Bile acids impact the microbiota, host, and C. difficile dynamics providing insight into mechanisms of efficacy of FMTs and microbiota-focused therapeutics

ABSTRACT Clostridioides difficile is a major nosocomial pathogen, causing significant morbidity and mortality worldwide. Antibiotic usage, a major risk factor for Clostridioides difficile infection (CDI), disrupts the gut microbiota, allowing C. difficile to proliferate and cause infection, and can often lead to recurrent CDI (rCDI). Fecal microbiota transplantation (FMT) and live biotherapeutic products (LBPs) have emerged as effective treatments for rCDI and aim to restore colonization resistance provided by a healthy gut microbiota. However, much is still unknown about the mechanisms mediating their success. Bile acids, extensively modified by gut microbes, affect C. difficile’s germination, growth, and toxin production while also shaping the gut microbiota and influencing host immune responses. Additionally, microbial interactions, such as nutrient competition and cross-feeding, contribute to colonization resistance against C. difficile and may contribute to the success of microbiota-focused therapeutics. Bile acids as well as other microbial mediated interactions could have implications for other diseases being treated with microbiota-focused therapeutics. This review focuses on the intricate interplay between bile acid modifications, microbial ecology, and host responses with a focus on C. difficile, hoping to shed light on how to move forward with the development of new microbiota mediated therapeutic strategies to combat rCDI and other intestinal diseases.


Introduction
Clostridioides difficile is a Gram-positive, sporeforming, anaerobic pathogen that causes C. difficile infection (CDI), a major healthcareassociated infection with significant morbidity and mortality worldwide.It is classified as an urgent threat by the Centers for Disease Control and Prevention (CDC), based on gaps in research and treatment despite aggressive actions being taken. 1 In 2017, there were approximately 462,100 cases and 20,500 deaths related to CDI. 2 Approximately, half of these cases were classified as community-associated as opposed to healthcareassociated, although community-associated CDI represents only about a fifth of in-hospital deaths. 2 A major risk factor for CDI is antibiotic usage, which kills protective gut microbes that are able to provide colonization resistance. 3,4The standard of care for patients with CDI is treatment with vancomycin or fidaxomicin, with a preference for fidaxomicin due to its preservation of the gut microbiota and reduced incidence of recurrence. 5reatment with vancomycin and to a much lesser degree fidaxomicin continues to alter the gut microbiota allowing C. difficile to reestablish infection, leading to recurrent CDI (rCDI) in approximately 25% of patients after initial treatment. 3,6he definition for rCDI is recurrence of diarrhea and a positive C. difficile test within 8 weeks of completing CDI directed therapy. 7The risk of recurrence can increase by 20-25% for subsequent recurrences. 8,9Fecal microbiota transplantation (FMT), a treatment derived from healthy donor stool, has emerged as an effective therapy for rCDI, with cure rates exceeding 90%. 6The goal of this therapy is to restore colonization resistance that is lost after antibiotic treatment.
Recently, the FDA approved the first microbiota-derived therapeutics for the treatment of rCDI.These microbiota-focused therapeutics source healthy stool in such a way as to reduce the risk of introducing harmful bacteria and aim to standardize the processing of stool for FMT. 10,11roducts such as Rebyota and VOWST are derived from healthy donor stool, with VOWST having an additional ethanol purification to select for spore forming bacteria. 10,11Live biotherapeutic products (LBPs) composed of bacterial consortia, such as MET-2 and VE303, are also being investigated. 12,13As of February 2024, MET-2 is in phase 1 clinical trials for rCDI and ulcerative colitis (UC), while VE303 is in phase 2 clinical trials for rCDI.However, the specific mechanisms underlying the efficacy of FMTs and LBPs as well as the mechanisms through which gut microbes provide colonization resistance are still an area of active investigation.
A proposed mechanism that we hypothesize contributes to the efficacy of microbial-based therapies is microbiota mediated alterations in the gut bile acid pool (Figure 1(a)), which impacts C. difficile, the microbiota, and the host.Bile acids exert antimicrobial effects against C. difficile and other members of the gut microbiota through different mechanisms such as disruption of the bacterial cell membrane. 4,22,23Microbial-derived secondary bile acids are able to inhibit different stages of the C. difficile life cycle, such as spore germination, vegetative outgrowth, toxin expression, production, and activity 4,17,19,22,24,25 (Figure 1b).Bile acids also impact the host immune response through interactions with nuclear receptors like farnesoid X receptor (FXR), Takeda G-protein receptor 5 (TGR5), pregnane X receptor (PXR), and retinoic acid-related orphan receptor γT (RORγT). 26Bile acid-induced activation of these receptors plays a role in immunity, specifically the differentiation of Th 17 and T reg cells. 27here are also non-bile acid mediated microbial interactions through which the gut microbiota may protect against C. difficile colonization, including the production of metabolites and exchange of nutrients between C. difficile and members of the gut microbiota.Other metabolites, such as shortchain fatty acids (SCFAs), including acetate, propionate, and butyrate return after FMT. 280][31] Butyrate was also found to increase C. difficile toxin production and sporulation. 31dditionally, many members of the gut microbiota are able to compete for nutrients, such as amino acids important for Stickland fermentation with C. difficile. 32Whereas others like Enterococcus are able to provide C. difficile with nutrients it requires for growth, including ornithine. 33n this review, we will focus on how bile acid modifications carried out by members of the gut microbiota impact the pathogen, microbiome, and host response with a lens on C. difficile.We will also consider the role of competition of nutrients and metabolite production by the microbiota.We will highlight more recent work that provides evidence for how members of the gut microbiota maybe contributing to the efficacy of FMTs and LBPs for the treatment of rCDI and other intestinal diseases.

