Abstract
Introduction
The pathogenesis of skin and soft tissue infections (SSTIs) remains incompletely understood, particularly regarding the causal relationships between circulating inflammatory proteins (CIPs), metabolites, and infection susceptibility. Mendelian randomization (MR) provides an approach to investigate potential causal associations while reducing confounding and reverse causation.
Methods
We conducted a bidirectional two-sample MR study integrating proteomic and metabolomic data. Genetic instruments for 91 CIPs were obtained from a proteomic GWAS (14,824 participants), SSTI summary statistics were obtained from the FinnGen consortium (R10 release), and 1400 plasma metabolites were analyzed from the GWAS Catalog. The inverse-variance weighted (IVW) method was used as the primary analysis, with sensitivity analyses performed to evaluate robustness. MR-based mediation analysis was conducted to assess metabolite-mediated pathways.
Results
Genetic evidence supported potential causal associations between six CIPs and SSTI risk. Higher genetically predicted IL-12B levels were associated with increased SSTI risk (OR = 1.036, 95% CI: 1.001–1.073, p = 0.044), whereas higher levels of CD40L (OR = 0.954, 95% CI: 0.914–0.996, p = 0.032), CX3CL1 (OR = 0.930, 95% CI: 0.869–0.996, p = 0.038), IL-10 (OR = 0.892, 95% CI: 0.842–0.946, p = 0.0001), IL-13 (OR = 0.935, 95% CI: 0.889–0.985, p = 0.012), and TNF-related activation-induced cytokine (OR = 0.952, 95% CI: 0.916–0.989, p = 0.011) were associated with reduced risk. Reverse MR showed no evidence of SSTI influencing these proteins. Mediation analysis identified an inconsistent mediation pathway involving the threonine-to-pyruvate ratio, which accounted for −7.79% (p = 0.028) of the total IL-10 protective association.
Discussion
This study provides genetic evidence supporting potential roles of inflammatory proteins and metabolites in SSTI susceptibility and highlights a counteracting metabolic pathway related to IL-10. The findings expand understanding of host immunometabolic mechanisms underlying SSTI risk.
Conclusion
Our findings provide genetic evidence supporting potential causal relationships between CIPs, metabolites, and SSTI susceptibility. Further functional studies are required to validate the identified immunometabolic pathways.
Keywords: skin and subcutaneous tissue infections, circulating inflammatory proteins, serum metabolites, mendelian randomization, inverse-variance weighted method, mediation analysis
Introduction
Skin and subcutaneous tissue infections are among the most prevalent bacterial infections encountered in clinical practice, posing a substantial global health and economic burden.1–4 Ranging from uncomplicated cellulitis to life-threatening necrotizing fasciitis, these infections frequently lead to hospitalization, surgical intervention, and long-term morbidity.2,5,6 While antibiotic therapy remains the cornerstone of treatment, the escalating threats of antimicrobial resistance, treatment failure, and infection recurrence underscore the limitations of a purely antimicrobial approach.7 Consequently, a deeper understanding of the host-specific factors that govern susceptibility to SSTI is imperative.
An individual’s systemic inflammatory state is increasingly recognized as a key determinant of infection risk. Chronic or dysregulated inflammation is thought to create a permissive environment for microbial invasion by compromising cutaneous barrier integrity and altering local immune responses. Pro-inflammatory cytokines such as TNF-α and IL-6 can disrupt epidermal function, dysregulate antimicrobial peptide production, and recruit or modulate innate immune cells in ways that may ultimately favor pathogen colonization.8,9 For instance, in Staphylococcus aureus (S. aureus) infected atopic dermatitis, elevated skin-derived chemokines like CXCL10 and CCL17 drive the recruitment of T-cell subsets that can amplify inflammation and tissue damage.10–13 However, in observational studies, it remains difficult to distinguish whether such inflammatory markers causally contribute to infection susceptibility or are merely epiphenomena confounded by factors like comorbidities, lifestyle, or socioeconomic status.
Beyond inflammation, the host metabolic landscape has emerged as a critical regulator of immune competence and pathogen clearance. Advances in metabolomics have revealed that systemic metabolism and immune function are intricately linked, a concept known as immunometabolism.14,15 Circulating metabolites, including short-chain fatty acids, amino acids, and their derivatives, can directly influence the phenotype, proliferation, and antimicrobial functions of key effector cells like neutrophils and macrophages.16 For instance, certain metabolites can fuel oxidative burst pathways or serve as signaling molecules to polarize immune responses.17,18 This nexus between metabolism and immunity suggests a compelling mechanistic hypothesis: systemic inflammation may predispose individuals to SSTIs not only through direct tissue effects but also indirectly by reprogramming host metabolism, thereby shaping an immune-metabolic state that is either conducive or resistant to infection.
