Metabolic dysregulation plays a critical role in polycystic ovary syndrome (PCOS), yet the involvement of energy metabolism-related genes (ERGs) remains incompletely understood. This study employs integrative bioinformatics and experimental validation to identify key ERGs as potential biomarkers and therapeutic targets in PCOS. This study aimed to screen and validate ERGs associated with PCOS by integrating gene expression datasets (GSE34526 and GSE193123) and to explore their molecular mechanisms and therapeutic implications. Differentially expressed genes (DEGs) were identified from two GEO datasets and intersected with ERGs from the Reactome database. Functional enrichment, subcellular localization, transcriptional and post-transcriptional regulatory networks, and drug prediction analyses were performed. Clinical validation of candidate biomarkers was conducted using RT-qPCR on granulosa cells from PCOS patients and controls. A total of 159 common DEGs were identified, including three key biomarkers: CD44 (upregulated), HS2ST1, and GPC4 (downregulated). CD44 was localized primarily to the plasma membrane, while HS2ST1 and GPC4 were localized to the Golgi apparatus and nucleus/plasma membrane, respectively. Regulatory network analyses revealed multiple transcription factors, kinases, and miRNAs associated with these biomarkers. Drug prediction identified bisphenol A as a compound linked to all three genes. RT-qPCR confirmed significant upregulation of CD44 in clinical PCOS samples (P<0.05). This study identifies and validates three ERG-associated biomarkers with potential diagnostic and therapeutic utility in PCOS. The integration of bioinformatics and clinical validation underscores the translational relevance of these findings for advancing biotechnological applications in PCOS management.
Zhang et al. (Tue,) studied this question.