Bile acids and C. difficile
4 Their role during digestion is to act as surfactants, allowing for the absorption of dietary fats and vitamins.Once they reach the terminal ileum, ~95% of bile acids are reabsorbed through enterohepatic circulation. 35Gut bacteria modify bile acids during intestinal transit, thereby changing the chemistry and diversity of the bile acid pool.Modifications made by the microbiota include deconjugation of the conjugated amino acid, reconjugation or a swap of the conjugated amino acid, dehydrogenation, dehydroxylation, and epimerization.Changes made to the sterol core, rather than the conjugated amino acid, mark the conversion of a primary bile acid to a secondary bile acid.In-depth reviews that explore bile acid modifications in humans and animals can be seen here. 36,37arious bile acids directly impact multiple stages of the C. difficile life cycle in vitro (Figure 1(b)).Bile acids such as taurocholate (TCA) are germinants of C. difficile spores 14− 34], while other bile acids such as chenodeoxycholate (CDCA) inhibit spore germination 14,15,17,22,38 (Figure 1b).Microbially derived secondary bile acids are able to inhibit C. difficile spore germination, vegetative cell growth, and toxin activity 17,19,20 (Figure 1b).The mechanisms through which bile acids inhibit growth are attributed to their antimicrobial and detergent-like properties, which change based on how they are modified. 34C. difficile is exquisitely sensitive to many of these modifications.Additionally, conjugated and unconjugated forms  [14][15][16] is inhibited by various bile acids (red box). 17,18outgrowth of vegetative C. difficile is impacted by a variety of bile acids (red box). 17,18the production of toxin is inhibited by bile acids (red box) through reducing expression of toxin or toxin activity. 15,17,19bile acids also bind directly to C. difficile toxin, reducing its toxicity in the host (red box). 21abbreviations: AA, amino acid; Ala, alanine; BSH, bile salt hydrolase; bai, bile acid inducible; CA, cholate; CDCA, chenodeoxycholate; DCA, deoxycholate; glu, glutamate; gly, glycine; HDCA, hyodeoxycholate; his, histidine; HSDH, hydroxysteroid dehydrogenase; iDCA, isodeoxycholate; iaLCA, isoallolithocholate; iLCA, isolithocholate; LCA, lithocholate; phe, phenylalanine; ser, serine; tau, taurine; trp, tryptophan; tyr, Tyrosine; UDCA, ursodeoxycholate; αMCA, α-muricholate; βMCA, β-muricholate; ωMCA, ω-muricholate; 3-oxo LCA, 3-oxolithocholate.
of the primary bile acids, cholate (CA) and CDCA, as well as secondary bile acids, deoxycholate (DCA) and lithocholate (LCA), can bind to C. difficile toxin TcdB mitigating its cell toxicity 21 (Figure 1 (b)).The combined repetitive oligopeptide (CROP) region of TcdB is essential for bile acid binding, which causes major conformational changes when bound. 21][41][42][43] Bile acids that are modified by the gut microbiota are associated with successful FMTs, potentially indicating their importance in rCDI (Table 1). 18,28,54,55The gut microbiota encodes a variety of bile acid altering enzymes that act on either the sterol core or the amino acid conjugated to it, including, but not limited to, bile salt hydrolases (BSHs), hydroxysteroid dehydrogenases (HSDHs), and those in the bile acid inducible (bai) operon (Figure 1(a)).These modifications and specifically how they help or hinder C. difficile are discussed in more detail below.

Bile salt hydrolases (BSHs)
BSHs are encoded by a variety of microbial Phyla including Bacillota, Bacteroidota, Actinomycetota, and Euryarchaeota. 56BSHs are traditionally known to cleave host-conjugated amino acids, either glycine or taurine, from conjugated bile acids. 57This first step is often a prerequisite for other bile acid modifications. 57he simple conversion of conjugated bile acids to deconjugated bile acids increases the inhibition of C. difficile vegetative growth, and the bile acid is more inhibitory as the conversions get more complex. 15,58istorically, it was proposed that the purpose of microbial BSHs was to detoxify conjugated bile acids, thereby providing a fitness benefit for colonization in the harsh gut environment by the microbes that Table 1.Efficacy of FMT, microbiota-focused therapeutics, and LBPs on rCDI and their impact on the bile acid pool and the microbiome.

Intervention
Impact on rCDI Impact on bile acid pool Impact on microbiome References FMT 81-93% prevention of C. difficile recurrence in human studies.
• Increase in secondary bile acids including DCA, LCA, and HDCA.
• Reduction in primary MCBAs, loss of histidine, serine, and valine conjugates.
• Increase in secondary MCBAs, increase in glutamate, tryptophan, and tyrosine conjugates.
• Increase in alpha diversity.• Significant reduction in TCA, GCA, and CA.
• Significant increase in DCA and LCA. 48Reduction in conjugated and deconjugated primary bile acids.
• Increase in alpha diversity.
12 encode them. 59Additionally, it was suggested that BSHs provide nutrients in the form of liberated amino acids (either taurine or glycine) for the microbes that encode them or the surrounding gut microbiota. 60However, recent studies leveraging both Gram-positive Lactobacillus and Gram-negative Bacteroides found that bile acids become more toxic to bacteria after deconjugation. 61,62Lactobacillus gasseri and Lactobacillus acidophilus each have a taurinespecific BSH and a glycine-specific BSH. 61Knockouts of these BSHs resulted in increased growth of these strains in vitro in the presence of certain bile acids including conjugated forms of CA, particularly in L. gasseri. 61Utilizing these knockout strains, the production of deconjugated bile acids by these BSHs decreased the membrane integrity of both microbes. 61BSH knockouts of L. gasseri and L. acidophilus also outcompeted their wild-type counterparts in gnotobiotic mouse models. 61BSH knockout strains of Bacteroides thetaiotaomicron behave similarly.In a recent paper, different combinations of B. thetaiotaomicron's bile acid altering enzymes, including two BSHs and one HSDH, were knocked out, and the resulting phenotypes were evaluated. 62SH knockouts of B. thetaiotaomicron showed an increase in growth in the presence of different bile acids, particularly in conjugated forms of DCA, when compared to wildtype. 62BSH knockout strains of B. thetaiotaomicron also showed broad changes in their transcriptome including carbohydrate, amino acid, lipid, and energy metabolism when compared to wildtype. 62Counter to the dogma that BSHs liberate amino acids to use as a nutrient, B. thetaiotaomicron altered the expression of its polysaccharide utilization loci (PULs) in the presence of bile acids rather than amino acid metabolism. 62The specific PUL that changes in expression was dependent on the bile acid B. thetaiotaomicron encountered. 62Most bile acids increased expression of a starch utilization PUL in B. thetaiotaomicron, while the secondary bile acid DCA had the broadest impact by activating PULs involved in the degradation of α-mannans, host and dietary glycans, mucins, rhamnogalacturonan I and II, and starch. 62These findings indicate that BSH mediated changes to bile acids modify the toxicity of bile acids as well as signaling bacteria to utilize different nutrients in the gut.
