Fate of Ingested Clostridium difficile Spores in Mice

Clostridium difficile infection (CDI) is a leading cause of antibiotic-associated diarrhea, a major nosocomial complication. The infective form of C. difficile is the spore, a dormant and resistant structure that forms under stress. Although spore germination is the first committed step in CDI onset, the temporal and spatial distribution of ingested C. difficile spores is not clearly understood. We recently reported that CamSA, a synthetic bile salt analog, inhibits C. difficile spore germination in vitro and in vivo. In this study, we took advantage of the anti-germination activity of bile salts to determine the fate of ingested C. difficile spores. We tested four different bile salts for efficacy in preventing CDI. Since CamSA was the only anti-germinant tested able to prevent signs of CDI, we characterized CamSa’s in vitro stability, distribution, and cytotoxicity. We report that CamSA is stable to simulated gastrointestinal (GI) environments, but will be degraded by members of the natural microbiota found in a healthy gut. Our data suggest that CamSA will not be systemically available, but instead will be localized to the GI tract. Since in vitro pharmacological parameters were acceptable, CamSA was used to probe the mouse model of CDI. By varying the timing of CamSA dosage, we estimated that C. difficile spores germinated and established infection less than 10 hours after ingestion. We also showed that ingested C. difficile spores rapidly transited through the GI tract and accumulated in the colon and cecum of CamSA-treated mice. From there, C. difficile spores were slowly shed over a 96-hour period. To our knowledge, this is the first report of using molecular probes to obtain disease progression information for C. difficile infection.


Introduction
Clostridium difficile infection (CDI) is the major identifiable cause of antibiotic-associated diarrhea in hospitals [1]. In the US alone, CDI develops in over 500,000 patients with up to 20,000 deaths per year [2]. The yearly health care burden has been estimated to be greater than $3 billion.
The infective agent of CDI is the C. difficile spore, a hardy structure formed under nutrient deprivation [3]. In a healthy gut, indigenous microbes form a protective barrier against C. difficile colonization of the gastrointestinal (GI) tract, but this protective function can be weakened by antibiotic therapy [4]. Under these favorable conditions, C. difficile spores interact with small molecule germinants, triggering a series of events committing the spore to germinate into toxin producing bacteria [5].
Since spore germination is the first committed step in CDI, understanding the behavior of spores in the GI tract of the host is a necessary first step in infection control [1]. Taurocholate, a natural bile salt, and glycine, an amino acid, were shown to activate C. difficile spore germination [6]. We have reported that C. difficile spores bind taurocholate and glycine through a complex mechanism [7]. Using kinetic analysis, we showed that unknown receptor homo-and heterocomplexes are formed. Others and we also showed that chenodeoxycholate, another natural bile salt, is a competitive inhibitor of C. difficile spore germination [7,8,9,10]. These findings strongly implicate the presence of unidentified proteinaceous germination receptor(s) that C. difficile uses to bind small molecules to activate spore germination.
Analogs of taurocholate and glycine were used as chemical probes to determine structure activity relationships for germinant binding and activation of germination of C. difficile spores in vitro [8]. The putative germination machinery of C. difficile seems to contain unique binding sites for alkyl, aromatic, and basic amino acids as co-germinants whereas the binding region for bile salts is restricted to taurocholate analogs [8]. We reported that a metabenzene sulfonic acid derivative of taurocholate, CamSA, is a strong competitive inhibitor of taurocholate-mediated C. difficile spore germination in vitro. Even more, a single 50 mg/kg dose of CamSA prevented CDI in mice without any observable toxicity [11]. Our results support a mechanism whereby the antigermination effect of CamSA is responsible for preventing CDI signs.
Although the germination of C. difficile spores has been studied in vitro, the in vivo fate of ingested spores is not clear [1]. Determining the timing of ingested spore germination will allow assessing the time window when patients are at risk of developing CDI. Furthermore, determining the transit time of ingested spores through the GI tract will allow defining whether ingested spores contribute to CDI relapse.
Understanding the fate of ingested spores has been hampered by the rapid CDI progression from spore challenge to clinical endpoint in the hamster model of CDI [12]. This is further complicated by the ability of C. difficile vegetative cells to resporulate in the intestine of the animal host [13]. Indeed, previous works have not been able to distinguish between ingested spores and spores formed in the gut of infected animals [14].
In the current study, we tested the ability of four bile salt analogs as in vivo inhibitors of C. difficile spore germination. Since CamSA had the best biological activity, we further characterized CamSA's stability, distribution and cytotoxicity in vitro. Finally, we used CamSA as a probe to estimate the transit time of ingested C. difficile spores and the timing of CDI onset in mice. With this information we proposed a model that describes the spatial and temporal fate of ingested C. difficile spores.

