Abstract
Sepsis is one of the leading causes of death among seriously ill patients worldwide, affecting more than 30 million people annually and accounting for 1-2% of hospitalizations. By analyzing gene expression omnibus (GEO) data set, our team explored the relationship between m6A methylation gene and poor prognosis of sepsis. The purpose of this present study is to examine new detection markers for patients with poor prognosis, provide theoretical basis for timely intervention and improve the survival rate of patients. First, GSE54514 transcriptome data were extracted from the GEO database 31 patients with sepsis related death and 72 sepsis survivors. Key genes were screened from differentially expressed genes (DEGs), least absolute shrinkage and selection operator (LSAAO) and random forest (RF). And then, METTL3, WTAP and RBM15 were further verified by quantitative reverse transcription PCR (qRT-PCR). The constructed nomogram model showed high accuracy in predicting death. These three genes are mainly involved in chemokine signaling pathway, differentiation of monocytes and T cells, and phagocytosis of immune cells. The analysis showed that a high m6A score subtype is linked to lower immunosuppression and higher survival rates in clinical samples, suggesting better immune responses and outcomes for these patients. Finally, the protective effect of METTL3 in sepsis was demonstrated in mouse sepsis model applied with METTL3 inhibitor, by conducting cell flow cytometry analysis, enzyme-linked immunosorbent assay (ELISA) and hematoxylin-eosin (HE) staining. In conclusion, these findings provide potential biomarkers and targets for early precision diagnosis and treatment.