1. Obese and normal-weight children display a different plasma metabolic profile as measured with 1 H-NMR spectroscopy Bervoets Liene 1,2,3, Massa Guy 2, Reekmans Gunter 3 and Adriaensens Peter 3 1 Faculty of Medicine and Life Sciences, Hasselt University, Hasselt, Belgium 2 Department of Paediatrics, Jessa Hospital, Hasselt, Belgium 3 Institute for Materials Research (IMO), Hasselt University, Diepenbeek, Belgium Contact information: liene.bervoets@uhasselt.beliene.bervoets@uhasselt.be 1 H-NMR investigation and statistical comparison of the plasma metabolic profile of obese and normal-weight children in order to gain information regarding the involved biochemical pathways and to define obesity-related biomarkers. Childhood obesity is a major health problem worldwide. Obese children are at high risk to develop co-morbidities such as cardiovascular dysfunction, type 2 diabetes, pulmonary, hepatic and renal complications. To improve current prevention and treatment strategies for childhood obesity, a proper understanding of obesity-related pathophysiological mechanisms is required. Metabolomics is increasingly used as a tool for the study of obesity. Subjects This study is part of the Limburg Clinical Research Program (LCRP) UHasselt-ZOL-Jessa, supported by the foundation Limburg Sterk Merk, Hasselt University, Ziekenhuis Oost-Limburg and Jessa Hospital. Samples are stored at the University Biobank Limburg (UBiLim). 53 overweight or obese28 normal-weight 81 children between 8 and 18 years 1 H-NMR spectroscopy Statistical analysis Differences in general characteristics between the two study groups were analyzed using the independent samples t test for scale variables and Chi square test for nominal variables. Statistical significance was assessed at the 5% level. The integration value of 110 spectral regions were normalized to the total integration area (except for TSP, water, glucose and fructose). Multivariate analysis was performed by means of OPLS- DA using SIMCA-P+ 12 (version 12.0, Umetrics, Umeå, Sweden). 1 H-NMR spectroscopy 400 MHz Agilent/Varian Inova spectrometer Fasting plasma sample 13 000 g 4 min 4°C 200 µl plasma + 600 µl D2O + TSP Table 1 | General characteristics of the study population. F: female; M: male; N: native; A: allochthonous. Figure 2 | S-line plot for the OPLS-DA model between obese (positive) and normal-weight (negative) children. The colored scale bar indicates the importance of metabolite variations in discriminating between obese and normal-weight children. unsaturated lipids arginine myo-inositol glucose α-ketoglutarate cysteine asparagine citrate NI proline lipids (VLDL,LDL) lactate Multivariate analysis of 1 H-NMR data regarding plasma metabolite concentrations reveals a separation of obese and normal-weight children. The obese metabolite profile showed an increase in unsaturated lipids, proline, lactate, lipids (VLDL and LDL) and a non-identified compound, besides a decrease in glucose, arginine, myo-inositol, α-ketoglutarate,cysteine, asparagine and citrate Figure 1 | OPLS-DA score plot of the model discriminating between obese (green) and normal- weight (blue) children. The horizontal axis corresponds to between class variability and the vertical axis to within class variability. R2X=0.81; R2Y=0.76; Q2=0.69 To our knowledge, this is the first study in which 1 H-NMR spectroscopy is used as a tool to study childhood obesity. Our findings show that obese children clearly display a different plasma metabolic profile as compared to normal-weight children. Obese children have elevated concentrations of lipids, VLDL, LDL, unsaturated lipids and myo-inositol in their plasma, suggesting an increased fat synthesis. The additional reduced glucose concentration suggests a high rate of glucose consumption for fat synthesis. Furthermore, several metabolites important in energy and amino acid metabolism differentiate between obese and normal-weight children. Future research will focus on a large sample population in order to define obesity-related biomarkers. There were more boys (p=0.006) and children of allochthonous origin (p=0.002) in the obese study group. Age was not different between the two groups (p=0.886). BMI was higher in obese children compared to normal-weight children (p<0.001).