Defining Genetic Underpinnings of Cellular Diversity and Vulnerability in the Context of Parkinson's Disease

Abstract

Parkinson’s disease (PD) is a severe, progressive neurodegenerative movement disorder. Currently, there is no cure, nor effective treatments for prevention or modification of disease progression. Thus, novel approaches are required to understand etiological contributors to cell dysfunction and death in PD to enable rational drug design. Recently, huge advancements have been made in the technology involved in collection, storage, and processing of high-volume genetic information. Genome-wide association studies (GWAS) have been integral in identifying genomic loci with common, disease risk-associated variation. However, annotation and prioritization of GWAS-identified risk variants is far from straightforward, and the gap between the identification of risk-associated variants and defining etiological mechanisms of disease is vast. Bridging this gap will require a richer understanding of where risk-associated mutations initiate dysfunction: what cell types are most affected by disease-associated mutations? Development of single-cell sequencing techniques has revealed unexplored levels of cellular heterogeneity within tissues once thought to be genetically uniform. This dissertation presents several approaches for contextualizing risk-associated genetic variation within a framework of cellular heterogeneity to understand cell-type-specific etiology. First, we focus on a single cell type vulnerable to dysfunction and loss in PD: the midbrain dopaminergic neuron. Multiplex fluorescent in situ hybridization was used to neuroanatomically localize dopaminergic subpopulations markers identified in a mouse brain single-cell sequencing database and to assess nigrostriatal dopaminergic subpopulations in two rodent models of sporadic PD. Next, we identify patterns of risk-associated genes across multiple cell types by bioinformatically clustering transcripts by cell-type-specific expression. We follow up on these analyses with empirical validation in multiple model systems and use gene ontology and protein-protein-interaction modeling to further explore cell-type-specific etiology. Altogether, this dissertation investigates how to leverage single-cell transcriptomic data to best predict cell-type-specific vulnerability to PD pathogenesis across diverse cell types in the brain. Future studies can use this information to construct cellular models to identify convergent targets for drug discovery efforts and determine the cell types most likely affected based on an individual’s genetic makeup.