|Home | About | Journals | Submit | Contact Us | Français|
Recently genetic variation in the DAPK1 and ABCA2 genes has been reported to be associated with late-onset and early-onset Alzheimer’s disease (AD), respectively. We examined the most significant two single-nucleotide polymorphisms (SNPs) in DAPK1 in a large case–control cohort of late-onset subjects and matched controls and one of the most significiant SNPs in ABCA2 in a small set of early-onset subjects as well. We did not detect associations with AD for any variation.
Alzheimer’s disease (AD) is a devastating neurodegenerative disorder with complex environmental and genetic etiology. Variation in four genes has been consistently associated with risk of AD: APP, PSEN1 and PSEN2 with early-onset AD (EOAD) and APOE with late-onset AD (LOAD) (Kamboh, 2004). Linkage studies have identified regions on several chromosomes, including chromosome 9 (Kamboh, 2004). Recently, variation in two genes on chromosome 9q34, DAPK1 (death-associated protein kinase 1) (Li et al., 2006) and ABCA2 (ATP-binding cassette, sub-family A, member 2) (Macé et al., 2005; Wollmer et al., 2006) has been associated with AD.
Li et al. (2006) reported an association between DAPK1/rs4878104 (c.63−26878C>T) and rs4877365 (c.62+41638G>A) and LOAD. Two other groups (Macé et al., 2005; Wollmer et al., 2006) reported associations between ABCA2/rs908832 (c.2037T>C) and EOAD.
Here we examined three single-nucleotide polymorphisms (SNPs) in a large case–control population of LOAD subjects and age-matched controls and a small series of EOAD cases for ABCA2/rs908832, in an effort to replicate those findings. (See Supplementary Material for case–control description and genotyping methodology.)
Genotype and allele frequencies for the three markers are presented in Table 1. Both DAPK1 SNPs were in significant linkage disequilibrium (D′ = 0.946, r2 = 0.633). Differences between frequencies in cases and controls did not differ significantly for either DAPK1/rs4878104 (genotype p = 0.224; allele p = 0.383) or DAPK1/rs4877365 (genotype p = 0.993; allele p = 0.928). Our power to detect Li and colleagues’ (2006) odds-ratio of 0.87 for rs4878104 and 0.80 for rs4877365 in their combined sample was 43.2% and 91.6%, respectively. DAPK1 is located 3.9 Mb away from the ubiquilin 1 gene (UBQLN1); because UBQLN1 has also been associated with AD risk, we performed a post hoc analysis of interactions between these two genes and found no interactions (see Supplementary Material).
For ABCA2, there were no statistically significant differences between LOAD and controls (genotypes, p = 0.315; alleles, p = 0.409), EOAD and controls (genotypes, p = 0.184; alleles, p = 0.184), or EOAD and LOAD (genotypes, p = 0.132; alleles, p = 0.372). Our study had 99.4% power to detect the odds-ratio of 3.82 reported by Macé et al. (2005) and 92.4% power to detect the odds-ratio of 6.82 reported by Wollmer et al. (2006).
The statistically significant results that Macé et al. (2005) found with the ABCA2/rs908832 SNP in their hypothesis-testing and total samples appears to be due solely to the EOAD subgroup (p = 5 × 10−5 in EOAD versus p = 0.65 in LOAD). Wollmer and colleagues (2006) examined three groups for this ABCA2 SNP: Swiss, Greeks and Japanese. In each group they found contrastive results. In the Swiss, they observed a significant association between EOAD and rs908832 (p = 0.008); in the Greeks, no significance (p = 0.4); and in the Japanese, the variation was monomorphic. In our Caucasian American sample, we observed no significant association either with EOAD or LOAD.
In summary, the DAPK1 and ABCA2 SNPs examined showed no significant association in our AD cohort. The genetics of Alzheimer’s disease, especially LOAD, has proven to be complicated, with many apparently spurious results. Pinning down causal variants, or variation correlated with such variants, under AD linkage peaks has been diffcult for several chromosomes, and chromosome 9 has been no exception.
This study was supported by National Institute on Aging grants AG13672 and AG05133.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
None of the authors has any actual or potential conflicts of interest that could inappropriately influence this work. This study was approved by the University of Pittsburgh Institutional Review Board.