and CD8+ T cell proliferation, in particular T helper 1 (Th1) cells (a subset of CD4+ effector T cells), in which vitamin D counteracts the release of cytokines, i.e., IL-2 or IFNγ, capable of activating macrophages, and IL-6, IL-8, IL-12, and TNFα, all molecules characterizing inflammaging [106,107]. The vitamin D-induced downregulation of the proinflammatory Th17 cell subset, suppression of dendritic cell (DC) maturation from monocytes, and impaired capacity to present antigens, occur together with Treg enhancement and an increase in anti-inflammatory cytokines such as IL-4, IL-5, IL-10, and CCL2. Altogether, these processes are supposed to be the main mechanisms underlying Th2 protolerogenic subset expansion, which, in turn, is able to mitigate inflammaging and autoimmune disorders [108,109,110,111,112,113]. In lymphocytes, vitamin D has been documented to inhibit IL-6, a critical factor in stimulating the Th17 cell subset, which plays a pivotal role in autoimmune reactions [114]. From in vitro studies on autoimmune diseases, the vitamin D-induced protolerogenic effect on T cells seems undeniable, but less effectiveness in vivo is hypothesized, maybe related to a different ability of T cells to modify their phenotype in response to vitamin D (more phenotypically committed, less responding to vitamin D) [115]. The precise biomolecular mechanism remains to be elucidated. Albeit the etiology of autoimmune diseases is complex and undeniably multifactorial, the role of hypovitaminosis D is acknowledged as highly impacting disease development [116]. In some autoimmune diseases, such as osteoarthritis, psoriasis-associated osteoporosis, and Guillain–Barré syndrome, a pathogenetic immunological crosstalk between vitamin D and IL-33 may be hypothesized, with the combination of hypovitaminosis D–IL-33/ST2 axis activation converging in deleterious effects. Hence, the neutralization of IL-33/ST2 signaling by vitamin D in immunocytes (T cells, DC) is suggested as a therapeutic strategy [117,118,119]. In this scenario, the Th17/Treg ratio tends to decrease simultaneously with a significant increase in the transcription factor fork head box (Fox)P3, the most reliable marker to date for Treg cells, and IL-10 [120]. Remarkably, IL-10 is the anti-inflammatory cytokine known to keep inflammaging and antigenic stress under control, thus representing a good molecular defense mechanism in the elderly [121]. In line with this concept, the poorer vitamin D status—i.e., due to reduced sunlight exposure, declined ability of the skin to produce vitamin D, malnourishment, or decreased vitamin D intake—observed in elderly people, likely contributes to the higher prevalence of a variety of age-related diseases associated with a compromised immune system [122,123,124,125]. As shown in COVID-19 patients, a wide-epigenetic T cell remodeling, promoting VDR expression and enzyme cytochrome P450, family 27, subfamily B, member 1 (CYP27B1) activation in autocrine/paracrine mode, likely underlies the transition from proinflammatory IFN-γ+ Th1 cells (via STAT3, c-JUN, and BACH2) to suppressive IL-10+ cells (via IL-6–STAT3 signaling) [126,127]. In addition to the effect on immunocytes, vitamin D deficiency increases endothelial senescence, allowing vascular dysfunction and atherosclerosis, both processes that increase in prevalence with aging. This effect is undeniably related to vitamin D’s anti-inflammatory action, as inflammation is a recognized trigger of atherosclerosis initiation, progression, and plaque and thrombus formation, also in the young [128]. However, the vitamin D-induced reduction in cholesterol uptake by macrophages and, in turn, suppression of foam-cell formation, emerges as a key event [129]. So far, hypovitaminosis D supports increased cellular senescence and arterial aging, characterized by the typical dysfunctions as a gradual loss of vascular smooth muscle cells’ contractility, and increased arterial permeability and intima thickness [130]. The delayed cellular senescence by vitamin D/VDR interaction includes different processes, including longer telomere lengths, an increased antioxidant effect via nuclear factor erythroid 2-related factor 2 (Nrf2) transcriptional regulation, decreased oxidative stress, DNA damage, and SASP, downregulation of p16, p53, and p21 (major regulators of the G1/S cell-cycle checkpoint), while upregulating Bmi1(polycomb ring finger oncogene, transcriptional suppressor), whose novel action is reported in cardiac regulation [131,132]. Another important aspect is the regulation of mitochondrial function. Vitamin D deficiency is associated with disorders of mitochondrial function, such as respiratory chain deregulation, with downregulation of mRNA and proteins involved in mitochondrial respiration, inhibition of sirtuin (SIRT) 1, which plays a pivotal role in mitochondrial biogenesis through PGC-1α, and, e.g., in brain aging delay [133,134]. The combination of vitamin D and curcumin, given as a supplement in a continuous way, has been recently hypothesized to counteract neurodegeneration [135]. SIRT signaling and lifespan seem deeply affected by dietary nutrient composition and resveratrol, which suppresses oxidant and inflammatory genes, altering promoter epigenetic status. The immune system is key to a host’s defense against pathogenic organisms. Aging is associated with changes in the immune system, with a decline in protective components (immunosenescence), increased susceptibility to infectious disease, and a chronic elevation in low-grade inflammation (inflammaging), increasing the risk of multiple noncommunicable diseases. Nutrition is a determinant of immune cell function and of the gut microbiota. In turn, the gut microbiota shapes and controls the immune and inflammatory responses. Many older people show changes in their gut microbiota. Age-related changes in immune competence, low-grade inflammation, and gut dysbiosis may be interlinked, and may relate, at least in part, to age-related changes in nutrition. A number of micronutrients (vitamins C, D, and E, and zinc and selenium) play roles in supporting the functions of many immune cell types. Some trials have reported that providing these micronutrients as individual supplements can reverse immune deficits in older people, and/or in those with insufficient intakes. There is inconsistent evidence that this will reduce the risk or severity of infections, including respiratory infections. Probiotic, prebiotic, or synbiotic strategies, that modulate the gut microbiota, especially by promoting the colonization of lactobacilli and bifidobacteria, have been demonstrated to modulate some immune and inflammatory biomarkers in older people, and in some cases, to reduce the risk and severity of gastrointestinal and respiratory infections, although, again, the evidence is inconsistent. Further research, with well-designed and well-powered trials, in at-risk older populations is required to be more certain about the role of micronutrients and of strategies that modify the gut microbiota–host relationship in protecting against infection, especially respiratory infection [136,137]. COX-2 expression has been shown to be dose-dependently inhibited by vitamin D in murine macrophages, through suppression of the protein kinase B (PKB) or Akt/NF-kB/COX-2 pathway [138]. The function of NF-kB is critical in cellular senescence and in inflammaging, since this factor is suggested to serve as a linkage between aging hallmarks in cell–cell communication, and in age-related pathophysiological mechanisms [139,140]. Indeed, while NF-κB unaltered signaling supports the correct interplay between immunocytes and non-immune cells, to maintain a functional host response, NF-kB hyperphosphorylation and enhanced activity are reported in different tissue aging (skin, hypothalamus, and cortical tissues) [141,142]. Similarly, NF-kB constitutive activation is reported in aged skeletal muscle [143]. To date, in vivo data in overweight/obese, but otherwise healthy, individuals failed to show any significant difference in NF-kB activity; however, the inadequacy of sample size to detect such a difference as a primary outcome is indicated by the authors as a possible reason for their null findings [144]. Nevertheless, the function of NF-kB as a regulator of inflammation and immunity continues to emerge as being critical for aging and age-related diseases [145]. NF-kB signaling hyperactivation is associated with pro-aging stimuli, and correlates with the development/progression of most aging-related diseases, whereas NF-kB inhibition can delay or even reverse aging processes [146,147,148,149]. Thus far, inflammaging likely primes inflammatory NF-kB signaling, which is involved in a variety of age-related tissue/organ dysfunctions. Interestingly, vitamin D can target hyperphosphorylation of this transcription factor in several cell types. The following paragraph focuses on this topic. NF-kB can be considered as the central crossroad where many paths and signals, either promoting or delaying aging, converge, by activating or inhibiting this transcriptional factor, respectively [147]. Biomolecular paths promoting aging such as insulin (I)/insulin growth factor (IGF)-1 signaling, activate NF-kB via the PI3K/AKT cascade and mTOR, a known pro-aging factor [146,150,151,152]. In addition, age-dependent DNA damage and telomere shortening are associated with NF-kB aberrant activation, and with increased levels of COX-2 and ROS [153]. NF-kB activation, besides telomere shortening via telomerase reverse transcriptase (TERT) catalytic subunit activation, intensifies inflammation by macrophage polarization to M1 phenotype, and IL-6 and TNFα upregulation [154,155], and increases SASP and cellular senescence [147]. At variance to this, pro-longevity factors, i.e., sirtuin and FOXO, repress NF-kB transcriptional activity, by directly interacting with the p65 subunit [156,157,158]. To date, chronic activation of NF-kB is found in several age-related diseases, e.g., atherosclerosis, osteoporosis, muscular atrophy, and neurodegeneration [159,160,161,162,163]. Quite remarkably, vitamin D can downregulate and suppress NF-kB expression and activity, significantly limiting inflammation. The mechanism whereby the vitamin D/VDR system targets NF-kB includes a physical interaction between VDR and IκB kinase β (IKKβ), which is enhanced by vitamin D, resulting in IκBα stabilization and p65/p50 nuclear translocation impairment [164]. The blockade of p65 translocation leads to the decrease or suppression of NF-kB activation and transcriptional activity, as shown in different cell types stimulated with TNFα—the prototypic cytokine inducing this nuclear factor—not only in immune cells, i.e., DC or B cells, but also in different types of organ resident cells, like human cardiac cells, skeletal muscle cells, thyrocytes, and nucleus pulposus cells [165,166,167,168,169,170]. Thus far, a well-functioning vitamin D/VDR system represents a helpful tool for tackling inflammation and aging, through the downregulation of the TNF-α/NF-kB/p65 signaling cascade in immune system cells and in different tissue cells. This effect seems particularly relevant at the level of striated cells, such as cardiomyocytes or skeletal muscle cells. It is acknowledged that aging per se is a risk factor for heart dysfunction and disease. Age-related cardiac reduced function reflects molecular and cellular modifications, both in non-cardiomyocyte-based components, i.e., vascular cells, fibroblasts, and extracellular matrix, and in cardiac cells, which undergo aberrant processes due to oxidative stress, inflammation, defects in metabolism, cellular repair, telomeres, alteration in gene expression, and post-translational modifications. The dogma that cardiomyocytes are postmitotic cells, terminally differentiated, and unable to undergo cell division, is overcome by the observation that a fraction of cells proliferate and divide in the hearts of the young, adults and the elderly, albeit the cardiomyocyte numbers undeniably decrease with aging, by necrosis [171]. The increased number of necrotic cells likely triggers repair-related aberrant inflammation, and a higher chance of autoantigen generation, due to oxidatively modified proteins [172]. Following oxi-inflamm-aging-induced cell death, garbage molecules such as lipoxidation products and advanced glycation end (AGE) products, accumulate from alterations in the protein degradation machinery and autophagy. Concerning autophagy and apoptosis, it should be mentioned that a functional vitamin D/VDR system—rather than a sufficient vitamin D level alone—can regulate senescence-induced signaling, either by genomic and non-genomic mechanisms, in immunocytes and non-immune cells, retaining the potentiality to be used as a helpful tool against diseases related to inflammation, oxidative stress, or cancer [173,174,175,176]. These processes would impact gene expression toward aging memorization and progress [177]. During cardiac physiological aging, ROS generation is associated with the activation of the inflammasome NOD-like receptor protein 3 (NLRP3), production of biologically active IL-1β and IL-18, and, in turn, pyroptosis, a process causing cell membrane pore formation/rupture and cell death [178,179,180,181]. The activation of NLRP3 is a multistep process, occurring either with or without a priming signal first, denoted as “canonical” or “noncanonical” activation, respectively [182,183,184]. The difference between the canonical/noncanonical mechanisms are exhaustively reported elsewhere, and are beyond the aim of this review. Independent of the mechanism of activation, an aberrant inflammasome is widely involved in aging, and in an extraordinary number of human age-related diseases. Of note, vitamin D/VDR signaling directly acts as a negative regulator of NLRP3 oligomerization/assembling/activation and IL-1β release, inhibiting pyroptosis [185,186,187]. There is evidence for IL-1β and IL-18 increase from VDR knockout macrophages, activated either through the canonical or noncanonical path [187]. Indeed, ligand-activated VDR attenuates NRLP3 deubiquitination, increasing the level of mitochondrial membrane uncoupling protein-2 (UCP2), which can directly prevent ROS production [188]. Albeit NLRP3 inflammasome activation is known to play a pivotal role in host defense from viral and bacterial infections, dysregulated or excessive activation of the inflammasome is associated with poor outcomes [189]. Similar to cardiac cells, skeletal muscle cells show age-dependent biomolecular dysregulation, associated with mitochondrial dysfunction, oxi-inflamm-aging, metabolic disturbance protein breakdown, accumulation of senescent cells, and tissue atrophy [190,191,192]. Age-related decline of skeletal muscle can lead to sarcopenia—not limited to muscle area/mass reduction, but extended to tissue functionality—and frailty [5,193]. In muscle as well, aging-related macroscopic changes reflect biomolecular and cellular modifications. Indeed, tissue changes and deficit are associated with an increase in several inflammatory cytokines, including TNFα, IL-1α/β, IL-6, IL-8, IFNγ, and their soluble receptors (sR) IL-1Ra, TNFαsR, and IL-6sR, referred to as “gerokines”, which altogether constitute the “aging secretome”, likely playing a causal role (rather than being simply markers) in inflammaging [19,194,195]. Even though there is still much to understand about how oxi-inflamm-aging translates into skeletal muscle decline, it is undeniable that the reduction in type II muscle fiber number and size, promoted by TNFα-induced apoptosis, and the unbalance in protein synthesis/degradation, induced by ROS and oxidative stress, merge into functional and mechanical muscle impairment [196,197]. The rise in TNFα or IL-1β increases fat mass toward an unfavorable muscle/fat ratio, which typically characterizes the elderly, especially those with sedentary habits [197]. To date, some intracellular cascades, particularly the TNFα/NF-kB/ubiquitin–proteasome system (UPS) cascade/Akt signaling, play a large part in skeletal muscle inflammaging, representing important targets to counteract skeletal muscle decline. Some strategies, such as protein supplementation, can reduce creatine kinase, but cannot affect circulating gerokines [198,199,200,201]. Another feature of aging muscle is the “differential” resistance to I regarding glucose, protein, and lipid metabolism. Indeed, in the elderly it is common that sensitivity to I concerning glucose is maintained, likely related to reduced utilization of peripheral glucose, while almost simultaneously the so-called “anabolic resistance” to I develops, due to protein synthesis/degradation imbalance (proteostasis loss), a condition preceding clinical manifestations [202,203,204,205,206]. The protective function exerted by vitamin D on skeletal muscle is acknowledged: i.e., vitamin D can directly interfere with and counteract the intracellular paths mediating tissue atrophy, such as proto-oncogene tyrosine-protein kinase Src/extracellular signal-regulated kinases 1 and 2 (ERK1/2)/Akt/forkhead box O3 (FOXO3), a signaling cascade, which upon activation, upregulates atrophic markers such as Atrogin-1 and MuRF1 [207]. An adequate intake of vitamin D can delay the aberrant processes associated with aging and age-related diseases, essentially restoring mitochondrial function and counteracting oxi-inflamm-aging. In myocytes, vitamin D can limit or even neutralize oxidative stress and ROS generation, and the accumulation and expression of garbage molecules like AGE/AGE receptor (RAGE), and high-fat diet-induced I resistance and myosteatosis [208,209,210]. Some experimental and human studies have shown that vitamin D is an effective antioxidant, via activation of the Nrf2-Keap1 antioxidant pathway (the main inducible defense against oxidative stress), showing, e.g., higher capacity than vitamin E to reduce zinc-induced oxidative stress in the central nervous system [211,212,213]. Of interest, in skeletal myocytes, vitamin D anti-aging effects are associated with the inhibition of NF-kB, the crossroad where different oxi-inflamm-aging/senescence paths merge, as previously reported [164,167,214]. Thus far, given the evidence on vitamin D’s role in regulating myocyte metabolism/mitochondrial function/ROS generation, vitamin D deficiency (below 25 nmol/l) is of great interest, particularly in the elderly, since this condition is common among community-dwelling elderly, and very common among institutionalized elderly [67]. To date, it should be mentioned that, while undeniably hypovitaminosis D enhances oxi-inflamm-aging and senescence, an overcorrection of vitamin D status may negatively impact on skeletal muscle cell metabolism, similarly to an overdose of antioxidants. Numerous investigations, indeed, have reported on the potential harmful effects of antioxidant (over)supplementation, especially on muscle fiber formation/muscle regeneration, metabolic homeostasis, and mitochondrial biogenesis [207,215,216,217,218]. The effect of vitamin D deficiency is classically related to reduced musculo-skeletal functions, and increased risk of disability in locomotion. Besides this undeniable effect, low levels of vitamin D per se strongly affect the aging process, as this molecule regulates cell homeostasis, counteracting oxi-inflamm-aging and cellular senescence with multitargeting actions. In this scenario, the challenge is to restore an adequate vitamin D level. Some years ago, fortified foods and drinks were introduced, to supplement diet, e.g., some cereals, plant-based beverages like soymilk, orange juice, and some yogurt and cheese. Fortified flour, cornflakes, and juices are more common in the USA, while in Europe, vitamin D enriched margarine, vegetable oil, and milk are more common [219,220,221]. However, from previous simulation studies, fortified foods seem not to be able to provide sufficient vitamin D to the elderly, therefore, individual supplementation is proposed, often in combination with calcium. A supply of about 20 µg vitamin D (800 IU) per day to people over 70, with 50 µg/day recommended as the safe level, and 600 IU, are recommended in subjects from 1 to 70 years of age. Supplementation with 20 µg vitamin D and 1000/1200 mg calcium increases the vitamin D level, suppresses secondary hyperparathyroidism (lowering the parathyroid hormone/PTH), and improves bone and muscle strength [67]. As found in several studies aimed at reducing osteoporotic fractures in the elderly, a lower amount of vitamin D is required (10 µg/day) in the presence of calcium supplementation (1000 mg) [222]. To date, research studies have explored the efficacy of vitamin D supplementation in the prevention and treatment of chronic conditions of aging, nevertheless the translation to standardized application is lacking, for a number of reasons. There is still an urgent need of clear indications as to what is the best dosage for a supplement, considering, e.g., the comorbidities. Moreover, the lack of a universally accepted standard for vitamin D assessment, i.e., due to variability in methods, strictly connects with the absence of clarity in vitamin D status definition (sufficiency/insufficiency/deficiency), based on a standardized reference range. Plasma 25(OH)D (the more stable analyte) is currently used to assess vitamin D status, but it is recognized that the circulating level of this metabolite varies depending upon season, latitude, clothing, dietary habits, race, pigmentation, skin thickness, sex, and age [67]. All the concerns about vitamin D determination and supplementation are exhaustively reported in the latest consensus statement from the 2nd International Conference on Controversies in Vitamin D [223]. So far, the role of vitamin D status in immunosenescence, inflammaging, and whole-body aging is based on scientifically documented data, and is fully recognized by the scientific literature, but still there are important limitations to translate knowledge into clinical practice, with important medical and socioeconomic consequences, considering the high and growing number of individuals aged 65 and older. The DO-HEALTH multicenter clinical trial, currently running in 2157 community-dwelling European men and women aged 70 and older, combines vitamin D (2000 IU/day) treatment with omega-3 fatty acids intake (1000 mg/day), and a 30-min physical activity (3 times/week home exercise). This trial addresses several health domains (cardiovascular, muscle, bone, brain, and immunity) and hopefully will help to implement clinical practice [224]. This review wants to focus as much attention as possible on these aspects, even though some limitations, such as lack of discussion on sex-dependent variability, or differences between active and sedentary elderly, are present. Given the importance of this topic, more accurate basic and clinical human research is necessary to approach vitamin D status determination, opening the way to future scenarios of personalized treatment.

PMC10002865

Alicja Rabiasz,Ewa Ziętkiewicz

Schmidtea mediterranea as a Model Organism to Study the Molecular Background of Human Motile Ciliopathies

24-02-2023

candidate genes,flatworms,motile cilia,planarians,primary ciliary dyskinesia (PCD),RNA interference (RNAi)

Cilia and flagella are evolutionarily conserved organelles that form protrusions on the surface of many growth-arrested or differentiated eukaryotic cells. Due to the structural and functional differences, cilia can be roughly classified as motile and non-motile (primary). Genetically determined dysfunction of motile cilia is the basis of primary ciliary dyskinesia (PCD), a heterogeneous ciliopathy affecting respiratory airways, fertility, and laterality. In the face of the still incomplete knowledge of PCD genetics and phenotype-genotype relations in PCD and the spectrum of PCD-like diseases, a continuous search for new causative genes is required. The use of model organisms has been a great part of the advances in understanding molecular mechanisms and the genetic basis of human diseases; the PCD spectrum is not different in this respect. The planarian model (Schmidtea mediterranea) has been intensely used to study regeneration processes, and—in the context of cilia—their evolution, assembly, and role in cell signaling. However, relatively little attention has been paid to the use of this simple and accessible model for studying the genetics of PCD and related diseases. The recent rapid development of the available planarian databases with detailed genomic and functional annotations prompted us to review the potential of the S. mediterranea model for studying human motile ciliopathies.

