Human genetics influences microbiome composition involved in asthma exacerbations despite inhaled corticosteroid treatment

Background: The upper-airway microbiome is involved in asthma exacerbations despite inhaled corticosteroid (ICS) treatment. Although human genetics regulates microbiome composition, its influence on asthma-related airway bacteria remains unknown. Objective: We sought to identify genes and biological pathways regulating airway-microbiome traits involved in asthma exacerbations and ICS response. Methods: Saliva, nasal, and pharyngeal samples from 257 European patients with asthma were analyzed. The association of 6,296,951 genetic variants with exacerbation-related microbiome traits despite ICS treatment was tested through microbiome genome-wide association studies. Variants with 1 × 10−4 < P < 1 × 10−6 were examined in gene-set enrichment analyses. Significant results were sought for replication in 114 African American and 158 Latino children with and without asthma. ICS-response–associated single nucleotide polymorphisms reported in the literature were evaluated as microbiome quantitative trait loci. Multiple comparisons were adjusted by the false discovery rate. Results: Genes associated with exacerbation-related airway-microbiome traits were enriched in asthma comorbidities development (ie, reflux esophagitis, obesity, and smoking), and were likely regulated by trichostatin A and the nuclear factor-κB, the glucocorticosteroid receptor, and CCAAT/enhancer-binding protein transcription factors (7.8 × 10−13 ≤ false discovery rate ≤ 0.022). Enrichment in smoking, trichostatin A, nuclear factor-κB, and glucocorticosteroid receptor were replicated in the saliva samples from diverse populations (4.42 × 10−9 ≤ P ≤ .008). The ICS-response–associated single nucleotide polymorphisms rs5995653 (APOBEC3B-APOBEC3C), rs6467778 (TRIM24), and rs5752429 (TPST2) were identified as microbiome quantitative trait loci of Streptococcus, Tannerella, and Campylobacter in the upper airway (0.027 ≤ false discovery rate ≤ 0.050). Conclusions: Genes associated with asthma exacerbation–related microbiome traits might influence asthma comorbidities. We reinforced the therapeutic interest of trichostatin A, nuclear factor-κB, the glucocorticosteroid receptor, and CCAAT/enhancer-binding protein in asthma exacerbations.


INTRODUCTION
The human microbiome involves all the microorganisms inhabiting the human body. 1 The elevated number of bacterial species and their genetic diversity at different body sites have aroused interest in investigating the influence of the microbiome on human diseases, especially in asthma and allergies. 1Previous studies reported that the airway and gut microbiome composition and microbial exposures during the lifespan contribute to asthma pathogenesis and treatment response. 1,2These microbial communities are influenced by host genetics and environmental factors, leading to high interindividual variation. 3However, little is known about host genetic variants modifying the composition of the upper-airway microbiome.[5][6] Microbiome genome-wide association studies (mbGWAS) have revealed single nucleotide polymorphisms (SNPs) regulating the microbiome composition, known as microbiome quantitative trait loci (mbQTLs).However, most mbGWAS have focused on analyzing the gut microbiome, and other asthma-relevant tissues have been barely examined. 4,7To date, only 4 mbGWAS have investigated the influence of host genetics on the upper-airway microbiome. 3,5,6,8These studies have reported that mucosal immunity genes and immunityrelated pathways are relevant to regulating the nasal microbiome. 3,8In addition, the heritability of the oral microbiome has been estimated at more than 50%, and the influence of host genetics on the oral microbiome is potentially higher than that of environmental factors. 5,6netheless, there is no study evaluating the effect of genetic loci on airway bacteria previously related to asthma pathogenesis. 1Recently, our group has identified multiple upper-airway microbiome biomarkers with a protective role for asthma exacerbations despite inhaled corticosteroid (ICS) treatment. 2We hypothesize that host genetics regulate the upper-airway microbiome diversity and composition involved in asthma exacerbations and ICS response.This study aimed to identify genetic variants and biological pathways regulating airway microbiome traits involved in asthma exacerbations despite ICS treatment.

