ABSTRACT
Background
Persistent inflammation is a crucial characteristic of idiopathic pulmonary fibrosis (IPF). Gut microbiota (GM) contribute to the occurrence and development of several pulmonary diseases through the “gut–lung axis.” The genetic role of GM in IPF and the mediating effect of circulating inflammatory proteins.
Methods
A single nucleotide polymorphism (SNP) was used as an instrumental variable (IV) for exposure to evaluate the causal relationship between exposure and outcome. A two‐step, two‐sample Mendelian randomization study mainly based on an “inverse variance weighted (IVW)” approach was performed to explore the causal relationship between GM and IPF mediated by circulating inflammatory proteins.
Results
The IVW way illustrated 12 taxa (Bacillales, Gastranaerophilales, Selenomonadales, Family XIII, Bacteroidaceae, Bacteroides, and Actinomyces, Bifidobacterium, Oscillibacter, Ruminococcus gnavus , Subdoligranulum, Veillonella) of GM and 8 circulating inflammatory proteins (CCL11, CXCL6, CXCL9, CCL8, CCL7, NRTN, STAMPB, and TGFa) had suggestive evidence of causality on IPF. The mediation MR demonstrated the causal pathway from Actinomyces to IPF was partly mediated by CCL11 (the mediation effect: 0.063, 95% CI [1.016–1.126]; p = 0.004) with a mediation proportion of 13.035%.
Conclusions
These findings may suggest a genetically predicted association between GM and IPF mediated by circulating inflammatory proteins.
Keywords: circulating inflammatory proteins, gut microbiota, idiopathic pulmonary fibrosis, Mendelian randomization
We evaluated the causal relationship between gut microbiota and idiopathic pulmonary fibrosis. We evaluated the causal relationship between 91 circulating inflammatory proteins and idiopathic pulmonary fibrosis. The mediation MR demonstrated the causal pathway from Actinomyces to IPF was partly mediated by CCL11.
Abbreviations
- CCL11
C–C motif chemokine ligand 11
- CI
confidence interval
- COPD
chronic obstructive pulmonary disease
- GM
gut microbiota
- GWAS
genome‐wide association study
- IPF
idiopathic pulmonary fibrosis
- IV
instrumental variable
- IVW
inverse variance weighted
- LD
linkage disequilibrium
- MR
Mendelian randomization
- MR‐PRESSO
MR‐Egger intercept test and outlier method
- OR
odds ratio
- SNP
single nucleotide polymorphism
1. Introduction
Idiopathic pulmonary fibrosis (IPF), a chronic progressive and incurable interstitial lung disease, is characterized by irreversible fibrotic destruction in the lung and irreversible decline in lung function and organ failure [1, 2]. The most common symptoms of IPF include cough and dyspnea, which would significantly decrease the quality of life [3]. Although epidemic data for IPF is rare, especially in low‐ and middle‐income countries, there are still some data indicating more than 8 and 28 cases per 100 000 people were diagnosed with IPF each year [4, 5]. In addition, the mortality of IPF is high, with the 5‐year survival after diagnosis less than 30% and the median survival limited to 2–3 years without treatment, which places a tremendous burden on society and personal life across the globe [4, 5]. Some risk factors like age, male sex, a positive history of cigarette smoking, and air pollution have been linked to the development of IPF [6]. However, the specific pathogenesis and pathogenic factors still need further exploration, which may contribute to enriching the treatment methods and improving the prognosis of IPF.
In recent years, a novel theory, “gut–lung axis,” has been proposed, indicating the long‐distance cross‐talk between lung and gut [7]. The gut microbiota (GM) is the core part of this theory. Moreover, much evidence has demonstrated that the imbalance of GM may be in connection with lung homeostasis and susceptibility to lung diseases [8]. For example, a study has proved that the dysfunction of GM in chronic obstructive pulmonary disease (COPD) patients was highly related to airway inflammation level, pulmonary function, and the progression of COPD [9]. In addition, a research showed that cannabidiol, a non‐psychoactive derivative of the cannabis plant, could exert an antifibrotic effect via reversing GM imbalance [10]. GM was associated with various lung diseases, but its role in IPF was still unknown.