Schmidtea mediterranea as a Model Organism to Study the Molecular Background of Human Motile Ciliopathies Cilia and flagella are evolutionarily conserved organelles that form protrusions on the surface of many growth-arrested or differentiated eukaryotic cells. Due to the structural and functional differences, cilia can be roughly classified as motile and non-motile (primary). Genetically determined dysfunction of motile cilia is the basis of primary ciliary dyskinesia (PCD), a heterogeneous ciliopathy affecting respiratory airways, fertility, and laterality. In the face of the still incomplete knowledge of PCD genetics and phenotype-genotype relations in PCD and the spectrum of PCD-like diseases, a continuous search for new causative genes is required. The use of model organisms has been a great part of the advances in understanding molecular mechanisms and the genetic basis of human diseases; the PCD spectrum is not different in this respect. The planarian model (Schmidtea mediterranea) has been intensely used to study regeneration processes, and—in the context of cilia—their evolution, assembly, and role in cell signaling. However, relatively little attention has been paid to the use of this simple and accessible model for studying the genetics of PCD and related diseases. The recent rapid development of the available planarian databases with detailed genomic and functional annotations prompted us to review the potential of the S. mediterranea model for studying human motile ciliopathies. Cilia and flagella are evolutionarily conserved organelles that form protrusions on the surface of many growth-arrested or differentiated eukaryotic cells, from simple unicellular organisms such as Chlamydomonas to specialized cells in higher animals—fish and mammals [1,2,3]. The cilium is anchored at the cell membrane via the basal body, which extends into the axoneme, the main part of the cilium protruding outside the cell; the transition zone between the basal body and axoneme has a function of ciliary gating [4]. Due to structural and functional differences, cilia can be roughly classified as motile and non-motile (also called primary or sensory), although their detailed classification is much more complicated [5,6]. Motile cilia are ancient organelles that were already present in the Last Eukaryotic Common Ancestor, LECA [2]. Their main function is associated with their ability to move; they also perform some sensory functions [5,7,8]. In unicellular organisms, such as Chlamydomonas, Paramecium, or Tetrahymena, cilia or flagella (structurally related to motile cilia, but much longer) are responsible for the movement of cells. In higher organisms, motile cilia (5–10 um long) form a dense carpet (hundreds of organelles per cell) on the apical side of multiciliated cells (MCCs), and through their coordinated planar beating are responsible for the flow of fluids covering epithelium [9]. Single flagella are responsible for sperm motility [10], and single motile cilia in the mammalian embryonal node regulate the directional flow of signals required for the establishment of left-right patterning [11]. Primary cilia found as singular entities on the surface of almost every differentiated metazoan cell type [1], are not in focus in this review, but it should be mentioned that they evolved from motile cilia. Due to the difference in their structure, they are immotile [12], but they play an important role in the reception and transmission of signals involved in the regulation of cellular processes essential for development and the maintenance of tissue homeostasis. The main signaling pathways coordinated by primary cilia include those regulated by Hedgehog (HH), G-protein-coupled receptors (GPCR), wingless (WNT), receptor-tyrosine kinases (RTKs), and TGFβ/BMP receptors [12,13,14,15]. In the axoneme of a typical motile cilium or flagellum (Figure 1), nine peripheral doublets of microtubules (A and B) surround two single central microtubules (C1 and C2). In the cross-section of the axoneme observed in the transmission electron microscope (TEM), this arrangement is known as the 9 × 2 + 2 pattern. Additional protein elements attached to the microtubules are distributed periodically along the axoneme length. Multiprotein complexes associated with the peripheral A microtubules form characteristic auxiliary structures: outer and inner dynein arms (ODA and IDA), ODA docking complexes, radial spokes (RS), and nexin-dynein regulatory complexes (N-DRC). Several detailed reviews describing the axonemal structure components have been published in recent years, e.g., [16,17,18]. Cilia movement is powered by ODAs and IDAs, which act as ATP-dependent molecular motors. They move in synchrony along the B microtubule of the adjacent doublet. The resulting mutual sliding of the peripheral doublets is restricted by the N-DRCs, which connect neighboring doublets; the net result is a bend of the cilium [26]. Radial spokes, attached to the peripheral A microtubules and transiently contacting the projections of the central microtubules, stabilize the structure and functionally connect the central apparatus to the dynein arms [27]. The cooperation of all these ultrastructural elements results in a coordinated planar beating of motile cilia, with a fast power stroke and a slower recovery stroke occurring in the same plane; for the review see, e.g., [18]. The ciliogenesis of motile cilia is a multi-step process, starting from the cell cycle exit and involving many signaling factors; it has been the subject of many excellent reviews, e.g., [28]. The inhibition of the NOTCH1 signal and activation of the MCIDAS-dependent pathway initiates the biogenesis process specific to MCCs [29]. In MCCs, the MCIDAS pathway orchestrates massive centriole amplification and their docking in the apical cell membrane (through the activation of cyclin O) [30]. The expression of the FOXJ1 transcription factor switches on the synthesis of proteins directly involved in the formation of motile cilia ultrastructure [31]. Cilia elongation and maintenance are possible thanks to the presence of a dedicated protein shuttle system, named intraflagellar transport (IFT), which involves the transport of molecules from the cell body through the basal body, transition zone to the tip of cilia and back [32,33]. Some of the multiprotein ciliary elements of motile cilia (e.g., dynein arms, ODA docking complexes) are preassembled in the cytoplasm in a process that requires the presence of proteins, which are physically not a part of the axonemal ultrastructure [34,35]. Genetically determined dysfunction of cilia is the cause of a large group of diseases, collectively referred to as ciliopathies. With the expanding knowledge of cilia biology and genetics, their number is now estimated as at least 35 [36]. The majority of ciliopathies are caused by the dysfunction of primary/sensory cilia [37]. Consistent with the presence of primary cilia on almost all cell types and their function in basic cellular pathways, their dysfunction affects multiple systems, including kidneys, brain, heart, skeleton, and eyes (e.g., polycystic kidney disease, PKD; nephronophthisis, NPHP; Bardet-Biedl syndrome, BBS; Joubert syndrome, JBTS; Meckel syndrome, MKS; Senior-Locken syndrome, SLSN; retinitis pigmentosa, RP); for the reviews see [36,37,38]. In this review, however, we focus on the diseases caused by the genetically determined defects of the structure and/or function of motile cilia. Primary ciliary dyskinesia, PCD (OMIM244400; population frequency of 1:10,000 to 1:20,000), is the flagship ciliopathy resulting from the dysfuntion of motile cilia [34,35,39,40,41,42,43]. It has to be emphasized that the word “primary” in the name of the disease refers to the fact that PCD is a genetically-based condition, as opposed to a “secondary” ciliary dyskinesia, where the dysfunction of motile cilia is caused by environmental factors. Typical clinical symptoms of PCD reflect the role of motile cilia in different parts of the human body. Defects of cilia on the apical surface of the epithelial cells lining the respiratory tract impair mucociliary clearance; as a result, PCD patients suffer from recurrent respiratory airway infections leading to chronic bronchopulmonary disease, recurrent sinusitis, rhinitis, otitis media, and bronchiectasis. The immotility of sperm flagella is a cause of male infertility, and the immotility of cilia on the epithelial cells lining fallopian tubes reduces fertility in females. Dysfunction of cilia present on the ependymal cells in the brain, which are responsible for cerebrospinal fluid flow, can lead to hydrocephalus. Finally, the defects of single motile cilia in the embryonic node impair the flow of morphogens and body patterning (situs), resulting in the randomization of body organ symmetry (situs inversus totalis in about 50% of PCD patients). Due to the largely nonspecific clinical symptoms of PCD and insufficient diagnostic methods, PCD patients are often diagnosed late. The range and severity of PCD symptoms depend on the mutated gene, reflecting different impacts of the dysfunctional protein on the ciliary structure and/or function [34,35,42,43,44,45]. Theoretically, the diagnostic problems can be overcome by applying genetic tests with high efficiency of mutation detection. However, the genetic basis of PCD is highly heterogeneous, reflecting a large number of proteins involved in the structure and function of motile cilia [46,47]. The inheritance of PCD in most families is autosomal recessive; the X-linked or autosomal dominant inheritance is rare. To date, ~50 genes have been reported to be involved in PCD pathogenesis (Table 1), and the involvement of more is still under investigation [34,35,39,40,42,44] (Table 2). The role of these genes in the axonemal structure, ciliary function, or in the biogenesis of motile cilia has been confirmed in animal models using a range of approaches, including analysis of mutant organisms, knockout of the gene in question by targeted mutation or gene knockdown using double-stranded RNA (dsRNA) or morpholinos. Pathogenic variants in many of the PCD genes result in obvious ultrastructural and/or functional defects of cilia, which can be readily visualized using transmission electron microscopy (TEM) or analysis of the ciliary beat [45,169,170]. The most frequent ultrastructural defects include the absence or shortening of ODA or both ODA and IDA, disorganization of microtubule arrangement, scarcity, or a complete lack of cilia. The defects of the ciliary beat can manifest as immotility, flickering, slow beating, or disturbed pattern of beating [169,170]. On the other hand, mutations identified in some of the genes, e.g., encoding central pair complex proteins, N-DRC proteins, or some of the dynein arm elements, have no clear effect on the axonemal structure or on the cilia beat frequency [102,119,123,139,147,171]. Importantly, even the use of high-throughput genome sequencing fails to detect mutations in known PCD genes in ~1/3 of the patients [34,35]. This may be due to the presence of unknown pathogenic variants, lying outside the most commonly studied coding sequences or impossible to detect with copy number insensitive techniques, as well as the presence of pathogenic mutations in yet-unidentified PCD genes. Moreover, the association of some candidate genes with PCD pathogenesis remains a matter of debate, especially when mutations have been described in single families. In addition, there is an expanding list of genes from the so-called PCD spectrum [172], in which mutations in cilia-related genes are associated with an atypical clinical picture of the disease, with a syndromic presentation [51,52,54,55,173] or without (or very mild) respiratory symptoms [10,102,174,175,176]. Classification and curation of gene-disease relationships involving PCD and related motile cilia disorders are currently the focus of the Motile Ciliopathy Gene Curation Expert Panel, a part of the ClinicalGenome consortium [https://www.clinicalgenome.org/ (accessed on 20 February 2023)] [177]. The overall result of the heterogeneity of heritable motile ciliopathies is that—to better characterize the molecular/genetic basis of the unsolved cases of PCD and to differentiate them from the PCD-like spectrum diseases—the search for new candidate genes potentially involved in the pathogenesis is still needed. The easy access to NGS-based genetic screening of PCD patients increases the chances to reveal new candidate genes. Validation of their impact on motile cilia structure and function requires functional studies. The use of primary cell cultures in such studies requires obtaining respiratory epithelial biopsies from patients with pathogenic variants in candidate genes; the amount of the biological material obtained this way is limited, and often insufficient for detailed biochemical and molecular analyses. Another approach, silencing candidate genes in the in vitro culture of healthy human respiratory epithelium (HRE), is not in a routine laboratory method (reviewed in [178]). Primary HRE cells have limited ability to proliferate in culture. While the recent development of conditionally reprogrammed HRE cell cultures increased the proliferative lifespan of these cells and their ability to differentiate, this model is very demanding and still not sufficiently robust and replicable [178]. This is especially important in functional studies, where genome modification is required to overexpress or silence specific candidate genes. Thanks to the high level of evolutionary conservation of motile cilia [3], a variety of model organisms have been successfully used for studying cilia biology. The same approach, which has allowed explaining the molecular basis of cilia assembly, structure, and function across Eukaryotic species, is widely used as a tool in the functional analysis of candidate genes underlying the pathogenesis of PCD and other cilia-related diseases. For almost half of the causative genes identified during over twenty years of PCD research, their involvement in motile cilia function has been revealed by earlier (non-PCD) forward genetics studies in model organisms, performed to study cilia biology–the identity of proteins, their ultrastructural localization, interactions, and function. The loss-of-function variants in these genes, when found among patients, have been directly associated with PCD pathogenesis. For another half of PCD genes, with deleterious variants identified during the genetic screening of patients, their involvement in motile cilia dysfunction has been confirmed by follow-up reverse genetics studies, involving candidate gene silencing in model organisms. The majority of all these studies had been based on a model of double-flagella unicellular alga, Chlamydomonas reinhardtii. DNAI1, the first gene identified as involved in the pathogenesis of PCD, has been earlier associated with the flagellar dysfunction caused by the mutation of IC78, DNAI1 homolog in a double-flagella unicellular alga, Ch. reinhardtii [88]. Other models, which had supported the role of then-candidate PCD genes in motile cilia dysfunction, include unicellular organisms (P. reinhardtii, Trypanosoma brucei, Tetrahymena thermophila), invertebrates (Drosophila melanogaster, Schmidtea mediterranea) or vertebrates (frog Xenopus leavis, fish Danio rerio/zebrafish, mouse, dog); reviewed in [179,180]. Among these model organisms considered in the context of PCD, relatively little attention has been put to the use of S. mediterranea. S. mediterranea is a representative of freshwater planarians, free-living invertebrates from the phylum Platyhelminthes (flatworms). These animals belong to the group of organisms that have three germ layers (endoderm, mesoderm, and ectoderm), bilateral symmetry, and tissues with separate organs. The manner of reproduction of the freshwater planarians varies across the species and can be exclusively asexual (by transverse fission), seasonally sexual, and exclusively sexual (by the cross-fertilization of hermaphrodites) [181]. Planarians achieved popularity due to their great ability to regenerate after amputation or injury. In some cases, a full organism can be rebuilt after several days from 1/279 piece of a single worm, although the regenerative abilities of planarians are different across the species [182,183,184,185]. This regenerative potential makes planarians practically immortal and enables researchers to use them as efficient model organisms in a variety of studies. There are several hundred species of planarians, but their use as animal models in molecular and genetic studies has been mostly limited to the Dugesiidae family (e.g., genera Dugesia, Girardia, Schmidtea), with the majority of research conducted using two species, S. mediterranea and Dugesia japonica [186]. D. japonica is favored for behavioral studies and toxicology screening, while S. mediterranea is popular and attractive for molecular experiments [187]. Two distinct strains of S. mediterranea exist in nature: a sexual strain (2 cm long) and an asexual strain (slightly shorter). Both are diploid, with four pairs of chromosomes (2n = 8); the asexual form results from a chromosome translocation between the sexual strain chromosomes 1 and 3 [181,188]. A planarian organism has a complex anatomy. The nervous system of flatworms is comprised of a bilobed ‘brain’ with different types of neurons and glia, and two longitudinal nerve cords connected by many transverse nerves [181,185]. Photo-, chemo- and rheoreceptors located at the front of the planarian’s body send signals to the brain, where they are processed, leading to behavioral responses [181,184]. Paired ‘eyes’ located on the planarian head allow the detection of light and shadow, and consist of two types of cells: pigmented optic cup cells, and photoreceptor neurons [189,190]. Due to the lack of respiratory and circulatory systems in planarians, oxygen is obtained and transported by diffusion. A centrally located pharynx is in charge of food intake and removal and is connected to a highly branched intestine, which circulates nutrients within the body. The excretory system (protonephridia) is responsible for the removal of waste products and osmoregulation [181]. Internal organs are surrounded by a mesenchymal tissue, parenchyma, consisting of adult pluripotent stem cells (neoblasts), which are essential for worms’ regeneration ability and comprise ~30% of the cells in the adult animal [185,191]. Planarians possess a set of muscle fibers, organized in longitudinal, diagonal, and circular orientations. The planarians body is covered with an epidermis; the ventral epidermis consists of a single layer of multiciliated cells (MCCs), and gland cells involved in the production and secretion of mucus, which is used by flatworms for protection, locomotion, catching food, and adhesion to substrates [181]. Although planarians do not fully reflect the complexity of the human organism, many of the annotated S. mediterranea’s genes have known orthologs (or at least homologs) in the human genome, and researchers increasingly use S. mediterranea in studies aiming to better understand aspects related to human development and function involving certain cell types or tissues. The maintenance of planarians is relatively easy and cheap, and does not require specialized equipment; only habitat conditions, such as temperature, darkness, feeding, and water culture, have to be provided, and many methodological guidelines have been published [e.g., [188,192,193]]. An important feature of using planarians as a model organism is the easy way to perform simple modifications of their gene expression. This can be achieved by knockdown/silencing genes of interest through RNA interference (RNAi) using double-stranded RNA (dsRNA). DsRNA can be administered to the worms by microinjection, by feeding them with dsRNA-containing bacteria, or with food mixed with free dsRNA [194,195,196]. The efficacy of gene silencing on the mRNA level can be evaluated using reverse transcription polymerase chain reaction (RT-PCR) or quantitative reverse transcription PCR (RT-qPCR), while the gene expression pattern of a silenced gene can be determined using whole-mount fluorescent in situ hybridization (FISH) and whole-mount in situ hybridization (WMISH). The phenotypic effect/s of gene silencing is typically observed within a week or two after implementing the RNAi procedure [188]. The genome of S. mediterranea has been well annotated, which makes this species more attractive than other planarians. The advances in single-cell RNA sequencing increased molecular knowledge about planarian stem cell differentiation, and have allowed for determining the transcriptomes for each cell type in S. mediterranea, and tracking the transcriptomic changes during the regeneration process [197,198,199,200]. The genomic and transcriptomic data are deposited in specialized databases and freely available to the research community [197,201,202,203,204]. The potential of using S. mediterranea as a model organism to study evolutionarily conserved motile cilia was first described in 2009, and its value has been confirmed in later publications [188,192,205,206]. Motile cilia are present in many planarian cell types (Figure 2). Multiple motile cilia (9 × 2 + 2) covering the apical side of MCCs (~80 per cell) in the planarian body epidermis beat in a synchronized way and are responsible for worms’ locomotion [188,207]. MCCs are also present in the epithelium that covers the feeding organ (pharynx) [188,207,208]. Specialized ciliated cells at the proximal end of protonephridia (so-called flame cells) play role in fluid ultrafiltration and circulation. Cilia with the same ultrastructure as motile ones are also found in the sensory neurons in planarians, although their ability to move is not clear [188,207]. Finally, sperm cells with flagella are present in the sexual strains of planarians [209,210]. In planarians, basal bodies are assembled de novo during terminal differentiation of ciliated cells from neoblast progenies, and never have the function of a centrosome [1,58,207]. While it is currently unclear how the flatworms generate multiple centrioles in cells that are initially centriole-free, RNAi experiments show that the known key factors of centriole duplication are crucial for their amplification [208]. Both ventral and pharyngeal epidermis are easily accessible and form cilia at high density and in known orientation [188]. The great advantage of using S. mediterranea as a model in motile cilia studies is that the effect of gene silencing on cilia function can be readily analyzed by recording the change in the speed of planarian locomotion. Importantly, defects that compromise the function and structure of the cilia are not detrimental to planarians, making them an ideal system for loss-of-function studies concerning cilia-related genes [192]. Under normal conditions, planarians move by the use of ciliated epithelium covering the ventral side of worms’ body (so-called gliding movement), while cilia-related gene silencing manifests in a so-called “inch-worming” movement that engages the muscles (waves of whole-body contraction and extension) [188,207]. This phenotype (inch-worming) is easily visible to even unaided human eyes; a stereoscope (with camera) makes it more precise and allows to record movies showing the movement of planarians, which can then be used to measure the distance traveled by worms (using ImageJ software) [192]. The motility impairment may be associated with edema, which results from the dysfunction of cilia in protonephridia [211]. The beating of cilia covering the lateral part of worms can be recorded using high-speed video camera microscopy (HSVM), and the cilia beating frequency, pattern, and synchrony can be analyzed in slow motion under a microscope. The number and length of cilia can be inspected using a fluorescence microscope after immunofluorescence staining (IF) with cilia-specific antibodies (acetylated alpha-tubulin, a marker of the axoneme). After RNAi, changes in the gene expression pattern of epidermal markers can be tracked using WMISH. In addition, the possibility to stimulate cell differentiation through worms’ fragmentation (cutting) that triggers the regeneration process allows for tracing changes in the gene expression during the differentiation of the neoblasts into ciliated cells. The effect of gene silencing on the ultrastructure of planarian cilia can be also examined using a transmission electron microscope (TEM). In addition, the flatworms bloat due to the inhibition of ciliary function in flame cells, which leads to defective osmoregulation and edema formation [207]. Planarians have been widely used as a model for studying signaling networks implicated in the maintenance of tissue homeostasis, regeneration, and polarity. A large number of studies were devoted to essential cellular pathways, including Wnt and Hedgehog signaling in establishing polarity [212], Akt signaling in tissue maintenance and regeneration [213], EGFR signaling in the regulation of excretory system [211]. Like in most bilaterally symmetric animals, canonical Wnt signal is transduced through frizzled receptor and with the help of disheveled stabilizes beta-catenin, which activates expression cascade controlling anterior/posterior axis during regeneration [212,214]. Wnt signals transduced through frizzled receptors to various non-canonical pathways (disheveled-dependent or Ca2+-dependent) control cell movement and planar cell polarity (apical positioning of the basal bodies of epithelial cells). Hedgehog signaling modulates Wnt/beta-catenin’s role in establishing the anterior/posterior axis; when Wnt signaling is low, heads develop, and when it is high, tails are formed [145,215]. The majority of these signaling networks have the ciliary context, linking various aspects of Hedgehog signaling, regeneration, and the biogenesis of cilia, e.g., [145,213,216,217]. RNAi-mediated silencing of a variety of genes in S. mediterranea has been used to explain/confirm the connection between the homologous genes, defects of the ciliary ultrastructure, and cilia dysfunction in other organisms. The explicit use of S. mediterranea as an animal model to elucidate the pathogenesis of motile ciliopathies (and in particular, the role of candidate PCD genes) includes relatively few studies, where the effect of gene silencing on motile cilia function has been examined using dsRNA-mediated knockdown of the planarian homologs of human candidate genes not previously linked to PCD. In many more S. mediterranea studies, the demonstrated phenotypic effects of motile cilia-related genes’ silencing strongly resemble those seen when PCD genes are mutated, but their involvement in the pathogenesis remains to be confirmed by finding deleterious variants in PCD patients. Deleterious variants of CFAP298 (C21orf59) [81], CCDC151 [109], and CFAP300 (C11orf70) [63] have been found in PCD patients. Knockdown of the planarian homologs of these genes has revealed the impaired locomotion phenotype in worms. The details of this phenotype have been explained using further assays. HSVM analysis of planarians with silenced CFAP300 (C11orf70) demonstrated changes in cilia motility pattern and lowered beat frequency, while TEM analysis of cilia in planarians with CFAP298 (C21orf59) and CCDC151 knockdown revealed ODA assembly defects of dynein arms and loss of ODA, respectively. The effects of these three genes’ silencing are consistent with the observations in other animal models. Ch. reinhardtii and zebrafish mutants lacking CCDC151 orthologues featured a loss of ODAs [111,218]; silencing of CCDC151 in zebrafish and mice was shown to alter ODA assembly [109]. Knockdown of CCDC298 in zebrafish and Ch. reinhardtii, and of CFAP300 in P. tetraurelia and Ch. reinhardtii resulted in a complete lack of ODA and IDA [61,81]. All three genes are presently considered PCD genes, involved in the assembly of dynein arms (CFAP298, CFAP300), and proper functioning of the ODA docking complex (CCD151). The role of FOXJ1 as the key transcription factor controlling motile cilia biogenesis has been reported in various FOXJ1-deficient model organisms, including mice [49], X. laevis, and zebrafish [50]. The S. mediterranea model has been used to demonstrate the conserved role of vertebrate FOXJ1. Among four FOXJ1 homologs found in planarians, silencing of FOXJ1-4 caused the absence of motile cilia, resulting in a characteristic inch-worming locomotion and edema formation [48]. The FOXJ1 involvement in PCD pathogenesis in humans has been demonstrated several years later, when dominant pathogenic variants in FOXJ1 were found in PCD patients with mild respiratory symptoms and hydrocephalus, caused by the severely reduced number of cilia per MCC due to defect in the apical docking of basal bodies [31]. Proteins essential to basal body assembly in S. mediterranea include orthologs of many conserved genes required for centriole assembly or function in humans. In planarians, depleting the ortholog of OFD1 (among other proteins) results in the decreased locomotion of knocked-down animals, apparently due to the inhibition of basal body docking [58]. A similar ciliary phenotype has been recently demonstrated in PCD patients with the disease caused by nonsense mutations in the few last exons of the OFD1 gene [55]. In humans, the truncation of the C-terminus of the protein causes PCD without severe neurological, skeletal, or renal symptoms characteristic for other OFD1-related syndromes associated with the loss of a larger part of the OFD1 protein cause syndromic diseases (e.g., oral-facial-digital syndrome type 1 or Joubert syndrome type 10). While the effect of the gene knockdown in planarians does not explain truncation size-dependent differences in human clinical phenotype, it corroborates the proposed mechanism for the ciliary phenotype in PCD patients, showing that apical docking of basal bodies in planarians and in humans employ, at least in part, the same molecular components. DAW1 (WDR69/ODA16) encodes a WD repeat protein, whose role as a dynein assembly factor has been shown in many model organisms. Depletion of DAW1 protein homologs results in ultrastructural defect characterized by the reduced number of ODAs in Ch. reinhardtii [219], zebrafish [220], and mouse [221]; Ch. reinhardtii studies have shown that DAW1 is involved in ODA transport through interaction with IFT46 protein [86,222,223]. The knockdown of DAW1 homolog in S. mediterranea results in shortened epidermal cilia and decreased abundance of ciliated protonephridia [85]. The recent finding of deleterious DAW1 variants in patients with disturbed laterality and respiratory symptoms has confirmed the predicted involvement of this gene in PCD pathogenesis, although only in patients whose cilia are characterized by subtle beating abnormalities [84]. Deleterious variants in two other genes, CFAP45/CCDC19/NESG1, and CFAP52/WDR16, have been found in human individuals whose clinical presentation, with situs inversus and asthenozoospermia, but only mild respiratory symptoms, did not allow for classifying them as classical PCD cases [151]. Earlier studies in Ch. reinhardtii have localized these two proteins in the lumen of the B microtubule of the peripheral doublet [24]. The knockdown of the planarian homologs resulted in significant impairment of planarian locomotion in viscous but not in a normal medium; TEM of the silenced worms has shown normal ciliary ultrastructure, consistent with TEM cross-sections of CFAP45- and CFAP52-deficient respiratory cilia from CRISPR-Cas9 generated mice or from humans with mutated genes [151]. Therefore, planarian results confirm the uncertain status of CFAP45 and CFAP52 as PCD genes. IC2 and LC1 are S. mediterranea homologs of human DNAI2 and DNAL1 genes, respectively, encoding integral components of ODA. Mutations in DNAI2(IC2) cause defects in ODA resulting in the reduction in ciliary beat frequency in Ch. reinhardtii, and are known to cause PCD in humans [90]. Mutations in DNAL1(LC1) disturb the proper function of ODA in Ch. reinhardtii [224], but the data supporting this gene’s role in human PCD are scarce [97,225]. The knockdown of either of these two genes in S. mediterranea severely decreases worms’ motility, due to the reduction in the ciliary beat frequency and coordination (metachronal synchrony). However, while TEM and IF reveal the loss of ODA in IC2-silenced planarians, no ODA defects are visible in LC1-silenced worms [91]. This is consistent with the still uncertain role of DNAL1 in PCD pathogenesis in humans. The ODA-docking complex is a microtubule-associated structure that targets ODA to its binding site on the axonemal microtubule [226]. In Ch. reinhardtii it contains three proteins, referred to as DC1, DC2, and DC3, of which DC1 and DC2 can assemble ODA in the absence of DC3 [227]. Ch. reinhardtii mutants with the loss of DC2 (a major subunit of the ODA-docking complex) have two flagella of normal length but show slow jerky swimming [228]. Two DC2 homologs, CCDC63 and CCDC114, function in ODA docking in vertebrates. Respiratory cilia in PCD patients with deleterious variants in CCDC114 have normal length, but lack ODAs due to the defects in ODA docking to microtubules [229]. In mice, in which CCDC63 (the testis-specific DC2 homolog) is knocked out, spermatozoa flagella are shortened, but ODAs remain unaffected, probably due to the compensation by overexpression of CCDC114 [230]. The knockdown of DC2 orthologue in S. mediterranea impairs worms’ locomotion due to the low-frequency, uncoordinated ciliary beating caused by the inefficient ODA docking; in addition, cilia density and length are decreased [231]. The importance of these findings for PCD pathogenesis remains to be explored. WDR92 is a highly conserved WD-repeat protein. Silencing of the planarian homolog of WDR92 results in a phenotype similar to those observed when acknowledged PCD genes are knocked down. Peristaltic contractions instead of smooth gliding of the worms reduced and uncoordinated the ciliary beat; in TEM analysis, partial loss of dynein arms, incomplete closure of the B-microtubule, and lack of normal central pair complex are observed [232]. WDR92 is required for the assembly of ODAs and IDAs in D. melanogaster and Ch. reinhardtii [233,234,235]. Based on these observations, WDR92 has been proposed to act as a part of a cytoplasmic chaperone required for the proper folding and stability of key axonemal components. So far, no pathogenic variants have been found in human PCD patients. IFT88 (Tg737) encodes a component of the IFT complex; its mutations in Ch. reinhardtii results in a lack of flagella, while in mice they cause shortening of primary cilia, as well as kidney and liver defects [236]. The importance of IFT88 in motile cilia biogenesis has been confirmed in the S. mediterranea model, where silencing of IFT88 significantly reduced planarians motility and caused the complete absence of cilia on the ventral surface of knocked-down animals [91]. Defects in IFT are likely to affect motile cilia in humans. Defects in the Tg737 gene in mice are very similar to those seen in humans with autosomal recessive polycystic kidney disease [237], but so far no pathogenic IFT88 variants have been reported in PCD patients. Ch. reinhardtii protein FAP163 is an intermediate dynein chain closely related to the FAP133 intermediate dynein chain that powers retrograde IFT required for the assembly of cilia. The functional role of FAP163 has been examined by the knockdown of the orthologous gene WD60 (FAP163) in S. mediterranea [238]. The silenced animals exhibited severely impaired movement (reduced velocity and inch-worming), resulting from a dramatic reduction in both the number and length of cilia. Cilia and ciliary stubs examined by TEM contained doublet microtubules and associated structures but had an enlarged diameter due to the presence of large quantities of amorphous electron-dense material located between the axonemal doublet microtubules and the ciliary membrane. These observations suggest that WD60(FAP163) is required for ciliary assembly. So far, no pathogenic variants have been found in human PCD patients. An interesting application of RNAi-mediated gene silencing in S. mediterranea concerns the analysis of planarian protonephridia as a model of pathological features of human cystic kidney diseases (CKDs), in which fluid-filled cysts develop from nephric tubules due to defective flow sensing, cell proliferation, and differentiation. In contrast to mammalian kidneys that contain only immotile sensory cilia, the excretory system of planarians is equipped with motile cilia that drive fluid flow into and through the tubules [239,240]. Interestingly, structure and function comparisons revealed that the combination of ultrafiltration and flow-associated filtrate modification is remarkably conserved between the planarian excretory system (flame cells) and the vertebrate nephrons (podocytes) [67]. The genome of S. mediterranea contains many genes that cause cysts when their equivalents are mutated in humans. Silencing of planarian homologs of human DNAH1 and LRRC50 genes resulted in abnormal worms locomotion due to the loss of cilia beating; animals also developed edema and formed protonephridial cysts [67]. These results suggest that cilia-driven fluid flow is crucial for maintaining cell homeostasis in planarian protonephridia and establish planarians as a novel and experimentally accessible invertebrate model for the study of human kidney pathologies. In the majority of the aforementioned studies, S. mediterranea was not the only organism used in the functional assessment of cilia-related genes, which are or can be considered candidate PCD genes. While the application of the planarian model in these studies may seem redundant, it can be seen that abundant work using S. mediterranea genes silencing has been performed to analyze the role of various proteins in the assembly, maintenance, and function of motile cilia, without immediate referring to PCD pathogenesis. These results are often used later, whenever deleterious variants in candidate PCD genes are found in human patients (see, e.g., the cases of FOXJ1 or DAW1). On the other hand, when a new candidate PCD gene comes into focus based on genetic screening in humans, using the planarian model is perhaps the most efficient way to perform preliminary functional studies. When compared to the most popular single-cell organisms (Ch. reinhardtii, P. tetraurelia, T. brucei, T. thermophila), S. mediterranea offers an advantage of studying the epidermis that closely resembles human epithelium with MCCs, and compared to fish or mammals allows a much faster, easier and more affordable alternative modeling. For many years, the use of the planarian model in the analysis of human candidate genes has been hampered by the lack of information on human-planarian orthologues. Recently, the rapid and extensive growth of the genomic, transcriptomic, phenotypic, and phylogenetic data generated by the planarian research community has alleviated this problem. In this review, the human gene names or aliases are used to facilitate comparison with the part of the text describing human cilia and ciliopathies. However, it should be emphasized that to establish a uniform method for naming genes and proteins and for describing RNAi experiments in S. mediterranea, nomenclature guidelines have been developed [241], where the main rule is that the name of the gene is preceded by a ‘Smed’ prefix. In addition, to enhance cross-platform and cross-species searchability, the Planarian Anatomy Ontology (PLANA), an extendable relational framework of defined S. mediterranea anatomical terms has been recently developed [242]. The rapid growth of the amount of genomic and functional data on S. mediterranea prompted the development of integration tools that would enable the collection and use of these data by the scientific community. In response to these needs, the SmedGD database has been developed by the Sánchez Alvarado team [201,202]. The data deposited in SmedGD refer to the S. mediterranea genome and include predicted and annotated genes, protein homologies, gene expression patterns, and RNAi phenotypes, among others. SmedGD has been a stand-alone web resource for 15 years. Recently, the S. mediterranea genome assembly from SmedGD has been transferred to SIMRbase at the Stowers Institute for Medical Research (https://simrbase.stowers.org/ (accessed on 20 February 2023)). Another high-quality S. mediterranea genome assembly can be found in the Planmine database (https://planmine.mpinat.mpg.de/planmine/begin.do (accessed on 20 February 2023)), developed in 2016 by the Rink team [199,203,243]. Originally, Planmine was based on the independently assembled transcriptomes from the Rink team and contributors from the planarian community. The updated database provides also genomic information, including a gene prediction set that assigns existing transcripts to defined genomic coordinates. In addition, Planmine uses recent datasets from the single-cell RNA-seq (e.g., from Digiworm resource and Planaria Single Cell Atlas), allowing for the expansion of the available gene expression information [197,198]. Both Digiworm and Planaria Single Cell Atlas refer to transcriptomes published by the Rink team, which makes these resources compatible with data in the Planmine database. Moreover, in contrast to other planarian databases, Planmine is also a resource of transcriptomes from other flatworm species. Planmine can be used to search for the planarian homologs of interesting transcripts and corresponding predicted genes. Planmine provides information about functional annotations (gene ontology), the best BLAST hits, expression patterns of homologs in planarian cell types, as well as phenotypes after RNAi of specified genes, among others. Planosphere (https://planosphere.stowers.org/ (accessed on 20 February 2023)) is a new website dedicated to S. mediterranea, which contains a collection of data and tools from the Sánchez Alvarado laboratory [242]. One of the Planosphere tools is “gene search”, which can be used to search homologs in S. mediterranea, and to define experimentally determined cell/tissue-specific gene expression patterns. In addition to transcriptomic and genomic data, Planosphere provides information about predicted protein sequences. This website is linked to the data deposited in the Planmine database and refers to data (e.g., from RNAseq) published by other teams. For researchers studying cilia or cilia-related diseases, it is important that RNAseq, together with other techniques, has allowed to characterize and reconstruct epidermal cell lineages, including the stages between neoblasts and fully differentiated epidermal cells [197,198,244,245,246]. Using Digiworm, researchers can check at which stage of epidermal lineage the gene of interest is expressed (it is possible for all cell/tissue types, e.g., protonephridia). The available web resources, especially Planmine, Planosphere, and Digiworm, provide researchers with a powerful tool to design experiments using S. mediterranea as a model organism. They can be used to simply explore their contents to better understand planarian biology. Importantly, they allow the cross-searching of databases devoted to other organisms, using gene names, sequences, annotation terms, etc. (the details of such searches differ among planarian websites). This facilitates using the growing planarian knowledge in applications related to studies of human ciliopathies.

PMC10002880

Maria Sofia Ciliento,Veronica Venturelli,Natale Schettini,Riccardo Bertola,Carlo Garaffoni,Giovanni Lanza,Roberta Gafà,Alessandro Borghi,Monica Corazza,Alen Zabotti,Sonia Missiroli,Caterina Boncompagni,Simone Patergnani,Mariasole Perrone,Carlotta Giorgi,Paolo Pinton,Marcello Govoni,Carlo Alberto Scirè,Alessandra Bortoluzzi,Ettore Silvagni

Evaluation of the Synovial Effects of Biological and Targeted Synthetic DMARDs in Patients with Psoriatic Arthritis: A Systematic Literature Review and Meta-Analysis

05-03-2023

psoriatic arthritis,targeted therapies,synovial biopsy,fibroblast-like synoviocytes,biological DMARDs,targeted synthetic DMARDs

The aims of this systematic literature review (SLR) were to identify the effects of approved biological and targeted synthetic disease modifying antirheumatic drugs (b/tsDMARDs) on synovial membrane of psoriatic arthritis (PsA) patients, and to determine the existence of histological/molecular biomarkers of response to therapy. A search was conducted on MEDLINE, Embase, Scopus, and Cochrane Library (PROSPERO:CRD42022304986) to retrieve data on longitudinal change of biomarkers in paired synovial biopsies and in vitro studies. A meta-analysis was conducted by adopting the standardized mean difference (SMD) as a measure of the effect. Twenty-two studies were included (19 longitudinal, 3 in vitro). In longitudinal studies, TNF inhibitors were the most used drugs, while, for in vitro studies, JAK inhibitors or adalimumab/secukinumab were assessed. The main technique used was immunohistochemistry (longitudinal studies). The meta-analysis showed a significant reduction in both CD3+ lymphocytes (SMD −0.85 [95% CI −1.23; −0.47]) and CD68+ macrophages (sublining, sl) (SMD −0.74 [−1.16; −0.32]) in synovial biopsies from patients treated for 4–12 weeks with bDMARDs. Reduction in CD3+ mostly correlated with clinical response. Despite heterogeneity among the biomarkers evaluated, the reduction in CD3+/CD68+sl cells during the first 3 months of treatment with TNF inhibitors represents the most consistent variation reported in the literature.

Evaluation of the Synovial Effects of Biological and Targeted Synthetic DMARDs in Patients with Psoriatic Arthritis: A Systematic Literature Review and Meta-Analysis The aims of this systematic literature review (SLR) were to identify the effects of approved biological and targeted synthetic disease modifying antirheumatic drugs (b/tsDMARDs) on synovial membrane of psoriatic arthritis (PsA) patients, and to determine the existence of histological/molecular biomarkers of response to therapy. A search was conducted on MEDLINE, Embase, Scopus, and Cochrane Library (PROSPERO:CRD42022304986) to retrieve data on longitudinal change of biomarkers in paired synovial biopsies and in vitro studies. A meta-analysis was conducted by adopting the standardized mean difference (SMD) as a measure of the effect. Twenty-two studies were included (19 longitudinal, 3 in vitro). In longitudinal studies, TNF inhibitors were the most used drugs, while, for in vitro studies, JAK inhibitors or adalimumab/secukinumab were assessed. The main technique used was immunohistochemistry (longitudinal studies). The meta-analysis showed a significant reduction in both CD3+ lymphocytes (SMD −0.85 [95% CI −1.23; −0.47]) and CD68+ macrophages (sublining, sl) (SMD −0.74 [−1.16; −0.32]) in synovial biopsies from patients treated for 4–12 weeks with bDMARDs. Reduction in CD3+ mostly correlated with clinical response. Despite heterogeneity among the biomarkers evaluated, the reduction in CD3+/CD68+sl cells during the first 3 months of treatment with TNF inhibitors represents the most consistent variation reported in the literature. Psoriatic disease is a chronic inflammatory disease characterized by complex clinical heterogeneity, affecting different sites, predominantly skin and nails, as well as peripheral joints, axial skeleton and entheses [1]. A severe dysregulation of several pro-inflammatory immune pathways plays a major role in the pathogenesis of the disease, along with genetic background and environmental factors [2,3]. Immune cells infiltrating target organs cause significant production of pro-inflammatory cytokines, including tumor necrosis factor-α (TNF-α) and interleukin (IL)-1β, IL-6, IL-22, IL-23, IL-17A, and IL-18, which cause perpetuation of inflammation [4]. Chronic synovitis is a key feature of psoriatic arthritis (PsA), and it is characterized by hyperplasia of fibroblast-like synoviocytes (FLS) in the intimal lining layer, inflammatory infiltrates of the synovial sublining, and neo-angiogenesis phenomena. Analysis of PsA synovial membrane has shown the presence of macrophages in the lining layer, while in sublining different types of immune cells can be found (macrophages, mast cells, polymorphonuclear cells, T cells, B cells and plasma cells), producing a wide range of cytokines [2,5]. The natural history of the process described above generates potentially progressive cartilage and bone damage. In recent years, growing knowledge of pathogenesis has allowed advances in the understanding of molecular pathways underlying psoriatic disease, and new therapies have been licensed [6,7]. Despite the suggestion that the treatment decision in PsA should be adapted to individual patient features, the management of psoriatic disease is far from precision medicine [8,9]. Indeed, there are no validated biomarkers able to predict the response to specific therapies [10], and the choice of the drug depends mostly on disease severity, extra-articular manifestations and endotypes of the disease, prognostic factors, prior treatment history, comorbidities, access to therapy, and patient’s preferences, similarly to other cognate diseases such as rheumatoid arthritis (RA) [11,12,13]. Research is, therefore, moving towards the development of new ways of personalizing treatments, to which the histological analysis of synovial membrane samples may contribute [14,15]. The longitudinal histological analysis of the effects on synovium biopsy samples of disease modifying antirheumatic drugs (DMARDs) and the studies of in vitro exposure of cells or tissues to drugs could advance the most influential effects of DMARDs on psoriatic synovial cells and could be applied to the goal of precision medicine. Supported by technological advances in synovial tissue processing, an increasing number of studies are focusing on cellular and molecular changes of synovium after treatments, with potentially relevant clinical implications [16]. A work gathering and summarizing systematically the latest evidence on this topic has not been published. Therefore, the aims of this systematic literature review (SLR) were (i) to identify the synovial effects of approved biological (b)/targeted synthetic (ts) DMARDs in PsA through the analysis of synovial biopsies, and (ii) to determine if these effects could be considered histological/molecular biomarkers of response to therapy. This SLR was performed to retrieve biomarkers modifications on synovial membrane and skin biopsies after bDMARD or tsDMARD administration (only drugs approved by the EMA for the systemic management of psoriasis and PsA were evaluated; design of studies evaluated: paired biopsies in longitudinal studies, in vitro studies). Here, we provide methods and data regarding the synovial effect of b/tsDMARDs. The SLR was conducted following the PRISMA 2020 Checklist (Supplementary Table S2) [17], and the protocol was registered in PROSPERO (CRD42022304986). First, the research question was translated into patients, intervention, comparator, outcome, study type (PICOs). The population was defined as patients with psoriasis or active peripheral PsA undergoing synovial biopsy. Intervention was defined as synovial biopsy in studies with bDMARDs or tsDMARDs approved for the systemic management of these conditions; for the outcomes, we evaluated (i) biomarker modifications on synovial membrane biopsies/in vitro cell cultures, and (ii) clinical response to systemic treatment (Supplementary Material 1.1). We included all published studies limited to humans, published in English or Italian, by 16 January 2022. Studies not considering synovial membrane biopsy (e.g., in vitro studies in cell lines or cells purchased from companies) or studies involving drugs different from b/tsDMARDs approved for the management of psoriasis or PsA were not included (Supplementary Material 1.2). A comprehensive search was conducted to find eligible articles in different electronic databases: PubMed, Embase, Scopus, and Cochrane Library (Supplementary Material 1.3). Records were imported into a bibliographic management software (Zotero) and articles appearing in more than one database were considered only once. Articles were selected based on (i) title and abstract and (ii) full text, by two pairs of independent reviewers (MSC, NS, VV, RB), classifying those studies meeting the inclusion criteria. Any disagreement between the reviewers regarding the eligibility of a particular study was resolved through discussion between reviewers; in case of persistent disagreement, a third referee was consulted (ES). Finally, we scrutinized the reference lists of the identified articles to find additional relevant studies. Data from included articles were extracted in pre-specified forms using a secure electronic data capture database for PsA patients (https://www.redcap.ospfe.it accessed on 16 August 2022) [18], hosted at the University Hospital of Ferrara, including general information on the article (title of publication, study design, year of publication), features of the population (baseline demographics, sample size, country, type of intervention, outcome measures) and, when available, standardized mean difference (SMD). The main outcome of interest was the modification of biomarkers following systemic or in vitro drugs, while, for biomarkers of response [19], the clinical response was considered as outcome. Results were presented in summary of evidence tables. Risk of Bias (RoB) of included studies was assessed using the Newcastle–Ottawa scale (NOS) for cohort and case–control studies [20], and the Cochrane RoB (RoB2) for randomized-controlled trials (RCTs) [21]. Discrepancies between the two pairs of reviewers were discussed with a third reviewer (ES). Descriptive results of the SLR were reported as mean and standard deviation (SD) for quantitative variables. Categorical variables were described as counts and percentages. Metanalyses were conducted to test variations of CD3+ lymphocytes and sublining CD68+ macrophages following bDMARD treatment in longitudinal studies. Mean and standard deviations were extrapolated from full-text articles. Missing data were imputed from median and range [22]. Raw data were requested from authors for a relevant article, but they were unavailable [23]. Standardized mean difference (SMD) was used as measure of effect to combine total synovial cell count with IHC semiquantitative scores. Cohen’s d was employed to calculate estimated effect sizes and was interpreted using the following conventions: small effect if ≥0.20, moderate effect if ≥0.50, and large effect if ≥0.80. Pooled SMD was estimated using a random-effects model with inverse variance method, and results were graphically presented with forest plots. Cochran’s Q and I2, respectively, were used to test and quantify between-study heterogeneity [24]. Publication bias was investigated with qualitative inspection of funnel plots and Begg’s test for assessing asymmetry [25]. Influence analyses were repeated after excluding each study once, while sensitivity analyses were performed by removing papers in which part of the sample also included patients with ankylosing spondylitis or other forms of SpA. Finally, a sensitivity analysis was conducted using unstandardized mean difference (MD) from articles in which a semiquantitative IHC score was adopted. Analyses were performed with Stata14 software (STATA Corporation, College Station, TX, USA) and RStudio© using the package ‘meta’ [26]. Out of 3111 non-duplicate articles evaluated, 22 were included (Figure 1), of which 19 were longitudinal and three in vitro, for a total of 365 patients with a mean age (standard deviation, SD) of 45.8 (4.3) years (Table 1). Five studies were RCTs, while the others were observational studies (cohort or case-series). Arthroscopy was the biopsy sampling technique used in almost all works. In longitudinal studies, infliximab, etanercept and adalimumab were the most frequently used drugs, while in in vitro studies the effects of JAK inhibitors (tofacitinib, upadacitinib) on synovial explants and FLS were evaluated, as well as adalimumab and secukinumab in FLS and CD4+ T cell co-cultures. Several laboratory outcomes were assessed. The main technique used in longitudinal studies was immunohistochemistry (IHC), often coupled with histology analysis. The average time for outcome assessment was 11.0 (6.4) weeks (longitudinal studies). Table 2 and Table 3 summarize the main findings of the SLR (longitudinal and in vitro studies), while Table 4 and Table 5 highlight RoB assessment for observational studies and RCTs, respectively. The effect of TNF-alfa inhibitors (TNFis) on synovial tissue using IHC was explored in 13 longitudinal studies, showing a modification in various types of inflammatory cells and molecules. The time of paired biopsy assessment varied between 48 h and 12 weeks. The most important finding was a decrease in CD3+ lymphocytes, observed in 10 studies, despite not always reaching statistical significance [23,29,30,31,32,33,34,35,36,39]. T lymphocytes (CD3+) represent one of the most important cellular populations in PsA synovitis, whose role is highlighted by the enrichment in T-cell-derived cytokines in the synovial fluid and inflamed synovium. Among T lymphocytes, CD4+ and CD8+ markers were globally reduced after treatment, as well as endothelial CD31+ cells and CD68+ macrophages, despite not being consistently reported across studies included in this SLR [23,27,29,31,32,33,34,35,36]. TNFis also interfered with the action of dendritic cells, responsible for antigen presentation and cytokine secretion, causing a reduction in C-type lectin domain family 9 member A (CLEC9A) [39], which is involved in the cross-presentation of antigens to CD8+ T cells [48]. Apart from cellular populations, IHC was adopted to unravel the effect of TNFis on cytokine expression, showing no effects of adalimumab on IL-17A, IL-17F, IL-17RA and IL-17RC [28]. Contrariwise, TNFis significantly reduced matrix metalloproteinases (MMPs) in synovial tissue, as demonstrated with adalimumab on MMP13 levels [23], or with etanercept on MMP3 and MMP9 [34], whereas MMP3 expression significantly differed between responders and non-responders to infliximab and etanercept [35]. Regarding adhesion molecules, a significant decrease in the levels of intercellular adhesion molecule 1 (ICAM-1) on synovial capillaries was observed in patients after combination therapy of infliximab and methotrexate [32], as well as a reduced expression of E-selectin [33] and vascular cell adhesion protein 1 (VCAM-1) [33]. Finally, TNFis demonstrated anti-angiogenic properties through significant reduction of αvβ3-positive neo-vessels [27,32] coupled with a decrease in other markers of neo-angiogenesis such as SDF-1+ vessels, vascular-endothelial growth factor (VEGF) and its receptor KDR/flk-1 (VEGFR-2). On the contrary, angiopoietin-2 showed a significant increase. The RANK/RANKL/osteoprotegerin (OPG) system, known to play a central role in bone resorption by promoting the maturation and activation of osteoclasts, was not significantly affected after 12 weeks of treatment with TNF-α blockade [30]. In fact, no significant differences were detected in synovial RANKL, OPG, and RANK expression, although a general decrease in the degree of cellular infiltration was observed. Despite the absence of effect on the global study population, TNFα blockade decreased the RANKL expression by FLS in a subset of patients with the best clinical response. Other key metabolic or pro-inflammatory processes involved in psoriatic synovitis were assessed by the retrieved articles exploiting laboratory techniques other than IHC. Cellular apoptosis was investigated using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) and caspase-3 staining, showing no increase in apoptosis following short-term TNFis treatment in longitudinal studies (48 h) [31]. Nuclear factor kappa B (NFκB) proteins, a group of transcription factors involved in inflammatory and immune responses able to interact with other transcription factors, including mitogen activated protein kinases (MAPKs) [49], were evaluated by Lories et al. [37], who analyzed protein expression levels of NFκB and the three main MAPKs, namely extracellular regulating kinase (ERK), the c-Jun-N-terminal kinase (JNK), and p38, before and after etanercept (6 months). By immunofluorescence staining and digital image analysis, etanercept acted by reducing NFκB, ERK, and JNK, but not p38 levels. A Spanish study conducted by the group of Cañete et al. [27] focused on the process of neo-vascularization in the synovium of psoriatic arthritis patients treated with infliximab after methotrexate (MTX) treatment failure. After 8 weeks, the vascular score showed a significant reduction, paired with IHC modifications suggesting a reduction in neo-vessels, as demonstrated by αvβ3+ markers. Similarly, VEGF and VEGFR-2 mRNA expression decreased, and the expression of Ang-2 increased after treatment. The gross histological evaluation permitted underlining a significant effect of infliximab or etanercept in reducing synovial lining layer thickness [33,34] and the number of blood vessels, along with a downregulation of follicular structures organization [33,38]. Another study, by Collins et al. [40], based on proteomics analysis, aimed at identifying protein expression differences in patients treated with TNFis (either etanercept or adalimumab) that could serve as potential biomarkers of treatment response. In their study, 119 different protein spots changed significantly following 4 weeks with etanercept (including haptoglobin, annexin A2, serum amyloid P, peroxiredoxin 6, serum albumin, Ig kappa chain C, fibrinogen beta chain), and 91 protein spots changed significantly following adalimumab treatment (including haptoglobin, serum albumin, ubiquitin conjugating enzyme E2, annexin A1, A2 and A6, serum amyloid P, heat shock cognate 71 kDa protein, fibrinogen beta chain, pyruvate kinase isozymesM1/M2, collagen alpha 3 and cathepsin B). Moreover, researchers identified 25 proteins that were differentially expressed between “good responders” and “poor responders” to adalimumab (annexin A1 and A2, serum albumin, haptoglobin, apolipoprotein A1, collagen alpha 3, actin, rho-GDP-dissociation inhibitor 2, alpha-1B-glycoprotein, 78 kD glucose-related protein, replication protein A, pyruvate kinase M1/M2, heat shock protein 70 kDa and 71 kDa, vimentin and lamin-B2). Four studies examined the adoption of non-TNF inhibitors in longitudinal studies. In the study by Szentpetery et al. [41], abatacept reduced FOXP3+ Treg expression (IHC) in the synovium over 6 months, an effect not confirmed in skin biopsies. No significant variations were reported for CD3+, CD8 or CD31 expression during the study period, despite a trend towards reduction in CD4+ cells. The study by Fiechter et al. [42] analyzed, in 24 patients with PsA and active knee or ankle arthritis, the IHC changes on synovial biopsy after the use of ustekinumab. After 12 weeks of systemic ustekinumab treatment, there was a numerical decrease in all infiltrating immune cells (CD3+, CD15, CD20) and a significant decrease in sublining CD68+ macrophages, as demonstrated by IHC analysis, as well as a reduction in MMP3 mRNA levels. Despite decreasing PsA synovial inflammation, patients with both clinical and ultrasound remission displayed a persistent synovial cellular infiltrate, suggesting residual histological inflammation under ustekinumab. Synovial TNF expression was unaffected by ustekinumab, as were levels of IL-6, IL-8 and IL-17. Instead, ustekinumab interfered with several chemotaxis and neo-angiogenesis pathways, and it remodulated wnt signaling (pro-chondrogenesis effect) and the PI3K-Akt-mTOR and MAPK-ERK pathways. Van Mens et al. [43], instead, evaluated the impact of secukinumab on the synovial immunopathology of 20 patients with peripheral spondyloarthritis (SpA), of which 13 had PsA. Along with the clinical benefit, at week 12, there was a significant reduction in CD15+ neutrophils and CD68 + sl macrophages as measured by IHC. When qPCR analysis was performed, the authors highlighted a significant reduction in IL-6, MMP3, and CCL20 mRNA expression, but not in IL-8. There was also a significant reduction in IL-17A mRNA, while the expression of TNF was unaffected. Chen et al. [44] also evaluated, through IHC analysis, the content of IL-17A inside the mast cells located in the synovial membrane during secukinumab treatment. While the percentage of all IL-17A-positive cells (non-mast cells) decreased, the IL17A content in mast cells increased. Five papers evaluated changes in CD3+ lymphocytes from baseline to follow-up in synovial tissue of patients treated with bDMARDs and included in quantitative analysis [29,31,33,34,41]. Fifty-three patients were included, with a mean (SD) age of 46.5 (5.4) years and a mean (SD) disease duration of 92.4 (40.2) months. Patients were exposed to TNFis in four of the five studies, while in one study they were treated with abatacept [41]. Random-effect meta-analysis revealed a significant reduction in CD3+ count after 4 to 12 weeks of bDMARD treatment (pooled standardized mean difference [SMD]= −0.85, 95% CI [−1.23; −0.47], p < 0.0001) (Figure 2). No significant between-study heterogeneity was found (I2 = 0%, Q = 1.86, p = 0.7614). In the sensitivity analysis, which excluded papers considering also the broader SpA population, the pooled effect size had similar effect direction and magnitude (pooled SMD −0.78, 95% CI [−1.32; −0.23] p = 0.0054) [33,34]. Influence analyses were performed after excluding each study once, as well as sensitivity analyses using unstandardized mean difference (MD) from articles in which a semiquantitative IHC score was adopted, with no significant variations from primary analyses. Additional details of sensitivity analyses and publication bias assessment, investigated with qualitative inspection of funnel plots and Begg’s test for assessing asymmetry, are shown in Supplementary Material 2.1 and 2.2, Supplementary Table S1 and Figure S1. A random-effect meta-analysis was conducted in five studies in which CD68+ macrophages in the sublining were quantified before and after 4 to 12 weeks of TNFis treatment (no other bDMARD was tested) [27,29,31,33,34]. Mean (SD) age was 47.5 (5.6) years, and mean disease duration was 90.4 (44.4) months. The results were consistent with a significant reduction in CD68+ levels (pooled SMD −0.74, 95% CI [−1.16; −0.32], p = 0.0005) without substantial heterogeneity (I2 = 22%, Q = 5.13, p = 0.2744) (Figure 3). The sensitivity analysis, excluding papers permitting the enrollment of other SpA populations, did not reconfirm a significant decline in CD68 + sl macrophage counts (SMD −0.70, 95% CI [−1.50; 0.09], p = 0.0841), with moderate level of heterogeneity (I2 = 55.1%, Q = 4.46, p = 0.1077) [33,34]. Additional details of sensitivity analyses and publication bias assessment are shown in Supplementary Materials 2.1 and 2.2, Supplementary Table S1 and Figure S2. Only three studies have assessed the effect of b/tsDMARDs through an in vitro design. The consequences of PsA FLS exposure to Janus kinase inhibitors (JAKis) have been analyzed in two cases (tofacitinib, upadacitinib) [45,46]. In the study by Gao et al. [45], the cells were not exogenously stimulated, and synovial explants were analyzed along with FLS, as opposed to the study of O’Brien et al. [46], which focused on FLS stimulated with oncostatin M (OSM), a known inducer of the JAK/signal transducer and activator of transcription (STAT) pathway. Both studies reported similar effects of JAKis in reducing the migration capacity of FLS. Moreover, using enzyme-linked immunosorbent assay (ELISA), the secretion of IL-6 and monocyte chemoattractant protein-1 (MCP-1) were reduced by currently approved JAKis in synovial explants and FLS cultures, respectively. In the study conducted by O’Brien et al. [46], several JAKis (upadacitinib, baricitinib, peficitinib, filgotinib; only the former is actually approved for PsA management) were evaluated. MCP-1 and IL-6 gene expressions were inhibited by baricitinib and upadacitinib. Furthermore, upadacitinib reduced ICAM-1 gene expression. With a cellular bioenergetic function analysis, a change in FLS overall metabolic profile was also observed, consisting of a rise in mitochondrial respiration processes with a concomitant decline in glycolysis. Apart from these, the study of Gao et al. [45] described a reduction in MMPs secretion in synovial explants and a decrease in IL-8 secretion after exposure to tofacitinib, a finding not confirmed with upadacitinib in FLS cultures. In both PsA explants and PsA FLS, tofacitinib was able to reduce NFkBp65 expression, while it reduced invasion and network formations of FLS. Another in vitro study, by Xu et al. [47], analyzed the effects of adalimumab and secukinumab on co-cultures of CD4+ T cells and FLS after stimulation with anti-CD3 and anti-CD28 for 72 h. The authors used ELISA and RT-PCR for laboratory readouts, and they found that secukinumab significantly reduced the production of IL-17A and IL-6, whereas adalimumab reduced TNF-alfa, MMP3 and MMP13. Both drugs reduced IL-8 and IL-1b secretion and their mRNA expressions. Of the included studies, 10 evaluated the clinical response to therapy, none with in vitro design. In three of these studies, all of which considered TNFis, the longitudinal reduction in CD3+ cells predicted the clinical response [23,29,35], while in one study with secukinumab, no correlation was found between changes in clinical scores and changes in cellular infiltrate, including CD3+ [43]. Other studies evaluated different parameters, such as the increase in IL-17A-positive mast cells [44], reduced ectopic lymphoid neogenesis [38], decreased RANKL [30], MMP13 [23] or MMP3 [35] expression, decreased infiltration of macrophages or polymorphonuclear cells [35], or the ratio of differentially expressed genes [42] or proteins [40] in responder patients compared to non-responders, suggesting that these biomarkers could be promising in antedating the effectiveness of systemic DMARDs. Regarding synovial histology, neither lining layer hyperplasia nor vascularity modifications correlated with disease activity over time [35]. Furthermore, no correlation was found between pre- and post-treatment measurements of NFκB and MAPK activation in synovial tissue and disease activity parameters [37]. This SLR aimed to deepen understanding of the synovial mechanisms of action of b/tsDMARDs approved for the systemic management of PsA, trying to evaluate their synovial effects in a standardized manner, exploring how this effect correlates with clinical response. Since no validated biomarker has yet entered clinical practice [10], the full knowledge of the synovial impact of such drugs is of value. This is particularly true since it is believed that identifying the appropriate treatment in the early stages of the disease is effective in achieving clinical remission, thus avoiding disease progression [8,50]. Our SLR highlighted a significant heterogeneity in laboratory techniques adopted to investigate the effects of DMARDs, with IHC emerging as the most used in longitudinal studies. Longitudinal reduction in CD3+ lymphocytes and CD68+ macrophages of the sublining represents the most common synovial modification following TNFis, with the first parameter emerging as a promising candidate for systemic treatment response prediction. However, the synovial effect of DMARDs in PsA is more wide-ranging than a mere anti-inflammatory effect (Figure 4). Our SLR aimed to identify two different types of studies, namely longitudinal and in vitro studies. In longitudinal studies (19 studies included), infliximab, etanercept and adalimumab were the most used drugs, while, for in vitro studies (three articles retrieved), tofacitinib and upadacitinib were investigated in FLS cultures, as well as in synovial explant cultures, and adalimumab and secukinumab in co-cultures of FLS and CD4+ T cells. Given this discrepancy, most of the evidence arises for TNFis and from longitudinal studies. Patients enrolled in the included studies were either naïve or previously exposed to bDMARDs, had a moderately long disease duration, and had mainly undergone arthroscopic synovial biopsy, with no US-guided synovial biopsy procedure performed. Since US-guided synovial biopsy is emerging as the technique of choice for synovial membrane analysis in chronic inflammatory arthritis [51,52,53], given its wide availability, the adoption of this technique is expected to increase in the following years, in line with similar experiences in longitudinal studies in RA [54,55]. Focusing on longitudinal studies, our meta-analysis suggested a significant downmodulating effect of several bDMARDs, mostly TNFis, in reducing CD3+ lymphocytes after 4–12 weeks of systemic treatment. This net effect remained significant considering only works assessing the cellular infiltrate in a semi-quantitative manner, and, similarly, after excluding works enrolling patients suffering from other forms of spondyloarthritis. When the same reasoning was applied to CD68 + sl variation, statistical significance was lost. In our opinion, this distinction applies specifically to PsA patients, and it might contribute to the distinction of PsA from RA, for which it is well-known that the reduction in CD68 + sl is one of the most well-characterized readouts of an effective treatment, resistant to placebo effects [56,57,58,59]. Moreover, longitudinal CD3+ reduction was the most frequently assessed modification related to systemic treatment response, suggesting this modification can be explored as a response biomarker. Regarding the design of the studies, it must be underlined that longitudinal studies carry some relevant limitations; the most relevant relates to the necessity for a patient to undergo separate biopsy procedures. This could be challenging for patients experiencing significant symptom relief following therapy, with a low rate of potential impact on clinical practice [60,61]. To this end, as demonstrated in patients with RA in clinical remission, the possibility of exploiting a less-invasive US-guided synovial biopsy procedure might be valuable [62]. Timelines for repeated biopsy assessment are not validated, and the numbers of drugs tested in these studies were necessarily low. Formally, a variation in a response biomarker should antedate the response to treatment; therefore, the utility of assessing synovial response after 12 weeks could be questioned. Moreover, synovial histological patterns, well-defined for RA [59,63], were not explored in PsA studies, and semi-quantitative scoring systems, advocated by OMERACT [64] and EULAR [65] to be used in synovial tissue research, were not homogeneously described in the included studies, with a significant heterogeneity in reporting that contributed to the low generalizability of the results, with no application of Krenn’s synovitis score [66]. As previously said, apart from the reduction in cellular components of the synovial inflammatory burden using IHC, longitudinal studies explored other relevant effects of bDMARDs, combining IHC with other relevant laboratory techniques, elucidating a more complex mechanism of action. Histologic evaluation was used to study the process of neovascularization in the synovium, affected by systemic drugs, as well as their effects on synovial lining layer thickness. As suggested by some authors [34], an effective treatment is mostly involved in modulation of ongoing inflammation in the short term, while the structural recovery might occur after a more prolonged treatment duration, and this cannot always be the case [67]. While apoptosis of cellular elements seemed not to be affected by TNFis in short-term studies, several reports pointed towards a reduction in MMP release and synthesis, expression of adhesion molecules such as ICAM-1, VCAM-1, E-selectin, responsible for white blood cell diapedesis and chemotaxis, NFκB and MAPK signaling, and pro-inflammatory cytokines transcription using bDMARDs, not restricted to TNFis. These aspects underline an environment with reduced cell trafficking, cell migration and/or tissue infiltration, possibly leading to less structural damage. It must be noted that bulk mRNA analysis was applied in the totality of studies assessing mRNA expression, with no data regarding single cell RNA sequencing [68]. The effect of DMARDs at the single cell level should be investigated in future studies. With in vitro studies, a net effect towards a reduction in cytokines or chemokines release mediated by JAKis or bDMARDs in FLS or synovial explant cultures was enhanced [45,46,47], with anti-invasive and anti-migratory actions on FLS, coupled with a metabolic shift from glycolysis to a more aerobic oxidative phosphorylation [45,46], similarly to that demonstrated in RA [69]. As a matter of fact, in vitro studies, especially when they involve non-homogeneous adoption of drug concentrations and time exposures, remain mostly mechanistic, and the real impact on generalizability of the results should always be fully considered. Given these observations and reasonings, the degree to which these kinds of works can have clinical impact in the mid-term remains to be fully elucidated. As underlined, both types of studies assessed (longitudinal and in vitro) face intrinsic limitations that prevent their full clinical application and, ideally, a biomarker of response should be retrieved before starting a systemic treatment, not after its adoption. To this end, current predictive approaches based on baseline synovial features/biomarkers (not addressed in this SLR) have failed in identifying biomarkers of response applicable to clinical practice in chronic inflammatory arthritis [10], but new and important results are emerging in RA [70,71], and it is expected that positive flags will follow also in the context of PsA. Moreover, only few articles reported data on paired synovial and skin biopsies, showing contradictory results regarding a possible unidirectional trend in the reduction of CD3+ cells in different tissues [32,41]. Whether the effect of b/tsDMARDs is directed primarily towards the synovium instead of other immune-competent organs is still under investigation. Our work has some limitations. For some articles, the cohort of patients was not restricted to PsA patients, since enrollment was allowed for subjects also suffering from other types of spondyloarthritis different from PsA, and separation of the results according to disease sub-categories was not always possible. However, we performed sensitivity analysis in our meta-analysis excluding studies involving SpA patients, underlining an effect that was even more specific to PsA for CD3+ cells. Full articles published after the deadline of our SLR did not vary substantially from the conclusions inferred by our SLR [72]. We also excluded animal models from our review, as well as cell lines, FLS purchased from companies, or synovial fluid analyses, and we did not focus on studies assessing remission under bDMARD treatment in which no baseline biopsy was performed [67]. Moreover, we did not focus on entheseal biopsies under b/tsDMARD treatment, and we assessed in longitudinal studies only drugs administered using the routes of administration formally approved by the EMA (e.g., intra-articular administration of TNFis was excluded) [73,74]. However, our work has some relevant strengths, as, to the best of our knowledge, this is the first SLR assessing the synovial effect of approved b/tsDMARDs in PsA, and the addition of a meta-analysis quantifies and strengthens the most important variations observed. Despite a high heterogeneity among the biomarkers evaluated, the reduction in CD3+ and CD68 + sl during the first 3 months of treatment with bDMARDs, mostly, but not restricted to, TNFis, represents the most consistent variation reported in the literature, but the synovial effect of DMARDs is much more variable than a pure anti-inflammatory effect. The possibility of easily collecting synovial material using US-guided synovial procedures might increase the information regarding the effect of the always-increasing number of DMARDs approved for the systemic management of this condition, and the possibility of comparing several drugs in head-to-head studies, even considering other key extra-synovial involvements such as skin and entheses, exploiting international collaborations and more sophisticated laboratory techniques or data synthesis methods, can change the actual knowledge of the synovial effects of these drugs and the actual treatment paradigm of a "trial and error" approach in psoriatic disease management.

