SNP based strategies to study candidate genes for Alzheimer’s disease

Abstract

Alzheimer’s disease (AD) is the most common form of dementia in the elderly. It is a genetically heterogeneous disease characterized by progressive cognitive decline and memory impairment. The rare familial form of AD is caused by three different genes called APP, PSEN1 and PSEN2. However, the predominant form of AD is a genetically complex disorder involving a combination of genetic factors. To date, the only risk factor identified for the complex form of AD is the APOE-epsilon 4 allele, but several susceptibility genes remain to be found.

This thesis outlines different strategies to use common genetic variation, in the form of single nucleotide polymorphisms (SNPs), to examine candidate genes and candidate regions for AD. Large-scale genotyping is a prerequisite for performing complex disease studies using SNPs. The validity and accuracy of a newly developed genotyping assay called Dynamic allele specific hybridization (DASH) was therefore investigated. DASH was shown to be a robust genotyping method, and was proven to work as well or better than several other available methods. 'the method was first implemented for a candidate gene association study of a promoter polymorphism in the TNFRSF6 gene. Significant association was found between this variant and early onset AD, indicating its possible role in disease etiology.

A large candidate pathway association study effort was then started testing for association between AD and 60 different SNPs. Genes were picked from four different pathways related to AD; oxidation, inflammation/apoptosis, amyloid interacting genes and a group of candidate genes previously showing significant association with AD. None of the markers showed significant disease association after correction for multiple testing. Although largely negative, these results high-lighted several methodological and study design issues related to association studies in general.

The most successful approach yet in dissecting complex disease using genetic variation has been to perform high resolution linkage disequilibrium (LD) mapping of regions indicated by linkage. Several independent research groups recently reported linkage peaks for AD on chromosome 10q We choose two regions under the 10q linkage peak for LD mapping studies. The first region contained the previously associated TNFRSF6 gene, and the other region included the insulin-degrading enzyme (IDE) gene, which has been shown to be involved in clearance of amyloid-beta. LD maps were created for all pair-wise markers in the two regions to determine the genetic LD structure. Haplotypes were estimated and haplotype tagging markers were chosen for further analysis.

Association analyses were performed for both single markers and haplotypes for case/control status as well as for quantitative traits related to the AD phenotype. Only weak significant signals were found for the TNFRSF6 gene. However, several significant associations were found for a large LD block including the IDE gene. The same haplotypes were always over-represented in cases compared to controls, or with more severe AD within the patient groups. These results indicate a role in AD for one of the three genes situated within the 276kb LD block including the IDE, KNSLI and HHEX genes. Further studies will now be required to identify the underlying risk alleles within the region.