A vital function of GM is to regulate immune and inflammatory levels [8]. Moreover, emerging shreds of evidence supported that abnormal immune and inflammatory cell levels contribute to the development of IPF due to their secretion of a large number of cytokines, chemokines, and mediators, which could be divided into two types: (1) proinflammatory, such as IL‐6, IL‐1, IL‐13, and TNF‐α, and (2) anti‐inflammatory, such as TGF‐β and IL‐10 [11, 12]. The proinflammatory factors activate fibroblasts and myofibroblasts and induce further fibrotic tissue remodeling to maintain fiber formation and attract immune cells to promote more inflammatory factors and differentiate fibroblasts into myofibroblasts, a positive feedback loop supporting fiber formation [13]. In the study, we aimed to determine the genetic effects of GM on IPF and whether it can be mediated by circulating inflammatory proteins.
2. Materials and Methods
2.1. Study Design
A two‐step Mendelian randomization (MR) study was performed to analyze the causal role of GM on IPF mediated by circulating inflammatory proteins. Firstly, a two‐sample MR was conducted to investigate the association of genetically predicted risk of IPF with GM and circulating inflammatory proteins. Then, the genetic effects of GM on circulating inflammatory proteins (genetically determined mediators) will be explored, and the mediation proportion will be calculated. Three basic assumptions should be followed in MR analysis: (1) Genetic variation has a solid association with exposure; (2) genetic variation should be independent of potential confounders; (3) these genetic tools should only affect outcomes via exposure. The overall design was shown in Figure 1.
FIGURE 1.
Overview of the MR study design. Three basic assumptions should be followed in MR analysis. (1) Genetic variation has solid association with exposure; (2) Genetic variation should be independent of potential confounders. (3) These genetic tools should only affect outcome via exposure. IPF, idiopathic pulmonary fibrosis; IV, instrumental variable; MR, Mendelian randomization; SNP, single nucleotide polymorphism.
2.2. Data Source
Single nucleotide polymorphism (SNP) was used as an instrumental variable (IV) for exposure to evaluate the causal relationship between exposure and outcome. All summary statistics were obtained from the GWAS Catalog (genome‐wide association studies; https://gwas.mrcieu.ac.uk/). The GWAS IDs for GM at five levels (phylum, class, order, family, genus) were from Ebi‐a‐GCST90016908 to Ebi‐a‐GCST90017118, including 14 306 samples [14]. The GWAS ID for IPF was Ebi‐a‐GCST90018120, including 437 235 control cases [15]. The accession numbers for circulating inflammatory proteins were from GCST90274758‐GCST90274848, including 14 735 cases [16].
2.3. Genetic IV Selection
According to the previous research [17], the parameter for identifying IVs of GM was set as p < 1e × 10−5. A more stringent cutoff value (p < 5e × 10−6) was used to select IVs of 91 circulating inflammatory proteins. The threshold for 91 circulating inflammatory proteins was stricter (p < 5e × 10−8). To reduce the bias caused by linkage disequilibrium (LD), we performed an LD clustering algorithm. IVs were independent as clumped using LD R 2 < 0.001 within ±10 000 kb distance in 1000 Genomes Project as reference pane. Then, we calculated the F statistics of SNPs with this formula: β divided by the square of the standard error. Finally, SNPs with F > 10 would be selected and searched in PhenoScanner V2 (http://www.phenoscanner.medschl.cam.ac.uk/) [18]. SNPs violating the second and third assumptions and having potential confounders and bypassing (e.g., age, sex, smoking, and other disease) would be removed. 2082, 1748, and 6 SNPs were selected as the IVs for GM, 91 circulating inflammatory proteins, and IPF, respectively (Table S1).