Recent work has indicated that BSHs may play an additional role in the production of bile acids conjugated with a variety of non-canonical amino acids, outside of glycine and taurine, collectively referred to as microbially conjugated bile acids (MCBAs) or bacterial bile acid amidates (BBAAs). 15,63,64A screen of 70 bacterial species identified 27 different species across multiple Phyla capable of conjugating amino acids to -CA, -CDCA, or -DCA. 65A recent survey of 654 unique BSHs from the microbiota in the Integrated Gene Catalog identified a selectivity loop that dictates substrate preference of BSHs in multiple taxa. 15he 'G-X-G' motif within the selectivity loop prefers taurine conjugated bile acids while the 'S-R-X' motif prefers glycine conjugated bile acids. 15urther characterization of electrostatic interactions of the selectivity loop implicates the 'Y-S-R-G' motif as a preference for aromatic MCBAs. 66A taurine-specific lactobacilli BSH cocktail was given to mice in a CDI model and was able to inhibit spore germination and growth of C. difficile in the small and large intestines.This was associated with the deconjugation of primary bile acids as well as the production of primary MCBAs (i.e., MCBAs with a primary bile acid sterol core). 15These primary MCBAs, specifically alanine, phenylalanine, serine, tryptophan, and tyrosine conjugated CA (Ala/Phe/Ser/Trp/Tyr-CA), glutamate, histidine, tryptophan conjugated CDCA (Glu/His/Trp-CDCA), and phenylalanine-βMCA (Phe-βMCA) inhibit TCA mediated spore germination 15 (Figure 1b).C. difficile vegetative growth was inhibited by Tyr-βMCA and Phe-βMCA. 15MCBAs also inhibit the expression of TcdA, specifically phenylalanine, and tyrosine conjugated CA (Phe/Tyr-CA) as well as phenylalanine and tyrosine conjugated βMCA (Phe/Tyr-βMCA). 15Host conjugated primary and secondary bile acids also impact different stages of the C. difficile lifecycle as seen in Figure 1(b).
Bacteria encoding BSHs are significantly associated with MCBA production, and most bacterial isolates that produce at least one MCBA have a bsh homologue. 63Unconjugated bile acids also upregulate transcription of bsh genes in a strain known to produce MCBAs, Bifidobacterium longum NCTC 11,818. 63BSH inhibitors were able to significantly decrease the production of MCBAs by B. longum in vitro and biochemical assays with the purified BSH encoded by this organism showed both acyltransferase (i.e., TCA to AA-CA) and reconjugation activity (i.e., CA to AA-CA). 63This novel activity of BSHs was also confirmed in a strain of Bacteroides fragilis using genetic knockouts and complementation, confirming this activity occurs in BSHs encoded across Actinomycetota and Bacteroidota. 63In newborn infants, the production of MCBAs increased as their microbiota developed from birth to 1 year and positively correlated with colonization of bsh encoding bacteria. 63n a different study, Clostridium perfringens BSH also demonstrated acyltransferase activity and reconjugation activity. 64This BSH produced MCBAs conjugated with most amino acids except for aspartate and proline. 64Further, in vitro analysis determined 19 of 29 strains of bacteria across Actinomycetia, Verrucomicrobia, Gammaproteobacteria, Bacilli, and Clostridia produced at least one MCBA. 64This activity was especially prevalent in the Lachnospiraceae Family. 44,64Hierarchical clustering of bacterial conjugation profiles indicated little association between phylogeny of these organisms and BSH activity. 64Sequence analysis indicated amino acid substitution at position 82 in the active site of BSHs, separate from the selectivity loop, may dictate the diversity of amino acids that BSHs can conjugate to bile acids to produce MCBAs. 64There is evidence that MCBAs enter enterohepatic circulation and therefore should have an opportunity to interact with a variety of host receptors. 64Though MCBA concentrations are lower than liver conjugated bile acids in healthy individuals, MCBAs have been observed in concentrations equal to or higher than primary bile acids in the stool of human bariatric surgery patients. 64,67In human bariatric surgery, a patient's total MCBA pool concentration was approximately 77.7 μM compared to the total primary conjugated and secondary bile acid pool concentration of approximately 34.2 μM. 64rimary MCBAs, are also enriched in patients with inflammatory bowel disease (IBD), and cystic fibrosis. 68,69In IBD, there is no change in the level of secondary MCBAs (i.e., MCBAs with a secondary bile acid sterol core).This is in contrast to rCDI patients before and after FMT, where there is a shift from high primary MCBAs pre-FMT to high secondary MCBAs post-FMT. 44The BSHs in highest abundance post-FMT were encoded by many members of the Lachnospiraceae Family. 44ifferences have been observed in how C. difficile, members of the gut microbiota, and the host respond to bile acids conjugated with taurine, as opposed to glycine, and now to non-canonical amino acids. 15,19,61,62In light of recent discoveries that challenge conventional wisdom, the precise role of BSHs encoded by the gut microbiota is an area of new investigation and could have major implications for the design of microbiota-focused therapeutics.