CamSA had no Observable Adverse Effects on Mice
To determine the acute toxicity of CamSA to mice, we used the fixed dose procedure [15]. No physical adverse effects or weight loss were observed when CamSA was administered for three consecutive days at doses up to saturating 300 mg/kg (Fig. S1). A 300 mg/kg dose of chenodeoxycholate caused immediate death, probably due to observed precipitation of chenodeoxycholate upon interaction with mouse saliva and gastric juice. Chenodeoxycholate at 50 mg/kg did not cause any observable side effects.

Prevention of CDI by Bile Salt Analogs
As previously reported, when mice were challenged with 10 8 CFU of C. difficile spores, severe CDI signs developed and all animals reached clinical endpoint by 48 hours post-challenge [16]. The large (10 8 CFUs) inoculum of spores ensured synchronized CDI onset and fast CDI sign progression. Mice treated with up to 300 mg/kg taurocholate or ethyl cholate also developed severe CDI and signs were undistinguishable from control DMSO-treated animals ( Fig. 1 and Fig. S2). Mice treated with 50 mg/kg chenodeoxycholate developed moderate to severe signs of CDI, but onset was delayed by 24 hours (Fig. S3). In contrast, all animals treated with 50 or 300 mg/kg CamSA showed no sign of CDI and were undistinguishable from non-challenged animals [11]. All asymptomatic animals remained free of CDI signs for at least 14 days post-challenge.

Stability of CamSA
CamSA is a taurocholate analog with an amide bond linking cholic acid to meta-aminobenzene sulfonic acid. To be effective, CamSA must survive the changing environments of the GI tract. To test for stability, CamSA was incubated in artificial gastric juice and intestinal juice. No degradation of CamSA was evident even after 24 hours incubation under both conditions (data not shown).
Bacterial bile salt hydrolases (BSHs) deconjugate primary and secondary bile salts [17]. B. longum and L. gasseri are two intestinal bacteria commonly used as test strains for BSH production. After incubation with a culture of B. longum for 24 hours, CamSA and taurocholate are both hydrolyzed to cholic acid at similar rates (Fig. 2). CamSA and taurocholate are less sensitive to degradation by BSHs secreted by L. gasseri. Less than 30% of either CamSA or taurocholate was hydrolyzed to cholic acid after 24 hours (Fig. 2). E. coli does not produce BSH and both CamSA and taurocholate were stable after 24 hour incubation with E. coli cultures (data not shown). CamSA was not degraded in growth medium alone.

Caco-2 Permeability of CamSA
To prevent C. difficile spores from germinating, CamSA needs to be retained in the intestinal lumen. Caco-2 monolayers serves as an in vitro surrogate assay for intestinal permeability, absorption, and metabolism [18]. CamSA was studied in a Caco-2 permeability assay and displayed an apical to basolateral apparent permeability coefficient (P app ) of ,10 26 cm/s and basolateral to apical P app of 10.9610 26 cm/s. The efflux ratio suggests that CamSA is a substrate for active transport (Table S1). In both assays, CamSA was recovered at 100% indicating low binding, accumulation, and metabolism by Caco-2 cells.  Effect of CamSA on Bacterial Growth E. coli, B. longum, and L. gasseri are indigenous mammalian gut bacteria and are continuously exposed to bile salts [17]. As expected, growth of these bacteria was unaffected by the presence of CamSA in the growth medium. C. difficile cells also grew normally in the presence of CamSA (Fig. S4).

Cytotoxicity of CamSA
Cell viability was qualitatively determined by visual observation of rounded/detached cells and trypan blue staining. CamSAtreated Vero, Caco-2, and macrophage cells appeared healthy and were undistinguishable from DMSO-treated cells (Fig. S5). Cell viability was also quantitatively determined by ATP production. Vero and Caco-2 cells treated with 50 or 200 mM CamSA produced ATP at similar levels to healthy control cells (Fig. 3).

