Unraveling the gut-liver axis in autoimmune liver disease overlap syndrome: A multi-omics perspective

Abstract

Autoimmune liver disease overlap syndrome (OS) is a rare and clinically significant condition that has received limited attention in microbiome research. In their recent study, Wang et al combined 16S rRNA sequencing with untargeted metabolomics to characterize the gut-liver axis in OS, identifying shared features of dysbiosis in autoimmune hepatitis (AIH) and primary biliary cholangitis (PBC), and unique signatures, including enrichment of Klebsiella and Escherichia and depletion of aromatic amino acids. In this letter, we critically appraise these findings, emphasizing that OS should be considered a distinct immunometabolic phenotype rather than a simple mixture of AIH and PBC. We discuss the potential mechanistic relevance of the Fusicatenibacter-tyrosine relationship, highlight the clinical implications of integrating microbiota-metabolite analyses, and outline the limitations that future studies must address.

Core Tip: Autoimmune liver disease overlap syndrome, involving features of both primary biliary cholangitis and autoimmune hepatitis, poses diagnostic and therapeutic challenges due to its complex pathophysiology. This letter highlights recent evidence linking gut microbiota alterations and serum metabolite changes to liver function in affected patients. The integration of microbial and metabolic data revealed novel interaction networks that may serve as biomarkers for disease differentiation and progression.

TO THE EDITOR

Autoimmune liver diseases are clinically heterogeneous and immunologically complex conditions that include primary biliary cholangitis (PBC), autoimmune hepatitis (AIH), and their less common but clinically significant confluence: Autoimmune liver disease overlap syndrome (OS). This syndrome presents with combined serological and histological features of PBC and AIH, and it remains one of the most diagnostically challenging entities in hepatology. Patients with OS are at an elevated risk for advanced fibrosis, cirrhosis, and hepatocellular carcinoma, often experiencing more severe clinical trajectories than those with either PBC or AIH alone[1].

Despite advancements in the understanding and treatment of liver disease, the pathophysiology of the OS remains insufficiently understood. Recent developments in systems biology, particularly multi-omics integration (i.e., genomics, proteomics, and metabolomics), offer valuable insights into uncovering underlying mechanisms. In this context, the study by Wang et al[2] contributes significantly by identifying distinct gut microbiota and serum metabolite profiles in patients with autoimmune OS, distinguishing them from those with isolated PBC or AIH[2]. By correlating multi-omics profiles with liver function parameters, this study emphasizes the role of the gut-liver axis in immune-mediated hepatic injury.

What sets Wang et al's study apart is its integrative, network-level analysis of microbial and metabolic data[2]. The observed link between the microbial genus Fusicatenibacter and the amino acid L-tyrosine, in relation to liver enzymes such as aspartate aminotransferase, reflects a potentially functional microbiota-host interaction. Such interactions may not be merely epiphenomena but could reflect the underlying immunometabolic compensation. These insights resonate with findings in related fields, including nonalcoholic steatohepatitis and inflammatory bowel disease, in which gut dysbiosis and altered metabolite signaling are implicated in disease progression[3,4].

The purpose of this letter is to contextualize and critically reflect on the findings of Wang et al[2], explore their diagnostic and translational relevance, and highlight the emerging clinical implications of microbiome-metabolite-liver interactions in autoimmune liver diseases. This study aims to situate this work within the broader efforts to define microbiota-based biomarkers and propose future research directions that may inform therapeutic strategies.

CRITICAL APPRAISAL OF THE STUDY

With the rising interest in gut-liver axis mechanisms, the study by Wang et al[2], published in the World Journal of Gastroenterology in 2025, contributes meaningfully to the growing literature examining microbiota and metabolomic alterations in liver autoimmune disorders. This appraisal evaluates the study’s design, methodology, findings, interpretations, and broader implications while also addressing its strengths and limitations in line with current expectations for systems-level research in hepatology.

Wang et al[2] conducted a retrospective observational study at Beijing Youan Hospital between September 2019 and September 2022, enrolling 32 patients: 16 with PBC, 9 with AIH, and 7 with OS. Diagnoses were confirmed using international guidelines, including histological verification of OS, to ensure high diagnostic rigor. Fecal and fasting serum samples were collected from all the participants. The gut microbiota was profiled using 16S rRNA sequencing on the Illumina HiSeq 2500 platform, with taxonomic classification via SILVA and Greengenes databases, and diversity indices calculated for group comparisons. Serum metabolomics was performed using untargeted liquid chromatography-mass spectrometry/mass spectrometry with compound annotation based on the mzCloud, Kyoto Encyclopedia of Genes and Genomes, and Human Metabolome databases. Differential metabolites were identified using variable importance in projection scores, fold-change thresholds, and statistical significance testing. To visualize patterns, principal component analysis and partial least squares discriminant analysis (PLS-DA) were applied. Importantly, multi-omics integration combined microbial and metabolite profiles with liver function tests through correlation analyses, producing a network model that identified key taxa-metabolite-liver interactions. This approach allowed the authors to move beyond descriptive profiling to explore the functional links between gut ecology, systemic metabolism, and hepatic injury in autoimmune liver disease.

