LOXL4, CREB5 and steroid hormone biosynthesis pathways are involved in type 1 diabetes with polycystic ovary-like changes.

Type 1 Diabetes Mellitus (T1DM) is related to increased Polycystic Ovary Syndrome (PCOS) prevalence in women. However, the molecular mechanisms have not been fully elucidated. A meta-analysis was conducted to estimate the prevalence of PCOS in women with T1DM. Then a mouse model of T1DM with polycystic ovaries was established using streptozotocin (STZ). RNA sequencing (RNA-seq) was performed on ovaries to identify differentially expressed genes (DEGs) and enriched pathways. Machine learning algorithms (LASSO, Random Forest, and SVM-RFE) were applied to pinpoint key genes linking T1DM and polycystic ovary-like changes. Meta-analysis revealed a 25 % PCOS prevalence in T1DM (95 % CI: 0.19-0.32). T1DM mice exhibited reduced ovarian weight and size, decreased serum estradiol levels, and an increased number of antral follicles compared to controls. RNA-seq analysis revealed a significant enrichment of DEGs in the steroid hormone biosynthesis pathway, involving Cyp19a1 and Cyp17a1. Machine learning analysis highlighted LOXL4 and CREB5 as potential key mediators, with receiver operating characteristic (ROC) area under the curve (AUC) values of 0.724 and 0.771, respectively. This study highlights a key role for steroid hormone biosynthesis in T1DM-associated polycystic ovary-like changes through the identification of enriched DEGs. The identification of LOXL4 and CREB5 as novel mediators offers new insights into the molecular pathogenesis of T1DM-associated polycystic ovary-like changes.