Mitochondrial related genome-wide mendelian randomization identifies putatively causal genes in the pathogenesis of sepsis

Background The dysfunction of mitochondria has been associated with the development of sepsis, but the specific mitochondrial-related genes and their roles in sepsis have not been fully elucidated. We employed Mendelian randomization and colocalization analysis to investigate the association between mitochondrial-related genes and sepsis by integrating multi-omics data. Methods Summary-level data on mitochondrial gene methylation, expression, and protein abundance levels were obtained from corresponding studies on methylation, expression, and protein quantitative trait loci, respectively. Genetic associations with sepsis were obtained from the GWAS catalog database. We utilized the MitoCarta3.0 database, which contains an updated list of 1,136 human mitochondrial genes, to identify mitochondrial genes. To assess the associations between mitochondrial gene-related molecular features and sepsis, we conducted summary-data-based Mendelian randomization analysis. Additionally, we performed colocalization analysis to determine whether the identified signal pairs shared a causal genetic variant. Findings After integrating the multi-omics data between mQTL-eQTL and eQTL-pQTL, we identified FIS1 as having tier 1 evidence for its association with sepsis. Methylation of cg01299997 in FIS1 was found to be associated with lower expression of FIS1, an increased risk of sepsis, and a positive role in cg01299997 methylation. Furthermore, NUDT2, IMMP2L, LYRM4, MRPL10, MRPL17, MTIF3, and TFAM genes were associated with sepsis risk with tier 2 evidence. Both gene expression and protein abundance levels of NUDT2 were observed to be associated with an increased risk of sepsis. Additionally, ATP5MC1 and VWA8 genes were associated with sepsis risk with tier 3 evidence. Among these tertiary genes, ATP5MC1 gene expression level showed a negative correlation (PPH4=0.9242), while the gene expression level of VWA8 exhibited a positive correlation (PPH4=0.7270). Interpretations We found that the mitochondrial FIS1, NUDT2, IMMP2L, LYRM4, MRPL10, MRPL17, MTIF3, TFAM, ATP5MC1 and VWA8 gene was putatively associated with sepsis risk with evidence from multi-omics levels. This study identified mitochondrial genes in relation to sepsis, which may enhance the understanding of the pathogenic mechanisms of sepsis development. Funding This work was supported by the Wuxi Health Commission Scientific Research Project [grant number No. Z202102].