Epigenome‐wide association studies in asthma: A systematic review

Abstract

Objective Asthma is a common chronic respiratory airway disease influenced by environmental factors and possibly their interaction with the human genome causing epigenetic changes. Epigenome‐wide association studies (EWAS) have mainly investigated DNA methylation and its association with disease or traits, exposure factors or gene expression. This systematic review aimed to identify all EWAS assessing differentially methylated sites associated with asthma in humans. Design Structured systematic literature search following PRISMA guidelines, Newcastle‐Ottawa Scale (NOS) for cohort studies was used for bias assessment. Data Sources We searched PubMed and Embase databases from 2005 to 2019. Eligibility Criteria Epigenome‐wide association studies testing association between differential methylation and asthma in humans. Results Overall, we identified 16 EWAS studies complying with our search criteria. Twelve studies were conducted on children, and 10 were conducted on sample sizes <150 subjects. Four hundred and nineteen CpGs were reported in children studies after correction for multiple testing. In the adult studies, thousands of differentially methylated sites were identified. Differential methylation in inflammatory‐related genes correlated with higher levels of gene expressions of inflammatory modulators in asthma. Differentially methylated genes associated with asthma included SMAD3, SERPINC1, PROK1, IL13, RUNX3 and TIGIT. Forty‐one CpGs were replicated at least once in blood samples, and 28 CpGs were replicated in nasal samples. Conclusion Although many differentially methylated CpGs in genes known to be involved in asthma have been identified in EWAS to date, we conclude that further studies of larger sample sizes and analyses of differential methylation between different phenotypes are needed in order to comprehensively evaluate the role of epigenetic factors in the pathophysiology and heterogeneity of asthma, and the potential clinical utility to predict or classify patients with asthma.

Call for papers for special issue in 2020

“…Also, a relative hypermethylation of cg04256470 (688 bases upstream of CORT) was associated not needing oral corticosteroid in all cohorts. These are interesting, robust associations which may help to elucidate why some children with asthma do better than others and echo other studies …”

“…These are interesting, robust associations which may help to elucidate why some children with asthma do better than others and echo other studies. 3,4 Ballardini et al have assessed the size of the association between childhood atopic dermatitis and asthma with childhood behaviour. 5 They analysed data from the PROmotion of Breastfeeding Intervention Trial (PROBIT) cohort where participants were followed up to 6.5 years of age.…”

Recent findings in the genetics and epigenetics of asthma and allergy

“…The so-far largest cross-sectional EWAS (631 cases and 2231 controls) using nine different cohorts increased the number of differentially methylated CpGs associated with asthma to 179 CpGs and 36 regions [27]. In general, there is significant overlap of the findings in EWAS analyzing asthma or atopy [31]. EWAS have shown to explain better the variation in a phenotype than GWAS as demonstrated for, e.g., circulating IgE levels [35] and levels of asthma related proteins such as CHI3L1 are partly mediated by DNA methylation changes, but not genetic variation [36].…”

“…In general, there is significant overlap of the findings in EWAS analyzing asthma or atopy [31]. Overall, despite the huge advances in the last years, there is still considerable heterogeneity in the published studies.…”

“…Overall, despite the huge advances in the last years, there is still considerable heterogeneity in the published studies. A recent review on EWAS studies in asthma identified among the thousands of CpGs associated with asthma in recent years, that only 41 of the associations were identified in at least one other study [31]. The epigenetic landscape is specific for a given cell thus requiring careful selection of the cell type of relevance for a given biomedical question as well as taking potential confounding effects caused by differential cell composition between, e.g., patients and controls into account [39].…”