Increased dietary protein stimulates amino acid catabolism via the gut microbiota and secondary bile acid production

Abstract

Excess amino acids from a protein-rich diet are mainly catabolized in the liver. However, it is still unclear to what extent the gut microbiota may be involved in the mechanisms governing this catabolism. Therefore, the aim of this study was to investigate whether consumption of different dietary protein concentrations induces changes in the taxonomy of the gut microbiota, which may contribute to the regulation of hepatic amino acid catabolism. Consumption of a high-protein diet caused overexpression of HIF-1α in the colon and increase in mitochondrial activity, creating a more anaerobic environment that was associated with changes in the taxonomy of the gut microbiota promoting an increase in the synthesis of secondary bile acids, increased secretion of pancreatic glucagon. This effect was demonstrated in pancreatic islets, where secondary bile acids stimulated the expression of the PC2 enzyme that promotes glucagon formation. The increase in circulating glucagon was associated with an induction of the expression of hepatic amino acid-degrading enzymes, an effect attenuated by antibiotics. Thus, high protein intake in mice and humans induced the increase of different species in the gut microbiota with the capacity to produce secondary bile acids leading to an increase in secondary bile acids and glucagon levels, promoting amino acid catabolism.

Keywords: Gut microbiota; amino acid catabolism; glucagon; high-protein diet; secondary bile acids.

Graphical abstract

Figure 1.

Mice fed increasing concentrations of dietary protein had decreased body fat mass and increased energy expenditure. (a) Body weight gain, (b) energy intake, (c) fat mass and lean mass, (d) respiratory exchange ratio (RER), (e) oxygen consumption, and (f) heat after dietary protein consumption. LP (low protein), NP (normal protein), HP (high protein). Data are expressed as mean ± SEM. Statistical analyses were performed by two-way ANOVA followed by Tukey’s post hoc test (for C and E-F). Multiple comparisons are shown in lowercase (a > b), n=5-8.

Figure 2.

Differential gene expression changes in the colon after feeding different protein concentrations. (a) Principal component analysis (PCA) plot based on differential mRNA expression in mice fed different protein concentrations. Relative abundance of mRNAs with significant changes: (b) autophagy related 7 (Atg7), (c) monocarboxylate transporter 4 (MCT4 or Slc16a3), (d) hypoxia inducible factor-1 (Hif-1α), (e) hepatocyte nuclear factor 4a (HNF-4A), (f) Mothers against decapentaplegic homolog 3 (Smad3), (g) Nuclear receptor coactivator 2 (Ncoa2), (h) Aly/REF export factor (Alyref), (i) Forkhead box O3 (Foxo 3). (j) Inhibitor of differentiation 1 (Id1), (k) B-cell lymphoma 2 (Bcl-2), and (l) B-cell lymphoma extra large (Bcl-xL). LP (low protein), NP (normal protein), HP (high protein). Statistical analyses were performed by two-way ANOVA followed by Tukey’s post hoc test (B-L). Multiple comparisons are summarized with lowercase letters (a > b). n=5.

Figure 3.

Colon morphology, mitochondrial function, and hypoxic state were altered in mice fed a high-protein diet. (a) Histologic sections of ascending colon from animals fed different protein concentrations. (b) Crypt length analyzed by automated morphometric analysis (hematoxylin-eosin stained micrographs, magnification ×100) (n=8-10). (c and d) Mitochondrial respiratory states of isolated colonic mitochondria. Values were obtained from oxygen respiratory rate (OCR), (n=4). (e) Colonic hypoxic microenvironment of mice fed different protein concentrations injected with pimonidazole. (f) Relative fluorescence units of colon, immunofluorescence staining was analyzed with imageJ (n=3) for hypoxia assessment. LP (low protein), NP (normal protein), HP (high protein). Data are expressed as mean ± SEM. Densitometric statistical analyses were performed by two-way ANOVA followed by Tukey’s post hoc test. Multiple comparisons are summarized with lowercase letters (a > b) (for B and F). p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 4 (for D).

Figure 4.

Gut microbiota is modified by dietary protein concentration in mice. (a) Comparative alpha diversity of gut microbiota, (b) Principal coordinate analysis of beta diversity. Taxonomy of gut microbiota at the level of (c) phylum and (d) genus; (e) linear discriminant analysis (LDA) shows significant differences in the relative abundance of bacterial species in mice fed different protein concentrations. (f) Relative mRNA abundance of P. distasonis. LP (low protein), NP (normal protein), HP (high protein). Statistical analyses were performed by two-way ANOVA followed by Tukey’s post hoc test (for A and D). Multiple comparisons are indicated by lowercase letters (a > b > c); n=8-10. Data are expressed as mean ± SEM.

Figure 5.

