Identification of the key ferroptosis-related genes involved in sepsis progression and experimental validation in vivo

Background: Ferroptosis has a vital role in sepsis, but the mechanism is not known. Understanding the mechanism of ferroptosis during sepsis will aid in developing improved therapeutic strategies. Methods: We used the Gene Expression Omnibus database and FerrDb database to obtain ferroptosis-related differentially expressed genes (DEGs) between sepsis patients and healthy volunteers (HVs). Analyses of PPI networks, functional enrichment, as well as use of the MCODE algorithm were used to identify key ferroptosis-related DEGs. Expression of key ferroptosis-related DEGs was verified using: GSE57065 and GSE65682 datasets; rats in which ferroptosis was induced with erastin; sepsis-induced acute lung injury (siALI) rats. The effects of acupoint catgut embedding (ACE) on ferroptosis and expression of key ferroptosis-related DEGs in the lungs of siALI rats were also observed. A Cox proportional hazard model was used to verify the effect of key ferroptosis-related DEGs on the survival of sepsis patients. Cytoscape was used to construct ceRNA networks and gene–transcription factor networks. Results: Between sepsis patients and HVs, we identified 33 ferroptosis-related DEGs. According to analyses of PPI networks and the MCODE algorithm, we obtained four modules, of which the most significant module contained nine ferroptosis-related DEGs. Functional-enrichment analyses showed that four of the nine DEGs were enriched in the MAPK signaling pathway: MAPK14, VEGFA, TGFBR1, and DUSP1. We verified expression of these four genes in GSE57065 and GSE65682 datasets and ferroptosis rats. In addition, expression of these four genes and that of the oxidative-stress indicators GSSG and MDA was upregulated, and glutathione peroxidase-4 (GPX4) expression was downregulated, in siALI rats, but ACE reversed these changes. The Cox proportional hazard model showed that survival of sepsis patients in the high-risk group was shorter than that in the low-risk group. We found that the XIST−hsa-let-7b-5p−TGFBR1/DUSP1 ceRNA network and transcription factor E2F1 may be important regulators of these four DEGs. Conclusion: Our results suggest that MAPK14, VEGFA, TGFBR1, and DUSP1 may be key regulatory targets of ferroptosis in sepsis, and that ACE pretreatment may be antioxidant treatment for sepsis and alleviate ferroptosis. These findings provide a basis for further ferroptosis-related study in sepsis and provide new targets for its treatment.