Genome-wide scan of Alzheimer disease cohort identifies genetic loci associated with human brain metabolite levels.

Abstract

Although metabolome-wide association study (MWAS) has been performed in a wide range of tissue types, such as serum, plasma, urine, saliva, and cerebrospinal fluid (CSF), it has not been conducted on any brain tissue. Here we seek to expand our knowledge on the genetic influence of brain metabolism and on the contribution of metabolism abnormality to AD with the metabolomics approach. We aim to identify metabolite quantitative trait loci (metabQTLs) with the largest AD brain cohort available. We performed the first brain MWAS with 460 parietal cortex brains from the Knight-ADRC. 880 metabolites were measured by the powerful non-targeted Metabolon platform (HD4). European individuals were selected, and the analyses were adjusted for age at death, sex, genotype array methods, and genetic components. We identified 31 locus-metabolite associations in 30 metabolites with genome-wide significance. 7 associations were significant after study-wide Bonferroni correction. We are performing functional characterization and replication leveraging datasets from ROSMAP brain cohort, WADRC CSF cohort, and other tissues' cohorts. The strongest locus-metabolite association (p=5.2e-104), found in N6-methyllysine in the discovery phase, was first identified in CSF MWAS (Panyard et al., biorxiv, 2020). The first brain metabolome-wide association study discovered the importance of genetic influence on brain metabolite levels. Following expectation, brain metabQTLs can be replicated in the CSF study, implicating that brain and CSF share genetic features in modulating metabolism.