Predicting diabetic kidney disease with serum metabolomics and gut microbiota

作者：Yuyun Hu, Xue Ni, Qinghuo Chen, Yihui Qu, Kanan Chen, Gaohui Zhu, Minqiao Zhang, Ningjie Xu, Xu Bai, Jing Wang, Yanhong Ma, Qun Luo, Kedan Cai · 发表于：Scientific Reports · 年份：2025 · DOI：10.1038/s41598-025-91281-9 · 被引用次数：17 · 研究领域：Diet and metabolism studies、Gut microbiota and health、Liver Disease Diagnosis and Treatment

This study aims to identify biomarkers for reliably predicting diabetic kidney disease (DKD), systematically characterize serum metabolites and gut microbiota in DKD patients, and investigate the correlation between differential serum metabolites and gut microbiota. From September 2021 to January 2023, 90 subjects were recruited: 30 with DKD, 30 with type 2 diabetes mellitus (T2DM), and 30 normal controls (NCs). Serum metabolites, including 180 different metabolites, were analyzed using untargeted metabolomics UPLC-MS/MS, and gut microbiota were assessed via 16S rRNA sequencing. Differential metabolites were identified through univariate (t-test or Mann–Whitney U-test, P < 0.05) and multivariate analyses (OPLS-DA model, VIP > 1, P < 0.05), followed by selection using the least absolute shrinkage and selection operator (LASSO). The selected overlapping serum metabolites, along with DKD-associated differential gut microbiota, were used to develop a logistic regression prediction model for DKD based on six markers. In the DKD group compared to the DM and NC groups, 39 and 60 differential serum metabolites were identified, respectively (VIP > 1, P < 0.01). Among these, 36 serum metabolites, including alpha-Hydroxyisobutyric acid, were significantly elevated in DKD patients compared to those with DM. Of these, 28 metabolites showed a negative correlation with estimated glomerular filtration rate (eGFR), while 29 showed a positive correlation with urine albumin creatinine ratio (UA...