Huntington’s disease (HD) is a devastating inherited neurodegenerative disease caused by a CAG repeat expansion in the Huntingtin (HTT) gene. The HD disease-causing mutation affects numerous cellular processes, contributing to overall dysregulated cellular homeostasis in the central nervous system, particularly in the striatum and cortex of the brain. While HD is predominately characterized by the loss of neurons, other non-neuronal cell types have been demonstrated to be significantly affected by mutant HTT; however, characterization of non-cell autonomous effects caused by HTT CAG repeat expansion in non-neuronal cell types was largely lacking prior to initiating this dissertation. Here, I describe my investigation into the cell-autonomous affects caused by mutant HTT through patient induced pluripotent stem cell (iPSC) modeling of two critical cellular regulators of brain homeostasis—brain microvascular endothelial cells (BMECs) and astrocytes. Assessing the dysregulation of HD BMECs and astrocytes provides a clearer understanding of disease pathogenesis and can provide additional knowledge to aid in the development of effective therapeutic interventions needed for HD patients.

This dissertation describes several transcriptomic analyses to provide mechanistic insights into pathways that underlie dysregulated HD iPSCs-derived BMECs (iBMECs) and iPSC-derived astrocytes (iAstros). First, bulk RNA-sequencing (RNA-seq) demonstrated aberrant WNT activation in HD iBMECs that when inhibited pharmacologically, prevented angiogenic deficits and improved paracellular barrier function in iBMECs. Secondly, single-nuclei RNA-seq highlighted inhibited synaptogenesis and glutamate receptor signaling states in both HD iAstros and HD mouse model astrocytes as well as a novel activated actin cytoskeletal signaling cell state unique in HD human astrocytes. HD iBMEC and HD astrocyte studies demonstrated dysfunctional properties that may be due to aberrant developmental trajectories. These data further support the hypothesis that mutant HTT induces dysregulated non-neuronal cell types that promote dysfunctional properties unique to each cell type, contributing to HD pathogenesis. Together, my analyses provide novel insights into molecular mechanisms that contribute to disease pathogenesis, and this knowledge can be used to inform future therapeutic investigations for HD.

Huntington’s disease (HD) is a neurodegenerative disorder caused by an expanded CAG repeat mutation in the Huntingtin (HTT) gene. The mutation impacts neuronal protein homeostasis and cortical/striatal circuitry. SUMOylation, a post-translational modification with broad cellular effects, directly modifies the Huntingtin protein (HTT), along with other key neuronal and synaptic proteins. Furthermore, disruptions in protein homeostasis lead to the accumulation of mutant Huntingtin (mHTT) protein and SUMOylated proteins, accompanied by dysregulated synaptic-related proteins and pathways. We investigated proteome-wide changes in striatal protein SUMOylation and SUMO-associated proteins in the context of HD using the R6/2 transgenic mouse model. We performed SUMO protein-enrichment from HD and non-transgenic (NT) tissue, followed by mass spectrometry, to assess changes within the SUMO-associated protein network in HD. I identified both known and potentially novel SUMOylated and/or SUMO-associated proteins, including those necessary for presynaptic function and cytomatrix at the active zone scaffolding, cytoskeleton organization, and glutamatergic signaling. A network-based analysis identified altered pathways in HD striata to include clathrin-mediated endocytosis signaling, synaptogenesis signaling, synaptic long-term potentiation, and SNARE signaling. A novel protein target of SUMOylation, protein interacting with C kinase (PICK1) was validated by an in vitro SUMOylation assay. Furthermore, the metabotropic glutamate receptor 7 (mGluR7) – a Group III mGluR and modulator of glutamatergic signaling (a core neurotransmission pathway disrupted in HD) – was SUMO-enriched. To evaluate functional measures of neuronal activity in HD cells in vitro, and to evaluate how modulation of SUMOylation via reduction of the E3 SUMO ligase, protein inhibitor of activated STAT1 (PIAS1) may alter mGluR localization, we conducted a receptor internalization assay in primary neuronal cultures. We found decreased mGluR7 internalization in R6/2 primary striatal neurons compared to NT cells, and siRNA-mediated knockdown of PIAS1 prevented this HD phenotype. In addition, microelectrode array analysis on primary neuronal cultures indicated early timepoint hyperactivity in several measurements of neuronal activity within HD cortical neurons, while later timepoints demonstrated deficits. HD phenotypes were rescued at select timepoints following knockdown of PIAS1. Taken together, my findings provide a mouse brain SUMO-ome resource, and demonstrate one of the first assessments of SUMO proteomics in the context of neurodegeneration. We find that significant alterations occur within the post-translational landscape and SUMO-protein interactions for synaptic proteins in HD.

