Introduction
Primary biliary cholangitis (PBC), formerly known as primary biliary cirrhosis, is a chronic progressive autoimmune disorder predominantly affecting women aged 40–70 years, though recent epidemiological data have shown a gradual increase in male incidence.1 PBC is characterized by immune-mediated destruction of small intrahepatic bile ducts, leading to chronic nonsuppurative cholangitis, ductopenia, and persistent cholestasis. Diagnostic hallmarks include sustained elevation of cholestatic liver enzymes (particularly alkaline phosphatase and γ-glutamyl transpeptidase) and the presence of pathognomonic autoantibodies such as anti-mitochondrial antibodies (AMA), anti-sp110, and anti-gp210.2 The clinical presentation of early-stage PBC is often insidious, with nonspecific manifestations including pruritus, chronic fatigue, and the xerophthalmia-xerostomia complex. Without timely diagnosis and intervention, the disease typically progresses through stages of biliary fibrosis, ultimately culminating in cirrhosis and hepatic failure.1 Ursodeoxycholic acid (UDCA) remains the first-line therapeutic agent, demonstrating significant improvement in long-term prognosis.3 However, approximately 40% of patients exhibit a suboptimal biochemical response to UDCA monotherapy.4
The precise etiology of PBC remains incompletely elucidated, with current hypotheses proposing an intricate interplay between genetic predisposition and environmental triggers.5,6 Genome-wide association studies have identified multiple susceptibility loci, primarily in immunomodulatory pathways, especially those linked to human leukocyte antigen Class II antigen presentation and interleukin (IL)-12 signaling. Studies exploring the relationship between genetic susceptibility and the microbiota reveal that human leukocyte antigen Class II genes may influence gut microbiota composition in PBC patients.7 However, these genetic aberrations are not PBC-specific but common across various autoimmune diseases, suggesting that genetic factors primarily regulate disease susceptibility rather than directly initiating pathogenesis.8–10 Environmental contributors include diverse exogenous agents such as cosmetic chemicals (hair dyes, perfumes, nail polish), tobacco smoke, xenobiotic estrogens, octynoic acid (a structural analog of lipoic acid found in mitochondrial enzymes),11,12 and urinary tract infections.6,13 The common conclusion of these studies is that environmental triggers may disrupt immune tolerance through molecular mimicry of the structure of the E2 subunit of the pyruvate dehydrogenase complex (PDC-E2) in biliary epithelial cells, ultimately leading to AMA-mediated destruction of small bile ducts.
The gut microbiota, a complex ecosystem comprising trillions of microorganisms, plays a pivotal role in maintaining host homeostasis. A well-balanced gut microbiota not only fosters the establishment of immune tolerance to commensal microorganisms during developmental stages but also serves as a biological barrier against exogenous and endogenous pathogens.14,15 Moreover, the microbial community participates in the biosynthesis of essential nutrients such as vitamins, short-chain fatty acids (SCFAs), and amino acids.16,17 Notably, gut microbiota-derived metabolites mediate communication by engaging specific host signaling receptors, thereby regulating systemic physiological functions.18 Previous studies have shown that the gut microbiota is closely associated with liver diseases such as alcoholic liver disease, non-alcoholic fatty liver disease, autoimmune liver disease, and primary sclerosing cholangitis.19–29 Gut dysbiosis contributes to or exacerbates liver injury through multiple mechanisms, including disruption of bile acid (BA) homeostasis,30 immune dysregulation,14,18 impairment of intestinal barrier integrity, and translocation of microbial metabolites.31,32
In this review, we describe how the gut microbiota affects the occurrence and development of PBC, which is of great significance for the treatment of this disease.
We conducted searches on both PubMed and Web of Science in February 2025 using the following search terms: primary biliary cholangitis OR primary biliary cirrhosis OR PBC OR autoimmune liver disease; gut microbiota OR intestinal microbiome OR intestinal microbiota; bile acid OR pathogenesis OR mechanism OR therapy OR treatment. Furthermore, we manually screened additional relevant articles from the bibliographies of retrieved articles.
