Background Gestational diabetes mellitus (GDM) is a pregnancy−related disorder characterized by inflammatory dysregulation that disrupts maternal–fetal immune homeostasis, yet the contribution of mitochondrial dysfunction to this pro−inflammatory state remains incompletely understood. Methods This study combined transcriptomic data obtained from the GEO repository and mitochondrial gene lists from MitoCarta3.0 to pinpoint mitochondrial-related genes (Mito-RGs) exhibiting differential expression in GDM. Machine learning algorithms, including the least absolute shrinkage and selection operator (LASSO), random forest (RF), and extreme gradient boosting (XGBoost), were applied to identify hub Mito-RGs. Gene set variation analysis (GSVA) and gene set enrichment analysis (GSEA) were performed to identify enriched pathways in various cell types. A predictive nomogram for GDM was developed based on Mito-RGs scores. Experimental validation was conducted in human placental tissues and a GDM mouse model to confirm hub gene expression. Results DHRS2, STX17, and TIMM44 were identified as hub Mito-RGs involved in GDM. Scores based on these genes formed the basis of a nomogram with strong predictive performance for GDM. Single-cell RNA sequencing data indicated that GDM placental tissues exhibited higher proportions of epithelial cells, macrophages, and NK cells, alongside a significant reduction in tissue stem cells. Glycolysis and hypoxia-related pathways were enriched in epithelial and stem cells, whereas inflammatory and immune−activation pathways were predominantly enriched in macrophages, indicating pro−inflammatory remodeling of the placental immune microenvironment. Immunohistochemistry confirmed significantly elevated DHRS2 protein levels in placentas from GDM patients and GDM mouse models. Conclusions These findings emphasize the critical impact of mitochondrial dysfunction on the pro−inflammatory reprogramming of the placental immune microenvironment in GDM, providing potential targets for anti−inflammatory and immunometabolic interventions.
Zhao et al. (Thu,) studied this question.