[P4–080]: CANDIDATE GENE STUDIES AND GWAS SUGGEST SUBSTANTIAL GENETIC INFLUENCE ON DEFICITS IN OLFACTORY IDENTIFICATION AMONG PERSONS AT RISK OF AD

The olfactory network is vulnerable to AD pathology. We recently reported that deficits in olfactory identification (OI) are associated with CSF AD biomarkers. Past genome-wide association studies (GWAS) suggested that MAPT (encodes microtubule-associated protein tau) is a susceptibility locus for OI impairment (Dong, 2015). The CDK5 signalling pathway (involves tau phosphorylation) was also strongly associated with OI. Because genetic influences on OI may illuminate underlying structures’ susceptibility to neurodegeneration, we investigated association of polymorphisms with OI. Using a candidate gene approach in individuals at risk of AD, we examined association of OI with neurodegeneration-associated SNPs rs199443 in MAPT and rs10984186 in CDK5RAP2, focusing on OI – SNP associations reported by Dong (2015, 2016) and several other AD-related SNPs (Lambert, 2013). Pyrosequencing techniques and a 2.5M-SNP GeneChip provided data for these analyses and an exploratory GWAS for SNPs associated with OI. We studied the family-history-positive PREVENT-AD cohort of cognitively normal individuals aged 55–83 using the 40-item University of Pennsylvania Smell Identification Test (UPSIT). DNA was extracted from blood. Automated genotyping relied on the Illumina Infinium Omni 2.5M-8 GeneChip. We used Matlab to evaluate the differences in OI across polymorphisms and in linear regression for allele-dosage studies. We identified SNPs associated with OI using PLINK and Haploview. Genome-wide data were available for 136 participants. After quality-control 2.28 million SNPs were available for analysis. We found a modest association of OI impairment with rs10984186 in CDK5RAP2 (Kruskal-Wallis, p<0.05) and a trend for rs199443 in MAPT (p=0.07; Figs 1 & 2). Allelic dose response analysis further clarified these correlations (CDK5RAP2 β=0.0521, p=0.07; MAPT β=0.701, p=0.0238). Several SNPs identified by Dong (2015, 2016) and Lambert (2013) showed no significant association. The GWAS identified 26 SNPs with genome-wide significance (P < 5 × 10−8; Fig 3), of which 22 survived adjustment for age, gender, and APOE ε4 status (Fig 4). In a sample at risk of AD we found limited association of OI impairment with polymorphisms in MAPT and CDK5RAP2. A GWAS identified several new SNPs related to OI, including three on chromosome 14 and one on chromosome 3 with p-values <10–12. UPSIT and the MAPT rs199443 polymorphisms. The C allele is the risk allele is associated with increased MAPT expression level in the frontal cortex [2](Dong, 2015). Presence of the C allele is associated with lower OI in older adults at risk of AD. Participants included were 81 CC, 44 CT, 10 TT. A Kruskal-Wallis revealed a trend for group differences in OI (X2(2)=5.29, p=0.071, df=2, n=135). Linear regression suggested that a minor (“protective”) allele dose response (F(1,133)=3.7683, p=0.0543, R2=0.0508). The dosage ranges from 0 to 2, where 0 doesn't have a copy of the minor allele (T allele) and 2 as 2 copies of the minor allele. UPSIT = University of Pennsylvania Smell Identification Test. MAPT = microtubule-associated protein tau. UPSIT and the CDK5RAP2 rs10984186 polymorphisms. This genotyping was pyrosequenced separately from the gene chip, thus we have data on 141 individuals. CDK5 regulatory subunit-associated protein 2 (CDK5RAP2) binds to the CDK5 complex, which is involved in Tau phosphorylation. Participants included were 77 GG, 53 AG, 12 AA. A Kruskal-Wallis revealed there were significant group differences in OI (X2(2)= 6.13, p=0.0466, df=2, n=141). Linear regression suggested that the minor allele dose response predicts lower OI (F(2,139)=3.61, p=0.0297, R2=0.0494). The dosage ranges from 0 to 2, where 0 doesn't have a copy of the minor allele (A allele) and 2 as 2 copies of the minor allele. UPSIT = University of Pennsylvania Smell Identification Test. Manhattan plots of the genome-wide association study of olfactory identification. The blue line indicates the genome-wide suggestive threshold (1*10ˆ-5), The red line indicates the genome-wide significance threshold (5*10ˆ-8). There were 26 SNPs that reached the genome-wide significance threshold. Note that 4 SNPs on chromosome 3 and 14 had a P<5*10ˆ-12. Manhattan plots of the genome-wide association study of olfactory identification after adjusting for age, gender, and APOE 4. The blue line indicates the genome-wide suggestive threshold (1*10ˆ-5), The red line indicates the genome-wide significance threshold (5*10ˆ-8). We found that 22 SNPs survived adjustment for age, gender, and APOE 4 carrier status and identified 9 new SPNs for a total of 31 SPNs. Note that 4 SNPs on chromosome 3 and 14 conserved a P<5*10ˆ-12 after adjusting for covariates.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.