Exploring the association between gut microbiota metabolites and hepatocellular carcinoma via network pharmacology

Introduction

Hepatocellular carcinoma (HCC), a major threat particularly in those with chronic liver disease, ranks sixth in global cancer incidence and fourth in cancer-related mortality. It exhibits marked geographical disparities (higher in Asia and Africa) and a rising prevalence of non-alcoholic fatty liver disease (NAFLD)-related cases, posing a heavy global healthcare burden (1,2). Notably, HCC has substantial etiological heterogeneity; stratifying its etiologies is not merely a research need but a prerequisite for precise clinical management, as different etiologies drive distinct pathogenic pathways, clinical features, and treatment responses—established risk factors including HBV/HCV infection associate with viral genome integration and persistent hepatic inflammation, while emerging drivers including metabolic syndrome, gut microbiota dysbiosis links to hepatic steatosis and immune microenvironment dysfunction (3). Current HCC therapies include curative options (surgical resection, liver transplantation, restricted by strict criteria), minimally invasive local ablation for small tumors, chemotherapy/transarterial chemoembolization (mainstays for advanced/unresectable cases, hampered by toxicity and drug resistance), and immune checkpoint inhibitors (ICIs) with recent breakthroughs but variable efficacy across subgroups (4). Despite these advances, HCC’s “silent progression” leads to over 60% of patients being diagnosed at advanced, incurable stages; the lack of etiological stratification often results in “one-size-fits-all” treatments failing to address subgroup characteristics, underscoring the urgent need to stratify HCC causes, decipher etiology-specific mechanisms (especially gut microbiota-related ones, a key HCC regulator), and identify targeted strategies.

The gut microbiota, often recognized as an underappreciated organ system, has garnered increasing attention over the past few decades (5). The human gastrointestinal tract harbors trillions of microorganisms, including bacteria, fungi, and viruses, with bacteria making up the largest share. The gut microbiota is diverse yet relatively stable, and its shared core microbiome is dominated by two major phyla: Bacteroidetes and Firmicutes (6). This microbial community comprises more than 114 bacterial species and has a total weight of approximately 1–2 kg, hence being dubbed the human body’s “second genome”. The gut microbiota plays a vital role in maintaining human health by contributing to nutrient absorption, energy metabolism, immune regulation, and maintenance of the intestinal barrier (7). Furthermore, beneficial gut bacteria can exert immunosuppressive effects by modulating the host’s immune responses (8).

Notably, the “gut-liver axis” has emerged as a pivotal paradigm for deciphering HCC pathogenesis, with preclinical models and clinical cohort studies confirming its role in bridging gut microbial dynamics to HCC initiation and progression (9). A key mechanism involves gut microbiota-derived short-chain fatty acids (SCFAs)—the major end products of dietary fiber fermentation by anaerobic bacteria—transported via the portal vein (the primary vascular connection between gut and liver) to hepatic tissues. Specifically, SCFAs (particularly acetate) have been shown to regulate hepatic immune homeostasis by inhibiting histone deacetylase (HDAC) activity; dysregulated SCFA levels (often reduced in HCC-associated gut dysbiosis) weaken this protective effect, thereby facilitating HCC progression (10). Furthermore, gut dysbiosis (characterized by reduced beneficial bacteria such as Bifidobacterium and increased pathogenic taxa like Escherichia coli) exacerbates HCC development by disrupting intestinal barrier integrity: lipopolysaccharides (LPS) secreted by pathogenic bacteria permeate the impaired intestinal epithelium into the portal circulation, then bind to Toll-like receptor 4 (TLR4) on hepatic cells to activate the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling cascade. This activation triggers sustained hepatic inflammation and fosters a pro-tumor microenvironment (TME), further accelerating HCC progression (11). Such findings collectively highlight the multifaceted role of the gut microbiota in HCC biology, as underscored by the “gut-liver axis” framework.

