Department of Medicine, Duke University School of Medicine, Durham, NC, USA.
Department of Population Health Sciences, Duke University School of Medicine, Durham, NC, USA.
Metabolomics. 2021 Dec 20;18(1):5. doi: 10.1007/s11306-021-01858-4.
Urine tricarboxylic acid (TCA) cycle organic anions (OAs) are elevated in diabetes and may be biomarkers for diabetic kidney disease (DKD) progression.
We assessed associations of 10 urine TCA cycle OAs with estimated glomerular filtration rate (eGFR) and eGFR slope.
This study is ancillary to the Simultaneous Risk Factor Control Using Telehealth to SlOw Progression of Diabetic Kidney Disease (STOP-DKD) Trial-a randomized trial of pharmacist-led medication and behavior management in 281 patients with early to moderate DKD at Duke from 2014 to 2015. We used linear mixed models to assess associations of urine TCA cycle OAs with outcomes and modelled TCA cycle OAs as: (1) the average of z-scores for each OA; and (2) principal component (PC) scores derived by principal component analysis (PCA). Untargeted urine metabolomics were added for additional discovery.
Among 132 participants with 24 h urine samples (50% men; 58% Black; mean age 64 years [SD 9]; mean eGFR 74 ml/min/1.73m [SD 21] and median urine albumin-to-creatinine [UACR] 20 mg/g [IQR 8-95]), PCA identified 3 OA metabolite PCs. Malate, fumarate, pyruvate, α-ketoglutarate, lactate, succinate and citrate/isocitrate loaded positively on PC1; methylsuccinate, ethylmalonate and succinate loaded positively on PC2; and methylmalonate, ethylmalonate and citrate/isocitrate loaded negatively on PC3. Over a median follow-up of 1.8 years (IQR, 1.2 to 2.2), higher average OA z-score was strongly associated with higher eGFR after covariate adjustment (p = 0.01), but not with eGFR slope (p = 0.9). Higher PC3, but not other PCs, was associated with lower eGFR (p < 0.001). Conditional random forests and smooth clipped absolute deviation models confirmed methylmalonate, citrate/isocitrate, and ethylmalonate, and added lactate as top ranked metabolites in models of baseline eGFR (R-squared 0.32 and 0.33, respectively). Untargeted urine metabolites confirmed association of urine TCA cycle OAs with kidney function.
Thus, lower urine TCA cycle OAs, most notably lower methylmalonate, ethylmalonate and citrate/isocitrate, are potential indicators of kidney impairment in early stage DKD.
尿三羧酸（TCA）循环有机酸（OA）在糖尿病中升高，可能是糖尿病肾病（DKD）进展的生物标志物。
我们评估了 10 种尿 TCA 循环 OA 与估计肾小球滤过率（eGFR）和 eGFR 斜率的相关性。
本研究是同时使用远程医疗进行危险因素控制以减缓糖尿病肾病进展（STOP-DKD）试验的辅助研究 - 2014 年至 2015 年期间，在杜克大学对 281 名早期至中度 DKD 患者进行药剂师主导的药物和行为管理的随机试验。我们使用线性混合模型来评估尿 TCA 循环 OA 与结局的相关性，并将尿 TCA 循环 OA 建模为：（1）每个 OA 的 z 分数的平均值；（2）通过主成分分析（PCA）得出的主成分（PC）得分。非靶向尿液代谢组学用于进一步发现。
在 132 名有 24 小时尿液样本的参与者中（50%为男性；58%为黑人；平均年龄 64 岁[SD 9]；平均 eGFR 为 74ml/min/1.73m[SD 21]和中位数尿白蛋白与肌酐比[UACR]20mg/g[IQR 8-95]），PCA 确定了 3 种 OA 代谢物 PC。苹果酸、延胡索酸、丙酮酸、α-酮戊二酸、乳酸、琥珀酸和柠檬酸盐/异柠檬酸在 PC1 上呈正加载；甲基琥珀酸、乙基丙二酸和琥珀酸在 PC2 上呈正加载；而甲基丙二酸、乙基丙二酸和柠檬酸盐/异柠檬酸在 PC3 上呈负加载。在中位数为 1.8 年（IQR，1.2 至 2.2）的随访期间，在调整协变量后，较高的 OA z 分数平均值与较高的 eGFR 密切相关（p=0.01），但与 eGFR 斜率无关（p=0.9）。较高的 PC3，但不是其他 PC，与较低的 eGFR 相关（p<0.001）。条件随机森林和光滑裁剪绝对偏差模型证实，甲基丙二酸、柠檬酸盐/异柠檬酸和乙基丙二酸，以及乳酸作为基线 eGFR 模型中排名最高的代谢物（R 平方分别为 0.32 和 0.33）。非靶向尿液代谢物证实了尿 TCA 循环 OA 与肾功能的关联。
因此，较低的尿 TCA 循环 OA，尤其是较低的甲基丙二酸、乙基丙二酸和柠檬酸盐/异柠檬酸，可能是早期 DKD 肾脏损害的潜在指标。
