Multi-scale MRI-based Diffeomorphometry of Pathology and Molecular Signatures Associated with Entorhinal Cortex Atrophy in Alzheimer's Disease

Johns Hopkins University, Baltimore MD

Investigators

Abstract

Understanding the unique vulnerability of certain cell populations in the brain to Alzheimer's disease (AD) - a debilitating, chronic disease affecting 6.2M Americans - has proven an elusive goal. Amyloid beta (Aβ) plaques and neurofibrillary tau tangle (NFT) pathology are hallmarks of AD that first become apparent in particularly vulnerable regions of the brain such as the entorhinal cortex (ERC) decades before symptom onset. Understanding how AD pathology affects specific brain regions, including a more complete understanding of the pathologic and molecular changes in selectively vulnerable areas is critical for the development and timing of potential AD treatments. The selective vulnerability of the medial temporal lobe (MTL) is well established, however, a more global picture of how tau pathology affects the mesial temporal lobe, its precise relationship to regional atrophy, and how this pathology is related to molecular changes associated with AD dementia remains unclear. To achieve this goal, we propose to integrate high-field post- mortem imaging, histological and spatial transcriptomic sections from the same individuals in a common coordinate system and learn how gene expression differences in the ERC are associated with early regional susceptibility to AD pathology. The central goal of the proposal is to link the spatial and temporal disease progression of MR markers at mm scale to the molecular substrates of tau pathology and spatial transcriptomics in layer 2 ERC. Uncovering the correlation of neuronal transcriptional changes to NFT pathology and to macroscopic shape change is necessary to understanding the biological link between these different scales and measures of AD pathology. This linkage has so far remained elusive because of the gaps in spatial scale between clinical MRI markers with histological and cellular assessments. We propose to close these gaps by introducing 100 μm high-field postmortem dense MRI reconstructions of the MTL allowing us to combine clinically-derived ERC population disease progression markers with histopathological and molecular profiling in these coordinates. In Aim 1 of this proposal, we will establish a protocol for postmortem high-field MRI and histological analyses of ERC to integrate microscale digital tau-pathology with Mai-Paxinos coordinates and population-based atlas mapping. We draw on ongoing comprehensive and longitudinal studies of aging including BIOCARD and ADRC studies which include cognitive measures, in vivo MRI scans, and postmortem protocols. In Aim 2, we will identify spatial gene expression signatures in layer 2 ERC neurons by anatomic location in Mai-Paxinos coordinates and quantify spatial transcriptional changes associated with regional AD pathology registered to MR measures and clinical outcomes. Integration of regional shape changes by MRI with pathologic and transcriptional information will shed light on the interaction of AD pathology with vulnerable ERC regions and underlying cell type and transcriptional signatures, and will provide the basis for the potential development of new MRI-based biomarkers of AD tau pathology which would be tested in future studies using in vivo clinical MRI.