CamSA Protection of Vero and Caco-2 Cells
Spent media from outgrowing C. difficile spores killed Vero cells in a dose-dependent manner (Fig. 4). These data are consistent with previous reports indicating that vegetative C. difficile secretes cell-killing toxins during growth [19]. When C. difficile spores were incubated in medium containing 200 mM CamSA, bacterial growth was reduced but not eliminated. As expected, spent media from CamSA-treated bacterial cultures were less effective at killing epithelial cells. Similar results were observed for Caco-2 cell cultures (data not shown).

Timing of CDI Onset
To determine the onset of CDI in mice, animals were challenged with C. difficile spores and treated with 300 mg/kg CamSA between 0 and 12 hours post-challenge. All animals treated with CamSA up to 6 hours post-challenge were fully protected from CDI. In contrast, all animals treated with CamSA at 9 or 12 hours post-challenge developed severe CDI undistinguishable from untreated mice and reached the clinical endpoint 48 hours post infection (Figs. 5A and 5B).
Similar to previous reports, GI contents from animals with CDI signs contained almost exclusively C. difficile vegetative cells [13]. These animals started to excrete large amounts (.10610 5 CFUs) of vegetative cells reaching a maximum between 8 and 10 hours post spore challenge (Fig. 5C). Although some C. difficile spores were excreted by diseased animals, the amounts were negligible (,10% of vegetative CFUs) compared to the high amount of excreted vegetative cells.

Recovery of C. difficile Cells and Spores from Intestines and Feces of CamSA-treated Mice
Similar to the hamster CDI model [13], ingested C. difficile spores narrowly localized to the cecum and colon of CamSA treated mice at every time point tested. A negligible amount of C. difficile was discovered in the small intestine and stomach (Fig. S6). C. difficile spores remained in the cecum and colon for 72 hours after spore challenge (Fig. 6A). By 96 hours, the amount of spores recovered from the cecum and colon of CamSA treated animals decreased almost tenfold, from greater than 12610 5 to less than 2610 5 CFUs.
Consistent with the results from intestinal content, the feces of CamSA-treated animals contained almost exclusively spores (Fig. 6B). In these animals, excretion of ingested C. difficile spores started 2 hours post-challenge and continued until at least 96 hours post-challenge. In fact, by 120 hours post-challenge, the sum of excreted C. difficile spores was quantitatively identical to the number of spores given by gavage.

