Genetic Associations of Clonal Hematopoiesis With Cardioembolic Stroke: Insights From Genome-Wide Mendelian Randomization, Bulk RNA, Single-Cell RNA Sequencing.
Ischemic stroke (IS), a major global health concern, is associated with aging-related clonal hematopoiesis of indeterminate potential (CHIP), though their mechanistic connection remains unclear. This study explores the causal CHIP-IS relationship, key genetic drivers, and potential therapies.
Genetic markers for CHIP were selected as instrumental variables and analyzed through bidirectional two-sample Mendelian randomization (MR) using GWAS data from IS cohorts. Functional annotation of significant loci was performed via FUMA, while transcriptomic datasets from GEO underwent differential expression analysis, weighted gene co-expression network construction, and machine learning-driven biomarker discovery. Protein-protein interaction networks and single-cell RNA sequencing (scRNA-seq) were employed to elucidate cellular mechanisms.
MR analysis revealed a significant causal association between CHIP and cardioembolic stroke (CES) risk (OR = 70.15, 95% CI = 2.03-2428.52, p = 0.02). PARP1 and CD3G emerged as hub genes connecting CHIP to IS pathogenesis, validated through multi-omics integration. Fourteen feature genes were identified, and potential therapeutic drugs targeting this pathway were discovered. scRNA-seq analysis further demonstrated downregulation of CD3G in T cells post-IS, disrupting immune cell communication and differentiation.
This study provides robust genetic evidence for CHIP-mediated predisposition to CES and identifies PARP1 and CD3G as critical therapeutic targets. The integration of machine learning and single-cell genomics offers novel insights into immune dysregulation in IS, paving the way for precision prevention strategies in CHIP patients.
Genetic markers for CHIP were selected as instrumental variables and analyzed through bidirectional two-sample Mendelian randomization (MR) using GWAS data from IS cohorts. Functional annotation of significant loci was performed via FUMA, while transcriptomic datasets from GEO underwent differential expression analysis, weighted gene co-expression network construction, and machine learning-driven biomarker discovery. Protein-protein interaction networks and single-cell RNA sequencing (scRNA-seq) were employed to elucidate cellular mechanisms.
MR analysis revealed a significant causal association between CHIP and cardioembolic stroke (CES) risk (OR = 70.15, 95% CI = 2.03-2428.52, p = 0.02). PARP1 and CD3G emerged as hub genes connecting CHIP to IS pathogenesis, validated through multi-omics integration. Fourteen feature genes were identified, and potential therapeutic drugs targeting this pathway were discovered. scRNA-seq analysis further demonstrated downregulation of CD3G in T cells post-IS, disrupting immune cell communication and differentiation.
This study provides robust genetic evidence for CHIP-mediated predisposition to CES and identifies PARP1 and CD3G as critical therapeutic targets. The integration of machine learning and single-cell genomics offers novel insights into immune dysregulation in IS, paving the way for precision prevention strategies in CHIP patients.