| Summary: | <p>Abstract</p> <p>Background</p> <p>Rheumatoid arthritis (RA) and Alzheimer's disease (AD) are inversely associated. To test the hypothesis that genetic elements associated with increased RA risk are associated with decreased AD risk, we evaluated RA genetic risk factors recently identified in genome-wide association studies (GWAS) for their association with AD in a two-stage, case-control analysis.</p> <p>Results</p> <p>In our Stage 1 analysis of ~800 AD and ~1,200 non-AD individuals, three of seventeen RA-associated SNPs were nominally associated with AD (p < 0.05) with one SNP, rs2837960, retaining significance after correction for multiple testing (p = 0.03). The rs2837960_G (minor) allele, which is associated with increased RA risk, was associated with increased AD risk. Analysis of these three SNPs in a Stage 2 population, consisting of ~1,100 AD and ~2,600 non-AD individuals, did not confirm their association with AD. Analysis of Stage 1 and 2 combined suggested that rs2837960 shows a trend for association with AD. When the Stage 2 population was age-matched for the Stage 1 population, rs2837960 exhibited a non-significant trend with AD. Combined analysis of Stage 1 and the age-matched Stage 2 subset showed a significant association of rs2837960 with AD (p = 0.002, OR 1.24) that retained significance following correction for age, sex and APOE (p = 0.02, OR = 1.20). Rs2837960 is near <it>BACE2</it>, which encodes an aspartic protease capable of processing the AD-associated amyloid precursor protein. Testing for an association between rs2837960 and the expression of <it>BACE2 </it>isoforms in human brain, we observed a trend between rs2837960 and the total expression of <it>BACE2 </it>and the expression of a <it>BACE2 </it>transcript lacking exon 7 (p = 0.07 and 0.10, respectively).</p> <p>Conclusions</p> <p>RA-associated SNPs are generally not associated with AD. Moreover, rs2837960_G is associated with increased risk of both RA and, in individuals less than 80 years of age, with AD. Overall, these results contest the hypothesis that genetic variants associated with RA confer protection against AD. Further investigation of rs2837960 is necessary to elucidate the mechanism by which rs2837960 contributes to both AD and RA risk, likely via modulation of <it>BACE2 </it>expression.</p>
|
|---|