Traditional observational studies are limited by residual confounding, reverse causation, and the difficulty of disentangling correlated inflammatory and metabolic pathways.19 Mendelian randomization, which leverages germline genetic variants as proxies for lifelong exposure levels, offers a powerful approach to infer causality and overcome these limitations.20,21 Conventional epidemiological approaches, including cross-sectional studies, prospective cohort studies, and multivariable regression analyses, have substantially contributed to identifying associations between circulating biomarkers and disease susceptibility. However, these approaches cannot completely exclude residual confounding caused by unmeasured environmental, lifestyle, or clinical factors. Moreover, because inflammatory responses and metabolic alterations may occur as consequences of infection, conventional observational studies are often unable to distinguish causal determinants from disease-related changes.22 Mendelian randomization (MR) provides an alternative framework by using genetic variants associated with exposures as instrumental variables. Since genetic variants are randomly allocated during conception and generally precede disease onset, MR can partially mitigate confounding and reverse causation, thereby improving causal inference.23 Furthermore, integrating proteomic and metabolomic data within an MR framework enables systematic investigation of potential upstream inflammatory drivers, downstream metabolic pathways, and their causal relationships with SSTI susceptibility.24 Integrating proteomic and metabolomic MR within a mediation framework provides a unique opportunity to dissect upstream inflammatory drivers, downstream metabolic consequences, and their combined influence on SSTI risk.
To address these gaps, we conducted a bidirectional two-sample MR study integrating CIPs and metabolomic datasets. MR leverages genetic variants as instrumental variables to strengthen causal inference by reducing the influence of confounding and reverse causation.25,26 Our objectives were threefold: (i) to assess the causal relationships between CIPs and SSTI risk, (ii) to identify serum metabolites with causal effects on SSTI susceptibility, and (iii) to investigate whether metabolites mediate the effects of inflammatory proteins on SSTI, thereby uncovering immune-metabolic axes relevant to infection pathogenesis.
Materials and Methods
Data Sources
Summary statistics for genome-wide association studies (GWAS) pertaining to 91 CIPs were sourced from a proteomic investigation of 14,824 individuals of European ancestry. These data are publicly available for download from the study website (https://www.phpc.cam.ac.uk/ceu/proteins) and the EBI GWAS Catalog (accession range GCST90274758–GCST90274848). Genetic association estimates for SSTI were obtained from the ffinngen_R10_L12_INFECT_SKIN dataset, comprising 23,514 cases and 388,667 controls. Summary statistics for 1091 blood metabolites and 309 metabolite ratios (accessions Ebi-a-GCST90199621 to Ebi-a-GCST90201020) were retrieved from the GWAS Catalog. All datasets are derived from European-ancestry cohorts, and no sample overlap exists between the exposure and outcome GWAS sources used in this analysis.27,28
Study Design
A two-stage, bidirectional MR framework was implemented to investigate: (i) the causal relationships between circulating inflammatory proteins (CIPs) and SSTI risk, and (ii) the potential mediating effects of serum metabolites. The analysis proceeded in three phases. First, bidirectional two-sample MR was applied to assess causal effects between CIPs and SSTI, identifying candidate CIPs significantly associated with the outcome. Second, two-sample MR was used to evaluate the causal effects of serum metabolites on SSTI risk. Finally, for significant CIP-metabolite-outcome pairs, we estimated the causal effect of the CIP on the candidate metabolite and quantified the proportion of the total CIP effect on SSTI mediated through the metabolite pathway. An overview of the design is provided in Figure 1.
Figure 1.
Study design and analytical workflow of the two-stage bidirectional Mendelian randomization analysis. Genetic variants were used as instrumental variables to investigate the relationships between circulating inflammatory proteins (CIPs), serum metabolites, and susceptibility to SSTIs. The analysis included bidirectional Mendelian randomization between CIPs and SSTIs, Mendelian randomization of serum metabolites on SSTI risk, and mediation analysis to quantify the indirect effects of candidate metabolites. Red crosses indicate the Mendelian randomization assumptions represented in the diagram: genetic variants should be independent of potential confounders and should influence the outcome only through the exposure, with no alternative direct pathway.
Abbreviations: SNPs, single-nucleotide polymorphisms; CIPs, circulating inflammatory proteins; MR, Mendelian randomization; SSTI, skin and soft-tissue infection.