Despite these advances, critical knowledge gaps persist in HCC research: Which core microbial taxa—such as beneficial Bifidobacterium and pathogenic Escherichia coli—and key metabolites (including SCFAs, LPS, and secondary bile acid derivatives) govern gut-liver axis homeostasis during HCC progression? Can gut microbiota profiles, detected via fecal or serum samples, serve as non-invasive diagnostic biomarkers for early HCC or predictors of post-treatment recurrence? Systematic investigation of the bidirectional HCC-gut microbiota crosstalk under the gut-liver axis framework is therefore indispensable. Such studies not only promise to unveil novel pathological mechanisms—including how microbial metabolites regulate hepatic inflammation and TME remodeling—but also provide actionable HCC management strategies: probiotic supplementation (such as Lacticaseibacillus rhamnosus) to restore intestinal barrier function, microbiota-targeted dietary interventions (such as fiber-rich diets) to boost SCFA production, and development of drugs targeting microbial metabolite-signaling axes (such as inhibitors of TLR4 to block LPS-induced activation of NF-κB). These approaches offer an interdisciplinary solution to unmet clinical needs in HCC prevention, early detection, and overcoming treatment resistance.

Our work aims to address these gaps by dissecting key microbial-metabolic regulators of HCC-gut crosstalk, with the study workflow shown in Figure 1. We present this article in accordance with the STREGA reporting checklist (available at https://tgh.amegroups.com/article/view/10.21037/tgh-2025-143/rc).

Figure 1 The flowchart of our study. GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; M-S-M-T, microbiota-substrate-metabolite-target; PPI, protein-protein interaction.

Methods

Identification of metabolites and genes

Gut microbiota metabolites and their associated genes were first retrieved from the gutMGene v2.0 database (12). Next, these metabolites were converted to the Simplified Molecular Input Line Entry System (SMILES) format via the PubChem database (13). Potential target genes of the metabolites were subsequently predicted using two databases—STP and SEA—with species restricted to Homo sapiens (14,15). Genes obtained from these two prediction approaches were visualized and analyzed using a Venn diagram, and the union of genes from STP and SEA was chosen as the final set of candidate metabolite target genes. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Identification of disease related genes

Using “Hepatocellular carcinoma” as the search keyword, HCC-related genes were collected from three databases—GeneCards, OMIM, and CTD (16-18). For the GeneCards database, only genes with a relevance score ≥10 were kept for subsequent analysis. The union of disease-associated genes across the three databases was identified through Venn diagram analysis, and then designated as the final set of HCC-related genes.

Protein-protein interaction (PPI) network construction and analysis

The final HCC-related genes and metabolite target genes underwent intersection analysis with metabolite-related genes using a Venn diagram, yielding the set of key genes. These key genes were then uploaded to the STRING database, with the interaction score threshold set to a combined score >0.4 to construct a PPI network (19). Subsequently, the constructed PPI network was visualized, and hub genes were identified through network topology analysis.

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis

Overlapping genes from the preliminary screening were used as the analysis objects, and GO and KEGG enrichment analyses were performed using the “clusterProfiler” package in R. Specifically, GO analysis covers three main categories: biological processes (BP), cellular components (CC), and molecular functions (MF), whereas KEGG analysis focuses on associations with signaling pathways.

During the analysis, the Benjamini-Hochberg method was used to adjust the P values of enrichment results. Screening thresholds were set as follows: adjusted P value <0.05, false discovery rate (FDR) <0.05, and at least 5 core genes per enriched term—aiming to exclude low-reliability enrichment outcomes. Finally, the “enrichplot”, “ggplot2”, and “circlize” packages in R were used to visualize significantly enriched GO terms and KEGG pathways; we also employed Cytoscape software to visualize the associations between certain key genes and pathways.

The evaluation of drug-likeness and toxicity

The SwissADME and ADMETlab platforms were used to evaluate the pharmacokinetic characteristics and toxicity profiles of key metabolites (20). Three metabolites associated with the three hub genes were assessed for their drug-likeness and toxicological traits.

High-potential candidate drug molecules were screened using Lipinski’s Rule of Five, including the following criteria: molecular weight (MW) <500 Da, hydrogen bond donors (HBD) <5, Moriguchi octanol-water partition coefficient (MlogP) <4.15, and polar surface area (PSA) <140 Ų—these criteria were applied to verify the metabolites’ suitability for clinical application. For the toxicity analysis results, risk levels were classified according to specific thresholds: values ≤0.3 were defined as low risk, values ranging from 0.3 to 0.7 as moderate risk, and values ≥0.7 as high risk.

