Metabolic adaptation to a high-fat diet is associated with a change in the gut microbiota

Abstract

Objective: The gut microbiota, which is considered a causal factor in metabolic diseases as shown best in animals, is under the dual influence of the host genome and nutritional environment. This study investigated whether the gut microbiota per se, aside from changes in genetic background and diet, could sign different metabolic phenotypes in mice.

Methods: The unique animal model of metabolic adaptation was used, whereby C57Bl/6 male mice fed a high-fat carbohydrate-free diet (HFD) became either diabetic (HFD diabetic, HFD-D) or resisted diabetes (HFD diabetes-resistant, HFD-DR). Pyrosequencing of the gut microbiota was carried out to profile the gut microbial community of different metabolic phenotypes. Inflammation, gut permeability, features of white adipose tissue, liver and skeletal muscle were studied. Furthermore, to modify the gut microbiota directly, an additional group of mice was given a gluco-oligosaccharide (GOS)-supplemented HFD (HFD+GOS).

Results: Despite the mice having the same genetic background and nutritional status, a gut microbial profile specific to each metabolic phenotype was identified. The HFD-D gut microbial profile was associated with increased gut permeability linked to increased endotoxaemia and to a dramatic increase in cell number in the stroma vascular fraction from visceral white adipose tissue. Most of the physiological characteristics of the HFD-fed mice were modulated when gut microbiota was intentionally modified by GOS dietary fibres.

Conclusions: The gut microbiota is a signature of the metabolic phenotypes independent of differences in host genetic background and diet.

Figure 1

Metabolic adaptation after 3 months of high-fat diet (HFD) and change in phenotype by supplementation of HFD with gluco-oligosaccharides (GOS). (A) Timeline of the experimental protocol. (B) Intraperitoneal glucose tolerance test (IPGTT) in HFD-fed mice that became diabetic (HFD-D, closed squares, n=24), diabetes-resistant (HFD-DR, open squares, n=10) or fed a GOS-supplemented HFD (HFD+GOS, closed triangles, n=10). Data are shown as mean ± SEM; *p<0.05, ***p<0.001 (HFD-D vs HFD-DR) and ^§§§p<0.001 (HFD-D vs HFD+GOS) (two-ANOVA and Bonferroni post test). (C) Glycaemic index; ***p<0.001 (unpaired Student t test).

Figure 2

Caecum microbial profiles vary according to metabolic phenotypes. Pyrosequencing analysis of (A, B) phyla and (C–E) taxa families in mice fed a high-fat diet (HFD) which became diabetic (HFD-D) or diabetes-resistant (HFD-DR) or mice fed HFD supplemented with gluco-oligosaccharides (HFD+GOS). Data are shown as a percentage of the total identified sequences per group.

Figure 3

Caecum microbial genera of different metabolic phenotypes. (A-E) Pyrosequencing of genera in mice fed a high-fat diet (HFD) that became diabetic (HFD-D), diabetes-resistant (HFD-DR) or mice fed a diet supplemented with gluco-oligosaccharides (HFD+GOS). Data are shown as a percentage of the total identified sequences per group.

Figure 4

Cluster identification of gut microbial profiles of the different metabolic phenotypes: (A) phyla, (B) families, (C) genera and (D) overall taxa shown according to metabolic phenotypes for mice fed a high-fat diet (HFD) that became diabetic (HFD-D, closed squares), diabetes-resistant mice (HFD-DR, open squares) and mice fed a diet supplemented with gluco-oligosaccharides (HFD+GOS, closed triangles). Pearson tree analysis was performed to cluster groups (top) and taxa (left side) of each heat map.

Figure 5

Plasma inflammation and gut paracellular permeability. (A) Fasting plasma lipopolysaccharide (LPS) levels, (B) cytokine concentrations in plasma and (C–E) intestinal paracellular permeability in the ileum (C), caecum (D) and colon (E) in mice fed a high-fat diet (HFD) that became diabetic (HFD-D), diabetes-resistant (HFD-DR) and in mice fed a diet supplemented with gluco-oligosaccharides (HFD+GOS). Data are shown as mean ± SEM; *p<0.05, **p<0.01 (unpaired Student t test; n=6–12 per group). IL, interleukin; PAI-1, plasminogen activator inhibitor 1; TNFα, tumour necrosis factor α.

Figure 6

Cell architecture and inflammation in visceral white adipose tissue (WAT). (A) Adipocyte area and distribution, (B) mean adipocyte area, (C) total cell count of the stroma vascular fraction (SVF)/g visceral WAT, (D) endothelial (CD31+), preadipocytes (CD34+), macrophages (F4/80/CD11b+), total lymphocytes and T cells (CD3+), (E) macrophage immunostaining and number (%) per adipocyte count, (F) western blot analysis of phosphorylated and total proteins involved in inflammatory pathways in mice fed a high-fat diet (HFD) which became diabetic (HFD-D), diabetes-resistant (HFD-DR) and mice fed a diet supplemented with gluco-oligosaccharides (HFD+GOS). Data are shown as mean ± SEM; *p<0.05, **p<0.01, ***p<0.001 (unpaired Student t test; n=5–12 per group).

Figure 7

Liver weight, inflammation, insulin and energy pathways during metabolic adaptation. (A) Liver weight; (B) TNFα, IL-6 and PAI-1 mRNA concentrations; (C–E) western blot analysis of phosphorylated and total proteins involved in (C) insulin signalling, (D) energy metabolism and (E) inflammation from mice fed a high-fat diet (HFD) which became diabetic (HFD-D), diabetes-resistant (HFD-DR) or mice fed a diet supplemented with gluco-oligosaccharides (HFD+GOS). Data are shown as mean ± SEM; *p<0.05, ***p<0.001 (unpaired Student t test; n=4 per group). IL, interleukin; PAI-1, plasminogen activator inhibitor 1; TNFα, tumour necrosis factor α.