Selection of Genetic Instrumental Variables (IVs)
In Mendelian randomization analysis, genetic variants are used as instrumental variables to estimate the potential causal effects of exposures on outcomes. In this study, SNPs associated with circulating inflammatory proteins or serum metabolites were selected as proxies for genetically predicted exposure levels. Because genetic variants are randomly allocated during meiosis and determined before disease development, they are less likely to be influenced by environmental factors or disease progression. Therefore, SNP-based instrumental variable analysis provides an approach to reduce potential confounding and reverse causation, allowing investigation of whether genetically predicted inflammatory proteins and metabolites are associated with SSTI susceptibility. Single nucleotide polymorphisms (SNPs) were selected as instrumental variables (IVs) based on trait-specific significance thresholds: p < 1×10−5 for circulating CIPs and serum metabolites, while a more stringent threshold of p < 5×10−8 was applied for SSTI to ensure robust IV-outcome associations. To eliminate linkage disequilibrium and ensure IV independence, SNPs were clustered with a clumping distance of 10,000 kb and a threshold of R2 < 0.001.29 The robustness of IVs for exposure traits was evaluated using F-statistics, with values greater than 10 indicating a strong and reliable correlation between IVs and the corresponding exposures (CIPs, metabolites, or SSTI). Additionally, PhenScanner V2 was utilized to screen all selected SNPs, aiming to identify and exclude potential confounders that might interfere with the causal inference between exposures and outcomes.
Ethical Considerations
Ethical approval was obtained from the Institutional Review Board of Tangdu Hospital, Fourth Military Medical University (K202410-04).
Statistics Analysis
Analyses were conducted in R (v4.4.0) utilizing the TwoSampleMR, ieugwasr, MR-PRESSO, and MRInstruments packages. The primary causal estimate was derived using the IVW method under a random-effects model. Sensitivity analyses included the weighted median and MR-Egger regression methods to evaluate robustness to horizontal pleiotropy. Associations with p < 0.05 were considered nominally significant and were interpreted together with the direction and consistency of effect estimates across complementary MR methods and sensitivity analyses. Given the exploratory nature of the large-scale screening of inflammatory proteins and metabolites, no formal Bonferroni or false discovery rate correction was applied. For the metabolomic analysis, a secondary threshold of p < 0.01 in the IVW analysis was used to prioritize candidate metabolites for subsequent pathway and mediation analyses; this threshold was used for prioritization rather than as a formal multiple-testing correction. Leave-one-out analysis was performed to determine if any single SNP disproportionately drove the causal estimate.29 The mediation effect of a serum metabolite on the CIP–SSTI pathway was calculated using the product-of-coefficients method, with standard errors approximated via the delta method.
Results
Total Effect of CIP on SSTI
The two-sample MR analysis was conducted to assess the causal effects of six circulating CIPs on the risk of SSTI.
The IVW method, serving as the primary analysis, identified six inflammatory proteins showing nominally significant genetic associations with SSTI risk (p < 0.05) (Figure 2A). Notably, elevated genetically predicted levels of five proteins were associated with a reduced risk of SSTI: CD40L receptor (OR = 0.954, 95% CI: 0.914–0.996, p = 0.032), CX3CL1 (Fractalkine) (OR = 0.930, 95% CI: 0.869–0.996, p = 0.038), IL-10 (OR = 0.892, 95% CI: 0.842–0.946, p = 0.0001), IL-13 (OR = 0.935, 95% CI: 0.889–0.985, p = 0.012), and TNF-related activation-induced cytokine (OR = 0.952, 95% CI: 0.916–0.989, p = 0.011). In contrast, higher levels of IL-12B were associated with an increased risk of SSTI (OR = 1.036, 95% CI: 1.001–1.073, p = 0.044).
Figure 2.
Total effect of CIP on SSTI. (A) Forest plot from forward MR analysis illustrating the causal relationships of 6 CIPs on SSTI. CI: Confidence Interval; OR: odds ratios; nsnp: number of single nucleotide polymorphisms. (B) Scatter plot depicting genetic associations between 6 CIPs and SSTI. (C) Leave-one-out SNP results between 6 inflammatory factors and SSTI in the forward MR analysis.
Sensitivity analyses largely supported the robustness of the primary findings for several factors. For CD40L receptor and TNF-related activation-induced cytokine, the weighted median and weighted mode estimates were directionally consistent and statistically significant. The association for IL-10 was strongly supported by the IVW method and showed a concordant, significant effect in MR-Egger regression (OR = 0.856, p = 0.017), though other sensitivity methods were not significant. The significant result for CX3CL1 and IL-13 were isolated to the IVW method, as all other sensitivity analyses yielded non-significant estimates, suggesting a need for cautious interpretation (Figure 2A).
A scatter plot illustrating the causal relationships between inflammatory proteins and SSTI is presented in (Figure 2B). Leave-one-out sensitivity analysis demonstrated that no single nucleotide polymorphism (SNP) significantly disrupted the overall effect of inflammatory proteins on SSTI (Figure 2C). Heterogeneity was detected solely in IL-10, indicated by a Cochran’s Q derived p-value of less than 0.05; however, the causal estimate was deemed acceptable when applying the random-effects IVW method (Table S1). The MR-Egger intercept p-value exceeded 0.05, indicating that no horizontal pleiotropy was present (Table S2).
