Comparative analysis and modeling of the severity of steatohepatitis in DDC-treated mouse strains

PLoS One. 2014 Oct 27;9(10):e111006. doi: 10.1371/journal.pone.0111006. eCollection 2014.

Abstract

Background: Non-alcoholic fatty liver disease (NAFLD) has a broad spectrum of disease states ranging from mild steatosis characterized by an abnormal retention of lipids within liver cells to steatohepatitis (NASH) showing fat accumulation, inflammation, ballooning and degradation of hepatocytes, and fibrosis. Ultimately, steatohepatitis can result in liver cirrhosis and hepatocellular carcinoma.

Methodology and results: In this study we have analyzed three different mouse strains, A/J, C57BL/6J, and PWD/PhJ, that show different degrees of steatohepatitis when administered a 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC) containing diet. RNA-Seq gene expression analysis, protein analysis and metabolic profiling were applied to identify differentially expressed genes/proteins and perturbed metabolite levels of mouse liver samples upon DDC-treatment. Pathway analysis revealed alteration of arachidonic acid (AA) and S-adenosylmethionine (SAMe) metabolism upon other pathways. To understand metabolic changes of arachidonic acid metabolism in the light of disease expression profiles a kinetic model of this pathway was developed and optimized according to metabolite levels. Subsequently, the model was used to study in silico effects of potential drug targets for steatohepatitis.

Conclusions: We identified AA/eicosanoid metabolism as highly perturbed in DDC-induced mice using a combination of an experimental and in silico approach. Our analysis of the AA/eicosanoid metabolic pathway suggests that 5-hydroxyeicosatetraenoic acid (5-HETE), 15-hydroxyeicosatetraenoic acid (15-HETE) and prostaglandin D2 (PGD2) are perturbed in DDC mice. We further demonstrate that a dynamic model can be used for qualitative prediction of metabolic changes based on transcriptomics data in a disease-related context. Furthermore, SAMe metabolism was identified as being perturbed due to DDC treatment. Several genes as well as some metabolites of this module show differences between A/J and C57BL/6J on the one hand and PWD/PhJ on the other.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Animals
  • Cluster Analysis
  • Disease Models, Animal
  • Fatty Liver / chemically induced
  • Fatty Liver / genetics*
  • Fatty Liver / metabolism*
  • Fatty Liver / pathology
  • Gene Expression Profiling
  • Gene Expression Regulation
  • Metabolic Networks and Pathways
  • Metabolome
  • Metabolomics
  • Mice
  • Non-alcoholic Fatty Liver Disease
  • Phenotype
  • Proteomics
  • Pyridines / administration & dosage
  • Pyridines / adverse effects
  • Severity of Illness Index
  • Signal Transduction

Substances

  • 3,5-diethoxycarbonyl-1,4-dihydrocollidine
  • Pyridines

Grant support

This work was funded by a grant of the Austrian Nationalstiftung and the Austria Wirtschaftsservice GmbH in the framework of the IMGuS research program (Institute for Medical Genome Research and Systems Biology, Vienna) and by the Max Planck Society. Co-author Ingrid Osprian was employed by BIOCRATES Life Sciences AG during the runtime of the IMGuS project where she did the metabolic analysis presented in this manuscript. BIOCRATES Life Sciences AG provided support in the form of salary for author Ingrid Osprian, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific role of this author is articulated in the ‘author contributions’ section.