Statistical analysis

All statistical analyses and graphical visualizations were conducted with R software (version 4.4.1) and Cytoscape software (version 3.10.3), with statistical significance set at P<0.05.

Results

The identifications of key genes of gut microbiota metabolites intervening HCC

A total of 226 metabolites with accessible structural information (in SMILES format) and 154 metabolite-associated genes were obtained from the gutMGene database. Using the STP and SEA platforms, 797 and 1,291 target genes were predicted for the 226 metabolites, respectively. Venn diagram analysis revealed a union of 1,518 genes, which was defined as the final set of potential targets for gut microbiota metabolites (Figure 2A).

Figure 2 Identification of gut microbiota metabolites and HCC-related genes. (A) Target genes of metabolites predicted by the STP and SEA databases. (B) HCC-related genes predicted by CTD, GeneCards, and OMIM. (C) Key genes screening. (D) The network of gut-targets-HCC. HCC, hepatocellular carcinoma.

From the CTD database, 535 HCC-related genes were obtained; 1,644 genes were retrieved from the GeneCards database, and 74 genes were obtained from the OMIM database. The union of these three gene sets was defined as potential pathogenic genes for HCC (Figure 2B). Subsequently, the intersection of the obtained metabolite target genes, HCC-related genes, and metabolite-associated genes was calculated, ultimately resulting in 52 overlapping genes (Figure 2C). A network illustrating interactions between gut microbiota, key genes, and HCC was subsequently constructed to visualize the regulatory associations (Figure 2D).

PPI network construction and analysis

The 52 key genes were submitted to the STRING database to construct a PPI network (Figure 3A), and the network was visualized using Cytoscape software—this showed 52 nodes and 593 edges (Figure 3B). According to degree centrality (DC) values, the top three hub genes (IL6, TNF, and TP53) were recognized as the core genes modulated by gut microbiota metabolites.

Figure 3 PPI network analysis. (A) The PPI network of the key genes. (B) The visualization of the PPI network. PPI, protein-protein interaction.

GO and KEGG enrichment analysis

GO and KEGG enrichment analyses showed that the 52 key genes were mainly enriched in three GO categories: BP, such as “response to molecule of bacterial origin”; CC, including “membrane raft”; and MF, encompassing “DNA-binding transcription factor binding” (Figure 4A-4C).

Figure 4 GO and KEGG enrichment analysis of key genes. GO enrichment analysis: (A) bar plot; (B) bubble plot; (C) circular plot. KEGG enrichment analysis: (D) bar plot; (E) bubble plot; (F) the network of targets-pathways. The V shape represents pathway; the rectangle shape represents target. GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Furthermore, these key genes were also significantly enriched in key signaling pathways, such as the PI3K-Akt signaling pathway, MAPK signaling pathway, and Toll-like receptor signaling pathway (Figure 4D,4E). These findings provide insights into the potential mechanisms through which these key genes mediate HCC progression. Finally, a target-pathway interaction network was built to visualize the associations between selected enriched pathways and key genes, where pathways are represented as V-shapes and genes as rectangles (Figure 4F).

The microbiota-substrate-metabolite-target (M-S-M-T) network analysis

Ultimately, a M-S-M-T network was constructed to visualize the intricate interactions between gut microbiota, their metabolic substrates, metabolites, and targets. This network identified three hub genes (IL6, TNF, and TP53), as well as their connections to 8 metabolites, 5 substrates, and 20 gut microbial taxa (Figure 5).

Figure 5 The network of microbiota-substrate-metabolites-target. The red color represents targets; the blue color represents metabolites; the green color represents substrates and the orange color represents gut microbiota.

In this network diagram, targets are marked in red, metabolites in blue, substrates in green, and gut microbiota in orange. Notably, IL6 showed the strongest connectivity with metabolites and substrates. All microbial taxa included in the network were clearly annotated, and these taxa could serve as potential therapeutic targets for HCC.

