Role of bile in infectious disease: the gall of 7α-dehydroxylating gut bacteria.

Summary

Local antibiotics and quorum sensors produced by gut bacteria regulate microbiota community structure and guard against pathogens. In this issue of Cell Chemical Biology, Kang et al. guide us through a reciprocal host-antimicrobial interplay that redefines our understanding of Clostridioides (Clostridium) difficile pathogenesis.

C. difficile is an opportunistic pathogen that causes life threatening colitis in patients administered antibiotics (Smits et al., 2016). In several seminal publications, the susceptibility to C. difficile infection (CDI) is correlated with perturbed fecal bile salt profiles and recent studies have implicated these cholesterol-derivatives as key regulators of C. difficile life cycle and virulence in the host. In this prior body of work, members of the colonic microbiota that convert the primary bile salt cholate to deoxycholate via the 7α-dehydroxylation pathway strongly correlate with an environment that is protected against C. difficile invasion (Buffie et al., 2015; Solbach et al., 2018; Theriot et al., 2016). It was shown over 50 years ago that during antibiotic treatment, the abundance of deoxycholate in the human colon is dramatically reduced, thus reducing the abundance of a metabolite toxic for C. difficile growth (Hamilton, 1963). Conversely, antibiotic treatment leads to an accumulation of cholate, possibly favoring conditions for C. difficile spore germination (Smits et al., 2016). Due to the effects of bile salts on C. difficile physiology and virulence, a patient’s bile salt status is being considered for precision-based microbiota therapy in infected patients or as preventative treatment in high risk individuals. In these scenarios, the central premise in returning a patient to a CDI-resistant state is the restoration of the production of deoxycholate from cholate by 7α-dehydroxylation gut bacteria thereby toxifying the colonic environment and preventing C. difficile colonization.

In this issue, Kang et al. (Kang et al., 2018) contributes significantly to this concept by demonstrating that the major protective action of 7α-dehydroxylating bacteria (e.g., Clostridium scindens or Clostridium sordellii) is through the production of tryptophan based antibiotics (1-acetyl-β-carboline and turbomycin A, respectively), combined with secondary bile salt conversion. The authors’ find that the production of 1-acetyl-β-carboline by C. scindens inhibits C. difficile growth in vitro and that the presence of cholate is required for its toxic effects during co-culture (Kang et al., 2018). The authors go on to demonstrate that there is a dose-dependent increase in the toxicity of turbomycin A or 1-acetyl-β-carboline toward C. difficile with increasing amounts of deoxycholate or lithocholate, but not cholate (chenodeoxycholate, the primary bile salt of lithocholate was not tested) (Kang et al., 2018). Their results demonstrate that 7α-dehydroxylating bacteria produce natural antibiotics that inhibit the growth of C. difficile in a healthy colonic environment that contains secondary bile salts (Figure 1).

Figure 1:

Summary figure illustrating the antimicrobial interplay between 7α-dehydroxylating gut bacteria and C. difficile under steady state conditions (top panel) and following antibiotics (bottom panel). Abbreviations: CA, cholate; CDCA, chenodeoxycholate; DCA, deoxycholate; LCA, lithocholate

The study findings are particularly noteworthy in light of several reports that implicate secondary bile salt production by the colonic microbiome as the mechanism by which a host is protected against CDI (Buffie et al., 2015; Solbach et al., 2018; Theriot et al., 2016). Despite the lack of clear mechanistic studies, this hypothesis has gained almost dogma-like status in the C. difficile community. The most compelling data in support of this hypothesis was first provided by Buffie and colleagues (Buffie et al., 2015). In this important study, the authors’ found that the presence of the bile acid inducible operon (Bai) in the colonic microbiome strongly correlates with an environment that is protected against CDI in both humans and mice. The authors went on to demonstrate that colonization of antibiotic-treated mice with a consortium of bacteria-associated with low CDI risk in patients and mice (including C. scindens) protected against mild experimental disease (Buffie et al., 2015). Moreover, the load of C. difficile was increased upon treatment of mouse samples with cholestyramine, a strong ionic resin that binds C. difficile toxin, antibiotics and bile salts, but demonstrates poor efficacy in CDI patients and severe experimental models (Kurtz et al, 2001), ). The authors concluded that secondary bile salt production by C. scindens (or others) was the principal mechanism by which a colonic microbiome protects against CDI and the cholestyramine-mediated removal of the secondary bile salts results in loss of protection by C. scindens. This may be true, but for different reasons. This anionic resin may indeed bind to and sequester the secondary bile salts produced by C. scindens. This binding may then have reduced the toxicity of 1-acetyl-β-carboline to C. difficile vegetative growth. Based on these two seminal studies, C. scindens (and possibly related organisms) is clearly important in providing resistance against C. difficile invasion by providing a “one-two punch” to C. difficile growth.