Hydroxysteroid dehydrogenases (HSDHs)
HSDHs that act on bile acids are encoded by a variety of organisms including Bacillota, Pseudomonadota, Bacteroidota, Actinomycetota, and Archaea. 44,70,71HSDHs convert hydroxyl groups to ketones at positions 3, 7, or 12 in either the α or β orientation, and 7α-HSDHs are the most studied. 72,73Although they can be components of a larger pathway, such as the bai operon, these enzymes are also found independently in some organisms. 74In some cases, HSDHs have activity on conjugated bile acids. 62It is hypothesized that the keto bile acids produced by HSDHs are further modified to produce epimers or remove the hydroxyl group, and thus are often referred to as intermediates.Recently, advances in mass spectrometry have allowed for the detection of these keto bile acids and epimers in urine and serum, indicating these bile acids enter enterohepatic circulation and may be interacting with host receptors. 75pimers of LCA arising from modifications of HSDHs, including 3-oxo LCA, isoLCA (iLCA), alloLCA (aLCA), 3-oxo aLCA, and isoalloLCA (iaLCA) have been found to be enriched in the stool of centenarians, implicating their role in sustained health. 20These bile acids are bactericidal and cause damage similar to β-lactam antibiotics. 20hese bile acids are also able to decrease carriage of C. difficile and other Gram-positive pathogens invivo. 20Colonization of a prolific producer of these epimers, Odoribacteraceae St21, significantly increased fecal iaLCA levels while reducing C. difficile shedding to non-detectable levels in a mouse model genetically modified to have similar bile acid profiles to humans. 20n vitro studies have shown that 3-oxo LCA, iLCA, and iaLCA effectively inhibit the growth of C. difficile 19,20 (Figure 1(b)).As each modification is made, there is a decrease in minimum inhibitory concentration (MIC) of C. difficile.While CDCA exhibits an MIC of 1.25 mM, this value drops to 0.1 mM with the conversion to 3-oxo LCA, and even further to 0.03 mM and 0.02 mM with the subsequent conversions to iLCA and iaLCA, respectively. 19Even at subinhibitory concentrations, iLCA and iaLCA significantly reduce toxin expression by C. difficile. 19Moreover, these bile acids, along with their precursor LCA, diminish the toxin activity of C. difficile in vitro. 19iLCA and iaLCA also exhibit a more pronounced impact on the growth of C. difficile compared to other commensal gut microbes, particularly Gramnegatives, while causing minimal effects on host cell viability. 19SDHs can also function on cholesterol-derived steroids.Gut microbe-encoded 3β-HSDH can degrade estradiol and testosterone, leading to depression in patients. 76,77Encoding HSDH provides a range of fitness benefits to microbes, including detoxifying bile acids and being able to use bile acids as electron donors for the electron transport chain. 72,78,79In patients undergoing FMT for rCDI, significantly higher levels of HSDH-modified bile acids such as 3-oxo LCA, iLCA, and 12-oxo LCA were observed post-FMT compared to pre-FMT. 44andom Forest Analysis showed that these bile acids were among the most significantly different and important metabolites. 44As our understanding of HSDHs evolves, these enzymes continue to be crucial for altering cholesterol-derived biomolecules in the gut.

The bile acid inducible (bai) operon
The bai operon is encoded by <1% of metagenomeassembled genomes (MAGs) identified in the gut of healthy humans. 80acillota, specifically Ruminococcaceae and Lachnospiraceae, are the most abundant bai operon-encoding bacteria. 80,81he bai operon comprises a series of genes involved in the removal of the 7α-hydroxyl group from bile acids.This operon is comprised of six core genes that carry out 7α-dehydroxylation: baiB, baiA, baiCD, baiE, baiF, and baiH, as well as baiG which encodes a transporter to bring bile acids into the cell. 82,835][86] The removal of the 7α-hydroxyl group results in the conversion of unconjugated primary bile acids, such as CA and CDCA, into the secondary bile acids DCA and LCA, respectively.Recent findings have identified some of the accessory genes associated with the bai operon, namely baiJ and baiP, as having an important role in the 7α-dehydroxylation of CDCA. 87It is known that organisms encoding the bai operon carry out 7α-dehydroxylation, however there are still many unknowns regarding the regulation and the substrate specificity of the different enzymes in this operon. 88ntibiotic-induced changes to the microbiota can modulate colonization resistance against C. difficile in mice and in humans.Often, these changes are due to the loss of 7⍺-dehydroxylating bacteria such as Clostridium scindens ATCC 35704, which has been associated with resistance to C. difficile in a secondary bile acid-dependent manner. 89Additionally, C. scindens ATCC 35704 has been shown to inhibit CDI in patients undergoing hematopoietic stem cell therapy, the radiation from which alters the gut microbiota in a way similar to antibiotics. 89While antibiotics alter the structure of the gut microbiota, they also alter the gut metabolome.Alongside antibiotic-induced decreases in secondary bile acids, glucose, free fatty acids, and dipeptides, there are increases in primary bile acids, amino acids, and sugar alcohols. 4The metabolites that increased post antibiotics in mice were able to support C. difficile TCA-mediated spore germination and vegetative growth with amino acids C. difficile is auxotrophic for, and sugars including mannitol, fructose, sorbitol, raffinose, and stachyose in vitro. 4These findings were further confirmed in a mouse model where broad-spectrum antibiotics resulted in a significant loss of secondary bile acids and members of the Lachnospiraceae and Ruminococcaceae, and an increase in C. difficile germination and outgrowth ex vivo. 22he inhibition of multiple stages of the C. difficile life cycle by secondary bile acids has been reviewed previously. 25,40Briefly, secondary bile acids produced by the bai operon including DCA, iDCA, LCA, iLCA, and hyodeoxycholate (HDCA) were found to inhibit TCA-mediated spore germination, growth, and toxin activity in several clinically relevant strains of C. difficile invitro 17,22 (Figure 1(b)).Work has also been done to determine whether bai-encoding bacteria can produce sufficient quantities of secondary bile acids to inhibit C. difficile in vitro.Four strains of Lachnospiraceae, C. scindens VPI 12708, C. scindens ATCC 35704, Clostridium hiranonis TO-931, and Clostridium hylemonae TN-271 were characterized for their ability to inhibit C. difficile in a series of in vitro assays. 90While C. difficile outcompeted these strains in co-culture, supernatants of both C. scindens strains grown in CA were able to produce enough DCA to inhibit C. difficile growth. 90This was associated with CA induced increased expression of the bai genes in strains of C. scindens, and this was not observed with supernatants of C. hiranonis or C. hylemonae grown in CA. 90 In human studies, bai operon genes are found in significantly higher abundance in patients without CDI than those with CDI. 91Secondary bile acid profiles are also distinct in CDI patients, which have low secondary bile acid abundance compared to noncolonized or asymptomatically colonized C. difficile patients, which had higher levels of secondary bile acids. 92Post-FMT samples in patients being treated for rCDI had a significant increase in baiA genes, primarily those encoded by the Lachnospiraceae Family. 44Post-FMT samples also had a significant increase in secondary bile acids, specifically DCA, LCA, HDCA, epideoxycholate (EDCA), 3-oxo LCA, iLCA, and 12-oxo LCA. 44Of these secondary bile acids that returned post-FMT, targeted metabolomics identified the return of many MCBAs with DCA as the sterol core (Table 1). 44Evaluating the abundance of other secondary bile acid amidates (i.e., MCBAs with a secondary bile acid core), such as those with an LCA sterol core, is still difficult due to the lack of synthesized standards.
