Premium
F1‐02‐03: Metabolites Associated with Cognitive Function in the Rotterdam Study and Erasmus Rucphen Family Study
Author(s) -
Duijn Cornelia M.,
van der Lee Sven J.,
Ikram M. Arfan,
Hofman Albert,
Hankemeier Thomas,
Amin Najaf,
Demirkan Ayse
Publication year - 2016
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2016.06.277
Subject(s) - rotterdam study , cognition , medicine , population , dyslipidemia , triglyceride , endocrinology , psychology , cholesterol , obesity , cohort study , psychiatry , environmental health
and metabolites from several metabolic pathways including glycerophospholipid and sphingolipid pathways are perturbed in Alzheimer’s disease (AD) brains, CSF, and/or plasma [1-3] and may be related to disease severity. In addition to environment, genetic variation plays an important role in levels of metabolites [4-5]. Therefore, we investigated the role of genetic variation on serum metabolites associated with key AD relevant phenotypes in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort.Methods: Quality-controlled baseline serum metabolites and GWAS (Illumina Human610-Quad) were used in the study. Standard QC and imputation to 1000 Genome reference panel were performed for GWAS data.Metabolites were pre-adjusted for sex, age, and several medications if they were associated with levels of metabolites and selected when they were significantly associated with AD relevant phenotypes. Analysis was restricted to 673 non-Hispanic Caucasian ADNI participants (control1⁄4181, MCI1⁄4331, AD1⁄4161) after removing 69 participants with non-fasting serum samples. Analysis with 49 targeted metabolites was conducted using an additive genetic model including the first principal component to correct for potential population substructure. Results: Five analytes were significantly associated with SNPs on chromosome 1 or 11 after Bonferroni correction (p<1.02x10) and four analytes showed nominal genome-wide significant associations (p<5x10) (Figure 1). Genes associated with medium-chain acylcarnitines and phosphatidylcholine species play important roles in fatty acid beta oxidation and unsaturation of fatty acids. Two out of nine metabolites showed novel genetic associations. Conclusions: This study demonstrates the power of joint analysis combining genetics and metabolomics to investigate genetic risk factors for AD. Further analyses of metabolomics and genetics with contemporaneous neuroimaging endophenotypes is underway to investigate mechanisms underlying perturbed metabolites and the joint impact of genes and metabolites on brain structures and function. [1] Han et al. PLoS One (2011). [2] Kaddurah-Daouk et al. Translational psychiatry (2013). [3] Mapstone et al. Nature Med. (2014). [4] Kettunen et al. Nature Genetics (2012) [5] Shin et al. Nature Genet. (2014).