2.4. Statistical Analysis
All statistical analyses were performed using R 4.3.1 (https://www.r‐project.org). The packages: “TwoSampleMR,” “ieugwasr,” and “VariantAnnotation” were applied to execute the two‐sample MR study with five modes: “Weighted median,” “MR Egger,” “Simple mode,” “Weighted mode,” and “Inverse variance weighted (IVW).” The IVW approach was the most precise and robust way and served as the primary mean to estimate the causal effect of exposure on outcome with p < 0.05 as the cutoff value. The degree of heterogeneity was tested using Cochran's Q statistic. If the null heterogeneity were rejected, the random effects would be instead of the fixed‐effect model. The MR pleiotropy residual sum and outlier method (MR‐PRESSO) combined with the MR‐Egger intercept test were applied to check the evidence of horizontal pleiotropy using the R package “MR‐PRESSO” [19, 20]. The MR‐PRESSO approach would adjust for horizontal pleiotropy by eliminating outliers and comparing the significance of the differences between the original and corrected data. The leave‐one‐out analysis was also employed to determine if a single SNP drove or influenced the MR analysis. The “product of coefficients” method was used to determine how circulating inflammatory proteins‐mediated GM indirectly affect the risk of IPF. β (impact of GM on circulating inflammatory proteins) * β (effect of circulating inflammatory proteins on IPF) is the indirect effect (β). The overall effect (β) = β (impact of GM on IPF). The mediated fraction is the indirect effect (β)/the overall effect (β). Using the delta approach, standard errors for the indirect effects were calculated.
3. Results
3.1. Total Effect of GM on IPF
The results of IVW analyses revealed that at the order level, Bacillales (odds ratio (OR) = 1.316, 95% confidence interval (CI) [1.006, 1.722], p = 0.045), Gastranaerophilales (OR = 1.441, 95% CI [1.019, 2.036], p = 0.039), and Selenomonadales (OR = 0.536, 95% CI [0.337, 0.941], p = 0.029) had robust causal associations with the risk of IPF; at the family level, Bacteroidaceae (OR = 1.976, 95% CI [1.040, 3.755], p = 0.038) and Family XIII (OR = 0.380, 95% CI [0.193, 0.748], p = 0.005) had; at the genus level, Bacteroides (OR = 1.976 95% CI [1.040, 3.755], p = 0.038), Bifidobacterium (OR = 0.639, 95% CI [0.431, 0.947], p = 0.026), Oscillibacter (OR = 0.571, 95% CI [0.405, 0.806], p = 0.001), Ruminococcus gnavus (OR = 0.681, 95% CI [0.494, 0.940], p = 0.019), Actinomyces (OR = 2.280, 95% CI [1.174, 4.427], p = 0.015), Subdoligranulum (OR = 0.586, 95% CI [0.353, 0.974], p = 0.039), and Veillonella (OR = 0.546, 95% CI [0.304, 0.982], p = 0.043) (Figure 2). No significant evidence of heterogeneity across these results was found using Cochrane Q statistics (Table S2). In addition, no definitive horizontal pleiotropy was detected with MR‐PRESSO and MR‐Egger intercept tests (Tables S3 and S4). In the leave‐one‐out sensitivity analysis, no single SNP seriously violated the overall effect of GM on IPF.
FIGURE 2.
Main results (IVW method) of the causal effect of GM on IPF. CI, confidence interval; GM, gut microbiota; IPF, idiopathic pulmonary fibrosis; IV, instrumental variable; IVW, inverse variance weighted; OR, odds ratio; SNP, single nucleotide polymorphism.
3.2. Effect of Circulating Inflammatory Proteins on IPF
The MR analysis using the IVW method demonstrated suggestive evidence of causality of 8 circulating inflammatory proteins on IPF (Figure 3): CCL11 (OR = 1.377, 95% CI [1.040, 1.822], p = 0.025), CXCL6 (OR = 1.725, 95% CI [1.044, 2.851], p = 0.033), CXCL9 (OR = 1.576, 95% CI [1.052, 2.362], p = 0.028), CCL8 (OR = 1.322, 95% CI [1.076, 1.623], p = 0.008), CCL7 (OR = 1.488, 95% CI [1.083, 2.044], p = 0.014), NRTN (OR = 1.395, 95% CI [1.011, 1.924], p = 0.043), STAMPB (OR = 1.771, 95% CI [1.013, 3.096], p = 0.045), TGFa (OR = 1.581, 95% CI [1.083, 2.310], p = 0.018). No apparent heterogeneity and horizontal pleiotropy were observed (Tables [Link], [Link]–S4). In addition, the leave‐one‐out analysis revealed a single SNP did not drive the overall estimates.
FIGURE 3.
Main results (IVW method) of the causal effect of circulating inflammatory proteins on IPF. CI, confidence interval; IPF, idiopathic pulmonary fibrosis; IV, instrumental variable; IVW, inverse variance weighted; OR, odds ratio; SNP, single nucleotide polymorphism.
