Spatial signatures of brain health and vulnerability in aging and Alzheimer's disease - PROJECT SUMMARY / ABSTRACT For decades, the analysis of Alzheimer’s brain pathology has remained, by and large, unchanged, with the detection of Aβ plaques and tau tangles representing the hallmarks features of Alzheimer’s disease (AD) that are intimately linked to cognitive decline. This traditional analysis of AD brain is very qualitative and limited by the slow and tedious “one-at-a-time” approach to detect individual markers of interest in the brain. This lack of innovation has posed major limitations in achieving more rapid, comprehensive, and quantitative views of dementia. Recently, a technology was developed called cytometry time-of-flight (CyTOF) imaging mass cytometry (IMC) and its potential for translational applications, especially neurodegenerative diseases, is remarkable. The reason for this is that we can now analyze all cell types and structures in a single experimental design. We can also quantify and mine these data to provide spatial insights into how the brain degenerates over time to establish a “brain signature” of aging and dementia. Such multiplexed analysis of a large panel of human brain samples will provide rapid quantitative insights that will inform our understanding of AD pathophysiology. In this study, we developed a “brain vulnerability and resilience” panel suitable for probing mouse and human brains, with particular emphasis on immune factors that have been implicated in AD. By probing their abundance, cell-type specific accumulation, and spatial proximity in relation to plaques and tangles, we will provide global insight into AD progression. We hypothesize that the ensemble of spatial information, gleaned from 45+ simultaneous markers of interest, will point to specific microglia subpopulations that drive or suppress brain pathology. In Aim 1, we will generate a comprehensive brain signature of aging and AD by employing IMC on a panel of human brains to derive in situ spatial information of brain health and resilience, providing a thorough assessment of disease trajectories. The inclusion of normal aged brains will also allow us to delineate the earliest stages of age-related brain changes, a timeframe that is more amenable to eventual therapeutic interventions. We will then validate the emergence of these spatial signatures using well characterized AD mouse models that show either slow- or fast-progression towards cognitive decline. In Aim 2, we will develop an automated bioinformatics approach for multi-sample IMC data to segment brain cell types and define their cellular microenvironments. We will also develop machine learning approaches to summarize the spatial patterns and predict clinical outcomes. This proposal is a highly innovative merging of unique experimental and computational strengths; it is significant since it represents a conceptual leap in how we image, visualize, and quantify brain data; and it has enormous clinical implications, since it could highlight multiplexed imaging coupled with predictive modeling to determine which patients transition from the normal to diseased state. Given that immune mediators often circulate peripherally, our study provides a foundation for the use of IMC to discover future biomarkers as readouts for immune dysfunction.
