Jiangwei Jia Bo Liu ^* Xin Wang ^* Fenglan Ji ^* Fuchun Wen ^* Huibo Xu ^* Tao Ding ^*

• Jilin Academy of Chinese Medicine Sciences, Changchun, China

Background: Diabetic retinopathy (DR) is one of the common chronic complications of diabetes mellitus, which has developed into the leading cause of irreversible visual impairment in adults worldwide. Compound Qilian Tablets (CQLT) is a Traditional Chinese Medicine (TCM) developed for treating DR, but its mechanism is still unclear. This study explored the mechanism of action of CQLT in treating DR through metabolomics and intestinal microbiota. Methods: Histopathologic examination of the pancreas and retina of Zucker diabetic fatty (ZDF) rats and immunohistochemistry were used to determine the expression levels of retinal nerve damage indicators ionized calcium binding adaptor molecule-1 (Iba-1) and glial fibrillary acidic protein (GFAP). Rat fecal samples were tested by LC-MS metabolomics to search for potential biomarkers and metabolic pathways for CQLT treatment of DR. Characteristic nucleic acid sequences of rat intestinal microbiota from each group were revealed using 16S rDNA technology to explore key microbes and related pathways for CQLT treatment of DR. At the same time, we investigated the effect of CQLT on the gluconeogenic pathway. Results: After CQLT intervention, islet cell status was improved, Iba-1 and GFAP expression were significantly decreased, and abnormal retinal microvascular proliferation and exudation were ameliorated. Metabolomics results showed that CQLT reversed 20 differential metabolites that were abnormally altered in DR rats. Intestinal microbiota analysis showed that treatment with CQLT improved the abundance and diversity of intestinal flora. Functional annotation of metabolites and intestinal flora revealed that glycolysis/gluconeogenesis, alanine, aspartate and glutamate metabolism, starch and sucrose metabolism were the main pathways for CQLT in treating DR. According to the results of correlation analysis, there were significant correlations between Iba-1, GFAP, and intestinal microbiota and metabolites affected by CQLT. In addition, we found that CQLT effectively inhibited the gluconeogenesis process in diabetic mice. Conclusion: In conclusion, CQLT could potentially reshape intestinal microbiota composition and regulate metabolite profiles to protect retinal morphology and function, thereby ameliorating the progression of DR.