Fermented dairy products modulate Citrobacter rodentium-induced colonic hyperplasia

Abstract

We evaluated the protective effects of fermented dairy products (FDPs) in an infection model, using the mouse pathogen Citrobacter rodentium (CR). Treatment of mice with FDP formulas A, B, and C or a control product did not affect CR colonization, organ specificity, or attaching and effacing lesion formation. Fermented dairy product A (FDP-A), but neither the supernatant from FDP-A nor β-irradiated (IR) FDP-A, caused a significant reduction in colonic crypt hyperplasia and CR-associated pathology. Profiling the gut microbiota revealed that IR-FDP-A promoted higher levels of phylotypes belonging to Alcaligenaceae and a decrease in Lachnospiraceae (Ruminococcus) during CR infection. Conversely, FDP-A prevented a decrease in Ruminococcus and increased Turicibacteraceae (Turicibacter). Importantly, loss of Ruminococcus and Turicibacter has been associated with susceptibility to dextran sodium sulfate-induced colitis. Our results demonstrate that viable bacteria in FDP-A reduced CR-induced colonic crypt hyperplasia and prevented the loss of key bacterial genera that may contribute to disease pathology.

Keywords: C. rodentium; DLIT-µCT; bioluminescence imaging; fermented dairy products; microbiota; probiotic.

Figure 1.

A, Quantification of Citrobacter rodentium (CR) colony-forming units (CFU) from stools over 8 days postinfection (p.i.) in the different treatment groups. B, In vivo optical imaging of a bioluminescent (BL) CR infection from 3 representative mice per treatment at days 4 and 8 p.i. C, Electron microscopy of terminal colon at day 8 p.i. reveals epithelial cell death (*) and attaching and effacing (A/E) lesions (arrowheads), irrespective of the fermented dairy product (FDP) treatment used. Abbreviations: BB, brush border; CFU, colony-forming units; CP, control product; SEM, scanning electron microscopy; TEM, transmission electron microscopy; U, untreated and uninfected.

Figure 2.

Histological analysis of Citrobacter rodentium (CR)–infected mice following treatment with (A) control product (CP), (B) fermented dairy product (FDP) A, (C) FDP-B, (D) FDP-C, (E) no treatment (Citrobacter rodentium [CR]), and (F) untreated and uninfected (U). FDP-A and FDP-B reduced lymphocyte accumulation in the lamina propria (arrowheads). Scale bar = 100 µm. G, Treatment of mice with FDP-A significantly reduced crypt hyperplasia compared with CP-treated mice (***P < .0001). H, Histological damage score demonstrating that FDP-A significantly reduced disease pathology compared with CR- and CP-treated groups (***P < .0001). FDP-A significantly reduced CR-associated pathology compared with FDP-B (***P < .0001). P values in red have been calculated compared with CP; P values in black have been compared to the CR group (*P < .05); P values in green are comparisons between FDPs.

Figure 3.

Quantification of crypt hyperplasia following treatment of mice with (A) fermented dairy product (FDP) A or (B) β-irradiated (IR) FDP-A, (C) no treatment (Citrobacter rodentium [CR]), and (D) untreated uninfected (U). E, FDP-A significantly reduced crypt hyperplasia compared with IR-FDP-A (***P < .0001) and qualitatively reduced lymphocyte accumulation in the lamina propria (arrowheads). IR-FDP-A treatment did not significantly reduce crypt hyperplasia compared with CR-infected mice (P > .05). F, Histological damage score demonstrating that FDP-A, but not IR-FDP-A, significantly reduced CR-associated pathology (***P < .0001). G, Treatment with supernatant from S-FDP-A significantly increased crypt hyperplasia compared with CR-treated or untreated and uninfected (U) mice (*P < .05). H, Histological damage score demonstrated that S-FDP-A treatment did not significantly alter CR associated pathology compared with CR-treated mice (P > .05). Scale bar = 100 µm.

Figure 4.

Indirect immunofluorescence using anti β-intimin (green) and anti–Lactobacillus rhamnosus (red). A and B, Fermented dairy product (FDP)–A treated and Citrobacter rodentium (CR) infected. (C and D), Irradiated FDP-A treated and CR infected. (E and F), No treatment and CR-infected and (G and H) untreated and uninfected (U). CR and Lactobacillus rhamnosus were observed associated to the epithelial layer lining the lumen of the colon. Lactobacillus rhamnosus was only found in the intestinal lumen, or associated with colonic epithelial cells following FDP-A treatment. The distribution of CR associated with the colonic mucosa was not affected by FDP treatment. Scale bars = 20 µm (left column) and 10 µm (right column).

Figure 5.

Microbiota profiling of stool samples by 16S rRNA gene pyrosequencing and quantitative polymerase chain reaction (qPCR). Samples were taken before treatment (day –10 [d-10]), before infection (day 0 [d0]), and 8 days postinfection (d8 p.i.). A, Evaluation of colonization of 4 strains present within fermented dairy product (FDP) A and β-irradiated (IR) FDP-A by qPCR. Taxa abundance profiles at (B) the phylum level and (C) the family level. D and E, Representation of 2D and 1D principal coordinates analysis of weighted Unifrac distances. F, Discriminant genera identified by sparse partial least squares multivariate analyses. Blue-colored boxes indicate discriminant genera in the corresponding comparisons (i) between day –10 and day 0 for both products, (ii) between day 0 and day 8 p.i. for both products, (iii) between FDPs for change after consumption (day 0 – day –10), or (iv) after infection (day 8 p.i. – day 0). Bold black frames focus on results described in the text, and stars in the blue boxes indicates significant population differences identified using a nonparametric Kruskal–Wallis test. “Other” indicates sequences that could not be attributed to a lower taxonomic level; g_ indicates sequences assigned to bacteria unclassified at the genus level.

Figure 6.

Taxa abundance profiles according to treatment groups and time points in cell equivalents per gram of feces were calculated using quantitative polymerase chain reaction. Abbreviations: FDP, fermented dairy product; IR, β-irradiated; p.i. postinfection.