Identification and validation of potential biomarkers related to oxidative stress and inflammation in ischemic stroke

Previous studies speculated that oxidative stress and inflammation might affect the pathogenesis of ischemic stroke (IS). This study explored the utility of oxidative stress (OS)-related genes (OSRGs) and inflammation-related genes (IRGs) in IS. IS and control samples were downloaded from public databases, and IRGs and OSRGs were obtained from previous research. The differentially expressed genes (DEGs) between the IS and control groups were identified by differential expression analysis. Subsequently, candidate genes were identified utilizing the intersection of DEGs, IRGs, and OSRGs. Then, machine learning algorithms and gene expression verification were utilized to obtain potential biomarkers. Furthermore, a nomogram was constructed, and gene set enrichment analysis (GSEA) was performed. Molecular regulatory network construction, drug prediction, and molecular docking were subsequently performed. Finally, the expression of the potential biomarkers in clinical samples was evaluated by reverse transcription quantitative polymerase chain reaction (RT-qPCR), and a mouse single-cell RNA sequencing (scRNA-seq) dataset was used as a complementary exploratory analysis to examine their cell type-specific expression patterns and identify candidate cell populations potentially associated with IS. In total, 596 DEGs were identified, and 6 candidate genes were highlighted. Ultimately, TNFSF10 , TLR2 , ADM , ABCA1 , and HIF1A were obtained as potential biomarkers. The nomogram had good predictive ability for IS. The GSEA results illustrated that all potential biomarkers excluding HIF1A were enriched in complement and coagulation cascades. The molecular regulatory network included 146 transcription factors, such as MYC . The microRNAs related to potential biomarkers included miR-6809-3p and miR-1276. Moreover, the drugs with high affinity for the potential biomarkers included diprovocim-1 and trichostatin A. Finally, the RT-qPCR results were consistent with the bioinformatic findings, while analysis of the mouse scRNA-seq dataset suggested that microglia and central nervous system-associated macrophages (CAMs) may represent candidate cell populations associated with the identified biomarkers. This study identified five potential biomarkers ( TNFSF10 , TLR2 , ADM , ABCA1 , and HIF1A ) with potentially significant effects on IS.