1. Introduction

Human papillomavirus (HPV) infection is one of the most prevalent sexually transmitted infections globally and poses a substantial threat to public health [1]. HPV can infect human epidermis and mucous membranes, and serves as a major etiological factor for various epithelial tumors, such as cervical cancer, oropharyngeal cancer, and vulvar cancer. Nearly 90% of cervical cancers are directly attributed to persistent HPV infection [2,3]. Statistics show that there are more than 660,000 new cases of cervical cancer worldwide each year, with nearly 348,000 deaths, which poses a serious threat to women’s health [4]. HPV can achieve immune evasion by reshaping the host immune microenvironment, or influence the infection process by interfering with host metabolic pathways [5–7]. While prophylactic vaccines have reduced primary infections in young populations, existing interventions remain limited for already infected individuals, and there is a lack of effective biomarkers to monitor infection progression, especially those linking metabolic perturbations to disease dynamics [8].

In the field of HPV infection, existing studies have predominantly focused on changes in the vaginal microbiota [9]. The vaginal microbiota is undoubtedly the primary determinant of the local microenvironment in HPV infection. While the relationship between HPV and the vaginal microbiota was known to all, the vaginal microbiota primarily reflects the local cervical microenvironment. In contrast, the gut microbiota serves as the body’s largest immune reservoir and a central regulator of systemic metabolism. Its metabolites can enter the circulation and modulate immune responses at distant mucosal sites, including the cervix [10]. Therefore, exploring gut-derived metabolites may help elucidate the gut-cervix axis and its involvement in HPV infection.

However, recent studies have indicated that there is a correlation between the gut microbiota and HPV infection [11]. In recent years, the gut microbiota, referred to as the “second genome” of the human body, includes a diverse community of bacteria, fungi, and viruses, with Firmicutes and Bacteroidetes being the predominant phyla [12]. These microorganisms live in a cooperatively beneficial relationship with the host and play a crucial role in supporting the functional balance of the microbiome [13]. They are involved in processes such as host nutrient absorption, immune regulation, metabolism, and disease progression [14,15]. In a healthy state, they maintain a dynamic balance. Once this balance is disrupted beyond the body’s self-repair capacity, it can exert detrimental effects on the host, including chronic inflammation, autoimmune diseases, and even tumorigenesis [16]. A study found that Bacteroidetes is the main dominant bacterial taxon in the gut of patients with HPV-related cervical cancer [17]. Metabolites of the gut microbiome also play a role in HPV infection. Hydrogen sulfide, a metabolite produced by Gemella, is associated with intestinal barrier dysfunction and systemic inflammation. It can contribute to the persistent development of HPV infection by promoting immune dysregulation and viral persistence [18]. The gut microbiota can also regulate estrogen levels by secreting β-glucuronidase, which indirectly affects the relative abundance of Lactobacillus in the vagina, thereby influencing the progression of HPV infection [19]. Although the beneficial effects of gut microbial metabolites have been extensively studied, their pharmacological mechanisms on HPV have not been fully understood due to the complexity of metabolite-related targets and wide-ranging signaling pathways. Network pharmacology, a systems biology-based in silico approach, offers a new direction for investigating the interactions between targets, metabolites, and signaling pathways, thereby advancing our understanding of the mechanisms involved in HPV infection.

Using network pharmacology, this study investigated the relationships among gut microbiota, gut microbiome metabolites, targets, and HPV infection. The aim was to identify key gut microbiota, their metabolites, targets, and signaling pathways involved in the progression of HPV infection, and to explore gut microbiome metabolites with potential relevance to HPV infection. Based on the limited availability of genotype-specific data in public databases, we focused on HPV infection broadly without stratification by genotype, aiming to initially identify core targets and pathways that may be commonly involved in HPV pathogenesis.

2. Methods

2.1. Study design

Using network pharmacology, this study explored the potential regulatory mechanisms of gut microbial metabolites on the progression of human papillomavirus (HPV) infection. A complex regulatory network between gut microbiota and their metabolites, and HPV infection was constructed, and gut microbial metabolites with therapeutic potential were screened. This work provides a new research direction for the development of intervention strategies against HPV infection (Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flow chart of the study.