The evaluation of drug-likeness and toxicity

To ensure the therapeutic safety of key metabolites, toxicity profiling was conducted—this serves as a critical prerequisite for clinical translation. Analysis via SwissADME showed that multiple compounds conformed to Lipinski’s Rule of Five, demonstrating favorable drug-likeness characteristics (Table 1).

Further results from ADMETlab indicated that these gut microbial metabolites exhibited variations in risk parameters, including carcinogenicity, hERG inhibition, human hepatotoxicity (H-HT), and drug-induced liver injury (DILI) (Table 2). Our study identified three gut-derived metabolites—arachidonic acid, docosahexaenoic acid (DHA), and linoleic acid—with minimal major toxicity, which stand out as core drug candidates for HCC. These findings suggest that a subset of the selected metabolites possesses acceptable toxicological profiles and clinical translation potential, albeit requiring further experimental validation.

Discussion

This study adopted a systematic network pharmacology framework that integrating multi-database target mining, PPI network analysis, and pharmacotoxicological validation to uncover the regulatory mechanisms of gut microbiota in HCC. This approach fills a key gap in current research: traditional investigations into gut-HCC crosstalk typically rely on correlative metagenome-wide association studies (MWAS) or single-target analyses, and fail to capture the multi-dimensional, interactive characteristics of microbial metabolism and cancer progression.

By retrieving 226 structurally defined gut microbiota metabolites (in SMILES format) and 154 metabolite-associated genes from the gutMGene v2.0 database, we conducted target prediction using STP and SEA (with species limited to Homo sapiens), ultimately identifying a union of 1,518 metabolite targets. This dual-platform screening approach reduced false negatives and guaranteed comprehensive coverage of potential metabolite-gene interactions—a major advantage over single-database workflows, which frequently overlook weak yet biologically meaningful regulatory associations. This is especially crucial in gut microbiota research, where metabolite effects are often pleiotropic and context-specific.

To link these metabolite targets to HCC pathogenesis—a disease marked by complex genetic dysregulation and progressive therapeutic resistance—we collected disease-associated genes from GeneCards (relevance score ≥10), OMIM, and CTD. This resulted in a union of HCC-associated genes, with 535 derived from CTD, 1,644 from GeneCards, and 74 from OMIM. The subsequent intersection of metabolite targets, HCC genes, and metabolite-associated genes yielded 52 key genes—a core subset that serves as the functional “crossroads” connecting gut microbiota metabolism and HCC progression. A PPI network was constructed for these 52 genes using the STRING database (combined score >0.4), generating 52 nodes and 593 edges; topological analysis further identified IL6, TNF, and TP53 as hub genes.

IL6 acts as a critical bridge between gut microbial imbalance and HCC initiation—gut dysbiosis-induced intestinal barrier impairment promotes systemic translocation of microbial antigens, sustaining IL6-signal transducer and activator of transcription 3 (STAT3) signaling to accelerate hepatocyte malignant transformation (21); TNF, a pivotal pro-inflammatory cytokine, drives HCC’s inflammatory TME, with its expression significantly upregulated by gut microbial metabolites (such as secondary bile acids) via activating the TLR4-NF-κB cascade (22); loss-of-function mutations in TP53 (a core tumor suppressor) are closely linked to advanced HCC progression, and gut dysbiosis—like pathogenic Escherichia coli overgrowth—exacerbates TP53 inactivation by amplifying gut-derived LPS-mediated hepatic oxidative stress and DNA damage (23). Notably, this study extends prior research by contextualizing IL6, TNF, and TP53 within the gut microecosystem framework. Earlier studies focused mainly on these genes’ intrinsic oncogenic/tumor-suppressive roles in hepatic cells, while the present work integrates their regulatory functions with gut microbial metabolism and barrier integrity—an analytical angle not fully addressed in previous HCC research.

