OR31-04 Androgen Signaling Regulates Female Fertility Through Modulation of H3K27me3 in Granulosa Cells

Abstract

Androgens play a major role in female fertility and in women’s health, in general. However, the underlying mechanism and downstream physiological targets of androgen are not fully understood. Androgen actions are mediated by “nuclear” transcriptional signals or “extra-nuclear” kinase actions. Primary androgen receptor (AR) target genes are those at which AR occupies a genomic androgen response element (ARE) and regulates gene transcription. Subsequently, androgens also regulate other genes in an AR-ARE independent fashion, involving membrane-initiated androgen signaling. In the last couple of years, we have reported that androgens may also influence gene expression through histone modifications. We have found that H3K27me3 (tri-methyl lysine 27 histone3) is a downstream target of androgen actions. H3K27me3 is a gene silencing mark, regulated by Enhancer of Zeste Homologue 2 (Ezh2), a histone methyltransferase that promotes tri-methylation of K27 and androgens inhibit the expression and activity of Ezh2. In this study, we report that androgens also regulate the expression of a histone demethylase called Jumonji domain containing protein 3 (JMJD3/KDM6B), that is responsible of removing the H3K27me3 mark. We find that in granulosa cells (GCs), androgen through the PI3K/Akt pathway, in a transcription-independent fashion, increases hypoxia-inducible factor 1 alpha (HIF1α) protein levels, which in turn induce JMJD3 expression. ChIP studies reveal increased binding of HIF1α on the JMJD3 promoter region with DHT treatment. To understand the global impact of androgens on ovarian gene expression and the contribution of androgen-induced decrease of H3K27me3, we have performed RNA-seq and ChIP-seq analysis with H3K27me3 antibody in primary mouse GCs treated with DHT or vehicle. Results show 190 significantly differentially expressed genes in DHT treated sample vs vehicle, out of which 129 and 61 genes were significantly up- and downregulated, respectively. Moreover, comparison of the RNA-seq and ChIP-seq data reveals that a number of upregulated genes have significantly lower H3K27me3 enrichment in the enhancer and/or promoter region in the DHT treated samples vs vehicle. This establishes that in the ovary, androgen-induced modulation of H3K27me3 mark through regulation Ezh2 and JMJD3 expression/activity is an important regulatory mechanism for ovarian gene expression. To delineate the physiological importance of JMJD3 in normal ovarian function and female fertility, we have also developed a GC-specific JMJD3 knockout mice. We find that GC-specific JMJD3KO mice are sub-fertile with longer estrous cycles and dysregulated follicular development. This phenotype is very similar to GC-specific ARKO mice. Thus, we propose that one of the critical actions of androgens in regulating follicular function and female fertility is through the regulation of JMJD3 expression.