Microbiota-focused therapeutics have also been observed to restore the secondary bile acid pool.rCDI patients treated with Rebyota showed a shift from dominance of primary bile acids to secondary bile acids, specifically driven by an increase in DCA and LCA (Table 1). 93Many spore-forming bacteria encode the bai operon, highlighting its role in products that select for spores such as VOWST or communities of spore formers such as VE303.Secondary bile acids and the bacteria that produce them play an important role in restoring colonization resistance against CDI, and it is important that these organisms and the enzymes they encode are further biochemically characterized.

Bile acids and the gut microbiota
Inhibition of growth by bile acids is not unique to C. difficile.Primary and secondary bile acids also exert varying effects on microbial growth across different taxa.As stated previously, bile acids inhibit bacteria by disrupting their membrane integrity. 4,22,23However, the concentration of bile acid required to disrupt membrane integrity can vary greatly by species.For example, glycocholate (GCA), which is generated by the host, has a MIC of 20 mM for B. thetaiotaomicron, Lactobacillus acidophilus, Lactobacillus gasseri, and Staphylococcus aureus. 61,62,94However, once GCA is converted to CA by BSHs, the MIC changes to 10 mM, 5 mM, 2.5 mM, and 20 mM, respectively. 61,62,94When CA is converted to DCA, via 7α-dehydroxylating bacteria, the MIC changes again to 0.625 mM, 10 mM, 10 mM, and 1 mM, respectively. 61,62,94The largest difference in MIC among the strains listed here was observed in the 7α-dehydroxylation of CA to DCA.This variation in response to secondary bile acids highlights how small changes in the bile acid pool can impact the fitness of members of the microbiome.
A recent study using gnotobiotic mice used a consortium of over 100 strains, denoted hCom1a, to demonstrate nuances of niche establishment with and without C. scindens ATCC 35704, and C. hylemonae DSM 15053, the only microbes in this consortium that perform 7αdehydroxylation.By simply removing these two organisms, >100-fold changes in the relative abundance of eight strains in the consortia were observed. 95The removal of the 7αdehydroxylating strains not only resulted in the elimination of 7α-dehydroxylated bile acids, but also an increase in other modifications to the 7-hydroxyl groups such as dehydrogenation and epimerization. 95The taxonomic shift associated with this was very specific, with decreases in Ruminococcus, Roseburia, and Eubacteria and increases in Veillonella, Clostridium, and Dorea.These changes appear to be species specific, as many other organisms at the same genus level in this consortium were not affected to such a degree.
Investigating genes involved in 7αdehydroxylation, namely the bai operon, has been difficult due to the lack of genetic tools for the microbes that encode it.However, recent work has been able to clone a bai operon into C. sporogenes and determined a minimal gene set required to perform 7α-dehydroxylation. 82This allowed for characterization of each enzyme in the reductive arm of 7α-dehydroxylation. 82This opens up the possibility for future studies to include strains with the appropriate background as a control, rather than removing 7αdehydroxylating bacteria or using closely related species.
In fact, most bile acids directly modify the structure of the gut microbial community.Mice exogenously fed host-derived primary bile acids, TCA and GCA, had large changes to the microbial composition of their gut, marked by a significant decrease in the abundance of Verrucomicrobia. 96In addition, mice fed the microbial-derived secondary bile acid UDCA had a significant increase in Lachnospiraceae as well as increases in Lactobacillus, Clostridiales, and unclassified Bacillota alongside decreases in Ruminococcaceae. 97The administration of bile acids to mice significantly influenced both the bile acid pool and the synthesis of host bile acids. 96,98Bile acids given therapeutically to humans are also associated with changes to the microbiota.Patients with colorectal adenomatous polyps treated with UDCA had significant shifts in their microbial community composition, marked by an increase in Faecalibacterium prausnitzii coupled with a decrease in Ruminococcus gnavus. 99CBAs can also differentially impact the microbiota.A prolific MCBA producer Enterocloster bolteae grows better in most MCBAs with CA core than unconjugated CA, 64 although this was not the case for all MCBA producers.For example, Phe-CA decreased the growth of the MCBA producer Lacrimispora indolis while increasing the growth of another MCBA producer of Lactiplantibacillus plantarum, but the opposite was true of TCA. 64BA dependent inhibition was extended to a total of 18 strains and the inhibition of each species was dependent on the amino acid conjugated to CA in vitro. 64Of the MCBAs, Phe-CA and Leu-CA had the strongest levels of inhibition for the most species, stronger than TCA and GCA, indicating that MCBA generation increases the toxicity of bile acids. 64Generally, the greater the hydrophobicity of the conjugated amino acid in the MCBA the greater the antimicrobial properties. 64pecies dependent inhibition by MCBAs was also confirmed in vivo.Mice fed the MCBAs Phe-CA and Ser-CA had significant shifts in the beta diversity of their fecal microbiota, though this effect was subtle and lost when comparing between groups at individual timepoints. 64Changes to the chemical structure of bile acids by gut microbes play an important role in shaping gut ecology at the species level and are an important factor when considering new therapies for CDI and other intestinal diseases.