Dysbiosis of gut microbiota in PBC
Table 1 summarizes recent studies on the gut microbiota in PBC.27,33–45 Several studies from China and Japan have found significant differences in the composition of gut microbiota between PBC patients and healthy controls. Overall, the diversity and species abundance of gut microbiota in PBC patients were significantly decreased. Alterations in the intestinal flora of PBC patients are mainly characterized by reduced abundance of beneficial bacteria (e.g., Faecalibacterium, Sutterella, and Oscillospira) and increased proliferation of potential pathogens (e.g., Veillonella, Streptococcus, and Klebsiella).27,33–37 In addition to comparisons between PBC patients and healthy controls, differences in gut microbiota have also been observed between PBC patients with different disease statuses. Lammert et al. found that the abundance of Weissella was significantly higher in patients with progressive liver fibrosis than in PBC patients without liver fibrosis.38 Shi et al. reported that compared with PBC patients with mild liver function damage and Albumin-Bilirubin (ALBI) Grade 1, those with ALBI Grades 2 and 3 showed a significant increase in Streptococcus and Enterococcaceae abundance and a reduction in Lachnospira. The former are mostly pathogenic bacteria, while the latter is an intestinal commensal involved in dietary fiber decomposition.39 Additionally, Zhou et al. compared the intestinal flora of anti-gp210-positive and anti-gp210-negative PBC patients. Anti-gp210-positive patients, who are associated with worse prognosis, showed a significant reduction in Oscillospiraceae abundance.39 These findings suggest that gut microbiota changes in PBC patients are involved in disease progression. However, alterations in the gut microbiota cannot explain pruritus in PBC patients; as a nonspecific clinical symptom, pruritus correlates positively with serum BA concentration and autotaxin levels.40
Table 1Summary of studies about gut microbiota in PBC
Researchers (year, country) | Participant | Sample | Result |
---|
Kitahata et al.27 (2021, Japan); Zhou et al.35 (2023, China); Wang et al.34 (2024, China); Abe et al.36 (2018, Japan); Chen et al.33 (2020, China); Furukawa et al.37 (2020, Japan); Tang et al.41 (2018, China); Wang et al.34 (2024, China); Lv et al.45 (2016, China) | PBC patients and HC | Small intestinal MAM, stool, Salivary | PBC vs. HC: Gut microbiota: Diversity, richness, and evenness↓. ↑: Bacteroidetes, Proteobacteria, Actinobacteria, Verrucomicrobia, Fusobacteria, Acidimicrobita, Lactobacillales, Enterobacteriales and Enterobacteriaceae, Sphingomonadaceae, Yersiniaceae, Lachnospiracea_incertae_sedis, ucg_010, Pseudomonas, Serratia, Escherichia, Bacteroides, Megamonas, Klebsiella, Haemophilus, Veillonella, Clostridium, Lactobacillus, Streptococcus; ↓: Firmicutes, Clostridiales, Ruminococcaceae, Clostridia, Sutterellaceae, Oscillospiraceae, Bacteroidetes spp, Christensenellaceae, Faecalibacterium, Parasutterella, Coprococcus, Sutterella, Oscillospira. Oral microbe: ↑: Eubacterium, Veillonella. ↓: Fusobacterium. PBC patients with cirrhosis vs. HC: ↑: Proteobacteria, Enterobacteriaceae, Neisseriaceae, Spirochaetaceae, Veillonella, Streptococcus, Klebsiella, Actinobacillus pleuropneumoniae, Anaeroglobus geminatus, Enterobacter asburiae, Haemophilus parainfluenzae, Megasphaera micronuciformis, Paraprevotella clara; ↓: Firmicutes, Acidobacteria, Faecalibacterium, Gemmiger, Streptococcus, Lachnobacterium, Bacteroides eggerthii, Ruminococcus bromii |