Study workflow integrating multi-database screening, network analysis, enrichment analysis, and molecular docking to identify HPV-related gut microbial metabolites.

https://doi.org/10.1371/journal.pone.0346716.g001

3. Results

3.1. Targets of gut microbiota metabolites acting on HPV infection

The gutMGene V2.0 database encompasses 282 human gut microbial species, 278 gut microbiota-derived metabolites, and 238 host targets. These retrieved metabolites were uploaded as SMILES strings to the SEA and STP databases, resulting in 1,397 and 1,316 predicted targets, respectively. Intersection analysis of these datasets identified 925 overlapping targets (Fig 2A).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Identification of potential targets shared between gut microbiota metabolites and HPV infection.

(A) 925 overlapping targets of gut microbial metabolites between SEA and STP databases; (B) 193 common targets between 925 metabolite targets and HPV infection-related targets; (C) 43 common targets between 193 gut microbiota metabolism-HPV infection targets and gut microbiota targets; (D) PPI network from STRING. The network has 43 nodes and 412 edges; (E) Ten core targets of the PPI network. The CytoHubba plugin was used to identify core targets based on the Degree algorithm. The top 10 genes with the highest Degree values were selected as core targets and highlighted in red. In the network visualization, node size and color intensity are proportional to the Degree value, with larger and darker nodes representing higher Degree scores.

https://doi.org/10.1371/journal.pone.0346716.g002

Targets were compiled and deduplicated from the GeneCards Database, OMIM Database, Therapeutic Target Database, and DisGeNET Database, resulting in 927 targets. These targets were cross-referenced with the 925 targets of the aforementioned metabolites, yielding 193 significant targets (Fig 2B). Finally, these 193 significant targets were further intersected with 238 host targets from the gutMGene database, ultimately identifying 43 key targets of gut microbial metabolites involved in persistent HPV infection (Fig 2C).

3.3. GO enrichment analysis

To further evaluate the potential mechanisms of gut microbiome metabolites in therapeutic strategies for HPV infection, the enrichment analysis on the 43 key targets was performed. GO enrichment analysis of these 43 key targets was conducted using the STRING database, with a screening criterion of FDR < 0.05 (S1 Table in S1 Data). The GO analysis results revealed that the significantly enriched Biological Process terms were predominantly associated with cellular stress responses and immune recognition, including response to lipopolysaccharide, cellular response to lipopolysaccharide, cellular response to reactive oxygen species, cellular response to chemical stress, and response to oxidative stress. The cellular components involve NF-kappaB p50/p65 complex, transcription factor AP-1 complex, membrane raft, transcription regulator complex, and ruffle. The molecular functions cover activities such as heme binding, RNA polymerase II-specific DNA-binding transcription factor binding, transcription coregulator binding, NAD(P)H-dependent monooxygenase activity, and DNA-binding transcription factor binding (Fig 3A).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. GO and KEGG enrichment analyses of gut microbiota metabolite-related targets.

(A) GO enrichment analysis of 43 key targets. Bubble size represents gene count. Color gradient reflects -log10(FDR), with darker colors indicating higher statistical significance. (B) KEGG pathway enrichment analysis of 10 core targets. Bubble size indicates gene count. Color represents FDR significance, with lighter colors (mint) corresponding to lower FDR values (more significant enrichment). Pathways are ranked by enrichment signal score.

https://doi.org/10.1371/journal.pone.0346716.g003

3.5. Construction of MMTS network and screening of key components

MMTS network analysis was performed using Cytoscape 3.7.2 software, which identified a total of 130 nodes (97 gut microbiota, 13 gut microbial metabolites, 10 core targets, and 10 pathways) and 225 edges (Fig 4). Network analysis revealed that the most critical signaling pathways include: the Toll-like receptor signaling pathway, which connects with 6 targets (IL1B, NFKB1, JUN, TLR4, IL6, AKT1) and 7 metabolites; the IL-17 signaling pathway, linked to 6 targets (IL1B, NFKB1, CASP3, PTGS2, JUN, IL6) and 8 metabolites; the C-type lectin receptor signaling pathway, associated with 6 targets (NFKB1, IL1B, PTGS2, JUN, IL6, AKT1) and 7 metabolites; and the TNF signaling pathway, connected to 7 targets (IL1B, NFKB1, CASP3, PTGS2, JUN, IL6, AKT1) and 9 metabolites (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Gut microbiota metabolites and core targets of major signaling pathways.

https://doi.org/10.1371/journal.pone.0346716.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. MMTS network.

