Comprehensive metabolic profiling of diabetic retinopathy

Diabetic retinopathy (DR) is an important complication of diabetes mellitus and a prevalent blind-causing ophthalmic disease. Despite years of efforts, rapid and accurate diagnosis of DR remains a challenging task. Metabolomics has been used as a diagnostic tool for disease progression and therapy monitoring. In this study, retinal tissues were collected from diabetic mice and age-matched non-diabetic mice. An unbiased metabolic profiling was performed to identify the altered metabolites and metabolic pathways in DR. 311 differential metabolites were identified between diabetic retinas and non-diabetic retinas under the criteria of variable importance in projection (VIP) > 1 and P < 0.05. These differential metabolites were highly enriched in purine metabolism, amino acid metabolism, glycerophospholipid metabolism, and pantaothenate and CoA biosynthesis. We then evaluated the sensitivity and specificity of purine metabolites as the candidate biomarkers for DR through the area under the receiver-operating characteristic curves (AUC-ROCs). Compared with other purine metabolites, adenosine, guanine, and inosine had higher sensitivity, specificity, and accuracy for DR prediction. In conclusion, this study sheds new light on the metabolic mechanism of DR, which can facilitate clinical diagnosis, therapy, and prognosis of DR in the future.