In conclusion, this MR analysis provides genetic evidence supporting a putative causal role for specific inflammatory pathways in SSTI susceptibility. IL-10 emerged as the most robust protective factor, while IL-12B may represent a risk factor. The findings for CD40L receptor and TNF-related activation-induced cytokine also appear robust, whereas the evidence for CX3CL1 and IL-13 are less consistent.
The Causative Impact of SSTI on CIPs
To evaluate potential reverse causation, a reverse MR analysis was conducted to evaluate the effect of genetic predisposition to SSTI on the circulating levels of the six significant CIPs.
Results from the IVW analysis revealed no statistically significant causal effects of SSTI on the levels of any of the analyzed inflammatory proteins (all p > 0.05). Specifically, genetic instruments for SSTI did not significantly alter levels of CD40L receptor (OR = 0.968, 95% CI: 0.887–1.057, p = 0.473), CX3CL1 (OR = 0.988, 95% CI: 0.898–1.087, p = 0.803), IL-10 (OR = 0.934, 95% CI: 0.837–1.041, p = 0.216), IL-12B (OR = 1.045, 95% CI: 0.935–1.168, p = 0.439), IL-13 (OR = 0.976, 95% CI: 0.883–1.080, p = 0.640), or TNF-related activation-induced cytokine (OR = 0.910, 95% CI: 0.816–1.014, p = 0.088) (Figure 3). The details regarding heterogeneity and horizontal pleiotropy are provided in Tables S3 and S4.
Figure 3.
The causative impact of SSTI on CIPs. Forest plot from reverse MR analysis displaying the reverse MR analysis of SSTI and 6 CIPs.
Abbreviations: CI, Confidence Interval; OR, odds ratios; nsnp, number of single nucleotide polymorphisms.
The lack of significant associations in this reverse direction analysis strengthens the validity of the primary (forward) MR findings. It suggests that the observed associations in the forward analysis (eg, the protective effect of IL-10) are unlikely to be driven by reverse causation, thereby supporting the inference that genetically predicted inflammatory protein levels may influence SSTI susceptibility, rather than being a consequence of the disease.
Effect of Serum Metabolites on SSTI
To explore the causal role of host metabolism in SSTI susceptibility, we performed a two-sample MR analysis of 1400 serum metabolites and metabolite ratios. Initially, 63 metabolites showed nominally significant associations with SSTI at p < 0.05. To prioritize candidates for subsequent analyses, a secondary IVW threshold of p < 0.01 was applied, resulting in six candidate metabolites (Table S5).
Genetically predicted levels of all six candidate metabolites were significantly associated with SSTI risk in the primary IVW analysis (Figure 4A). Higher levels of Ascorbic acid 3-sulfate, Deoxycholic acid 12-sulfate, Flavin adenine dinucleotide (FAD), and the Cortisone to 4-cholesten-3-one ratio were associated with a reduced risk of SSTI (OR range: 0.95–0.97). Conversely, elevated levels of 5-hydroxymethyl-2-furoylcarnitine and the Threonine to pyruvate ratio were associated with an increased risk (OR = 1.066 and 1.084, respectively).
Figure 4.
Effect of serum metabolites on SSTI. (A) Forest plot of MR analysis which show the causality of serum metabolites on SSTI. (B) The scatter plot of causal relationships between serum metabolites and SSTI. (C) Leave-one-out of SNPs results between SSTI and serum metabolites in MR analysis.
Abbreviations: CI, Confidence Interval; OR, odds ratios; nsnp, number of single nucleotide polymorphism.
The scatter plot of the causal relationship between serum metabolites and SSTI is shown in Figure 4B. Leave-one-out sensitivity analysis indicated that no single SNP significantly disrupted the overall impact of serum metabolites on SSTI (Figure 4C), and no evidence of heterogeneity or horizontal pleiotropy was observed (Tables S6 and S7).
In summary, through a stringent selection process, we identified a shortlist of six serum metabolites with putative causal effects on SSTI risk. The most robust of these candidates provide a basis for investigating potential mediating pathways between inflammatory factors and SSTI susceptibility.
Effect of CIPs on Serum Metabolites
To clarify whether metabolic pathways mediate the associations between inflammatory factors and SSTI, further analyses were conducted to examine the causal relationships between the six key inflammatory factors and the six core candidate metabolites (Figure 5).
Figure 5.
Forest plot of MR analysis which show the mediate effect of Threonine to pyruvate ratio of IL-10 levels on SSTI.
Abbreviations: CI, Confidence Interval; OR, odds ratios; nsnp, number of single nucleotide polymorphism.
