MODEL-AD: The Disease Modeling Project
Poster Jan 11, 2018
M.Sasner1, A. Oblak2, H. Williams1, G. Howell1, B.T. Lamb2 and the MODEL-AD consortium
The Alzheimer’s Disease (AD) patient population consists almost entirely (~98%) of the late-onset form of AD (LOAD); however, most mouse models used to study AD are based on familial AD mutations in APP, PSEN1 or PSEN2. The Model Organism Development and Evaluation for Late-onset AD (MODEL-AD) Center was created to develop, characterize, and distribute more precise preclinical models for AD.
The Disease Modeling Project (DMP) of MODEL-AD will use CRISPR genome editing to generate at least 40 novel mouse models that carry various combinations of human risk alleles for LOAD that have been identified by the Bioinformatics and Data Management Core (BDMC). In the early years of the Center, we will prioritize understanding known GWAS variants (ABCA7 and CR1) as well as novel variants identified by BDMC analyses of the AD Neuroimaging Initiative dataset (IL1RAP).
We will utilize a two-phase screening strategy. At least 24 models will undergo high capacity screening at 12 months of age, which includes functional assays and AD-related pathology. The most promising models will advance to the deep phenotyping phase, which will occur independently at Indiana University and The Jackson Laboratory to ensure reproducibility of data. In the deep phenotype screening, cohorts of mice will be evaluated at three time points (2, 6, and 12 months). We will prioritize clinically relevant endpoints including in vivo imaging, blood and tissue biomarkers, and genomics, and compare these to more traditionally used endpoints such as battery of functional assays. All data will be provided to the BDMC to determine the predictive potential for each mouse model, which will then be prioritized and selected for the preclinical testing core (PTC). To test our pipeline, we will deep phenotype models currently used for studying familial AD (APP/PS1, 5XFAD, hTau). In addition, we will also deep phenotype a new model of LOAD that we have created. This model carries the two greatest genetic risk factors for LOAD, APOE4 and the R47H variant of TREM2. Our strategy closely integrates human and mouse data, with the aim that these new AD models will show a high degree of clinical translatability for preclinical testing of new therapeutic targets.
All models created will be made available at the earliest opportunity through the JAX AD Mouse Mutant Resource. All protocols and data will be made available via the SAGE Synapse portal (www.synapse.org); we anticipate that we will be able to provide not only improved models, but also validation protocols and data so that the research community can efficiently adopt these new models. For more information, see model-ad.org.