ABSTRACT
Genetic variation, both single-nucleotide variations and copy number variations (CNV), contribute to changes in gene expression. In some cases these variations are meaningfully correlated with disease states. We hypothesized that in a genetically heterogeneous disorder such as sporadic Alzheimer's disease (AD), utilizing gene expression as a quantitative trait and CNVs as a genetic marker map within the same individuals in the context of case-control status may increase the power to detect relevant loci. Using this approach an 8-kb deletion was identified that contains a PAX6-binding site on chr2q33.3 upstream of CREB1 encoding the cAMP responsive element-binding protein1 transcription factor. The association of the CNV to AD was confirmed by a case-control association study consisting of the Texas Alzheimer Research and Care Consortium and NIA-LOAD Family Study data sets.
Subject(s)
Alzheimer Disease/genetics , Cyclic AMP Response Element-Binding Protein/genetics , Adult , Aged , Aged, 80 and over , Alzheimer Disease/metabolism , Case-Control Studies , Cyclic AMP Response Element-Binding Protein/metabolism , DNA Copy Number Variations , Eye Proteins/metabolism , Female , Gene Expression Profiling , Genetic Predisposition to Disease , Genotype , Homeodomain Proteins/metabolism , Humans , Male , Middle Aged , PAX6 Transcription Factor , Paired Box Transcription Factors/metabolism , Repressor Proteins/metabolismABSTRACT
OBJECTIVES: Copy number variants (CNVs) have been recognized as a source of genetic variation that contributes to disease phenotypes. Alzheimer disease (AD) has high heritability for occurrence and age at onset (AAO). We performed a cases-only genome-wide CNV association study for age at onset of AD. METHODS: The discovery case series (n = 40 subjects with AD) was evaluated using array comparative genome hybridization (aCGH). A replication case series (n = 507 subjects with AD) was evaluated using Affymetrix array (n = 243) and multiplex ligation-dependent probe amplification (n = 264). Hazard models related onset age to CNV. RESULTS: The discovery sample identified a chromosomal segment on 14q11.2 (19.3-19.5 Mb, NCBI build 36, UCSC hg18 March 2006) as a region of interest (genome-wide adjusted p = 0.032) for association with AAO of AD. This region encompasses a cluster of olfactory receptors. The replication sample confirmed the association (p = 0.035). The association was found for each APOE4 gene dosage (0, 1, and 2). CONCLUSION: High copy number in the olfactory receptor region on 14q11.2 is associated with younger age at onset of AD.