Bile acids and the host response
Bile acids not only interact with the microbiota but also with the host.These interactions are reviewed in Collins et al. 2023. 100During enterohepatic circulation, bile acids can directly interact with the host through nuclear receptors, such as FXR and PXR, which regulate the production of bile acids in the host.2][103][104] TGR5 also interacts with bile acids. 103,105Among these, FXR and TGR5 are the only receptors whose known agonists are solely bile acids, and they both react differently depending on which bile acid is bound. 101,103The differentialbinding ability of bile acids to host receptors has been reviewed by Fiorruci et al. 2021. 104XR has differential activation depending on the bile acid it encounters in the order of CDCA>CA>LCA>DCA. 104,106Both murine and human FXR are activated by MCBAs, again with varying responses to each bile acid, the highest activation is with CDCA conjugates. 63ncreased FXR signaling is associated with successful FMT, though FXR signaling does not appear to be significantly altered in patients with primary CDI. 32,45XR activation regulates the production of bile acids by the host through controlling the expression of the gene that encodes the rate limiting enzyme of host bile acid synthesis: Cyp7a1. 101XR may also modulate the nutritional environment in the gut, specifically the amount of taurine present through increase of host taurine synthesis and bile acid amidation. 107This plays an important role in modulating blood cholesterol levels and the amount of bile acid present in the gut.Diet can also modulate bile acid levels, for example dietary polyphenols modulate bile acid metabolism and signaling pathways. 108Not only is FXR important in regulating bile acid synthesis, but it has been shown to be important in modifying the immune response.FXR can activate the NLRP3 inflammasome in inflammatory macrophages, thereby modulating aspects of the innate immune system. 109XR is also able to stabilize the transcriptional regulator, NCoR, on the NF-kB binding site of IL-1B, preventing its transcription and reducing the inflammatory response. 110Certain diseases, such as IBD, also have higher expression of FXR increasing sensitivity to fiber-induced changes to microbiotaderived bile acids such as increased CA and CDCA. 111Treatments targeted at modifying the microbiota to restore bile acid metabolism will also directly impact the host through receptors such as FXR.
Bile acids, while toxic to certain microbes such as C. difficile, can also be toxic to the host.PXR can be activated by various xenobiotics including pharmaceuticals, nutraceuticals, dietary factors, environmental chemicals, as well as secondary bile acids. 112PXR regulates the gut liver axis through bidirectional interactions with the gut microbiome under different conditions, to reduce the amount of toxic bile acids.MCBAs, particularly those conjugated to glutamate, have been observed to interact with PXR in mouse organoids and human cell culture reporter models. 63ORγT is particularly interesting with regard to CDI as this bile acid receptor modulates T cells and is activated differently depending on which bile acid is present. 113The microbial-derived bile acid isoDCA promotes the generation of colonic FOXP3 regulatory T cells expressing RORγT through interaction with dendritic cells. 113ORγT also plays a role in the differentiation of Th 17 and T reg cells in response to microbially derived bile acids. 27Diet also plays a role in RORγT signaling. 102In mice, a change in diet due to weaning alters RORγT signaling and was associated with an increase in Clostridia, a class of organisms rich in 7α-dehydroxylation of bile acids. 114Diet is also important as malnutrition has also been observed to increase the number of cells expressing RORγT. 115Increases in RORγT cells have been observed in mouse models of FMT for IBD. 116everaging the microbiota to alter bile acids to modulate nuclear receptors may be able to impact inflammation and immunity in patients with CDI.However, it is essential to acknowledge that bile acids are not always beneficial to the host.Certain bile acids or an excess of bile acid such as CDCA, DCA, and LCA have been observed to increase inflammation in the gut, so it will be important to be able to control the concentration delivered. 117- 119Inflammation, including that resulting from CDI, has wide reaching impacts including increasing host-derived microbial nutrients in the gut, impacting the colonization of microbes, and affecting the metabolome. 119,120n-bile acid dependent microbial interactions and C. difficile Cross feeding, or the production of essential nutrients for C. difficile by gut microbes, is important to consider when studying how C. difficile establishes colonization.Amino acids are some of the most important nutrients to consider in cross-feeding, as C. difficile is auxotrophic for six amino acids: cysteine, isoleucine, leucine, proline, tryptophan, and valine. 121Most cases of cross feeding involve C. difficile using nutrients produced by other gut microbes while generating minimal nutrients in return. 122In silico models predict that C. difficile consumes amino acids produced by commensal organisms such as Bifidobacterium longum, including proline, glutamate, leucine, tyrosine, alanine, serine, glutamine, and methionine while only providing aspartate in return. 122Enterococci cross feeds C. difficile with amino acids including leucine and ornithine, resulting in a high C. difficile load and worse disease in a mouse model of CDI. 33hese findings were mirrored in CDI patients colonized with vancomycin-resistant Enterococcus (VRE). 33While Enterococci is involved in cross feeding, it can also limit the amino acids available to C. difficile.Competition for arginine between Enterococcus faecalis OG1RF and C. difficile causes stress in C. difficile, which in turn increases virulence factor expression and promotes disease. 33. sardiniense also cross feeds C. difficile with amino acids including ornithine, leading to an increase in pathogen biomass and disease in a mouse model.123 Host factors also play an important role in nutrient-mediated support of C. difficile.The gut environment of pre-FMT rCDI patients contains higher levels of host-associated acylcarnitines, which other members of the gut microbiota can use for growth.124 This suggests that C. difficile may be engaging in cross-feeding by liberating host molecules via toxin production.44,124 Hydroxyproline is the main ingredient of collagen and is an amino acid-rich nutrient that C. difficile likely liberates from the host via toxin mediated inflammation.Toxin mediated inflammation disrupted collagen networks, which supported C. difficile growth.120 Hydroxyproline is converted to proline, a key nutrient for C. difficile, by hypD and proC, genes widely found in many commensals and in C. difficile. 125he presence of hydroxyproline affects the metabolic gene expression of both C. difficile and commensal Clostridia strains including C. scindens VPI 12708, C. hylemonae TN 271, and C. hiranonis TO 931, suggesting that it influences nutrient competition and adaptation within the gut environment.125 Specifically, proline reductase genes are upregulated in C. difficile, while hydroxyproline is present in vitro.125 This increase in proline reductase genes was not consistent across the commensal species tested indicating hydroxyproline may be preferred by specific species.125 A mouse model comparing wild type C. difficile to hypD knockout C. difficile noted changes in the microbiota, mainly as an increase in Lachnospiraceae.125 Nutrient competition between members of the gut microbiota and C. difficile is another mechanism by which colonization resistance can be maintained.Gut communities with members that compete with pathogens for key nutrients have been shown to restore colonization resistance against Klebsiella pneumoniae and Salmonella enterica serovar Typhimurium in a mouse model.126 Since C. difficile and many other Bacillota use Stickland fermentation to generate energy, substrates such as glycine, isoleucine, leucine, proline, and hydroxyproline are often the cause of competition between microbes.121,[127][128][129] In a mouse model of CDI, competition for Stickland metabolites between C. difficile and commensals like C. hiranonis 10542, Clostridium leptum ATCC 29065, or C. scindens VPI 12708 is enough to prevent CDI-related weight loss.32 As these organisms encode a bai operon, which may be producing inhibitory bile acids, the ability for these organisms to compete with C. difficile and prevent CDI-related weight loss was validated in a knockout mouse that does not produce CA. 32 here was also an increase in Stickland metabolites, namely 5-aminovalerate which is a byproduct of proline fermentation, in germfree mice monoassociated with the commensal strains.32 In another study, monocolonization of germ-free mice with the C. scindens ATCC 35704 prior to C. difficile challenge delayed clinical signs of disease and colonic damage for a few weeks, and did not prevent CDI. While these strains contain the bai operon and, therefore, may produce inhibitory secondary bile acids, other strains such as Paraclostridium bifermentans that do not encode the bai operon also compete with C. difficile, and increases survival in a CDI mouse model, indicating that competition for nutrients between C. difficile and its close relatives may prevent or reduce disease.32,123 Non-toxigenic strains of C. difficile are also able to colonize the gut of rCDI patients and reduce CDI recurrence, likely by competing for the same nutrients.132 Considering that many of these Bacillota are lost during antibiotic treatment, they are a population which therapies often seek to reestablish with the hope of increasing colonization resistance.For example, a strain of C. scindens ATCC 35704 produces an antimicrobial alkaloid derived from L-tryptophan and oxaloacetaldehyde, 1-acetyl-β-carboline, that inhibits C. difficile.133 Another microbial metabolite, Urolithin A, has been observed to reduce the expression of C. difficile toxin and repair epithelial damage.134 These metabolites and the bacteria that produce them are important to consider when designing novel therapeutics against C. difficile.

