5-Aza-2'-Deoxycytidine (5-Aza-dC, Decitabine) Inhibits Collagen Type I and III Expression in TGF-ß1-Treated Equine Endometrial Fibroblasts.

Endometrosis negatively affects endometrial function and fertility in mares, due to excessive deposition of type I (COL1) and type III (COL3) collagens. The pro-fibrotic transforming growth factor (TGF-ß1) induces myofibroblast differentiation, characterized by α-smooth muscle actin (α-SMA) expression, and collagen synthesis. In humans, fibrosis has been linked to epigenetic mechanisms. To the best of our knowledge, this has not been described in mare endometrium. Therefore, this study aimed to investigate the in vitro epigenetic regulation in TGF-ß1-treated mare endometrial fibroblasts and the use of 5-aza-2'-deoxycytidine (5-aza-dC), an epigenetic modifier, as a putative treatment option for endometrial fibrosis. Methods and Results: The in vitro effects of TGF-ß1 and of 5-aza-dC on DNA methyltransferases (DNMT1, DNMT3A, and DNMT3B), COL1A1, COL3A1, and α-SMA transcripts were analyzed in endometrial fibroblasts, and COL1 and COL3 secretion in a co-culture medium. TGF-ß1 upregulated DNMT3A transcripts and collagen secretion. In TGF-ß1-treated endometrial fibroblasts, DNA methylation inhibitor 5-aza-dC decreased collagen transcripts and secretion, but not α-SMA transcripts. Conclusion: These findings suggest a possible role of epigenetic mechanisms during equine endometrial fibrogenesis. The in vitro effect of 5-aza-dC on collagen reduction in TGF-ß1-treated fibroblasts highlights this epigenetic involvement. This may pave the way to different therapeutic approaches for endometrosis.

Metallopeptidades 2 and 9 genes epigenetically modulate equine endometrial fibrosis.

Endometrium type I (COL1) and III (COL3) collagen accumulation, periglandular fibrosis and mare infertility characterize endometrosis. Metalloproteinase-2 (MMP-2), MMP-9 and tissue inhibitors of metalloproteinases (TIMP-1 and TIMP-2) are involved in collagen turnover. Since epigenetic changes may control fibroproliferative diseases, we hypothesized that epigenetic mechanisms could modulate equine endometrosis. Epigenetic changes can be reversed and therefore extremely promising for therapeutic use. Methylation pattern analysis of a particular gene zone is used to detect epigenetic changes. DNA methylation commonly mediates gene repression. Thus, this study aimed to evaluate if the transcription of some genes involved in equine endometrosis was altered with endometrial fibrosis, and if the observed changes were epigenetically modulated, through DNA methylation analysis. Endometrial biopsies collected from cyclic mares were histologically classified (Kenney and Doig category I, n = 6; category IIA, n = 6; category IIB, n = 6 and category III, n = 6). Transcription of COL1A1, COL1A2, COL3A1, MMP2, MMP9, TIMP1, and TIMP2 genes and DNA methylation pattern by pyrosequencing of COL1A1, MMP2, MMP9, TIMP1 genes were evaluated. Both MMP2 and MMP9 transcripts decreased with fibrosis, when compared with healthy endometrium (category I) (P < 0.05). TIMP1 transcripts were higher in category III, when compared to category I endometrium (P < 0.05). No differences were found for COL1A1, COL1A2, COL3A1 and TIMP2 transcripts between endometrial categories. There were higher methylation levels of (i) COL1A1 in category IIB (P < 0.05) and III (P < 0.01), when compared to category I; (ii) MMP2 in category III, when compared to category I (P < 0.001) and IIA (P < 0.05); and (iii) MMP9 in category III, when compared to category I and IIA (P < 0.05). No differences in TIMP1 methylation levels were observed between endometrial categories. The hypermethylation of MMP2 and MMP9, but not of COL1A1 genes, occurred simultaneously with a decrease in their mRNA levels, with endometrial fibrosis, suggesting that this hypermethylation is responsible for repressing their transcription. Our results show that endometrosis is epigenetically modulated by anti-fibrotic genes (MMP2 and MMP9) inhibition, rather than fibrotic genes activation and therefore, might be promising targets for therapeutic use.

Collagen Type III as a Possible Blood Biomarker of Fibrosis in Equine Endometrium.

