Serum metabolomic profiling highlights pathways associated with liver fat content in a general population sample.

M Koch, S Freitag-Wolf, S Schlesinger, J Borggrefe, J R Hov, M K Jensen, J Pick, M R P Markus, T Höpfner, G Jacobs, S Siegert, A Artati, G Kastenmüller, W Römisch-Margl, J Adamski, T Illig, M Nothnagel, T H Karlsen, S Schreiber, A Franke, M Krawczak, U Nöthlings, W Lieb

Journal: European journal of clinical nutrition 2018;71(8):995-1001

PMID: 28378853

Full Text on Nature Publishing Group View on PubMed

Abstract

BACKGROUND/OBJECTIVES

Fatty liver disease (FLD) is an important intermediate trait along the cardiometabolic disease spectrum and strongly associates with type 2 diabetes. Knowledge of biological pathways implicated in FLD is limited. An untargeted metabolomic approach might unravel novel pathways related to FLD.

SUBJECTS/METHODS

In a population-based sample (n=555) from Northern Germany, liver fat content was quantified as liver signal intensity using magnetic resonance imaging. Serum metabolites were determined using a non-targeted approach. Partial least squares regression was applied to derive a metabolomic score, explaining variation in serum metabolites and liver signal intensity. Associations of the metabolomic score with liver signal intensity and FLD were investigated in multivariable-adjusted robust linear and logistic regression models, respectively. Metabolites with a variable importance in the projection >1 were entered in in silico overrepresentation and pathway analyses.

RESULTS

In univariate analysis, the metabolomics score explained 23.9% variation in liver signal intensity. A 1-unit increment in the metabolomic score was positively associated with FLD (n=219; odds ratio: 1.36; 95% confidence interval: 1.27-1.45) adjusting for age, sex, education, smoking and physical activity. A simplified score based on the 15 metabolites with highest variable importance in the projection statistic showed similar associations. Overrepresentation and pathway analyses highlighted branched-chain amino acids and derived gamma-glutamyl dipeptides as significant correlates of FLD.

CONCLUSIONS

A serum metabolomic profile was associated with FLD and liver fat content. We identified a simplified metabolomics score, which should be evaluated in prospective studies.

Address: Institute of Epidemiology, Kiel University, Kiel, Germany.; Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA, USA.; Institute of Medical Informatics and Statistics, Kiel University, Kiel, Germany.; Institute of Epidemiology, Kiel University, Kiel, Germany.; Department of Radiology, University of Cologne, Cologne, MA, USA.; Division of Cancer Medicine, Department of Transplantation Medicine, Surgery and Transplantation, Norwegian PSC Research Center, Oslo University Hospital, Rikshospitalet, Oslo, Norway.; K.G. Jebsen Inflammation Research Center, Institute of Clinical Medicine, University of Oslo, Oslo, Norway.; Division of Cancer Medicine, Surgery and Transplantation, Research Institute of Internal Medicine, Oslo University Hospital, Oslo, Norway.; Division of Cancer Medicine, Section of Gastroenterology, Department of Transplantation Medicine, Surgery and Transplantation, Oslo University Hospital, Rikshospitalet, Oslo, Norway.; Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA, USA.; Nutritional Epidemiology, Department of Nutrition and Food Science, Rheinische Friedrich-Wilhelms-University Bonn, Bonn, Germany.; Department of Internal Medicine B, University Medicine Greifswald, Greifswald, Germany.; DZHK (German Centre for Cardiovascular Research), Partner Site Greifswald, Greifswald, Germany.; Department of Study of Health in Pomerania/Clinical-Epidemiological Research, Institute for Community Medicine, University Medicine Greifswald, Greifswald, Germany.; Institute of Epidemiology, Kiel University, Kiel, Germany.; PopGen Biobank, University Medical Center Schleswig-Holstein, Kiel, Germany.; Cologne Center for Genomics, University of Cologne, Cologne, Germany.; Institute of Experimental Genetics, Genome Analysis Center, Helmholtz Zentrum München, Neuherberg, Germany.; Institute of Bioinformatics and Systems Biology, Helmholtz Zentrum München, Neuherberg, Germany.; Deutsches Zentrum für Diabetesforschung (DZD), German Centere for Diabetes Research, Neuherberg, Germany.; Institute of Bioinformatics and Systems Biology, Helmholtz Zentrum München, Neuherberg, Germany.; Institute of Experimental Genetics, Genome Analysis Center, Helmholtz Zentrum München, Neuherberg, Germany.; Deutsches Zentrum für Diabetesforschung (DZD), German Centere for Diabetes Research, Neuherberg, Germany.; Experimental Genetics, Technical University of Munich, Freising, Germany.; Hannover Unified Biobank, Hannover Medical School, Hannover, Germany.; Institute for Human Genetics, Hannover Medical School, Hanover, Germany.; Institute of Clinical Molecular Biology, Christian-Albrechts-University of Kiel, Kiel, Germany.; Institute of Internal Medicine I, University Medical Center Schleswig-Holstein, Kiel, Germany.; Institute of Clinical Molecular Biology, Christian-Albrechts-University of Kiel, Kiel, Germany.

DOI: 10.1038/ejcn.2017.43

Link outs

Free resources

Medical:

https://medlineplus.gov/fattyliverdisease.html

Other Literature Sources:

https://scite.ai/reports/28378853

Subscription / membership required

Full Text Sources:

MeSH Terms: Aged, Alcohol Drinking, Biological Specimen Banks, Biomarkers, Cohort Studies, Computational Biology, Cross-Sectional Studies, Dipeptides, Expert Systems, Fatty Liver, Alcoholic, Female, Glutamic Acid, Humans, Lipid Metabolism, Liver, Magnetic Resonance Imaging, Male, Metabolomics, Middle Aged, Non-alcoholic Fatty Liver Disease, Self Report, Severity of Illness Index