Comparative sequencing of candidate genes in complex disease

Abstract

Complex multi-factorial diseases such as cardiovascular, metabolic, neurological and respiratory disorders affect a great number of people across the world. In the post-Human Genome Sequencing era, genome wide association studies are increasingly viable alternatives to linkage approaches in locating disease genes. In addition, positional and hypothesis-driven candidate genes can be assessed for their role in susceptibility to sporadic common diseases. The efficiency and success of these approaches depend on knowledge of DNA variation and linkage disequilibrium. This thesis describes the comparative sequencing of regions across the candidate genes HTR2C, MA OA MAOB, IDE, KIF11 and HHEX to illustrate the importance of understanding the fine scale nucleotide and LD distribution for improvements in association study design with Obsesty, Depression and Alzheimer's disease.

In HTR2C, recombination between the commonly used nsSNP marker, Cys23Ser, and the promoter was observed (Paper I). Furthermore, nucleotide and haplotype analysis showed that gene conversion in the promoter contributed to the complexity of LD. If the functional promoter polymorphisms act in the susceptibility to serotonergic-related phenotypes, this work suggests that the unknown structure of LD across HTR2C could have been an issue in previous association studies using Cys23Ser. Support for HTR2C promoter polymorphisms in obesity was provided by the associations of promoter haplotype, TA 13GCG ( P>0.0001) with BMI 30 > kgm-2 and promoter SNP, -995G>A, (P = 0.01) with serum-leptin/%body fat (Paper I). Suggestive, but not significant, effects were observed by the HTR2C promoter haplotype GGCC in depression (Paper IV).

Sequence variation was scarce in the AL40 regions studied. This contributes to the hypothesis that these genes are under selective pressure and that much of the variants in public databases for MAOA/B could be population specific (Paper II). Lack of MAO variation was reflected in the poor validation rate of SNPs and LD complex structure in a Swedish twin sample group. While MAOB SNPs were found to correlate with depressive state in elderly Swedish Twins, no association was observed with polymorphisms in this gene and trbcactivity (Paper II). Conversely, low trbc-activity was found to associate with MAOA SNPs and haplotypes. A potentially additive effect on the risk for depression per MAO haplotype was observed. In a larger sample no significant associations were found with any of the MAO SNPs are haplotypes (Paper IV). However, a trend towards departures from HWE between the genes may suggest that this region warrents further sequencing to identify potential regulatory mechanisms of MAO expression.

A wealth of polymorphisms was found in re-sequencing IDE, KIF11, HHEX and conserved regions within a haplotype block associated with Alzheimer's Disease. However, no significant associations between IDE and KIF11 SNPs, and Alzheimer's disease were observed. These works demonstrate the advantages of re-sequencing in providing a better understanding of the various genetic factors influencing studies of polymorphisms with complex diseases. With advances in technology and throughput, sequencing will become instrumental in the location of disease genes and the identification of causative polymorphisms.