PMC10002906

36725208

Aaron M Walsh,John Leech,Curtis Huttenhower,Hue Delhomme-Nguyen,Fiona Crispie,Christian Chervaux,Paul D Cotter

Integrated molecular approaches for fermented food microbiome research

01-02-2023

fermented foods,microbiome,high-throughput sequencing,metagenomics,transcriptomics,metabolomics

Abstract Molecular technologies, including high-throughput sequencing, have expanded our perception of the microbial world. Unprecedented insights into the composition and function of microbial communities have generated large interest, with numerous landmark studies published in recent years relating the important roles of microbiomes and the environment—especially diet and nutrition—in human, animal, and global health. As such, food microbiomes represent an important cross-over between the environment and host. This is especially true of fermented food microbiomes, which actively introduce microbial metabolites and, to a lesser extent, live microbes into the human gut. Here, we discuss the history of fermented foods, and examine how molecular approaches have advanced research of these fermented foods over the past decade. We highlight how various molecular approaches have helped us to understand the ways in which microbes shape the qualities of these products, and we summarize the impacts of consuming fermented foods on the gut. Finally, we explore how advances in bioinformatics could be leveraged to enhance our understanding of fermented foods. This review highlights how integrated molecular approaches are changing our understanding of the microbial communities associated with food fermentation, the creation of unique food products, and their influences on the human microbiome and health.

Integrated molecular approaches for fermented food microbiome research Molecular technologies, including high-throughput sequencing, have expanded our perception of the microbial world. Unprecedented insights into the composition and function of microbial communities have generated large interest, with numerous landmark studies published in recent years relating the important roles of microbiomes and the environment—especially diet and nutrition—in human, animal, and global health. As such, food microbiomes represent an important cross-over between the environment and host. This is especially true of fermented food microbiomes, which actively introduce microbial metabolites and, to a lesser extent, live microbes into the human gut. Here, we discuss the history of fermented foods, and examine how molecular approaches have advanced research of these fermented foods over the past decade. We highlight how various molecular approaches have helped us to understand the ways in which microbes shape the qualities of these products, and we summarize the impacts of consuming fermented foods on the gut. Finally, we explore how advances in bioinformatics could be leveraged to enhance our understanding of fermented foods. This review highlights how integrated molecular approaches are changing our understanding of the microbial communities associated with food fermentation, the creation of unique food products, and their influences on the human microbiome and health. Microbial exposures are critical for health, both with respect to the human microbiome and from external, environmental organisms. Food microbiomes are uniquely positioned to span both of these ecologies. This is particularly true of fermented foods, which are defined as those made through desired microbial growth and enzymatic conversions of food components (Marco et al. 2021). These foods consist of >5000 global varieties (Tamang and Kailasapathy 2010), many of which are unique to specific locations and populations. Generally, fermented foods are grouped based on the food substrate, e.g. dairy, vegetables, and meat. However, within these groupings there can be many subcategories, and corresponding differences in the associated microbiota (Leech et al. 2020). Indeed, even the microbiota of particular families of fermented foods can vary, e.g. temporally or spatially, with consequences for the underlying basic biology, product quality (Walsh et al. 2016), and induced phenotypes (van de Wouw et al. 2020). Fermentations can be driven by microorganisms that are endogenous to the raw food substrate or the production environment, i.e. spontaneous fermentations, or driven by microorganisms that are added to the food (Fig. 1). Traditionally, the latter was often achieved using ‘backslopping’, which involves adding a fraction of the previously fermented food to new food substrates to instigate a new fermentation. This is one example of a broader approach to the use of undefined cultures, which are consortia of the microorganisms that are responsible for initiating the fermentation, i.e. starters, as well as adjuncts, which do not contribute initially but become important later in the process. In the 19th century, the realization that microorganisms are responsible for fermentation led to the isolation of strains from foods (Caplice and Fitzgerald 1999). Subsequently, fermented foods have been produced in industry using these strains as defined starters, i.e. known cultures that are added to foods in specific quantities, and the optimization of these processes has been the subject of extensive research (Gibson et al. 2017, Hatti-Kaul et al. 2018). Over the last 20 years, molecular profiling techniques have revolutionized our understanding of the microbiota of fermented foods (Walsh et al. 2017). Genome-based characterization of individual starters and adjuncts provided valuable initial insights, while characterization of spoilage and pathogenic microorganisms provided information as to how to better control these contaminants (Garrigues et al. 2013). Marker gene and metagenomic sequencing techniques have proliferated in recent years (Fig. S1), with the latter approach in particular making it possible to study the broader microbial ecologies of these foods (De Filippis et al. 2017), facilitating greater understanding of the taxonomic composition of the microorganisms present and of their associated genetic arsenal, responsible for the associated biochemical changes in an environment. It is important to acknowledge the limits of high-throughput DNA sequencing (HTS), which include strain-level characterization of certain taxa or detection of less abundant microbes in a given food. Most recently, transcriptomics have begun to provide information on, how, and when relevant microbial genes are expressed (De Filippis et al. 2018; Fig. S1). Other molecular approaches add further layers of information regarding fermentations’ biochemical environments. These include proteomics (Carrasco-Castilla et al. 2012) for identifying important enzymes and metabolomics for profiling volatile compounds or other metabolites (Rizo et al. 2020). Studies deploying combinations of these technologies are increasing. These are critical for understanding, targeting, and developing previously unharnessed or under-utilized properties and bioactivities of fermented foods. This is particularly important in light of the current resurgence in popularity of artisanal fermented foods in Western society, largely driven by a health-conscious market. This review addresses how combining information from metagenomic, transcriptomic, and other molecular studies can be used to address important research questions of relevance to fermented foods, and indeed of broader significance to microbiology, while also improving upon the characteristics that consumers and industry seek to fully realize the potential of these foods. Food fermentation is the result of the biological activity of microbes present within food matrices (Marco et al. 2021). HTS enables high-quality culture-independent characterization of microbial communities, including those present in fermented foods (Leech et al. 2020). Three different HTS approaches have been used to characterize the microbiota of fermented foods: amplicon sequencing, whole metagenome shotgun sequencing, and metatranscriptomics (also known as RNA-Seq). We have compiled a comprehensive list of studies to have used HTS to characterize the microbiota of fermented foods that includes those published from 2009 until September 2019 (Table S1). Amplicon sequencing has been the most frequently used HTS approach for the characterization of the microbiota of fermented foods (Cao et al. 2017, De Filippis et al. 2017; Fig. S1). Although it has yielded many novel insights into the microbial diversity in these foods (Kergourlay et al. 2015), it has some inherent limitations, including the absence of functional information. Additionally, in general, short-read amplicon sequencing is limited to genus-level classification, although it has been demonstrated that long-read amplicon sequencing can achieve (sub)species-level classification (Karst et al. 2021). Shotgun metagenomics yields considerably more information than amplicon sequencing, including the functional profile of the microbiome (Leech et al. 2020), but at a higher cost. Strain-level identification is also possible with shotgun metagenomics but this is often only suitable for dominant species. It can also be challenging to disentangle mixtures of strains from the same species within a sample, although improved tools are starting to make this achievable (Vicedomini et al. 2021, van Dijk et al. 2022). Notably, long-read shotgun metagenomics could facilitate improved strain-level analysis, and tools such as Strainberry have already been developed that use long reads to recover strain-resolved genomes from metagenomes (Vicedomini et al. 2021). Metatranscriptomics is more costly again, and fermented foods can be difficult to extract high-quality mRNA. However, it can be a powerful tool for examining foods as it allows the examination of gene expression. Each sequencing approach mentioned above can be used in conjunction with other omics methods, such as metabolomics or proteomics, to achieve multi-omic (Franzosa et al. 2015) analyses of fermented foods (Fig. 2). Such analyses are employed to link changes in the proportions, functional potential, or gene expression of microbes with biochemical changes that occur during food fermentations. Here, we review studies that have used high-throughput sequencing, with an emphasis on shotgun metagenomics, metatranscriptomics, or multi-omic approaches that have been used to analyse several common fermented foods. We discuss the ways in which the information gained from such analyses might be applied to enhance food qualities such as flavour. Additionally, we explore the potential for novel bioinformatics or computational biology methods to further our understanding of food fermentations. A schematic outline of these approaches is presented in Fig. 2. A cornucopia of fermented foods have been characterized using HTS, initially by amplicon sequencing (Kim et al. 2002), followed later by shotgun metagenomics (Jung et al. 2011), propelling our understanding of these products. The majority of studies have focused on cataloging what taxa are present in fermented foods. Like any other microbial ecosystem, HTS can reveal many taxa that cannot be found by current culture-based techniques, therefore revealing deeper insights into these communities. Taxonomic profiling of fermented foods has revealed a large degree of interchangeability between species/strains fermenting the same food/substrate, highlighting the variable nature of fermented food microbiomes. Longitudinal studies that examine the community dynamics over the course of a fermentation reveal the periods at which certain taxa dominate, and pairing with metabolomics data, the contributions of specific taxa for the progression of the fermentation can be uncovered (Walsh et al. 2016). Such approaches are vital for understanding the contribution of the overall community to the final product. More recently, several studies have examined large numbers of fermented foods to examine broader patterns of microbial ecology. Walsh et al. (2020) examined 184 cheeses, including 77 new samples and 107 samples in publicly available databases. The study included volatile data on the 77 new samples. By examining multiple samples simultaneously, the authors uncovered that strain-level differences within the cheese microbiome had a significant impact on the resulting volatile profile. The authors also uncovered antimicrobial resistance genes, but with a low risk of transfer between microbes. This study is also a great example of the power of HTS, as the interactions between viruses and prokaryotes were examined. Analyses of CRISPR and anti-CRISPR proteins revealed the importance of phage for transferring genetic information between bacteria. Another recent study examined the microbial diversity between sourdough starter cultures from four different continents (Landis et al. 2021). Interestingly, the study did not see a geographic effect between sourdoughs from different countries. The study uncovered strong co-occurrence patterns between different microbes, highlighting the importance of a microbial-mediated community structure. The study also revealed the importance of acetic acid bacteria and their contribution to the flavour and dough-rise of sourdough bread. Finally, Leech et al. (2020) examined 58 artisanal-produced fermented foods from eight different countries. Again, no geographic signal was discovered across the 58 food microbiomes. The authors did uncover strong patterns in microbial ecology across the different substrates used, showing that dairy-based, salt-based (foods fermented in 2% saline, e.g. sauerkraut), and sugar-based (foods containing high sucrose at the beginning of fermentation, e.g. kombucha) fermented foods had different communities and functional profiles to each other. The study also revealed that dairy-based samples had lower alpha diversity than salt-based and sugar-based fermented foods. These three studies are a demonstration of the utility of HTS for looking at broader patterns across fermented foods, and how these insights can be useful for the commercial application of these ecological insights. A 2014 study by Wolfe et al. pioneered the use of fermented foods as models to explore the forces that might shape microbiota. The authors reported that microorganisms from cheese rinds were easy to culture (Wolfe et al. 2014). Importantly, they were able to recreate the cheese rind microbiota in vitro. In this study, HTS revealed that moisture influenced the composition of these communities; although this finding was unsurprising, it proved that fermented foods can be used as models to assess the impact of abiotic factors on microbiota. Subsequently, the use of cheeses as models has been extended to study several processes that might mould microbiota, including horizontal gene transfer (HGT). Bonham et al. sequenced the genomes of 165 bacteria isolated from cheese rinds, and they reported that HGT was frequent between these genomes (Bonham et al. 2017). Notably, many of the transferred genes were involved in the acquisition of nutrients from the environment, including iron, which is limiting on cheese rinds; such findings highlight the importance of the availability of nutrients on microbiota. Cheese models have also been used to study interactions between the members of microbiota. Morin et al. grew mutants of Escherichia coli in cheese rinds to identify genes that were involved in interactions between microorganisms in this environment (Morin et al. 2018). The mutants were generated using an approach called Random Barcode Transposon-site Sequencing (RB-Tn-Seq), which measures the fitness of genes under a given condition by counting the abundance of random barcodes at transposon insertion sites. The mutants were co-cultured separately with each member of a three-species community (Hafnia alvei + Geotrichum candidum + Penicillium camemberti), and they were also grown with the entire community simultaneously. RB-Tn-Seq indicated that many of the interactions that occurred when mutants were grown as part of a pair did not occur when the mutants were grown as part of the community. In other words, the way in which strains interacted differed depending on the context in which they were grown. Cheese models have highlighted the importance of cross-kingdom interactions in fermented foods. For example, studies have shown that the release of siderophores by fungi can influence the growth of bacteria on cheese rinds. Kastman et al. observed that the establishment of Staphylococcus equorum on cheese rinds was assisted by the mould Scopulariopsis, and RNA-Seq suggested that this might have been explained by the release of iron by the mould, which was utilized by the bacterium (Kastman et al. 2016). Similarly, Pierce et al. used RB-Tn-Seq to show that fungi could influence the growth of bacteria on cheese rinds by modulating the availability of cofactors (Pierce and Dutton 2022). Intriguingly, Zhang et al. discovered that Serratia on cheese rinds can travel along the hyphae of the fungi Mucor (Zhang et al. 2018). To determine the mechanism by which this phenomenon occurred, they employed transposon mutagenesis to generate Serratia mutants, and Whole Genome Sequencing (WGS) of Serratia mutants revealed that flagella were essential for dispersal along the hyphae. Elsewhere, Cosetta et al. used metabolomics to show that volatiles produced by fungi promoted the growth of Vibrio in cheese rinds (Cosetta et al. 2020). Aside from cheese models, the kefir microbiome has been proposed as another model for studying microbial communities. Recently, Blasche et al. used integrated approaches to unravel the interactions between species in kefir over the course of fermentation (Blasche et al. 2021). 16S rRNA gene sequencing analysis revealed that Lactobacillus kefiranofaciens dominated in the kefir but, surprisingly, it was found that L. kefiranofaciens was unable to grow in milk by itself. Consequently, the authors postulated that L. kefiranofaciens needed to cooperate with other members of the community to propagate during fermentation. Indeed, when L. kefiranofaciens was co-cultured with Leuconostoc mesenteroides, cross-feeding between the two species was observed, wherein L. kefiranofaciens made amino acids available for L. mesenteroides, which in turn made lactate available for L. kefiranofaciens. Here, we discuss the use of HTS approaches to predict what volatiles are produced by microorganisms, and we examine the use of integrated and experimental approaches that combine HTS with metabolomics to understand the relationship between the microbes and volatiles. We refer to these two approaches as (a) correlative/predictive (Fig. 2B, C, and D) and (b) causative/experimental. The correlative/predictive approach leverages the findings of research from the 20th century that established the enzymology of the development of flavours during fermentation (Smit et al. 2005). Specifically, HTS enables us to detect genes that are associated with the production of volatiles in foods. For example, analysis of the koumiss microbiome, a traditional fermented milk product consumed in parts of Central Asia, revealed that it contained genes that are important for flavour, including those associated with proteolysis (Yao et al. 2017). In dairy products, proteolysis of caseins releases amino acids that can serve as precursors to volatiles. The koumiss microbiome also contained an aminotransferase involved in the transaminase pathway that initiates the formation of volatiles (e.g. aldehydes, acids, alcohols, esters, etc.) or lyases that are involved in the production of sulphur compounds. The Cojita microbiome, a Mexican cheese, contained aldehyde dehydrogenases that convert aldehydes to acids, alcohol dehydrogenases that can convert aldehydes to alcohols, and genes involved in lipolysis, a source of volatiles (Escobar-Zepeda et al. 2016). Metatranscriptomics analysis of an industrial Camembert-type cheese over a 77-day ripening period indicated that genes involved in the production of sulphur compounds were expressed mostly by Geotrichum candidum, while those involved in lipolysis were expressed mostly by Penicillium camemberti, which suggested that the contributions of these fungi to the flavour of the cheese were distinct (Lessard et al. 2014). Interestingly, metatranscriptomics analysis of a traditional Italian Caciocavallo Silano cheese revealed that nonstarter lactic acid bacteria (LAB) contributed to amino acid metabolism during ripening (De Filippis et al. 2016). Shotgun metagenomic analysis of the cheese rind microbiome revealed that Pseudoalteromonas species contained methionine gamma-lyase, which is involved in the production of sulphur compounds. Only Brevibacterium linens was previously described to produce this enzyme in cheese (Yeluri Jonnala et al. 2018). In a recent meta-analysis of the cheese microbiome, metagenome-assembled genomes were found belonging to species that had not been characterized before, several of which were predicted to secrete compounds that might influence the quality of cheese, including acetate (Walsh et al. 2020). Notably, in this study, Walsh et al. also integrated metagenomics with metabolomics, and it was found that strain-level variation corresponded to differences in the volatilome, the set of volatile compounds present in a sample, which is consistent with results reported elsewhere (Niccum et al. 2020). For example, two strains of B. linens showed distinct patterns of correlation with butanoic acid, wherein one strain was significantly positively correlated with the compound, whereas the second strain was significantly negatively correlated with it. Such findings highlight that species-level correlations need to be interpreted with caution. HTS has been particularly valuable when applied in spontaneously fermented foods. Metatranscriptomics analysis of kimchi revealed that only Leuconostoc species expressed mannitol dehydrogenase genes during fermentation, and thus were solely responsible for mannitol production (Jung et al. 2013). Additionally, genes involved in the production of the acetoin, diacetyl, and 2,3-butanediol were highly expressed in L. mesenteroides, further highlighting its importance with respect to flavour development in kimchi (Chun et al. 2017). An extension of the predictive approaches described above is metagenome-scale metabolic modelling, a method which uses the metagenome to predict which enzymes, and ultimately metabolites, may be produced by the microbiome (Magnúsdóttir and Thiele 2018). It has been demonstrated that such an approach accurately predicted the metabolites produced by the gut microbiota in obese humans (Shoaie et al. 2015). Given the relative simplicity of fermented food microbiota compared to, e.g. the communities found in the human gut, it is plausible that metagenome-scale metabolic modelling may be applied to these communities to predict the production of flavour compounds. A challenge posed by modelling is the requirement for curation to ensure the quality of models, which, until recently, was done manually (Henry et al. 2010). However, with the release of CarveMe (Machado et al. 2018), it is now possible to automatically construct genome-scale metabolic models for the members of a community. CarveMe might be used to construct models from genomes that are recovered from metagenomes, using tools like MetaBAT (Kang et al. 2015), to predict the volatiles produced by the entire community or by different combinations of its members. Such an approach may lead to the large-scale identification of candidates as starters, but it might also inform efforts to tailor ingredients for the development of flavours of interest. The integration of HTS with metabolomics has been employed to explore the potential contributions of microorganisms to the flavour of fermented foods. Typically, this has involved correlating the abundances of taxa with the concentrations of metabolites, and most studies to use this approach have integrated amplicon sequencing with metabolomics to examine the contribution of genera to volatiles. In a study of Chinese liquor fermentations, the approach revealed that microorganisms that originated from the surfaces of the facility in which the beverages were produced were correlated with the levels of metabolites in the products (Bokulich et al. 2014). Other studies have integrated shotgun metagenomics with metabolomics to examine the contributions of species to volatiles. For example, analysis of kefir over the course of 24-hour fermentations revealed that Lactobacillus kefiranofaciens correlated with carboxylic acids, ketones and esters. In contrast, Leuconostoc mesenteroides correlated with 2,3-butanedione and acetic acid (Walsh et al. 2016). These correlations suggested a causal relationship between the microbiota and the flavour of kefir, which was supported by experimental evidence, spiking milk kefir with both organisms separately. Sensory analysis validated these findings by showing that a kefir with a high relative abundance of L. mesenteroides had a likeable buttery flavour, whereas another kefir with a high relative abundance of L. kefiranofaciens had a less likeable but fruitier flavour. Similarly, an integrated approach was employed to characterize surface-ripened cheeses during a 30-day ripening period (Bertuzzi et al. 2018). Again, strong correlations were identified between the relative abundances of individual species and the levels of flavour compounds in the cheeses. Importantly, these correlations were supported by evidence from prior studies that had shown that these species can produce such compounds. Interestingly, Staphylococcus xylosus, which had only previously been associated with sulphur compounds in meats, was also found to correlate with sulphur compounds in these cheeses. It is crucial to recall the adage that ‘correlation is not causation’ when interpreting findings from integrated approaches. Noecker et al. reported that in simulated microbiome–metabolome datasets, wherein the producers of each metabolite were known, correlation analysis produced false-positives in 50% of cases, and identified relationships between unrelated variables (Noecker et al. 2016). However, researchers can take factors into account to mitigate this issue when interpreting correlations, including: (a) the likelihood that a correlation is valid increases if that correlation persists across subsets of data, and (b) if a microorganism contains genes associated with the production of a metabolite. Correlations should also be validated experimentally. As previously mentioned, strains of the same species can influence flavour differently (Pereira et al. 2017, Niccum et al. 2020, Walsh et al. 2020, Jimenez-Lorenzo et al. 2021). Strain-level characterization can be a challenge but several tools have been released that enable strain-level characterization of the microbiome (Truong et al. 2017, Olm et al. 2021, Vicedomini et al. 2021, van Dijk et al. 2022). The integration of these tools with metabolomics represents an opportunity to evolve the integrated approach discussed above (Walsh et al. 2020). Potentially, such an improvement might help in the identification of strains that produce volatiles of interest, thus facilitating: (a) the selection of starters, and (b) the construction of models that can predict the metabolome of a sample based on the strains that are present in that sample to track the progress of fermentation. Ultimately, these approaches can only help us to predict the role of microbes in the development of flavour, and experimentation would be required to test the validity of these predictions. For example, such an experiment might involve (1) spiking a fermented food with an inoculum of a given microbe, and (2) performing sensory analysis to determine if the microbe caused a change in flavour. A Zhenjiang vinegar study is a great example of the experimental approach where shotgun metagenomics combined with pathway analysis indicated that A. pasteurianus, in addition to Lactobacilli species, had the ability to synthesize acetoin (Wang et al. 2016). Co cultures of isolated bacteria from the vinegar were found to produce more acetoin than mono-cultures in vitro and in situ when inoculated in vinegar. Also, in Chinese liquor, metatranscriptomics was used to measure the expression of genes associated with the production of two sulphur compounds, 3-(methylthio)-1-propanol and dimethyl sulphide. Saccharomyces expressed every gene necessary to produce both compounds, while Lactobacilli expressed genes involved in recycling methionine, a precursor to these compounds. Saccharomyces cerevisiae and L. buchnerii were isolated from the liquor and cultivated in monoculture or co-culture. While L. buchnerii monocultures produced neither compound, co-cultures produced significantly more of the sulphur compounds than S. cerevisiae monocultures, thus confirming a synergistic relationship between these species. Finally, as mentioned above, experimental evidence confirmed the contribution of specific kefir bacteria to the flavour profile (Walsh et al. 2016). To date, too few studies have adopted such an approach, but it is crucial that the field moves in this direction to translate our improved understanding of the fermented food microbiome into action. As discussed above, microorganisms can produce metabolites that improve the organoleptic qualities of fermented foods. The opposite is also true: spoilage can be caused by overfermentation, which occurs when microbial metabolites (e.g. acids, alcohols) are excessively produced. Often, this can be rectified by modifying the parameters of fermentation (e.g. duration, temperature). Alternatively, spoilage can be caused by microbial contaminants. In Chinese rice wine, shotgun metagenomics provided strong evidence that Levilactobacillus brevis caused spoilage of the beverage (Hong et al. 2016). Taxonomic analysis revealed that this species was prevalent in spoiled wine, while functional analysis revealed the presence of genes involved in biotin biosynthesis and short-chain fatty acid production, which were thought to contribute to off-flavours (Hong et al. 2016). In nunu, a yoghurt-like, milk-based fermented product widely consumed in West Africa, Enterobacteriaceae genes for the biosynthesis of putrescine, which causes an unpleasant odour, were identified (Walsh et al. 2017). In fermented seafood, several studies have observed that levels of Halanaerobium corresponded with increases in spoilage metabolites, including methylamines (Lee et al. 2014, 2015). Other defects in the qualities of fermented foods, such as pigments that discolour cheeses, are costly for producers. Shotgun metagenomics revealed that Thermus thermophilus, which is not associated with the typical cheese microbiota, has been found to be enriched in cheeses with a pink discolouration defect (Quigley et al. 2016). Carotenoid biosynthesis genes were correspondingly enriched in those cheeses. It was observed that the pinking defect was reproduced in experimental cheeses inoculated with T. thermophilus (Quigley et al. 2016). Similarly, Psychrobacter has been linked to the purpling of cheese (Kamelamela et al. 2018). Analysis of Psychrobacter genomes recovered from cheeses revealed that these bacteria contained an enzyme, which can result in the production of the pigment indigo. While spoilage during fermentation is unpleasant and can have economic consequences, in extreme cases, specific pathogens and negative microbial biochemistry can pose a true health hazard. In addition to the insights that genomics has provided into various food pathogens, HTS can also be used to detect pathogens in fermented foods and/or associated production facilities. Indeed, some such studies have detected Enterobacteriaceae in African fermented milk products (Walsh et al. 2017, Parker et al. 2018) and, in one instance, shotgun metagenomics of nunu samples highlighted the presence of pathogenic strains, similar to strains that had caused illness (Walsh et al. 2017). It is crucial to note that the sensitivity of shotgun metagenomics for the detection of pathogens is lower than that of methods such as qPCR (Andersen et al. 2017), which has a limit of detection of 104–105 CFU/ml (Hazards et al. 2019). Instead, as currently employed, shotgun metagenomics is perhaps more suited for identifying the source of outbreaks by analysing suspected foods (Buytaers et al. 2021). HTS has also been used to identify producers of histamine, a biogenic amine that can be harmful to some consumers, in cheese and other fermented foods (O’Sullivan et al. 2015). Also of potential concern is the presence of bacteria with antibiotic resistance genes (ARGs), especially ARGs present on mobile elements, which could in principle transfer to members of the gut microbiota after ingestion (Maeusli et al. 2020). Analysis of the resistome of cheeses revealed that ARGs were not present on the plasmids of LAB, but were detected on the plasmids of Enterobacteriaceae (Walsh et al. 2020), which are generally microbial contaminants. Thus, improving hygiene during cheese manufacture might reduce reservoirs of resistance in the food. Finally, throughout history, fermentation has been used as a means of preserving foods. Fermentation continues to contribute to shelf-stability today as a means to avoid or decrease reliance on chemical preservatives and to preserve foods in regions where refrigeration is limited. During fermentation, microorganisms secrete compounds (e.g. acids, alcohols, diacetyl, bacteriocins, and others) that inhibit the growth of contaminants (including spoilers and pathogens) (Ross et al. 2002). Notably, HTS has been used to screen a variety of fermented food microbiomes for bacteriocin genes (Leech et al. 2020, Walsh et al. 2020). Approaches available for bioprospecting of bacteriocin genes in fermented food microbiomes include detection by homology to known bacteriocin gene clusters (Hammami et al. 2010), Hidden Markov models (HMM)-based tools (Morton et al. 2015, van Heel et al. 2018), or machine learning (Hamid and Friedberg 2019). The application of these can contribute to the selection of starters that produce bacteriocins targeting undesirable microorganisms. In addition to food quality considerations for consumers, fermented foods have great potential in maintaining and improving human health (Fig. 3) (Soedamah-Muthu et al. 2013, Eussen et al. 2016, Wastyk et al. 2021). This has long been of interest, but the great diversity and variability both of fermentation microbiomes and of the human microbiome have made it difficult to design and target benefits in a precision manner. Molecular techniques now allow us to engineer not just the fermentation process, but the resulting live cell and chemical effects in human hosts and populations. At a population scale, fermented foods, or at least those containing living microorganisms, occupy a unique position with respect to long-term human health, as one of the only practical ways to ‘chronically’ deliver beneficial live microbes concordantly with microbial chemical products outside of a specifically therapeutic context (Rezac et al. 2018). That is, while live cell therapies and fecal microbiota transplants are under intense study for treatment of acute conditions (Kelly et al. 2014, Youngster et al. 2016), they are not generally appropriate for regular, long-term use. Other types of dietary or prebiotic interventions are practical for maintenance of microbes already present in the gut (Scott et al. 2008, Cotillard et al. 2013), but they are not capable of reliably introducing new ones unless part of synbiotics. As food fermentation microbes can have beneficial effects on the gut biochemical environment during transit, but do not generally engraft stably after short exposures (Derrien and van Hylckama Vlieg 2015), adherence to regular consumption must be part of any practical ‘wellness maintenance’ applications of fermented foods. Despite this potential, to date, there has been a shortage of large-scale investigations into the relationship between fermented food consumption and with gut microbiome modulation or, indeed, health biomarkers. It is also important to appreciate that the microbiomes and metabolomes, of specific fermented foods can differ greatly and thus the findings from studies with specific fermented foods cannot be extrapolated across to other fermented foods. Smaller cohort studies have been carried out on a range of foods, most commonly dairy, with variable results (Marco et al. 2017, Markowiak and Śliżewska 2017, Stiemsma et al. 2020). Consistent benefits have been shown for circulating lipid metabolism and corresponding cardiometabolic health in particular (Soedamah-Muthu et al. 2013, Chen et al. 2014, Lim et al. 2015, Eussen et al. 2016), while gut-local conditions such as Inflammatory Bowel Disease (IBD) (Bengmark 2007, Geier et al. 2007), Irritable Bowel Syndrome (IBS) (Laatikainen et al. 2016), or diarrhoea (Parvez et al. 2006, Nagata et al. 2016) often respond in a much more population- and study-specific manner. Perhaps one of the most critical, least studied links between fermented foods and health outcomes is their role as part of the ‘disappearing microbiome’ (Segata 2015). During the period of time during which fermented foods became less popular in Western society, there was a corresponding increase in the application of other food preservation approaches. As a consequence, and as introduced above, the shelf life of many products across the globe has been extended considerably over the past century (Boor et al. 2017). As with other changes in early life microbial exposure—e.g. livestock (Kim et al. 2019), antibiotics (Langdon et al. 2016), breastfeeding (Bäckhed et al. 2015), and Caesarian section (Bokulich et al. 2016)—it is possible that the decreased exposure to microbes, either within fermented foods or ‘unwanted’ growth in the form of nonharmful food spoilage, may have unintended consequences on immune development over the course of a lifetime (Olivares et al. 2006). Indeed, a recent study examining the effect of consuming fermented foods found an increase in microbial diversity and an improvement of the inflammatory status of the participants who increased their daily fermented food intake (Wastyk et al. 2021). Despite this promise, a great deal of additional study is required in this area. There are frequent misunderstandings about the expected influences of fermented foods, and associated microorganisms, on the gut microbiome. In most cases, fermentation-associated microorganisms evolved to, by definition, ferment food substrates rather than persist in the human gut (Bachmann et al. 2012). However, it has been shown that particular species of LAB found in foods are often closely related to those found in the gut (Pasolli et al. 2020). As a large bolus of live microorganisms, the presence and molecular effects of fermentation-associated microorganisms are transiently quite apparent via HTS or metabolomics (Zhang et al. 2016). Although not always the case, these transient effects can have quantifiable beneficial influences on resident gut microbiota structure, metabolism, or host immunity, inflammation or the gut brain axis (Derrien and van Hylckama Vlieg 2015, van de Wouw et al. 2020, Wastyk et al. 2021). In human subjects, live microorganisms from fermented foods can remain metabolically active in the gut, even without long-term engraftment, although the health consequences are less well understood (David et al. 2014). In limited cases, individual microbes do persist (Zhang et al. 2016, Milani et al. 2019) or transfer genetic material (Hehemann et al. 2010). However, although they may influence the dynamics of the process (Derrien and van Hylckama Vlieg 2015), in general, studies (or at least studies of faecal microbiomes) indicate that fermented food strains do not colonize (Fig. 3). In addition to the limitations of relying on faecal samples, it should also be noted that the majority of these examples derive from fermented milk products, with only extremely limited data generated to date for other fermented foods (Han et al. 2015, Abbondio et al. 2019, Jung et al. 2019). Specific examples of the consequences of transient impacts of fermented food consumptions, and in some cases its underlying molecular mechanisms, have been studied both in human subjects and animal models. In one of the earliest HTS-based examples, while consumption of a fermented milk product did not change gut microbial composition in gnotobiotic mice, it transiently shifted carbohydrate utilization towards an upregulation of plant glycan catabolism, possibly due to contributions from Bifidobacterium animalis subsp. lactis (McNulty et al. 2011). Similar effects proved to be locally anti-inflammatory due to mechanisms as diverse as resident ecological disruption (Veiga et al. 2010), lipid metabolism (Bourrie et al. 2018), serotonergic signalling (van de Wouw et al. 2020), quorum sensing inhibition (Rooks et al. 2017), and reactive oxygen mitigation (Ballal et al. 2015) in mouse models. At a molecular level, while breaking down complex compounds to simple molecules, the microorganisms responsible for fermentation synthesize enzymes, vitamins, essential amino acids, bioactive components, and potentially remove undesirable compounds (allergens, antinutritional factors). This leads to changes in texture and interaction between macro- and micronutriments. As a consequence, fermentation can improve the nutritional qualities and, in turn, health benefits of foods (Tamang et al. 2016, Şanlier et al. 2019). Kimchi is an example of a nutritionally enriched food with high levels of vitamins (e.g. vitamin C, β-carotene, and vitamin B), minerals, dietary fibers, and other bioactive compounds such as capsaicin, allyl compounds, gingerol, isothiocyanate, and chlorophyll. While the quantity of many of the vitamins are enhanced due to the biosynthesis of these vitamins through fermentation, other nutritional benefits are realized through increased bioavailability due to fermentation. These active compounds have, in turn, been associated with many health benefits attributed to kimchi (Park et al. 2014). The same is true for koumiss (Dhewa et al. 2015), miso (Watanabe 2013), and many other fermented foods. In all cases, specific microbiome components are responsible for the biotransformations, but in some cases, the specific microbes and enzymatic reactions have been elucidated thanks to insights provided by metagenomics and metatransciptomics. In kimchi, metatranscriptomics revealed that genes associated with folate biosynthesis were expressed by Latilactobacillus sakei, while genes associated with riboflavin biosynthesis were expressed by Leuconostoc mesenteroides (Jung et al. 2013). In fermented mung beans, Rhizopus induces high amounts of free amino acid and γ-amino butyric acid and has been proposed to have antidiabetic and antioxidant properties (Yeap et al. 2015). In kombucha, genes involved in the biosynthesis of B-vitamins were present in Komagataeibacter rhaeticus (Arıkan et al. 2020) but as with the flavour-enhancing biochemical mechanisms above, much work remains to identify nutrition-enhancing pathways in the general case. Another family of well-characterized nutrition-enhancing biotransformations includes the enhancement of antioxidant potential in fermented foods. Surprisingly, due to this process, some cheeses can have antioxidant levels close to that of vegetables or fruit juices (Fardet and Rock 2018). In kombucha, changes in the microbiota were found to correspond with increases in the levels of antioxidants, which may have arisen from microorganisms degrading polyphenols in the tea (Chakravorty et al. 2016). Furthermore, in tempeh, polyphenol, and isoflavone contents were significantly enhanced due to Rhizopus oligosporus activities (Kuligowski et al. 2017). Although it is not clear in each case what caused these enhancements, correlations within the data set point towards deglucosidation as a mechanism in some strains, but not in others. Information on the genetic profile and gene expression of these strains via metagenomics and metatranscriptomics can supplement such studies to better understand the strain-level differences and mechanisms for food enhancement. Fermentation also provides a natural means to reduce undesirable compounds. Lactose (of concern to lactose-intolerant individuals) is commonly reduced to free glucose, galactose, and/or lactate; antinutrients such as phytate and tannins are degraded to release and enhance bioavailability of minerals such as iron, zinc, and calcium (Blandino et al. 2003, Poutanen et al. 2009); protease inhibitors and lectins are reduced in favour of protein absorption (Nkhata et al. 2018); beta-galactosides responsible for gut discomfort (e.g. stachyose and raffinose) are hydrolysed (Mukherjee et al. 2016); and toxic substrates present in raw materials can be eliminated. For the latter, examples include reductions in aflatoxin B1 levels in Ogi (a Nigerian fermented sorghum porridge) (Dada and Muller 1983), and removal of cyanide from cassava fermented with S. cerevisiae (Iyayi and Losel 2000). While the reduction of aflatoxins in fermented foods is mainly attributed to cellular binding (both viable and heat killed) and growth inhibition of aflatoxin-producing Aspergillus spp., other toxicity-mitigating mechanisms are emerging. For instance, Huang et al. (2017) showed that L. plantarum may reduce toxicity by altering gene expression in the liver of the consumer. Some of our most basic gaps regarding fermented food microbiomes have been dictated by the assays used to study them to date. The majority of studies have used amplicon sequencing, i.e. 16S rRNA or ITS sequencing to determine the composition of the bacteriome and mycobiome, respectively. The adoption of shotgun metagenomics has been slow relative to other microbial community types, and among studies that have adopted the approach, the greatest focus has remained on taxonomic profiling of the bacteriome, with corresponding approaches for the mycobiome being limited by the lower numbers of food-relevant eukaryotic genomes in reference databases and/or the higher complexity of eukaryotic genomes (Table S1, Fig. S1). The application of shotgun metagenomics to study viromes also remains a relatively emerging area, despite the importance of phage in fermented foods (O’Sullivan et al. 2015). In the dairy industry, for example, phage infection remains the biggest cause of fermentation failure and leads to significant economic loss (Samson and Moineau 2013). However, phages can also be used as antimicrobial agents in food production, and thus be used to prevent contamination (Fernández et al. 2017). As in most microbial communities, fermented food viromes are technically challenging to study. They are best accessed through the enrichment of viral particles, but protocols for the extraction of viral nucleic acids from fermented foods are not yet standardized. Chloroform-based preparations adapted from other environments have been effective for cheeses (Dugat-Bony et al. 2020), while Muhammed et al. (2017) optimized ultracentrifugation for the purification of viral particles from dairies. Subsequent in silico analysis of viral sequences is notoriously challenging due to their great diversity, and under-representation in reference databases, limiting their analytical tractability even in the best case (Garmaeva et al. 2019, Wang 2020). Therefore, going forward it is imperative to expand reference databases, which might be achieved by sequencing isolates (Lagier et al. 2018) or recovering genomes from metagenomes in silico (Chen et al. 2020). Long-read sequencing has emerged as a promising technology that could be used for the characterization of the virome. Notably, using long-read metagenome assembly, Somerville et al. (2019) were able to recover phage genomes from natural whey cultures used in the production of Swiss Gruyère cheese. Since long-read sequencers generate reads longer than many phage genomes, long-read sequencing can also be used to recover complete phage genomes from samples without the requirement for assembly (Beaulaurier et al. 2020). Shotgun metagenomics is further hampered by the prevalence of genes with unknown functions in reference databases. Over the past decade, the number of available genomes has exploded, while the annotation of these genomes has stagnated (Salzberg 2019). If we do not know the function of genes in the fermented food microbiome, shotgun metagenomics by itself cannot explain the roles of microorganisms in fermentation, completely. Improving reference databases will be important to address this issue. This might be achieved by annotating the genomes of isolates from fermented foods, e.g. CRISPR interference (Wang et al. 2018) or Tn-Seq (Van Opijnen et al. 2009). Additionally, the utilization of multi-omics (including metabolomics and metaproteomics) in parallel might prove useful for inferring the function of members of fermented food microbiomes. One approach would be to replicate communities in vitro and investigate what happens to the metabolome or metaproteome when a specific strain is added to or removed from the community. The implementation of such approaches has the potential to maximize our understanding of the biology of fermented foods microbiomes. It should be noted that high-throughput metaproteomics is still somewhat challenging and, in the context of fermented foods, is further complicated by the frequently high levels of background proteins present in the food substrate. If such technical hurdles can be overcome, the rewards have the potential to be particularly great. Once the field is able to fill gaps relating to the composition of a broader variety of fermented food microbial systems and how they function, the logical progression is to influence and modify those functions towards desirable, beneficial outcomes. This is especially the case for products using ‘undefined starters’, which contain complex mixtures of microorganisms that can vary with time. In fermented foods, variability in the microbiota equates to variability in quality and, potentially, health benefits. The challenge for industry is to meet this demand while replicating the products authentically. We propose two avenues that producers might take to achieve this, i.e. controlling abiotic factors to shape the microbiota or using defined microbiota. The first strategy can be used to forcibly shape or select for desirable properties even when beginning from a nominally undefined community. As mentioned above, Wolfe et al. (2014) demonstrated that moisture influenced the convergence of cheese rind microbiota, and temperature can act as a similar, simple selective pressure. Indeed, De Filippis et al. (2016) showed that an increase in temperature accelerated the rate at which cheeses ripened by inducing an increase in the expression of genes involved in the development of flavour. Controlling the availability of nutrients can also affect fermented food microbiota. Analysis of kefir revealed that Lactobacillus kefiranofaciens decreased during fermentation. Interestingly, L. kefiranofaciens lacked genes involved in the biosynthesis of tyrosine, and its decrease coincided with a reduction in the concentrations of tyrosine during fermentation (Walsh et al. 2016). Overall, these studies suggest that HTS can inform efforts to optimize fermentations in ways that favour the growth of microorganisms with the phenotype of interest, by modifying abiotic factors. The second option, i.e. the use of defined starters, is already standard industrial practice, since it ensures the consistency of products. This sometimes comes at the cost of desirable properties lost due to the limitations of these controlled communities. Defined microbiota can be considered as a possible extension of this approach whereby strains identified as being key components of the undefined starters, on the basis of HTS-related insights into properties relating to flavour, tractability, and health benefits, could be targeted for isolation and combined. With respect to production, especially at scale, it will also be important to choose combinations of strains that will grow well together. In limited cases, it has already been demonstrated that the cheese rind microbiota can be replicated in vitro (Wolfe 2018), and therefore it is plausible that this can be done with engineered communities for other fermented food microbiota. Owing to the great variety of fermented foods, health benefit claims across different regulators and countries are a challenge. Beyond this, we also suffer from gaps in our basic biological knowledge of the potential health benefits of many fermented foods. These health benefits may derive from many sources, likely in combination: microorganisms in these foods, the metabolites they produce, or their impact on the raw food substrate on the host directly or indirectly through the resident human gut microbiome. Untangling the specific effects of microorganisms and metabolites/food bioactives can be facilitated through parallel investigations to determine the health benefits of foods from which the microorganisms have been removed (e.g. through centrifugation, filtration, or pasteurization) or the microbial cultures when not grown in the food substrate. These experiments, along with microbial genetic modifications and whole organism knock-ins/knock-outs from fermentation communities, can readily be performed in animal models, as performed by Bourrie et al. (2021), and in some cases in human subjects, given the wide range of strains already approved for food use. Multi-omic approaches to analyse host responses to fermented foods can also play a major role in elucidating health benefits. While population-scale observational studies are useful for investigating potential health benefits (Taylor et al. 2020), ultimately, there is a need for truly randomized, blinded, placebo-controlled trials using fermented foods, which have been largely absent to date. It will be impossible to standardize precision health or wellness maintenance based on fermented foods, however, without a much more substantial understanding of the underlying biochemical and microbial mechanisms. The ability to probe the composition and functionalities of the microbial populations present in fermented foods using the integrated approaches discussed above affords us an unprecedented opportunity to optimize various attributes of these foods. It is important that future studies validate observed correlations experimentally so that these insights can be translated. Importantly, integrated approaches have also helped to establish fermented foods as models for studying the interaction between microorganism within microbiomes, and we expect that technological improvements, such as enhancements relating to long-read sequencing, will help us to e.g. explore the role of phage in these microbiomes. Finally, integrated approaches are beginning to provide insights into impact of fermented foods on our health, although more studies are needed to support some of the claims attributed to these products. In conclusion, while integrated approaches have become central to the field of microbiology as a whole, they have, and are likely to continue to be, particularly transformative with respect to the microbiomes of fermented foods. Click here for additional data file.

