Assessment of relationships between mitochondrial proteins and ischemic stroke: a bidirectional Mendelian randomization study

Abstract

Background Increasing evidence suggests an association between mitochondrial function and ischemic stroke (IS). However, whether this association might be causal or explained by reverse causal association/residual confounding is unclear. Therefore, we designed this study to evaluate the causal association of mitochondrial function with IS risk. Methods Mitochondrial proteins were considered the exposure factor, and IS was considered the outcome variable. Exposures and outcomes were obtained from the IEU Open GWAS database. First, we obtained 66 mitochondrial protein genome-wide association studies data sets from European populations, as well as IS data. We then performed two-sample Mendelian randomization (MR) analysis to determine associations between mitochondrial proteins and IS. We additionally performed bidirectional MR analysis to examine the directions of the causal associations. Results IVW indicated that three mitochondrial proteins were associated with IS: ribosome-recycling factor (mtRRF) was negatively associated with IS [OR = 0.93, 95%CI (0.88–0.98), P = 0.005]; malonyl-CoA decarboxylase (MLYCD) was negatively associated with IS [OR = 0.89, 95%CI (0.82–0.97), P = 0.005]; and mitochondrial Lon protease homolog (LONP1) was positively associated with IS [OR = 1.06, 95%CI (1.02–1.10), P = 0.004]. Sensitivity analysis indicated no evidence of reverse causality, pleiotropy, or heterogeneity, thus suggesting that MR was an effective method for causal inference in this study. Conclusion Our MR analysis indicated that three mitochondrial proteins are causally associated with IS, and may aid in early detection and prevention of IS at the microscopic molecular level. Our findings provide new insights into IS microscopic mechanisms and clinical research.