The network comprises 130 nodes and 225 edges, illustrating the complex relationships between gut microbiota, their metabolites, core targets, and signaling pathways. Blue diamonds represented gut microbiota; Pink triangles represented gut microbiome metabolites; Yellow circle represented targets; Green V-shaped represented Signaling pathways. Node size is proportional to the degree score, with larger nodes indicating greater connectivity within the network.

https://doi.org/10.1371/journal.pone.0346716.g004

3.6. In silico identification of drug-likeness

The ADME parameters of 13 metabolites were analyzed. However, due to the excessive length of the SMILES format for Vancomycin, relevant physicochemical property results could not be obtained from the SwissADME platform. Therefore, a literature review was conducted, which indicated that this metabolite possesses drug-like properties. Additionally, analysis showed that the remaining 12 metabolites all conform to Lipinski’s rule, indicating favorable drug-like physicochemical properties. This suggests that these metabolites may possess the potential to influence HPV infection (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. The physicochemical properties of the metabolites from gut microbiota.

https://doi.org/10.1371/journal.pone.0346716.t002

3.8. Molecular docking validation

In molecular docking, binding free energy serves as a fundamental metric for quantifying intermolecular affinities. A docking binding energy less than 0 kcal/mol indicates that the receptor and ligand can bind spontaneously, suggesting a favorable binding potential. Notably, a binding affinity below −5.0 kcal/mol is generally regarded as a threshold for robust binding potential. A lower binding energy indicates superior conformational stability and higher occupancy of the binding site, which collectively enhance the likelihood of triggering downstream biochemical cascades.

In silico analysis revealed that succinate exhibits robust binding affinity for IL-1B, yielding a docking score of −5.66 kcal/mol (Fig 5), which marginally outperformed the established positive control, quercetin (−5.55 kcal/mol; S1 Fig in S1 Data.). This suggests a favorable interaction between the metabolite and the target protein. To ensure methodological precision, a re-docking validation was performed using the IL-1B crystal structure (PDB: 5R87). The predicted binding pose of the native ligand showed excellent congruence with its experimental orientation, with a RMSD of 0.509 Å. This value, well below the standard 2.0 Å threshold, confirms the fidelity of the force field and the predictive accuracy of our docking parameters.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Molecular docking assays for the core target IL1B-succinate pair.

Succinate binds to the active site of IL1B, forming hydrogen bonds with GLN-197, VAL-248, GLU-141, and LEU-142.

https://doi.org/10.1371/journal.pone.0346716.g005

4. Discussion

HPVs represent one of the most genetically diverse families of viruses infecting humans. Fortunately, among the numerous HPV subtypes, only a small subset of high-risk types possess oncogenic potential. Persistent infection with these high-risk types can lead to epithelial malignancies, including cervical and oropharyngeal cancers, which remain leading causes of cancer-related mortality among women globally [36]. Current clinical interventions for HPV infection remain notably limited: for susceptible populations, invasive treatments such as loop electrosurgical excision, conization, and cryotherapy are commonly employed, or reliance is placed on interferon-α, therapeutic vaccination, with few oral medication options [37]. As “invisible regulators” of host health, gut microbiota and their metabolites may serve as potential intervention targets by modulating the pathogenesis of HPV infection. Therefore, network pharmacology was employed to comprehensively elucidate the complex interactions of gut microbial metabolites in regulating the MMTS network during HPV infection, aiming to identify potential biomarkers for intervening in the progression of HPV infection. Our findings suggest that succinate may represent a metabolite of interest warranting further experimental investigation.

