Genetic and epigenetic characterisation of respiratory disease and immune responses

Abstract

Background: Asthma is the most common chronic respiratory disease in the world. Genome-wide association study (GWAS) asthma-associated SNPs from large GWAS studies only explain a fraction of the genetic heritability. Likely causes of the missing heritability are broad phenotype definitions and gene-environment interactions (GxE). The mechanisms underlying GxE in asthma are still poorly understood. Enrichment analysis of GWAS-associated SNPs for cell-type specific functional regions suggested that immune cell-types, and especially CD4+ T-cells, play an important role in the asthma pathogenesis. The aim of this thesis was to increase our understanding of asthma genetics and GxE by using novel datasets and techniques coupled with more precise asthma definitions and cell-types. Methods: We measured cytokine responses (n=28) to stimuli including live viruses and bacteria (n=14) in PBMC from children in the Manchester Asthma and Allergy Study (MAAS) cohort at ages 11 (n=332) and 16 (n=317) years. Linear regression was used to identify genetic variants associated with cytokine responses - response cytokine-QTL (re-cQTL). In a MAAS subset, we isolated CD4+ T-cells to investigate differences in gene expression (RNA-seq n=45) and chromatin interaction (promoter capture Hi-C n=15) between asthmatics and non-asthmatics using Weighted Gene Co-expression Network Analysis (WGCNA). Finally, the five cohorts from the Study Team for Early Life Asthma Research (STELAR) consortium (n=8390) were used to study the GxE between the 17q21 asthma-risk variant rs2305480 and early life pet ownership in relation to Latent Class Analysis (LCA)-derived wheezing classes (multinomial logistic regression) and asthma binary phenotypes (logistic regression). Results: A total of 151 and 127 re-cQTL were identified at age 11 and age 16, respectively. TLR1 missense variant rs5743618 was identified as a significant re-cQTL for IFN-gamma (pAge11=1.36x10^-08; pAge16=4.67x10^-18) and IL-13 (pAge11=8.77x10^-12; pAge16=1.47x10^-09) levels in response to PAM3CSK4 at both ages. Apart from rs5743618, all other re-cQTL showed age-specific associations. Both sets of re-cQTL showed a significant enrichment for GWAS-associated SNPs for childhood asthma, but not adult asthma. WGCNA analysis of RNA-seq data identified three modules (sets of co-expressed genes) associated with asthma at FDR 10%. The modules showed a significant enrichment for the GO biological processes of response to virus (increased in asthma) and antigen processing/presentation (decreased in asthma). Coupling WGCNA with bulk RNA-seq cell decomposition a significant association for the modules with naive and memory regulatory T-cells was also identified. The module's hub gene (gene with the highest number of connections in the network) ENTPD1, a gene important in Treg activation, showed significantly lower gene expression and chromatin interaction in asthmatics compared to controls. Meta-analysis of the STELAR cohorts identified a significant association of rs2305480 with increased risk of wheeze while no association was found between dog and cat ownership for either asthma or wheeze when genetics was not taken into account. A significant GxE interaction between rs2305480 and dog ownership (p=0.02), but not cat ownership (p=0.93), was identified in relation to the LCA class Persistent wheeze. Children homozygous for the asthma-risk allele and a dog in early life were at no increased risk of wheeze. Conclusions: In heterogeneous diseases, studying genetic associations in a cell-type and stimulation specific context can identify associations missed by GWAS studies while chromatin conformation capture techniques can increase our understanding of gene regulation in disease. Using phenotypes derived from LCA models in GxE studies can identify new interactions missed by broad binary definitions.

Date of Award	1 Aug 2022
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Magnus Rattray (Supervisor), Angela Simpson (Supervisor) & John Curtin (Supervisor)