Discussion
Germination of C. difficile spores is believed to be the first step in establishing CDI [1]. However, determining the fate of ingested C. difficile spores is challenging since spores can germinate and the resulting cells can then re-sporulate perpetuating a cycle of disease. Anti-germinants prevent CDI by effectively freezing ingested spores in their dormant state. We took advantage of this anti-  The resulting spent media were added to Vero cell cultures and incubated for 24 hours. Cell viability was determined with the CellTiter Glo viability kit. The luminescence signal from untreated cells was set as 100% cell viability. Percent survival for other conditions was calculated relative to untreated cells. Error bars represent standard deviations from at least five independent measurements. doi:10.1371/journal.pone.0072620.g004 germinant property to study the temporal and spatial distribution of ingested C. difficile spores.
Following our success using the anti-germination properties of CamSA to protect mice from CDI, we tested taurocholate (a natural germination enhancer), chenodeoxycholate (a natural germination inhibitor), and ethylcholate (a commercially available germination inhibitor) as prophylactics of CDI. Ethyl cholate inhibits C. difficile spore germination with a half maximal inhibitory concentration (IC 50 ) of 8.2 mM, seven-fold more potent than CamSA in vitro (data not shown). Chenodeoxycholate, on the other hand, inhibits C. difficile spore germination with an IC 50 of 235 mM, approximately five-fold less potent than CamSA [20].
As previously reported, CamSA was a prophylactic of murine CDI [11]. In this study, we were able to prevent CDI with saturating concentrations of CamSA without any overt toxic effects. This is consistent with the low toxicity observed for taurocholate and cholate [21]. In fact, CamSA had anti-CDI activity at concentrations that were at least six times lower than its toxic threshold. Furthermore, CamSA-treated animals did not show signs of CDI even 14 days post-challenge. At 300 mg/kg concentration, chenodeoxycholate showed acute toxicity probably due to low solubility in biological fluids. At lower concentrations, chenodeoxycholate was not able to prevent CDI. Animals treated with 50 mg/kg chenodeoxycholate show the same CDI sign patterns as animals treated with 5 mg/kg CamSA [11]. Another potent inhibitor of C. difficile spore germination in vitro, ethylcholate, did not protect mice from CDI even at 300 mg/kg. Similarly, taurocholate, a germinant, was unable prevent CDI. CamSA was the only compound tested that was effective in preventing CDI. Hence, we further characterized CamSA's in vitro pharmacological properties.
To be effective as in vivo probes, anti-germination compounds must be stable to the variable GI tract environments. CamSA is stable to all tested GI tract microenvironments except incubation with the BSH-producing bifidobacteria. Antibiotics disrupt the normal microflora dynamics of the gut allowing for outgrowth of C. difficile [22]. A recent study shows that antibiotic cocktails shift the normal murine gut bacterial population to a preponderance of Lactobacilli [23]. Since BSHs produced by Lactobacilli were not Since CDI is an intestinal infection, anti-germination compounds require low oral bioavailability and high GI tract stability for maximum efficacy. In vitro assays suggest that CamSA will be retained in the intestinal tract of mammals (Table S1). The efflux ratio of CamSA suggests that it is a substrate for Pgp [24]. Hence, the efficacy of CamSA might be in part due to retention in the GI tract due to Pgp mediated excretion of the drug back into the intestinal lumen. The poor bioavailability of CamSA will also reduce toxic effects to other organs since CamSA is unlikely to circulate outside of the intestinal lumen. Tox-ADME analyses also suggest low metabolism of CamSA by intestinal epithelial cells and further supports that CamSA will remain stable in the intestinal lumen.
Because the natural microbiota is key to resist C. difficile infection, anti-germination compounds should not damage this natural barrier [22]. Indeed, CamSA did not affect growth of commensal bacteria, B. longum, L. gasseri, or E. coli. Although CamSA inhibits spore germination in vitro, it does not affect C. difficile vegetative growth (Fig. S4). These data support the proposed mechanism that CamSA inhibits spore germination and does not act as an antibacterial agent. Anti-germination compounds should also show low toxicity toward mammalian cells. Indeed, CamSA did not affect the viability of two different mammalian epithelial cell lines nor did it affect macrophage immune cells.
Since toxin is only secreted by metabolically active C. difficile cells, halting germination should result in less toxin production. Tissue culture protection experiments rely on decreasing the accumulation of toxins secreted into media during C. difficile vegetative growth [19]. CamSA inhibits C. difficile spore germination in vitro and hence protects mammalian cells by reducing the number of toxin-producing bacteria.
Previous works have not been able to distinguish between ingested C. difficile spores and spores that were produced after colonization of the host's GI tract. Indeed, enumeration of C. difficile in feces and GI content from infected hamsters yields mixtures of vegetative cells and spores [13,14]. Furthermore, enumerated spore loads were higher than the original inoculum. These data suggest that ingested C. difficile spores germinated and re-sporulated during colonization of the hamster gut. Due to the fast progression of CDI in untreated mice, we could only determine bacterial loads for the first 48 hours after challenge. Even then, we could not distinguish whether spores recovered from diseased animals came from the original inoculum or from re-sporulation in the intestines.
Since CamSA shows favorable in vitro pharmacological properties and can block C. difficile spore germination in vivo, we were able to follow the fate of ingested C. difficile spores without interference from germination and/or re-sporulation. CamSA was effective in preventing CDI when administered up to six hours following spore challenge. In contrast, CamSA was ineffective when administered nine hours post-challenge, even at the highest concentration tested. This narrow three-hour window correlates C. difficile spore germination with maximum C. difficile vegetative cell shedding in symptomatic mice. These data suggest that a fraction of germinated C. difficile cells are excreted soon after germination, while the remaining C. difficile vegetative cells lead to CDI onset.
Human CDI shows more heterogeneous symptoms and longer disease progression than rodent models [12]. CamSA treatment allowed us to observe the behavior of ungerminated spores for a period extended beyond the normal clinical endpoint of CDIdiseased mice. Ingested C. difficile spores were quantitatively recovered from feces, cecum, and colon contents of CamSAtreated mice. Interestingly, ingested C. difficile spores started to be shed soon after challenge, but part of the population remained in the lower GI for up to four days.
The mechanism of dormant C. difficile spore accumulation in the lower intestine is not understood, but suggests that ingested C. difficile spores can form a transitory reservoir that is slowly released from the lower intestine. A possibility is that a small fraction of C. difficile spores enter a superdormant state that helps them to be retained in the intestine [25]. Although the amount of unattached spores in the intestines is small, it is tempting to speculate that these spore reservoirs serve as a focal point for CDI relapse.
CamSA's anti-germination activity can be used to address mechanistic details about CDI initiation (Fig. 7). The sum of our data suggests that ingested spores rapidly transit through the GI tract (Fig. 6B) and accumulate in the lower intestine ( Fig. 6A and  Fig. S6). Six to nine hours after ingestion enough C. difficile spores germinate to establish infection (Fig. 5A). C. difficile vegetative cells start shedding almost immediately after germination and continue throughout the infection (Fig. 5C). In contrast, dormant C. difficile spores are slowly shed over a four day period (Fig. 6B). The timing of C. difficile spore germination and the persistence of ungerminated spores in the lower intestine can have profound implication in the prophylactic treatment of CDI.