3.3. Effect of GM on Circulating Inflammatory Proteins
Next, we investigated the genetic relation between 12 GM and 8 circulating inflammatory proteins. The results showed only the causal role of Actinomyces on CCL11 was significant (OR = 1.221, 95% CI [1.085, 1.373], p < 0.001; Figure 4). The p values derived from Cochran's Q statistic were 0.305 and 0.280, hinting no heterogeneity was determined (Table S2). Moreover, the MR‐Egger intercepts test and MR‐PRESSO confirmed no directional horizontal pleiotropy between exposure and outcome (Tables S3 and S4). The leave‐one‐out analysis showed that no particular SNPs substantially impacted the MR analysis.
FIGURE 4.
The MR analysis of the causal effect of Actinomyces on CCL11. CI, confidence interval; IV, instrumental variable; IVW, inverse variance weighted; OR, odds ratio; SNP, single nucleotide polymorphism.
3.4. The Mediation Effect of GM on IPF
Previously, we identified the causal effect of Actinomyces on IPF and CCL11 as well as the effect of CCL11 on IPF, suggesting potential mediation of Actinomyces to IPF causality via CCL11. Thus, we performed a mediation analysis. In the causal pathway from Actinomyces to IPF, the mediation effect of CCL11 was estimated at 0.063 (95% CI [1.016–1.126]; p = 0.004) with a mediation proportion of 13.035% (Table 1).
TABLE 1.
Mediation effect of GM on IPF via inflammatory protein.
| Mediator | Total effect | Direct effect A | Direct effect B | Mediation effect | p | Mediated Proportion (%) |
|---|---|---|---|---|---|---|
| β (95% CI) | β (95% CI) | β (95% CI) | β (95% CI) | |||
| CCL11 |
0.487 (1.174–4.427) |
0.199 (1.085–1.373) |
0.319 (1.040–1.822) |
0.063 (1.016–1.126) |
0.004 | 13.035 |
Note: Total effect: The causal role of gut microbiota on IPF. Direct effect A: The causal role of GM on inflammatory protein. Direct effect B: The causal role of inflammatory protein on IPF. β (indirect effect) = β (direct effect A) * β (direct effect B). The mediated proportion = β (indirect effect)/β (total effect).
Abbreviations: CI, confidence interval; GM, gut microbiota; IPF, idiopathic pulmonary fibrosis.
3.5. A Reverse MR Analysis
Finally, we performed a reverse MR analysis. The results demonstrated weak evidence of a potential causal effect of IPF on Actinomyces (OR = 0.998, 95% CI [0.975, 1.021], p = 0.849) and CCL11 (OR = 0.994, 95% CI [0.975, 1.014], p = 0.576) (Figure 5). Similar results were observed in the genetic role of CCL11 in Actinomyces (OR = 0.947, 95% CI [0.869, 1.032], p = 0.215). Furthermore, no evidence for heterogeneity and horizontal pleiotropy was observed (Tables [Link], [Link]–S4).
FIGURE 5.
A reverse MR analysis (IVW method) showed the causal effect of IPF on Actinomyces and CCL11 as well as CCL11 on Actinomyces. CI, confidence interval; GM, gut microbiota; IPF, idiopathic pulmonary fibrosis; IV, instrumental variable; IVW, inverse variance weighted; OR, odds ratio; SNP, single nucleotide polymorphism.
4. Discussion
In the present study, we executed a two‐sample MR analysis to seek the causal relationship between GM and IPF. In the past decade, numerous references have proved that GM contributes to the occurrence and development of several pulmonary diseases, like COPD, through the “gut–lung axis” [8]. The current MR study illustrated the increased proportion of five taxa (Bacillales, Gastranaerophilales, Bacteroidaceae, Bacteroides, and Actinomyces) would enhance the risk of IPF; whereas, the decreased abundance of seven taxa (Selenomonadales, Family XIII, Bifidobacterium, Oscillibacter, R. gnavus , Subdoligranulum, Veillonella) would reduce. Regulating immune and inflammatory levels is one of the fundamental biological functions that GM plays. Furthermore, persistent inflammation is a key characteristic of IPF, widely involved in various important pathogenic processes of IPF [11]. The IVW approach showed 8 circulating inflammatory proteins (CCL11, CXCL6, CXCL9, CCL8, CCL7, NRTN, STAMPB, and TGFa) had disadvantageous effects on IPF. The evidence above hinted that GM may increase the development risk of IPF via circulating inflammatory proteins. Hence, we implemented a mediation analysis to validate the hypothesis. Among the 12 taxa of GM, only Actinomyces and CCL11 had a significant connection. The proportion of indirect effects of CCL11 was 0.063, accounting for 13.035% of the total effect. Those data indicated that CCL11 was a critical moderator in the causal pathway from Actinomyces to IPF.
