SNPs associated with CSF tau levels modify rate of progression in Alzheimer's disease

Abstract

Background: Diagnostic misclassification has consistently been reported in complex diseases such as AD. The ‘‘gold standard’’ for a definitive diagnosis of AD is performed by neuropathological examination post-mortem. We stratified our AD families previously reported in a genome-wide scan, into a subset of 24 families with at least one autopsy-confirmed AD case per family. Methods: The total sample set included 78 patients and 60 unaffected siblings and spouses. The range of cases per family was 2-8 (mean 3.5) and the average AAO was 68.8 (range 42-84) years. A linkage analysis of the 24 pedigrees was performed showing a significant spt LOD score of 4.4 in 8q24, followed by fine mapping under the linkage peak. First by microsatellites in the same 24 families with additional siblings and secondly by gene-wide SNP genotyping of 22 selected genes in an extended family material consisting of 30 families (the original 24 plus six novel families).The SNP data were first subjected to a family-based association analysis using PLINK, followed by linkage analysis using a combination of SNP genotypes and the already acquired microsatellite genotypes. We have also sequenced the translated parts of two candidate genes in the linked region; NDRG1 and KHDRBS3. Results: A suggestive linkage was obtained at peak marker rs6577853 (mpt LOD 1⁄4 2.4) and a spt LOD of 3.2 at D8S1746. Twelve SNPs with p-values less than 0.05 were obtained in the PLINK analysis; however none of the SNPs survived correction for multiple testing. The lowest obtained p-value was recorded for the intronic tagSNP rs2252696 in the SLA/TG region. No mutation was identified in the sequence analysis of NDRG1 and KHDRBS3. Recombination events were observed under the linkage peak in five of the 30 analyzed families. These recombinations can be utilized to identify the smallest shared chromosomal interval; in this case a 3.2 Mb region flanked by markers D8S256 and D8S1761. Conclusions: A handful genes are located in the interval between the peak markers obtained in the mpt and spt linkage analysis, supported by the observed recombinations, and none of the genes have previously been implicated in AD.