Browsing by Author "MODEL-AD consortium"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Characterization of the A1527G variant of ABCA7 in an animal model for late‐onset Alzheimer’s disease
    (Wiley, 2025-01-03) Bernabe, Cristian S.; Kotredes, Kevin P.; Pandey, Ravi S.; Carter, Gregory W.; Sasner, Michael; Oblak, Adrian L.; Howell, Gareth R.; Lamb, Bruce T.; Territo, Paul R.; MODEL-AD consortium; Medicine, School of Medicine
    Background: Genome‐wide association studies (GWAS) identified the ATP binding cassette subfamily A member 7 (ABCA7) gene as increasing risk for Alzheimer’s disease (AD). ABC proteins transport various molecules across extra and intra‐cellular membranes. ABCA7 is part of the ABC1 subfamily and is expressed in brain cells including neurons, astrocytes, microglia, endothelial cells and pericytes. However, the mechanisms by which variations in ABCA7 increase risk for AD are not known. Method: The IU/JAX/PITT MODEL‐AD Center identified the A1527G variant in ABCA7 (ABCA7*A1527G) as a putative LOAD risk factor. CRISPR/CAS9 was first used to introduce Abca7*A1527G variant to B6.APOE4.Trem2*R47H (LOAD1) mice to assess the transcriptional profiling on brain hemispheres from different ages. The Abca7*A1527G was then incorporated into B6.APOE4.Trem2*R47H.hAb (LOAD2) mice to further evaluate its contribution to LOAD. Female and male LOAD2.Abca7*A1527G and LOAD2 mice were characterized at 4, 12, and 24 months using the following phenotyping pipeline: behavior, PET/CT, multi‐omics, fluid biomarkers, electrophysiology, cognition, and neuropathology. Result: Brain transcriptional profiling showed that Abca7*A1527G induced changes in gene expression that are similar to some of those observed in human AD (e.g., granulocyte/neutrophil migration, and insulin receptor signaling). LOAD2.Abca7*A1527G showed no aging cognitive deficit but did show significant sex‐ and region‐dependent increases in brain glycolysis paralleled by reduced tissue perfusion yielding progressive age‐related uncoupled phenotypes between 4‐12 and 4‐24 months. While multi‐resolution consensus clustering of regional covariance matrices revealed an increase in cluster number and organization in LOAD2.Abca7*A1527G over LOAD2 for both sexes at 4 months, the cluster number and complexity were reduced by 24 months. Importantly, LOAD2.Abca7*A1527G, but not LOAD2, displayed a similar age‐dependent reduction in cluster number for both sexes. Consistent with the uncoupled phenotype, IL6, IL10, and TNFα were elevated in plasma with genotype, but were not age dependent. Conversely, brain levels of IL4, IL12, TNFα, and CXCL1 were decreased, whereas IL2 and IL10 were elevated in LOAD2.Abca7*A1527G relative to LOAD2. Lastly, assessment of plasma levels of Ab40‐Ab42 revealed an age‐dependent increase in both genotypes. Conclusion: Data collected to date support a model whereby variations in ABCA7 exert risk for AD through interactions between cerebrovasculature, microglia, and peripheral immune cells.
  • Thumbnail Image
    Item
    LOAD1 and LOAD2: Longitudinal characterization of mouse models carrying human‐relevant risk factors for late‐onset Alzheimer’s disease
    (Wiley, 2025-01-03) Bernabe, Cristian S.; Kotredes, Kevin P.; Pandey, Ravi S.; Carter, Gregory W.; Sasner, Michael; Oblak, Adrian L.; Howell, Gareth R.; Lamb, Bruce T.; Territo, Paul R.; MODEL-AD consortium; Medicine, School of Medicine
    Background: Alzheimer’s disease (AD) is the most common form of dementia, yet the effectiveness of disease‐modifying interventions is inconclusive. Although exceptional progress in our understanding of AD neuropathology has been made via transgenic mouse models bearing familial mutations, they often fail to recapitulate the disease progression of late‐onset AD (LOAD). To address this, MODEL‐AD has developed LOAD1 and LOAD2 mouse models which carry the most common human‐relevant risk factors for AD. In‐depth, longitudinal characterization through aging will reveal useful insights to develop novel treatments for LOAD. Method: APOEε4 and Trem2*R47H, two risk factors for LOAD, were incorporated into C57BL/6J mice to produce the double homozygous LOAD1 model, whereas LOAD2 also contains humanized amyloid‐beta (Aβ) yielding a triple homozygous model. Cohorts of LOAD1 and LOAD2 mice were aged on multiple sites to 4‐, 12‐, 18‐, and 24‐month timepoints and both sexes were characterized using behavior, PET/CT, cytokines/Aβ40‐42 immunoassays, and astrocyte and microglia immunohistochemistry. Result: Although aging LOAD1 and LOAD2 mice did not display significant cognitive deficits, there was a genotype‐dependent increase of plasma levels of Ab40‐Ab42. Both sexes across genotypes showed significant region‐dependent increases in brain glycolysis and tissue perfusion between 4 and 24 months. Consistent with the increased brain metabolism and perfusion neurovascular coupling phenotypes, immunopathology analyses revealed an age‐dependent increased number of astrocytes across genotypes that was restricted to cortical regions. Similarly, the total number of activated microglia was slightly elevated in cortical regions of aging mice, even though the results were only marginally significant. Lastly, aged LOAD1 mice displayed increased brain levels of the pro‐inflammatory cytokine IL‐12p70 when compared to LOAD2. Longitudinal analyses of brain and plasma cytokines are still in progress. Conclusion: LOAD1 and LOAD2 PET/CT analyses revealed phenotypes which are in line with imaging profiles of patients at prodromal stages of AD. Combined with the astrogliosis, these strains are promising venues that can be used to test early disease‐modifying therapeutic targets and can also serve as platform to incorporate additional human‐relevant AD risk factors.