Actinomyces is a genus of bacteria that includes a wide range of species and is found in both human and animal bodies in addition to soil microbial ecosystems [21]. Actinomyces belong to Gram‐positive bacilli that are facultatively anaerobic or microaerophilic rods. Actinomycosis is the most prevalent infection linked to actinomycosis [22]. Sulfur particles, or little yellow clumps, can induce persistent granulomatous infections once an actinomycete has infected the host. Previously, a study has shown that the main commensals of gut Actinomyces in severe COVID‐19 patients were increased and closely related to the severity of the condition [23]. Furthermore, gut Actinomyces could secrete membrane vesicles rich in lipoteichoic acid, which promotes inflammation signaling through TLR2 [24], which exhibited high‐density distribution on the surface of alveolar macrophages in the lungs and could regulate the secretion of proinflammatory cytokines. In addition, Actinomyces was found to induce inflammatory cell death, including pyroptosis, with this effect being partially mediated by the TLR2 pathway [25, 26] The abundance of Actinomyces increased after gastrectomy. After restoring intestinal homeostasis, early postoperative inflammation was reduced, and immune function was enhanced [27]. Additionally, another literature suggested that increased gut Actinomyces may be related to host immunity related to tuberculosis status. The above data indicated that gut Actinomyces may provide biological functions for regulating host immunity and inflammatory response [28].
C–C motif chemokine ligand 11 (CCL11), one of several chemokine genes clustered on the q arm of chromosome 17, is a secreted protein superfamily involved in immune regulation and inflammatory processes, with chemotactic activity towards eosinophils [29, 30]. It is believed to be associated with eosinophilic inflammatory diseases such as atopic dermatitis and allergic rhinitis [29]. The expression of CCL11 in IPF was increased approximately twofold and highly related to pulmonary function. Moreover, inhibiting the release of CCL11 from T helper 2 cells could control the transition from an acute inflammatory response to chronic fibrosis [31]. However, there is currently no research reporting the specific role of CCL11 in IPF. CCL11 can promote the binding of Th2‐related cytokines to TLR, promoting the production of immune cells (eosinophils, macrophages, T cells, and B cells) [32]. This process can be inhibited by Th1‐related interferons. CCL11 can also induce fibroblast production, eosinophil infiltration, and Th2 immune response through different pathways [33]. CCL11 can directly enhance the extracellular matrix generation and microscopic proliferation ability of human lung fibroblasts in vitro, and blocking CCL11 signaling could ameliorate fibrosis in mice [34, 35]. In the study, we assessed the role of the pathway from Actinomyces to CCL11 on IPF. In future research, utilizing these target antibodies as therapeutic methods for liver fibrosis models will certainly add more translational value to current findings.
In the current study, we explored the causal role of GM on IPF and the mediation effect of circulating inflammatory proteins with a two‐sample, two‐step MR study. MR analysis is a powerful statistical strategy based on whole genome association analysis to infer causal relationships between exposure and outcome in epidemiology, which could simulate randomized controlled trials with a lower cost and less risk of a reverse causal effect. However, there were still some shortcomings: (1) All data were gathered from the public dataset, so we cannot evaluate all of the potential horizontal pleiotropy and heterogeneity. (2) Because the GWAS data used are mainly based on the European population, the generalizability of this conclusion is limited, and it is necessary to validate this conclusion in other populations. (3) We used different threshold criteria (p < 1 × 10−5 for GM; p < 5 × 10−8 for inflammatory proteins). Some SNPs that are weakly associated with exposure factors may be influenced by other unknown factors, thereby affecting the reliability of causal effects. (4) The results showed only the pathway: Actinomyces‐CCL11‐IPF was significant. Herein, we only analyzed the role of 211 bacteria and 91 circulating inflammatory proteins on IPF, and only 14 306 samples for bacteria and 14 735 cases for circulating inflammatory proteins were collected, which may lead to bias in results. Hence, the study including more GM and inflammatory proteins, as well as larger sample sizes, is necessary. (5) CCL11 partly mediated the pathway from GM to IPF. Nevertheless, other mediators we have yet to study may also exist, requiring further investigation.