Materials
Bile salts were purchased from Sigma-Aldrich Corporation (St. Louis, MO) or were synthesized in the Abel-Santos laboratory [8]. All bile salts were dissolved in DMSO prior to use. Artificial gastric juice, intestinal juice and PRO disks were purchased from Fisher Scientific (Pittsburg, PA). Thin layer chromatography (TLC) silica gel 60 F 254 plates were purchased from EMD Chemicals (Gibbstown, NJ). Cell Titer-Glo luminescent cell viability assay kit was obtained from Promega Corporation (Madison, WI). Clostridium difficile selective agar plates were purchased from BD Biosciences (Franklin, Lakes, NJ). All bacterial strains used in this study were purchased from ATCC (Manassas, VA) and grown as suggested.

Animals
This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was reviewed and approved by the Institutional Animal Care and Use Committee at the University of Nevada, Las Vegas (Permit Number: R0411-266). Weaned Female C57BL/6N mice were purchased from Harlan laboratories (Indianapolis, IN). Animals were housed in groups of 5 mice per cage in the University of Nevada, Las Vegas animal care facility. All cages, water, food and bedding were autoclaved prior to contact with animals. Upon arrival, mice were allowed to acclimate for one week prior to experimentation. Post-challenge animal manipulations were performed in a biosafety level 2 laminar flow hood.

Acute Toxicity of CamSA in Mice
To determine acute toxicity of CamSA, we used the Fixed Dose Procedure (FDP) [15]. CamSA was dissolved in DMSO to a concentration of 100 mg/ml. Groups of five mice were treated by oral gavage for three consecutive days with 50 mg/kg body weight of CamSA or chenodeoxycholate. A control group was administered DMSO. Weight changes were recorded daily and mice were observed for adverse reactions such as vomiting, diarrhea, hair loss, hunched posture, weight loss, and lethargy. Other groups of mice were treated with 300 mg/kg CamSA or chenodeoxycholate and observed as above.

Prevention of CDI by Bile Salt Analogs
C. difficile 630 (ATCC BAA-1382) were prepared as previously described [11]. Purified C. difficile spores of 630 strain were used to challenge mice as published [16,26]. Briefly, an antibiotic cocktail containing kanamycin (0.4 mg/ml), gentamycin (0.035 mg/ml), colistin (850 U/ml), metronidazole (0.215 mg/ml), and vancomycin (0.045 mg/ml) was prepared in autoclaved water and sterile filtered. For three consecutive days, mice were allowed to drink the antibiotic cocktail ad libitum. The antibiotic water was refreshed daily. After three days of antibiotic water, all mice received autoclaved water for the remainder of the experiment. A single dose of clindamycin (10 mg/kg) was administered by intraperitoneal (IP) injection on the fourth day (24 hours before C. difficile challenge). At this time, groups of five antibiotic-treated mice received neat DMSO, 300 mg/kg CamSA, 50 mg/kg chenodeoxycholate, 300 mg/kg taurocholate, or 300 mg/kg ethylcholate oral gavage. The day of challenge, animals received 10 8 CFUs of C. difficile spores by oral gavage. One hour post challenge, animals received a second dose of the corresponding bile salt or DMSO. A third dose of bile salt or DMSO was administered 24 hours postchallenge. All animals were observed twice daily for signs of CDI. Disease signs were scored using the following rubric: pink anogenital area (score of 1), red anogenital area (score of 2), lethargy (score of 1), diarrhea/increase in soiled bedding (score of 1), wet tail (score of 2), hunchback posture (score of 2), 8-15% loss of body weight (score of 1), .15% loss of body weight (score of 2). Animals scoring 2 or less were undistinguishable from non-infected controls and were considered non-diseased. Animals scoring 3-4 were considered to have mild CDI with signs consisting of pink anogenital area, lethargy, an increase of soiled bedding and minor weight loss. Animals scoring 5-6 were considered to have moderate CDI with signs consisting of mild CDI signs plus red anogenital area and hunchback posture. Animals scoring .6 were considered to have severe CDI and were immediately euthanized. These animals displayed signs described above plus wet tail and severe weight loss. Asymptomatic animals were monitored for up to 14 days post-challenge to monitor CDI onset delay.