PMC10002918

Urszula Demkow

Molecular Mechanisms of Neutrophil Extracellular Trap (NETs) Degradation

03-03-2023

neutrophil extracellular traps,degradation,DNAses,macrophages,autoimmunity,thrombosis,cancer

Although many studies have been exploring the mechanisms driving NETs formation, much less attention has been paid to the degradation and elimination of these structures. The NETs clearance and the effective removal of extracellular DNA, enzymatic proteins (neutrophil elastase, proteinase 3, myeloperoxidase) or histones are necessary to maintain tissue homeostasis, to prevent inflammation and to avoid the presentation of self-antigens. The persistence and overabundance of DNA fibers in the circulation and tissues may have dramatic consequences for a host leading to the development of various systemic and local damage. NETs are cleaved by a concerted action of extracellular and secreted deoxyribonucleases (DNases) followed by intracellular degradation by macrophages. NETs accumulation depends on the ability of DNase I and DNAse II to hydrolyze DNA. Furthermore, the macrophages actively engulf NETs and this event is facilitated by the preprocessing of NETs by DNase I. The purpose of this review is to present and discuss the current knowledge about the mechanisms of NETs degradation and its role in the pathogenesis of thrombosis, autoimmune diseases, cancer and severe infections, as well as to discuss the possibilities for potential therapeutic interventions. Several anti-NETs approaches had therapeutic effects in animal models of cancer and autoimmune diseases; nevertheless, the development of new drugs for patients needs further study for an effective development of clinical compounds that are able to target NETs.