Collagen pathological deposition in equine endometrium (endometrosis) is responsible for infertility. Kenney and Doig's endometrial biopsy histopathological classification is the gold standard method for endometrosis evaluation, whereby blood biomarkers identification would be less invasive and could provide additional information regarding endometrosis diagnosis and fertility prognosis. This study aimed to identify blood biomarkers for endometrosis diagnosis (42 mares were used in experiment 1), and fertility assessment (50 mares were used in experiment 2). Reproductive examination, endometrial biopsy histopathological classification (Kenney and Doig) and blood collection were performed. Endometrium and serum collagen type I (COL1) and type III (COL3), and hydroxyproline concentrations were measured (ELISA). Serum COL3 cut-off value of 60.9 ng/mL allowed healthy endometria (category I) differentiation from endometria with degenerative/fibrotic lesions (categories IIA, IIB or III) with 100% specificity and 75.9% sensitivity. This cut-off value enabled category I + IIA differentiation from IIB + III (76% specificity, 81% sensitivity), and category III differentiation from others (65% specificity, 92.3% sensitivity). COL1 and hydroxyproline were not valid as blood biomarkers. Serum COL3 cut-off value of 146 ng/mL differentiated fertile from infertile mares (82.4% specificity, 55.6% sensitivity), and was not correlated with mares' age. Only COL3 may prove useful as a diagnostic aid in mares with endometrial fibrosis and as a fertility indicator.

Collagen and Microvascularization in Placentas From Young and Older Mares.

In older mares, increasing collagen fibers (fibrosis) in the endometrium and oviduct predisposes to sub-fertility and infertility. In this study, (i) gene transcription of collagen (qPCR: COL1A1, COL1A2, COL3A1, COL5A1); (ii) total collagen protein (hydroxyproline); (iii) collagen distribution (Picrosirius red staining; polarized light microscopy); and (iv) microvascular density (Periodic acid-Schiff staining), were evaluated in mares' placenta, and related to mares age, and placenta and neonate weights. Samples were collected from the gravid horn, non-gravid horn, and body of the placenta from younger (n = 7), and older mares (n = 9) of different breeds. Transcripts of COL1A1, COL3A1 and COL5A1, total collagen protein, chorionic plate connective tissue thickness, and microvascularization increased in the gravid horn of older mares' placentas, compared to the youngest (P < 0.05). Although in other species placenta fibrosis may indicate placental insufficiency and reduced neonate weight, this was not observed here. It appears that older fertile mares, with more parities, may develop a heavier, more vascularized functional placenta with more collagen, throughout a longer gestation, which enables the delivery of heavier foals. Thus, these features might represent morphological and physiological adaptations of older fertile mares' placentas to provide the appropriate nutrition to the equine fetus.

Collagens and DNA methyltransferases in mare endometrosis.

Inflammation and fibroproliferative diseases may be modulated by epigenetic changes. Therefore, we suggest that epigenetic mechanisms could be involved in equine endometrosis pathogenesis. DNA methylation is one of the methods to evaluate epigenetics, through the transcription of methyltransferases (DNMT1, DNMT3A, DNMT3B). The correlation between DNMTs and collagen (COL) transcripts was assessed for the different Kenney and Doig's (Current Therapy in Theriogenology. Philadelphia: WB Saunders; 1986) endometrium categories. Endometrial biopsies were randomly collected from cyclic mares. Histological classification (category I, n = 13; II A, n = 17; II B, n = 12; and III, n = 7) and evaluation of COL1A2, COL3A1 and DNMTs transcripts by qPCR, were performed. Data were analysed by one-way analysis of variance (ANOVA), Kruskal-Wallis test and Pearson correlation. As mares aged, there was an increase in endometrium fibrosis (p < .01), and in DNMT1 mRNA (p < .001). Considering DNMT3B transcripts for each category, there was an increase with fibrosis (p < .05). No changes were observed for DNMT1 and DNMT3A transcripts. However, DNMT3A mRNA levels were the highest in all categories (p < .01). In category I endometrium, a positive correlation was observed for transcripts of all DNMTs in both COLs (p < .01). In category IIA, this correlation was also maintained for all DNMTs transcripts in COL1A2 (p < .05), but only for DNMT3B in COL3A1 (p < .05). In category IIB, there was a positive correlation between DNMT3B and COL3A1 (p < .05). In category III, a positive correlation was only observed between DNMT3B and COL3A1 (p < .05). Our results suggest that there is a disturbance in COLs and DNMTs correlation during fibrosis.