Antibiotic-Induced Alterations of the Gut Microbiota Alter Secondary Bile Acid Production and Allow for Clostridium difficile Spore Germination and Outgrowth in the Large Intestine

Abstract

It is hypothesized that the depletion of microbial members responsible for converting primary bile acids into secondary bile acids reduces resistance to Clostridium difficile colonization. To date, inhibition of C. difficile growth by secondary bile acids has only been shown in vitro. Using targeted bile acid metabolomics, we sought to define the physiologically relevant concentrations of primary and secondary bile acids present in the murine small and large intestinal tracts and how these impact C. difficile dynamics. We treated mice with a variety of antibiotics to create distinct microbial and metabolic (bile acid) environments and directly tested their ability to support or inhibit C. difficile spore germination and outgrowth ex vivo. Susceptibility to C. difficile in the large intestine was observed only after specific broad-spectrum antibiotic treatment (cefoperazone, clindamycin, and vancomycin) and was accompanied by a significant loss of secondary bile acids (deoxycholate, lithocholate, ursodeoxycholate, hyodeoxycholate, and ω-muricholate). These changes were correlated to the loss of specific microbiota community members, the Lachnospiraceae and Ruminococcaceae families. Additionally, physiological concentrations of secondary bile acids present during C. difficile resistance were able to inhibit spore germination and outgrowth in vitro. Interestingly, we observed that C. difficile spore germination and outgrowth were supported constantly in murine small intestinal content regardless of antibiotic perturbation, suggesting that targeting growth of C. difficile will prove most important for future therapeutics and that antibiotic-related changes are organ specific. Understanding how the gut microbiota regulates bile acids throughout the intestine will aid the development of future therapies for C. difficile infection and other metabolically relevant disorders such as obesity and diabetes. IMPORTANCE Antibiotics alter the gastrointestinal microbiota, allowing for Clostridium difficile infection, which is a significant public health problem. Changes in the structure of the gut microbiota alter the metabolome, specifically the production of secondary bile acids. Specific bile acids are able to initiate C. difficile spore germination and also inhibit C. difficile growth in vitro, although no study to date has defined physiologically relevant bile acids in the gastrointestinal tract. In this study, we define the bile acids C. difficile spores encounter in the small and large intestines before and after various antibiotic treatments. Antibiotics that alter the gut microbiota and deplete secondary bile acid production allow C. difficile colonization, representing a mechanism of colonization resistance. Multiple secondary bile acids in the large intestine were able to inhibit C. difficile spore germination and growth at physiological concentrations and represent new targets to combat C. difficile in the large intestine.

Keywords: Clostridium difficile; antibiotics; bile acids; colonization resistance; microbiota.

Figures

FIG 1

Antibiotic treatment scheme and experimental design. (A) C57BL/6 mice were treated with various antibiotics that would result in different microbial and metabolic environments. Red circles represent the time of necropsy for each group (n = 7 to 10 mice per group). (B) At the time of necropsy, ileal and cecal contents were collected. Ex vivo germination and outgrowth of C. difficile spores were measured in paired ileal and cecal contents from all treatment groups as well as by targeted bile acid analysis. In vitro spore germination and growth studies were done using relevant in vivo ileal and cecal bile acid concentrations. Microbiome analysis was also done to understand the relationship between gut bacteria and bile acids using correlation analysis. Abbreviations: noabx, no antibiotic; cef, cefoperazone; 1wk to 6wk, number of weeks off cefoperazone; 6wkC, 6 weeks off cefoperazone plus an intraperitoneal administration (IP) of clindamycin; vanco, vancomycin; metro, metronidazole; kana, kanamycin; clinda, clindamycin.