Molecular Mechanisms of Neutrophil Extracellular Trap (NETs) Degradation Although many studies have been exploring the mechanisms driving NETs formation, much less attention has been paid to the degradation and elimination of these structures. The NETs clearance and the effective removal of extracellular DNA, enzymatic proteins (neutrophil elastase, proteinase 3, myeloperoxidase) or histones are necessary to maintain tissue homeostasis, to prevent inflammation and to avoid the presentation of self-antigens. The persistence and overabundance of DNA fibers in the circulation and tissues may have dramatic consequences for a host leading to the development of various systemic and local damage. NETs are cleaved by a concerted action of extracellular and secreted deoxyribonucleases (DNases) followed by intracellular degradation by macrophages. NETs accumulation depends on the ability of DNase I and DNAse II to hydrolyze DNA. Furthermore, the macrophages actively engulf NETs and this event is facilitated by the preprocessing of NETs by DNase I. The purpose of this review is to present and discuss the current knowledge about the mechanisms of NETs degradation and its role in the pathogenesis of thrombosis, autoimmune diseases, cancer and severe infections, as well as to discuss the possibilities for potential therapeutic interventions. Several anti-NETs approaches had therapeutic effects in animal models of cancer and autoimmune diseases; nevertheless, the development of new drugs for patients needs further study for an effective development of clinical compounds that are able to target NETs. Neutrophil extracellular traps (NETs) are web-like structures built from chromatin fibers decorated with antimicrobial enzymes and histones, serving as a trap, immobilizing and killing microorganisms, and therefore limiting their spread [1]. These structures, crucial for the proper functioning of the immune system, were described in 2004 by Arturo Zychlinsky and his group [1]. The clinical relevance and high impact of this phenomenon on thousands of physiological and pathological processes confirm that this is a Nobel prize-worthy discovery. As we previously described in detail, NETs release by neutrophils occurs primarily through a cell death mechanism termed NETosis [2]. This process begins with neutrophil activation by microbial products including endotoxins, with the support of platelets, proinflammatory cytokines or other danger signals. Several molecules bind to neutrophil receptors (such as Toll-like receptors (TLRs) and complement receptors) to activate neutrophils and trigger NETosis. Various factors of microbial origin, such as the size of microorganism and ROS, are regulators of NETosis—further reviewed in [2,3]. A next step is the activation of nicotinamide-adenine-dinucleotide-phosphate (NADPH) oxidase and intracellular granular proteases, followed by histones citrullination and chromatin decondensation [1]. Citrullination is mediated by the enzyme peptidylarginine deiminase 4 (PAD4), which removes positive charges from core histones by converting arginine residues to citrullines, thereby weakening the interaction between histones and DNA [1,2]. Next, the nuclear envelope of neutrophils breaks, and decondensed nuclear chromatin is expulsed, mixing with cytoplasmic and granule components, including neutrophil elastase, myeloperoxidase, cathelicidin antimicrobial peptide (LL37), high mobility group protein B1 (HMGPB1) and cathepsin G and proteinase 3 [1,2]. Subsequently, the cell membrane permeabilizes, NETs are released out of the cell and neutrophil dies in the process of NETosis [1,2,3]. An alternative mechanism maintaining structural integrity of granulocyte, referred as non-suicidal NETosis, starts from the blebbing of the nuclear envelope and fast exportation of microvesicles containing NETs outside of the cell in the absence of nuclear destruction. This is a fast reaction of the infiltrating granulocytes attracted to the sites of infection [1,2]. The neutrophils stay alive and retain the ability to combat bacteria after expulsion of their DNA. As a third mechanism, NETs can be generated from mitochondrial DNA, previously described by our group in [2,3]. In this process, mitochondria move to the cell surface and expel NETs. NETs are important contributors to the neutrophil antimicrobial response in tissues and vessels. NETs release and their degradation by DNases must be tightly regulated to prevent excessive inflammatory reactions [1,2]. Both the overproduction and the defects of the NETs clearance were found to promote numerous pathologies [2,3]. Neutrophils die at inflamed tissues undergoing netosis or one of the various other cell death mechanisms as apoptosis, necrosis, necroptosis, pyroptosis or autophagy [1,2,3,4]. The interplay of these processes is necessary for combatting foreign invaders and accounts for a further resolution of the inflammation [2,3]. All cell death mechanisms contribute to the regulation of neutrophils number, but also guaranties the degradation of their cargo and regulate the production of pro- and anti-inflammatory mediators [2,5]. The dysregulation of neutrophil death occurs in various pathological conditions such as sepsis and ARDS [3,5]. It is assumed that netosis, apoptosis and autophagy sharply rely on NADPH oxidase function and ROS production [2,3]. The redox disbalance within neutrophil likely accelerates the induction of the death machinery [2]. Following a stimulus, the generation of ROS is a prerequisite for autophagic, apoptotic or netotic processes and can be activated as a consequence of elevated ROS levels [2,3,4,5]. Moreover, stimulation of neutrophils with Toll-like receptor (TLR) agonists, PMA or phagocytosis of microorganisms activates netotic machinery or other type of cell death [1,2,3]. Further study is needed to better understand the molecular mechanisms regulating neutrophil death decisions. In particular, efforts should be made to gain insight into the starting points and upstream events of neutrophil netosis, autophagy, apoptosis or simply necrosis. NETs not only play a key role as a host defense mechanism against local and systemic infections, but if overproduced and persistent, exacerbate acute and chronic infectious diseases and participate in a variety of non-infectious conditions, all reviewed in [1,2,3,4,5]. NETs formation has been linked to an extraordinarily broad range of biological events. Netting neutrophils have the capacity to actively participate in multiple cellular and molecular cascades by releasing the cargo of mediators, including histones, metalloproteinases, cytokines, free DNA, proteases and ROS [1,2]. The NETs contribute to an overactivation of immune cells, the generation of thrombi in the circulation, endothelial and epithelial cells damage, vascular and bronchial occlusion, local tissue destruction, amplification of the vicious circle of the inflammatory response, etc., all processes are discussed in detail in our previous reviews [3,4,5,6]. The NETs interact with dendritic cells and macrophages, which in turn release interleukin 1 β (IL-1β) and interferon α (IFNα). NETs can also activate T-cell to release IFNα and IFNγ, deeply discussed in [3]. DNA decorated with histones and proteases, by disturbing homeostasis of the immune system, is involved in the pathogenesis of various inflammatory diseases such as psoriasis, rheumatoid arthritis, granulomatosis with polyangiitis, systemic lupus erythematosus (SLE), preeclampsia, cystic fibrosis, chronic otitis media, atherosclerosis, stroke, pancreatitis or severe COVID-19 [4,6,7]. NETs are also implicated in various non-inflammatory pathological processes, such as coagulation disorders, cancer, diabetes and wound healing [4,6,8]. As soon as the very first publications describing the NETs appeared, it was recognized that this structure is a potent source of various autoantigens which may induce autoimmune reaction and contribute to the development of autoimmune diseases [2,9]. Moreover, NETs components may act as damage-associated molecular patterns (DAMPs), and opposite DAMPs are able to induce NETs formation, generating a vicious circle of inflammation exaggerating organ damage and causing remote organ injury in the course of chronic inflammatory processes [10]. NETs have been implicated in numerous autoimmune disorders, including both systemic and local diseases, which may affect different organs (kidneys, joints, skin, blood vessels, lungs, central and peripheral nervous system) [2,6,9]. The accumulation of NETs and its components in the circulation correlates with the formation of anti-double-stranded DNA (dsDNA), anti-nucleosomes and anti-histones antibodies being considered a pathogenic factor for SLE [11]. The immune complexes built from these materials and immunoglobulins may depose in the glomeruli and cause lupus nephritis (LN). NETs are also engaged in the pathological processes in anti-neutrophil cytoplasmic antibodies (ANCA)—associated vasculitis, psoriasis and gout [2,9]. Elevated levels of circulating NETs markers were observed in multiple sclerosis [6]. In addition, elevated NETs components were found in peripheral blood, synovial fluid, rheumatoid nodules and skin of rheumatoid arthritis patients, and the NETs markers were positively associated with the concentration of anti-citrullinated protein antibodies (ACPA) [9]. Moreover, a majority of monoclonal antibodies found in synovial fluid and serum from rheumatoid arthritis patients reacts with citrullinated proteins (histones H2A/H2B, fibrinogen and vimentin) [11]. The citrullination of various proteins is a prominent feature of rheumatoid arthritis but, concomitantly, it plays an important role in the process of NETs formation [12]. The single nucleotide polymorphism in the gene encoding a protein tyrosine phosphatase (PTPN22) at position 1858 resulting in a missense mutation that converts an arginine a tryptophan was strongly associated with rheumatoid arthritis and excessive citrullination [13]. Chang et al. confirmed that the modification of C1858T disrupted the interaction between PTPN22 PAD4, followed by enhanced citrullination and exuberant NETs formation [13]. As described above, the citrullination of histones by PAD4 and the activation of the Raf-MEK-ERK signaling pathway have been described as necessary for their respective effects on histone degradation and expression of antiapoptotic pathways, subsequently leading to the release of decondensed chromatin DNA [1,2,13]. All components of NETs (DNA, histones and proteases) display procoagulant properties in the vascular compartment and in the surrounding tissues [5,8]. NETs promote venous, arterial and microvessels thrombosis by activating platelet adhesion and aggregation, providing a physical scaffold for thrombus formation from platelets and fibrin and being a trap for erythrocytes, all occluding the capillaries [8]. Remnants of NETs (dsDNA, myeloperoxidase-DNA complexes and citrullinated histones) activate coagulation cascade by increasing the protease activity of coagulation factors including thrombin [14]. NETs—derived dsDNA—directly activate the extrinsic pathway of coagulation, while NETs remnants promote thrombosis by the induction of tissue factor release from activated platelets and monocytes to initiate the intrinsic pathway as described in detail in [8]. Histones impair the function of coagulation inhibitors including thrombomodulin, thus promoting thrombin generation [15]. NETs are also required for the propagation of thrombi by binding and activating factor XII [5,8]. Neutrophil elastase promote coagulation by inactivating tissue factor pathway inhibitors, thus further increasing coagulation and fibrin deposition in vivo [8]. NETs aggregates can also occlude other tubular structures such as the bile and pancreatic ducts, provoking alterations of organ function and inflammation known as neutrophil extracellular trap-driven occlusive diseases [16]. Neutrophils and their products, including NETs, strongly contribute to acute lung injury, multi-organ damage and mortality in COVID-19, as reviewed by Szturmowicz and Demkow [5]. The markers of NETs formation, such as circulating DNA, nucleosomes, citrullinated histones, neutrophil elastase activity or myeloperoxidase-DNA complexes were found in sera of COVID-19 patients at a higher level as compared to healthy donors [5,16]. Moreover, the concentration of those markers significantly decreased in the recovery phase of COVID-19 [5,16]. Endothelial and pulmonary alveoli epithelial cell injury, as well as the disruption of alveolar-capillary barrier, a hallmark of severe pulmonary COVID-19, have been reported to be caused by NETs and their components [5,17]. The DNA threads form large conglomerates causing local obstruction of the small bronchi, and together with neutrophil elastase, are responsible for the overproduction of mucus by goblet cells of surface epithelia [5]. An excess of NETs promote the production of proinflammatory cytokines in SARS-CoV-2 pulmonary disease, leading to cytokine storm and, in consequence, to diffuse alveolar damage [5]. Dysregulated NETs formation in severe COVID-19 is responsible for the immunothrombosis of poor prognostic significance [5]. Zuo et al. found a strong correlation between neutrophil-activation markers/NETs and D-dimer (fibrin degradation product) in patients with thrombotic complications of COVID-19 [8]. The above-mentioned discoveries point to the fact that NETs are key pathogenic mechanisms in COVID-19 [8]. Of note are the findings that NETs production is associated with various other disseminated infections including sepsis. NETs are an important structure preventing the dissemination of microorganisms [2,5]. On the other hand, overproduction and persistence of NETs may activate an immune response that is destructive to the host tissues [2,5]. In the course of sepsis, NETs production is also triggered, by various pro-inflammatory mediators and activated cells: platelets, endothelial cells, tumor necrosis fact alpha (TNF-α), interleukin-8 (IL-8), nitric oxide and various autoantibodies [5]. NETs components, in particular histones, DNA fibers and antimicrobial proteins significantly contribute to lethality in sepsis [5]. All these associations between NETs and sepsis have been described in detail by Gierlikowska and Demkow [3]. NETs emerged as important players in contributing to tumor growth and metastasis formation—all these processes are described in detail by Demkow [4]. NETs have the ability to modulate the evasion capacities of the tumor cells [4,18]. To summarize, NETs awaken dormant cancer cells, promote cancer cell extravasation, enhance proliferation and migration of cancer and regulate the tumor microenvironment by degrading the extracellular matrix through the secretion of proteases providing a niche for metastatic tumor [4]. Moreover, NETs initiate the mesenchymal transition of the epithelial cells and potentiate migratory and invasive abilities of cancer cells. Circulating tumor cells, when entrapped by NETs fibers, can be sequestered and brought to distant organs forming lymphatic or hematogenous metastases [4,18]. Among the NET-driven tumorigenic activities, NETs directly affect the characteristics of tumor cells through activating signals, thus enhancing the invasiveness of cancer cells [4]. Furthermore, as mentioned above, NETs fuel cancer-associated thrombosis. Finally, NETs surround the primary tumor forming a barrier blocking the access of cytotoxic T cells and natural killer cells, thereby facilitating immune escape from the immunosurveillance [4]. The latter effect is not opposing the previously described mechanisms as it is responsible for the development of an immunosuppressive microenvironment fueling tumor growth, thus allowing progression and metastasis. It has been widely recognized that NETs can contribute to the pathogenic mechanism of various diseases affecting the central nervous system, such as ischemic stroke or systemic sclerosis as currently described by Manda-Handzlik and Demkow [6]. Ischemic stroke is usually caused by local thrombosis in the brain circulation or migration of peripheral clot responsible for vascular occlusion blocking the oxygen supply of the brain. NETs further promote secondary thrombosis, extending the period of ischemia. It is also postulated that the no-reflow phenomenon, impairing t-PA-induced thrombolysis, may be attributed to the NETs conglomerates entrapping platelets and activating intrinsic coagulation pathway in the brain capillaries [6,19,20]. The NETs clearance is necessary to maintain the correct balance between NETs formation and degradation [2,4]. The effective removal of extracellular DNA is crucial for tissue homeostasis, the prevention of inflammation and to avoid the presentation of auto-antigens [2]. Although many researchers have been exploring the process of NETs generation and pathophysiology, the knowledge on their degradation and the restitution of NETs-injured tissues is scarce [4,5]. Haider et al. suggested that NETs are cleaved by a concerted action of extracellular and secreted DNases followed by intracellular degradation by macrophages [21]. The cleavage with DNases plays a major role among physiological processes maintaining a low concentration of circulating free DNA. As DNA is the main component of NETs, DNases emerged as fundamental enzymes that breakdown NETs in vivo [22]. The extracellular DNases hydrolyzing circulating DNA comprise of the two families: DNase I, DNase II, exhibiting slightly different biochemical properties but partially redundant roles. DNases hydrolyze phosphodiester bonds of DNA molecules. The primary evolutionary role of DNases is suggested to degrade bacterial DNA [23]. The DNase I family consists of four members: DNase I, DNase1L1, DNase1L2 and DNase1L3, while the DNase II family includes DNase II α, DNase II β and L-DNase II [24]. The ability to hydrolyze DNA is common for both families. DNases are expressed across multiple tissues [24]. The degradation of DNA by DNase1 and DNases1L3 is the rate-limiting factor for NETs accumulation. DNase1 and DNase1L3 cleare NETs in blood vessels in the course of sepsis or sterile neutrophilia [24]. All except one are encoded by DNase I and DNase II, while the putative gene coding L-DNase II is SERPINB1 [24]. DNase I, mainly produced by the pancreas and kidneys, is the major nuclease present in the blood and other body fluids that cleaves extracellular dsDNA into fragments with 30-hydroxy and 50-phospho ends [24]. The structure and sequence of the DNA substrate affects the kinetics of hydrolysis—DNase I cleaves double-stranded DNA (dsDNA) 100–500 times faster than single-stranded DNA (ssDNA) [24]. DNase II digest phosphodiester backbone of DNA resulting in the formation of two fragments with 30-phospho and 50-hydroxy ends. This enzyme has the highest activity in the absence of divalent cations and at acid pH. DNase II resides in lysosomes of various cells including macrophages, and in multiple tissues, pointing to the role of this enzyme in the hydrolysis of phagocytosed fragments of exogenous DNA, mainly derived from apoptotic cells [25]. Nagata and coworkers have confirmed that DNAses play an important role during apoptosis and its deficiency activates innate immune response [26]. The same group found that DNase II-deficient mice develop polyarthritis attributable to an overproduction of TNF-α by macrophages accumulating undigested DNA [27]. Conversely, Ferrera et al. did not observe excessive accumulation of NETs-derived DNA in macrophages nor TNF-α release as a result of DNase II silencing. Moreover, these authors claim that DNase II plays a role in the detection of NETs-derived DNA in cells costimulated via TLRs [28]. The evidence has accumulated that, apart from DNase family, there are other enzymes dismantling the NETs structure such as 3′-exonucleases (TREX1 and TREX2) [24]. TREX1 cleaves DNA fragments remaining in the course of DNA replication, apoptosis, netosis, DNA repair and recombination pathways. The 3′ to 5′ exonucleases-dependent DNA fragmentation results in the release of DNA 3′ termini necessary for downstream events critical for DNA repair or replication, i.e., the excision of modified, mismatched, fragmented, damaged or even normal nucleotides [24]. What is more, the 3′ to 5′ proofreading of DNA synthesis represent the most important mechanism securing genome stability. If 3′ exonuclease activity fails, the cell cycle defects, genome instability and enhanced radiation sensitivity results in mutagenic DNA changes promoting cancerogenesis [29,30]. A relevant role of TREX family in the process of NETs degradation is related to its potential to destroy oxidized DNA which is resistant to DNAses I and II. As oxidative stress is an important mechanism in the process of NETs formation, the oxidized form of DNA is largely present and exposed in NETs. Furthermore TREX1, activates the cGAS–STING intracellular pathway through a BAK/BAX-dependent process, leading to misbalance in type I interferon synthesis and immune dysregulation/autoimmunity [22,31]. The clinical consequences of TREX1 deficiency was described by Morita et al. in TREX1 null mice [30]. The knock-out mice presented interferon-dependent autoimmune response resulting in inflammatory myocarditis progressing into dilated cardiomyopathy with fatal consequences [30]. Mutations in the gene encoding TREX1 were further associated with common and rare autoimmune and inflammatory conditions [31]. The intracellular degradation of NETs by macrophages is also dependent on TREX1 function [32]. It was also demonstrated that dendritic cells degrade NETs using DNase1L3. In the light of these observations, it can be assumed that a concerted action of all mechanisms involved and extracellular DNA degradation is necessary to maintain the homeostasis of the immune system [32]. The effects of TREX1 and TREX2 are clearly distinguishable. TREX2 supports the genome integrity of keratinocytes playing a role in DNA damage removal and degradation of removed fragments [29]. Recent evidence strongly supports the opinion that TREX2 complex is involved in the transcription processes and nuclear messenger RNA transport in mammalian cells [33]. The work of Farrera et al. suggests that DNase I in physiological concentrations is not sufficient to completely degrade NETs, pointing to an additional mechanism necessary for the decomposition of this structure. These authors highlight a prominent role of macrophages in NETs degradation [28]. Macrophages and neutrophils are important cells of an innate immune response and act in cooperation. The interaction between polymorphonuclear cells (PMN) and macrophages has been suggested as a crucial mechanism modulating inflammation in the course of many pathological conditions [34]. Macrophages helps eliminate damaged cells and debris in their microenvironment. Moreover, macrophages scavenge foreign invaders or apoptotic/necrotic cells protecting the organism from potential danger signals [35]. They are strategically located and have the ability to uptake and process infectious agents and many other particles [34,35]. The macropinocytosis and endocytosis of DNA fragments assure a counterbalance of NETs generation and degradation, necessary for the maintenance of proper homeostasis [34,35]. Macrophages, being key modulators of extracellular DNA degradation, phagocytose NETs elements without giving rise to an inflammatory reaction, however, if preactivated with microbial products such as LPS, they secrete proinflammatory cytokines such as IL-1β, interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) and begin effective antigen presentation [36]. Proinflammatory stimulation of macrophages and dendritic cells prime these cells for enhanced uptake and breakdown of NETs. [32,37]. The same directionality of effect was observed by Farrera et al. who confirms that macrophages are capable of the efficient clearance of NETs by taking up the extracellular DNA [28]. This process is facilitated by the extracellular digestion of large fragments of NETs by DNase I secreted by macrophages, as well as by the opsonization of NETs with complement factor 1q (C1q) [28]. The blocking of macropinocytosis in mice bearing a thrombus led to prolonged resolution of the clot; moreover, the NETs amount inside the thrombus was increased [28]. Preprocessing of NETs by DNase I and/or opsonization with C1q facilitated their clearance by macrophages in a cytochalasin D-dependent manner [28]. These authors have also shown that transfection of NETs or NET DNA inside macrophages stimulates the production of interferons, whereas the normal uptake of NETs by macrophages is immunologically silent, i.e., is not inducing the production of any mediators of the immune response [28]. The experiments with chloroquine proved that NETs undergo degradation in lysosomes, however the involvement of other cell compartments is not excluded [28]. Haider et al. aimed to determine the capacity of macrophages to degrade NETs and to identify the mechanism of endocytosis pathway as well as to investigate whether polarization of macrophages may change the kinetics of uptake and degradation [21]. These authors also provided evidence that local macrophage density in tissue sample from human aortic aneurysm is inversely associated with the presence of NETs in the tissue [21]. Haider et al. claim that an effective degradation of naked DNA or oligonucleotides, as well as the NET-degrading ability of polarized macrophages activated by proinflammatory stimuli (LPS + IFN-γ) is augmented. Long-term polarization by LPS and IFN-γ increased amounts of DNase1L3 and DNase 2 in macrophages [21]. These results are consistent with a previous observation by Farrera et al. showing that unpolarized macrophages using DNase1 were able to cut NETs into smaller fragments [28]. These authors already demonstrated that the knockdown of DNase 2 does not inhibit NETs degradation by unpolarized macrophages and that cytoplasmic TREX1 activity is needed for effective NETs breakdown by macrophages. [28] Haider et al. also claim that degradation of NETs by both unpolarized and polarized macrophages, as well as by their respective conditioned media, is abrogated by an inhibitor of DNase activity—EDTA [21]. Haider et al. further characterized the repertoire of DNases in human unpolarized and polarized macrophages and showed that the main secreted form of deoxyribonuclease in macrophages is DNase 1L3 [21]. Another discovery of this group led to a conclusion that DNase IL-1 intensively localized in the filopodia of activated macrophages. This might, at least partially, explain an increase in NETs degradation ability of proinflammatory macrophages [21]. Moreover, Haider and coworkers suggest that the intensive degradation of intracellular DNA is only partly due to the enhanced production of DNase. Another complementary process required for the effective clearance of NETs in the thrombi is macropinocytosis [21]. The experiments on murine thrombosis model using Sytox green-labeled DNA provided further evidence that preventing micropinocytosis by imipramine (a selective inhibitor of micropinocytosis) increased the presence of NETs components in the thrombi found in vasculature and concomitantly decreased fibrinolysis, supporting the statement that macropinocytosis is an important mechanism playing a role in the uptake of NETs by macrophages, both in vitro as well as in vivo [21]. The inhibition of macropinocytosis by imipramine resulted in longer and wider thrombi with increased NETs content [21]. The blockade of phagocytosis through inhibiting actin polymerization or phagosome-lysosome fusion also reduced NETs breakdown [21]. Previous observations suggested a potential mechanism of an activation of macropinocytosis dependent on the stimulation of Toll-like receptors in the macrophages [38]. In addition, macropinocytosis has been described to be altered in differently polarized macrophages [39]. Additionally, Haider et al. continued their experiments on human model using samples from aortic aneurysm patients who underwent surgery [21]. Of note was the finding that NETs are involved in the pathogenesis of aneurysms (inflammation, infiltration with macrophages, destruction of vascular wall and formation of trombus) and are thus prominently found in human arterial tissue [21]. Haider et al. showed that local macrophage density in human aortic aneurysms was negatively associated with surrounding NETs in the intraluminal thrombi as well as in the vessel wall [21]. Li et al. further explored the role of macrophages in NETs degradation process using the model of hepatocellular carcinoma (HCC) [40]. They found that diabetes-induced NETosis boosted HCC invasion in a NETs DNA-dependent manner. They confirmed that deficient DNASE1L3 expression in tumor tissues is a key cause responsible for the impairment of NETs DNA removal [40]. The resulting accumulation of NETs cause DNA-primed HCC cells to invade by activating the cGAS-ncNF-κB signaling pathway [40]. These observations were further confirmed by Wang et al. who showed that expression of DNASE1L3 is very low in HCC tissues, which may create a NETs DNA-rich microenvironment, thereby promoting cancer invasion and/or metastasis [41]. All the above-mentioned mechanisms of NETs degradation are presented in Figure 1. The effective clearance of NETs prevents overactivation of the immune system with concomitant thrombosis [2,5]. The inefficient dismantling of NETs may potentially serve as a source of immunogens derived from these structures, i.e., DNA, histones, enzymes and other NETs components [6,11,12,13]. Recently, different studies highlighted the link between NETs clearance defects and clinically relevant autoimmune disorders, especially SLE and vasculitis [6,22]. Overall, DNase activity is required to prevent the spontaneous formation of intravascular thrombi containing NETs [14]. All mechanisms involved in NETs degradation, as described above, can be impaired. Low DNase activity and functional impairment can be caused by the generation of anti-DNase inhibitors (and/or anti-DNAse autoantibodies) or mutations occurring in DNAses genes [42]. Genetic mutations affecting DNASE1, DNASE2, DNASE1IL3 and TREX were described [24]. DNase1 and DNase1-like 3 are independently expressed and thus provide dual host protection against the deleterious effects of intravascular NETs [24]. In vivo studies using DNASE-knocked-out mice confirmed the direct correlation between DNase activity and autoimmune diseases [43]. Knockout mice lacking both deoxyribonucleases rapidly died from multiorgan failure due to rapid occlusion of blood vessels with NETs containing clots [43]. In patients with severe bacterial infections, vascular occlusions were invoked by a defect in NETs removal ex vivo manifested as the formation of intravascular NETs—bearing thrombi [7]. Another mechanism that may lead to DNase functional impairment is the presence of circulating DNase inhibitors or the generation of anti-DNase antibodies [42]. A decade ago, Hakkim et al. first focused on the central role of DNase I for disassembling NETs, and then correlated the functional defects of DNase I with the impaired degradation of NETs in a subset of patients with SLE [22]. They further showed that, in selected patients named as ‘non-degraders’, a balance between NETs production and degradation was restored by the sera of healthy donors or discarding antibodies from SLE patients serum [22]. In the light of these observations, the authors hypothesized that the presence of anti-DNase I antibodies or DNases I inhibitors in the sera of SLE patients is responsible for the disease flares and kidney involvement [22]. A strong association between the reduction of DNases activity and the accumulation of NETs in autoimmune conditions was reported [22]. Insufficient production of DNase I (mutations occurring in DNase1 and DNase1L3 genes) or a decrease of its activity (DNase inhibitors or the generation of anti-DNase antibodies preventing the enzyme access to NETs) result in an inefficient degradation of free-circulating DNA and could determine the production of anti-nuclear autoantibodies (ANA) associated with SLE and LN being both a biomarker and a pathogenic factor contributing to the development of this condition [22,44]. The inverse correlation between circulating DNase1L3 and the formation of antichromatin and anti-dsDNA antibodies, with clinically relevant SLE-like disease and renal involvement, was also confirmed in animal studies [45]. DNASE1L3-deficient mice develop a typical lupus syndrome and have been widely used to support a direct implication of DNASE 1L3 in SLE/LN [45]. Yasumoto et al. presented two cases of patients with SLE and autoimmune glomerulonephritis bearing stop codon mutations in exon 2 of DNASE1 [46]. The patients with genetic deletion of DNase I had high levels of anti-DNA antibodies and low levels of circulating DNase I, as well as IgG and (complement factor 3) C3 glomerular deposition [42]. In LN, the removal of DNA, and consequently of NETs, may be impaired for different reasons, including key actionable mutations in genes encoding the DNases [42,45]. A second mechanism that may lead to DNase functional impairment is the presence of DNase inhibitors in the sera of patients with low DNase activity [45], or the generation of anti-DNase antibodies [42]. The loss-of-function mutations in genes encoding nucleases is considered as an important mechanism determining the development of autoimmunity [42]. DNase I-knocked out mice presented with typical symptoms of SLE, including presence of ANA, aggregation of immune complexes in kidneys, development of glomerulonephritis and further death [44,47]. Congruently, a causal relationship in human studies between mutations in DNAse I are linked to SLE, and a direct correlation between low activity of DNase I and SLE is confirmed [48]. Low DNase I activity is implicated in multiple systemic and organ-specific autoimmune diseases including thyroid autoimmunity, Sjogren’s syndrome and severe inflammatory bowel diseases [49]. It has been appreciated that low DNase activity is both a biomarker and a pathogenic factor in SLE [24]. Hakkim et al. discovered that impaired ability to clear NETs by SLE patients may account for the pathogenesis of LN [22]. Both mechanisms were implicated: the presence of anti-NETs antibodies and DNase1 inhibitors. Impairment of DNase1 function and failure to dismantle NETs are correlated with kidney involvement [22]. The same directionality of effect was observed by Bruschi et al. who tested NETs profiles in SLE patients and discovered that circulating NETs markers increased in 216 SLE patients, half of which had incident LN [50]. These authors found a significant correlation between high NETs marker levels, high anti-dsDNA antibody levels or low C3 activity and the presence of LN associated with either high anti-dsDNA antibody-circulating levels or low C3 activity. DNase activity was found to be selectively decreased in patients with LN compared to patients with SLE without kidney involvement and to the healthy controls, despite similar serum levels of DNASE I [50]. More recently, Hartl et al. provided evidence for the direct implication of anti-DNase antibodies in the pathogenesis of SLE in humans complicated by different organ pathologies [51]. They have also explored the mechanism of this association discovering that IgG autoantibodies to DNase 1L3 (but not to DNAse I) in serum are responsible for a decrease in enzyme activity in 50% of patients with LN as compared to patients with uncomplicated SLE or healthy controls [51]. In LN, DNase1L3 activity was also lower in patients with active proteinuria compared to those in remission. In accordance with the fact that DNASE 1L3 mutations are rare and could not account for the diminished DNase1L3 activity in 50% of the patients, an autoimmune mechanism was suggested [51]. These scientists tested the ability of autoantibodies to DNase 1L3 to lower the activity of the enzyme and found that the high and specific binding of IgG to DNase 1L3 in the serum of patients with LN correlated with diseases activity [51]. Consistently, no binding to DNase I was observed [51]. Overall, the findings by Hartl et al. support the statement that anti-DNase 1L3 antibodies are responsible for the inhibition of this enzyme activity in patients with LN [51]. The TREX1 disease-causing alterations include mutations and SNPs, and cause varied TREX1 dysfunction that might play a previously unanticipated role explaining the multiple clinical symptoms resulting from persistent oxidized DNA, as mentioned above, leading to enhanced type I interferon synthesis and immune dysregulation [22,31]. This mechanism links TREX1 deficiency with persistent NETs—dependent inflammation and autoimmunity [31]. Loss of function mutations in TREX1, both inherited and de novo, cause a spectrum of nucleic acid-mediated immune activation disease, including Aicardi–Goutieres syndrome, familial chilblain lupus and retinal vasculopathy with cerebral leukodystrophy and SLE [31,52]. These genetic discoveries have established a causal relationship between TREX1 mutation and autoimmune diseases [53]. Overall, deletions or mutations of any DNASEs, although rare or ultrarare, are always associated with a chronic inflammatory condition accompanied by the autoimmune glomerulonephritis [42,54,55]. Leffler et al. described three children with homozygous mutations in DNASE2 associated with a decreased degradation of NETs [55]. All three patients had similar clinical phenotype: membranoproliferative glomerulonephritis, fibrosing hepatitis and recurrent fever [55]. None of the patients fulfilled the clinical criteria of SLE and the serum levels of anti-DNA antibodies were variable [55]. All cases were compatible with an IFN-mediated inflammatory disease that also characterized SLE [55]. The pediatric onset of monogenic familial SLE with glomerulonephritis and very high anti-dsDNA antibodies is evoked by mutations of DNASEIL3 [55]. Additionally, these conditions can be manifested as urticarial vasculitis syndrome and hypocomplementemia, further progressing to severe SLE [56]. As another example of polymorphic changes in DNASE1L3 (rs35677470), it was linked to the family of autoimmune connective tissue diseases such as scleroderma, SLE or rheumatoid arthritis [57]. All these antoimmune conditions are present with functional defects of NETs degradation. Persistent NETs start a cascade of adaptive immune responses and complement activation and the deposition of NET-specific autoantibodies, creating a vicious circle of failed degradation and immune stimulation directly implicated in the pathogenesis of SLE [42,54,55,56]. Similarly, the presence of anti-DNase antibodies produced in the response to persistent NETs was described to be associated with microscopic polyangiitis (MPA) [58,59]. MPA patients had decreased DNase I activity in sera. Both IgG depletion from myeloperoxidase-ANCA (MPO-ANCA)-associated MPA sera and the supplementation of DNase I synergistically restored NET degradation [59]. This review focuses on the NETs—degrading mechanisms, suggesting a new way to design novel therapeutics for the management of a diverse set of NETs-dependent indications. The above-mentioned observations support the statement that the digestion of extracellular nucleoproteins may have a significant potential for the prevention and treatment of PMN-mediated disorders, including autoimmune diseases, exaggerated inflammatory reactions, severe infections and cancer. Further investigations on the inhibition of NETosis pathway as well as NETs degrading drugs provide potential therapeutic avenues for autoimmune diseases, especially SLE. An interesting option is also the combination of classical and anti-NETs intervention. A recent review by Mutua and Gershwin summarized the current knowledge on potential anti-NETs therapeutics [60]. Certain widely applied SLE therapeutics, such as tacrolimus, cyclosporine A and chloroquine, are targeting NETs components or interfering with mechanisms of NETs formation [61]. With the recent advances in the knowledge of how to inhibit or degrade NETs, several approaches to develop strategies to NET-targeting can be considered. DNAses are the most important enzymes dismantling NETs DNA. Gupta and Kaplan demonstrated that the administration of DNase 1 diminished SLE activity in mice [62]. They showed that TAK-242, a TLR4 inhibitor, decreased NETs formation, suggesting a therapeutic effect on autoimmune diseases [62]. In addition, PF1355, an inhibitor of MPO, limits the progression of autoimmune vasculitis in mice [62]. The modulation of either the NET production or the DNA removal appears as two possible effective strategies in SLE/LN treatment, and a balance of the two approaches may produce a synergy. On the other hand, blocking NET production may fail and, in some cases, may negatively impact the general clinical status and severe infectious complications. Blocking NET production is still an experimental area of investigation and further studies are warranted to explore this therapeutic option [50]. According to Pagnoux et al., the increase of DNase due to removing or blocking the synthesis of the circulating autoantibodies decreases the concentrations of circulating chromatin in SLE patients and propose plasmapheresis to decrease autoantibody levels [63]. Therapeutic plasma exchange has been widely used in many autoimmune disorders; however, further studies are needed to confirm its efficacy in NETs-dependent conditions [63]. On the basis of the reviewed studies, we may suggest that the blockade or the selective depletion of anti-DNase autoantibodies, or other strategies aimed at reducing NETs formation, could create a potential therapeutic option to prevent the progression of SLE and LN. Novel approaches to correct NETs-related tissue damage focused on the use of a recombinant human DNase-1 (dornase alpha—mucolytic agent applied in cystic fibrosis). De Buhr et al. observed the ability of DNase to degrade NETs in the lungs of calves infected with bovine respiratory syncytial virus [64]. Park et al. confirmed the effectiveness of DNase-1 coated nanospheres as modulators of NETs-associated complication of severe infection in mice [65]. Consistent observations were noted in SARS-CoV-2 patients [66]. The experimental use of DNase -1 coated melanine-like nanospheres on the plasma of COVID19 patients resulted in the significant reduction of NETs and MPO activity, as well as the decrease of the cytokines IL-1β, IL-6 and TNFα, involved in NETs vicious circle [66]. Nevertheless, the NETs remnants may be responsible for the development of bacterial superinfection in COVID-19 patients [5]. Thus potential benefits of DNase containing products in SARS-CoV-2 infection have to be confirmed by further studies. The other naturally occurring molecule, reducing pathological NETs activity is alpha-1-antitripsin (AAT), a neutrophil elastase inhibitor [67]. AAT binds extracellular IL-8, reducing the neutrophils’ influx to the inflammatory site and augments neutrophil superoxide production, inhibiting the activity of neutrophil elastase [67]. The other beneficial effects of AAT depend on the inhibition of endothelial cells apoptosis and thrombin generation [67]. These properties of the drug may be important in reducing the NETosis and immunothrombosis in the course of SARS-CoV-2 infection. Moreover, AAT expresses the natural anti-SARS-CoV-2 activity as inhibitors of S-protein cleavage [5]. During acute-phase reaction, especially in the course of severe infections, circulating AAT levels increase [67]. Moreover, the individuals diagnosed with AAT deficiency were more prone to the development of uncontrolled infections. Vianello et al., while looking for predictors of severe SARS-CoV-2 disease in Italian population, proved the geographic co-localization of AAT deficiency and SARS-CoV-2 infections [68]. Thus, it is possible that the patients with severe SARS-CoV-2 disease could benefit from therapeutic AAT administration [67]. It is also demonstrated that the direct inhibition of the process of NETosis can prevent COVID-19 exacerbation. As another example, recombinant DNases may play a very important role as a potential drug in monogenic SLE. It is also demonstrated that DNase I digesting the NETs can destruct the scaffold of clot formation, suggesting the potential therapeutic role of the enzyme in the development of NETs-dependent thrombosis [23]. Gupta and Kaplan observed that calcineurin inhibitors blocking calcium mobilization required for NETosis (cyclosporine A and tacrolimus) are effective medications for SLE patients [62]. Furthermore, N-acetyl cysteine (NAC), a potent ROS scavenger, confers inhibiting effects of NETs extrusion because of its sharp reliance on ROS production, while exerting therapeutic effects in autoimmune diseases. NAC was effective in the therapy of SLE patients as confirmed by two clinical studies [62]. The evidence has also accumulated that Mito TEMPO, a specific inhibitor of ROS production, hindered NETosis and concomitantly decreased activity of SLE in mice. Moreover, the pharmacological inhibition of PAD activity attenuated the clinical course and reduced organ damage in the mice model of SLE and RA [62]. Similarly, again on the mice model, the inhibition of NET formation by Cl-amidine inhibited arterial thrombosis and diminished vascular damage [69]. Furumoto et al. described an inhibitory effect of tofacitinib on NETs, combined with an amelioration of vascular damage in the course of murine lupus [70]. Consistent with the antidiabetic drug metformin, inhibiting the NETs DNA-pDC-IFNα pathway reduced the risk of SLE exacerbations and corticosteroid dose in SLE patients [71]. Furthermore, Handono et al. observed the protective effect of vitamin D3 on NETs-dependent endothelial damage in SLE patients by blocking the externalization of neutrophil elastase [72]. Finally, anti-NETs therapy is believed to prevent the awakening of dormant cancer cells to inhibit the spreading of tumors as well as the formation of metastases [4]. The present review highlights complex interactions between the generation and degradation of NETs. Focusing on NETs degradation mechanisms may provide novel insights into the therapy of cancer, severe infections including COVID-19, or autoimmune diseases and many others [73]. The overproduction of NETs confirmed by high levels of circulating NETs markers or the presence of NETs in tissue samples may stand for the identification of patients who could benefit from NET-targeting therapy [74]. NETs degrading drugs may supplement other therapeutic regimes applied to prevent or treat cancer, autoimmunity and immunothrombosis. As mentioned above, numerous researchers have developed promising concepts on anti-NETs strategy [4,5,6]. The potential benefits of destroying NETs in vivo encourage further research. Several anti-NETs approaches had therapeutic effects on animal models of cancer and autoimmune diseases; nevertheless, the development of new drugs for patients needs further study and more time necessary for the effective development of clinical compounds able to target NETs [4,5,75,76]. Both options, either to dismantle formed NETs, or to block their production, require further study to enable clinicians to be more confident to use those drugs. Such strategies and underlying molecular mechanisms are at the preliminary phase and further data to explore their therapeutic potential and potential severe side-effects are highly anticipated. On the other hand, the risk of systemic infections in NETs-depleted patients may limit clinical applications in anti-NETs therapy and further study is warranted to investigate this issue. Targeting NETs is a worthwhile strategy in contemporary medicine that can be envisioned thanks to the ground-breaking discovery of Brinkmann et al. [1].

PMC10002922

Beatriz Linillos-Pradillo,Lisa Rancan,Sergio D. Paredes,Margret Schlumpf,Walter Lichtensteiger,Elena Vara,Jesús Á. F. Tresguerres

Low Dose of BPA Induces Liver Injury through Oxidative Stress, Inflammation and Apoptosis in Long–Evans Lactating Rats and Its Perinatal Effect on Female PND6 Offspring

26-02-2023

bisphenol A,oxidative stress,inflammation,apoptosis,liver injury,perinatal offspring

Bisphenol A (BPA) is a phenolic compound used in plastics elaboration for food protection or packaging. BPA-monomers can be released into the food chain, resulting in continuous and ubiquitous low-dose human exposure. This exposure during prenatal development is especially critical and could lead to alterations in ontogeny of tissues increasing the risk of developing diseases in adulthood. The aim was to evaluate whether BPA administration (0.036 mg/kg b.w./day and 3.42 mg/kg b.w./day) to pregnant rats could induce liver injury by generating oxidative stress, inflammation and apoptosis, and whether these effects may be observed in female postnatal day-6 (PND6) offspring. Antioxidant enzymes (CAT, SOD, GR, GPx and GST), glutathione system (GSH/GSSG) and lipid-DNA damage markers (MDA, LPO, NO, 8-OHdG) were measured using colorimetric methods. Inducers of oxidative stress (HO-1d, iNOS, eNOS), inflammation (IL-1β) and apoptosis (AIF, BAX, Bcl-2 and BCL-XL) were measured by qRT-PCR and Western blotting in liver of lactating dams and offspring. Hepatic serum markers and histology were performed. Low dose of BPA caused liver injury in lactating dams and had a perinatal effect in female PND6 offspring by increasing oxidative stress levels, triggering an inflammatory response and apoptosis pathways in the organ responsible for detoxification of this endocrine disruptor.

