Endocrine Abstracts

Yu Pei and Sanjiv Risal are joint first authors. Elisabet Stener-Victorin and Qiaolin Deng are joint corresponding authors. Polycystic ovary syndrome (PCOS) affects around 15% of women of reproductive age and the key feature of the syndrome is hyperandrogenism. To understand the complex pathophysiology of PCOS, more than 30 PCOS-like animal models have been developed to mimic certain pathophysiological features. So far, not much is known about the common and unique molecular and cellular features among the most used PCOS-like mouse models. The overall aim of this study is to compare three hyperandrogenic PCOS-like mice models that are commonly used to understand the molecular pathology across inflicted tissues: hypothalamus, subcutaneous adipose tissue, ovary, and metaphase II (MII) oocytes among the prenatal androgenized (PNA) model (F1 offspring), the prepubertal androgenized (PPA) model, and the theca-cell specific nerve growth factor overexpressing mouse model (17 NF). We performed bulk RNA sequencing of tissues and single-cell RNA sequencing of MII oocytes across the models and identified differential expressed genes (DEGs) in each tissue and model. The greatest number of DEGs was found in the ovary, followed by adipose tissue whereas hypothalamus is least affected in all models. In addition, ovary and adipose tissue were most affected in PPA model compared to other models. We found several common DEGs were involved in lipid metabolism and steroid hormone biogenies among ovary, adipose tissue and hypothalamus in all models. Moreover, we conducted weighted gene correlation network analysis (WGCNA) method to identify functional correlated gene modules across all models and revealed common biological pathways for hub genes including gonadal development, cell-cell communication, hormonal metabolism and lipid metabolism in ovary and adipose tissue. Notably, greatest transcriptional alteration was also observed in MII oocytes in the PNA model, with DEGs in gonad development and germ cell development, likely due to fetal programming effects. Currently, we are investigating crosstalk between tissues and oocytes and comparing these finding with relevant human tissues.