Rational design and standardization of microbiota-focused therapeutics to combat rCDI
While FMT serves as a last resort in treating patients with rCDI, the lack of knowledge of specific mechanisms that mediate its success impedes the development of microbiota-focused therapeutics.Defining which bacteria are needed for a successful FMT is a challenge.We are still not sure if the ability of donor strains to inhabit the recipient after an FMT, a process called engraftment, is required for successful treatment.A recent meta-analysis found that the clinical success of FMTs for rCDI was associated with donor strain engraftment and convergence of microbial species abundance. 135Although this process is taxonomically biased, some taxa have more difficulty engrafting than others. 136However, engraftment is not always required for successful treatment with FMTs or LBPs.Some microbiota-focused therapeutics and LBPs used to treat rCDI do not always show complete engraftment observe increases in beneficial taxa even if they are not included in the formulation. 51,53Additionally, benefits from probiotic strains, such as lactobacilli, can be observed even in the absence of engraftment, as they are passing through, but still providing afunction. 33The complexities of FMT, including the intricacies of engraftment and variable roles of individual bacterial taxa, underscore the need for further research to decipher the mechanisms behind its success.
Attempts to systematically reestablish the functions that are lost after antibiotic treatment with regard to CDI are ongoing.Rebyota (RBX2660) is a product produced by Ferring Pharmaceuticals that achieved FDA approval in 2022 and became the first microbiota-based therapeutic to enter the market.Rebyota is pathogen screened feces that has been filtered after suspension in a solution of saline and polyethylene glycol 3350 mix. 11,137,138ditional screens ensuring a minimal dose of Bacteroides and a maximum dose of polyethylene glycol 3350 were also performed. 11,137,138This product is similar to FMT as it utilizes minimal processing thereby keeping the microbiome of the donor mostly intact.Patients treated with Rebyota have similar changes to the bile acid pool as those observed in FMT studies, a reduction of primary bile acids with an increase in secondary bile acids (Table 1). 48,49While this is a critical first step in standardizing the process of introducing beneficial bacteria to treat rCDI, there are still potentially unnecessary bacteria in this preparation.VOWST, another microbiota-focused therapeutic previously known as SER-109, produced by SERES Therapeutics also attained FDA approval in 2023.VOWST is produced using an ethanol-based purification method to enrich spore forming cells, namely Bacillota. 10,51This method has a similar success rate of 88% for the treatment of rCDI compared to Rebyota 87.1% in their respective clinical trials. 50,52This is similar to the success rates reported for FMT which range from 81% to 93%. 6,46VOWST was able to restore the presence of secondary bile acids after treatment, similar to Rebyota and FMTs (Table 1). 51Patients treated with VOWST also had an increase in the amount of Bacteroides, which was absent from the treatment administered. 51This highlights that a successful microbiota-focused therapeutic does not necessarily have to contain all of the bacteria necessary to treat rCDI.The bacteria administered might play an important role in changing the existing gut ecology, which provides resistance against C. difficile.
