PROJECT SUMMARY
Alzheimer’s disease is the most common cause of dementia in the elderly, but there are a number of other
related dementias that exhibit substantial overlap in the behavioral, cognitive, and neuropathological
manifestations of the disease. In fact, the majority of dementia cases likely arise from the co-occurrence of one
or more of these AD and AD-related pathologies, with very few individuals exhibiting ‘pure’ Alzheimer’s
pathology (e.g., only amyloid plaques). This complexity makes diagnosis and therapeutic development
challenging, a problem exacerbated by a paucity of accurate animal models for ADRD that faithfully
recapitulate the full spectrum of the molecular, cellular, cognitive, and behavioral pathologies of these
dementias. In response to PAR-19-167, we will create a panel of genetically diverse knock-in mice harboring
known mutations associated with AD and several related dementias using precise genomic editing to ensure
biologically-relevant gene expression patterns and levels. In Aim 1, we will use CRISPR/Cas9 to create mice
carrying combinations of disease-causing mutations in App, Psen1, Mapt, Tardbp, and Snca to produce a set
of ‘core’ strains we expect to better capture the complexity of ADRD. To capture the role of genetic background
in disease risk, we will then cross these ‘core’ mice to four genetic backgrounds known to promote
susceptibility or resilience of ADRD (DBA/2J, FVB/NJ, WSB/EiJ, and C57Bl/6J). We will then leverage our
expertise in high-throughput mouse neurobehavioral phenotyping to screen 16 new ADRD strains to identify
the lines that best model ADRD. In Aim 2, we will use our deep phenotyping pipeline to fully characterize our
top strains across the entire spectrum of ADRD-related symptoms, including both cognitive and non-cognitive
domains. We will also use high-field MRI, histopathological measurements, and molecular phenotypes to
assess effects on brain structure, extent of neuropathologies, and impact on gene networks and pathways
associated with disease. Finally, in Aim 3, we will validate our new models for use in basic science and
preclinical studies by determining concordance between mouse and human data and use network modeling
approaches to identify early drivers of disease that predict late-stage outcomes in humans. This project will
produce much-needed new models for AD and related dementias that will greatly enhance our understanding
of the pathological mechanisms underlying these diseases. Finally, all of the models produced here will be
distributed to the community via the JAX Repository. We will also make all of the phenotyping data publicly
available using resources such as Mouse Phenome Database, GeneWeaver, and Synapse.