GO and KEGG enrichment analyses further clarified the functional relevance of the 52 key genes, shedding light on how these genes connect gut dysbiosis to the core biological features of HCC. In the BP category, enrichment in “response to molecule of bacterial origin” highlights the key genes’ role in mediating gut microbiota signals to HCC cells. These signals—primarily components of pathogenic gut bacteria such as Klebsiella pneumoniae that translocate across the impaired gut barrier to the liver—depend on the key genes to activate downstream NF-κB inflammatory signaling, a core pathway linking gut microbial imbalance to dysregulated hepatocyte proliferation and sustained liver inflammation (24). This finding confirms the HCC TME’s reliance on microbial cues, with key genes acting as molecular bridges for pro-tumor signals.

CC enrichment in “membrane raft” identifies the HCC cell surface as a critical site for gut microbiota signal interactions. Membrane rafts are cholesterol-enriched lipid domains that recruit key signaling proteins like TLR4; studies have confirmed gut-derived bacterial components bind to TLR4 on HCC cell membranes to initiate signal transduction, a process closely tied to tumor progression (25). The key genes’ association with this component suggests they localize to rafts, facilitating signal transduction from extracellular microbial cues to intracellular regulatory networks, which explains why gut dysbiosis correlates with enhanced HCC aggressiveness.

In the MF category, enrichment in “DNA-binding transcription factor binding” indicates the key genes regulate HCC transcriptional programs. Relevant studies have shown gut microbiota-derived metabolites such as butyrate can remodel chromatin accessibility at gene enhancers, recruiting transcription factors like STAT4 to bind target genes and modulate their expression—an interaction mechanism central to HCC immune microenvironment regulation (26). This positions the key genes as central mediators of “gut microbiota-transcription factor-HCC” crosstalk, directly shaping HCC cell fate via transcriptional regulation.

KEGG pathway enrichment analysis revealed that the 52 key genes were significantly enriched in the PI3K-Akt signaling pathway, underscoring their role in HCC progression. The PI3K-Akt pathway is a well-established driver of HCC, with its aberrant activation boosting HCC cell survival and proliferation. Notably, gut microbiota dysbiosis contributes to this activation: Catenibacterium mitsuokai—a bacterium enriched in HCC patients’ feces and tumors—translocates to the liver and secretes quinolinic acid, which binds to the TIE2 receptor on hepatocytes to trigger PI3K-Akt activation. The 52 key genes’ enrichment here suggests they regulate pathway activity, mediating the pro-tumor effects of this gut bacterium on HCC (25).

Enrichment of the 52 key genes in the MAPK signaling pathway further clarifies gut microbiota-HCC crosstalk. The MAPK pathway transduces extracellular signals to drive HCC invasiveness and metastasis. During gut dysbiosis, LPS from Gram-negative bacteria translocate to the liver via a compromised intestinal barrier, binding TLR4 on hepatocytes to activate MAPK signaling—an interaction linked to upregulated invasion-related gene expression. The 52 key genes’ presence in this pathway indicates they participate in LPS-induced MAPK signal transduction, linking gut microbial imbalance to heightened HCC aggressiveness (27).

The 52 key genes’ enrichment in the Toll-like receptor (TLR) signaling pathway strengthens the mechanistic link between gut microbiota and HCC. TLRs sense microbial components, and their hepatic activation fuels chronic inflammation—a precursor to HCC. Klebsiella pneumoniae, enriched in HCC patients’ gut and liver, translocates via the gut-liver axis and binds TLR4 on hepatocytes, triggering downstream inflammatory signaling to shape a pro-TME. The 52 key genes here likely mediate microbial signal transmission to HCC cells, regulating TLR pathway activation and inflammatory TME formation (24).

The M-S-M-T network—incorporating 20 gut microbial taxa, 5 substrates, 8 metabolites, and 3 hub genes—visually delineates these multi-layered interactive associations, while clearly labeling microbial taxa with potential as therapeutic targets. This network overcomes a critical limitation in earlier HCC-focused probiotic research: while multiple in vitro studies have reported decreased HCC cell viability after probiotic supplementation, most were unable to define the specific microbial taxa, metabolites, or genes underpinning this effect. This clarity in linking a given microbial taxon to distinct metabolic pathways and target genes breaks the “black box” constraint of prior probiotic studies, which often relied on broad formulations without mechanistic specificity. With these observable interactions, the network enables the design of more targeted clinical interventions—such as prioritizing these annotated microbial taxa for probiotic applications—rather than using non-specific microbial mixtures, thereby enhancing the translatability of gut microbiota-driven approaches to HCC management.

