Identification and validation of ferroptosis-related genes for diabetic retinopathy

Abstract

Diabetic retinopathy (DR) is a leading cause of blindness, and ferroptosis may be an essential component of the pathological process of DR. In this study, we aimed to screen five hub genes (TLR4, CAV1, HMOX1, TP53, and IL-1B) using bioinformatics analysis and experimentally verify their expression and effects on ferroptosis and cell function. The online Gene Expression Omnibus microarray expression profiling datasets GSE60436 and GSE1025485 were selected for investigation. Ferroptosis-related genes that might be differentially expressed in DR were identified. Then, Gene Ontology (GO) enrichment, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment, and protein-protein interaction (PPI) network analyses were conducted to characterize the differentially-expressed ferroptosis-related genes. After tissue-specific analyses and external dataset validation of hub genes, the mRNA and protein levels of hub genes in retinal microvascular endothelial cells (HRMECs) symbiotic with high glucose were verified using real-time quantitative PCR (qRT-PCR) and immunocytochemistry (ICC). Finally, hub genes were knocked down using siRNA, and changes in ferroptosis and cell function were observed. Based on the differential expression analysis, 19 ferroptosis-related genes were identified. GO and KEGG enrichment analyses showed that ferroptosis-related genes were significantly enriched in reactive oxygen species metabolic processes, necrotic cell death, hypoxia responses, iron ion responses, positive regulation of cell migration involved in sprouting angiogenesis, NF-kappa B signaling pathway, ferroptosis, fluid shear stress, and atherosclerosis. Subsequently, PPI network analysis and critical module construction were used to identify five hub genes. Based on bioinformatics analysis of mRNA microarrays, qRT-PCR confirmed higher mRNA expression of five genes in the DR model, and immunocytochemistry confirmed their higher protein expression. Finally, siRNA interference was used to verify the effects of five genes on ferroptosis and cell function. Based on bioinformatics analysis, five potential genes related to ferroptosis were identified, and their upregulation may affect the onset or progression of DR. This study sheds new light on the pathogenesis of DR.

Keywords: Bioinformatics analysis; Diabetic retinopathy; Ferroptosis; HRMECs; Oxidative stress.