Lammert et al.38 (2021, USA); Zhou et al.35 (2023, China); Hegade et al.40 (2019, UK); Shi et al.39 (2024, China) | PBC patients | Stool | Patients with advanced fibrosis vs. Patients without fibrosis: Microbe: Diversity↓, ↑: Weissella. gp210-(+) vs. gp210-(-): ↓: Oscillospiraceae. Patients with pruritus vs. asymptomatic individuals: Gut microbiota: similar. (ALBI Grade 2 & Grade 3) vs. ALBI Grade 1: ↓: Clostridia, β- Proteobacteria, Erysipelotrichia, Lachnospira. ↑: Lactobacillales, Streptococcus, Enterococcaceae. |
Chen et al.33 (2020, China); Tang et al.41 (2018, China); Wang et al.34 (2024, China); Furukawa et al.37 (2020, Japan); Han et al.42 (2024, China); Han et al.43 (2022, China); Liu et al.44 (2024, China) | PBC patients with or without UDCA treatment | stool | After treatment vs. Before treatment: ↑: Bilophila spp (e.g., Bilophila wadsworthia), Bacteroidetes spp, Sutterella spp, and Oscillospira spp. ↓: Haemophilus spp, Streptococcus spp, Pseudomonas spp. Non-responder vs. Responder: Diversity: ↓. ↑: Faecalibacterium, Ruminococcus, Gemmiger, Blautia, Anaerostipes, Coprococcus, Holdemania. Clostridia-low microbiomes vs. Clostridia-high. UDCA non-response rate: ↑. |
Conversely, PBC treatment can also affect gut microbiota composition. Chen, Tang, and colleagues conducted a 12-month trial of UDCA monotherapy in PBC patients. After treatment, the abundances of Bilophila spp., Bacteroidetes spp., Sutterella spp., and Oscillospira spp. significantly increased, while those of pathogenic microorganisms such as Haemophilus spp., Streptococcus spp., and Pseudomonas spp. significantly decreased.33,41 Wang, Han, et al. evaluated biochemical responses to UDCA treatment and found that, compared with complete responders, patients with no or poor response had significantly higher abundances of Faecalibacterium, Ruminococcus, Gemmiger, Blautia, Anaerostipes, Coprococcus, and Holdemania in their gut.42–43,46 In a prospective study, Liu et al. reported that PBC patients with lower baseline abundance of Clostridia had a higher rate of non-response to UDCA monotherapy, suggesting that gut microbial characteristics influence treatment outcomes. These findings demonstrate that gut microbiota has potential as a tool for evaluating drug efficacy.44
Gut microbiota influences the onset and progression of PBC through effects on immune responses, intestinal barrier function, and BA circulation
Transplantation of fecal microbiota from PBC patients into mice induced PBC-like alterations in serum parameters and liver tissue, while administration of quadruple antibiotics reversed splenomegaly and liver inflammation in PBC-model mice,47,48 indicating a direct role of gut microbiota in disease pathogenesis. Although human clinical trials on fecal microbiota transplantation (FMT) are lacking, numerous retrospective and prospective studies have identified distinct gut microbiota profiles between PBC patients and healthy individuals. Metabolomics analyses reveal that these intestinal flora changes correlate with dysregulation of secondary metabolites, including reduced levels of protective metabolites (e.g., secondary BAs and SCFAs) and accumulation of pro-inflammatory metabolites (e.g., primary BAs).33,41,45 Regarding disease onset, gut microbial dysbiosis promotes the production of autoimmune antibodies, triggering immune dysfunction. During disease progression, gut microbiota dysbiosis impacts the host through intestinal barrier disruption, metabolite dysregulation, and bacterial translocation. Notably, gut microbiota dysbiosis is closely linked to BA metabolic dysregulation. These processes ultimately result in bile duct injury and cholestasis, leading to liver fibrosis.
In this paper, we describe how the gut microbiota contributes to disease pathogenesis through four mechanisms: immune dysregulation, intestinal barrier damage, BA metabolic dysregulation, and cholestasis.