To assess the clinical applicability of gut microbiota-derived metabolites for HCC management, three candidate metabolites were subjected to a two-phase screening workflow: druggability evaluation via the SwissADME platform and toxicity analysis with the ADMETlab tool. Only metabolites that passed both assessments were included, with their key characteristics and safety profiles documented in the associated tables. This screening framework prioritizes fundamental biological accessibility (through druggability) and secondarily focuses on minimizing clinical risks (through toxicity), laying a targeted foundation for advancing HCC intervention approaches.

Notably, three gut-derived metabolites—arachidonic acid, DHA, and linoleic acid—with minimal major toxicity stand out as core drug candidates for HCC. Their discovery carries significant implications for future research and clinical practice, which can be elaborated as follows:

First, they address unmet therapeutic needs by offering advantages over synthetic drugs, including fewer off-target effects and the ability to target pathways relevant to HCC. For instance, linoleic acid is metabolized by gut bacteria such as Roseburia spp. into conjugated linoleic acid (CLA) precursors, which modulate the liver microenvironment to counteract pro-tumor inflammatory responses (28); arachidonic acid, as a substrate for cyclooxygenase-2 (COX-2) regulated by gut microbiota, links microbial dysbiosis to prostaglandin E2 (PGE2)-mediated immune suppression in HCC—making targeting it a feasible anti-tumor strategy (27). Second, they enable innovative combination therapeutic strategies: their gut-derived nature is consistent with evidence that probiotics can boost endogenous synthesis of such metabolites, addressing the instability issue of single-agent therapies. This aligns with the growing focus on gut microbiome-based HCC care. Third, they advance precision medicine by matching specific patient subgroups: DHA, which is free of hepatotoxicity, benefits patients on polypharmacy protocols (common in HCC patients with comorbid liver disease); derivatives of linoleic acid may be suitable for patients with depleted gut Roseburia spp. populations; meanwhile, modulators of arachidonic Acid target COX-2-high inflammatory tumors. Fourth, they redefine drug development paradigms by circumventing high-risk early-stage development, cutting down translation timelines and costs—thus bridging the gap between in silico predictions and clinical use for HCC.

Despite these strengths, this study has limitations inherent to network pharmacology and in silico research. First, reliance on public databases including gutMGene and GeneCards means identified gut metabolite-gene-microbiota interactions require experimental validation. This step is essential for verifying network predictions, as database-derived regulatory links frequently lack direct experimental evidence— a gap addressed by standard workflows combining in silico screening with in vitro and in vivo verification (29). Second, analysis focused solely on coding genes, omitting non-coding RNAs (miRNAs, sRNAs) that mediate gut microbiota-metabolite regulatory cascades. Gut bacteria actively produce non-coding RNAs to modulate metabolic pathways critical for host colonization and metabolite synthesis, and these molecules are increasingly recognized as key mediators of microbe-host crosstalk. Their exclusion narrows the network’s scope by overlooking a well-documented layer of regulatory complexity (30). Third, the metabolite-microbiota-gene-tumor network ignores inter-individual gut microbiota variability in HCC patients. Diet, age, and medications drive this variability, which directly alters metabolite production and host responses. Such differences can obscure consistent metabolite-gene associations, necessitating prospective cohort validation to enhance clinical applicability (31).

In conclusion, this study elucidates a functional gut microbiota-metabolite-gene axis in HCC, with IL6, TNF, and TP53 serving as central regulators and specific gut-derived metabolites as promising therapeutic candidates. Through the integration of rigorous target screening, network analysis, and pharmacotoxicological assessment, our work offers a framework for developing microbiota-targeted therapeutic strategies. Beyond specific findings, this study advances the wider field of microbial oncology by demonstrating how network pharmacology can bridge correlative microbiome data to mechanistic understanding. It founds a basis for “microbiome-guided precision oncology” in HCC, and future research built on these findings could reshape HCC treatment—by leveraging the gut microbiota as both a therapeutic target and a tool for personalized medicine.