Stability of CamSA in Artificial Gastric and Intestinal Juices
CamSA was analyzed for stability in simulated gastric and intestinal juices as published [27]. Briefly, 100 mg CamSA was added to 1 ml of either artificial intestinal or artificial gastric juice and incubated at 37uC. Aliquots were taken at 4, 8, 12, and 24 hours. Samples (1 ml) were spotted on silica TLC plates and allowed to air dry. Plates were developed with 75% ethyl acetate/ methanol. TLC plates were visualized by spraying with a 10% wt/ vol phosphomolybdic acid (PMA)/ethanol solution followed by heating at 100uC for 2 minutes. Quantification of CamSA was determined using a GE Healthcare Typhoon 9410 Variable Mode Imager and analyzed using ImageQuant TL 5.2 software.

Stability of CamSA after Incubation with Bile Salt Hydrolase (BSH) Producing Bacteria
Following previous procedures [28], Escherichia coli DH5a, Bifidobacterium longum (ATCC BAA-999) and Lactobacillus gasseri (ATCC 33323) were incubated for 24 hours at 37uC. Bacterial cultures were then adjusted to an optical density of 1.0 with fresh media supplemented with 6 mM CamSA or taurocholate and incubated at 37uC. Samples of spent media were taken at 4, 8, 12, and 24 hours. Bile salt concentration was monitored by TLC as above. Percent conjugated bile salts were derived by dividing the intensity of TLC spots obtained at different times by the intensity of the TLC spot obtained before incubation.

In vitro Permeability Assays
Caco-2 permeability assays of CamSA were performed by Apredica, LLC (Watertown, MA). Briefly, CamSA was dissolved in DMSO and added to Caco-2 cell cultures to 10 mM final concentration. CamSA was analyzed for both apical to basolateral permeability and basolateral to apical permeability across a Caco-2 cell monolayer. After a 2 hour incubation, CamSA concentrations in the apical and basolateral sides of the Caco-2 monolayers were determined by HPLC-MS. An in vitro ADME-Tox test was also conducted to estimate the percent recovery of CamSA from either the apical to basolateral permeability or basolateral to apical permeability.

Effect of CamSA on Bacterial Growth
Laboratory strains of E. coli DH5a, B. longum, L. gasseri, and C. difficile were individually inoculated from freezer stock onto appropriate agar medium as directed by ATCC. Plates were incubated overnight at 37uC either aerobically (L. gasseri and E. coli) or anaerobically (C. difficile and B. longum). Anaerobic conditions consisted of a 5% CO 2 , 10% H 2 , and 85% N 2 environment. Single cell clones were carefully selected and used to inoculate 5 mL of liquid medium. Inoculated broth was shaken at 37uC for approximately four hours until optical density at 580 nm reached 0.8 representing exponential phase of growth. Bacteria were sub-cultured (1:100) into fresh media and individually supplemented with 10 mM CamSA or taurocholate. An increase of optical density at 580 nm (OD 580 ) was used to measure exponential bacterial growth. The OD 580 was recorded at 0, 1, 2, 3, 4, 6, and 8 hours post subculture inoculation. Growth inhibition was determined by comparing optical density of bile salt-treated cultures with untreated control cultures.