The PPI network identified ten core targets (IL6, AKT1, IL1B, CASP3, NFKB1, EGFR, PPARG, JUN, PTGS2, and TLR4), which are deeply involved in the immune dysregulation, sustained inflammation, and aberrant cell proliferation that characterize HPV infection progression. Among these ten targets, IL6, IL1B, PTGS2, and NFKB1 act as central nodes in pro-inflammatory signaling. IL6, a pleiotropic cytokine, is elevated in HPV-infected cervical tissues and fuels inflammation and malignant transformation, partly via STAT3 activation [38,39]. Similarly, the pro-inflammatory cytokine IL1B can amplify inflammatory cascades through the NF-κB pathway, accelerating virus-induced lesion development [40–43]. PTGS2 encodes cyclooxygenase-2, which amplifies inflammatory responses through lipid mediator synthesis and is induced by inflammatory stimuli in various cell types, thereby potentially contributing to HPV-associated pathogenesis [44]. NFKB1, encoding the p50 subunit of NF-κB, is a critical regulator of inflammation and immunity in HPV infection, and its downregulation can be exploited by the virus to evade host immune responses, fostering viral persistence [45,46]. AKT1, EGFR, and CASP3 are crucial for regulating cell survival and proliferation. The serine/threonine kinase AKT1 is a key component of the PI3K pathway and regulates multiple cellular processes, including proliferation, migration, and invasion [47,48]. It is frequently upregulated in HPV-driven cancers, where it promotes oncogenic progression [49–51]. The E5 protein of HPV can directly bind to and activate EGFR, a tyrosine kinase receptor that promotes cell proliferation and differentiation, highlighting a direct viral mechanism to manipulate host cell machinery [52,53]. Conversely, CASP3, an executive molecule in apoptosis, is often downregulated as HPV lesions progress, allowing infected and transformed cells to evade programmed cell death [54–56]. TLR4, PPARG, JUN modulate the cellular environment. TLR4, a mediator of inflammation and tumorigenesis, can promote tumor growth and shape an immunosuppressive microenvironment post-HPV infection [57–59]. PPARG, a nuclear receptor involved in metabolism and immune regulation, can have its activity disrupted by HPV, leading to metabolic alterations like lipid accumulation [60]. JUN, a component of the activator protein-1 transcription factor, participates in the processes of cervical cell transformation [61]. In summary, this multifaceted target network highlights the potential of gut microbial metabolites to disrupt key pathological processes in HPV infection through coordinated effects on inflammation, cell survival, and the immune microenvironment.

GO enrichment analysis revealed that the 43 key targets were primarily involved in microbial signal recognition and cellular stress responses, including responses to lipopolysaccharide and reactive oxygen species. These processes were closely associated with membrane rafts, the NF-κB p50/p65 complex, and the AP-1 transcription factor complex. Membrane rafts serve as platforms for lipopolysaccharide-induced Toll-like receptor 4 signaling, which activates NF-κB and AP-1 to drive inflammatory gene expression [62]. Previous studies have shown that the HPV 16 E5 oncoprotein upregulates COX-2 through these transcription factors, contributing to cervical epithelial transformation [63]. Concurrently, the enrichment of response to oxidative stress and cellular response to reactive oxygen species reflects the redox imbalance in HPV pathogenesis. HPV E6 and E7 oncoproteins disrupt antioxidant defenses, leading to reactive oxygen species accumulation, DNA damage, and genomic instability that facilitate viral integration [64]. The enriched molecular functions, including heme binding and NAD(P)H-dependent monooxygenase activity, implicate cytochrome P450 enzymes in modulating intracellular redox status [65]. Collectively, these findings suggest that gut microbial metabolites may influence HPV infection by regulating lipopolysaccharide-mediated inflammatory signaling, oxidative stress responses, and their associated transcriptional complexes. KEGG enrichment analysis revealed that gut microbiota and their metabolites primarily exert regulatory effects on HPV infection mechanisms through five signaling pathways: the TNF signaling pathway, Toll-like receptor signaling pathway, C-type lectin receptor signaling pathway, and IL-17 signaling pathway in diabetic complications. Activation of the TNF signaling pathway can induce the secretion of proinflammatory cytokines, creating an inflammatory microenvironment conducive to persistent HPV infection [66]. Inhibition of the Toll-like receptor signaling pathway impairs host antiviral immunity, facilitating immune evasion of HPV [67]. The IL-17 signaling pathway, closely linked to immunity and inflammation, can accelerate lesion progression by promoting cell proliferation [68,69]. These pathways play critical roles in immune regulation, inflammatory amplification, and malignant transformation during HPV infection.

