UNDERSTANDING THE CELLULAR AND GENE-REGULATORY MECHANISMS UNDERLYING THE MESENCHYMAL TRANSITION OF EPENDYMOMA TUMOR CELLS USING OMICS APPROACHES
Degree type
Graduate group
Discipline
Medical Sciences
Biology
Subject
deconvolution
mesenchymal transition
microglia
omics
single-cell data
Funder
Grant number
License
Copyright date
Distributor
Related resources
Author
Contributor
Abstract
Childhood ependymoma is a cancer of the central nervous system with a chronic relapsing pattern. With little evidence that adjuvant chemotherapies extend patient survival, surgical resection followed by radiotherapy remains the best treatment for this disease. In children, the most common and aggressive ependymoma subgroup is posterior fossa ependymoma type A (PFA) which has a 10-year progression-free survival rate < 25% and 10-year overall survival rate of 55%. Single-cell transcriptomic analyses of these tumors have identified tumor-derived cell lineages that resemble developmental radial glia cells and a tumor cell population with a mesenchymal gene expression profile that is associated with poor prognosis. However, little is known about the cellular and molecular mechanisms that enable the initiation, maintenance, and differentiation of these tumor cells. In this dissertation, we use transcriptomic (Chapter 2) and chromatin accessibility (Chapter 3) data to identify the cell types, signaling pathways, transcription factors, and enhancers associated with PFA tumor cell dynamics. We find that tumor cells that recapitulate neurodevelopmental programs are driven by WNT/β-catenin signaling while the mesenchymal phenotype is associated with TGF-β and TNF-α/NFκB signaling and the activation of transcription factors like NFkB, AP-1, MAF/BACH, MYC, and HIF1A. We are also the first to use single-cell omics data to characterize the ependymal tumor microenvironment and identify pro-inflammatory microglia as an important mediator of the mesenchymal transition of ependymoma tumor cells. Building upon on the computational methods used to analyze these data, we developed ConDecon, a clustering-independent deconvolution method for inferring high-resolution cell abundances from bulk tissues using single-cell data as reference (Chapter 4). Applying ConDecon to the transcriptomic data of pediatric ependymal tumors from the posterior fossa, we find that as tumor cells transition into a mesenchymal phenotype, microglia gain a neurodegenerative inflammatory phenotype. Altogether, these investigations improve our understanding of the oncogenic mechanisms regulated pediatric posterior fossa ependymoma and identify promising targets for therapeutic intervention.