Wang et al[2] reported that patients with OS exhibited markedly reduced microbial diversity compared to those with isolated PBC or AIH, with depletion of beneficial taxa such as Faecalibacterium and Bacteroides and enrichment of pro-inflammatory genera including Klebsiella and Escherichia. At the phylum level, OS was characterized by lower Firmicutes and Bacteroidetes but elevated Proteobacteria and Verrucomicrobia, consistent with a dysbiotic gut profile. Metabolomic analysis identified 26 significantly altered serum metabolites, notably elevated were two compounds with potential anti-inflammatory or immune-regulatory roles [pentadecanoic acid and 5-aminoimidazole-4-carboxamide-1-β-D-ribofuranoside (AICAR)], and reduced levels of aromatic amino acids such as L-tyrosine and L-phenylalanine, which were negatively correlated with the liver injury markers: Aspartate aminotransferase and total bilirubin. Multivariate analyses (PLS-DA, orthogonal partial least squares-discriminant analysis) confirmed distinct clustering of OS patients from PBC and AIH, though with limited predictive power. Importantly, integrative network analysis revealed strong associations between specific microbes and metabolites; for example, Fusicatenibacter correlated inversely with liver dysfunction indices, suggesting a protective role, whereas taxa such as Akkermansia and Parabacteroides were closely linked with aromatic amino acid metabolism. Collectively, these findings support the idea that OS is not a simple overlap of PBC and AIH, but a distinct immunometabolic entity shaped by gut-liver axis disruptions. These observations provide a foundation for a broader discussion of their mechanistic and clinical significance, particularly in the context of gut-microbiota-metabolite interactions.

These findings highlight that the microbial-metabolite interactions observed in OS may have functional consequences rather than representing secondary epiphenomena. One striking example is the inverse relationship between Fusicatenibacter and serum L-tyrosine levels. Fusicatenibacter is a short-chain fatty acid (SCFA)-producing genus known to harbor microbial isozymes that participate in aromatic amino acid metabolism. The reduced abundance of this taxon could therefore limit microbial contributions to tyrosine availability, which in turn may influence host immune regulation. Tyrosine is a critical substrate for catecholamine and neurotransmitter biosynthesis, and its depletion may alter the signaling pathways that modulate hepatic immune tolerance. In autoimmune OS, such depletion could exacerbate inflammatory responses by impairing regulatory T-cell activity and enhancing pro-inflammatory cytokine production. This suggests that the observed tyrosine deficit is not merely a metabolic byproduct but may reflect a pathogenic mechanism linking gut dysbiosis with liver injury. Beyond the mechanistic insights, these patterns have potential clinical applications. Serum metabolites such as pentadecanoic acid and AICAR, together with microbial signatures including Fusicatenibacter and Akkermansia, could form the basis of non-invasive biomarker panels to differentiate OS from isolated PBC or AIH. In addition, the depletion of beneficial taxa and accumulation of pro-inflammatory organisms, such as Klebsiella, support the rationale for microbiota-directed interventions, including diet modification, probiotics, or fecal microbiota transplantation. By framing OS as a distinct immunometabolic phenotype defined by specific microbiota-metabolite relationships, this study encourages a shift toward precision medicine approaches in hepatology.

STRENGTHS AND LIMITATIONS OF THE STUDY

The study by Wang et al[2] had several methodological strengths. Diagnostic accuracy was ensured using internationally accepted clinical and histological criteria for OS, reducing the risk of patient misclassification. The combination of 16S rRNA sequencing with untargeted serum metabolomics provided a robust multi-omics framework that captured both microbial alterations and systemic metabolic changes. The integration of network-based correlation analyses further strengthened this work by linking specific taxa to serum metabolites and clinical markers of liver injury, thereby moving beyond descriptive profiling to explore functional relationships. In addition, the use of multiple reference databases for metabolite annotation increased the reliability of compound identification and enhanced the interpretive value of the results.

However, several limitations constrain the generalizability and mechanistic depth of the findings. Microbial analysis was restricted to the genus level, preventing the identification of strain-specific functions that may critically shape host-microbe interactions. Future studies should incorporate metagenomic or metatranscriptomic sequencing to provide strain-level functional insights. The cross-sectional design limited the ability to infer causality or assess dynamic changes in microbiota-metabolite networks over time, which could be addressed by longitudinal cohort designs that track microbial and metabolic profiles during disease progression and treatment. The relatively small cohort size, drawn from a single institution, further constrains statistical power and external validity, highlighting the need for multi-center studies across diverse and multi-ethnic populations. Moreover, the absence of dietary and lifestyle information represents a missed opportunity, as these factors strongly influence gut microbial composition and metabolic outputs; incorporating standardized dietary assessments or diet-controlled cohorts would strengthen future analyses. Together, these improvements would enhance the interpretive power of multi-omics research in OS and support more robust conclusions regarding causality and clinical applicability.

