A proteome‐derived temporal model of neurodegeneration in older brains. (December 2021)

• Record Type: Journal Article
• Authors: Heath, Laura M; Gockley, Jake; Cary, Gregory A.; Wiley, Jesse C; Poehlman, William L.; Carter, Gregory W; Greenwood, Anna K; Mangravite, Lara M
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 3
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.057845
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: We previously developed a temporal model for unobserved molecular changes occurring during late‐onset Alzheimer's Disease (AD) for RNA‐Seq data. Here we apply this method to proteomics data from an AD case‐control cohort and show that the model's estimate of disease progression is associated with clinical and neuropathological endpoints, and identify changes in the proteome across stages of AD. Methods: Manifold learning defines an order across samples based on their similarity of expression. This ordering estimates pseudotime, an inference of molecular disease progression, quantitatively measured as the distance of each sample from the start of the inferred trajectory. We applied this approach to TMT‐proteomics data from human postmortem brain samples from the ROS/MAP study within the Accelerating Medicine Partnership‐AD consortia (AMP‐AD) (N=400). We examined associations between pseudotime and neuropathological and clinical endpoints. We performed differential expression analysis on tree branches to identify progression‐specific pathways, and used gene set enrichment analysis to identify significant GO terms from a set that have been curated into 15 distinct AD‐relevant biological domains. Results: Pseudotime estimates were significantly associated with LOAD status (females, p=3.67x10‐7; males, p=1.03x10‐3), such that "early" (low pseudotime) samples are enriched for controls, and "late" (high pseudotime) samples are enriched for cases. Pseudotime was associated with cognitive diagnosis (females, p=2.46x10‐5; males, p=1.14x10‐3), Braak stage (females, p=1.48x10‐5; males, p=1.52x10‐3), and CERAD neuritic plaque score (females, p=1.17x10‐11; males, p=1.32x10‐4). Proteins involved in immune response were upregulated, while mitochondrial metabolism was downregulated, consistently across pseudotime. We observed differences between branches in biological domains encompassing vascular function, structural stabilization, lipid metabolism, apoptosis regulation, and epigenetics. Synaptic dysfunction was downregulated during mid‐to‐late pseudotime in women, but not in men. Conclusions: This approach to identifying biological changes associated with AD progression has now been applied to two data modalities, and despite substantial differences in data used to order trajectories, pseudotime was associated with multiple measures of AD progression in both studies. We also uncovered proteome‐specific changes, providing opportunities to glean new insights about genetic drivers of AD across disease severity.