Abstract The long-term goal of this project is to develop, characterize, and validate genetically modified marmoset models of sporadic Alzheimer's Disease (AD) that will serve as tools for investigating molecular and cellular disease mechanisms, and for identifying therapeutic targets. AD is the most common cause of dementia, with the majority of cases (~95%) appearing to be sporadic, which is caused by complex interactions between multiple gene variants and environmental factors. Numerous models of AD have been developed (e.g., mice); these models are primarily focused on familial forms of AD and have enabled significant progress toward understanding AD, but they fail to recapitulate the full spectrum of molecular, cellular, behavioral, and cognitive pathologies observed in AD and provide poor predictive value when trying to translate findings to human clinical trials. Several lines of evidence suggest that marmosets may effectively bridge the gap between mice and humans for both basic and translational neuroscience research. First, marmosets and humans have very similar brain structures, cognitive/social behavioral repertoires, metabolism, and immune functions. Second, compared to other primates, marmosets have short lifespan, small body size, and high reproductive power. Finally, gene editing tools are now available to generate various types of genetically modified marmosets. Apolipoprotein E (APOE) is the strongest risk factor for late-onset, sporadic AD and also increases the age- dependent risk of monogenic familial AD and incidence of AD in women. There are three APOE alleles in humans with the APOE*ε4 allele conferring increased risk and the APOE*ε2 allele conferring decreased risk relative to the common APOE*ε3 allele. APOE isoforms differentially modulate both amyloid-β (Aβ)-dependent and Aβ-independent pathways to affect brain homeostasis, including tau-mediated neurodegeneration, microglial inflammation, lipid transport, synaptic integrity, glucose metabolism and cerebrovascular function. These three APOE alleles differ with regard to cysteine (C) and arginine (R) amino acids at positions 112 and 158 (C112/C158 in APOE2; C112/R158 in APOE3; and R112/R158 in APOE4). Several lines of evidence demonstrate that R61 is critical for APOE4-mediated AD risk. Marmoset APOE (mAPOE) is encoded by a single allele and contains the equivalent of R112 and R158, but lacks the critical R61 equivalent (it contains T instead), suggesting that it behaves like human APOE3. To test this hypothesis, mAPOE T61R mutant protein was generated to test the effect on inflammatory responses induced by lipopolysaccharides in microglial cells. The mAPOE T61R variant resulted in a more robust response compared to wild type mAPOE. Similar results were found when human APOE4 was used (APOE4>APOE3). Taken together, these preliminary results indicate that mAPOE T61R is functionally equivalent to human APOE4. Further, several pairs of CRISPR gRNAs for generating an APOE null mutation have been identified. To investigate the role of APOE in sporadic AD in the marmoset, APOE null and T61R mutant marmosets will be generated and characterized.

Narrative The long-term goal of this project is to develop, characterize and validate genetically modified marmosets that exhibit brain pathology and/or clinical/behavioral signs compatible with sporadic Alzheimer's Disease (AD). These marmoset models are aimed at informing human AD across the full life span or throughout the disease- relevant stages and serve as tools to interrogate molecular disease mechanisms and identify therapeutic targets. Specifically, as APOE4 allele is the most important risk factor for sporadic AD, marmosets carrying a complete loss of APOE or a point mutation mimicking APOE4 will be generated.