Interactions Between Diet and the Intestinal Microbiota Alter Intestinal Permeability and Colitis Severity in Mice

Abstract

Background & aims: It is not clear how the complex interactions between diet and the intestinal microbiota affect development of mucosal inflammation or inflammatory bowel disease. We investigated interactions between dietary ingredients, nutrients, and the microbiota in specific pathogen-free (SPF) and germ-free (GF) mice given more than 40 unique diets; we quantified individual and synergistic effects of dietary macronutrients and the microbiota on intestinal health and development of colitis.

Methods: C56BL/6J SPF and GF mice were placed on custom diets containing different concentrations and sources of protein, fat, digestible carbohydrates, and indigestible carbohydrates (fiber). After 1 week, SPF and GF mice were given dextran sulfate sodium (DSS) to induce colitis. Disease severity was determined based on the percent weight change from baseline, and modeled as a function of the concentration of each macronutrient in the diet. In unchallenged mice, we measured intestinal permeability by feeding mice labeled dextran and measuring levels in blood. Feces were collected and microbiota were analyzed by 16S rDNA sequencing. We collected colons from mice and performed transcriptome analyses.

Results: Fecal microbiota varied with diet; the concentration of protein and fiber had the strongest effect on colitis development. Among 9 fiber sources tested, psyllium, pectin, and cellulose fiber reduced the severity of colitis in SPF mice, whereas methylcellulose increased severity. Increasing dietary protein increased the density of the fecal microbiota and the severity of colitis in SPF mice, but not in GF mice or mice given antibiotics. Psyllium fiber reduced the severity of colitis through microbiota-dependent and microbiota-independent mechanisms. Combinatorial perturbations to dietary casein protein and psyllium fiber in parallel accounted for most variation in gut microbial density and intestinal permeability in unchallenged mice, as well as the severity of DSS-induced colitis; changes in 1 ingredient could be offset by changes in another.

Conclusions: In an analysis of the effects of different dietary components and the gut microbiota on mice with and without DSS-induced colitis, we found complex mixtures of nutrients affect intestinal permeability, gut microbial density, and development of intestinal inflammation.

Keywords: IBD; Microbiota; Mouse Models; Systems Biology.

Figure 1. Dietary screening demonstrated that protein and fiber have the greatest effect upon DSS colitis

(A) Mice were fed one of thirty-two unique diets for a week before giving 3% DSS in drinking water. Disease severity was measured as weight loss. Stepwise regression and lasso were used to identify (B) macronutrients and (C) dietary ingredients that contributed to differences in weight change at 4 and 7 days after beginning 3% DSS. Factors significantly associated with increased and decreased colitis severity are shown in red and blue respectively.

Figure 2. Casein promotes colitis and psyllium protects from colitis

Mice fed (A) LC diet had less severe weight loss after one week on DSS than those fed a HC diet, while mice consuming the (B) PSY diet had less severe disease than those consuming the CEL diet. (C) After 4 days of DSS, fecal lipocalin-2 was almost undetectable for the PSY diet and highly increased in the HC diet. After 7 days of DSS or at the humane endpoint, (D) TNF-α and (E) IL-6 from whole colonic explants demonstrated increases in HC and decreases in PSY. (F) Representative H&E staining and (G) histopathological scoring of colons indicate that HC and CEL drive more severe disease. The mean ± SEM are plotted for each time point and group (AC, G). Points represent individual animals (D–E). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001

Figure 3. The effects of casein upon health and colitis are microbiota-dependent

(A) In the DSS colitis model, HC diet reduces survival relative to LC diet in SPF mice but not in GF animals. Alpha diversity measured by (B) Faith’s phylogenetic diversity rarified to 10k reads per sample demonstrated decreased microbial diversity in the HC diet. Furthermore, mice on HC diet had (C) decreased barrier function in healthy SPF mice, while barrier was not influence by casein in GF mice. In panes D–F, animals were given one of 9 controlled diets with different casein concentrations (TD.09049-TD.09057) with varying amounts of nutrients for a week and then given DSS. (D) Increased gut microbial density, driven by casein concentration, was associated with more severe disease in terms of both (E) weight change on DSS at d7 and (F) TNF-α. Lines represent surviving number of animals (A). SPF animals were given 3% DSS and GF animals were given 2% DSS. Black (SPF) and gray (GF) points represent individual animals (B–C). The mean ± SEM are plotted for each point (D–F). *p<0.05, **p<0.01

Figure 4. Dietary psyllium has both microbiota-dependent and microbiota-independent effects upon host health and colitis

The HPSY diet led to (A) increased survival in both SPF and GF mice with the greatest difference observed in SPF animals. Alpha diversity measured by (B) Faith’s phylogenetic diversity rarified to 7k reads per sample demonstrated increased microbial diversity in the HPSY diet. Mice fed HPSY had increased colonic (C) butyrate, isobutyrate (p=0.028, t-test; Table S5), and valerate (p<0.0001, t-test; Table S5), which was coupled with (D) an increase in colonic Tregs in SPF animals. In GF animals, Tregs were increased via oral supplementation of SCFA in the drinking water. (E) To determine the dietary pretreatment time necessary to obtain the protective benefit of psyllium in DSS colitis, mice were initially given the HCEL diet and were transferred to the HPSY diet at various time points. Administering the HPSY diet at least two days prior to the administration of DSS was sufficient to provide the same protection as a week of diet. (F) To determine the duration of the benefit of psyllium in DSS colitis, mice were initially given the HPSY diet and were switched to the HCEL diet at different time points. The replacement of psyllium with cellulose almost immediately halted the protective benefits of psyllium in mouse survival. (G) The HCEL diet was associated with increased intestinal permeability with a larger barrier defect in SPF relative to GF mice. SPF animals were given 3% DSS and GF animals were given 2% DSS. Lines represent surviving number of animals (A, E–F). Black (SPF) and grey (GF) points represent individual animals (B–D, G). **p<0.01, ***p<0.001, ****p<0.0001.