Gut microbiota dysbiosis and immune disorders
A characteristic diagnostic feature of PBC is the presence of AMA targeting the PDC-E2, which is widely expressed in mitochondria across various cell types. Intriguingly, strains of rough Enterobacteriaceae exhibit cross-reactivity with AMA. Subsequent studies have identified the cross-reactive epitope as the Escherichia coli 2-oxoglutarate dehydrogenase complex, which shares structural homology with human PDC-E2. This molecular mimicry implies a microbial role in PBC pathogenesis, whereby E. coli infection disrupts immune tolerance and promotes disease onset through autoantibody induction.49–51 Epidemiological evidence supports this hypothesis: PBC patients demonstrate a higher incidence of refractory urinary tract infections, with urinary tract infection onset frequently preceding PBC diagnosis.52,53 Experimental validation in non-obese diabetic, B6 (Idd10/Idd18) mice revealed that E. coli infection induces autoantibodies against murine PDC-E2 and promotes PBC-like histopathology, mechanistically linking enterobacterial infection to AMA production via molecular mimicry, ultimately driving autoimmune bile duct injury.54 Jochen Mattner et al. infected mice with Novosphingobium aromaticivorans via intravenous injection to induce AMA and trigger chronic T cell-mediated autoimmunity against small bile ducts.55Serratia, Yersiniaceae, and Escherichia coli all belong to the order Enterobacteriales; they exhibit similar antigenic structures and are significantly enriched in the gut microbiota of PBC patients, with their abundance positively correlated with serum IgG levels.34,35,41,45 Additionally, Pseudomonas overgrowth in the ileal mucosa has been linked to antigenic structures cross-reacting with AMA,27 collectively supporting the “molecular mimicry” hypothesis. However, the changes in gut microbiota observed in these studies do not establish causal relationships. Whether alterations in gut microbiota contribute directly to autoantibody production remains to be confirmed by larger controlled trials.
A large number of lymphocytes infiltrate the portal vein area around damaged bile ducts in PBC patients, suggesting that after antigen exposure, autoimmune responses contribute to disease progression.56 Antigen-presenting cells (such as dendritic cells) present autoantigens and induce CD4+ and CD8+ T cells to target the immunodominant epitope of PDC-E2 on biliary epithelial cells, triggering an autoimmune response.55,57,58 Among these, CD103+ tissue-resident memory T cells are the predominant reactive CD8+ T cells and play a key pathogenic role in PBC.59 Additionally, plasma cells secrete AMA that specifically targets the lipid acyl domain of PDC-E2 in bile duct epithelial cells (BECs), forming immune complexes that stimulate liver macrophage aggregation and amplify duct damage.60
Gut-colonizing bacteria such as Bacteroides, Ruminococcus, Faecalibacterium, Clostridium, and Anaerostipes break down dietary fiber or complex carbohydrates to produce SCFAs such as acetate, propionate, and butyrate. These microbial metabolites exhibit immunomodulatory and anti-inflammatory properties through multiple pathways. As endogenous ligands for free fatty acid receptors 2/3 expressed on immune cells, SCFAs mediate systemic immune regulation. Acting as signal transduction molecules, SCFAs bind to various cognate G-protein coupled receptors and regulate immune cells such as neutrophils, macrophages, dendritic cells, mast cells, and lymphocytes.61 In the liver, butyrate attenuates inflammation by suppressing nuclear factor kappa B activation in Kupffer cells, thereby reducing tumor necrosis factor-α, IL-5, and myeloperoxidase production. Furthermore, propionate and butyrate regulate T cell and dendritic cell functions through histone deacetylase (HDAC) inhibition.62–64
Many studies have consistently demonstrated reduced abundance of SCFA-producing taxa in PBC patients and UDCA non-responders.34–37,39,41,44,45 However, these investigations primarily established correlative relationships without elucidating the underlying molecular mechanisms. Notably, Wang et al. demonstrated that butyrate enhances the immunosuppression of myeloid-derived suppressor cells via the HDAC pathway and alleviates bile duct injury in PBC mice.46 Nevertheless, the immunomodulatory roles of other SCFAs (e.g., acetate and propionate) in PBC pathogenesis remain to be experimentally validated.