Pharmacological similarity and toxicological analyses of 13 gut microbial metabolites identified succinate as the most promising candidate related to HPV infection progression. Molecular docking further revealed that succinate exhibits a favorable binding capacity to IL1B, suggesting a potential role in IL1B-mediated inflammatory cascades. However, this computational finding requires cautious interpretation. Succinate is a small endogenous metabolite whose well-established immunomodulatory effects are primarily mediated through its specific cell surface receptor, SUCNR1 [70,71]. Activation of SUCNR1 by succinate can trigger pertussis toxin-sensitive Gi/Gq pathways, a mechanism that may relate to the pertussis disease pathway identified in our KEGG enrichment analysis [70]. In the cervical microenvironment during HPV infection, SUCNR1 is activated in macrophages, leading to altered metabolism and succinate accumulation. The released extracellular succinate further promotes IL1B production, amplifying the inflammatory response [72]. Thus, while our docking analysis raises the possibility of a direct succinate-IL1B interaction, its functional relevance remains to be experimentally validated. Alternatively, succinate may exert its effects on IL1B primarily through the canonical SUCNR1-mediated pathway, which is consistent with its established role in immune regulation.

In addition to succinate, other short-chain fatty acids (SCFAs) such as butyrate, propionate, and acetate also emerged from our screening. SCFAs exert immunomodulatory effects through multiple mechanisms, including activation of G protein-coupled receptors (GPCRs) and inhibition of histone deacetylases [73]. In neutrophils, propionate, acetate, and butyrate are the main SCFAs produced by fermentation by bacteria [74]. Bacteria of the phylum Bacteroides are strong SCFA producers and could be beneficial during viral infections, since Bacteroidetes produce high levels of acetate and propionate, whereas Firmicutes produce more butyrate [75]. Propionate contributes to epithelial barrier integrity and regulates inflammatory responses [76]. Acetate acts via its cognate receptor FFAR2 to potentiate innate immunity. In neutrophils, acetate-FFAR2 signaling enhances their recruitment to inflamed sites, activates the inflammasome, and promotes IL-1B secretion [77]. Butyrate reinforces the barrier function, exerts anti-inflammatory effects, and inhibits tumor growth through cell cycle arrest and apoptosis [78,79]. However, recent evidence indicates that sodium butyrate may suppress humoral immunity by inhibiting follicular T helper cell differentiation [80]. These findings highlight the complexity of butyrate’s role in immunity, which may vary with concentration, cellular context, and the specific arm of the immune response being engaged [81]. Regarding safety profiles, our toxicological evaluation indicated that among all metabolites screened, succinate emerged as particularly significant due to its favorable drug-likeness and acceptable toxicological profile. Among the short-chain fatty acids, butyrate and propionate exhibit favorable safety, with skin sensitization being the only predicted adverse effect. Acetate, however, showed potential for both skin sensitization and drug-induced liver injury, warranting caution. The differential effects and safety profiles of these SCFAs suggest they may hold translational potential in HPV management. Future studies should explore whether modulating SCFA-producing gut microbiota could influence the cervical immune microenvironment through the gut-cervix axis.

