Comprehensive Analysis of circRNA-Associated-ceRNA Networks in Human Corneal Endothelial Dysfunction

Abstract

Purpose: Circular RNAs (circRNAs) are a novel class of endogenous noncoding RNAs that regulate gene expression through the competitive endogenous RNA (ceRNA) mechanism. CircRNA-associated-ceRNA networks are closely related to oxidative stress-related diseases. Oxidative stress-induced dysfunction of the corneal endothelium (CE) is a major pathological feature in many corneal diseases. This study was aimed to analyze circRNA-associated-ceRNA networks in oxidative stress-induced CE dysfunction.

Methods: A CE dysfunction model was established using human corneal endothelial cells (HCECs) treated with H 2 O 2 at a concentration of 250 μM for 4 hours at 37°C. High-throughput sequencing was conducted to determine the expression profiles of circRNA, miRNA, and mRNA. Bioinformatic analyses, including Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes analysis, were conducted to identify the potential biological modules and pathologic pathways of dysregulated circRNAs. CircRNA-associated-ceRNA networks were established based on the data of sequencing and bioinformatic analyses.

Results: We obtained 108 differentially expressed circRNAs, including 77 upregulated and 31 downregulated circRNAs. GO analysis suggested that dysregulated circRNAs were mainly targeted to protein quality control for misfolded or incompletely synthesized proteins (biologic process), nuclear chromatin (cellular component), and ubiquitin protein ligase binding (molecular function). GO terms related to CE functions responding to oxidative stress were also identified. Kyoto Encyclopedia of Genes and Genomes pathway analysis indicated that dysregulated circRNAs were mostly enriched in the adherens junction pathway. Network analysis identified several potential therapeutic targets for CE dysfunction.

Conclusions: CircRNAs are significantly dysregulated in HCECs under oxidative stress. The circRNA-associated-ceRNA networks are closely related to HCEC functions. Targeting these networks might provide novel therapies for CE dysfunction.