Mendelian randomization analysis of causal associations between immune cell subsets and stroke and glioma risk with fine-mapping of candidate genes

Abstract

Post-operative cerebral infarction represents a severe complication following glioma surgery, yet the molecular mechanisms underlying this association and its potential relationship with tumorigenesis remain unclear. This study aims to explore the molecular connections between cerebral infarction and glioma through two-sample Mendelian randomization (2SMR) and summary-data-based Mendelian randomization (SMR) integrating peripheral blood expression quantitative trait loci (eQTL) data from the CAGE consortium, providing genetic evidence for understanding post-surgical ischemic complications. A two-sample Mendelian randomization (2SMR) design was employed to systematically evaluate causal associations between 731 immune cell subsets and glioblastoma (GBM) risk using GWAS data from the GWAS Catalog (GCST90001391 to GCST90002121) and FinnGen consortium. Summary-data based Mendelian randomization (SMR) integrated CAGE peripheral blood eQTL data with GWAS data for both glioma and cerebral infarction to identify shared risk genes. An oxygen–glucose deprivation (OGD) model was established in BV2 microglial cells, and qRT-PCR was used to detect expression changes of four key genes (HAX1, LIME1, TPM3, and CCL2) under ischemic stress conditions. After false discovery rate (FDR) correction, no immune cell subset showed a statistically significant association with GBM risk. SMR analysis identified 8 probe-gene associations significantly associated with glioma risk, all passing FDR correction and HEIDI test filtering. HAX1 and TPM3 demonstrated dual protective effects (glioma OR = 0.314 and 0.299; cerebral infarction OR = 0.342 and 0.456), while LIME1 enhanced risk for both diseases (glioma OR = 2.169; cerebral infarction OR = 2.341) and CCL2 increased cerebral infarction risk (OR = 1.654). BV2 cell OGD experiments confirmed moderate upregulation of protective genes HAX1 and TPM3 (3.45-fold and 4.22-fold, P < 0.001), contrasted with dramatic increases in risk genes LIME1 and CCL2 (11.62-fold and 9.00-fold, P < 0.0001). This study provides the first genetic and cellular evidence of shared molecular pathological foundations between cerebral infarction and glioma. Ischemic stress-triggered inflammatory cascades simultaneously promote acute stroke damage and create a pro-tumorigenic microenvironment.