Novel mechanism underlying the APOE ε2 protective effect for Alzheimer disease implicated by integrative genome and transcriptome analysis

• Published in: Alzheimer's & dementia Vol. 16
• Main Authors: Panitch, Rebecca, Zhu, Congcong, Hu, Junming, Chung, Jaeyoon, Stein, Thor D., Farrer, Lindsay A., Jun, Gyungah R.
• Format: Journal Article
• Language: English
• Published: 01.12.2020
• ISSN: 1552-5260; 1552-5279
• DOI: 10.1002/alz.040065

Background The APOE ε4 allele is the strongest genetic risk factor for late onset Alzheimer disease (AD) and the ε2 allele is known to mitigate risk of AD. While potential mechanisms accounting for AD risk associated with ε4 have been extensively studied, the mechanism underlying the protective effect of ε2 is largely unknown. Methods We evaluated transcriptome‐wide (TW) differential gene expression (DGE) analysis in autopsied brains from 568 participants (339 AD, 229 controls) of the Religious Orders Study and Memory and Aging Project (ROSMAP), 162 subjects (82 AD, 80 controls) in the Mayo Clinic Study of Aging (MAYO), and 193 participants combined from the Framingham Heart Study and Boston University Alzheimer’s Disease Center (FHS/BUADC:64 AD, 129 controls). Results were combined across datasets by meta‐analysis. Analyses were focused on APOE 2/3 (E23) persons including 42 AD cases and 72 controls. The TW significance level was set at p < 10−6. Co‐expressed gene‐networks were created separately in AD cases and controls using Weighted Co‐expression Network Analysis. To validate AD enriched networks, we conducted gene enrichment analysis using lists from our top ranked DEGs and AD genes from a recent genome‐wide association study by Kunkle et al. In addition, we tested association of expression profiles from the top‐ranked DEGs with neuropathological traits in ROSMAP and FHS/BUADC. Results C4A (P = 1.9 × 10−5), GFAP (P = 3.8 × 10−5), and C4B (P = 4.8 × 10−5) were the most significant DEGs in the E23 group and TW significant in the total sample. The top‐ranked co‐expressed gene‐network in the E23 AD cases contained complement pathway genes such as C4A, C4B, and CR1, with significant enrichment for DEGs from the E23 subgroup (P = 0.05), DEGs from the total sample (P = 4.3 × 10−32), and known AD risk genes (P = 1.2 × 10−5). In FHS/BUADC, C4A expression was significantly associated with increased pTau181/tTau ratio (P = 0.01) and pTau231/tTau ratio (P = 1.0 × 10−5). Conclusion The complement pathway was previously implicated in AD, but to our knowledge this is the first biological link between APOE ε2 and the complement pathway. Together with a recent study showing direct binding between ApoE and C1q, our findings suggest that further studies of the interaction of APOE ε2 with complement pathway genes may lead to novel therapeutic targets for AD.