Integrated bioinformatic analysis and experimental validation identify IDH2 as a key mitochondrial energy metabolism-ferroptosis biomarker in ischemic stroke

Ischemic stroke (IS) is implicated in disruptions of mitochondrial energy metabolism and ferroptosis. However, the underlying mechanisms remain elusive. Three stroke-related transcriptome datasets (GSE22255, GSE58294, and GSE16561) were retrieved from the GEO database. A series of bioinformatic analyses were conducted, including differential expression analysis, functional enrichment, and protein–protein interaction (PPI) network construction. We further applied random forest modeling and single-cell sequencing to screen mitochondrial energy metabolism- and ferroptosis-related genes. The diagnostic value and immune correlations of these genes were further validated. We constructed an IS model in HT22 cells with OGD/R induction and employed IDH2 overexpression by transfection for subsequent functional validation. 8 differentially expressed genes (DEGs) were identified (MGST1, ACSL1, PDK4, IDH2, ECH1, GPX4, BCAT2, PARK7). Functional enrichment analysis revealed their involvement in pathways like pyruvate metabolism and immune response regulation. In the random forest model, IDH2 showed a promising diagnostic value in the external validation set (AUC = 0.682). Single-cell analysis identified specific expression of IDH2 in glial cells and MGST1 in astrocytes, with both genes showing a notable correlation to macrophage infiltration. Notably, IDH2 expression was markedly reduced in IS cell models, and its overexpression significantly improved mitochondrial energy metabolism and inhibited ferroptosis. This study elucidated the critical roles of mitochondrial energy metabolism–ferroptosis–related genes in IS and highlights IDH2 as a key candidate for disease diagnosis and potential therapeutic intervention. And its overexpression significantly improved mitochondrial energy metabolism and inhibited ferroptosis, suggesting a protective mechanistic role.