5. Conclusions
Herein, we concluded that GM and circulating inflammatory proteins had a genetically predicted association with IPF. Of note, the causal effect of Actinomyces on IPF was, to a large extent, mediated by CCL11.
Author Contributions
Quanteng Hu and Bo Zhang designed the study and revised the manuscript. Hongyu Zhu and Caihua Chen analyzed the data, performed the experiment, and wrote the manuscript. Haixie Guo was involved in data collection and statistical analysis and reviewed the manuscript. All authors read and approved the final manuscript.
Ethics Statement
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: The selection and harmonization of IVs.
Table S2: The heterogeneity test using Cochrane Q statistics.
Table S3: The horizontal pleiotropy test using MR‐Egger intercept test.
Table S4: The horizontal pleiotropy test using MR‐PRESSO.
Acknowledgments
The authors have nothing to report.
Zhu H., Chen C., Guo H., Zhang B., and Hu Q., “The Role of Gut Microbiota on Idiopathic Pulmonary Fibrosis Mediated by Circulating Inflammatory Proteins: A Two‐Step, Two‐Sample Mendelian Randomization Study,” The Clinical Respiratory Journal 19, no. 9 (2025): e70120, 10.1111/crj.70120.
Funding: The authors received no specific funding for this work.
Hongyu Zhu and Caihua Chen contributed equally to this work.
Contributor Information
Bo Zhang, Email: zhbo@enzemed.com.
Quanteng Hu, Email: huqt@enzemed.com.
Data Availability Statement
The datasets analyzed during the present study were obtained from GWAS Catalog (genome‐wide association studies; https://gwas.mrcieu.ac.uk/). The GWAS ids for GM at five levels (phylum, class, order, family, genus) were from Ebi‐a‐GCST90016908 to Ebi‐a‐GCST90017118. The GWAS ID for IPF was Ebi‐a‐GCST90018120. The accession numbers for circulating inflammatory proteins were from GCST90274758–GCST90274848, including 14 735 cases.
References
- 1. Lederer D. J. and Martinez F. J., “Idiopathic Pulmonary Fibrosis,” New England Journal of Medicine 378, no. 19 (2018): 1811–1823. [DOI] [PubMed] [Google Scholar]
- 2. Raghu G., Remy‐Jardin M., Richeldi L., et al., “Idiopathic Pulmonary Fibrosis (an Update) and Progressive Pulmonary Fibrosis in Adults: An Official ATS/ERS/JRS/ALAT Clinical Practice Guideline,” American Journal of Respiratory and Critical Care Medicine 205, no. 9 (2022): e18–e47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Ley B. and Collard H. R., “Epidemiology of Idiopathic Pulmonary Fibrosis,” Clinical Epidemiology 5 (2013): 483–492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Mortimer K. M., Bartels D. B., Hartmann N., et al., “Characterizing Health Outcomes in Idiopathic Pulmonary Fibrosis Using US Health Claims Data,” Respiration 99, no. 2 (2020): 108–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Hilberg O., Hoffmann‐Vold A. M., Smithm V., et al., “Epidemiology of Interstitial Lung Diseases and Their Progressive‐Fibrosing Behaviour in Six European Countries,” ERJ Open Research 8, no. 1 (2022): 00597‐2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Podolanczuk A. J., Thomson C. C., Remy‐Jardin M., et al., “Idiopathic Pulmonary Fibrosis: State of the Art for 2023,” European Respiratory Journal 61, no. 4 (2023): 2200957. [DOI] [PubMed] [Google Scholar]
- 7. Nallanchakravarthula S., Amruta N., and Ramamurthy C., “Cancer Microbiome; Opportunities and Challenges,” Endocrine, Metabolic & Immune Disorders Drug Targets 21 (2021): 215–229. [DOI] [PubMed] [Google Scholar]