While the trajectory of development for products like Rebyota or VOWST is a top-down approach starting with healthy donor feces, LBPs composed of defined consortia aim to build an effective set of microbes from the bottom up.The Nubiyota product, Microbial Ecosystem Therapeutic-2 (MET-2), is comprised of 40 strains of bacteria and has shown promise in phase 1 clinical trials to treat rCDI as well as phase 2 clinical trials for depression and general anxiety disorder. 12,139Similar to FMT and other LBPs, MET-2 is associated with an increase in alpha diversity post-treatment (Table 1). 12VE303, developed by Vedanta Biosciences, is a formulation of eight Clostridia strains.In a recent phase 2 trial, this consortia achieved a rCDI remission rate of 86.2% in the high-dose group, an outcome consistent with FMT and other microbiota-focused therapeutics. 13After treatment with VE303 in patients who recently received antibiotics there was an increase in deconjugated bile acids and secondary bile acids, as well as a decrease in conjugated primary bile acids, indicating a return of microbes encoding bile acid altering enzymes (Table 1). 53In a first of its kind dose-response study on vancomycin-treated patients receiving VE303, the microbiota recovered significantly quicker in high-dose cohorts (8.0 × 10 9 CFU/day) vs low-dose cohorts (≤4.0 × 10 9 CFU/day). 53 common theme among LBPs is smaller communities being biased toward Bacillota in which most 7α-dehydroxylation occurs. 90,140crobiota-focused therapies beyond rCDI FMTs and LBPs have shown promise in other diseases outside of rCDI.Perhaps, the most direct translational diseases include other bowel and liver diseases such as IBD, including Ulcerative Colitis (UC) and Crohn's disease (CD), and NAFLD that are all associated with shifts in the bile acid pool. 141,142The guidelines set by the American Gastroenterological Association (AGA) state FMTs are recommended for the treatment of rCDI in immunocompetent and mildly immunocompromised adults, while microbiota-based therapeutics such as Rebyota and VOWST are recommended for use only in immunocompetent adults, however, no microbiota-based therapeutic is recommended for IBD at this time. 7Understanding the overlap between rCDI and other intestinal diseases will aid in our understanding of how to best leverage microbiotabased therapeutics.IBD is marked by increases in primary bile acids such as CA and CDCA, alongside decreases in abundance of secondary bile acids such as DCA and LCA, similar to changes observed in rCDI. 44,141Changes to the serum bile acid pool in NAFLD also include an increase in primary bile acids and secondary bile acids as well as a decrease in conjugated bile acids overall in serum. 142ncreased levels of secondary bile acids and SCFAs were found in UC patients who responded to FMT in a randomized trial. 143An open label clinical trial comparing gastroscopy and colonoscopy for delivery of FMTs to treat CD achieved 66.7% remission rates and increased the taxonomic diversity in the gut. 144In cases where CDI co-occurs with IBD, FMTs have been observed to have a positive effect on both diseases at the same time. 145iver-related diseases such as NAFLD, including nonalcoholic fatty liver (NAFL) and nonalcoholic steatohepatitis (NASH) have shown promise with microbiota-focused therapies.A clinical trial using a consortium of eight probiotic strains, called VSL#3, including Streptococcus thermophilus, Bifidobacterium breve, two strains of Bifidobacteria animalis, Lactobacillus acidophilus, Lactobacillus plantarum, Lactobacillus paracasei, and Lactobacillus helveticus improved NAFLD. 146FMTs have also been evaluated for other diseases in which gut bacterial bile acid metabolism is also thought to be important, such as diabetes mellitus. 47he gut-brain axis is another active area of research in the context of bile acid metabolism.Bile acids share structural similarities with other cholesterol-derived hormones, and as such research in bile acid modifications may be applicable to these structurally similar molecules.One such instance is polycystic ovarian syndrome (PCOS), which is characterized by an excess of androgens. 147In studies using mice colonized with microbiomes from women with PCOS, it was observed that there was a disruption of ovarian function coupled with altered bile acid metabolism compared to mice colonized with a control microbiome. 147As mentioned previously, there are also links to depression, with gut encoded HSDHs being implicated, as well as MET-2 the defined consortia in clinical trials used to treat depression. 76,77,139A reduction in symptoms associated with autism spectrum disorder (ASD) has also been observed alongside changes in blood neurotransmitter levels in open label clinical trials. 148Validation of the open-label trials, ideally through double blinded randomized controlled trials will be critical in further evaluating the efficacy of FMT for these diseases.Other neurological disorders such as Alzheimer's, stroke, epilepsy, Tourette Syndrome, diabetic neuropathy, and Guillain-Barre are also being investigated, although limited evidence for efficacy is currently available. 149

Conclusions and future directions
C. difficile is a major healthcare-associated pathogen with significant morbidity and mortality.CDI is associated with antibiotic usage, which disrupts the native gut microbiota, diminishing colonization resistance against C. difficile.A deeper understanding of the underlying mechanisms facilitating FMTs and LBPs, which aim to reinstate colonization resistance, will allow for targeted approaches to their development.We have reviewed C. difficile's high sensitivity to bile acids, and the microbial modifications that enhance this susceptibility.These modifications, mediated by the gut microbiota through enzymes like BSHs, HSDHs, and those in the bai operon play a pivotal role in altering bile acid profiles.Moreover, these modifications not only influence C. difficile, but the composition of the microbiota and host responses as well.Beyond bile acid modifications, other mechanisms such as cross-feeding, nutrient blocking, and the production of antimicrobial metabolites warrant consideration for optimizing FMT and LBP therapies.Ongoing efforts in this optimization process include the FDA approved Rebyota, a product comprised pathogen screened feces, and VOWST, a product comprised feces enriched for spore-forming bacteria, as well as ongoing clinical trials using defined bacterial consortia such as VE303 and MET-2.These products have been shown to alter the gut microbial community as well as the bile acid pool reflecting changes after an FMT.FMTs and LBPs have demonstrated potential efficacy in treating various other diseases such as IBD including UC and CD, NAFLD, PCOS, ASD, Alzheimer's, stroke, epilepsy, Tourette syndrome, diabetic neuropathy, and Guillain-Barre syndrome.
The diversity of modifications to the bile acid pool has exploded recently to include MCBAs and newer modifications are continuing to be discovered. 150xploring the impact of these novel bile acids in the context of rCDI prevention and treatment is one of the newest uncharted territories, warranting the need for comprehensive studies.Host produced and microbial-derived bile acids play a critical role in inhibiting C. difficile spore germination, growth, and toxin production, and continued evaluation of each bile acid's role in these processes will be critical in developing therapies for rCDI.By leveraging the different antimicrobial effects bile acids have on different members of the microbiota, new therapies have the potential to shape not only pathogen dynamics, but commensal dynamics.This highlights the need to not only evaluate how bile acids impact pathogens, but also commensal organisms in the gut.Alongside these changes to the microbiome, bile acids also interact with host receptors and can mediate immune responses, highlighting their importance in developing therapies for a variety of diseases.More research is needed to understand how MCBAs and other non-canonical modified bile acids interact with host receptors and the consequences this has for host immunity and the gut microbiota.Other than bile acid mediated interactions with C. difficile, the microbiome, and the host, competition for nutrients and cross-feeding in the gut is becoming an important mechanism to evaluate while developing microbiotafocused therapeutics.Investigating the intricate connections between these elements will be important for developing successful microbial-focused therapeutics.

Disclosure statement
C.M.T. consults for Vedanta Bioscicens, Inc., Summit Therapeutics, and Ferring Pharmaceuticals, Inc. and is on the Scientic Advisory Board for Ancilia Biosciences.

FundingA
.S.M was funded by the Molecular Biotechnology Training Program at NCSU (NIH NCSU MBTP T32 GM133366).C.M. T. was funded by the National Institute of General Medical Sciences of the National Institutes of Health under award number [R35GM149222].