The IVW analysis demonstrated a significant causal relationship between genetically elevated IL-10 levels and increased Threonine to pyruvate ratio levels (OR = 1.116, 95% CI: 1.026–1.214, p = 0.011). For CD40L and cortisone to 4-cholesten-3-one ratio, IVW analysis also identified a significant association (OR = 0.932, 95% CI: 0.872–0.996, p = 0.037); however, this association was excluded from subsequent interpretation as the effect directions of OR values across different analytical approaches failed to meet the predefined criterion. Specifically, four out of five MR analytical methods (MR Egger, Weighted median, IVW, Weighted mode) yielded OR values < 1 for the CD40L-cortisone to 4-cholesten-3-one ratio pair (ranging from 0.932 to 0.940), while the Simple mode method returned an OR value of 1.075 (> 1), leading to inconsistent effect directions (Table S10). No other analyses revealed significant effects, nor did we detect heterogeneity or horizontal pleiotropy (Tables S8 and S9). This positive association indicates that the genetic predisposition to higher IL-10 levels leads to an increase in this specific metabolic ratio.
Mediation Effect of CIPs on SSTI
Mendelian randomization-based mediation analysis identified an inconsistent mediation pattern involving IL-10, the threonine-to-pyruvate ratio, and SSTI risk. The estimated proportion of the total protective effect of genetically predicted IL-10 on SSTI that was attributable to the indirect pathway through the threonine-to-pyruvate ratio was −7.719% (95% CI: −14.8% to −0.837%, p = 0.028).
Decomposition of the total, direct, and indirect effects provided insight into this finding. The total effect analysis indicated that genetically predicted higher IL-10 levels were associated with a reduced risk of SSTI (β = −0.1139). In the mediation framework, IL-10 was positively associated with the threonine-to-pyruvate ratio (Path A: β = 0.1095), whereas a higher threonine-to-pyruvate ratio was associated with increased SSTI risk (Path B: β = 0.0811). Consequently, the indirect effect through the threonine-to-pyruvate ratio was positive (β = 0.00888), operating in the opposite direction to the overall protective effect of IL-10 (Table 1).
Table 1.
MR-Based Mediation Analysis of the Association Between Genetically Predicted IL-10, the Threonine-to-Pyruvate Ratio, and SSTI Risk
| Total Effect | Direct Effect A | Direct Effect B | Mediation Effect | P value | Mediated Proportion (%) |
|---|---|---|---|---|---|
| β (95% CI) | β (95% CI) | β (95% CI) | β (95% CI) | 0.02811 | −7.79% (−0.837%, −14.8%) |
| −0.1139 (−0.1717, −0.05599) | 0.1095 (0.02533, 0.1936) | 0.08106 (0.04426, 0.1178) | 0.00888 (0.000953, 0.0168) |
Therefore, the negative mediated proportion should be interpreted as an inconsistent mediation effect rather than conventional mediation. Specifically, the threonine-to-pyruvate ratio represents a counteracting indirect pathway that partially offsets, rather than explains, the protective association between genetically predicted IL-10 and SSTI risk. These findings suggest that while IL-10 has an overall protective association with SSTI susceptibility, the identified metabolic pathway may act in an opposing direction and that additional biological pathways may contribute to IL-10-mediated protection.
Discussion
This MR study provides genetic evidence that specific CIPs and serum metabolites contribute causally to susceptibility to SSTI. Among the six proteins identified, IL-10 emerged as the most consistent protective factor, whereas IL-12B was associated with increased risk. Mediation analysis further revealed that the threonine-to-pyruvate ratio partially counteracted IL-10’s protective effect, suggesting that IL-10’s benefit operates through immunomodulatory or tissue-repair pathways independent of this metabolic axis.
Our findings align with prior experimental studies demonstrating IL-10’s role in limiting tissue damage during cutaneous infection. IL-10 suppresses excessive inflammatory responses by inhibiting dendritic cell activation and T-cell proliferation, thereby preserving barrier integrity and promoting resolution.30,31 Animal models of S aureus infection have shown that IL-10 deficiency leads to uncontrolled inflammation and impaired healing, consistent with our genetic evidence.32–34 However, the observed counteractive metabolic effect highlights the complexity of cytokine-induced immunometabolic regulation. Recent metabolomic studies indicate that threonine metabolism is a key regulator of innate immune cell function, influencing processes such as macrophage polarization and protein synthesis essential for immune responses.35,36 Our discovery that IL-10 causally elevates the threonine-to-pyruvate ratio—a itself a risk factor for SSTI—suggests a plausible immunometabolic antagonism: IL-10 may inadvertently foster a threonine-related metabolic state that partially counteracts its own anti-inflammatory and tissue-protective benefits, potentially through modulating the functional balance of neutrophils and macrophages at the site of infection.