Though this exciting work by Kang et al. (Kang et al., 2018) builds upon the hypothesized mechanism of C. scindens-mediated protection against CDI, there are some limitations to their study. The antibiotics produced by C. scindens and C. sordellii generally have modest minimum inhibitory concentrations and their efficacy is likely dependent on the presence of secondary bile salts in the colon. Proof-of concept for this synergistic antimicrobial mechanism needs to be demonstrated in vivo, especially since these bacteria may occupy different intestinal niches. This may be relevant for C. sordelii which is reported to co-associate with C. difficile in patients. Given the lack of developed genetic systems for C. scindens, it would be a tall task to demonstrate, conclusively, that 7α-dehydroxylation of primary bile salts and the production of 1-acetyl-β-carboline, together, toxify the colonic environment such that C. difficile cannot colonize. However, C. sordellii is genetically tractable and could be used to tease apart the impact of turbomycin A with secondary bile salt production. Moreover, the use of gnotobiotic mice fed 1-acetyl-β-carboline or turbomycin A with secondary bile salts, alone, may be straightforward.

Surprisingly, when C. difficile and C. scindens are grown in co-culture in the absence of cholate, C. difficile suppressed C. scindens growth by producing cyclic dipeptides – potentially explaining why the colonic microbiome has difficulty in recovering to a pre-antibiotic state during CDI without anti-C. difficile therapeutic intervention. This work adds a significant new disease mechanism(s) that could explain how C. difficile interacts with the microbiota community structure to generate its own privileged niche to facilitate growth and toxin production. This concept has been eluded to previously by Lawley et al. describing the C. difficile epidemic 027 ribotype as a pathogen capable of directly modulating the host microbiota community to maintain an infective “super-shedder” state (Lawley et al., 2012). Identification of proline based cyclic-dipeptides in genus Clostridium is interesting in light that several of these antimicrobials are previously reported for soil and marine organisms, but they have not been described in anaerobic gut bacteria. The finding that C. difficile produces proline-based cyclic-dipeptides is especially intriguing with proline utilization in C. difficile growth being well established (Bouillaut et al., 2015). The use of Stickland metabolism by C. difficile as a source of energy has been described in several studies (Bouillaut et al., 2015). During Stickland metabolism, bacteria oxidatively deaminate and/or decarboxylate several different amino acids. In the reductive branch, glycine and/or proline are reduced by the selenium-containing glycine and proline reductases to acetate or 5-aminovalerate, respectively, to regenerate the NAD+ pool (Bouillaut et al., 2015). Proline is a highly relevant amino acid for C. difficile growth and a recent study by Battaglioli et al. (Battaglioli et al., 2018) found that the proline reductase is required for a robust infection in gnotobiotic mice colonized with a dysbiotic patient microbiota that is susceptible to CDI. These findings combined indicate that proline metabolism by C. difficile impacts multiple facets of the C. difficile lifestyle – from colonization and growth to the production of toxic molecules to potentiate a C. difficile growth advantage (Bouillaut et al., 2015; Kang et al., 2018).

Conclusions

Kang et al. elegantly illustrate the yin and yang of antimicrobial resistance that has been successfully exploited in the natural design of the human microbiome evasion of C. difficile. This concept is appealing since the synergistic actions of secondary bile salts on antimicrobial activity may have widespread therapeutic application by enhancing the efficacy of standard-of-care antibiotics. Such a prototypic therapeutic approach may be particularly useful in the compromised host with aberrant secondary bile salt profiles, as is the case in patients with liver disease patients who are highly susceptible to CDI.