Abstract / Summary
Background/objective: Blood transcriptomic responses to stroke may nominate genes relevant to myocardial infarction (MI), but association does not establish shared causality. This study prioritized stroke-associated blood transcriptomic candidates for MI and generated preliminary hypotheses on mechanisms and druggable targets. Methods: Three stroke blood transcriptomic datasets (GSE16561, GSE22255, GSE58294) were integrated, retaining one sample per patient, yielding 149 independent samples (82 stroke, 67 controls), with disease-protected ComBat correction. We applied weighted gene co-expression network analysis (WGCNA), cis-expression quantitative trait locus (eQTL)-based two-sample Mendelian randomization (MR) with Benjamini-Hochberg correction across 322 primary tests, colocalization, gene set enrichment analysis (GSEA), consensus clustering, and CIBERSORT deconvolution, plus network sensitivity analyses. External comparisons used mouse MI (GSE304090) and human brain endothelial oxygen-glucose deprivation (OGD) (GSE311086) datasets. Molecular docking and dynamics computationally prioritized compounds, including Compound Congrong Yi Zhi Capsule (CCYC) components. Experimental assessment used in vitro neonatal rat cardiomyocyte OGD. Results: WGCNA identified a stroke-associated brown module (685 genes); its eigengene correlated with stroke status (r = 0.287, 95% CI 0.133–0.428), but attenuation occurred after omitting GSE58294 (r = 0.141, 95% CI −0.054–0.325). Seventeen module genes showed nominal inverse-variance-weighted associations with MI, including TM9SF2, ROPN1L, NPL and MANSC1; none survived correction (smallest q = 0.086 for ROPN1L). Regional colocalization was inconclusive. GSEA implicated immune and metabolic pathways, including ferroptosis and lysosomal signaling; clustering suggested differential immune infiltration with residual cohort dependence. Docking predicted binding for chlortetracycline-TM9SF2 (−8.8 kcal/mol) and (R)-N-methylcoclaurine-TM9SF2 (−8.2 kcal/mol); single-trajectory dynamics suggested stable configurations for three targets. In OGD cardiomyocytes, 24-h (R)-N-methylcoclaurine treatment was associated with qPCR expression changes in candidate genes. Conclusions: This exploratory study prioritizes stroke-associated blood transcriptomic network genes with nominal MI genetic support and highlights immune/metabolic pathways as candidate mechanisms. Findings are hypothesis-generating, do not establish shared causality or validated target engagement, and require experimental and clinical validation. Further validation is essential.