The identification of IL-12B as a risk factor supports existing literature linking IL-12-driven Th1 responses to tissue-destructive inflammation.37 Conversely, protective associations for CD40L receptor, CX3CL1, IL-13, and TRANCE suggest that diverse immunoregulatory pathways contribute to host defense.38,39 Notably, the weaker consistency of CX3CL1 and IL-13 across sensitivity analyses indicates that their roles may be context-dependent, requiring further validation.
Beyond host-derived inflammatory and metabolic factors, microbial characteristics represent another critical determinant of SSTI susceptibility and clinical progression. SSTIs are caused by a diverse range of pathogens, among which Staphylococcus aureus and Streptococcus pyogenes are predominant Gram-positive organisms, while Gram-negative bacteria and polymicrobial infections contribute to more severe and complicated cases. Pathogen-specific virulence factors, including toxins, adhesins, immune evasion mechanisms, and biofilm formation capacity, can influence bacterial colonization, tissue invasion, and persistence within infected sites. In addition, antimicrobial resistance profiles of causative microorganisms have become increasingly important determinants of treatment response and disease burden.
The present MR study primarily focused on host genetic susceptibility factors and identified inflammatory and metabolic pathways associated with SSTI risk. These findings should therefore be interpreted within the broader context of host–pathogen interactions. Genetic predisposition affecting immune regulation and metabolic homeostasis may influence the host’s ability to recognize, contain, and resolve microbial invasion, whereas pathogen-specific properties determine the characteristics of infection establishment and progression. Future studies integrating host genomics, microbial genomics, and longitudinal infection data may provide a more comprehensive framework for understanding SSTI pathogenesis and identifying individualized therapeutic strategies.
Several limitations warrant consideration. First, all GWAS datasets were derived from European ancestry cohorts, limiting generalizability to other populations. Second, the statistical power of metabolite GWAS remains modest compared to proteomic datasets, which may have constrained the detection of additional mediating pathways. Third, MR cannot fully capture context-dependent effects such as cytokine thresholds, tissue-specific expression, or temporal dynamics during infection. Functional validation in experimental models will be essential to confirm the mechanistic pathways suggested by our genetic analyses. In addition, because a relatively large number of inflammatory proteins and metabolites were screened, the possibility of false-positive findings due to multiple testing cannot be excluded. No formal Bonferroni or false discovery rate correction was applied in the present exploratory analysis. Therefore, associations identified at nominal significance thresholds should be considered suggestive rather than definitive and require replication in independent datasets and functional validation.
Despite these limitations, our study highlights the importance of integrating immunological and metabolic perspectives in SSTI pathogenesis. The discovery that IL-10’s protective effect is partially offset by a metabolic shift emphasizes the need for future research to explore cytokine-metabolite interactions in greater depth. Expanding analyses to multi-ethnic cohorts, incorporating longitudinal metabolomic profiling, and conducting mechanistic studies in infection models will be critical next steps. Ultimately, targeting metabolic pathways that synergize with protective cytokines such as IL-10 may offer new strategies for adjunctive therapies against SSTI, particularly in the era of rising antimicrobial resistance.
Conclusion
This MR study provides genetic evidence supporting potential causal relationships between circulating inflammatory proteins, serum metabolites, and SSTI susceptibility. IL-10 emerged as a potential protective factor, while the threonine-to-pyruvate ratio showed an opposing-direction indirect effect consistent with inconsistent mediation. These findings highlight potential immunometabolic pathways involved in SSTI susceptibility and provide a foundation for further experimental validation.
Funding Statement
This study was supported by the Special Support Program for High Level Talents in Shaanxi Province, and the Shaanxi Provincial Health Scientific Research Innovation Platform Project (2025PT-07) to Weiwei Qin, Tangdu Hospital Subject Promotion Project (2025XJRH025) and Xinfei Project of the Air Force Medical University to Zhuo Wan.
Abbreviations
SSTIs, Skin and Subcutaneous Tissue Infections; CIP, Circulating Inflammatory Protein; GWAS, Genome-Wide Association Study; IVW, Inverse Variance Weighted; MR, Mendelian Randomization; SNP, Single Nucleotide Polymorphism; S. aureus, Staphylococcus aureus; IVs, Instrumental Variables; CX3CL1, Fractalkine; IL-10, interleukin-10; IL-12B, interleukin-12 subunit beta; IL-13, interleukin-13.
Data Sharing Statement
All data generated or analyzed during this study are included in this published article.
Ethics Approval and Consent to Participate
Ethical approval was obtained from the Institutional Review Board of Tangdu Hospital, Fourth Military Medical University (K202410-04).
Disclosure
The authors declare no conflict of interest, financial or otherwise.