Gut microbiota dysbiosis and intestinal barrier destruction
Alterations in gut microbiota composition and their metabolites impact the liver by disrupting the first line of defense in the gut-liver axis—intestinal barrier integrity. Kitahata et al. revealed ileal mucosa-associated microbiota dysbiosis in PBC patients, characterized by overgrowth of Sphingomonadaceae and Pseudomonas, alongside reduced abundance of the TM7 phylum and Leptotrichia genera. Notably, Sphingomonadaceae enrichment was undetectable in fecal microbiota, suggesting that ileal mucosal sampling may provide more direct insights into barrier-disrupting taxa than fecal microbiota analysis. However, methodological challenges in ileal mucosal biopsy limited validation across broader populations.27 Functional analyses of enriched gut microbiota in PBC patients revealed aberrant activation of the “bacterial invasion of epithelial cells” pathway, correlating with proliferation of Enterobacteriaceae and Klebsiaceae. These bacteria can translocate to mesenteric lymph nodes and the liver, directly triggering infections and exacerbating barrier permeability.41 Studies by Wang et al. and Lv et al. identified significantly elevated levels of Enterobacteriaceae and Klebsiaceae in PBC patients. Notably, the abundance of Klebsiella was positively correlated with cytokine IL-2, implicating its pro-inflammatory potential.34,45
Under physiological conditions, the host establishes gradients of oxygen and nitrate concentrations, allowing the small and large intestines to be predominantly colonized by aerobic and anaerobic bacteria, respectively, which govern energy metabolism.65 UDCA non-responsive PBC patients exhibit a respiratory mode shift: downregulation of anaerobic fermentation pathways (e.g., butyrate and methane production) and upregulation of aerobic pathways (e.g., nitrate respiration and oxidative phosphorylation).44 Disruption of this respiratory mode transition not only signifies intestinal barrier damage but also drives shifts in microbial composition and metabolite profiles.
As mentioned earlier, gut microbiota-derived metabolites such as butyrate act as anti-inflammatory agents by binding to free fatty acid receptors 2 to enhance regulatory T cell function and inhibit HDAC6/9 activity, thereby promoting anti-inflammatory mediators (e.g., IL-10 and Foxp3).66,67 Butyrate deficiency impairs mucosal repair, and fecal metagenomic/metabolomic analyses of anti-gp210-positive PBC patients with poor clinical prognosis demonstrate significantly reduced abundance of butyrate-producing Faecalibacterium compared to gp210-negative patients, indicating exacerbated intestinal barrier injury.41 Other protective metabolites, including tryptophan derivatives (e.g., indole-3-propionic acid and indole-3-aldehyde) and B vitamins (e.g., biotin and pyridoxal), maintain epithelial tight junctions. Multiple controlled studies report diminished abundance of protective metabolite-producing genera (e.g., Clostridium) in PBC patients.34,44 Intestinal barrier dysfunction and increased permeability facilitate entry of pro-inflammatory metabolites and bacteria into the bloodstream and liver via circulation, causing damage to the liver and bile ducts.