COMPARISON WITH EXISTING LITERATURE

The study by Wang et al[2] represents a significant advance by applying an integrative multi-omics approach to OS, an entity that has rarely been investigated in prior microbiome research. Earlier studies have focused on either PBC or AIH individually, and several common themes have emerged in this literature. Liu et al[5] and Zhang et al[6] both demonstrated that SCFAs-producing bacteria, such as Faecalibacterium prausnitzii and Roseburia, were consistently depleted in PBC and AIH, respectively, while Yang et al[7] and Luxenburger et al[8] further showed enrichment of potentially pathogenic taxa including Escherichia-Shigella and accompanying loss of microbial diversity. These shifts are associated with altered immune regulation, including Th17/regulatory T cells imbalance, and impaired mucosal homeostasis. Complementary metabolomic studies, such as those by Hao et al[9] in PBC and Sun et al[10] in AIH, reported reductions in fecal SCFAs and perturbations in nucleotide and lipid metabolism, reinforcing the concept of a disrupted gut-liver-immune axis.

Multi-omics investigations have expanded this understanding by linking microbial composition to metabolic function. Trujillo-Gonzalez et al[11] identified disruptions in amino acid and SCFA metabolism across PBC and PSC, whereas Liu et al[12] associated microbiota shifts in PBC with bile acid metabolism and treatment response. Zeng et al[13] reported that reduced microbial network stability correlated with disease severity in chronic liver disease, a finding echoed by Wang et al[2], who extended the approach by employing network centrality metrics to pinpoint influential taxa-metabolite interactions. Similarly, Tang et al[14] and Bezirtzoglou et al[15] highlighted the role of bile acids and drug-microbiota interactions in modulating autoimmune liver injury, although their work did not integrate network-level analyses.

Within this context, Wang et al[2] uniquely contributed by focusing on OS, where they observed shared features of dysbiosis, such as reduced alpha diversity and depletion of commensal taxa, as well as distinctive signatures, including enrichment of Klebsiella and Escherichia and depletion of Akkermansia. Their metabolomic data also revealed marked reductions in aromatic amino acids, particularly L-tyrosine and L-phenylalanine, along with elevations in pentadecanoic acid and AICAR, which are metabolites not consistently described in isolated PBC or AIH cohorts. These observations differentiate OS from related autoimmune hepatopathies, suggesting that they represent not merely an intermediate phenotype, but also a distinct immunometabolic entity. By integrating redundancy analysis, microbe-metabolite correlations, and network centrality mapping, Wang et al[2] advanced the field beyond prior studies that examined either microbial composition or metabolite profiles in isolation, positioning OS as a valuable new focus for biomarker discovery and precision medicine approaches in hepatology.

CLINICAL IMPLICATIONS AND FUTURE DIRECTIONS

Clinical implications

OS poses diagnostic and therapeutic challenges due to its hybrid features of AIH and PBC. Wang et al[2] advanced the field by showing that the OS is not simply a mixture of the two diseases, but instead exhibits distinct microbial and metabolic signatures. Two metabolites, pentadecanoic acid and AICAR, are notable for their translational potential. Both are functionally linked to microbial activity and hepatic inflammation, suggesting that they can serve as accessible biomarkers for early diagnosis or therapeutic monitoring. Incorporating such metabolites into clinical panels may help stratify patients whose biochemical test results remain ambiguous.

Beyond diagnostics, the observed depletion of protective taxa such as Akkermansia and enrichment of pro-inflammatory genera, such as Klebsiella, highlights the rationale for microbiota-directed therapies. Prebiotics, probiotics, and fecal microbiota transplantation warrant investigation as adjunctive treatments for immunosuppression. Importantly, the network analyses used by Wang et al[2] suggest that therapeutic strategies should target not only individual taxa but also microbe-metabolite interactions that drive disease progression. This systems-level perspective could enable precise interventions aimed at restoring ecological and metabolic balance rather than treating organ damage in isolation.

Future directions

Translating these insights requires longitudinal multi-omics studies to establish the causality between dysbiosis, metabolite shifts, and hepatic dysfunction. Functional assays are needed to clarify how pentadecanoic acid and AICAR influence hepatocyte injury and immune regulation. Larger, multi-center cohorts with ethnically diverse patients and detailed dietary data are needed to validate signatures beyond a single population. Finally, clinical trials should test microbiota- and metabolite-modulating therapies whether through dietary interventions, targeted probiotics, or metabolite-mimetic drugs as complements to immunosuppressive regimens.

Together, these directions move clinical hepatology toward an integrated care model that accounts for the immune, microbial, and metabolic dimensions. This approach holds promise for earlier diagnosis, better disease stratification, and more personalized therapies for complex autoimmune liver conditions.

CONCLUSION

The study by Wang et al[2] provided key insights into the pathophysiology of autoimmune liver disease OS by integrating microbiome and metabolomic data. These findings highlight the contributory role of gut-microbiota-host metabolic interactions, expanding the understanding of disease mechanisms beyond traditional immunological paradigms. This work underscores the utility of multi-omics approaches in refining diagnostics, risk stratification, and the identification of novel therapeutic targets. While further validation is warranted, this study supports a shift toward more integrated, mechanism-based strategies for managing complex autoimmune liver diseases.