Abstracts

Abstract

Background: Eosinophils play a key role in the allergic response in asthma by the release of cytotoxic molecules such as eosinophil cationic protein (ECP) and eosinophil-derived neurotoxin (EDN) that generate epithelium damages. Objective: We aimed to identify genetic variants influencing ECP and EDN levels in asthma-ascertained families of European ancestry. Methods: We conducted univariate and bivariate genome-wide association analyses of these proteins in 1,018 subjects from the EGEA study with follow-up in 153 subjects from SLSJ study and performed meta-analysis to combine evidence from the two datasets. We then conducted Bayesian statistical fine-mapping together with in silico quantitative trait locus and functional annotation analyses to identify credible sets of variants and target candidate genes. Results: We identified four genome-wide significant loci (P<5x10-8) including six distinct signals associated with ECP and/or EDN levels. These six signals were located on 14q11, 7p21, 1p31 and 9q22 chromosomal regions. Four of the six distinct signals were fine-mapped to small credible sets of putative causal variants (95% credible set size =10 SNPs). More particularly, the two signals on 7p21 locus each included one SNP with high posterior inclusion probability (PIP>0.7). The most likely candidate genes targeted by these SNPs were: RNASE2 and RNASE3 (14q11), AK4 (1p31), CTSL (9q22), and NDUFA4 (7p21). Conclusion: This study highlights the interest of joint analysis of biological phenotypes involved in the pathophysiological mechanisms of asthma to increase power to detect new loci and candidate genes.