The gut microbiota drives the impact of bile acids and fat source in diet on mouse metabolism

Abstract

Background: As the gut microbiota contributes to metabolic health, it is important to determine specific diet-microbiota interactions that influence host metabolism. Bile acids and dietary fat source can alter phenotypes of diet-induced obesity, but the interplay with intestinal microorganisms is unclear. Here, we investigated metabolic consequences of diets enriched in primary bile acids with or without addition of lard or palm oil, and studied gut microbiota structure and functions in mice.

Results: In combination with bile acids, dietary lard fed to male C57BL/6N mice for a period of 8 weeks enhanced fat mass accumulation in colonized, but not in germ-free mice when compared to palm oil. This was associated with impaired glucose tolerance, lower fasting insulin levels, lower counts of enteroendocrine cells, fatty liver, and elevated amounts of hepatic triglycerides, cholesteryl esters, and monounsaturated fatty acids. Lard- and bile acid-fed mice were characterized by shifts in dominant gut bacterial communities, including decreased relative abundances of Lachnospiraceae and increased occurrence of Desulfovibrionaceae and the species Clostridium lactatifermentans and Flintibacter butyricus. Metatranscriptomic analysis revealed shifts in microbial functions, including lipid and amino acid metabolism.

Conclusions: Caution is required when interpreting data from diet-induced obesity models due to varying effects of dietary fat source. Detrimental metabolic consequences of a diet enriched with lard and primary bile acids were dependent on microbial colonization of the host and were linked to hepatic lipid rearrangements and to alterations of dominant bacterial communities in the cecum.

Keywords: 16S rRNA gene amplicon sequencing; Bile acids; Diet-induced obesity; Dietary fat; Germ-free mice; Gut microbiota; Lard; Lipidomics; Metabolic diseases; Metatranscriptomics.

Fig. 1

Impact of experimental feedings and microbial colonization on mouse metabolism. a Final body weight (at the age of 18 weeks after 8 weeks of feeding). b WAT mass for all groups and representative pictures of respective fat depots collected from SPF mice fed the palm oil- or lard-based diets. c Corresponding regression analysis of WAT mass and body weight. d Blood glucose concentrations during OGTT with corresponding areas under the curve. Color code for diets: CD, black; BA, red; PHB, dark blue; LHB, cyan blue. Symbols for colonization status: GF, diamonds; SPF, filled triangles. All mice used in the experiments are shown (group size varied as indicated below the x-axis). See the “Methods” section for description of statistical analyses. e Quantification of glucagon-like peptide (GLP) 1-positive cells in colonic tissue sections of SPF mice from the different feeding groups. At least three non-consecutive sections were stained from each mouse and quantified. Symbols represent average values from individual mice. Representative pictures of immunohistochemical staining acquired with a confocal microscope are shown (for the sake of space and appropriate size of images, picture for CD and BA are not shown but are equivalent to LHB group). Arrows indicate cells positive for GLP1. The black bars indicate 100 μm. ***p < 0.01, one-way ANOVA followed by the Tukey test (performed using Graph Pad Prism)

Fig. 2

Modulation of hepatic lipid profiles. a Liver to body weight ratio. b Triglyceride content. c Amounts, composition, and distribution of total fatty acids measured in fasted mice (n = 4–7 as indicated in the figure). d Alterations in lipid classes and species. Only lipids representing > 1% total amounts were considered for statistical analysis. Color code is as in Fig. 1. SAFA saturated fatty acids, MUFA monounsaturated fatty acids, PUFA polyunsaturated fatty acids, unsat unsaturated fatty acids, PC phosphatidylcholine, CE cholesteryl esters. See the “Methods” section for description of mass spectrometric measurements and for statistical analyses. e mRNA relative expression of Cd36 in the liver of mice. GAPDH was used as housekeeping gene for the normalization. GF mice fed the PHB diet were used as reference group. In all figure panels, stars indicate statistical significance as follows: *p < 0.05; **p < 0.01; ***p < 0.001 (two-way ANOVA followed by Holm-Sidak)

Fig. 3

Diet-induced alteration of cecal microbiota profiles. a Alpha-diversity shown as richness and Shannon effective counts. b Beta-diversity analysis via multidimensional scaling analysis of generalized UniFrac distances. The p value was obtained by PERMANOVA for testing the significance of separation between sample groups. c Boxplots of significantly altered taxonomic groups at the family level. Erysipelotrichaceae were detected in four of six CD-fed mice. d Relative abundances of dietary group-specific OTUs shown as a heat map. OTU sequences (ca. 450 bp of the V3/V4 region) were classified using EzTaxon. The range of relative abundances of each OTU is given in square brackets next to the corresponding OTU identification number. Statistics were performed and original graphs were generated in the R programming environment using Rhea [67]: *p < 0.05; **p < 0.01; ***p < 0.001. Number of mice: CD, 6; BA, 6; PHB, 7; LHB, 6

Fig. 4

Diet-induced shifts in the metatranscriptome of mouse cecal microbiota. a Heat map of the 1207 genes with differential expression levels between the four diets. Genes were selected according to adjusted p values ≤ 0.001 and absolute(log2FC) ≥ 5. Mice were grouped into two main clusters corresponding to the BA/CD diets or the HFDs supplemented with BA. b Heat map depicting the expression of 266 genes showing differential expression level between the lard- and palm oil-based HFD. Genes were selected according to adjusted p values ≤ 0.001 and absolute(log2FC) ≥ 2.5. c Main metabolic pathways with significantly different expression level between the two HFDs