RESULTS AND DISCUSSION
A schematic overview of the available data and workflow of this study is represented in Fig E1 (in the Online Repository available at www.jacionline.org).A total of 257 saliva, 232 pharyngeal, and 229 nasal samples from European individuals with asthma from the Genomics and Metagenomics of Asthma Severity (GEMAS) study were analyzed in the discovery phase. 9Their main demographic and clinical characteristics are summarized in Table I.Briefly, subjects had a median age of 39.0 years (interquartile range, 18.0-59.0years) and 61.5% were female.More than 88.0% of patients had severe persistent asthma and 25.8% had a poorly controlled disease.Furthermore, 20.4% of patients reported gastroesophageal reflux disease (GERD), 31.7% obesity, 69.3% atopy, and 27.7% recent antibiotic treatment, and 28.3% were smokers.
Human genome-wide genotypes were imputed using the TOPMed Reference Panel, and microbiome profiling was conducted by targeted sequencing of the 16S ribosomal RNA gene (V3-V4 region), as described elsewhere. 2mbGWAS were conducted to test for the association of 6,296,951 genetic variants with microbiome traits through regression models adjusted for age, sex, and ancestry.We aimed to identify genes and biological pathways associated with microbiome traits (3 alpha diversity indices and 18 bacterial genera) previously associated with asthma exacerbations despite ICS treatment in the GEMAS study. 2 Independent suggestive mbQTLs identified in a total of 24 mbGWAS (P < 1 × 10 −5 ) were included in a gene-set enrichment analysis.After multiple comparisons adjustment (false discovery rate < 0.05), we observed an enrichment in genes previously associated with major asthma comorbidities, including reflux esophagitis (a main consequence of GERD), 10 obesity, and smoking.Furthermore, gene-set enrichment analysis revealed that genes suggestively associated in the mbGWAS partially overlapped with genes whose expression is regulated by trichostatin A (TSA) and transcription factors, including the nuclear factor-κB (NF-κB), the glucocorticosteroid receptor or GR (encoded by the nuclear receptor subfamily 3, group C, member 1 or NR3C1 gene), and CCAAT/enhancer-binding proteins (CEBPs) (Fig 1 and Table II).The robustness of these findings was ensured by varying the input P-value threshold for genetic variants selection to include in the analysis (ie, P < 1 × 10 −6 and P < 1 × 10 −4 ).Enrichment results remained significant (P <.05) after varying this parameter, indicating that our findings are not dependent on the arbitrary P-value threshold for variant selection.Stratified analyses by biological sample showed that these enrichment terms were driven by microbiome traits from different body sites (Table II).
Significant results from the salivary microbiome were followed up for replication in 158 Latino children with and without asthma from the Genes-environments & Admixture in Latino Americans (GALA II) study and 114 African Americans from the Study of African Americans, Asthma, Genes & Environments (SAGE).Their characteristics are summarized in Table I.Bioinformatic analyses were conducted using similar procedures as in the discovery phase.Enrichment in smoking, TSA, NF-κB, and GR was replicated in the mbGWAS of saliva samples from both African American and Latino children with and without asthma (4.42 × 10 −9 ≤ P ≤ .008)(Table II).1][12] Different mechanisms have been suggested to explain how GERD affects asthma, including inflammatory lung injury and vagal nerve stimulation by gastric acid. 10Furthermore, T H 2 cytokines and eosinophilia might mediate the coexistence of asthma with eosinophilic esophagitis, another allergic disease that mimics GERD symptoms. 13However, inconsistent data have been reported about the effectiveness of GERD therapies in asthma, suggesting that the underlying mechanisms between GERD and asthma are not fully elucidated. 10Similarly, asthma and obesity are 2 highly heritable traits with shared mechanisms including genetic factors. 14he knowledge of the genetics of asthma and obesity is limited due to the polygenic character of these traits, 14 and it has been hypothesized that genetic polymorphisms might exert an effect on the obese asthma phenotype through other omic layers. 11A shared feature among these 3 comorbidities is their impact on the airway, salivary, and/or gut microbiome compositions. 11,12,15Moreover, bacterial dysbiosis is considered a link between GERD, obesity, and asthma. 11,15Our study provides novel insights into the shared influence of human genetics on the upper-airway microbiome composition involved in asthma exacerbations and the development of major asthma comorbidities.