References
- 1.Brothwell JA, Griesenauer B, Chen L, et al. Interactions of the skin pathogen haemophilus ducreyi with the human host. Front Immunol. 2021;11:615402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Phoenix G, Das S, Joshi M. Diagnosis and management of cellulitis. BMJ. 2012;345(aug07 2):e4955–12. doi: 10.1136/bmj.e4955 [DOI] [PubMed] [Google Scholar]
- 3.Collazos J, De la fuente B, Garcia A, et al. Cellulitis in adult patients: a large, multicenter, observational, prospective study of 606 episodes and analysis of the factors related to the response to treatment. PLoS One. 2018;13(9):e0204036. doi: 10.1371/journal.pone.0204036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Vella V, Derreumaux D, Aris E, et al. The incidence of skin and soft tissue infections in the United States and associated healthcare utilization between 2010 and 2020. Open Forum Infect Dis. 2024;11(6):ofae267. doi: 10.1093/ofid/ofae267 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Pollack CV, A AMIN, Ford WT, et al. Acute bacterial skin and skin structure infections (ABSSSI): practice guidelines for management and care transitions in the emergency department and hospital. J Emerg Med. 2015;48(4):508–519. doi: 10.1016/j.jemermed.2014.12.001 [DOI] [PubMed] [Google Scholar]
- 6.Kaye KS, Petty LA, Shorr AF, et al. Current epidemiology, etiology, and burden of acute skin infections in the United States. Clin Infect Dis. 2019;68(Suppl 3):S193–S9. doi: 10.1093/cid/ciz002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Goel A, Garima, Aggarwal N, et al. Skin and soft tissue infections: current advancement in epidemiology, pathogenesis and management. J Pure Appl Microbiol. 2023;17(1):89–111. doi: 10.22207/JPAM.17.1.50 [DOI] [Google Scholar]
- 8.Dong J, Wang W, Zhou W, et al. Immunomodulatory biomaterials for implant-associated infections: from conventional to advanced therapeutic strategies. Biomater Res. 2022;26(1):72. doi: 10.1186/s40824-022-00326-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kumar R, Kolloli A, Singh P, et al. The innate memory response of macrophages to Mycobacterium tuberculosis is shaped by the nature of the antigenic stimuli. Microbiol Spectr. 2024;12(8):e0047324. doi: 10.1128/spectrum.00473-24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Miller LS, Cho JS. Immunity against Staphylococcus aureus cutaneous infections. Nat Rev Immunol. 2011;11(8):505–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Clowry J, Dempsey DJ, Claxton TJ, et al. Distinct T cell signatures are associated with Staphylococcus aureus skin infection in pediatric atopic dermatitis. JCI Insight. 2024;9(9). doi: 10.1172/jci.insight.178789. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mbaye EA, Scariot D, Bakoshi A, et al. Subcutaneous nanotherapy for islet transplantation preserves functional protective immunity in a staphylococcus aureus skin infection model. J Immunol. 2025;214:vkaf283–605. [Google Scholar]
- 13.I ROY, Evans DB, Dwinell MB. Chemokines and chemokine receptors: update on utility and challenges for the clinician. Surgery. 2014;155(6):961–973. doi: 10.1016/j.surg.2014.02.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.O’neill LAJ, Kishton RJ, Rathmell J. A guide to immunometabolism for immunologists. Nat Rev Immunol. 2016;16(9):553–565. doi: 10.1038/nri.2016.70 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Xu W, Bradstreet TR, Zou Z, et al. Reprogramming aerobic metabolism mitigates Streptococcus pyogenes tissue damage in a mouse necrotizing skin infection model. Nat Commun. 2025;16(1):2559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Jefferson L, Lima PDA, Archer SL. Macrophage plasticity and glucose metabolism: the role of immunometabolism in pulmonary arterial hypertension. Clin Sci. 2025;138(19):1073–1094. doi: 10.1042/CS20257363 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Garcia-gimenez JL, Canovas-cervera I, Nacher-sendra E, et al. Oxidative stress and central metabolism pathways impact epigenetic modulation in inflammation and immune response. Free Radic Biol Med. 2025;233:378–399. doi: 10.1016/j.freeradbiomed.2025.04.004 [DOI] [PubMed] [Google Scholar]
- 18.Hu T, Liu CH, Lei M, et al. Metabolic regulation of the immune system in health and diseases: mechanisms and interventions. Signal Transduct Target Ther. 2024;9(1):268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Meuli L, Dick F. Understanding confounding in observational studies. Eur J Vasc Endovasc Surg. 2018;55(5):737. [DOI] [PubMed] [Google Scholar]