Impact of Dietary Protein Levels on Gut Microbiota Potential Functions in Mice. Phylogenetic analysis of communities to predict the functional metagenome and thus differentially active bacterial metabolic pathways, including (a) nitrogen metabolism, (b) histidine metabolism, and (c) secondary bile acid biosynthesis after consumption of different dietary protein concentrations. (d) Total fecal concentration of bile acids, (e) primary fecal bile acids concentration, (f) western blot and (g) densitometric analysis of FGF15 in ileum and colon, (H) fecal secondary bile acid concentrations, (I) bile acid profile by percentage. (j) serum bile acids concentration, (j) serum glucagon concentration, (l) stimulation of glucagon secretion, and (m) relative expression of prohormone convertase 2 (PC2) by secondary bile acids in pancreatic islets. Data are expressed as mean ± SEM. Statistical analyses were performed by one-way ANOVA followed by Tukey’s post hoc test. Multiple comparisons are summarized with lowercase letters (a > b > c). n=8-10. A, B, and C were analyzed using PICRUSt bioinformatics package software. Abbreviations: CA, cholate; CDCA, chenodeoxycholate; DCA, deoxycholate; LCA, lithocholate; UDCA, ursodeoxycholate.

Figure 6.

Expression of amino acid catabolizing enzymes (AACE) in mouse liver and increased postprandial urea depend on the amount of dietary protein. Densitometric analysis and blots of protein abundance of (a) SDS, (b) CPS1 and (c) HAL, GAPDH was the structural protein control. (d, e and f) Measurement of postprandial biochemical parameters. Data are presented as mean ± SEM. Statistical analyses were performed by two-way ANOVA followed by Tukey’s post hoc test (A-J). Multiple comparisons are summarized with small letters (a ≠ b ≠ c). n=5-10.

Figure 7.

Antibiotic depletion of the gut microbiota reduced the effect of dietary protein concentration on reducing body fat mass and increasing energy expenditure. (a) Body weight gain, (b) fat mass and lean mass, (c) energy intake, (d) respiratory exchange ratio (RER), (e) oxygen consumption, and (f) heat after dietary protein consumption. LP (low protein), NP (normal protein), HP (high protein). Data are presented as mean ± SEM. Statistical analyses were performed by two-way ANOVA followed by Tukey’s post hoc test (for c and e,f). Multiple comparisons are summarized with small letters (a ≠ b ≠ c). n=5-6.

Figure 8.

Gut microbiota depletion with antibiotics modifies the effect of dietary protein concentration on nitrogen metabolism, total secondary bile acids, and serum glucagon concentration. (a) Comparative alpha diversity in the gut microbiota, (b) principal coordinate analysis of beta diversity, (c) genus level, (d) histological sections of ascending colon collected from the indicated group of animals fed different protein concentrations and treated with antibiotic cocktail given in water consisting of antibiotic ampicillin salt/sodium (1 g mL-1) and neomycin (0.5 g mL-1) (n=10). (f) Nitrogen metabolism, (g) Relative abundance of secondary bile acid biosynthesis, (h) Serum bile acid concentration, (i) Fecal total secondary bile acids, (j) Serum glucagon concentration, Relative expression of (k) Histidase (HAL), (l) Serine dehydratase (SDS), and (m) Glutaminase. Data are presented as mean ± SEM. Statistical analysis was performed by two-way ANOVA followed by Tukey’s post hoc test (for A and E-I), n=5-10. Multiple comparisons are indicated by small letters (a ≠ b ≠ c). LP (low protein), NP (normal protein), HP (high protein). F and G were analyzed by PICRUSt.

Figure 9.

Circulating concentrations of amino acids were modified by the antibiotic treatment in mice fed different concentrations of dietary protein. (a) Valine, (b) Leucine, (c) Isoleucine, (d) Serine, (e) Histidine, (f) Glutamine, and (g) Arginine. Data are presented as mean ± SEM. Statistical analyses were performed by two-way ANOVA followed by Tukey’s post-hoc test; statistical significance was set at p < 0.05.

Figure 10.

Consumption of a high-protein diet altered the taxonomy of the gut microbiota, increasing secondary bile acids and circulating glucagon in humans. (a) Comparative alpha diversity of the gut microbiota. Taxonomy of the gut microbiota at the (b) phylum level and (c) genus level. (d) Linear discriminant analysis (LDA) of the gut microbiota at the species level. Concentration of fecal secondary bile acids (e) lithocholic and (f) deoxycholic. (g) Serum glucagon concentration, (h) Serum bile acids concentration. Data are presented as mean ± SEM. Statistical analyses were performed by one-way ANOVA followed by Tukey’s post-hoc test; statistical significance was set at p < 0.05.