From the perspective of the MMTS network, succinate production is linked to the metabolic activities of diverse gut microbiota. It is primarily derived from Succinivibrio dextrinosolvens and Succinatimonas hippei (Proteobacteria), as well as Bacteroides thetaiotaomicron and Parabacteroides distasonis (Bacteroidota). Members of the family Veillonellaceae and Acidipropionibacterium acidipropionici (Firmicutes) may also contribute to succinate homeostasis through indirect metabolic pathways [82]. Changes in the abundance of these microbiota may influence the host’s local immune microenvironment through the “gut-cervical axis”. On the one hand, succinate could reach cervical tissues via the bloodstream, and through targeting IL1B, activate signaling pathways, including the TNF signaling pathway, Toll-like receptor signaling pathway, C-type lectin receptor signaling pathway, and IL-17 signaling pathway. This provides a theoretical basis for hypothesizing that succinate, once reaching cervical tissues via the bloodstream, could influence local immune responses [83]. However, how succinate is transported from the gut to the cervix, its local concentration, and its direct immunomodulatory effects in the cervical microenvironment remain unclear. Thus, while our findings support a potential “gut-cervix axis” in HPV infection, this concept requires direct validation, particularly regarding systemic metabolite trafficking and cervical immune regulation. On the other hand, dysbiotic mediators released by imbalanced gut microbiota can induce hormonal imbalance, leading to abnormal estrogen metabolism and disrupting the vaginal microecology, which could indirectly affect the outcome of HPV infection [19,84]. Taken together, our findings position succinate as a potential metabolic marker in HPV progression. Further research is warranted to elucidate the functional role of succinate in HPV pathogenesis and to determine whether modulating its levels could represent a viable strategy for intervention.

Higher dietary fiber intake has been associated with a reduced risk of HPV infection, and certain probiotics have shown promise in clearing HPV-related lesions [85]. These benefits may be mediated through prebiotic dietary factors such as inulin and fructooligosaccharides, which serve as key substrates for gut bacteria and are fermented into SCFAs, including acetate, propionate, and butyrate [86]. Building on this, targeting succinate-producing bacteria identified in our MMTS network, namely Proteobacteria, Bacteroidota, and Firmicutes, through probiotics or dietary modification may help reduce succinate levels at the source [87]. The bacterial strains identified as producers of succinate and other bioactive compounds in our MMTS network may themselves hold potential as next-generation probiotics. In addition, disrupting the interaction between succinate and IL1B could be worth considering. Developing small-molecule inhibitors that specifically block this interaction may inhibit IL1B-mediated inflammatory pathways. Currently, IL1B targeted therapies such as anakinra, an IL1 receptor antagonist, have been approved for inflammatory diseases [88], and their potential application in HPV-related persistent inflammation warrants further investigation. Another angle is SUCNR1, the receptor through which succinate signals [70,72]. Blocking SUCNR1 activation might prevent some of the immune dysfunction seen in the cervical microenvironment, like overactive macrophages and excessive cytokine release. Since SUCNR1 antagonists have shown potential in metabolic disease, it’s not a stretch to think they might be repurposed for HPV [89]. However, any of these strategies would need to be weighed carefully against safety concerns.

5. Conclusion

This study systematically mapped gut microbial metabolites relevant to HPV infection, identifying 10 core inflammatory and immune regulatory targets (IL6, AKT1, IL1B, CASP3, NFKB1, EGFR, PPARG, JUN, PTGS2, and TLR4) and their associated pathways, including TNF, Toll-like receptor, C-type lectin receptor, and IL-17 signaling pathways. Succinate stood out among the 13 screened metabolites for its combination of a favorable safety profile and predicted binding to IL1B (−5.66 kcal/mol). The identification of succinate-producing bacteria (Proteobacteria, Bacteroidota, Firmicutes) further links gut microbial activity to the proposed gut-cervix axis. These findings provide a testable framework for investigating how specific gut-derived metabolites might influence cervical immunity, and offer prioritized targets and pathways for future experimental validation.

6. Limitations

Despite the progress made in this study, several limitations must be acknowledged. First, the host targets were retrieved from public databases using the general keyword “HPV infection”, which may introduce inherent biases such as incomplete annotation and publication bias. Moreover, the obtained data were not stratified by viral genotype, patient gender, or age groups. Although our study identified universal core targets, their specific relevance to high-risk types such as HPV16 and HPV18 requires further investigation. Second, while we assessed drug-likeness, toxicological profiles, and performed molecular docking simulations to predict metabolite-target interactions, these findings are based on computational analyses. It is important to note that the associations identified between gut microbial metabolites and HPV infection targets are correlational rather than causal, and their directionality requires experimental validation to confirm the biological relevance and therapeutic potential of these interactions. Future studies utilizing stratified clinical cohorts and experimental approaches will be essential to substantiate these findings.

Supporting information