APOE Isoform-Dependent Self-Association Measured by a Split-Luciferase Complementation Assay: Differential Effects of Disease-Risk and Protective Variants

Apolipoprotein E (ApoE) is the principal lipid transport protein in the central nervous system and the strongest genetic modifier of late-onset Alzheimers disease (AD) risk. The three common isoforms, ApoE2, ApoE3, and ApoE4, differ in their propensity to self-associate, with ApoE4 forming oligomers more readily than ApoE3 or ApoE2. This enhanced self-association is proposed to reduce the pool of lipid-competent monomeric ApoE4 available for cholesterol transport and amyloid-{beta} clearance, contributing to AD pathogenesis. Here we describe a quantitative, cell-based split-luciferase complementation assay for ApoE self-association using the NanoBiT system, in which SmBiT- and LgBiT-tagged ApoE produced by HEK293 cells are combined and luminescence is measured. ApoE4 shows significantly enhanced self-association relative to ApoE3, while ApoE2 is no different from ApoE3. Testing a panel of naturally occurring and engineered variants demonstrates that the C-terminal self-association interface is the primary determinant of isoform-specific differences: two APOE {varepsilon}3-backbone C-terminal variants, Jacksonville (V236E) and W276C, both reduce self-association below ApoE3 levels, while the APOE {varepsilon}4-backbone protective variant R251G and the engineered domain-interaction probe R61T both reduce ApoE4 self-association to the level of ApoE3. In contrast, the Christchurch variant (R136S), the African-ancestry risk variant R145C, and the Admixed American risk variant R189C do not alter self-association. These findings establish a sensitive cell-based assay for ApoE self-association and highlight the C-terminal domain as a potential therapeutic target for normalizing ApoE4 function.