Low Dose of BPA Induces Liver Injury through Oxidative Stress, Inflammation and Apoptosis in Long–Evans Lactating Rats and Its Perinatal Effect on Female PND6 Offspring Bisphenol A (BPA) is a phenolic compound used in plastics elaboration for food protection or packaging. BPA-monomers can be released into the food chain, resulting in continuous and ubiquitous low-dose human exposure. This exposure during prenatal development is especially critical and could lead to alterations in ontogeny of tissues increasing the risk of developing diseases in adulthood. The aim was to evaluate whether BPA administration (0.036 mg/kg b.w./day and 3.42 mg/kg b.w./day) to pregnant rats could induce liver injury by generating oxidative stress, inflammation and apoptosis, and whether these effects may be observed in female postnatal day-6 (PND6) offspring. Antioxidant enzymes (CAT, SOD, GR, GPx and GST), glutathione system (GSH/GSSG) and lipid-DNA damage markers (MDA, LPO, NO, 8-OHdG) were measured using colorimetric methods. Inducers of oxidative stress (HO-1d, iNOS, eNOS), inflammation (IL-1β) and apoptosis (AIF, BAX, Bcl-2 and BCL-XL) were measured by qRT-PCR and Western blotting in liver of lactating dams and offspring. Hepatic serum markers and histology were performed. Low dose of BPA caused liver injury in lactating dams and had a perinatal effect in female PND6 offspring by increasing oxidative stress levels, triggering an inflammatory response and apoptosis pathways in the organ responsible for detoxification of this endocrine disruptor. Bisphenol A [BPA; 2,2-bis (4-hydroxyphenyl)] is a synthetic xenoestrogen compound with a high prevalence in our environment [1,2]. BPA is not hazardous in its polymeric form but is unstable in acidic and basic solutions and when exposed to ultraviolet light. These conditions can convert/transform polymeric BPA into monomeric forms [1]. It is used mainly in the food industry as a monomer in the manufacture of polycarbonate plastics and epoxy resins such as plastic food or beverage containers and in the coating of cans, protecting the contents from direct contact with the metal surface [3,4,5,6,7,8], but also for certain paper products. BPA residues can migrate into the food, beverages or environment due to high temperatures, causing people to inevitably be exposed to BPA in their daily lives [3,6,9,10]. The main source of human exposure is through ingestion [5,9,11], while transdermal absorption and inhalation would be possible through secondary routes of exposure [3,5,9]. BPA can act as an endocrine disruptor showing effects that are similar to those of estrogenic and thyroid hormones. Due to continuous exposure, it can cause health problems in humans, including endocrine, reproductive and metabolic effects, cardiovascular disorders and cancer, so that it has been considered a risk for public health [2,6,9,11]. BPA is absorbed from the small intestine and reaches the liver through the blood, this organ being responsible for its metabolism into its glucuronic acid-conjugated form. Therefore, there is a very real possibility of the presence of a higher concentration and toxicity of this compound in the liver [12]. BPA has also been observed to play a major role in inflammation; as Moon et al. [13] reported, it increases the expression of pro-inflammatory cytokines such as IL-6 and TNF-α. In addition, it also induces an increase in oxidative stress by decreasing antioxidant enzymes [1,3,4,5,7,8] and significantly compromises mitochondrial function [14]. BPA is also able to inhibit cytochrome P450 isoforms in the rat liver [15,16,17]. Other in vivo experimental studies have shown that exposure to BPA can also cause liver disease, including steatosis [18], liver tumors [19] and the metabolic syndrome [20]. In previous studies, BPA has been detected in the human placenta [21], umbilical cord blood [22], amniotic fluid [17,23], fetal liver [24] and breast milk [25] as well as in human serum and urine [26]. Hence, since BPA was found in the previously mentioned tissues, as well as at birth [27], exposure to this compound during prenatal life is probable. However, the effect of BPA on the offspring is still poorly understood. The aim of this study was to evaluate whether BPA administration during pregnancy is able to induce liver damage in lactating rats by affecting the oxidant/antioxidant balance through the induction of oxidative stress, increasing inflammation and triggering apoptosis. Moreover, it was studied whether this effect can also be observed in female offspring at postnatal day 6 (PND6). In this study, clinical observations were made daily and body weight of females was monitored every 3–4 days. Regarding the general appearance, the animals did not show any alteration that could be perceived visually or any unexpected behavior. During the entire experiment it was not necessary to sacrifice any animal for signs of cadence or signs of pain or aggression. Considering all rats from day 24 (before starting mating) at equal weight, an increase in body weight was observed in all females until the maximum weight was reached at the end of pregnancy. No significant differences were observed in the groups of females treated with different doses of BPA in the diet compared to the control group (Figure 1A). There were also no significant differences in food consumption between the control and BPA treatment groups, monitored during the second week of premating and the second week of pregnancy (Figure 1B). Regarding reproduction data, in the case of control females (n = 10), eight female rats were pregnant and two females were not pregnant. In dams treated with BPA, 0.036 mg/kg/b.w./day group (low-dose BPA) (n = 9), it resulted in eight pregnant females and only one non-pregnant female. Regarding dams treated with BPA, 3.42 mg/kg/b.w./day group (high-dose BPA) (n = 8), six females were pregnant and two remained non-pregnant. Therefore, pregnancy was achieved in 22 females from a total of 27 females. The highest percentage of pregnancy was observed in the BPA low dose group (88.8%) followed by the control group (80%) and the lowest percentage of pregnancy was seen in the high-dose BPA group (75%) (Figure 1C). Considering the total number of offspring, the highest number of offspring (106 pups) was obtained in the BPA low dose group followed by the control group (90 pups) and the lowest number of offspring (72 pups) was seen in the high-dose BPA group. However, the mortality rate after birth was higher in the BPA low-dose group (14.1%) compared to the control animals (13.3%) and the BPA high-dose group (8.3%), which had the lowest number of dead pups. Among the living animals, the number of female offspring was 45 in the control group, 46 in the BPA low-dose group and 31 in the high-dose BPA group. All these animals were included in the study. In addition, the offspring of both BPA treatments had lower body weights compared to the control PND6 offspring, while no significant differences were observed between both doses of BPA (Figure 1E). No significant differences were observed between BPA and control groups in the body weight of the females during the entire experiment. In all experimental groups, a constant weight gain was observed, reaching the maximum at the end of pregnancy, as expected (Figure 1A). Female rats exposed to low and high doses of BPA were compared in terms of antioxidant enzyme activities and glutathione concentrations; which are endogenous antioxidant defense systems to prevent cellular damage measured in the liver (Figure 2). When lactating females were treated with low-dose BPA, all antioxidant enzyme activities such as catalase (CAT), superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione reductase (GR) and glutathione S-transferase (GST) were significantly decreased compared to the control group (Figure 2A–E, respectively). In addition, a decrease in reduced glutathione (GSH) concentration (Figure 2F) and an increase in oxidized glutathione (GSSG) concentration (Figure 2G) were observed in low BPA dose-treated dams. When lactating dams were treated with the high dose of BPA, decreased activities of antioxidant enzymes CAT, GPx and GST were observed in comparison to the control group (Figure 2A,C,E, respectively). GSSG concentration was also increased compared to the control (Figure 2G), but no significant differences were observed in the GSH concentration (Figure 2F). The GSSG/GSH ratio, a marker of oxidative stress, was significantly increased in dams exposed to the low dose of BPA compared to the control group and significant differences were also observed between treatment groups, resulting in higher levels of oxidative stress in the low-dose BPA group (Figure 2H). Data concerning measurements of oxidative damage and the gene expression of antioxidant enzymes (GPx, GR, GST, γ-glutamylcysteine synthetase (γ-GCS)) were tested in the liver of dams (Figure 3). In the low-dose BPA dams, GPx, GR, GST and γGCS gene expressions were down-regulated versus control dams (Figure 3A–D). Lipid peroxidation is a metabolic process that causes oxidative deterioration of lipids by reactive oxygen species (ROS). This process can degrade lipids within the cell membrane leading to cell damage and eventual cell death. In lactating dams treated with the low dose of BPA, we observed an increase in malondialdehyde (MDA) and lipid hydroperoxide (LPO), two products generated under oxidative stress situations, used to measure oxidative lipid damage (Figure 3E,F, respectively). In dams treated with the low dose of BPA, an increase in 8-oxo-2′-deoxyguanosine (8-OHdG), one of the main DNA oxidation products, used as a biomarker of oxidative DNA damage, was also observed (Figure 3I). Furthermore, an increase in adenosine triphosphate (ATP) energetic levels was observed in the low dose of BPA dams compared to the control group (Figure 3G). Nitric oxide (NO) plays a dual role in oxidative and antioxidant behavior. As an antioxidant, NO protects cells from oxidative stress. However, when produced in excess, it behaves as an important pro-oxidant factor. In this case, an increase in plasma NO metabolites of dams treated with the low dose of BPA was observed (Figure 3H). In addition, when the two BPA doses were compared, significant differences were observed in LPO concentrations, ATP levels and NO metabolites, being significantly lower than the results of the group treated with the high dose of BPA when compared to the low dose (Figure 3F,G,H, respectively). Results obtained for gene and protein expression of oxidative stress-inducing proteins: Heme oxygenase-1 (HO-1d) and NOS isoforms: Inducible nitric oxide synthase (iNOS) and endothelial nitric oxide synthase (eNOS) in the livers of dams exposed to the two doses of BPA are shown in Figure 4. In dams treated with the low dose of BPA, an up-regulation in gene and protein expressions of HO-1d and iNOS compared to the control group was observed (Figure 4A–D). In dams treated with the high dose of BPA, a significant increase in iNOS gene and protein expressions was observed versus the control dams (Figure 4C,D, respectively). However, no differences in eNOS gene and protein expressions were observed among the groups (Figure 4E,F, respectively). When the two doses of BPA were compared, the only oxidative stress-inducing protein that showed significant differences among groups was HO-1d, which was significantly higher in the low dose of BPA group compared to the high dose one (Figure 4B). The results of gene and protein expressions of inflammatory markers such as interleukin-1-β (IL1β) and apoptosis markers: Apoptosis-inducing factor (AIF), Bcl-2-associated X protein (BAX), B-cell lymphoma (BCL-2) and B-cell lymphoma-extra large (BCL-XL) are shown in Figure 5. The proinflammatory cytokine IL1β showed a significant increase in gene and protein expressions in dams treated with the low dose of BPA as compared with the control group (Figure 5A,B, respectively). Regarding the proapoptotic molecules, AIF gene and protein expressions were up-regulated in dams treated with the low dose of BPA (Figure 5C,D, respectively) compared with the control group. BAX gene expression was up-regulated in the low dose of BPA treated dams (Figure 5E). In dams treated with high dose of BPA, an increase in AIF gene expression was observed compared to control (Figure 5C). Considering the antiapoptotic molecules, BCL-2 and BCL-XL protein expressions were down-regulated in dams treated with the low dose of BPA (Figure 5F,G, respectively). When the two doses of BPA were compared, a significant difference was found in the protein expression of the proapoptotic molecule AIF, the low dose of BPA group being significantly higher than the high-dose group (Figure 5D). In addition, the protein expression of BCL-XL, an antiapoptotic molecule, was significantly lower in the low dose of BPA group compared to the high dose one (Figure 5G). Representative protein blots for each tested marker are shown in Figure 5H. Antioxidant enzyme activities and glutathione concentrations were determined in the livers of female PND6 pups to determine the effect of perinatal exposure to low and high doses of BPA (Figure 6). When PND6 offspring were perinatally exposed to low dose of BPA, all antioxidant enzyme activities (CAT, SOD, GPx, GR and GST) were significantly decreased compared to the control group (Figure 6A–E). In addition, a decrease in reduced glutathione (GSH) concentration (Figure 6F) and an increase in oxidized glutathione (GSSG) concentration (Figure 6G) were observed in the low dose of BPA offspring. These same effects were observed in lactating dams exposed to the low dose of BPA (Figure 2). When PND6 offspring were perinatally exposed to the high dose of BPA, decreased activities of antioxidant enzymes SOD and GST were observed in comparison to the control group (Figure 6B,E, respectively). GSH concentration decreased in comparison to the control group whereas no significant changes were observed in GSSG concentration (Figure 6G). As observed in lactating dams, GSSG/GSH ratio increased in offspring exposed to low dose of BPA as compared to the control group (Figure 6H). When antioxidant enzyme activities and glutathione concentrations were compared between treated groups, a significant increase in GPx was observed in the high dose of BPA group compared to the low dose one (Figure 6C). On the contrary, an imbalance between GSSG and GSH levels was observed in the low dose of BPA group, resulting in a higher ratio as a marker of oxidative stress compared to the high dose one (Figure 6H). The transcriptional levels of antioxidant enzymes (GPx, GR, GST, γGCS) and markers of oxidative damage in the liver of female PND6 offspring are shown in Figure 7. GPx, GR, GST and γGCS gene expressions were down-regulated in low-dose-PND6 offspring versus control offspring (Figure 7A–D). In PND6 offspring perinatally exposed to a low dose of BPA, an increase in MDA, LPO and 8-OHdG content compared to the control group was observed (Figure 7E,F,I, respectively). ATP energy levels increased in low-dose-BPA offspring compared to the control group (Figure 7G). These results suggest that perinatal exposure to low doses of BPA increased oxidative damage of lipids and DNA in offspring, as it was observed in dams exposed to low doses of BPA (Figure 3E,F,G,I). Furthermore, an increase in plasma NO metabolites of low-dose-BPA offspring was observed (Figure 7H). In offspring exposed to high doses of BPA, an increase in 8-OHdG was observed, showing oxidative DNA damage (Figure 7I). When the two BPA doses were compared, significant differences were observed in GST gene expression, which was significantly higher in the high-dose-PND6 group compared to the low dose one (Figure 7C). In addition, LPO concentrations and ATP levels showed significant differences between groups, the high-dose-PND6 group being significantly lower compared to the low dose one (Figure 7F,G, respectively). These results are similar to those observed in dams, where significant differences were also observed between treated groups the high dose of BPA group being the one that showed significantly lower levels (Figure 3F,G, respectively). In PND6 offspring exposed to low doses of BPA, an up-regulation in gene and protein expressions of HO-1d and iNOS compared to the control group was observed (Figure 8A–D). Regarding HO-1d gene expression and iNOS protein expression, significant differences were also observed between treatment groups, where the low-dose-PND6 group showed significantly higher expressions than the high dose group (Figure 8A,D, respectively). However, no differences in eNOS gene and protein expressions were observed among groups (Figure 8E,F, respectively); these results were similar to those observed in dams exposed to low doses of BPA (Figure 4). The results of gene and protein expressions of inflammatory markers IL1β and apoptosis markers AIF, BAX, BCL-2 and BCL-XL are shown in Figure 9. The proinflammatory cytokine IL1β showed a significant increase in gene and protein expressions in offspring treated with low doses of BPA as compared to the control group (Figure 9A,B, respectively). Regarding the proapoptotic molecules, AIF gene and protein expressions were up-regulated in offspring exposed to low doses of BPA compared to the control group (Figure 9C,D, respectively). When AIF gene and protein expressions were analyzed between treated groups, significant differences were found. In both gene and protein expressions, the low-dose-PND6 offspring showed significantly higher values than the high-dose group, whose results were not different from the control group (Figure 9C,D, respectively). BAX gene expression was up-regulated in low dose of BPA offspring (Figure 9E). The anti-apoptotic markers BCL-2 and BCL-XL significantly reduced their protein expression in low-dose-BPA offspring versus the control group (Figure 9F,G, respectively). This imbalance between pro-apoptotic and anti-apoptotic family members shown in low-dose-BPA pups was also observed in lactating dams (Figure 5F,G, respectively). Representative protein blots for each tested marker are shown in Figure 9H. Hematoxylin and eosin staining was used to analyze the effect of BPA on liver injury; this is shown in Figure 10. In the livers of dams exposed to BPA, no changes were observed in cellular structure compared to control hepatocyte images (Figure 10A). However, histological staining showed that BPA administration increased nucleus aggregation and infiltration of inflammatory cells in PND6 offspring liver tissue compared to control pups. In addition, lower concentration of BPA had a noteworthy impact on liver injury compared to higher doses in PND6 offspring (Figure 10B). Results of hepatic serum marker assessment indicated that dams treated with lose-dose BPA exhibited liver injury manifested by a significant rise in the levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALP) when compared to the control (Figure 10C,D, respectively). The serum levels of gamma glutamyl transpeptidase (GGT) did not show any significant change in animals receiving BPA in comparison with the control group (Figure 10E). Bisphenol A is one of the most widely used industrial chemicals worldwide. Trasande et al. [28] reported that 90% of the general population has detectable levels of BPA. These BPA levels are 70 times higher in occupationally exposed individuals than in environmentally exposed populations [29]; therefore, BPA exposure is considered an unavoidable and concerning situation. Since BPA exposure occurs mainly through ingestion, in the present study the effects of BPA administered by the oral route at two different concentrations were evaluated: a low dose of 0.036 mg/kg/b.w./day and an almost 100-fold higher dose of 3.42 mg/kg/b.w./day, in the liver of pregnant dams and their perinatal effect in PND6 offspring. The liver is the main organ responsible for the metabolism of BPA through conjugation by the liver enzyme uridyl diphosphate glucuronyl transferase (UDPGT) to a less toxic compound called bisphenol A-glucuronide (BPAG) [30,31]; this being the main pathway for the detoxification of this xenobiotic. A smaller amount of BPA reacts with sulfate giving rise to BPA-sulfate (BPAS) [32] or can be oxidized to a catechol followed by further transformation to an O-quinone (4,5-bisphenol-O-quinone) [33]. The catechol-O-quinone couple is capable of redox cycling with generation of ROS and oxidative stress [14]. ROS are cytotoxic agents that cause oxidative damage by attacking the cell membrane but also DNA. The liver has an endogenous antioxidant defense system to prevent cellular damage such as antioxidant enzymes and the glutathione system. The activities of antioxidant enzymes CAT, SOD, GPx and GR were decreased in the livers of dams treated with low doses of BPA compared to the control group. There was also a decrease in the enzymatic activities of CAT and GPx in dams treated with high doses of BPA, without any significant difference with the low doses of BPA group. SOD generally dismutates the superoxide anion radical into hydrogen peroxide, which is degraded by CAT using GSH. The reduction in CAT activity may reflect the inability to remove H2O2 produced after BPA exposure [15,34]. GST protects the cell by conjugating glutathione (GSH) to electrophilic substrates, generating less reactive and more soluble compounds, being a detoxifying enzyme involved in the metabolism of many xenobiotics [35]. Exposure to both doses of BPA showed a significant reduction in GST activity, reflecting an inability to detoxify this compound. Exposure to low BPA dose also reflects lower gene expression levels of the γGCS that catalyzes the first step in glutathione synthesis, resulting in low cellular levels of glutathione. Glutathione provides a first line of defense against ROS as it can scavenge free radicals and reduce H2O2 formation in the cell. BPA produces several quinols and semiquinone intermediates that can react with glutathione producing glutathione conjugates, which, in turn, increase oxidative stress levels [7]. GPx utilizes reduced glutathione (GSH) to remove peroxides produced by oxidative stress [36]. On the other hand, GR reduces oxidized glutathione (GSSG) back to GSH using NADPH [7]. In the present study, the GSH depletion shown in the dams treated with low BPA dose along with NADPH oxidation and altered redox homeostasis seems to play an important role in the disruption of antioxidant defense, leading to elevated levels of oxidative stress in liver cells. Oxidative stress produces free radicals that can easily react with cell membrane lipids, proteins and nucleic acids, thus initiating a chain of reactions leading to the production of lipid peroxides [37] and DNA damage [38]. In our results, a significant increase in MDA levels was observed with low doses of BPA and an increase in LPO with both doses of BPA in dams, in this case being higher in the low dose of BPA group compared to the high dose one. In turn, exposure to low doses of BPA led to an increase in oxidative damage to DNA as shown by the increased values of 8-OHdG in the livers of pregnant dams. We observed that exposure to low BPA dose induced liver damage in rats, affecting the oxidant/antioxidant balance and causing liver injury. Our results are in agreement with many others, such as Acaroz et al. [3] who demonstrated decreased SOD and CAT enzymatic activities and GSH levels in Wistar albino rats exposed to BPA at different oral doses (5, 10 and 20 mg/kg). In another study using a dose of 25 mg/kg in rats for 50 days, an increase in MDA levels and a decrease in GSH levels and SOD and CAT activities in kidney, brain and testis tissues was found [39]. Bindhumol et al. [15] also showed a reduction of antioxidant enzymes (SOD, CAT, GR, GPx) in the mitochondrial and microsome-rich fractions of the liver; while H2O2 and MDA levels increased in Wistar rats treated with BPA doses ranging from 0.2 to 20 µg/kg. The same occurred in the study by Hassan et al. [40] where antioxidant activities were decreased at doses of 50 mg/kg of BPA in rat livers. To investigate the involvement of BPA in cellular oxidative stress, eNOS, iNOS and HO-1d were tested as mediators of this process. Vascular function mainly depends on the balance between synthesis/degradation of nitric oxide (NO). NO produced by eNOS is a result of a physiological response that plays an important role in mediating many processes such as vasodilation, immunity and neurotransmission. In our results, we observed no difference in eNOS gene and protein expression in both treatment groups. However, elevated plasma NO levels and higher gene and protein expression of iNOS were observed in dams treated with low BPA dose compared with the control group and high BPA dose. An increase in the synthesis of NO produced by iNOS causes vascular dysfunction and its iNOS activation may have some detrimental effects for liver function. NO is a potent oxidant and a nitrating agent capable of attacking and modifying proteins, lipids and DNA, as well as decreasing antioxidant defenses [41]. Regarding heme oxygenase (HO), which participates in the metabolism of the heme group of hemoproteins, two isoforms have been characterized: one inducible (HO-1d) and one constitutive (HO-2d). The inducible isoform, HO-1d, is expressed under various stimuli, such as oxidative stress and cytokines such as TNFα; being a reliable marker of a proinflammatory and prooxidant state [42]. Our results showed an increase in gene and protein expression of HO-1d on low BPA dose treatment in dams. Its induction following increased oxidative stress could act as a cellular defense mechanism to prevent progression of liver fibrosis. Kazemi et al. [1] showed an increase in HO-1d gene expression with a BPA dose-dependent profile in liver cells. High levels of oxidative stress have been linked to inflammatory processes. In this study, dams treated with low BPA dose increased gene and protein levels of the proinflammatory cytokine (IL-1β). In accordance with our results, Acaroz et al. [3] showed that BPA exposure at 25 mg/kg in male Wistar rats increased the expression of proinflammatory cytokines such as TNF-α, IL-6 and IL-1β and decreased anti-inflammatory/antifibrotic cytokine (IL-10). Elswefy et al. [8] administered 50 mg/kg of BPA to rats orally for eight weeks and reported that its administration significantly increased the serum level of IL-1β and reduced the level of IL-10. This increase in proinflammatory cytokines induced liver inflammation by transporting mononuclear and polymorphonuclear leukocytes to inflamed tissues [43]. In our study, no structural changes of hepatocytes were noticeable yet after BPA administration in the liver of the dams. However, in the histological study by Kazemi et al. [44] it was demonstrated that oral administration of BPA by gavage at low doses induced liver injury in male adult rats. Liver tissue damage can be assessed by serum liver markers. In our study, a marked increase in AST and ALT was observed with the low BPA dose, indicating tissue damage in the liver. This is consistent with the study by Ijaz et al. [45] where a substantial increase in the levels of alanine aminotransferase (ALT), alkaline phosphatase (ALP) and aspartate aminotransferase (AST) was also observed in BPA-treated rats. This may be because overproduction of ROS damages the structural integrity of liver cells, which is manifested by an increase in hepatic serum markers [46]. However, it is not yet apparent in histological sections because it is at an early stage of involvement after seven weeks of BPA exposure. Previous studies found that BPA impairs hepatic mitochondrial function by releasing soluble factors into the cytosol [13,47]. This membrane permeabilization may be the initial stage of mitochondrial apoptosis [6]. One of the proapoptotic markers is AIF, which upon release into the cytosol, translocates to the nucleus where it triggers apoptotic pathways. In our study, elevated gene levels of AIF were found in the liver upon exposure to BPA. We also studied the mRNA expression of BAX, another factor that promotes apoptosis, which showed significantly elevated gene expression levels in dams treated with low BPA dose, whereas protein expression of Bcl-2 and BCL-XL, which are anti-apoptotic factors that protect the cell from various cytotoxic alterations, were found to be significantly decreased with the low dose of BPA treatment. Previous studies also showed increased pro-apoptotic protein caspase-3 and reduced anti-apoptotic protein BCL in the liver of male rats [6,8]. BPA weakened hepatocyte mitochondrial function and promoted cell apoptosis in the liver by up-regulating protein levels of Bax, cleaved caspase-3 and cleaved PARP1, while it down-regulated Bcl-2 in the liver using high doses of BPA [48]. Notably, cytochrome c, a key mediator of apoptosis through activation of caspases in the cytosol [49,50,51], was also found to be increased. Low BPA dose treatment showed elevated ATP levels in pregnant dams compared to the control group. This maintenance of sufficient ATP levels together with the release of pro-apoptotic factors causes liver cells to enter apoptosis [52]. The mechanism of BPA-induced apoptosis probably also involves an alteration in the expression ratio of pro-apoptotic and anti-apoptotic proteins of the BCL-2-associated X family (BAX) and BCL-2 in the outer mitochondrial membrane that modulates the release of proapoptotic factors [53,54]. Therefore, exposure of pregnant dams to low doses of BPA may exert toxic effects on liver cells through the formation of ROS, induction of inflammation and apoptosis. At high doses of BPA these effects are not as noticeable or significant in many of the parameters studied compared to the control group. This may be because there is a higher level of vulnerability in the liver towards low doses of BPA compared to other organs due to the initial metabolism of BPA by the liver [12,13,55]. BPA is considered a xenoestrogen, but not an estrogen mimic [56] due to its ability to bind to the classical nuclear estrogen receptors (ER) ERα and ERβ [57]; although compared to 17β-estradiol the affinity is about 10,000 times lower for ERα and 1000 times weaker than the affinity for ERβ [58]. It is also able to bind to classical and non-classical membrane estrogen receptors [59], as well as to the G protein-coupled receptor 30 (GPR30) [60], and act through non-genomic pathways [61] and also as an activator of the thyroid hormone and androgen receptors [59,62]. This may explain that BPA, as an endocrine disruptor, such as some hormones, can follow non-monotonic dose–response curves (NMDR), showing more noticeable effects at low doses than at high doses [63]. The endocrine system is configured to respond to very low concentrations of hormones and a maximal biological response can be observed without a high receptor occupancy of this response. This could be due to the fact that response mechanisms become saturated before all receptors are occupied. This is consistent with a previous study that observed a non-monotonic relationship in pregnant Wistar rats exposed to BPA (50, 250 or 1250 μg/kg) and their offspring after weaning. Only the lowest dose of 50 μg/kg of BPA produced effects such as increased body weight, elevated serum insulin and impaired glucose tolerance in adult pups. However, this study exposed rats to normal or high-fat diets, which could also play a role in the response mechanisms. Rats exposed perinatally to the higher doses showed none of the adverse effects, regardless of diet [20]. Most scientific studies have focused on the effect of high doses of BPA in adults, but the effect of low BPA dose on perinatal exposure seems to be more important to take into consideration [64]. Exposure of pregnant dams to BPA is of concern to the developing fetus since it is able to cross the placenta and enter into cord blood and amniotic fluid. This is in addition to the presence of little or no fetal enzymatic activity at all of UDPGT to biotransform it into inactive BPAG [17]. Furthermore, the enzyme β-glucuronidase is highly active in the placenta and can further contribute to increase fetal exposure to free BPA by hydrolysis of conjugated BPA entering the fetal compartment [20,65]. BPA also binds to the estrogen-related receptor gamma (ERɣ), which is highly expressed in the placenta, facilitating the accumulation of BPA and thus increasing the exposure of the developing fetus to this compound causing potential harmful effects to the offspring at very low and sustained doses. A recent in vitro study showed that activation of the P2X7 receptor after incubation with BPA has been observed in human placental cells, leading to different pathways involved in producing preeclampsia and preterm delivery, through activation of the NLRP3 inflammasome and apoptosis [66]. Nishikawa et al. [67] showed that the presence of free BPA in the liver of fetal rats could be the result of direct transfer of free BPA into the maternal circulation via the placenta, in addition to the hydrolysis of BPAG in the fetal liver. In the present study, we observed a decrease in the activity levels of antioxidant enzymes CAT, SOD, GPx, GR and GST in PND6 pups exposed to the low dose of BPA. SOD and GST activities were also decreased in the offspring with the high dose of BPA. GSH was reduced in the offspring exposed to the low dose of BPA, with increased levels of oxidized GSH. In addition to decreased antioxidant enzyme activity, lipid peroxidation-associated damage (increased levels of MDA and LPO) was increased in the liver of offspring exposed to low dose, along with increased DNA oxidation in offspring exposed to both doses of BPA. This is consistent with a study in pregnant mice orally exposed to a dose of 100 ng/g BPA from PND7 to PND21, showing that perinatal BPA exposure could induce oxidative damage and alter normal metabolic profiles in the liver [68]. Lin et al. [69] showed that perinatal BPA exposure causes the development of non-alcoholic fatty liver disease (NAFLD) in the offspring of pregnant Sprague-Dawley rats that had access to water containing 1 or 10 μg/mL BPA from gestational day six (GD6) to PND21. BPA exposure is associated with up-regulation of lipogenic genes, dysregulation of autophagy and activation of the inflammatory response involving PI3K/Akt/mTOR and TLR4/NF-κB pathways. This oxidant/antioxidant imbalance also became noticeable here as gene and protein expression levels of oxidative stress-inducing proteins (HO-1d and iNOS) were increased in the offspring exposed to the low dose of BPA, along with elevated plasma NO levels. Increased proinflammatory cytokine IL-1β and proapoptotic factors AIF and BAX, with the subsequent decrease in antiapoptotic factors BCL-2 and BCL-XL, led to an induction of apoptosis in liver cells in the offspring perinatally exposed to the low dose of BPA. In our study, higher aggregation of nucleus and infiltration of inflammatory cells were observed in the liver of PND6 offspring treated with low dose BPA as compared to the high dose one. Santoro et al. [70] showed that the main histological alteration of the liver was a mild to moderate microvesicular steatosis in BPA-treated rats at 10–17 PND and 45–60 PND. Mild hepatocellular hypertrophy was observed in some BPA-exposed lactating or weaned animals. Furthermore, the expression of inflammatory cytokines, Sirt1, its natural antisense long non-coding RNA (Sirt1-AS LncRNA), and histone deacetylase 1 (Hdac1) were affected in exposed animals. Another study has shown susceptibility to NAFLD in adulthood following mitochondrial dysregulation upon perinatal exposure [20]. Jiang et al. [18] showed that perinatal BPA exposure contributes to the development of hepatic steatosis in male offspring at 3, 15 and 26 weeks when postnatally treated with 40 µg/kg BPA, and that this was mediated by impaired hepatic mitochondrial function. Therefore, exposure to low levels of endocrine disrupting chemical (EDC) BPA—these levels being easier to achieve in daily life—is of concern since it interferes with many metabolic processes and causes widespread damage to body tissues. The fact that lower levels of BPA are generally more effective than the higher doses is a very remarkable issue as previously described. Moreover, it should also be noted that the timing of BPA exposure may determine the long-term outcome, as earlier exposure points tend to exert a more severe effect [18]. Thus, fetuses and newborns are more sensitive than adults, and chemical exposure during critical developmental stages could cause irreversible long-term consequences [6,17,71]. In our study, similar effects were observed in perinatally exposed offspring as well as in their lactating dams after BPA exposure, being this a critical period influencing ontogenic development of various tissues and also increasing the risk of developing diseases later in adulthood. Further research is critical to understand the extent and effect of prenatal exposure to potentially toxic chemicals including BPA. Our study is subject to a series of technical limitations. First, since it is part of a bigger European Union’s Horizon 2020 Research and Innovation Programme project (ENDpoiNTs; grant number: 825759), at the time of extraction, livers had to be quickly divided and immediately frozen in liquid nitrogen. Therefore, the liver weight could not be measured. These data could have provided additional information regarding a possible hepatic injury. Another technical limitation regarding the analysis of liver functional enzymes is that this could only be performed with dams’ sera. In the case of PND6 offspring, the collected serum volume was insufficient to perform these chemical determinations. Hence, information about hepatic functionality in PND6 offspring is missing. Furthermore, for experimental design reasons, only female pups were included. It would have been interesting to compare the effects of the BPA administration also in male pups. Finally, an important limitation of the study is that, for technical limitations, BPA level determinations either in plasma or liver biopsies are missing. Furthermore, although special attention was paid to avoid any BPA contamination throughout the entire experiment, an interference of background BPA exposure with low-dose treatment may not be completely excluded. Twenty-seven female (eight weeks of age) and twelve male (ten weeks of age) Long–Evans rats (Janvier Labs, Le Genest-Saint-Isle, France) were used in the study. The animals were all housed and maintained in a well-ventilated room at 22 ± 2 °C, with automatic light cycles (12 h light/dark) and all had free access to diet and drinking water ad libitum. Rats were housed in special polypropylene cages (Sodispan Research, Coslada, Madrid, Spain) that were manufactured with the lowest chemical composition of Makrolon, a polycarbonate with bisphenol A. Water bottles were made of glass. Animals were randomly divided into three groups consisting of: (1) Control (non-treated) group—received chow with a corresponding concentration of corn oil (n = 10 females; n = 4 males); (2) bisphenol A (0.5 mg/kg chow) low-dose group—diet intake of 0.036 mg/kg body weight/day of BPA (n = 9 females; n = 4 males); and (3) bisphenol A (50 mg/kg chow) high-dose group—diet intake of 3.42 mg/kg body weight/day of BPA (n = 8 females; n = 4 males). High dose of BPA was chosen in the range of doses (2.5 mg/kg and 50 mg/kg) that consistently induced impairment learning and memory loss in rodents when administered in the perinatal period. While low dose was 100 times lower than that. The dose ingested by each rat was calculated based on the food intake data per animal which corresponded to 7.3% of body weight. BPA with purity >99% was purchased from Sigma Aldrich (Argovia, Switzerland) (CAS number 80-05-7; article number: 239658). It was dissolved in ethanol and then corn oil at a ratio of 10% ethanol and 90% corn oil. The chosen chow was purchased from Granovit (Argovia, Switzerland) corresponding to a diet of natural ingredients low in phytoestrogens (rather restricted concentrations so that the estrogenic effects were weak) (Granovit AG, Kaiseraugst, KLIBA NAFAG 3317.PX.L15). Rats were bred in special polypropylene cages (Sodispan Research, Coslada, Madrid, Spain) and glass drinking bottles were used to avoid the presence of substances that could also act as endocrine disruptors. A cylindrical environmental enrichment element was included. During the entire experiment (premating, mating, pregnancy, lactation), the control group cages were kept separate from the BPA-treated groups, to avoid any chance of spreading chow containing BPA. During premating, female and male rats were treated with the diet with their corresponding dose of BPA for two weeks. Control animals received the control diet. Mating phase took place between a male and a female from the same group, after checking that the female was in the estrus phase. The following morning, a check for sperm-positive vaginal smear or sperm-plug was carried out and the process was repeated all mornings for one week. Treatment was maintained during pregnancy. After birth, the lactating dams were kept in individual cages with their offspring and dietary treatment continued until PND6. During the entire period, body weight of females was recorded every 3–4 days and clinical observations were made daily. Furthermore, food consumption (FC) was calculated by weighing the food administered (FA) on the previous day and subtracting weight of food remaining (FR) (FC: FA-FR) during the second week of premating and second week of pregnancy. With 2 rats per cage to avoid isolation stress, we divided the total quantity consumed in the cage by 2 (FC: (FA-FR)/2). In addition to counting the number of pups on the day of birth, the number of dead pups was recorded and the sex ratio and pup weight on PND6 were identified. Lactating dams were sacrificed by decapitation using a guillotine. Female offspring were sacrificed at postnatal day 6 (PND6) by decapitation using scissors. The livers were collected and immediately frozen in liquid nitrogen and stored at −80 °C until analysis. Plasma samples were collected from the lactating dams and stored at −20 °C (Figure 11). Catalase (CAT), superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione reductase (GR) and glutathione-S-transferase (GST) activities were measured in the liver homogenate previously lysed with the corresponding buffer and analyzed spectrophotometrically according to the manufacturer’s instructions (Cayman Chemical; Ann Arbor, MI, USA). CAT activity was determined by the reaction with methanol in the presence of an optimal concentration of hydrogen peroxide (H2O2). The formaldehyde produced was measured spectrophotometrically with 4-amino-3-hydrazino-5-mercapto-1,2,4-triazole as the chromogen at 540 nm. SOD activity was assessed by measuring the dismutation of superoxide radicals generated by xanthine oxidase and hypoxanthine. The standard curve generated using this enzyme provides a means to accurately quantify the activity of all three types of SOD (Cu/Zn, Mn and FeSOD). GPx activity was measured spectrophotometrically; it was coupled to the oxidation of NADPH by GR. A GR assay kit was used to measure activity of this enzyme by quantifying the rate of NADPH oxidation. GST activity was determined spectrophotometrically by measuring formation of the conjugate of reduced glutathione (GSH) and 1-chloro-2,4-dinitrobenzene (CDNB) at 340 nm. Each sample was tested in triplicate. Enzyme activities were normalized according to liver protein content and were expressed as nmol/min/mg of protein except for SOD, which was expressed in U/mg of protein. Liver was homogenized in 50 mM phosphate buffer and 0.1 M EDTA, pH 8. Then, 10 µL of HClO4 was added per mL of homogenate and supernatants were used for the quantification of both reduced (GSH) and oxidized (GSSG) glutathione by o-phthalaldehyde (OPT) at pH 12 and pH 8, respectively, resulting in the formation of a fluorescent compound. Fluorescence was measured at 350 nm excitation and 420 nm emission. Results were expressed as nmol of GSSH and GSH per milligram of protein. Moreover, the GSSG/GSH ratio was calculated for each sample. Quantification of lipid peroxidation (LPO) was carried out in liver homogenate according to the manufacturer’s instructions (Cayman Chemical; Ann Arbor, MI, USA). Lipid Hydroperoxide Assay Kit measures the hydroperoxides directly utilizing the redox reactions with ferrous ions. The amount of lipid hydroperoxide was obtained from the linear regression of the standard curve substituting corrected absorbance values for each sample. LPO content was expressed as nmol/mg of tissue. This procedure eliminates any interference caused by hydrogen peroxide or endogenous ferric ions in the sample and provides a more sensitive and reliable assay for lipid peroxidation. Lipid peroxidation was also evaluated using a commercial kit (BioVision, Mountain View, CA, USA), which measures the reaction of malondialdehyde (MDA) with thiobarbituric acid (TBA) and the MDA-TBA adduct formation. Samples were resuspended in lysis buffer with the antioxidant butylated hydroxy-toluene (BHT) (0.1 mM) to prevent further formation of MDA during the preparation of the sample or during the heating step. Then, they were centrifuged at 3200× g for 30 min. Furthermore, 200 μL of supernatants from each sample were added to 600 μL TBA, and incubated at 95 °C for 60 min. Samples were cooled in ice for 10 min, and 300 μL of n-butanol were added (Sigma-Aldrich, Madrid, Spain) to create an organic phase in which the MDA molecules were to be placed. Samples were centrifuged and 200 μL of upper organic phase were collected and dispensed into a 96-well microplate for spectrophotometric measurement at 532 nm. Results were expressed as nmol TBARS/mg protein. The adenosine triphosphate (ATP) levels of liver tissue were determined using a colorimetric/fluorometric assay kit (Bio Vision, Milpitas, CA, USA) according to the manufacturer’s instructions. For the assay, 50 mg of liver tissue was used and the ATP content was calculated and expressed as nmol/mg of protein. Levels of nitric oxide metabolites (NOx) in plasma samples were measured by the Griess reaction as nitrite ion (NO2−) concentration after nitrate (NO3) reduction to NO2−. Briefly, after incubation of the plasma with Escherichia coli NO3 reductase and NADPH+ (37 °C, 30 min), 300 µL of Griess reagent (0.5% naphthylenediamine dihydrochloride, 5% sulfonilamide, 25% phosphoric acid (H3PO4)) (Sigma-Aldrich, Saint Louis, MO, USA) was added. The reaction was performed at 22 °C for 20 min, and the absorbance at 546 nm was measured, using sodium nitrite (NaNO2) solution as standard. Results were expressed as nmol/µL of plasma. Oxidative DNA damage was assessed by means of an ELISA kit consisting of a competitive assay for the quantitative measurement of 8-hydroxyguanosine (8-OHdG) (Cell Biolabs, Inc., San Diego, CA, USA). The unknown 8-OHG samples or 8-OHdG standards were first added to an 8-OHdG/BSA conjugate preabsorbed microplate. After a brief incubation, an anti-8-OHdG monoclonal antibody was added, followed by an HRP conjugated secondary antibody. The 8-OHdG content in unknown samples was determined by comparison with a predetermined 8-OHG standard curve. Finally, results were expressed as ng/mg DNA. The protein content of the same samples was evaluated following a bicinchoninic acid protein assay kit protocol (Sigma-Aldrich, Madrid, Spain) or by BCA Assay Pierce (Bio-Rad Laboratories, Hercules, CA, USA) using a BSA standard curve. mRNA expression of GPx, GR, GST, γ-glutamylcysteine synthetase (γGCS), heme oxygenase 1 (HO-1d), inducible nitric oxide synthase (iNOS), endothelial nitric oxide synthase (eNOS), interleukin-1-β (IL-1β), apoptosis-inducing factor (AIF) and Bcl-2-associated X protein (BAX) was measured using real time qRT-PCR. RNA was isolated from liver samples according to the method described by Chomczynski [72] using the TRI Reagent Kit (Molecular Research Center, Inc., Cincinnati, OH, USA) following the manufacturer’s protocol. The purity of the RNA was estimated by 1% agarose gel electrophoresis, and RNA concentrations and ratio 260/280 were determined by spectrophotometry BioDrop (Fisher scientific, Waltham, MA, USA). Reverse transcription of 2 mg of RNA for cDNA synthesis was performed using the StaRT Reverse Transcription Kit (AnyGenes, Paris, France). qRT-PCR was performed using a 7500 Fast Real Time PCR System thermal cycler (Applied Biosystems, Cambridge, MA, USA) with the TB Green Ex Taq (Tli RNase H Plus) (Takara Bio Inc., Shiga, Japan) and 300 nM concentrations of specific primers (Table 1). The qPCR amplification cycles were a 95 °C 10 min cycle, followed by 45 cycles at 95 °C for 10 s and at 60 °C for 30 s and finally melting curve analysis, following the recommendations of the manufacturer (95 °C for 10 s, 65 °C for 30 s and 95 °C for 0 s). Amplification of 18S mRNA was used as a loading control for each sample. The gene expression level was analyzed in triplicate for each sample. Relative changes in mRNA expression were calculated using the 2−∆∆CT method [73]. Western blotting was used to measure levels of HO-1d, iNOS, eNOS, IL-1β, AIF, Bcl-2 and B-cell lymphoma-extra large (BCL-XL). Briefly, liver samples, after homogenization with modified RIPA lysis buffer (PBS, Igepal, Sodium deoxycholate (D5670-5G), 10% SDS, PMSF, 0.5 M EDTA and 100 mM EGTA) to which protease inhibitor cocktail (#P-2714) (Sigma-Aldrich, Madrid, Spain), PMSF (#P7626, 1 mM), sodium orthovanadate (#S6506, 2 mM) and sodium pyrophosphate (#S6422, 20 mM) were added, were sonicated and boiled for 10 min at 100 °C in the ratio 1:1 with gel-loading buffer (100 mmol/L TrisHCl (pH 6.8), 4% SDS, 20% glycerol, bromophenol blue 0.1, 200 mmol/L dithiothreitol). Total protein equivalents (25 μg) for each sample were separated by SDS-PAGE by using 10% Mini-PROTEAN TGX Precast acrylamide gels (Bio-Rad Laboratories, Hercules, CA, USA) and were transferred onto a PVDF membrane using the Trans-Blot Turbo Transfer System (Bio-Rad Laboratories, CA, USA). The membrane was immediately placed into blocking buffer containing 5% non-fat milk in 20 mM Tris pH 7.5, 150 mM NaCl and 0.01% Tween-20. The blot was allowed to block at 37 °C for 1 h. The membrane was incubated with a rabbit polyclonal antibody (1:1000) (Table 2) for 12 h at 4 °C, followed by incubation with a goat anti-rabbit IgG secondary antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) (1:7000). Protein detection was performed using the Clarity Western ECL Substrate assay kit (Bio-Rad Laboratories, CA, USA) and ECL Plus (Amersham Life Science Inc., Buckinghamshire, UK) by chemiluminescence with the BioRad ChemiDoc MP Imaging System to determine the relative optical densities. Prestained protein markers were used for molecular weight determinations. Housekeeping gene GAPDH was used as loading control (1:5000) (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Proteins were quantified using BioRad Image Lab software v6.1 (Bio-Rad Laboratories, Hercules, CA, USA). Liver tissues were washed in 0.9% cold saline and fixed in a 10% formalin buffer solution for the histopathological assessment for 24 h. After fixation, samples were processed for embedding in paraffin. Serial sections (5 µm) were prepared using a rotary microtome Leica RM2125 RTS (Leica Biosystems, Wetzlar, Germany) for hematoxylin and eosin staining (H&E). The sections were stained with 0.1% hematoxylin (Ciba, Basel, Switzerland) for 5 min. Then slides were washed with tap water for 15 min and then a quick wash with hydrochloric alcohol (0.5% HCl in absolute ethanol) to remove excess staining on the sample (differentiation). The acid was neutralized by immersing the sections in tap water for 5 min and a final wash with distilled water. They were immersed in 0.1% eosin (Ciba, Basel, Switzerland) for 5 min. After washing with distilled water, tissue sections were dehydrated using ascending ethanol passages and finishing in xylol for 30 s. Tissue sections were cover slipped. Images were captured with Leica Microscope (Leica Biosystems, Wetzlar, Germany). Evaluation of aspartate aminotransferase (AST), alanine aminotransferase (ALP) and gamma glutamyl transpeptidase (GGT) levels in dams’ sera was determined by a veterinary laboratory (LAV Arturo Soria, Madrid, Spain) using a KONE sequential automatic autoanalyzer (Kemia Científica, Madrid, Spain). Differences between obtained values (mean ± SD) were assessed by one-way analysis of variance (ANOVA) followed by the Tukey–Kramer multiple comparison test or Bonferroni post-test to compare all pairs of means after testing for normal distribution. A confidence level of 95% (p < 0.05) was considered statistically significant. Statistics were calculated using Prism v7 (GraphPad Software Inc., La Jolla, CA, USA). Bisphenol A is a molecule capable of producing estrogenic effects and its continued exposure at low doses is unavoidable. It produces adverse effects on the body, including the liver, the main organ in charge of detoxifying the organism. In this study, it was observed that exposure of female Long–Evans rats to low doses of BPA during pregnancy and lactation increased the levels of oxidative stress in the liver, decreasing antioxidant activities and the glutathione system. This loss of homeostasis generated by the excessive accumulation of ROS and NOS caused an increase in inflammation, triggering cellular apoptosis pathways. The effect of perinatal BPA exposure on female offspring was also studied at PND6 and shows similar effects as found in the dams. Although alterations were observed at both doses of BPA, the maximum effect occurred with the low dose of BPA, resulting in an inverse dose–response relationship. We consider that this is especially important since in our everyday life we are constantly exposed to low doses of BPA, with it being present in many commonly used products.

