Diagnostic and Predictive Values of Ferroptosis-Related Genes in Child Sepsis

Front Immunol. 2022 Jun 30:13:881914. doi: 10.3389/fimmu.2022.881914. eCollection 2022.

Abstract

Background: Early diagnosis of sepsis in children was essential to reducing mortality. This study aimed to explore the value of ferroptosis-related genes in children with sepsis.

Methods: We screened the septic children microarray dataset from the GEO database and analyzed the ferroptosis-related differentially expressed genes (DEGs). A functional analysis of ferroptosis-related DEGs was performed. The protein-protein interaction network was used to identify hub genes. We explored the immune landscape of sepsis and controls. The value of hub genes in diagnosing sepsis was tested in the training (GSE26440) and validation sets (GSE13904), and ELISA was used to verify their diagnostic value in children with sepsis in our hospital.

Results: A total of 2,103 DEGs in GSE26440 were obtained, of which ferroptosis-related DEGs were 34. Enrichment analysis showed significant enrichment in the ferroptosis and hypoxia pathways (i.e., HIF-1 pathway). The top three genes (HMOX1, MAPK14, TLR4) were selected as hub genes. Immunological analysis suggested that 10 cell types (i.e., CD8/CD4 T cells) were lower in sepsis. Immune checkpoint-related genes CD274 (PD-L1), HAVCR2 (TIM3), and SIGLEC15 were overexpressed in sepsis. The AUROC for the diagnosis of sepsis for HMOX1 and TLR4 ranged from 0.77 to 0.81, while the AUROC of MAPK14 reached 0.935 and 0.941 in the training and validation sets. Serum ELISA results of HMOX1 and TLR4 showed no significant difference in differentiating sepsis. The AUROC of MAPK14 was 0.877. When the diagnostic threshold was 74.852 ng/ml, the sensitivity and specificity were 0.906 and 0.719, respectively.

Conclusion: Ferroptosis-related gene MAPK14 is of considerable value in the early diagnosis of sepsis in children.

Keywords: MAPK14; children; ferroptosis; immune landscape; sepsis.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Child
  • Computational Biology / methods
  • Ferroptosis* / genetics
  • Gene Ontology
  • Humans
  • Mitogen-Activated Protein Kinase 14*
  • Sepsis* / diagnosis
  • Sepsis* / genetics
  • Toll-Like Receptor 4

Substances

  • Toll-Like Receptor 4
  • Mitogen-Activated Protein Kinase 14