Premium
Gene, Sex and Diet Interact to Control the Tissue Metabolome
Author(s) -
Wells Ann,
Barrington William,
Threadgill David,
Dearth Stephen,
Campagna Shawn,
Saxton Arnold,
Voy Brynn
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.127.2
Subject(s) - metabolome , metabolite , adipose tissue , biology , strain (injury) , metabolomics , endocrinology , medicine , obesity , physiology , genetics , bioinformatics , anatomy
Individuals use diet to lose weight, combat disease and improve overall health, but one diet does not necessarily fit all. Genetic variation is an important determinant of dietary efficacy, but little is known about how diet and genetics interact at the tissue level to alter metabolism. The objective of this study is to determine how dietary regimens that are known to impact health impact tissue level metabolism, and how these responses are influenced by genetic background, using untargeted metabolomics. Ultimately, metabolite profiles will be associated with overlying system traits linked to obesity and its consequences. Four strains of mice (C57Bl/6J, FVB/NJ, A/J and NOD/LtJ), which included males and females, were fed one of five isocaloric diets (Research Diets Inc): Mediterranean (D12052702), ketogenic (D12052706), Japanese (D12052703), Western (D12052705) or standard chow (D12052701). Abdominal adipose, liver and muscle tissue metabolites were measured using liquid‐chromatography mass spectrometry. Statistically significant differences between groups were identified using the following model : metabolite = sex + diet + strain + sex*strain + sex * diet + sex * strain * diet + tissue weight + internal standard, where tissue weight and internal standard are covariates. ANOVA (FDR p‐value<. 05) showed that 116, 124, and 158 metabolites in adipose, liver and skeletal muscle, respectively, were significantly altered by sex, strain, diet, strain by diet, strain by sex, diet by sex, or sex by strain by diet interaction. Clustering algorithms were used to group mice based on “metabotypes”, phenotypes defined by distinct strain and diet dependent metabolite profiles. Unsupervised clustering revealed that metabolite profiles distinguish mice based on strain. Supervised partial least squares‐discriminant analysis showed that metabolite profiles, in adipose, muscle, and liver tissue, differentiated mice into three subgroups based on strain. A/J and C57Bl/6 were clustered independently, while FVB/NJ, NOD/LtJ were clustered together. Effect of diet on metabotype was most apparent in all three tissues with the ketogenic and western diet differentiating from the other four diets. The effects of sex and diet on all three tissues distinguished male and female Japanese diet from all other metabotypes. The effect of strain and sex on all three tissues discriminated A/J males and females from all other metabotypes. However, the effect strain and diet interactions on all three tissues were not apparent. Calculating the VIP scores for all potential interactions revealed that the importance of certain metabolites were unique to each interaction. Diet by sex and diet in adipose and muscle tissue showed hydroxyproline to be the most influential metabolite on clustering, while allantoate and N‐acetyl‐L‐ornithine were most influential to liver tissue. Collectively, these data illustrate that macronutrient composition and genetic background interact to alter tissue metabolism independent of caloric intake. Ongoing efforts are directed toward linking metabotypes to systems level traits that are relevant to obesity and metabolism. Support or Funding Information University of Tennessee Ag Research, and Graduate School of Genome Science and Technology