Genetic analysis of diagnostic and therapeutic potential for ferroptosis in postoperative sepsis

Ferroptosis is a new form of iron-dependent cell death that is closely associated with sepsis. However, few studies have investigated the diagnostic and therapeutic potential for ferroptosis-related genes (FRGs) among postoperative sepsis.

The GSE131761 dataset was used to identify differentially expressed FRGs (DE-FRGs). KEGG and GO analyses were subsequently performed. LASSO and SVM-RFE methods were applied for identifying genetic biomarkers for sepsis. Gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA) were applied for exploring the biological properties of the DEGs. CIBERSORT was applied to analyse immune cell infiltration. DGldb was employed for predicting potential target drugs for the DEGs. Competing endogenous RNA (ceRNA) networks were constructed to analyse the regulatory patterns of the DEGs. The expression of hub genes was validated based on GSE26440 dataset. The bioinformatics analysis was carried out with R software (version 4.1.2). Blood from sepsis patients and healthy controls was collected and the expression of hub genes was experimentally verified by real-time quantitative polymerase chain reaction (RT-qPCR).

38 sepsis-associated DE-FRGs were assessed via Gene Expression Omnibus (GEO) and Ferroptosis database (FerrDb), and the gene function analysis showed that they were closely related to inflammatory response and autophagy regulation. Subsequently, SVM-RFE and LASSO methods determined 7 marker genes. GSEA suggested that these marker genes may be involved in regulating several biological pathways. Furthermore, 52 gene-targeted drugs were identified in this study, the vast majority of which were associated with MAPK14. CIBERSORT analysis suggested that SLC38A1, MGST1, and MAPK14 may be involved in immune microenvironment alterations. We revealed the potential complex regulatory relationship by constructing a ceRNA network based on marker genes. Finally, 6 genes were validated in the validation set, with 5 of them further confirmed through RT-qPCR.

Seven genes associated with ferroptosis are screened from postoperative sepsis samples. The expression of these genes has high diagnostic validity for sepsis and may serve as potential diagnostic biomarkers. This study gives an entrance point to uncover the underlying mechanisms of sepsis.