Genetic association studies in allergic rhinitis

Abstract

Allergic rhinitis (AR) is a global health problem that causes major disability worldwide. Nasal obstruction, secretion and itching are characterizing features of the disease. The development and severity of AR are determined by a complex interaction between environmental and genetic factors and the heritability for AR has been estimated to be high. Genetic association studies are commonly used to investigate complex disease. Most association studies have focused on common variation in the form of single nucleotide polymorphisms (SNPs), where allele and genotype frequencies are compared between cases and controls. More than 100 SNPs have previously been associated with AR.

In the first study of this thesis, the general reproducibility of previous AR associations were evaluated. The overall result showed that very few of the investigated SNPs were associated with AR in the study populations. The study also indicated that odds ratios were inflated in most of the original studies, and concordance of risk alleles between the studies was low. Since the genetic background of asthma has been far more investigated compared to the genetics of AR and since asthma is closely related to AR, the second study investigated genetic variation in 21 highly replicated asthma-associated genes for associations also in AR. Only two genes were identified as potential links between AR and asthma, indicating the different genetic make-up of these diseases. Three meta studies of genome-wide association studies (mGWAS) recently identified a total of 47 index SNPs associated with different AR phenotypes.

In the third study, these SNPs were investigated in a replication study using the same SNPs and the same phenotype definitions as in the mGWAS. Two out of four loci (TLR1-TLR6 and HLA- DQA1-HLA-DQB1) identified by all three original studies were also detected in the replication study. This suggests a central role of these loci in the epidemiology of allergic disease. In addition, associations between genetic variation in the SSTR1-MIPOL1 and TSLP-SLC25A46 loci and age at which the allergic symptoms started was also identified. This was the first report of age at onset effects in AR. The Toll-like receptors (TLRs) have earlier been investigated for their involvement in different allergic diseases.

In the fourth study, common genetic variation in the TLR genes was investigated for association with AR in two ethnically different populations. The TLR7-TLR8 locus was identified as associated with AR in both populations in a sex-specific manner. In addition, weak associations were also observed for TLR1 and TLR6.

In the fifth study of this thesis, rare variation in both the coding sequences and the putative promoter regions of the TLR genes were investigated using sequence data from 288 AR patients and a European subset (EUR) of the 1000 Genomes project. The promotor region of TLR10 showed a significant accumulation of SNPs with minor allele frequency ≤ 1% in the AR population compared to the EUR population. Another potential accumulation was a nonsense mutation, S324* in TLR1, estimated to 5 copies in the AR population but none in the EUR populations. This indicates that both common and rare SNPs in the TLR genes contribute to AR. In summary, this thesis demonstrates that both rare and common variation are associated with AR and highlights the importance of the TLR10-TLR1-TLR6 locus in the development of the disease. It also emphazises the need for large and well-characterized populations in association and replication studies.