In addition, we observed an enrichment in genetic signatures related to TSA and genes regulated by NF-κB and the GR in lung inflammation bronchial epithelial cells.GR is the main mediator of the anti-inflammatory effect of glucocorticosteroids by interacting with transcriptional coactivators and corepressors (eg, histone deacetylase 2 and NF-κB). 16arge evidence supports that deficiencies in GR expression and activity are involved in steroid-resistant asthma. 16NF-κB is a proinflammatory transcription factor involved in airway inflammation in patients with asthma. 17Alterations in NF-κB-related pathways partially explained the heterogeneity response to asthma therapies, including ICS. 17 TSA is an inhibitor of histone deacetylases that regulates NF-κB-driven inflammatory gene transcription and has been demonstrated to reduce airway inflammation in murine asthma models. 18A previous meta-GWAS in European children with asthma exacerbations despite ICS treatment showed enrichment in a TSA genetic signature. 19Our findings reinforce the potential therapeutic use of TSA in asthma by regulating genes involved in ICS response and microbiome composition.Nevertheless, further studies are required to evaluate the safety and efficacy of TSA as an asthma treatment.In addition, we reported enrichment in DNA motifs for CEBP-α and CEBP-β, transcription factors related to the pathophysiology of asthma. 20The potential corticosteroid unresponsiveness and airway cell proliferation in asthma have been related to a lack of CEBP-α in bronchial cells in these patients. 20EBP members are implicated in corticosteroid response, and their expression patterns are regulated by glucocorticosteroids and bronchodilators. 20 addition, we assessed the potential role of SNPs previously associated with ICS response by GWAS as mbQTLs of microbiome traits involved in asthma exacerbations despite ICS treatment (see Table E1 in this article's Online Repository at www.jacionline.org).Among the 21 independent SNPs selected from the literature, 3 SNPs were identified as mbQTLs with a false discovery rate < 0.05 (Table III and Fig 2).The SNP rs5995653, located in the intergenic region of APOBEC3B-APOBEC3C, was associated with the relative abundance of Streptococcus in the nasal microbiome (β for the A allele, 0.34; SE, 0.11; P = 1.90 × 10 −3 ) and the presence of Tannerella in the pharyngeal microbiome (β for the A allele, −1.06; SE, 0.33; P = 1.31 × 10 −3 ).Moreover, the SNPs rs6467778 (TRIM24) (β for the A allele, -−0.36;SE, 0.11; P = 1.45 × 10 −3 ) and rs5752429 (TPST2) (β for the G allele, 0.26; SE, 0.09; P = 4.74 × 10 −3 ) were associated with the relative abundance of Campylobacter in the pharyngeal microbiome.All these associations remained robust in sensitivity analyses after adjusting for asthma exacerbations and potential confounders from the nasal and pharyngeal microbiome composition (all P < .05,Table III).
2][23] APOBEC3B-APOBEC3C are members of the cytidine deaminase gene family, highly expressed in the lower airways, and implicated in innate immunity and host defense against viral infections. 21TRIM24 regulates IL-1 receptor (IL-1R) expression in T cells, a protein whose expression in sputum is associated with severe asthma and participates in IL-1-mediated exacerbations. 24,25On the other hand, TPST2 encodes a sulfotransferase that regulates the sulfation of chemokine receptors involved in asthma T H 2 inflammation. 23Our findings suggest that ICS-responserelated APOBEC3B-APOBEC3C, TRIM24, and TPST2 genetic loci affect airway bacteria associated with asthma exacerbations.
This study has several strengths.First, we integrated microbiome data from 3 different asthma-relevant body sites with human genome-wide data to conduct the first mbGWAS in patients with persistent asthma.Second, we followed reference guidelines and recommendations for microbiome profiling to ensure the robustness of sequencing assays while controlling for many potential confounders in microbiome studies. 2 Third, the reliability and robustness of our findings were ensured by correcting for multiple comparisons, adjusting for covariates, replicating in independent populations, and conducting sensitivity and stratified analyses.Nevertheless, some limitations must be acknowledged.First, our sample size is limited to identify genome-wide significant associations.However, enrichment analyses were used as a powerful tool to identify plausible findings in the absence of genome-wide results. 8,19Second, although we reported evidence of replication in the salivary microbiome, we were not able to replicate the results from the nasal and pharyngeal samples because we only had access to studies with human genome-wide genotype data and 16S-ribosomal RNA-sequenced bacterial communities profiled in saliva samples.Third, the targeted metagenomic approach (16S ribosomal RNA) does not allow us to study the host genetic influence on specific bacterial species and other microorganisms involved in asthma exacerbations and ICS response.
In conclusion, genes suggestively associated with asthma exacerbation-related microbiome traits might have an influence on major asthma comorbidities development in diverse populations (ie, reflux esophagitis, obesity, and smoking).Those genetic loci are significantly more likely to be regulated by TSA and NF-κB, GR, and CEBP transcription factors than expected by chance.Finally, we reported that ICS-response-related genetic loci (APOBEC3B-APOBEC3C, TR1M24, and TPST2) are associated with airway bacteria related to asthma exacerbations.