- 8. Ananya F. N., Ahammed M. R., Fahem M. M., et al., “Association of Intestinal Microbial Dysbiosis With Chronic Obstructive Pulmonary Disease,” Cureus 13, no. 11 (2021): e19343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Li N., Yi X., Chen C., et al., “The Gut Microbiome as a Potential Source of Non‐Invasive Biomarkers of Chronic Obstructive Pulmonary Disease,” Frontiers in Microbiology 14 (2023): 1173614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Sun M., Zhang F., Lu F., et al., “Integrating Fecal Metabolomics and Intestinal Microbiota to Study the Mechanism of Cannabidiol in the Treatment of Idiopathic Pulmonary Fibrosis,” Frontiers in Pharmacology 15 (2024): 1358626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Mutsaers S. E., Miles T., Prêle C. M., and Hoyne G. F., “Emerging Role of Immune Cells as Drivers of Pulmonary Fibrosis,” Pharmacology & Therapeutics 252 (2023): 108562. [DOI] [PubMed] [Google Scholar]
- 12. Xu Y., Lan P., and Wang T., “The Role of Immune Cells in the Pathogenesis of Idiopathic Pulmonary Fibrosis,” Medicina (Kaunas, Lithuania) 59, no. 11 (2023): 1984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Van‐Geffen C., Deißler A., Quante M., Renz H., Hartl D., and Kolahian S., “Regulatory Immune Cells in Idiopathic Pulmonary Fibrosis: Friends or Foes?,” Frontiers in Immunology 12 (2021): 663203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kurilshikov A., Medina‐Gomez C., Bacigalupe R., et al., “Large‐Scale Association Analyses Identify Host Factors Influencing Human Gut Microbiome Composition,” Nature Genetics 53, no. 2 (2021): 156–165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Duckworth A., Gibbons M. A., Allen R. J., et al., “Telomere Length and Risk of Idiopathic Pulmonary Fibrosis and Chronic Obstructive Pulmonary Disease: A Mendelian Randomisation Study,” Lancet Respiratory Medicine 9, no. 3 (2021): 285–294. [DOI] [PubMed] [Google Scholar]
- 16. Zhao J. H., Stacey D., Eriksson N., et al., “Genetics of Circulating Inflammatory Proteins Identifies Drivers of Immune‐Mediated Disease Risk and Therapeutic Targets,” Nature Immunology 24, no. 9 (2023): 1540–1551. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Chen J., Yu X., Wu X., Chai K., and Wang S., “Causal Relationships Between Gut Microbiota, Immune Cell, and Non‐Small Cell Lung Cancer: A Two‐Step, Two‐Sample Mendelian Randomization Study,” Journal of Cancer 15, no. 7 (2024): 1890–1897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Kamat M. A., Blackshaw J. A., Young R., et al., “Phenoscanner V2: An Expanded Tool for Searching Human Genotype‐Phenotype Associations,” Bioinformatics 35, no. 22 (2019): 4851–4853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Burgess S. and Thompson S. G., “Interpreting Findings From Mendelian Randomization Using the MR‐Egger Method,” European Journal of Epidemiology 32, no. 5 (2017): 377–389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Lee K. and Lim C. Y., “Mendelian Randomization Analysis in Observational Epidemiology,” Journal of Lipid and Atherosclerosis 8, no. 2 (2019): 67–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Dioguardi M., Quarta C., Alovisi M., et al., “Microbial Association With Genus Actinomyces in Primary and Secondary Endodontic Lesions, Review,” Antibiotics (Basel) 9, no. 8 (2020): 433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Cerrato A., Zanette G., Boccuto M., Angelini A., Valente M., and Bacci C., “ Actinomyces and MRONJ: A Retrospective Study and a Literature Review,” Journal of Stomatology Oral and Maxillofacial Surgery 122, no. 5 (2021): 499–504. [DOI] [PubMed] [Google Scholar]
- 23. Shimizu K., Hirata H., Tokuhira N., et al., “Dysbiosis of Gut Microbiota in Patients With Severe COVID‐19,” Acute Medical and Surgical Care 11, no. 1 (2024): e923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Miyakawa Y., Otsuka M., Shibata C., et al., “Gut Bacteria‐Derived Membrane Vesicles Induce Colonic Dysplasia by Inducing DNA Damage in Colon Epithelial Cells,” Cellular and Molecular Gastroenterology and Hepatology 17, no. 5 (2024): 745–767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Shimada E., Kataoka H., Miyazawa Y., et al., “Lipoproteins of Actinomyces viscosus Induce Inflammatory Responses Through TLR2 in Human Gingival Epithelial Cells and Macrophages,” Microbes and Infection 14, no. 5 (2012): 916–921. [DOI] [PubMed] [Google Scholar]