Cytotoxicity of CamSA
Vero cells, Caco-2 cells, and murine macrophages J774A.1 were seeded in complete medium (minimum essential medium (MEM), 10% fetal bovine serum (FBS), and 1% penicillin/streptomycin). Cells were grown at 37uC with 5% CO 2 . Cells were detached by incubation with 1 mM EDTA-trypsin for 5 minutes. Complete medium was then added and monolayers lifted with a cell scraper. Cells were recovered at 8006g for 5 minutes at room temperature and the cell pellet was resuspended in fresh complete medium. A sample of cell suspension was treated with trypan blue to determine background non-viable cells [29]. Culture cells were plated in 12-well or 96-well tissue culture plates at a density of 10 5 cells/ml and allowed to attach overnight. Spent medium was removed and fresh complete medium supplemented with 50 or 200 mM CamSA was added. As negative control, cells were treated with complete media supplemented with DMSO. As positive control, cells were treated with complete media supplemented with 10% EtOH. Plates were incubated overnight as described above.
The 12-well plates containing cell cultures were used as a qualitative method to determine cytotoxicity by visual observation of morphological changes, such as cell rounding. Cell viability was also determined by trypan blue dye exclusion staining. The 96-well plates cultured with mammalian cells were used as a quantitative method to determine cytotoxicity using the CellTiter Glo Luminescent cell viability assay. This assay quantitates the concentration of ATP, which indicates metabolically active cells [30]. After overnight CamSA-treatment, the 96-well plates were equilibrated to room temperature for 30 minutes before addition of the CellTiter-Glo reagent. Luminescence was read with an integration time of 1 second per well using a Tecan Infinite 200 plate reader and iControl software. The luminescence signals from cell cultures supplemented with DMSO only were set as 100% cell viability. Percent survival for other conditions was calculated relative to these untreated cells.

C. difficile Toxin-induced Cell Death
C. difficile strain 630 spores were washed five times with nanopure water, heat activated at 68uC for 30 minutes and washed five more times with water. Spore pellets were resuspended in 0.1 M sodium phosphate buffer supplemented with 0.5% sodium bicarbonate (pH 6.0) to an OD 580 of 1.0. Spores were diluted five-fold in BHI broth supplemented with 6 mM taurocholate/12 mM glycine (germination medium), germination medium supplemented with 50 mM CamSA, or germination medium supplemented with 200 mM CamSA. Spore suspensions were incubated anaerobically at 37uC overnight. The following day, C. difficile cells and spores were removed by centrifugation and spent media filter sterilized. In parallel, Vero and Caco-2 cells were cultured in 96-well plates, as described above. The mammalian cell cultures were then exposed to varying concentrations of sterile spent media for 24 hours. Cell viability was determined as before with the CellTiter Glo viability kit. The luminescence signals from cell cultures supplemented with 0% spent media were set as 100% cell viability. Percent survival for other conditions was calculated relative to these untreated cells.

Onset of CDI Signs in Mice
Antibiotic treated mice were challenged by oral gavage with 10 8 CFUs of C. difficile strain 630 spores. Individual groups of five mice were treated with a single 300 mg/kg dose of CamSA at 0, 6, 9 or 12 hours post-spore challenge. A second 300 mg/kg dose of CamSA was administered 24 hours after the first dose. Mice were observed for signs of CDI twice daily and scored accordingly.

Enumeration of C. difficile Vegetative Cells and Spores
Spore challenged animals were treated with 0 or 300 mg/kg CamSA. Cages from infected animals were changed and feces were collected at different time points. Selected animals were sacrificed at different time points and their GI tract observed for signs of disease. Gastrointestinal tracts were removed in blocks and GI tract contents were flushed with autoclaved water. Feces and intestinal contents were weighed and homogenized in autoclaved water. Aliquots of the fecal suspensions were heated to 68uC for 30 minutes. Heated and unheated feces and GI tract contents were serially diluted in water and plated on Clostridium difficile selective agar (CDSA). Plates were incubated anaerobically for 48 hours and colonies were counted to enumerate colony-forming units (CFUs). CFUs obtained from unheated samples represent the sum of C. difficile vegetative cells and spores. CFUs obtained from heated samples represent the number of C. difficile spores only. The presence of C. difficile was verified by PRO disk.

Statistical Analysis
Standard deviations represent at least three independent measures, unless otherwise stated. Recovered CFU and recovered spores represent mean values from a pool of five animals. Student's unpaired t-test was used to determine the significance of difference of means.   Author Contributions