Dissertation/Thesis Abstract

The author has requested that access to this graduate work be delayed until 2019-05-19. After this date, this graduate work will be available on an open access basis.
Multiscale Modeling of Oligodendrocytes and Myelin in Alzheimer's Disease
by McKenzie, Andrew Thomas, Ph.D., Icahn School of Medicine at Mount Sinai, 2017, 263; 10284858
Abstract (Summary)

Several lines of evidence from neuroimaging, neuropathologic, and model system approaches suggest that the dysregulation of oligodendrocytes (OLs) and myelination may play a critical role in the pathogenesis of Alzheimer’s disease (AD). In this thesis, I utilized novel systems-wide approaches to test the hypothesis that OL dysregulation and dysmyelination makes a primary contribution to axonal injury and AD pathogenesis, as opposed to being a consequence of axon and neuronal degeneration. I first analyzed genetic and neuroimaging data, finding that regional measures of AD pathophysiology in late-life are inversely correlated with early-life myelination rates. This result suggests that slow myelination in development may causally contribute to AD-associated degeneration during aging, possibly consequent to overall decreased trophic and metabolic support over time. Because most of the published data was generated using postmortem brain tissue, with the inherent complexity of differences in cell-type composition, I built two critical tools to decouple cell type proportion changes from changes in the levels of individual molecules or transcripts. The first tool was used for analyzing and deconvoluting brain cell type gene expression signatures, including OLs and OL precursor cells (OPCs). Since it is also critical to identify differences in network-level interactions within OLs in AD, the second computation tool leverages permutation tests to uncover gene-gene and gene-trait differences in correlation. Both these tools were applied to the study of OL-associated gene modules, from both RNA and protein expression data, in AD. OL module genes were enriched in AD risk factor genes, and there was a loss of correlation between the OL module genes at the protein level. A knockout mouse model of a key driver in the OL module, CNP, mimicked myelin-associated gene dysregulation seen in the hippocampus of AD patients. Finally, relative cell type proportion differences and associated network-level changes were studied in AD and control brains. There was an alteration in the proportion of OL lineage cells in AD in the frontal white matter and temporal cortex grey matter. The structure of the network relative to the AD downregulated hub gene PSEN1 was recapitulated by gene expression dysregulation of mice with an AD-causative Psen1 M146V/WT mutation, which also led to a deficit in intermediate stage OL lineage cells. Overall, our multiscale, network-based approaches provide evidence for several genetic and gene expression pathways through which dys-homeostasis of OLs may contribute to AD pathogenesis.

Indexing (document details)
Advisor: Zhang, Bin, Casaccia, Patrizia
Commitee: Cai, Dongming, De Jager, Philip, Goate, Alison, Houten, Sander, Roussos, Panagiotis
School: Icahn School of Medicine at Mount Sinai
Department: Genetics and Genomic Sciences
School Location: United States -- New York
Source: DAI-B 78/10(E), Dissertation Abstracts International
Subjects: Neurosciences, Genetics, Aging
Keywords: Alzheimer's disease, Myelin, Network biology, Oligodendrocyte, RNAseq
Publication Number: 10284858
ISBN: 9781369795271