O4–01–01: Association of genetic variants with cerebrospinal fluid protein levels of ACE, MMP3 and other proteins and risk for Alzheimer's disease

This is the authors' abstract. We don't add key points for this paper.

Studying the genetic basis of quantitative traits related to human disease phenotypes has proven to be an effective approach to finding disease modifying genetic markers. We performed genome-wide association studies of 68 human disease-related proteins measured in the cerebrospinal fluid and evaluated significant findings for association with Alzheimer's disease (AD) risk. Phenotypes were measured using the Rules Based Medicine Human DiscoveryMAP in 293 samples from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and the 265 samples from the Knight Alzheimer's Disease Research Center at Washington University (WU). Standard QC including transformations to approximate normality, missingness and HWE filters and population stratification were assessed. Association between genetic markers and cerebrospinal fluid protein levels was assessed using linear regression after adjustment for age, gender and the principle components from EigenSoft analysis. Analysis was performed in each sample independently; results were combined in a metaanalysis using METAL. We applied additional filters to find a set of associated markers that are the highest priority for follow-up. SNPs from these loci were tested for association with AD status in data from the Alzheimer's Disease Genetics Consortium. 69 phenotypes were analyzed after QC filtering. We 12 identified candidate loci for 11 phenotypes including cis-effects for Angiotensin Converting Enzyme (rs4968782 P=3E-12), Interleukin 6 Receptor (rs61812598 P=5E-63) and Matrix Metalloproteinase 3 (rs679620 P=5E-44) and trans effects for CD-40 antigen (rs7250371 P=7E-09), Chemokine CC-4 (P=8E-22; P=5E-09), Macrophage Inflammatory Protein-1 beta (rs6808835 P=1E-13), Monocyte Chemotactic Protein 1 (rs2228467 P=3E-18), Osteopontin (rs11076196 P=2E-08), Prostatic Acid Phosphatase (rs3844501 P=2E-23), Sortilin (rs1708487 P=3E-09) and Trefoil Factor 3 (rs7280100 P=1E-08). Of these loci, SNPs in ACE (p=0.003), MMP3 (p=0.01) and PAP (0.03) showed association with AD risk. We have identified 12 loci with strong and replicable genetic associations with levels of 11 different proteins in the cerebrospinal fluid. Associations with AD status for ACE, MMP3 and PAP indicate a possible role for these markers in AD risk. These findings illustrate the utility of an endophenotype based approach to understanding disease risk and identify these genes and important therapeutic targets for modulating risk for AD and other disorders related to these proteins.