Identification and Validation of Genes Related to RNA Methylation Modification in Diabetic Retinopathy

Purpose: To identify and validate the differentially expressed genes related to RNA methylation modification in diabetic retinopathy.Methods: The data sets GSE12610 and GSE111465 related to diabetic retinopathy in the Gene Expression Omnibus were selected. The R software package was used to identify differentially expressed genes related to RNA methylation modification in diabetic retinopathy. Protein-protein interaction network was constructed to explore the interactions between proteins and predict proteins. Then, Gene Ontology annotation analysis and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis were used to analyze the potential enrichment pathways and clarify the biological functions of these genes. In addition, the correlation between them and immune cells was visualized, and receiver operating characteristic curves were drawn to evaluate the diagnostic performance of each one of them for diabetic retinopathy. To verify the differentially expressed genes, the mRNA expression of rat retinal vascular endothelial cells cultured in low and high glucose medium separately were detected by RT-qPCR.Results: The expression of Lrpprc, Nsun4, Nsun6 and Trdmt1 were significantly up-regulated in diabetic retinopathy samples, while the expression of Cbll1, Hnrnpc, Mettl3 and Wtap were significantly down-regulated. Differentially expressed genes were mainly enriched in the RNA-methylation-medication pathways and biological function. The results of immune infiltration analysis proved that eosinophils aggregated more in diabetic group, while T cells follicular helper aggregated more in normal samples. These genes of Cbll1 (AUC = 0.986), Hnrnpc (AUC = 0.819), Lrpprc (AUC = 0.806), Mettl3 (AUC = 0.917), Nsun4 (AUC = 0.819), Nsun6 (AUC = 0.819), Trdmt1 (AUC = 0.972) and Wtap (AUC = 0.972) were respectively used as the diagnostic basis of diabetic retinopathy. According to the RT-qPCR results, the expression of Mettl3 was significantly down-regulated (p < 0.0005) in cells cultured in high glucose, while Trdmt1 (p < 0.05), Nsun4 (p < 0.05) and Nsun6 (p < 0.05) were significantly up-regulated.Conclusion: Differentially expressed genes such as Mettl3, Nsun4, Nsun6, and Trdmt1 should be conducted to explore, and the role of RNA methylation in the process of diabetic retinopathy would be revealed in-depth.