Supplementary Material
Refer to Web version on PubMed Central for supplementary material.

DISCLOSURE STATEMENT
This study was funded by the Spanish Ministry of Science and Innovation (MICIN), grant number SAF2017-83417R awarded by MCIN/AEI/10.13039/501100011033and "ERDF A way of making Europe," to M.P.-Y.and F.L.-D.M.P.-Y.and J.V. were also supported by CIBER-Consorcio Centro de Investigación Biomédica en Red-(CIBERES), Instituto de Salud Carlos III (ISCIII), and European Regional Development Fund (ERDF) (grant no.CB06/06/1088).This project was also partially funded by grant support from GlaxoSmithKline (Spain) through an agreement with Fundación Canaria Instituto de Investigación Sanitaria de Canarias (FIISC) to M.P.-Y.and J.M.H.-P., Asociación Canaria de Neumología y Cirugía Torácica (NEUMOCAN), and an agreement between the Spanish Ministry of Science, Innovation, and Universities and Universidad de La Laguna (ULL Disclosure of potential conflict of interest: The authors declare that they have no competing interests or other interests that might be perceived to influence the interpretation of the article.No supporting institution may gain or lose financially through this publication.M. Pino-Yanes and F. Lorenzo-Diaz report grants and support from the Spanish Ministry of Science and Innovation (MCIN/AEI/10.13039/50110001103)and "ERDF A way of making Europe."J. Perez-Garcia reports a fellowship from the Spanish Ministry of Universities.A. Espuela-Ortiz reports a fellowship from the MICIN and Universidad de La Laguna (ULL).M. Pino-Yanes and J. M. Hernández-Pérez report grant support from GlaxoSmithKline (Spain) through Fundación Canaria Instituto de Investigación Sanitaria de Canarias (FIISC).J. Villar reports public/academic grants from the Instituto de Salud Carlos III, Madrid, Spain, and the European Regional Development Fund. E. G. Burchard reports grants from the National Institutes of Health, the Tobacco-Related Disease Research Program, the Sandler Family Foundation, the American Asthma Foundation, the Amos Medical Faculty Development Program from the Robert Wood Johnson Foundation, and the Harry Wm. and Diana V. Hind Distinguished Professorship in Pharmaceutical Sciences II.

•
Genes associated with exacerbation-related microbiome traits are related to the susceptibility to major asthma comorbidities (ie, reflux esophagitis, obesity, and smoking).
• TSA and multiple transcription factors have a potential therapeutic interest in asthma exacerbations treatment by targeting genes that regulate exacerbationrelated airway bacteria.
• ICS-response-related genetic loci (APOBEC3B-APOBEC3C, TRIM24, and TPST2) affect airway bacteria associated with asthma exacerbations.Bar plots of the most relevant significant gene-set enrichment results for each database.The −log 10 of the P value adjusted by the false discovery rate (FDR) is represented on the x-axis.
Immunol.Author manuscript; available in PMC 2023 September 26.during patient recruitment.Descriptive statistics are represented by the median (interquartile range) for continuous variables and the count (proportion) for categorical variables.
). M.P.-Y. was also supported by a grant from the Ramón y Cajal Program (grant no.RYC-2015-17205), by MCIN/AEI/ 10.13039/501100011033, and by the European Social Fund "ESF Investing in your future."J.P.-G. was funded by the fellowship FPU19/02175 (Formación de Profesorado Universitario Program) from the Spanish Ministry of Universities.A.E.-O.reports funding from the Spanish Ministry of Science, Innovation, and Universities (MICIU) and ULL, under the M-ULL program.E.M.-G. was supported by a fellowship (TESIS2022010045) awarded by the Board of Economy, Industry, Trade, and Knowledge of the Canarian government, with a European Social Fund cofinancing rate managed by the Canarian Agency for Research, Innovation, Society and Information (ACIISI).Genotyping was performed at the Spanish National Cancer Research Centre, in the Human Genotyping lab, a member of CeGen, PRB3, and is supported by grant number PT17/0019, of the PE I+D+i 2013-2016, funded by ISCIII and ERDF.This research was funded in part by the Sandler Family Foundation, the American Asthma Foundation, the Robert Wood Johnson Clinical Harold Amos Medical Faculty Development Program, the Harry Wm. and Diana V. Hind Distinguished Professor in Pharmaceutical Sciences II, National Institutes of Health (grant nos.R01HL117004, R01Hl128439, R01HL135156, and X01HL134589), the National Institute of Environmental Health Sciences (grant nos.R01ES015794 and R21ES24844), the National Institute on Minority Health and Health Disparities (grant nos.P60MD006902, RL5GM118984, and R01MD010443), and the Tobacco-Related Disease Research Program under award number 24RT-0025 and 27IR-0030 to E.G.-B.The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

TABLE I .
Clinical and demographic characteristics of study populations

TABLE II .
Summary results of the main significant GSEA resultsOnly the top 3 significant results for each database in the discovery phase are summarized (a complete list is described in the Online Repository).
*ChIP-seq assays in human bronchial epithelial cells from human lung inflammation models are reported in this table.J Allergy Clin Immunol.Author manuscript; available in PMC 2023 September 26.

TABLE III .
Summary statistics and sensitivity analyses of SNPs previously associated with ICS response identified as mbQTLsThe main model was adjusted for age, sex, and ancestry.‡ Main model including as covariate asthma exacerbations in the past 6 months despite ICS treatment.§ Main model including the following covariates: antibiotic use, sequencing pool, sampling season, body mass index, and smoke in the past 30 min for nasal samples; and antibiotic treatment, cavities, and liquid intake in the past 30 min for pharyngeal samples.