Integrating Transcriptomic Data and Mendelian Randomization Analyses Reveals Potentially Novel Sepsis-related Targets.

Background: Sepsis remains a leading cause of global morbidity and mortality.

Objective: To identify candidate biomarkers that may be mechanistically related to the pathogenesis of sepsis.

Methods: The Gene Expression Omnibus database was leveraged to identify differentially expressed genes (DEGs) between the healthy control and septicemia groups. Genes causally related to sepsis were probed through the integration of GWAS and expression quantitative trait loci (eQTL) data in a two-sample Mendelian randomization (MR) analysis. A set of key sepsis-related genes was then selected based on the overlap between these putative causal genes and the DEGs. These genes were then subjected to enrichment analyses, testing set validation, and analyses of their expression dynamics in clinical samples.

Results: An examination of the overlap between 228 sepsis-related DEGs identified in the training dataset and 275 candidate causal genes linked to sepsis derived from the MR analysis led to the selection of four overlapping (SLC22A15, IL5RA, HDC, and SLC46A2) that may play a key role in sepsis. Enrichment analyses indicated that these genes were involved in the regulation of histidine metabolism and immune/inflammatory responses. In immune cell infiltration analyses, these genes were positively correlated with inflammatory response activation and the suppression of adaptive immunity. Consistent findings were obtained through qPCR verification in clinical samples.

Conclusion: These results offer potential insight into the mechanisms that govern septicemia and thus suggest a promising series of candidates that may be amenable to targeting to prevent or treat sepsis more effectively.