Abstract
PURPOSE: The aim of our study was to investigate a comprehensive profile of streptozotocin (STZ)-induced early diabetic retinopathy (DR) mice to identify a risk scoring signature based on micorRNAs (miRNAs) for early DR diagnosis. METHODS: RNA sequencing was performed to obtain the gene expression profile of retinal pigment epithelium (RPE) in early STZ-induced mice. Differentially expressed genes (DEGs) were determined with log2|fold change (FC)| > 1 and p value < 0.05. Functional analysis was carried out based on gene ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis and the protein-protein interaction (PPI) network. We predicted the potential miRNAs via online tools and ROC curves were then conducted. Three potential miRNAs with AUC > 0.7 were explored via public datasets and a formula was further established to evaluate DR severity. RESULTS: In total, 298 DEGs (200 up-regulating and 98 down-regulating) were obtained through RNA sequencing. Hsa-miR-26a-5p, hsa-miR-129-2-3p and hsa-miR-217 were three predicted miRNAs with AUC > 0.7, suggesting their potential to distinguish healthy controls from early DR. The formula of DR severity score = 19.257 - 0.004 × hsa-miR-217 + 5.09 × 10(-5) × hsa-miR-26a-5p - 0.003 × hsa-miR-129-2-3p was established based on regression analysis. CONCLUSIONS: In the present study, we investigated the candidate genes and molecular mechanisms based on RPE sequencing in early DR mice models. Hsa-miR-26a-5p, hsa-miR-129-2-3p and hsa-miR-217 could work as biomarkers for early DR diagnosis and DR severity prediction, which was beneficial for DR early intervention and treatment.