Gut microbiota dysbiosis, BA dysregulation, and cholestasis
Under normal circumstances, BECs are in a quiescent state and form a protective alkaline bicarbonate “umbrella” through anion exchanger 2.68 When anion exchanger 2 is deficient or the alkaline bicarbonate umbrella is damaged, BECs can be stimulated by exogenous or endogenous factors, leading to a series of phenotypic changes. They actively proliferate and release chemokines (such as ICAM-1, tumor necrosis factor-α, CCL2, CXCL9, CXCL10, etc.) that recruit various inflammatory cells, including natural killer cells, natural killer T cells, liver-resident Th1-like cells, mucosa-associated invariant T cells, and hepatic stellate cells, amplifying local inflammation.69 This response represents a self-protective mechanism by which BECs coordinate proliferation and apoptosis to repair damage and remodel the biliary tract. However, if the stimulating factors persist, chronic inflammation leads to destruction and disappearance of small bile ducts, eventually progressing to liver fibrosis.67,70 In a mouse model of drug-induced autoimmune cholangitis, alterations in intestinal flora induce BECs to recruit CD8+ T cells and produce inflammatory damage by secreting chemokines via the Toll-like receptor 2 pathway.71 In another study, PBC mice fed p-Cresol sulfate, a metabolite produced by Clostridium metabolism of tyrosine, showed that p-Cresol sulfate inhibits Kupffer cell immune responses by polarizing Kupffer cells and alleviating bile duct inflammation.72 This illustrates that gut dysbiosis and its metabolites can enter the liver via the mesenteric circulation, stimulate BECs, and induce their phenotypic changes.73
BAs are bioactive molecules regulating metabolic and immune functions. Figure 1 illustrates the enterohepatic circulation. Primary BAs (cholic acid, chenodeoxycholic acid) are synthesized in hepatocytes from cholesterol by the rate-limiting enzyme cholesterol 7α-hydroxylase, which maintains cholesterol homeostasis and protects against hepatic inflammation and fibrosis.74 Free BAs are conjugated with glycine or taurine to form conjugated BAs (e.g., glycocholic acid, taurocholic acid, glycochenodeoxycholic acid, taurochenodeoxycholic acid) and secreted into the intestine, where they facilitate lipid and fat-soluble vitamin absorption, modulate fatty acid metabolism, mediate cholesterol excretion, and regulate gut microbiota to maintain immune balance and intestinal barrier integrity. Approximately 95% of BAs are reabsorbed in the ileum via the apical sodium-dependent bile acid transporter and enter the portal circulation. They are ultimately reabsorbed into hepatocytes via sodium taurocholate co-transporting polypeptide and organic anion-transporting polypeptide 1/4, completing enterohepatic circulation.75–77 Unabsorbed BAs are metabolized by gut microbiota (e.g., Lachnospiraceae, Ruminococcaceae, and Blautia) through 7α-dehydroxylation to generate secondary BAs such as deoxycholic acid and lithocholic acid, enhancing BA pool diversity and hydrophobicity to promote fecal excretion.30,75,78,79
BAs interact with receptors such as farnesoid X receptor (FXR), Takeda G protein-coupled receptor 5, pregnane X receptor, and vitamin D receptor to regulate BA synthesis. For instance, secondary BAs activate ileal FXR, suppressing cholesterol 7α-hydroxylase expression and downregulating hepatic BA production via negative feedback.26,78 Thus, intact BA cycling depends on microbiota-host equilibrium. Impaired microbial generation of secondary BAs or inhibited FXR signaling disrupts enterohepatic circulation, leading to accumulation of hydrophobic and toxic BAs (e.g., deoxycholic acid). Elevated BA concentrations damage cholangiocyte barriers, induce mitochondrial swelling and apoptosis, and ultimately cause bile duct injury and cholestasis.26,31,80 Notably, PBC patients with low Clostridium abundance exhibit marked reductions in secondary BAs.44 Kyoto Encyclopedia of Genes and Genomes pathway analysis confirms dysregulated microbial secondary BA synthesis in PBC, strongly associated with altered BA hydrophobicity and toxic accumulation.35
Dysbiosis not only aggravates cholestasis but is also exacerbated by cholestatic conditions. In patients with advanced liver fibrosis, gut microbiota displays reduced biodiversity and diminished capacity to convert primary BAs into secondary forms, correlating with altered abundance of key genera (e.g., Weissella).38,81 Abundance of protective metabolite-producing genera (e.g., Clostridium) is also decreased in PBC patients.82 In bile duct ligation-induced cholestatic mice, gut microbial diversity is significantly reduced, accompanied by depletion of hepatoprotective metabolites (e.g., biotin, spermidine, arginine, and ornithine), which negatively correlate with pro-inflammatory cytokines (IL-6, IL-23, MCP-1). These functional losses are linked to depletion of beneficial genera (Anaeroplasma, Cyanobacteria, Eubacterium, Lachnoclostridium) and expansion of pathogens (Escherichia, Enterococcus).83