- 20.Kun H, Zhou Q, Wu H, et al. Genetic susceptibility to cerebral palsy involves complement system-mediated neuronal development and plasticity pathway. Sci Rep. 2025;15(1):23088. doi: 10.1038/s41598-025-07178-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Xue S, Wu K, He Q, et al. Plasma metabolites and endometrial cancer: elucidating causal associations through Mendelian randomization. Discov Oncol. 2025;16(1):1740. doi: 10.1007/s12672-025-03585-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Smith GD, Ebrahim S. Mendelian randomization: prospects, potentials, and limitations. Int J Epidemiol. 2004;33(1):30–42. doi: 10.1093/ije/dyh132 [DOI] [PubMed] [Google Scholar]
- 23.Little M. Mendelian randomization: methods for using genetic variants in causal estimation. J Royal Stat Soc Series A. 2018;181(2):549–550. doi: 10.1111/rssa.12343 [DOI] [Google Scholar]
- 24.De ruiter SC, Tschiderer L, Grobbee DE, et al. Mendelian randomisation: concepts, opportunities, challenges, and future directions. Glob Heart. 2025;20(1):54. doi: 10.5334/gh.1438 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Bowden J, Holmes MV. Meta‐analysis andMendelianrandomization: a review. Res Synth Meth. 2019;10(4):486–496. doi: 10.1002/jrsm.1346 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Carter AR, Sanderson E, Hammerton G, et al. Mendelian randomisation for mediation analysis: current methods and challenges for implementation. Eur J Epidemiol. 2021;36(5):465–478. doi: 10.1007/s10654-021-00757-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wu J. Genetically predicted Circulating inflammatory proteins mediate the association between plasma lipidome and kidney stones: a Mendelian randomization study. Urolithiasis. 2025;54(1):6. doi: 10.1007/s00240-025-01905-y [DOI] [PubMed] [Google Scholar]
- 28.Zhao JH, Stacey D, Eriksson N, et al. Genetics of circulating inflammatory proteins identifies drivers of immune-mediated disease risk and therapeutic targets. Nat Immunol. 2023;24(9):1540–1551. doi: 10.1038/s41590-023-01588-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ji Y, Gao X, Liu L, et al. Mediating effects of plasma metabolites in inflammatory protein-lymphoma causality: a mendelian randomization study. Curr Med Chem. 2025;33. doi: 10.2174/0109298673382735250904071151 [DOI] [PubMed] [Google Scholar]
- 30.Ouyang W, Rutz S, Crellin NK, et al. Regulation and functions of the IL-10 family of cytokines in inflammation and disease. Ann Rev Immunol. 2011;29(1):71–109. doi: 10.1146/annurev-immunol-031210-101312 [DOI] [PubMed] [Google Scholar]
- 31.Saraiva M, O’garra A. The regulation of IL-10 production by immune cells. Nat Rev Immunol. 2010;10(3):170–181. doi: 10.1038/nri2711 [DOI] [PubMed] [Google Scholar]
- 32.Kelly AM, Leech JM, Doyle SL, et al. Staphylococcus aureus-induced immunosuppression mediated by IL-10 and IL-27 facilitates nasal colonisation. PLoS Pathog. 2022;18(7):e1010647. doi: 10.1371/journal.ppat.1010647 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kelly AM, Mccarthy KN, Claxton TJ, et al. IL-10 inhibition during immunization improves vaccine-induced protection against Staphylococcus aureus infection. JCI Insight. 2024;9(13). doi: 10.1172/jci.insight.178216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Leech JM, Lacey KA, Mulcahy ME, et al. IL-10 plays opposing roles during staphylococcus aureus systemic and localized infections. J Immunol. 2017;198(6):2352–2365. doi: 10.4049/jimmunol.1601018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Kelly B, O’neill LA. Metabolic reprogramming in macrophages and dendritic cells in innate immunity. Cell Res. 2015;25(7):771–784. doi: 10.1038/cr.2015.68 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Liu PS, Wang H, Li X, et al. alpha-ketoglutarate orchestrates macrophage activation through metabolic and epigenetic reprogramming. Nat Immunol. 2017;18(9):985–994. doi: 10.1038/ni.3796 [DOI] [PubMed] [Google Scholar]
- 37.Teng MWL, Bowman EP, Mcelwee JJ, et al. IL-12 and IL-23 cytokines: from discovery to targeted therapies for immune-mediated inflammatory diseases. Nat Med. 2015;21(7):719–729. doi: 10.1038/nm.3895 [DOI] [PubMed] [Google Scholar]
- 38.Karnell JL, Rieder SA, Ettinger R, et al. Targeting the CD40-CD40L pathway in autoimmune diseases: humoral immunity and beyond. Adv Drug Delivery Rev. 2019;141:92–103. doi: 10.1016/j.addr.2018.12.005 [DOI] [PubMed] [Google Scholar]
- 39.Olmos calvo I, Fodor E, Kardos D, et al. A pilot clinical study of hyperacute serum treatment in osteoarthritic knee joint: cytokine changes and clinical effects. Curr Issues Mol Biol. 2021;43(2):637–649. doi: 10.3390/cimb43020046 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data generated or analyzed during this study are included in this published article.