PMC10002924

Lucrezia Principi,Erica Ferrini,Roberta Ciccimarra,Lisa Pagani,Clizia Chinello,Paolo Previtali,Andrew Smith,Gino Villetti,Matteo Zoboli,Francesca Ravanetti,Franco Fabio Stellari,Fulvio Magni,Isabella Piga

Proteomic Fingerprint of Lung Fibrosis Progression and Response to Therapy in Bleomycin-Induced Mouse Model

23-02-2023

lung fibrosis,bleomycin mouse model,proteomics,bottom-up mass spectrometry,nintedanib,Coronin-1A,lactate dehydrogenase B

Idiopathic pulmonary fibrosis (IPF) is a chronic lung disease characterized by the aberrant accumulation of extracellular matrix in the lungs. nintedanib is one of the two FDA-approved drugs for IPF treatment; however, the exact pathophysiological mechanisms of fibrosis progression and response to therapy are still poorly understood. In this work, the molecular fingerprint of fibrosis progression and response to nintedanib treatment have been investigated by mass spectrometry-based bottom-up proteomics in paraffin-embedded lung tissues from bleomycin-induced (BLM) pulmonary fibrosis mice. Our proteomics results unveiled that (i) samples clustered depending on the tissue fibrotic grade (mild, moderate, and severe) and not on the time course after BLM treatment; (ii) the dysregulation of different pathways involved in fibrosis progression such as the complement coagulation cascades, advanced glycation end products (AGEs) and their receptors (RAGEs) signaling, the extracellular matrix-receptor interaction, the regulation of actin cytoskeleton, and ribosomes; (iii) Coronin 1A (Coro1a) as the protein with the highest correlation when evaluating the progression of fibrosis, with an increased expression from mild to severe fibrosis; and (iv) a total of 10 differentially expressed proteins (padj-value ≤ 0.05 and Fold change ≤−1.5 or ≥1.5), whose abundance varied in the base of the severity of fibrosis (mild and moderate), were modulated by the antifibrotic treatment with nintedanib, reverting their trend. Notably, nintedanib significantly restored lactate dehydrogenase B (Ldhb) expression but not lactate dehydrogenase A (Ldha). Notwithstanding the need for further investigations to validate the roles of both Coro1a and Ldhb, our findings provide an extensive proteomic characterization with a strong relationship with histomorphometric measurements. These results unveil some biological processes in pulmonary fibrosis and drug-mediated fibrosis therapy.

Proteomic Fingerprint of Lung Fibrosis Progression and Response to Therapy in Bleomycin-Induced Mouse Model Idiopathic pulmonary fibrosis (IPF) is a chronic lung disease characterized by the aberrant accumulation of extracellular matrix in the lungs. nintedanib is one of the two FDA-approved drugs for IPF treatment; however, the exact pathophysiological mechanisms of fibrosis progression and response to therapy are still poorly understood. In this work, the molecular fingerprint of fibrosis progression and response to nintedanib treatment have been investigated by mass spectrometry-based bottom-up proteomics in paraffin-embedded lung tissues from bleomycin-induced (BLM) pulmonary fibrosis mice. Our proteomics results unveiled that (i) samples clustered depending on the tissue fibrotic grade (mild, moderate, and severe) and not on the time course after BLM treatment; (ii) the dysregulation of different pathways involved in fibrosis progression such as the complement coagulation cascades, advanced glycation end products (AGEs) and their receptors (RAGEs) signaling, the extracellular matrix-receptor interaction, the regulation of actin cytoskeleton, and ribosomes; (iii) Coronin 1A (Coro1a) as the protein with the highest correlation when evaluating the progression of fibrosis, with an increased expression from mild to severe fibrosis; and (iv) a total of 10 differentially expressed proteins (padj-value ≤ 0.05 and Fold change ≤−1.5 or ≥1.5), whose abundance varied in the base of the severity of fibrosis (mild and moderate), were modulated by the antifibrotic treatment with nintedanib, reverting their trend. Notably, nintedanib significantly restored lactate dehydrogenase B (Ldhb) expression but not lactate dehydrogenase A (Ldha). Notwithstanding the need for further investigations to validate the roles of both Coro1a and Ldhb, our findings provide an extensive proteomic characterization with a strong relationship with histomorphometric measurements. These results unveil some biological processes in pulmonary fibrosis and drug-mediated fibrosis therapy. Idiopathic pulmonary fibrosis (IPF) is a chronic and progressive disease with a median survival time of 2–3 years from diagnosis [1]. The fibrotic lung is characterized by fibroblastic foci as well as the accumulation of collagen in the extracellular matrix (ECM) space. All of these features lead to the degeneration of alveolar architecture and the decline of respiratory function [2]. Up to now, disease diagnosis and the evaluation of its progression have been based solely on clinical (e.g., level of dyspnea, pulmonary hypertension), radiological (High-Resolution Computed Tomography (HR-CT)), and functional variables (e.g., forced vital capacity) [3]. The key events responsible for the initiation of fibrosis occur as a result of an abnormal wound-healing response, driven by alveolar epithelial injury and subsequent secretion of inflammatory mediators (cytokines and chemokines), which leads to platelet activation and inflammatory cell migration. Pro-fibrotic cytokines, such as transforming growth factor β (TGF-β), released by inflammatory cells, play a central role in fibrogenesis, modulating the recruitment and activation of fibroblasts, promoting collagen synthesis and ECM deposition, and driving the differentiation of fibroblasts into myofibroblasts. Nevertheless, the mechanisms underpinning IPF lung tissue remodeling, fibrosis onset, and progression involve complex signaling pathways that are still largely unknown [4]. Currently, there is no cure for IPF, and nintedanib is one of the two antifibrotic drugs approved for its treatment, slowing down the progression of the disease and preserving lung function. The drug is a triple tyrosine kinase inhibitor (TKI) that targets the vascular endothelial growth factor receptor (VEGFR), the platelet-derived growth factor receptor (PDGFR), and the fibroblast growth factor receptor (FGFR) [5]. Further exploration of the molecular mechanisms of action of nintedanib (NINT) is needed and could pave the way to its future use in the treatment of progressive fibrotic lung diseases other than IPF. During the last decade, high-throughput proteomics technologies have taken huge leaps in order to shed light on the mechanisms of human disease as well as to clarify the action of drugs, and the understanding of IPF lung fibrosis has already started to gain benefits from these approaches, highlighting the overall landscape of ECM proteins [6] and the role of ER and oxidative stress [7,8] in lung tissue fibrosis. Recently, Zheng et al. applied an integrative multi-omics approach combining transcriptomics and proteomics analysis in order to identify novel potential biomarkers of human IPF. They found butyrophilin-like-9 (BTNL9) and plasmolipin (PLLP) genes being downregulated in IPF patients compared to controls and suggested that both genes might play protective roles in IPF, reducing immune response, inhibiting ECM production, and enhancing endothelium regeneration [9]. Konigsberg and colleagues combined four omics datasets (protein-coding RNA, protein, DNA methylation, and noncoding RNA) from IPF and healthy lung tissues to construct a multi-omics network, and their results confirmed previously validated molecules (i.e., MMP7, AGER, COL17A1, TNXB, LAMC3, and RARA-AS1) and pathways already known to be dysregulated in IPF disease [10]. Nevertheless, despite underlining that the use of such a multi-omics strategy can provide a comprehensive characterization of IPF, both studies show a lack of consistency. In fact, Zheng et al.’s observations were based on a particularly small and rather heterogeneous cohort of solely end-stage IPF patients (n = 9 IPF patients and n = 9 healthy donors) [9], while Konigsberg et al., despite the higher number of patients involved (n = 24 IPF subjects and n = 14 control subjects) in the study, did not stratify them based on their fibrotic stage/disease severity [10], which is of paramount importance when seeking to better clarify the molecular mechanisms involved in the development and progression of IPF. Overall, the molecular mechanisms involved in IPF still remain unclear, and their investigation using human samples is not straightforward for two main reasons: on the one hand, the disease can present with a high degree of inter-patient heterogeneity, and on the other hand, surgical lung biopsy is recommended only in a small number of cases with an uncertain diagnosis. As a consequence, there is a reduced number of human tissue samples available for research purposes, and furthermore, there are also difficulties in finding healthy lung counterparts as control samples. Hence, the most frequently used and internationally recognized animal model to investigate IPF and potential therapies is the bleomycin (BLM) mouse model due to its high degree of reproducibility as well as its effectiveness in mimicking many aspects of human IPF [11,12]. While lung tissue from BLM-induced fibrosis mouse models is highly characterized from a histopathological and radiological point of view [13], little is still known about the molecular events involved in the progression of bleomycin-induced pulmonary fibrosis. Here, a label-free bottom-up proteomics approach based on nano liquid chromatography coupled with electrospray ionization tandem mass spectrometry (nLC-ESI-MS/MS) analysis was used to characterize the proteomic fingerprint of fibrotic lung tissue derived from bleomycin-treated mice. This was conducted in order to elucidate the molecular mechanisms of fibrosis progression and to highlight novel molecular targets modulated by the anti-fibrotic drug nintedanib. The present work investigated the molecular hallmarks of fibrosis progression and drug-mediated slowdown while characterizing the proteomic fingerprint of a BLM-induced pulmonary fibrosis mouse model. Prior to lung explantation, all the mice underwent micro-CT lung imaging to non-invasively monitor the progression of pulmonary fibrosis and the anti-fibrotic effect of nintedanib (Figure 1a). Since air represents the natural contrast agent in lung CT imaging, normal lung parenchyma appears darker with respect to more fibrotic portions of the lung. This resulted in it being as dense as the surrounding tissue due to excessive collagen deposition. As expected, a worsening of lung aeration was already observed at 14 days after BLM administration and was highlighted by a decrease in the air content at the end of expiration. A consequent increase in the % poorly aerated tissue, corresponding to the most fibrotic areas, was measured in BLM- and NINT-treated mice compared to healthy mice both at 14 and 21 days (Table S1), especially for mice categorized as severely fibrotic. Overall, however, NINT slowed down fibrosis development by partially restoring lung function when compared to vehicles with a higher percentage of fibrotic tissue (Table S1). After the last imaging session, lungs were harvested and paraffin-embedded for histological assessment of the fibrosis. Representative whole-lung slides for each group of treatments are reported in Figure 1a. The quantification of fibrosis was based on the Ashcroft score evaluation [14,15]. Representative fields with diverse degrees of fibrotic lesions, along with their match with the corresponding CT scan, have been highlighted in Figure 1b. As described above, subjects categorized as mild fibrosis present normal lungs with minimal thickening of the alveolar walls at the parenchyma level. Subjects labeled as moderate exhibit a thickening of the alveolar walls without obvious damage of the lung architecture and show heterogeneous areas of individual fibroproliferative masses. The subjects defined as severe show large areas of the confluence of fibroproliferative foci with clear damage and distortion of the pulmonary structure. The good correlation between micro-CT and the average Ashcroft score, presented in Figure 1c, strongly supports the idea that these two independent methods can be used as a double read-out to describe the overall condition of each sample analyzed and thus to stratify mice for subsequent proteomic analysis. Lung tissue punches were then collected from histologically healthy parenchyma (control samples) or fibrotic areas (BLM-treated and NINT-treated samples) and analyzed by a label-free, bottom-up nLC-ESI-MS/MS proteomic approach. There were three different keys to interpret proteomic data: (i) time course after bleomycin treatment (saline, BLM 14 days, and BLM 21 days); (ii) average histological fibrotic grade (mild, moderate, or severe); and (iii) anti-fibrotic treatment effect. A total of 1413 proteins were identified with at least one unique peptide, whereas 272 proteins with at least 2 unique peptides were quantified using a label-free approach. This subset of 272 proteins was further investigated using multivariate data analysis approaches, followed by statistical analysis aimed at performing pairwise comparative proteomic analysis, and finally, functional and pathway analysis (Figure 2a). The capability of the label-free, bottom-up proteomic approach to highlight differences in the proteomic fingerprint of non-fibrotic and fibrotic lung tissue samples was initially investigated by performing multivariate analysis, using both unsupervised principal component analysis (PCA, Figure 3a,b) and heatmaps, combined with hierarchical clustering analysis (HCA, Figure 3c,d) of the protein expression data matrix from the 272 quantified proteins. The first 3 components explained 66.6% of the total variance. In particular, the PCA score plots (Figure 3a) of the first three components highlighted that the five biological replicates from saline samples with mild fibrosis (in red) were clustered together and well separated from all the fibrotic samples with moderate or severe fibrosis and treated with bleomycin (14 and 21 days) or with the anti-fibrotic drug nintedanib. However, while investigating further the samples across the fourth (14.7% of the total variance) and fifth (7.4% of the total variance) PCs, the fibrotic samples could also be separated based on their fibrosis severity, going from moderate to severe, while nintedanib biological replicates were clustered together in between moderate and severe samples. Vimentin (Vim), actin (Actg1), hemoglobin (Hba), and fibronectin (Fn1) present in the loading plots (Figure 3b) were positively correlated with fibrotic samples along the first component and negatively correlated with carbonyl reductase [NADPH]2 (Cbr2), uteroglobin (Scgb1a1), collagen alpha-1(IV) chain (Col4a1), and the advanced glycosylation end product-specific receptor (Ager), which was strongly correlated with control samples. Vim and Fn1 overexpression have already been observed in IPF lung tissue [16]. The correlation between these two proteins and fibrotic samples could be explained since an active fibrotic process is ongoing. During the epithelial-mesenchymal transition process (EMT), the specific proteins of the epithelial cells decrease while those typical of the mesenchymal ones, such as actin, increase [16]. Accordingly, the loadings plot showed a positive correlation of Actg1 with fibrotic samples. In accordance with our results, Lee et al. demonstrated the preventive role of uteroglobin in the development of pulmonary fibrosis in uteroglobin knockout mice (UG-KO) [17]. In particular, they demonstrated that a lack of UG predisposes mice to readily develop pulmonary fibrosis, even when treated with a low dose of bleomycin [17]. Furthermore, similarly to our observation, low levels of Ager and Col4a1 in fibrotic samples have been previously observed in IPF at the proteomic, genetic, and transcriptomic levels [6,18,19,20]. The dataset was further explored by performing HCA using heatmap data visualization. Data autoscaling was applied, and the data were clustered by using the top 50 most statistically significant features (using a T-test/Analysis of Variance (ANOVA)) based on treatment (Figure 3c) and on the average histological fibrotic grade (Figure 3d). Interestingly, the HCA results showed that the more appropriate key to interpreting the proteomic mass spectrometry data was the average histological grade of the samples (mild, moderate, and severe) and that sample clusterization was not influenced by the different observation time points of the BLM mouse model (14 or 21 days). The heatmap obtained by clustering the data based on treatment (saline, bleomycin 14 days, bleomycin 21 days, and nintedanib 21 days, Figure 3c) showed that while all five biological replicates of saline samples were under the same tree and clearly separated from all the treated samples (BLM and NINT), on the other hand, samples exposed to BLM treatment for 14 and 21 days were not well-separated. In fact, sample 1871, exposed to BLM for 21 days, was clustered under the same tree as those samples exposed to BLM for 14 days. The answer to this discrepancy was obtained by looking at the average histological fibrotic grade of each sample (Table 1). Sample 1871 was the only one among those exposed to BLM for 21 days to indeed have a severe grade of fibrosis, similar to that of those samples exposed to BLM treatment for 14 days, while the other 4 samples exposed to BLM treatment for 21 days were all characterized by a moderate grade of fibrosis. As a consequence of that, the HCA of the heatmap obtained when stratifying the data based on the average histological grade of fibrosis (Figure 3d), clearly shows that (i) mild samples (light green-colored) were all clustered under the same tree and well separated from all fibrotic samples; (ii) all fibrotic samples were clustered together under a separate tree and further divided into severe (dark orange-colored) and moderate (light orange-colored). In this Ashcroft-based heatmap, sample 1943, with severe fibrosis, is clustered, however, under the tree of moderate fibrosis samples. Indeed, sample 1943 had an average Ashcroft of 4.66 (severe fibrosis, average Ashcroft > 4.5); however, when evaluating the distribution of the single Ashcroft values, no area with a single microscope score of 8 was present, only 6.25% of the area with a single microscope score of 7, and only 9.4% of the area with a single microscope score of 6 (Table 1), for a total of 15.6% of the tissue with scores 6 and 7. However, all the other samples with a severe grade of fibrosis had a total tissue area with a score of 6 or 7, ranging from 24% up to 70%. The score distribution in sample 1943 is hence borderline between moderate and severe samples. It has to be noted that despite tissue punches being collected with a depth of about 2 mm from homogenous areas of fibrotic tissue, the Ashcroft score was evaluated on the superficial histological tissue slice of each FFPE block. Accordingly, the comparisons of the two HCA using heatmap visualization highlighted that the most appropriate parameter to interpret our mass spectrometry proteomic data was the fibrotic severity of tissue. Based on the results of the multivariate data analysis and with the aim of investigating the proteomic alterations involved in fibrosis progression, samples were stratified based on the severity of their fibrosis grade. For this purpose, only saline samples and bleomycin-treated samples were investigated, while all nintedanib-treated samples were excluded from this step. A one-way ANOVA, using an adjusted p-value cut-off of 0.05, was applied to the dataset, and 53 proteins were found to be statistically significant among the 3 groups (Table S2). In addition to the one-way ANOVA test, correlation analysis was performed using the Pattern Search tool in Metaboanalyst 5.0 against four specific patterns of protein expression: proteins that increase in accordance with the severity of fibrosis (pattern 1-2-3), proteins that, on the contrary, decrease (3-2-1), and proteins that are high (1-2-1) or low (2-1-2) only in moderate fibrosis (Tables S3 and S4). All 53 statistically significant proteins also had statistically significant correlation patterns (p-value ≤ 0.05). In particular, 36 proteins had an increasing trend based on the severity of fibrosis, whereas 15 proteins had a decreasing trend, whereas only 2 proteins were higher in moderate fibrosis (Figure 4b–e). Among the 15 proteins with a decreased trend in fibrosis (3-2-1 pattern), hepatic Flavin-containing Monooxygenase 1 (Fmo1) presented the highest correlation pattern. Fmo1 is an enzyme that catalyzes the N-oxygenation of secondary and tertiary amines and is involved in the metabolism of drugs [21]. Hence, the difference observed between mild, moderate, and severe samples is probably related to the bleomycin treatment itself and does not correlate with the mechanisms of fibrosis progression. The second protein with the highest correlation with the pattern 3-2-1 was platelet glycoprotein 4 (Cd36) (Figure 4c). Cd36 is a multifunctional receptor found on the platelet membrane that can bind to distinct types of ligands: thrombospondin, fibronectin, collagen, and other proteins or lipids [22]. The binding of these ligands to Cd36 activates multiple cellular responses, including angiogenesis, an inflammatory response, and fatty acid metabolism [23,24,25]. Numerous studies have investigated the role of Cd36 in lung fibrosis [26,27,28]. Some of these studies have shown that mice with a lack of Cd36 had reduced lung fibrosis, but our results suggest a protective role for Cd36 against lung fibrosis [26,27]. Comparable results were observed in a recent study from 2021, where Wang et al. found a significant decrease in the expression of Cd36 in fibroblasts from fibrotic lungs at both protein and messenger ribonucleic acid (mRNA) levels [28]. Cd36 handles the internalization of collagen in the platelets and its degradation [28]. Hence, the downregulation of Cd36 in moderate and severe fibrosis could be related to the inhibition of collagen catabolism, which has, as a direct consequence, collagen accumulation in the fibrotic lung. On the other hand, among the 36 proteins with a statistically significant correlation pattern (1-2-3), Coro1a was the one with the highest correlation (Figure 4d). Moreover, it has to be noted that, among all the 53 statistically significant proteins highlighted by the one-way ANOVA, only Coro1a and Rpl29 were statistically significant and with an increased trend in fibrosis in all the comparisons: moderate vs. mild, severe vs. mild, and severe vs. moderate (Table S2). Coro1a, also known as Translational Activator of Cytochrome Oxidase (TACO), is part of the conserved family of coronins, actin cytoskeletal regulators that promote cellular mobility and other actin-dependent processes [29,30]. Recent studies have shown the involvement of Coro1a in inflammatory processes [31,32], in cancer [33,34], and in fibrotic diseases that share common pathways with IPF: chronic classical cardiomyopathy (CCC), characterized by inflammation and myocardial fibrosis; renal interstitial fibrosis (RIF); and cystic fibrosis [35,36,37]. Interestingly, similarly to our observation during lung fibrosis progression, Wu et al. demonstrated that the expression level of Coro1a was significantly higher in the fibrotic tubular cells in chronic kidney disease compared to controls and that the expression level of Coro1a was directly related to fibrosis severity, hence the more severe the fibrosis was, the higher the expression level of Coro1a [36]. Furthermore, the presence of the Coro1a marker on FFPE sections was determined through its expression levels (Figure 4f). A significant increase (p < 0.001) in the Coro1a signal was found among the moderate and severe samples compared to the mild group. In agreement with the proteomic data, a statistically significant difference was also detected with the immunofluorescence assay in the severe versus moderate groups (Figure 4g). Taken together, all these results suggest a possible role for Coro1a as a specific protein indicative of the fibrotic process, irrespective of the affected district (kidney, heart, or lung). Additionally, Lysozyme C-2 (Lyz) and Transferrin (Trf) were the only two proteins, among the 53 statistically found to be significant in the one-way ANOVA, with a statistically significant correlation pattern and increased expression in moderate fibrosis, while their expression reverted to baseline levels in severe fibrosis (Figure 4e and Table S4). In particular, from the post-hoc analysis, it was observed that Lyz2 was statistically significant and had a higher expression only in the comparison of moderate vs. severe, while Trf had a statistically significant altered expression also in the comparison of moderate vs. mild (Table 1). Interestingly, both proteins were found to be associated with alveolar macrophages. Lysozymes have primarily a bacteriolytic function, are secreted onto epithelial surfaces, and are found in the primary and secondary granules of neutrophils, as well as the granules of mononuclear phagocytes [38]. Lysozymes are also expressed in both alveolar (which reside in the alveoli) and interstitial (located within the lung parenchymal tissue) macrophages [39]. Moreover, lysozyme overexpression in fibrosis, in particular Lyz1, has already been detected in a bleomycin-induced lung fibrosis rat model [40]. On the other hand, Trf, one of the iron metabolism-related proteins, binds to free iron and transports it into cells. Transferrin-bound iron is imported via the transferrin receptor and exported via the iron exporter ferroportin, both of which are expressed on airway macrophages (AM) and the respiratory epithelium [41]. Pulmonary iron content is therefore tightly regulated, and alterations in iron metabolism have been associated with chronic lung disease, and patients with idiopathic pulmonary fibrosis have been reported to have numerous aspects of dysfunctional iron metabolism. Similarly, pulmonary iron levels increase in bleomycin-induced pulmonary fibrosis in mice [41]. The fibrotic process in the BLM mouse model is characterized by three different phases, which include the inflammatory, proliferative, and maturation phases [13]. Our proteomic results revealed that, in moderate samples, the proteins Lyz2 and Trf were highly expressed, while their expression was lower in severe fibrosis samples. This is most likely due to the fact that lungs with moderate fibrosis present a proteomic profile related to cellular activities involved in tissue remodeling, while in severe fibrotic lesions the main cellular activities are related to ECM deposition and inflammatory cell recruitment. All 53 proteins found to be statistically significant in the one-way ANOVA were submitted to Cytoscape, and the KEGG pathway database was used as a reference database to gain an overview of the pathways represented in our dataset and, at the same time, map the pathways involved in IPF progression (Figure 4a). Fibrosis progression displayed the alteration of proteins involved in several pathways: ribosome and coronavirus disease; complement and coagulation cascades; advanced glycation end products and receptor for advanced glycation end products (AGE-RAGE) signaling pathway; ECM-receptor interaction; regulation of actin cytoskeleton; endocytosis; tight junction; 5′ adenosine monophosphate-activated protein kinase (AMPK) signaling pathway; RNA transport; pyruvate metabolism; hypoxia-inducible factor-1 (HIF-1) signaling pathway; and phagosome. The ribosome/coronavirus disease pathway was activated, with all the proteins (Rpl7a, Rpl7, Rps9, Rps3a1, Rpl18, Rpl29, and Rpl13a) being all overexpressed (based on post-hoc analysis) in severe fibrotic samples compared to mild. Interestingly, post-hoc analysis highlighted that Rpl7a, Rpl7, Rps9, and Rpl13a were also overexpressed in the comparison of severe to moderate, while Rpl18 and Rps3a1 were overexpressed in the comparison of moderate to mild. The ribosomal protein Rpl29 was the only one with a statistically significant increasing trend in all comparisons from mild to severe fibrosis (moderate-mild, severe-mild, and severe-moderate). Ribosomal proteins are involved in the regulation of apoptosis, cell proliferation, neoplastic transformation, cell migration, and invasion, as well as tumorigenesis [42]. Dysregulated cell proliferation (i.e., fibroblast cells) is one of the key hallmarks of fibrosis, and the overexpression of the ribosome pathway in fibrotic tissues has already been observed in human IPF samples [6]. Similarly, complement and coagulation cascades, platelet activation, and neutrophil extracellular trap formation pathways were also activated and shared the proteins fibrinogen alpha, beta, and gamma chains (Fga, Fgb, and Fgg), which were all overexpressed in severe fibrotic samples (severe-mild and severe-moderate) as previously observed [43,44]. The bleomycin-induced lung fibrosis mouse model (C57BL/6) with double OA administration of BLM has already been characterized from a histological and radiological perspective. However, the proteomic characterization of the model is still lacking but necessary in order to better understand the molecular mechanisms involved in the fibrotic process [13]. Among all the 53 statistically significant proteins highlighted in the one-way ANOVA, 22 proteins (Coro1a, Rpl7a, Des, Fn1, Rpl29, Fgg, Fga, Rpl7, Hnrnpa2b, Eif4b, Fgb, Rps9, Map4, Eif3b, Rpl13a, Rrbp1, Elavl1, Vim, Hnrnpm, Psmb10, Aldh6a1, and Rbm3) were overexpressed, while 3 proteins (Trf, Lyz2, and Myh14) were under expressed in the comparison of severe vs. moderate. All these proteins together represent the proteomic signature of fibrosis progression, from moderate to severe fibrosis. On the other hand, 15 proteins (Coro1a, Anxa1, Rpl29, Cndp2, Nme2, Rpl18, Rnh1, Anxa2, Lcp1, Snx2, Anxa4, Cct8, Arpc1b, Ckmt1, and Rps3a1) were overexpressed, while 13 proteins (Ldhb, Fmo1, Rras2, Cd36, Cbr2, Aldh1a1, Efemp1, Rab10, Col4a1, Ehd2, Lamb3, Selenbp1, and Mdh2) were under expressed in both the comparisons moderate vs. mild and severe vs. mild. Hence, all these 28 proteins together represent the proteomic fingerprint of fibrosis onset from a non-fibrotic (mild) to a fibrotic (severe and moderate) state. Further statistical analysis was conducted focusing on pairwise comparisons (p-valueadj ≤ 0.05 and with a fold-change ≥ 2 or ≤−2) (Table S5). In particular, 60 proteins (48 up-regulated and 12 down-regulated) were found to be altered in moderate BLM vs. mild, while 72 proteins were altered in the comparison of severe BLM vs. mild (48 up-regulated and 24 down-regulated) (Figure 2b). Among them, 22 proteins were differentially expressed only in moderate BLM/mild, 34 were differentially expressed only in severe BLM/mild, and 38 were shared proteins. In order to perform functional enrichment analysis, the gene names associated with the proteins differentially expressed in the pairwise comparisons were imported into the String-db tool. In particular, functional enrichment analysis of the 38 common proteins between the two comparisons showed a protein-to-protein interaction (PPI) network with 38 nodes and 63 edges and a PPI enrichment p-value of 1.06 × 10−9. The blood coagulation fibrin clot formation (Fga, Fgg, and Fgb) biological process was enriched (strength 2.4) as well as myofibroblast tissue expression (strength 2.06—Eln and Fn1 genes). Elastin (Eln) and Fn1 are proteins secreted by different subtypes of stromal cells and were overexpressed in both moderate and severe fibrotic samples. On the other hand, from the PPI network obtained from the 22 proteins differentially expressed only in the moderate/mild comparison, emerged the enrichment of the biological processes associated with the regulation of reactive oxygen species biosynthesis (strength 1.59; Ddah2, Rac1, Hsp90ab1, and Eef1a1), heat shock factor 1 (HSF1) activation (strength 2.4; Hsp90ab1 and Eef1a1), the AGE-RAGE signaling pathway (strength 1.47; Col4a2, F3, and Rac1), embryonic fibroblasts (strength 1.19; Hsp90ab1, Hspa8, Uba1, Rpn1, and Eef1a1), and fibroblast tissue expression (strength 1.09; Hsp90ab1, Hspa8, Uba1, Rpn1, Sptbn1, and Eef1a1). Heat shock proteins (HSP) are a category of stress proteins and the HSP90 family is the most abundant of the HSPs, regulating myofibroblast differentiation and promoting ECM and collagen synthesis as well as production [45]. Similarly to our results, the activation of HSP90 was observed both in patients with IPF as well as in BLM-treated mice, in which it could have a regulatory function on stromal cells [46]. The 34 proteins differentially expressed only in the severe/mild comparison showed a PPI network with 34 nodes and 22 edges and a PPI enrichment p-value of 9.24 × 10−5. This PPI network showed the enrichment of bronchiolar epithelium tissue expression (strength 2.64), with Cyp2f2 and Scgb1a1 proteins being both downregulated in severe fibrosis. These data suggest a modification of the secretive function of bronchiolar epithelium that is mandatory for the protective barrier and preservation of proper airway function [47]. Furthermore, our proteomics results on the BLM mouse model were compared with two recent multi-omics human studies in order to highlight similarities and differences in terms of protein expression between mice and human subjects (Table S5) [9,10]. The 78 proteins found to be commonly differentially expressed in both the transcriptomics and proteomics datasets in the Zheng et al. study were compared to our results [9]. Ager, Selenbp1, and Limch1 were the only three proteins shared between ours and the dataset of Zheng et al. All three proteins were downregulated only in the comparison of severe vs. mild, in accordance with what was observed for IPF end-stage patients [9]. Interestingly, both Selenbp1 and Limch1 were included by Zheng et al. in the thirteen potential marker gene list with the most significant fold changes and adjusted p values [9]. These three proteins were also observed in the study of Konigsberg et al. having the same downregulated expression, with the exception of Limch1, which was upregulated in the poly RNA-seq dataset. A total of 56 proteins were in common between our study and that of Konigsberg et al., with Aldh1a1, Ager, Edh2, Cd36, Ehd4, Selendp1, Lamc2, and Ctnnd1 being downregulated while Vim, Hspa8, Des, Rps3, Ldha, Rps18, Khsrp, and Palld, were upregulated, respectively, in both mouse and human proteomics studies. It has to be noted that Ager, Ehd2, Ehd4, Selendp1, and Ctnnd1 were all coherently downregulated in both mouse and human proteomics and transcriptomics datasets [9,10]. These results represent a step forward in the proteomic characterization of the bleomycin-induced pulmonary fibrosis mouse model. The molecular mechanisms of action of nintedanib and its ability to slow down the progression of BLM-induced lung fibrosis were explored by comparing samples treated with NINT (NINT21d) with samples exposed to BLM for 21d; all samples in both groups had moderate fibrosis. Fifteen proteins were found to be differentially expressed in this pairwise comparison, with 14 up-regulated and 1 down-regulated (Figure 2b). In particular, 7 proteins were altered only in the comparison of Moderate NINT 21d/Moderate BLM 21d, whereas 2 proteins were common to all the three comparisons (Ldhb and Pdlim5) (Figure 5a,b). The pattern search tool of MetaboAnalyst v.5.0 was used in order to detect proteins with a statistically significant correlation pattern (p-value ≤ 0.05) with an increased or decreased expression (patterns 1-2-1 or 2-1-2) only in NINT 21d samples compared to mild and moderate BLM 21d. A total of 22 proteins were highlighted (Figure 5c). Among those proteins, only lactate dehydrogenase b (Ldhb), myosin heavy chain 11 (Myh11), and electron transfer flavoprotein subunit beta (Etfb) were statistically significant in the pairwise comparison moderate NINT 21d vs. moderate BLM 21d (p-valueadj ≤ 0.05 and FC ≥ 2 or FC ≤ −2). Rps26, Ddah2, Tnks1bp1, and Pkm were also statistically significant but with an FC ≥ 1.5 or FC ≤ −1.5 (Figure 5c). These 22 proteins were used to build the KEGG Pathway network (Figure 5d), underlining the alteration of pyruvate metabolism, oxidative phosphorylation, proteoglycans in cancer, the Mitogen-activated protein kinase (MAPK) signaling pathway, and the ribosome. Lowering the FC to 1.5 in both comparisons, moderate BLM 21d vs. mild and moderate NINT 21d vs. moderate BLM 21d, ten proteins (Tnc, Rps26, Myh11, Ldhb, Etfb, Ddah2, Tnks1bp1, Pkm, Ighm, and Acaa2) were inverted in their trend by the action of NINT (Figure 5e,f). Among them, Myh11 was the only one statistically significant within the FC range ≥ 2 or ≤−2 and included in the short list of 7 proteins specific for the comparison of moderate NINT vs. moderate BLM (Figure 5b). Myh11 was down-regulated (FC, −1.51) in the comparison of moderate BLM vs. mild, while it became up-regulated in the comparison of moderate NINT 21d vs. moderate BLM 21d (FC: 2.13). Myh11 is a protein that participates in muscle contraction, tight junction, and regulation of the actin cytoskeleton (Figure 5d) and converts chemical energy into mechanical energy through the hydrolysis of Adenosine Triphosphate (ATP) in Adenosine Diphosphate (ADP) [48]. Previous studies have demonstrated the downregulation of Myh11 in several types of cancers. A recent study demonstrated its downregulation even in lung cancer, highlighting its potential role as a novel drug target and prognostic indicator [49,50,51]. The molecular mechanisms of Myh11 remain, however, still unclear. The downregulation of Myh11 in the moderately fibrotic tissue suggests a decreased utilization of ATP by myosin, possibly leading to an accumulation of ATP in the cells. Both hypoxic conditions and inflammation, which have been hypothesized to be related to fibrosis formation, lead to increased extracellular ATP. Based on our results, Myh11 might be involved in this process [52]. It should be noted that in our results, Myh11 appeared to be statistically significant in the comparison of severe vs. mild but with an FC of only 1.22, while in the comparison of severe vs. moderate had an FC of 1.8. Hence, while Myh11 was downregulated in moderate fibrosis, on the other hand, its expression was restored in severe fibrosis (Figure S1). Nevertheless, the molecular mechanisms involved in Myh11 alteration due to fibrosis progression, as well as the modulatory effect of nintedanib, have yet to be further explored and clarified. Ldhb was the only protein being both statistically significant with an FC ≥ 2 or FC ≤ −2 shared among all the three pairwise comparisons as well as reverted by nintedanib action (Figure 5a,b). Lactate dehydrogenases, both A and B, are enzymes involved in the conversion pyruvate-lactate: Ldha has a higher affinity for pyruvate and converts pyruvate to lactate, as well as nicotinamide adenine dinucleotide (NAD) + hydrogen (H) (NADH) to NAD+, in anaerobic conditions; whereas Ldhb has a higher affinity for lactate, converting lactate to pyruvate, as well as NAD+ to NADH, in aerobic conditions. Previous studies have demonstrated the accumulation of lactic acid in IPF lung tissues compared to controls [53]. Our results showed that in bleomycin-induced IPF samples (both moderate and severe), Ldhb was down-regulated while Ldha was up-regulated and, interestingly, despite their similar structure, nintedanib modulated only Ldhb expression (Figure 6a). The dysregulation of lactate metabolism, the cells involved, and the exact mechanisms of action remain yet to be clarified. Danforth et al., in a recent study of 2021, have observed the alteration of lactate metabolism in human Alveolar type II Epithelial Cells (AEC2), and they hypothesize its central involvement in the development and progression of IPF [54]. Although the conversion of pyruvate to lactic acid takes place in physiologically low oxygen tension conditions, it can also still occur in aerobic conditions, known as the Warburg effect, which is typically observed in cancer cells [55]. Recently, a “reverse Warburg effect” has been proposed as a factor contributing to the pathogenesis of fibrosis, suggesting that aerobic glycolysis takes place in the fibroblasts while the secreted glycolytic metabolites influence the behavior of other cell types (such as epithelial cells and macrophages) [56]. Moreover, our results highlighted the deregulation of other proteins (Pyruvate kinase (Pkm), 3-ketoacyl-CoA thiolase (Acaa2), Etfb and Cd36) related to glycolytic pathway, β-oxidation of fatty acids and Tricarboxylic acid (TCA) cycle, suggesting the alteration of ATP metabolism in moderate fibrosis and their modulation by nintedanib (Figure 6b). Interestingly, the expression of Pkm, Acaa2, and Etfb, similar to what was observed for Myh11, was reverted in severe fibrosis (Figure S1). Trifluoroacetic acid (TFA), ammonium bicarbonate, trypsin from porcine pancreas (Proteomics Grade, BioReagent, Dimethylated), dithiothreitol (DTT), iodoacetamide (IAA), and formic acid (FA) were purchased from Sigma-Aldrich (Sigma-Aldrich Chemie Gmbh, Buchs, Switzerland). HPLC-grade water, acetonitrile (ACN), ethanol, and toluene were purchased from Honeywell (Honeywell Research Chemicals Riedel-de-Haën™, Seelze, Germany). RapiGest SF surfactant was purchased from Waters Corporation (Waters, Milford, MA, USA). ZipTips were purchased from EMD Millipore (Billerica, MA, USA). The study was conducted using male inbred C57BL/6J mice (Envigo, San Pietro al Natisone, Udine, Italy) aged 7 to 8 weeks. Prior to use, mice were acclimated to the local vivarium conditions (20–24 °C room temperature; 40–70% relative humidity; 12-h light-dark cycle) for at least 5 days, having free access to standard rodent chow and softened tap water. All procedures were in compliance with the principles outlined in the European Directive 2010/ 63 UE, Italian D.Lgs 26/2014, and the revised “Guide for the Care and Use of Laboratory Animals” (National Research Council Committee, US, 2011) [57]. Animal studies were performed in an AAALAC (Association for Assessment and Accreditation for Laboratory Animal Care) certified facility at Chiesi Farmaceutici and authorized by the Italian Ministry of Health with protocol number 809/2020-PR and by the internal AWB (Animal Welfare Body). A visual analog scale (0–10) for pain assessment was assessed daily by a designated veterinarian or trained technicians. VAS ≥ 6 and/or body weight loss ≥ 20% were considered humane endpoints, as were signs of dyspnea or apathy evaluated by a designated veterinarian. Pulmonary fibrosis was induced through oropharyngeal aspiration (OA) [13,58] of 10 μg/mouse bleomycin hydrochloride (Baxter, BLM group) diluted in 50 μL of saline, while vehicle mice received only 50 μL saline (Saline group). The OA procedure was performed at days 0 and 4 in mice lightly anesthetized with 2.5% isoflurane. After induction, high-calorie dietary supplements (recovery gel from Dietgel) and sterile sunflower seeds were added daily to the standard rodent chow in order to reduce body weight loss. At day 7, BLM-treated mice were randomly divided into 2 subgroups, receiving either nintedanib (60 mg/kg/day, Carbosynth Limited, Compton, UK) dissolved in Tween80 0.05% in saline (NINT group) or vehicle (Tween80 0.05% in saline), by gavage, daily for 2 weeks (Figure 7a). Prior to lung excision for proteomic analyses, micro-CT imaging was performed with a Quantum GX Micro-CT (PerkinElmer, Inc., Waltham, MA, USA). The system was calibrated monthly with standard phantoms to check the noise, uniformity, low contrast, and resolution [59]. Images were acquired with an intrinsic retrospective two-phase respiratory gating technique with the following parameters: 90 KV, 88 μA over a total angle of 360° for a total scan time of 4 min. The “high speed” scan mode resulted in two 3D datasets corresponding to the two different phases of the breathing cycle (inspiration and expiration), but the images and data reported in this work refer to the end-expiratory phase. An automated deep learning (DL)-based model previously published by Vincenzi et al. [60] was applied to rescale CT scans into Hounsfield units (HU), setting −1000 HU as the density of air and 0 HU as the density of water, and for the segmentation of the total lung volumes. Pre-clinical HU density ranges [61] were then applied to segmented lungs for the quantitative assessment of parenchymal lesions. Normo-aerated tissue [−860 HU; −435 HU] and poorly-aerated tissue [−435 HU; +121 HU] were defined and expressed as % of total lung volumes. Finally, 3D renderings were generated by using the Analyze software (Analyze 12.0; Copyright 1986–2017, Biomedical Imaging Resource, Mayo Clinic, Rochester, MN, USA). Following the in vivo procedures, subsets of saline-, BLM-, and NINT-treated mice were culled for the histological assessment of pulmonary fibrosis. The lungs were removed and inflated through the trachea by gentle infusion with 0.6 mL of 10% neutral buffered formalin and fixed for 24 h. Samples were then dehydrated in a graded ethanol series, cleared in xylene, and paraffin-embedded. Sections of 5 μm thickness were cut with a rotary microtome (Slee Cut 6062, Slee Medical, Mainz, Germany) and stained with hematoxylin and eosin (H&E) and Masson’s trichrome (Figure 7b). An immunofluorescence assay was performed to detect, in situ, the Coronin-1A (Coro1a) protein. For this purpose, an anti-Coro1a antibody (0.5 μg/mL; ab203698; Abcam, Cambridhe, UK) was used. The secondary antibody (Rhodamine Red-X donkey anti-rabbit IgG 711-295-152; Jackson Immunoresearch, Cambridgeshire, UK) was used at a 1:200 dilution. The nuclei were counterstained with DAPI, and the sections were mounted with a suitable mounting medium for fluorescence [62]. For analyses, slide images were acquired by a NanoZoomer S-60 digital slide scanner (NanoZoomer S60, Hamamatsu, Hamamatsu City, Japan). Several 10X fields were analyzed for each sample and morphological changes were graded semi-quantitatively according to the scale defined by Ashcroft [63] and modified by Hübner et al. [64] by two independent researchers blinded to the experimental design. An average Ashcroft () was derived for each histological sample as a descriptive indication of the overall frames. In addition, the frequency percentage for each Ashcroft score was also reported, as presented in Table 1, so that the most frequent grade could be immediately identified for each sample analyzed. After animal sacrifice, lung tissue samples were collected and histologically processed, and 19 formalin-fixed and paraffin-embedded (FFPE) tissue blocks were prepared. From each FFPE block, at least 5 tissue punches with a size of 2 × 3 mm were collected and stored at room temperature until the day of the analysis. At least 3 biological replicates (Table 1) for each group (saline, BLM 14 days, BLM 21 days, and NINT 21 days) were used for nLC-ESI-MS/MS analysis. In order to stratify the mice for the following proteomic analysis, both the average Ashcroft value () and the micro-CT % poorly aerated tissue (p) were considered. Subjects were divided into 3 categories defined as mild ( ≤ 3.5 and p ≤ 25%), moderate ( > 3.5 or ≤4.5 and p > 25%), and severe ( > 4.5 and p > 50%) (Table 1). Proteins were extracted from lung tissue punches for their identification and quantification by nLC-ESI-MS/MS [65]. Deparaffinization of the FFPE samples was performed by incubating the tissues for approximately 1 h at 65 °C, followed by 3 consecutive washes in toluene (4 min each while sonicating) and centrifugation (14,000× g rpm, 5 min). To ensure complete paraffin removal, these washing steps were repeated in triplicate. Subsequently, consecutive washes were performed while sonicating, in 100% (×2), 90%, and 70% of ethanol for 4 minutes each, as well as 100% water for 2 min, followed by centrifugation (14,000× g rpm, 5 min). Heat-induced antigen retrieval using a citrate buffer (10 mM, pH 6) at 97 °C for 45 min was performed, followed by centrifugation (14,000× g rpm, 5 min). Samples were washed while sonicating with 100% water for 2 min, followed by centrifugation (14,000× g rpm, 5 min). A 50 mM ammonium bicarbonate buffer solution was added to the samples, and, in order to enhance protein solubilization and digestion, RapiGestTM SF surfactant was added to each sample to a final concentration of 0.1%. After disulfide bond reduction (DTT 10 mM) and alkylation (IAA 15 mM), proteins were digested by adding 5 μg of trypsin (0.2 μg/μL), and samples were incubated overnight at 37 °C. The enzymatic reaction was stopped by acidification with TFA (pH < 2). Each sample was dried with a vacuum centrifugal evaporator (Hetovac, Savant) and resuspended in 50 µL of loading pump phase A (H2O:ACN:TFA 98:2:0.1), and protein content was determined by the NanoDrop assay (Thermo Scientific, Sunnyvale, CA, USA). Finally, desalting and concentration of the samples were performed using Ziptip™ µ-C18 pipette tips (Merck Millipore Ltd., Darmstadt, Germany), and, following the standard protocol provided by Millipore, peptides were eluted with a solution of 80% ACN and 0.1% FA, dried under vacuum using an Hetovac centrifuge, and resuspended in 50 µL of loading pump phase A. Tryptic peptides were then injected (volume of injection: 10 μL) into a Dionex UltiMate 3000 rapid separation (RS) LC nano system (Thermo Scientific, Germany), coupled on-line to an Impact HD™ Ultra High Resolution-QqTOF (Bruker Daltonics, Bremen, Germany). Each sample was injected in duplicate to minimize technical variability. Samples were loaded into a pre-column (Thermo Scientific, Acclaim PepMap 100, 100 µm × 2 cm, nanoViper, C18, 3 µm), followed by a 50 cm nano-column (Thermo Scientific, Acclaim PepMap RSLC, 75 µm × 50 cm, nanoViper, C18, 2 µm). The HPLC separation was performed at 40 °C and at a flow rate of 300 nL/min using a multistep gradient of 4 h from 4% to 98% of nanopump phase B (nanopump phase A being H2O w/0.1% FA and nanopump phase B being 80:20 ACN:H2O w/ 0.08% FA). The column was on-line interfaced to a nanoBoosterCaptiveSpray™ ESI source (Bruker Daltonics), where eluted peptides were ionized using heated nitrogen dry gas (T = 150 °C; 3 L/min) enriched with ACN. MS/MS spectra were generated by collision-induced dissociation, assisted by N2 functioning as a collision gas. Mass accuracy was improved by calibrating the instrument with a mix of ten standards with known masses (MMI-L Low Concentration Tuning Mix, Agilent Technologies, Santa Clara, CA, USA) by using a specific lock mass (m/z 1221.9906), and by an internal calibration based on a 15 min segment (10 mM sodium formate cluster solution) before the beginning of the gradient for every single run. Mass spectrometry data were acquired in data-dependent acquisition (DDA) modality, with automatic switching between full-scan MS and MS/MS acquisition. Acquisition parameters were set as already described [66]. The Compass DataAnalysis v4.1 software (Bruker Daltonics, Hamburg, Germany) was used to calibrate, deconvolute, and convert the acquired raw data prior to protein identification and quantification. Peaks Studio X-Plus (Bioinformatics Solutions Inc., Waterloo, ON, USA) was used for protein identification and label-free quantification analysis. The parameters were set as follows: trypsin as the enzyme, carbamidomethyl as a fixed modification, oxidation (M), and FFPE+12 and FFPE+30 as variable modifications. The precursor mass error and the fragment mass error tolerances were set at 20 ppm and 0.05 Da, respectively. The identification engine uses an in-house-constructed UniProt reference database (accessed July 2021, 565,254 sequences; 203,850,821 residues), and the taxonomy of Mus musculus was selected (17,089 sequences). For identification, an FDR ≤ 1% was applied to all the analyses. Proteins were considered identified if there was at least one unique significant peptide (p-value ≤ 0.05). Label-free quantification analysis on Peaks Studio X-Plus was performed, and both quality and average abundance filters were applied and equal to 5 and 1 × 10−5, respectively. Proteins were selected for quantification if they were identified with at least 2 unique peptides. The list of quantified proteins was then imported into Microsoft Excel, and the statistical analysis was performed using in-house Excel add-in software. The area of the top 3-peptides was used to calculate the relative abundance of each protein. In order to investigate fibrosis progression and the effect of nintedanib treatment, the comparisons of moderate/mild, severe/mild, and moderate NINT/moderate BLM were evaluated. The quantified proteins were filtered using the non-parametric Mann–Whitney U-test corrected for multiple testing using a Benjamini–Hochberg adjusted p-value ≤ 0.05 and a fold-change (FC) ≥ 2 or ≤−2 in order to further investigate only those statistically significant and altered in fibrosis progression and drug-mediated slowdown. In order to further explore those pathways involved in fibrosis progression and drug mediated slow-down, the fold change was lowered to FC ≥ 1.5 or FC ≤ −1.5. Statistical analysis and exploratory data analysis were further investigated using Metaboanalyst 5.0 [67] and clustvis web tools [68]. Functional analysis of the comparisons was performed using the Cytoscape 3.9.1 network analysis platform. Gene ontology (GO) annotation, including biological process (BP) and molecular function (MF), as well as pathway term clustering using the Kyoto Encyclopedia of Genes and Genomes (KEGG) [69], were carried out using ClueGO 2.5.9 and CluePedia 1.5.9 [70]. Identifiers were UniProt accession numbers, and the organism reference database was Mus musculus (ClueGO mapped date 13 May 2022). For all of the analysis, the default parameters were used. The String-db open-source platform (https://string-db.org/ accessed on 17 September 2023) was used in order to investigate protein-protein interaction (PPI) networks. This work represents a step forward in the comprehension and improved characterization of the bleomycin-induced pulmonary fibrosis mouse model. In particular, the proteomic fingerprint of fibrosis progression and response to nintedanib was unveiled from lung tissues. Although in this study a bulk analysis of lung tissue was performed, and spatial protein information was lost, our findings provide novel insights into the proteomic alterations involved in fibrosis progression and in drug-mediated slow-down. Coro1a could be considered as a putative biomarker of pulmonary fibrosis onset and progression. However, in the near future, further in vivo and in vitro studies are needed in order to determine Coro1a function and to decipher the underlying molecular mechanisms in lung fibrosis. Moreover, Ldhb expression, but not Ldha, was significantly restored during nintedanib treatment, suggesting Ldhb is a putative target of nintedanib action, but its role needs to be further investigated and validated in future research. The next step will be to integrate our nLC-ESI-MS/MS proteomics analysis with information regarding the spatial localization of proteins using single-cell matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI), allowing us to use more deeply investigate specific regions of interest (ROI) and hence to (i) explore the cell-to-cell communication jigsaw occurring during the fibrotic process; (ii) characterize the overall tissue proteomic profile based on the single microscope field histological fibrotic grade; and (iii) investigate the proteomic alterations related with different time courses after bleomycin treatment (i.e., comparing ROIs, having the same Ashcroft score, from samples exposed to BLM for 14 days with 21 days samples).

