148 FOLLICULAR FLUID microRNA SEQUENCES AS BIOMARKERS OF COMPETENT OOCYTES IN CATTLE

Abstract

Oocyte developmental competence is correlated with antral follicle count through ill-defined mechanisms. Oocytes from ovaries with fewer than 10 mid-antral follicles of 2 to 6 mm in diameter (low group) show reduced competence compared with those from ovaries with more than 10 follicles (high group). To unravel mechanisms underlying this phenomenon, this work explored the role of follicular fluid microRNAs (miRNAs, short non-coding RNAs regulating gene expression at the post-transcriptional level). A total of 3 pools of 300 µL of follicular fluid (FF) were collected from mid-antral follicles of low (L) and high (H) groups, respectively. Following miRNA extraction and library preparation, deep sequencing was carried out on Illumina HisEqn 2000 (Illumina Inc., San Diego, CA, USA). Differentially expressed miRNAs were identified with R package edgeR (http://bioconductor.org/packages/release/bioc/html/edgeR.html). Target genes of differentially expressed miRNAs were predicted with DIANA miRPath using homologous human miRNA and gene union options (P < 0.01). Gene ontology (GO) analysis was carried out by Cytoscape (http://www.cytoscape.org/) using a bovine database. In total, 1279 miRNAs were identified in FF: 805 ± 139 in L and 862 ± 36 in H (P > 0.05). We found that 27 miRNAs were differentially expressed (false discovery rate ≤0.001): 17 were up-regulated in L and 10 in H. Up-regulated miRNAs in L group were predicted to target 121 genes, 39 of which are specific for ovarian function (e.g. BCL2, FOXO3, KIT, TP53, and PTK2). The GO analysis indicated that these genes were kinases, anti-apoptotic and oncogenic factors, and enriched stress-activated MAPK cascade mediated by oxygen reactive species, G₁/S transition checkpoint, cellular response to interleukin-1, and negative regulation of cellular adhesion. Overexpressed miRNAs in H group were predicted to target 92 genes, 22 of which (e.g. MAPK, APC, JNK, PKA) are involved in folliculogenesis. These genes were represented by kinases, apoptotic and cytoskeleton remodelling factors, and enriched very important ovarian processes such as cell cycle, ephrin receptor, smoothened and phosphatidylinositol 3-kinase activity GO processes. Only 7 target genes were common between the 2 groups, 2 of which were important in ovarian functionality (CDKN1A and ITGA5). Interestingly, overexpressed miRNAs in both groups regulate several genes involved in processes apparently not related to folliculogenesis. Finally, regulation can be exerted also by low levels of specific miRNAs such as miR-320, which was reduced in L group and is known to be associated with premature ovarian senescence in women and decreased developmental potential of mouse oocytes. Our results indicate that the different oocyte quality is associated with a different miRNA blueprint, which may alter the expression of several genes relevant for intra- and extra-ovarian processes. Further studies will be necessary to determine if FF miRNAs can act outside the ovary and if their levels can be detected in the bloodstream thereby becoming possible noninvasive, real-time markers to determine oocyte quality in living animals.

This study was supported by FP7-KBBE-2012-FECUND-312097.