Screening and validation of key genes for pediatric sepsis based on bioinformatics analysis

Background: Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection, this study conducted bioinformatics analysis on sepsis-related data sets to identify potential biomarkers for pediatric sepsis. Methods: We obtained sepsis-related datasets (GSE145227 and GSE236713) from Gene Expression Omnibus (GEO) and screened for differentially expressed genes (DEG). Functional enrichment analysis of DEGs was performed using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). STRING was used to create protein-protein interaction networks (PPI), and core genes were screened using the Cytohubba plugin in Cytoscape software. Validation in an external dataset (GSE95233) was performed to obtain key genes. Subsequently, validation was performed in paediatric sepsis patients recruited from our hospital. Results: Gene bioinformatics analysis screened nine key genes, namely MPO, LTF, ELANE, DEFA4, CEACAM8, MMP8, MMP9, PRTN3 and TLR4. In the external validation dataset of GSE95233, MPO, LTF, ELANE, DEFA4, CEACAM8, MMP8, MMP9, PRTN3 and TLR4 were expressed at higher mRNA levels than in healthy controls. Real-time quantitative polymerase chain reaction (qRT-PCR) showed that LTF, ELANE, DEFA4, PRTN3, and TLR4 were significantly up-regulated in the peripheral blood of patients with pediatric sepsis compared to controls. Enzyme-linked immunosorbent assay (ELISA) showed that serum levels of PRTN3 in the group of pediatric sepsis patients [319.05 (188.45, 499.49ng/mL)] were significantly higher than those of the control group [137.44 (125.68 ± 281.66ng/mL)]. Conclusion: The five key genes: LTF, ELANE, DEFA4, PRTN3 and TLR4 may be potential diagnostic biomarkers for pediatric sepsis.