- 26. Wu Z., Takigawa H., Maruyama H., Nambu T., Mashimo C., and Okinaga T., “TLR2‐Dependent and Independent Pyroptosis in dTHP‐1 Cells Induced by Actinomyces oris MG‐1,” Biochemistry and Biophysics Reports 38 (2024): 101680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Cao W., Zheng C., Xu X., et al., “ Clostridium butyricum Potentially Improves Inflammation and Immunity Through Alteration of the Microbiota and Metabolism of Gastric Cancer Patients After Gastrectomy,” Frontiers in Immunology 13 (2022): 1076245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Huang H. L., Luo Y. C., Lu P. L., et al., “Gut Microbiota Composition Can Reflect Immune Responses of Latent Tuberculosis Infection in Patients With Poorly Controlled Diabetes,” Respiratory Research 24, no. 1 (2023): 11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Nazarinia D., Behzadifard M., Gholampour J., Karimi R., and Gholampour M., “Eotaxin‐1 (CCL11) in Neuroinflammatory Disorders and Possible Role in COVID‐19 Neurologic Complications,” Acta Neurologica Belgica 122, no. 4 (2024): 865–869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Mello J. D. C., Gomes L. E. M., Silva J. F., et al., “The Role of Chemokines and Adipokines as Biomarkers of Crohn's Disease Activity: A Systematic Review of the Literature,” American Journal of Translational Research 13, no. 8 (2021): 8561–8574. [PMC free article] [PubMed] [Google Scholar]
- 31. Dong J. and Ma Q., “In Vivo Activation of a T Helper 2‐Driven Innate Immune Response in Lung Fibrosis Induced by Multi‐Walled Carbon Nanotubes,” Archives of Toxicology 90, no. 9 (2016): 2231–2248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Zhao T., Wu X., Zhao X., Yao K., Li X., and Ni J., “Identification and Validation of Chemokine System‐Related Genes in Idiopathic Pulmonary Fibrosis,” Frontiers in Immunology 14 (2023): 1159856. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Zhang T., Huang Y., Ji X., Wu T., and Xiao P., “CCL11 (Eotaxin) Promotes the Advancement of Aging‐Related Cardiovascular Diseases,” Reviews in Cardiovascular Medicine 26, no. 2 (2025): 26020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Daldegan M. B., Teixeira M. M., and Talvani A., “Concentration of CCL11, CXCL8 and TNF‐Alpha in Sputum and Plasma of Patients Undergoing Asthma or Chronic Obstructive Pulmonary Disease Exacerbation,” Brazilian Journal of Medical and Biological Research 38, no. 9 (2005): 1359–1365. [DOI] [PubMed] [Google Scholar]
- 35. Zheng B., Keen K. J., Fritzler M. J., et al., “Circulating Cytokine Levels in Systemic Sclerosis Related Interstitial Lung Disease and Idiopathic Pulmonary Fibrosis,” Scientific Reports 13, no. 1 (2023): 6647. [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.
Supplementary Materials
Table S1: The selection and harmonization of IVs.
Table S2: The heterogeneity test using Cochrane Q statistics.
Table S3: The horizontal pleiotropy test using MR‐Egger intercept test.
Table S4: The horizontal pleiotropy test using MR‐PRESSO.
Data Availability Statement
The datasets analyzed during the present study were obtained from GWAS Catalog (genome‐wide association studies; https://gwas.mrcieu.ac.uk/). The GWAS ids for GM at five levels (phylum, class, order, family, genus) were from Ebi‐a‐GCST90016908 to Ebi‐a‐GCST90017118. The GWAS ID for IPF was Ebi‐a‐GCST90018120. The accession numbers for circulating inflammatory proteins were from GCST90274758–GCST90274848, including 14 735 cases.