Conclusions

This study adopted a network pharmacology method to explore the potential mechanisms by which gut microbiota metabolites act in HCC. The results offer a comprehensive insight into the involved key genes and how gut microbiota metabolites affect HCC via these genes. Furthermore, this study identified high-potential core genes (IL6, TNF, and TP53) and their associated metabolites, which may provide new directions for the diagnosis and treatment of HCC in subsequent studies.

Acknowledgments

The authors would like to thank the GutMgene, STP, SEA, GeneCards, OMIM, and CTD databases for the open-source data.

Footnote

Reporting Checklist: The authors have completed the STREGA reporting checklist. Available at https://tgh.amegroups.com/article/view/10.21037/tgh-2025-143/rc

Peer Review File: Available at https://tgh.amegroups.com/article/view/10.21037/tgh-2025-143/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tgh.amegroups.com/article/view/10.21037/tgh-2025-143/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Shen KY, Zhu Y, Xie SZ, et al. Immunosuppressive tumor microenvironment and immunotherapy of hepatocellular carcinoma: current status and prospectives. J Hematol Oncol 2024;17:25. [Crossref] [PubMed]
2. Simon TG, Roelstraete B, Hagström H, et al. Non-alcoholic fatty liver disease and incident major adverse cardiovascular events: results from a nationwide histology cohort. Gut 2022;71:1867-75. [Crossref] [PubMed]
3. Tilg H, Adolph TE, Trauner M. Gut-liver axis: Pathophysiological concepts and clinical implications. Cell Metab 2022;34:1700-18. [Crossref] [PubMed]
4. Vogel A, Meyer T, Sapisochin G, et al. Hepatocellular carcinoma. Lancet 2022;400:1345-62. [Crossref] [PubMed]
5. Ago T, Matsuo R, Hata J, et al. Insulin resistance and clinical outcomes after acute ischemic stroke. Neurology 2018;90:e1470-7. [Crossref] [PubMed]
6. Zhou B, Yuan Y, Zhang S, et al. Intestinal Flora and Disease Mutually Shape the Regional Immune System in the Intestinal Tract. Front Immunol 2020;11:575. [Crossref] [PubMed]
7. Tain YL, Hsu CN. Role of the Gut Microbiota in Children with Kidney Disease. Children (Basel) 2023;10:269. [Crossref] [PubMed]
8. Zou YT, Zhou J, Zhu JH, et al. Gut Microbiota Mediates the Protective Effects of Traditional Chinese Medicine Formula Qiong-Yu-Gao against Cisplatin-Induced Acute Kidney Injury. Microbiol Spectr 2022;10:e0075922. [Crossref] [PubMed]
9. Yadav S, Kumar R. Gut Microbiota and the Gut-liver Axis in Hepatocellular Carcinoma: A Comprehensive Review of Pathogenesis and Therapeutic Strategies. Anticancer Agents Med Chem 2025;25:1287-301. [Crossref] [PubMed]
10. McBrearty N, Arzumanyan A, Bichenkov E, et al. Short chain fatty acids delay the development of hepatocellular carcinoma in HBx transgenic mice. Neoplasia 2021;23:529-38. [Crossref] [PubMed]
11. Ram AK, Vairappan B, Srinivas BH. Nimbolide attenuates gut dysbiosis and prevents bacterial translocation by improving intestinal barrier integrity and ameliorating inflammation in hepatocellular carcinoma. Phytother Res 2022;36:2143-60. [Crossref] [PubMed]
12. Qi C, He G, Qian K, et al. gutMGene v2.0: an updated comprehensive database for target genes of gut microbes and microbial metabolites. Nucleic Acids Res 2025;53:D783-8. [Crossref] [PubMed]
13. Gao FY, Chen XF, Cui LX, et al. Gut microbiota mediates the pharmacokinetics of Zhi-zi-chi decoction for the personalized treatment of depression. J Ethnopharmacol 2023;302:115934. [Crossref] [PubMed]
