Defining shared/distinct genetic mechanisms for respiratory infections and subsequent respiratory disease

Respiratory infections refer to a range of respiratory diseases involving the respiratory tract which are characterised by generally short periods of infection and inflammation. While many individuals only experience benign, self-limiting infections, certain individuals are at risk of significant morbidity and mortality, with many lower respiratory infections, such as pneumonia, being some of the leading causes of death due to infectious disease worldwide. Environmental factors, such as smoking, have long been known to be risk factors for developing respiratory infections. Genetic factors may also play a role, however the biological mechanisms underlying existing statistical associations remain unclear. If respiratory infections had a degree of shared aetiology with related traits, then this may be informative for the development of novel therapeutics. In this thesis, genetic variants showing association with respiratory infection disease risk and/or frequency are investigated and the genome-wide genetic correlation between respiratory infections and other pulmonary traits is quantified.

A genome-wide case-control analysis of respiratory infection disease risk revealed 56 putatively associated signals at a p-value <5×10−6, one of which surpassed genome-wide significance (p-value <5×10−8). The genome-wide significant signal was located in PBX3, a transcription factor, and in a functional follow-up analysis, this signal was found to colocalise with PBX3-specific expression quantitative trait loci affecting expression in, among others, lung tissue, T cells and whole blood. Statistical replication in nine independent cohorts was not supportive of this signal, however.

The Extended Cohort for E-health, Environment and DNA (EXCEED) study of 10,000 individuals from Leicester, Leicestershire and Rutland was used to develop a primary care-based respiratory infection disease frequency phenotype, and was established as a resource for genetic studies.

A further genome-wide analysis of respiratory infection disease frequency revealed 51 putatively associated signals using the same criteria as above. One signal achieved genome-wide significance after correction for genome-wide inflation. This signal was located in SEMA3F-AS1, an RNA gene. Despite this signal being located in SEMA3F-AS1, the sentinel variant was associated with increased expression of the proximal RBM6 gene across a very large number of tissues and cell types. Socioeconomic factors were highlighted as being significantly correlated with respiratory infection frequency, which warrants further study.

Finally, there was significant, but not complete, genome-wide genetic correlation between the two respiratory infection analyses, suggesting a high degree of shared aetiology between risk and frequency, though there was little overlap in the top association signals from the two studies. With lung function, there was somewhat weak, yet highly significant, genome-wide genetic correlation with respiratory infection risk and frequency, highlighting shared aetiology between respiratory infections and measures of lung function.