PMC10002928

36824022

Hu Shaohua,Wang Yihui,Zhang Kaier,Bai Ying,Wang Xiaoyi,Zhao Hui,Di Guohu,Chen Peng

Aquaporin 5 maintains lens transparency by regulating the lysosomal pathway using circRNA

23-02-2023

aquaporin 5,autophagy,circRNA,HSF4,lens opacity,miRNA,mRNA

Abstract The lens is transparent, non‐vascular, elastic and wrapped in a transparent capsule. The lens oppacity of AQP5−/− mice was increased more than that of wild‐type (AQP5+/+) mice. In this study, we explored the potential functional role of circular RNAs (circRNAs) and transcription factor HSF4 in lens opacity in aquaporin 5 (AQP5) knockout (AQP5−/−) mice. Autophagy was impaired in the lens tissues of AQP5−/− mice. Autophagic lysosomes in lens epithelial cells of AQP5−/− mice were increased compared with AQP5+/+ mice, based on analysis by transmission electron microscopy. The genetic information of the mice lens was obtained by high‐throughput sequencing, and then the downstream genes were analysed. A circRNA‐miRNA‐mRNA network related to lysosomal pathway was constructed by the bioinformatics analysis of the differentially expressed circRNAs. Based on the prediction of the TargetScan website and the validation by dual luciferase reporter assay and RNA immunoprecipitation‐qPCR, we found that circRNA (Chr16: 33421321‐33468183+) inhibited the function of HSF4 by sponging microRNA (miR‐149‐5p), and it downregulated the normal expression of lysosome‐related mRNAs. The accumulation of autophagic lysosome may be one of the reasons for the abnormal development of the lens in AQP5−/− mice.

Aquaporin 5 maintains lens transparency by regulating the lysosomal pathway using circRNA The lens is transparent, non‐vascular, elastic and wrapped in a transparent capsule. The lens oppacity of AQP5−/− mice was increased more than that of wild‐type (AQP5+/+) mice. In this study, we explored the potential functional role of circular RNAs (circRNAs) and transcription factor HSF4 in lens opacity in aquaporin 5 (AQP5) knockout (AQP5−/−) mice. Autophagy was impaired in the lens tissues of AQP5−/− mice. Autophagic lysosomes in lens epithelial cells of AQP5−/− mice were increased compared with AQP5+/+ mice, based on analysis by transmission electron microscopy. The genetic information of the mice lens was obtained by high‐throughput sequencing, and then the downstream genes were analysed. A circRNA‐miRNA‐mRNA network related to lysosomal pathway was constructed by the bioinformatics analysis of the differentially expressed circRNAs. Based on the prediction of the TargetScan website and the validation by dual luciferase reporter assay and RNA immunoprecipitation‐qPCR, we found that circRNA (Chr16: 33421321‐33468183+) inhibited the function of HSF4 by sponging microRNA (miR‐149‐5p), and it downregulated the normal expression of lysosome‐related mRNAs. The accumulation of autophagic lysosome may be one of the reasons for the abnormal development of the lens in AQP5−/− mice. The lens is the main refractive structure of the eyeball. The transparency of the lens is important. The inner lens fibre is completely surrounded by the transparent lens capsule. A layer of epithelial cells is attached to the medical surface of the anterior lens capsule. When the epithelial cells reach the equator, they continue to elongate and bend, and become to lens fibres cells. Degradation of organelles begins in embryonic primary lens epithelial cells (LEC), and it continues in lens fibre cells after birth. , The transformation of lens epithelial cells into lens fibre cells occurs throughout life, but its level and speed gradually decrease with age. , Aquaporins (AQPs), water‐selective channel proteins, allow water to move rapidly across the plasma membrane. Three aquaporins are expressed in mammalian cornea (AQP1, AQP3 and AQP5) and three in the lens (AQP0, AQP1 and AQP5). AQP5 was expressed in lens epithelial and fibre cells at the gene , and protein , levels. The subcellular distribution may be orchestrated by its phosphorylation status. AQP5 on the cell membrane played an important role in maintaining lens transparency. In hyperglycaemia caused by diabetes, AQP5 can maintain lens homeostasis and transparency. In a previous study, we found that a novel AQP5 mutation (p. L51P) was related to congenital cataracts, and lens opacity appeared in AQP5 knockout mice. Autophagy is an intracellular process that maintains nutritional and energy balance by digesting cytoplasmic components or organelles. It has been proved that the development of cataract may be related to autophagy, especially macroautophagy. ATG5 mutations can lead to lens epithelial abnormality and cataract. Autophagy is a lysosome‐mediated degradation process. , Lysosomal hydrolases eventually degrade autophagic substrates. Microtubule‐associated protein 1A/1B‐light chain 3B (LC3B) and a multifunctional scaffolding protein p62/SQSTM1 (p62) are often used to assess autophagy. Lysosomal components are retrieved to replenish the lysosomal pool after cargo is degraded. In the lens epithelial and fibre cells, abnormal degradation of organelles caused by dysfunction of lysosomes can cause catract. In our study, aquaporin 5 knockout (AQP5 −/−) mice exhibited lens opacity and abnormal autophagy compared with AQP5 +/+ mice. Differentially expressed circRNAs were screened, and a circRNA‐miRNA‐mRNA network associated with lysosomes was constructed. Our result indicated that AQP5 deficiency may lead to abnormal lens development, and its mechanism may be related to the disorder of circRNA‐regulated autophagy. AQP5 −/− (C57BL/6 N) mice were produced through CRISPR/Cas9 technology (Cyagen Biosciences Inc. Guangzhou, China). All experiments were approved by the Animal Care and Use Committee of Qingdao University (Qingdao, China). Examination with an Ophthalmic slit lamp was performed at 1–12 months of age. Primary lens epithelial cell was prepared from mice aged between 5 and 6 weeks. The lens capsules were separated and digested with 1.5 mg/mL Dispase II enzyme at 37°C for 5 min. After neutralization with complete medium, the cells were centrifuged and suspended and inoculated in Dulbecco's modified Eagle media: Nutrient Mixture F‐12 (Biological Industries) (containing 5 μg/mL TGF‐β, and 2% fetal bovine serum). Lens capsules were isolated and fixed in the electron microscope fixation fluid. They were then fixed with 1% osmium tetroxide. After the samples were dehydrated and embedded, sections of 60–80 nm were made and stained with 2% uranyl acetate solution and lead citrate. The images were collected and analysed under a transmission electron microscope. Dispersion of lens capsules under a microscope was followed by extraction of the proteins with lysis buffer (Beyotime Biotechnology, Shanghai, China). Protein extract was separated by electrophoretic 10% or 12.5% SDS‐PAGE (EPizyme, Shanghai, China). Then, it was transferred to PVDF membranes. The blots were incubated with the primary antibodies of LC3II/I (Abcam, ab192890), P62 (Abclonal, A19700), Lamp1 (Abclonal, A16894), Cln5 (Abclonal, A12886), Gla (Abclonal, A1700) and Lipa (Abclonal, A6385). The blots were incubated with a secondary antibody (ZSGB‐Bio, ZB‐2301), and then they were visualized using enzyme‐linked chemiluminescence using the ECL kit (Applygen, Beijing, China). Immunofluorescence staining was performed on primary LEC and frozen lens sections. They were fixed in 4% paraformaldehyde and then permeated with 0.1% Triton X‐100 (T8200, Solarbio). The samples were stained with primary antibodies and then with secondary antibodies (Thermo Fisher Scientific, Rockford, IL, USA, A11015 and A21207). Nuclei were stained with 4′6‐diamidino‐2‐phenylindole (Beyotime Biotechnology, Shanghai, China). Total RNA was extracted from 6 lenses (combined into one sample, respectively) by using RNA extraction kit (Invitrogen, Carlsbeth, CA, USA). Transcriptome high‐throughput sequencing was performed by Cloud‐Seq Biotech (Shanghai, China). Purified RNA samples were constructed into mice lenses RNA libraries by using the Total RNA Library Preparation Kit (Illumina, San Diego, CA, USA). Then, the 10‐pM library was denatured and reverse‐transcribed into single‐stranded DNA molecules. Finally, the Illumina HiSeq sequencer was used for 150 cycles after setting the parameters. HiSeq 4000 sequencer (Illumina) was used to read the paired terminals and then the Q30 was used for quality control. High‐quality reads were mapped using STAR software (version 2.5.1b). The results were input into the DCC software (version 0.4.4) and then matched. Unpaired join points were compared with identify candidate circRNAs. Finally, the data were standardized using Edger software (version 3.16.5). Table 1 showed the preliminary analysis of high‐throughput sequencing results. The genes with |log10| > 2 and p‐value < 0.05 under the Q30 data set were contained in Table 1. Finally, the miRNAs were predicted by TargetScan (https://www.targetscan.org/) and miRanda (http://www.microrna.org/microrna/), the results were depicted in Tables 2, 3, 4. These together with circRNAs and mRNAs were constructed as the regulation network diagram of circRNA‐miRNA‐mRNA related to the lysosomal pathway. Tables 5 and 6 included primers of circRNAs and mRNAs. Primers for miRNA reverse transcription and qRT‐PCR are listed in Tables 7 and 8. Venn diagram and cluster heat map plot were drawn for circRNAs. Venn diagram, cluster heat map and volcano plot were drawn for mRNAs. High quality reads were uploaded to STAR software (version 2.5.1b) (http://www.bioinfo‐scrounger.com/). Through GO (http://www.geneontology.org) analysis the function of mRNA was analysed from three aspects: molecular function, biological processes and cell composition. KEGG (http://www.genome.jp/kegg) pathway enrichment analysis was used to explain the pathways of the differentially expressed genes. p < 0.05 was considered significant. The top 10 enhanced GO terms were ranked according to their p‐values. The expression difference of Chr16: 33421321‐33468183+, mmu‐miR‐149‐5p, and HSF4 were analysed using the EZ‐Magna RIP RNA‐Binding Protein Immunoprecipitation kit (Millipore, Bedford, MA United States). The lens was lysed into cell suspensions in RIP lysis buffer. The extract was then incubated with immunoprecipitation buffer containing magnetic beads conjugated with Anti‐Ago2 (ab186733, Abcam, Cambridge, MA, USA) or Anti‐IgG (ab181236, Abcam). The beads were collected and washed. The RNA complex was isolated by phenol–chloroform extraction. The enrichment levels of target genes were analysed by qRT‐PCR. Tables 5 and 6 show the primers used for selected circRNAs and mRNAs. The TargetScan database was employed to predict the possible binding sites of HSF4 and mmu‐miR‐149‐5p which might target the 3′UTR of HSF4 at three positions: 159–165, 500–506 and 517–523. To determine the interaction between mmu‐miR‐149‐5p and HSF4, the wild‐type and mutated target sequence of HSF4 were cloned into luciferase vector plasmids (Genome Editech, Shanghai, China). The constructs were wild‐type HSF4 (HSF4 WT) and HSF4 mutant 1 (159–165) (HSF4 MT1), HSF4 mutant 2 (500–506) (HSF4 MT2) and HSF4 mutant 3 (517–523) (HSF4 MT3). For the luciferase assay, HEK‐293 cells were co‐transfected with (1) HSF4 WT together with the NC mimics or miR‐149‐5p mimics; (2) HSF4 MT1 together with the NC mimics or miR‐149‐5p mimics; (3) HSF4 MT2 together with the NC mimics or miR‐149‐5p mimics; (4) HSF4 MT3 together with the NC mimics or miR‐149‐5p mimics. Lipofectamine 2000 (Invitrogen) was used to perform the transfection according to the manufacture's protocol. According to the instructions and prior to standardization of Renilla luciferase internal control, fireflies and Renilla luciferase activity were analysed using a dual luciferase reporting kit (Promega, Madison, WI, USA). Statistical analysis was performed using GraphPad Prism 8.0 (GraphPad Software Inc., La Jolla, CA, United States). The results were presented as mean ± SD. Comparison between two groups was assessed using Student's t‐test (p < 0.05 considered significant). All experiments were repeated three times. Through observation under slit lamp, it was found that the lens opacity of AQP5 −/− mice increased with age (Figure 1A). To investigate the possible mechanism of AQP5 deficiency on lens transparency, RNA was obtained from the lens of mice and sequencing was performed. A total of 2780 circRNAs were present in the lenses, of which 870 had not been reported before (Figure 1B). There were many types of circRNAs, with exons accounting for about 80% of the total circRNAs (Figure 1C). Majority of the circRNAs were on chromosomes 1–19 (Figure 1D). The size of circRNAs varied greatly, ranging from 128 nucleotides to more than 2000 nucleotides (Figure 1E). The total mean length was 3453 nt. Of the 2780 circRNAs identified; 1476 were detected only in AQP5 +/+ mice and 375 were checked only in AQP5 −/− mice (Figure 1F). 30 differentially expressed circRNAs were found in AQP5 −/− mice, of which 24 were downregulated and 6 were upregulated. Hierarchical clustering indicated significant differences in the expressions of circRNAs between AQP5 +/+ and AQP5 −/− mice (Figure 1G). The downregulation of 12 circRNAs was verified by Quantitative real‐time polymerase chain reaction (Figure 1H). We performed pathway enrichment analyses using GO and the KEGG. Cellular macromolecule metabolic process was the most abundant GO terms for downregulated circRNAs (Figure 2A), nucleus (Figure 2B) and organic cyclic compound binding (Figure 2C). The most relevant pathway was the MAPK signalling pathway (Figure 2D). A total of 15,792 mRNAs were found in the mouse lenses, of which 1956 were expressed in AQP5 +/+ mice, and 187 were expressed in AQP5 −/− mice (Figure 3A). 1214 differentially expressed mRNAs were screened from AQP5 −/− mice, of which 14 were significantly upregulated and 1200 were remarkably downregulated more than 2‐fold. Hierarchical cluster analysis showed that the mRNA expression profiles of the lens of both types of mice were significantly different (Figure 3B). The most abundant GO term for downregulated mRNAs were in response to metabolic processes (Figure 3C), membrane‐bounded organelle (Figure 3D) and binding (Figure 3E). KEGG analysis showed that the most relevant pathway of downregulated mRNAs was the lysosome (|log2 FC| ≥ 1 and p‐value < 0.05) (Figure 3F). The volcano plot showed that the differential expression of mRNAs in AQP5 +/+and AQP5 −/− mice was significant (|log2 FC| ≥ 1 and p‐value < 0.05) (Figure 3G). The scatter plot was built to assess the expression variation of mRNAs between the two groups (|log2 FC| ≥ 1 and p‐value < 0.05) (Figure 3H). A total of 12 circRNAs with reduced expression were selected as predictive miRNA binding sites. The method of mRNA selection was the same as that of circRNA screening, and 29 lysosomal associated mRNAs with decreased expression in microarray results were screened in the AQP5 −/− group. The 29 miRNAs were identified by TargetScan and miRanda. All Target circRNAs, miRNAs and mRNAs were selected, and then the network relationship diagram between circRNA‐miRNA and miRNA‐mRNA association pairs was constructed (Figure 4). A qRT‐PCR experiment was performed on 10 mRNAs of the lysosomal pathway (membrane proteins: Lamp1, Cln3, Cln5, Hgsnat and Lipaf; enzymes: Ctsb, Gla, Lipa, Gm2a, and Npc2). It showed that the relative expression levels of the 10 mRNAs were downregulated in AQP5 −/− group (Figure 5A). The RIP‐qPCR experiment showed that the enrichment of Chr4:150439343‐150534945+, Chr18:12871078‐12898301‐, Chr3: 59031546‐59042413+, Chr2:140042094‐140057499‐ and Chr6: 119920110‐119921028+ were elevated in the anti‐Ago2 group (Figure 5B). RIP‐qPCR experiment also showed that the enrichment of Lamp1, Cln3, Cln5, Litaf, Ctsb and Lipa was elevated in the anti‐Ago2 group (p < 0.05) (Figure 5C). The expression of lysosomal proteins Lamp1, Cln5, Gla and Lipa were less in the lenses of AQP5 −/− mice than that of AQP5+/+ mice (Figure 5D–F). Transmission electron microscopy showed that the structure of organelles in LEC of AQP5 +/+ mice were generally normal. However, an autophagic lysosome (ASS) was found in AQP5 −/− mice (Figure 6A). The results of immunofluorescence staining showed that LC3II/I and p62 were widely expressed in mouse lens epithelial and fibre cells (Figure 6B). In addition, the fluorescence of LC3II/I and p62 in the primary cultured lens epithelial cells of AQP5 −/− mice seemed to be brighter than that of AQP5 +/+ mice (Figure 6C). To quantitatively compare the expression levels of LC3II/I and p62, we performed a Western blot assay. LC3II/I and P62 were greatly increased in the lenses of AQP5 −/− mice compared with the AQP5 +/+ mice (Figure 6D,E). QRT‐PCR experiment showed that Chr16: 33421321‐33468183+ was decreased in AQP5 −/− mice (p < 0.05). RIP‐qPCR experiment showed that the enrichment of chr16:33421321‐33468183+ was elevated in the anti‐Ago2 group (p < 0.05) (Figure 7A). qRT‐PCR experiment also showed that HSF4 expression was decreased in AQP5 −/− mice (p < 0.05). RIP‐qPCR experiment showed that enrichment of HSF4 was elevated in the anti‐Ago2 group (p < 0.05) (Figure 7B). QRT‐PCR experiment showed that miR‐149‐5p was increased in the lens of AQP5 −/− mice (p < 0.05) (Figure 7C). With the aim of further exploring the mitigating effect of the downstream regulatory mechanism of HSF4 on lens opacity, the upstream miRNAs of HSF4 were screened by the TargetScan database. It revealed that miR‐149‐5p had better prediction results. Therefore, miR‐149‐5p was selected for further studies. Three mutation sites were designed according to the predicted three binding sites (Figure 7D). A luciferase reporter assay showed that, in the presence of HSF4‐WT‐UTR and miR‐149‐5p mimics, the relative luciferase activity was significantly antagonized (Figure 7E). Mutations of the miR‐149‐5p complementary sites in the 3′UTR of HSF4 (HSF4M1, HSF4M3, HSF4M3) abolished the suppressive effect of miR‐149‐5p through the disruption of the interaction between miR‐149‐5p and HSF4 (Figure 7E). The analysis showed that a regulatory pattern diagram of HSF4 was derived. Chr16: 33421321‐33468183+ probably adsorbed to miR‐149‐5p through the sponge mechanism, and miR‐149‐5p unidirectional regulate HSF4 expression finally targeting regulation of lysosomal mRNAs (Figure 7F). Thirteen human AQP subtypes have been identified. Current studies have shown that the functional differences in water permeability in different tissues and cells may be related to transcriptional regulation, post‐translational modification, protein stability and polarized membrane distribution among different AQP subtypes. AQP0, AQP1, and AQP5 are conserved in all mammalian lens examined thus far. , Seven AQP0 mutations that can cause cataract have been found, all of which lead to the inability of AQP0 to traffic to the plasma membrane. , These mutations include Arg233Lys, Arg33Cys, Asp150His, Glu134Gly, Thr138Arg, as well as C‐terminal truncation mutants, Δ213‐AQP0 and Tyr219stop. The expression of AQP1 increased gradually after birth, which was consistent with the increase in lens size during growth and development. The water permeability of lens epithelium in AQP1−/− mice was approximately three times lower. AQP5 is expressed in cornea, lacrimal gland, lens, lung pneumocyte type I cells, retina, salivary gland, pancreas and uterus. AQP5 is less abundant, but the water permeability of AQP5 is 20 times that of AQP0. We previously found that the AQP5 −/− mice developed lens opacity, which increased with age. In this study, we found that the number of autophagic lysosome in the lens epithelial cells of AQP5 −/− mice was more than the AQP5 +/+ mice. As development proceeds, the differentiation of fibre cells is accompanied by degrading membranous organelles such as nuclei, mitochondria, Golgi apparatus and endoplasmic reticulum, form organelle‐free zones, namely organelle‐free zones to achieve optical transmittance. Autophagosomes are elucidated from immunoelectron microscopy findings in lens epithelial cells. Later studies revealed autophagosomes existed in lens epithelial and fibre cells of mice, chicken and humans. P62 serves as a scaffold having binding sites for both ubiquitin and LC3B. LC3B and p62 are degraded by hydrolysis enzymes of lysosomes along with the cargo. , , The correlation between autophagy and lens development and cataract formation has been reported, but there are few studies on the relationship between AQP5 and autophagy in lens development. In the study, we found that extended outflow blockage led to extensive accumulation of p62 protein in the lens of AQP5−/− mice (Figure 6B–E), while the expression of Lamp1, Lipa, Gla and Cln5 decreased than those of AQP5 +/+ mice through Western blot and immunofluorescence staining (Figure 5D–F). The same result was found in the primary LEC. There is no evidence of how AQP5 causes the change of autophagy level. The relationship between AQP5 and autophagy needs to be clarified. CircRNAs, a novel class of noncoding RNAs, , can regulate target genes through miRNA sponge or RNA‐binding proteins. , CircRNAs participates in pathological process such as apoptosis, proliferation, activity and oxidative damage and may eventually cause cataract. Experimental studies have shown that they could be specifically several miRNAs, like miR‐15a, miR‐23b‐3p, miR‐34a‐5p, miR‐184 and miR‐211‐5p and regulate the apoptosis or oxidative stress of lens epithelial cells in the formation of catract. , The regulation of target mRNAs by miRNA is usually achieved by guiding the degradation or inhibiting their translation. More and more evidence show that circRNAs are involved in the occurrence, pathogenesis and progression of various ocular disorders. In our study, we found 12 downregulated circRNAs, 29 upregulated miRNAs and 29 downregulated mRNAs constructed a circRNA‐miRNA‐mRNA regulatory network. RIP‐qPCR experiments were used to prove the reliability of the network. For example, Chr4: 150439343‐150534945+, Chr18: 12871078‐12898301‐, Chr3: 59031546‐59042413+, Chr2: 140042094‐140057499‐ and Chr6: 119920110‐119921028+ were upregulated in RIP‐qPCR experiments. mRNAs (Lamp1, Cln3, Cln5, Litaf, Ctsb, Lipa) were also upregulated in RIP‐qPCR. Heat shock factor, a regulator of heat shock response, maintains the stability of the intracellular environment by protecting cells from environmental stress or stress related to cell proliferation and differentiation. Heat shock transcription factor 4 (HSF4) is closely associated with lens development. , It controls the expression of heat shock proteins (Hsps), alpha β‐crystallin and γ‐crystallin in lens tissue. Mutations of HSF4 were associated with autosomal dominant cataracts. In lens, lysosomes are involved in maintaining the homeostasis of LEC and the terminal differentiation of fibre cell. HSF4 may participate in protein and nuclear DNA quality control by regulating the alpha β‐crystallin‐associated lysosomal pathway. In this study, HSF4 was decreased in the lens of AQP5 −/− mice compared with that of AQP5 +/+ mice (Figure 7B). In addition, the expressions of Lamp1, Lipa, Gla and Cln5 were decreased in the lens and the primary lens epithelial of AQP5 −/− mice (Figure 6D–F). Through the transcription factor prediction assay, we found that HSF4, as a transcription factor, regulated many mRNAs in lysosomal pathways (Figure 4). Finally, it was found that HSF4 was a target gene of miR‐149‐5p (Figure 7A–C). In bioinformatics analysis Chr16: 33421321‐33468183+ can inhibit miR‐149‐5p by sponge mechanism (Figure 7D,E). Chr16: 33421321‐33468183+ may regulate lysosome associated mRNAs by targeting the miR‐149‐5p/HSF4 axis (Figure 7F). The present study discovered that deficiency of AQP5 causes early onset cataract in mice. AQP5 knockout may be related to altering the level of autophagy in the lens, in addition circRNA levels were significantly changed in the lens of AQP5 −/− mice. The data presented here indicate that AQP5 may coordinate downstream regulatory events through circRNA‐miRNA‐mRNA network and HSF4‐mediated lysosome expression, which participated in the pathogenesis of abnormal lens development. Hu Shaohua: Methodology (equal); writing – original draft (equal); writing – review and editing (equal). Wang Yihui: Project administration (equal); writing – original draft (equal). Zhang Kaier: Investigation (equal); methodology (equal). Bai Ying: Methodology (equal). Wang Xiaoyi: Investigation (equal); validation (equal). Zhao Hui: Data curation (equal); supervision (equal). Di Guohu: Data curation (equal); writing – original draft (equal); writing – review and editing (equal). Chen Peng: Funding acquisition (lead); project administration (lead). The authors declared that they have no conflict of interest. Click here for additional data file. Click here for additional data file.