14. Su S. Revealing the impact of gut microbiota-derived metabolites on depression through network pharmacology. Artif Cells Nanomed Biotechnol 2025;53:327-42. [Crossref] [PubMed]
15. Cao M, Huang P, Xu LS, et al. Analysis of gut microbiota-derived metabolites regulating pituitary neuroendocrine tumors through network pharmacology. Front Pharmacol 2024;15:1403864. [Crossref] [PubMed]
16. Huang J, Yang Q, Zeng L, et al. Multi-omics network toxicology reveals the role of benzo[a]pyrene in ovarian cancer: Integrating gut microbiota dynamics and Mendelian randomization. Ecotoxicol Environ Saf 2025;302:118601.
17. Zhang F, Xu W, Tang Q, et al. The identification of metabolites from gut microbiota in autism spectrum disorder via network pharmacology. Sci Rep 2025;15:31765. [Crossref] [PubMed]
18. Wang D, Lan J, Meng W, et al. Co-exposure to triclosan and polystyrene nanoplastics on neurodevelopmental toxicity and gut microbiota dysbiosis in zebrafish (Danio rerio). Environ Pollut 2025;381:126643. [Crossref] [PubMed]
19. Shi L, Chen L, Jin G, et al. Si-Ni Decoction as a Potential Treatment for Ulcerative Colitis: Modulation of Gut Microbiota and AKT1 Inhibition Through Network Pharmacology and in vivo Validation. J Inflamm Res 2025;18:6263-80. [Crossref] [PubMed]
20. Xiao B, Chen X, Zong R, et al. Network pharmacology-based insights into the role of gut microbiota metabolites in insulin resistance. Front Microbiol 2025;16:1617496. [Crossref] [PubMed]
21. Nenu I, Toadere TM, Topor I, et al. Interleukin-6 in Hepatocellular Carcinoma: A Dualistic Point of View. Biomedicines 2023;11:2623. [Crossref] [PubMed]
22. Kang Y, Cai Y, Yang Y. The Gut Microbiome and Hepatocellular Carcinoma: Implications for Early Diagnostic Biomarkers and Novel Therapies. Liver Cancer 2022;11:113-25. [Crossref] [PubMed]
23. Khor AHP, Koguchi T, Liu H, et al. Regulation of the innate immune response and gut microbiome by p53. Cancer Sci 2024;115:184-96. [Crossref] [PubMed]
24. Wang X, Fang Y, Liang W, et al. Gut-liver translocation of pathogen Klebsiella pneumoniae promotes hepatocellular carcinoma in mice. Nat Microbiol 2025;10:169-84. [Crossref] [PubMed]
25. Zhang Y, Liu W, Wong CC, et al. Catenibacteriummitsuokai promotes hepatocellular carcinogenesis by binding to hepatocytes and generating quinolinic acid. Cell Metab 2025;37:1998-2013.e7. [Crossref] [PubMed]
26. Zhang M, Huang X, Zhang Y, et al. Gut microbial metabolite butyrate suppresses hepatocellular carcinoma growth via CXCL11-dependent enhancement of natural killer cell infiltration. Gut Microbes 2025;17:2519706. [Crossref] [PubMed]
27. Xiao K, Li K, Xiao K, et al. Gut Microbiota and Hepatocellular Carcinoma: Metabolic Products and Immunotherapy Modulation. Cancer Med 2025;14:e70914. [Crossref] [PubMed]
28. Devillard E, McIntosh FM, Duncan SH, et al. Metabolism of linoleic acid by human gut bacteria: different routes for biosynthesis of conjugated linoleic acid. J Bacteriol 2007;189:2566-70. [Crossref] [PubMed]
29. Ma Y, Deng Y, Li N, et al. Network pharmacology analysis combined with experimental validation to explore the therapeutic mechanism of Schisandra Chinensis Mixture on diabetic nephropathy. J Ethnopharmacol 2023;302:115768. [Crossref] [PubMed]
30. Chen Z, Yang Y, Chen X, et al. An RNase III-processed sRNA coordinates sialic acid metabolism of Salmonella enterica during gut colonization. Proc Natl Acad Sci U S A 2025;122:e2414563122. [Crossref] [PubMed]
31. Healey GR, Murphy R, Brough L, et al. Interindividual variability in gut microbiota and host response to dietary interventions. Nutr Rev 2017;75:1059-80. [Crossref] [PubMed]