Bervoets Liene, Massa Guy, Guedens Wanda, Louis Evelyne, Noben Jean-Paul, Adriaensens Peter
Faculty of Medicine and Life Sciences, Hasselt University, Martelarenlaan 42, 3500 Hasselt, Belgium.
Department of Pediatrics, Jessa Hospital, Stadsomvaart 11, 3500 Hasselt, Belgium.
Diabetol Metab Syndr. 2017 Jun 26;9:48. doi: 10.1186/s13098-017-0246-9. eCollection 2017.
Type 1 diabetes mellitus (T1DM) is one of the most common pediatric diseases and its incidence is rising in many countries. Recently, it has been shown that metabolites other than glucose play an important role in insulin deficiency and the development of diabetes. The aim of our study was to look for discriminating variation in the concentrations of small-molecule metabolites in the plasma of T1DM children as compared to non-diabetic matched controls using proton nuclear magnetic resonance (H-NMR)-based metabolomics.
A cross-sectional study was set-up to examine the metabolic profile in fasting plasma samples from seven children with poorly controlled T1DM and seven non-diabetic controls aged 8-18 years, and matched for gender, age and BMI-SDS. The obtained plasma H-NMR spectra were rationally divided into 110 integration regions, representing the metabolic phenotype. These integration regions reflect the relative metabolite concentrations and were used as statistical variables to construct (train) a classification model in discriminating between T1DM patients and controls.
The total amount of variation explained by the model between the groups is 81.0% [RY(cum)] and within the groups is 75.8% [RX(cum)]. The predictive ability of the model [Q(cum)] obtained by cross-validation is 50.7%, indicating that the discrimination between the groups on the basis of the metabolic phenotype is valid. Besides the expected higher concentration of glucose, the relative concentrations of lipids (triglycerides, phospholipids and cholinated phospholipids) are clearly lower in the plasma of T1DM patients as compared to controls. Also the concentrations of the amino acids serine, tryptophan and cysteine are slightly decreased.
The present study demonstrates that metabolic profiling of plasma by H-NMR spectroscopy allows to discriminate between T1DM patients and controls. The metabolites that significantly differ between both groups might point to disturbances in biochemical pathways including (1) choline deficiency, (2) increased gluconeogenesis, and (3) glomerular hyperfiltration. Although the sample size of this study is still somewhat limited and a validation should be performed, the proof of principle looks promising and justifies a deeper investigation of the diagnostic possibilities of H-NMR metabolomics in follow-up studies. NCT03014908. Registered 06/01/2017. Retrospectively registered.
1型糖尿病（T1DM）是最常见的儿科疾病之一，在许多国家其发病率正在上升。最近的研究表明，除葡萄糖外的其他代谢物在胰岛素缺乏和糖尿病发展中起着重要作用。我们研究的目的是使用基于质子核磁共振（H-NMR）的代谢组学方法，寻找T1DM儿童与非糖尿病匹配对照组血浆中小分子代谢物浓度的差异变化。
开展一项横断面研究，以检测7名T1DM控制不佳的儿童和7名8至18岁非糖尿病对照者空腹血浆样本的代谢谱，这些对照者在性别、年龄和BMI-SDS方面相匹配。将获得的血浆H-NMR光谱合理地划分为110个积分区域，代表代谢表型。这些积分区域反映了相对代谢物浓度，并用作统计变量来构建（训练）区分T1DM患者和对照者的分类模型。
模型解释的两组间总变异量为81.0%[RY（累积）]，组内为75.8%[RX（累积）]。通过交叉验证获得的模型预测能力[Q（累积）]为50.7%，表明基于代谢表型区分两组是有效的。除了预期的较高葡萄糖浓度外，与对照组相比，T1DM患者血浆中脂质（甘油三酯、磷脂和胆碱化磷脂）的相对浓度明显较低。此外，氨基酸丝氨酸、色氨酸和半胱氨酸的浓度也略有降低。
本研究表明，通过H-NMR光谱对血浆进行代谢谱分析能够区分T1DM患者和对照者。两组之间显著不同的代谢物可能表明生化途径存在紊乱，包括（1）胆碱缺乏、（2）糖异生增加和（3）肾小球高滤过。尽管本研究的样本量仍然有限，需要进行验证，但原理证明看起来很有前景，有理由在后续研究中对H-NMR代谢组学的诊断可能性进行更深入的研究。NCT03014908。于2017年1月6日注册。回顾性注册。
