APOE‐stratified genome‐wide association analysis identifies novel Alzheimer disease candidate risk loci for African Americans. (December 2021)

• Record Type: Journal Article
• Authors: Kunkle, Brian W.; Jean‐Francois, Melissa N.; Hamilton‐Nelson, Kara L.; Schmidt, Michael A.; Naj, Adam C; Martin, Eden R; Vance, Jeffery M.; Cuccaro, Michael L.; Rajabli, Farid; Jun, Gyungah R; Wang, Li‐San; Farrer, Lindsay A.; Haines, Jonathan L.; Byrd, Goldie S.; Schellenberg, Gerard D.; Mayeux, Richard; Beecham, Gary W.; Pericak‐Vance, Margaret A.; Reitz, Christiane
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 3
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.056383
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: We recently published a genome‐wide association study (GWAS) for Alzheimer disease (AD) in African Americans (AA; 2, 784 cases, 5, 222 cognitively normal individuals) (Kunkle et al. JAMA Neurol. 2021). We now present a GWAS meta‐analysis of this AA data stratified by APOE ε4‐genotype. Method: All datasets were imputed to the African Genome Resource panel (∼93 million variants) using the Sanger imputation server. Single‐variant association analysis was conducted in APOE ε4‐negative (1, 060 cases, 3, 247 cognitively normal individuals) and APOE ε4‐positive (1, 450 cases, 1, 808 cognitively normal individuals) groups separately, adjusting for age, sex and principal components. Individual datasets were analyzed applying logistic regression for case‐control datasets and general estimating equations for family‐based datasets. Within‐study results were meta‐analyzed using METAL. Result: Following quality control, ∼14.4 million variants in the APOE ε4‐positive group and ∼14.2 million variants in the APOE ε4‐negative group with minor allele frequency >=0.01 were analyzed. We identified seven novel candidate loci at P<10 ‐6 for AD in AA. These include signals in the APOE ε4‐negative analysis centered at 16q24.1 (P=3.9x10 ‐7 ), 2q22.2 (P=4.3x10 ‐7 ), 1p36.32 (P=5.8x10 ‐7 ), 2p23.1 (P=8.2x10 ‐7 ), 8q21.3 (P=9.4x10 ‐7 ), and 4p15.1 (P=9.4x10 ‐7 ), and one signal in the APOE ε4‐positive analysis centered at 7p15.3 (P=9.6x10 ‐7 ). A scan for eQTL's within these loci found an eQTL in the 7p15.3 locus for the gene DECR1, involved in reaction to oxidative stress, in brain cerebellum (P=1.8x10 ‐5 ; GTEx.v7). Conclusion: We identified several potential novel candidate loci for AD in AA at P<10 ‐6 in groups of individuals with and without APOE ε4 alleles. In addition to DECR1, these loci contain genes and functions relevant to AD including: FOXL1 (key regulatory molecule associated with transcriptional changes in AD) and an association (P=2x10 ‐6 ) with hippocampal volume in ADNI at 16q24.1, LRP1B (APP binding partner; decreases APP processing to Aβ) at 2q22.2, and MMEL1 (clears Aβ; expression and activity are altered in mild‐cognitive impairment) at 1p36.32. Confirmation of these loci in larger studies of AA is necessary and could provide novel therapeutic targets for AD.