FIG 2

C. difficile ex vivo spore germination and outgrowth assays in murine ileal and cecal contents. Ex vivo germination and outgrowth of C. difficile spores were done in ileal (A) and cecal (B) contents collected from mice treated with various antibiotics. C. difficile VPI 10463 spores inoculated into the ileal contents of mice treated with or without antibiotics (noabx) allowed for spore germination and outgrowth after a 6-h period, whereas spores in non-antibiotic-treated cecal content did not. Only specific antibiotic treatments (cefoperazone [cef], 1 or 2 weeks off cefoperazone, 6 weeks off cefoperazone plus an intraperitoneal administration of clindamycin [6wkC], clindamycin [clinda], and vancomycin [vanco]) in the cecum supported spore germination and outgrowth. Black bars represent spores only, and gray bars represent spores and vegetative cells. Significance between groups was determined by Mann-Whitney nonparametric t test. Error bars represent the mean ± standard error of the mean (SEM) (*, P < 0.05; **, P < 0.01). n.s. not significant.

FIG 3

Targeted bile acid metabolomics of murine ileal and cecal contents. Bile acids were analyzed by LC-MS from paired ileal (A) and cecal (B) contents from ex vivo spore germination and outgrowth assays. A heat map shows the bile acid concentration present in micrograms per 100 mg of gut content, ranging from 1 to 120,000. The black boxes represent samples that did not reach significance in supporting spore germination and outgrowth of C. difficile spores in Fig. 2. (C) Bile acids present in cecal content that did not support spore germination and outgrowth in the black bars (resistance) or were able to support germination and outgrowth (susceptible) are compared. Significance between groups was determined by Mann-Whitney nonparametric t test. Error bars represent the mean ± SEM (****, P < 0.0001).

FIG 4

Secondary bile acids present during C. difficile resistance inhibit spore germination. In vitro spore germination inhibition assays were performed with C. difficile strain VPI 10463 to assess if the secondary bile acids ω-muricholate (ωMCA), hyodeoxycholate (HDCA), ursodeoxycholate (UDCA), lithocholate (LCA), and deoxycholate (DCA) were able to inhibit spore germination with known germinants TCA and DCA. Spores were incubated for 30 min in (A) BHI plus TCA (0.1%) or (B) BHI plus DCA (0.1%) supplemented with a range of secondary bile acids at relevant in vivo concentrations. Positive controls include BHI plus TCA (0.1%) or DCA (0.1%) alone with mock H2O and mock ethanol (EthOH) (red bars). A negative control was also used: BHI plus TCA supplemented with CDCA (0.04%) or BHI plus DCA supplemented with CDCA (0.04%), a known inhibitor of spore germination (blue bar). Negative controls include BHI alone without the addition of TCA or DCA. The data presented represent the mean ± standard deviation (SD) from triplicate experiments and were significant compared to the positive controls (A) TCA alone or (B) DCA alone (Student’s t test, *, P < 0.05; **, P < 0.01; ***, P < 0.001).

FIG 5

Secondary bile acids present during C. difficile resistance inhibit growth of C. difficile. (A) Growth rates (per hour) are shown. C. difficile was grown in BHI medium supplemented with various secondary bile acids with a range of in vivo concentrations (0.001% to 0.1%) present during resistance to C. difficile. The data presented represent the means ± SD from triplicate experiments and were significant compared to positive controls C. difficile and C. difficile with mock ethanol (EthOH) without bile acids (Students t test, **, P < 0.05; ***, P < 0.01). (B) Growth curve of representative secondary bile acids that significantly decreased the growth rate of C. difficile in panel A.

FIG 6

Correlation analysis of the gut microbiome and bile acids. Spearman’s correlation analysis was done with all 121 OTU (i.e., OTU that were greater than 1% of the total bacterial population) in the microbiome, color coded by phylum and grouped based on unsupervised clustering. All 26 bile acids detected were similarly clustered and are color coded based on structure. There were three distinct clusters of OTU (O1 to O3) and two distinct clusters of bile acids (B1 and B2). The heat map scale ranges from positively correlated, +0.7, to negatively correlated, −0.7.