Huntington’s disease (HD) is a devastating neurodegenerative disease that affects movement, psychiatric well-being and cognition (Group. et al., 1993; Huntington, 2003; Saudou & Humbert, 2016). Neuropathology includes morphological atrophy of the cortex and degeneration of the striatum, with selective loss of medium spiny neurons (MSNs). HD is caused by a CAG repeat expansion to ≥40 repeats in the first exon of the Huntingtin gene (HTT) which translates to form a mutant protein (mHTT) containing an expanded polyglutamine tract (Group. et al., 1993). Currently, there are no approved disease modifying treatments. Changes in RNA expression is an early, reproducible, and progressive feature of disease (Feyeux et al., 2012). However, to date underlying causal mechanisms of RNA dysregulation in HD remain elusive. HTT interactor studies have revealed interactions of HTT with RNA-binding proteins (RBPs) that regulate cellular processes such as RNA processing including RNA splicing, RNA localization, RNA Stability, and stress granule (SG) formation (Kaltenbach et al., 2007a; Ratovitski et al., 2012a; Shirasaki et al., 2012). A potential disruption of RBP interactions with HTT or dysregulation of RBP expression and localization due to the presence of the mutation may contribute to the transcriptional dysregulation observed in HD. To date, the field lacks a dedicated study into how HTT’s interaction with RBPs contributes to HD progression. My thesis focuses on understanding how HTT’s interaction with RBPs can regulate post-transcriptional RNA-processing. This work has targeted multiple HTT interactors and multiple RNA processes, in addition to developing a new molecular tool to study these HTT-dependent events. Together, this thesis established two mHTT dependent RBP phenotypes (Chapters 1 & 3), and developed a method to study HTT’s RBP interactors and their target RNA (Chapter 2).Chapter 1: Stress granules (SGs) have been established as a neuropathological feature of neurodegenerative diseases including Amyotrophic Lateral Sclerosis (ALS) and Alzheimer’s disease (AD) (Japtok et al., 2015; Wolozin & Ivanov, 2019a). SGs form during cellular stress and are a type of RNA granule comprised of RBPs and RNAs. The main SG protein G3BP1 has been shown to interact with HTT (Ratovitski et al., 2012b). Here, I helped study and characterize SGs in HD and our results identified increased and persistent SG formation in the HD mouse model R6/2 brain and human HD brain tissue. Chapter 2: Previous HTT interactor studies have been performed in yeast, mouse models, and non-neuronal cell types, and have not focused on RNA-binding proteins and the RNA transcripts they regulate. Here, I developed and optimized a system to identify RBP interactors of HTT and RNA that indirectly bind to HTT. This method uses mass-spectrometry combined with RNA-sequencing, and I have identified ~3000 HTT interactors along with differential binding preferences between HTT and mHTT in a susceptible neuronal population in HD. Chapter 3: My analysis of RNA sequencing data from the R6/2 model revealed upregulation of RNA-processing and identified the TAR DNA-binding protein (TARDBP/TDP-43) and METTL3 as contributing to the HD dysregulated transcriptome. Using RBP eCLIP-sequencing and m6A eCLIP-sequencing I have established a potential role for these two proteins in HD progression involving the regulation of alternative splicing and RNA stability. In addition, I identified the translocation of nuclear TDP-43 to the disease-associated cytoplasmic, phosphorylated TDP-43 in both the R6/2 HD mouse and HD patient brain tissues.

Loss of neurons is a key macroscopic feature that unites all neurodegenerative diseases including Huntington’s disease (HD) and Amyotrophic lateral sclerosis (ALS). These diseases affect distinct neuronal subtypes, e.g. striatum and cortex in HD and motor neurons in ALS. However, it is clear that common molecular mechanisms may underlie pathogenic features of each disease. Currently, there is no disease modifying treatment for HD or ALS. It is critical to elucidate the specific mechanisms leading to pathogenesis.

A single gene mutation, a CAG repeat expansion within the HD gene, causes HD. In contrast, 90% of all cases of ALS are sporadic and of the known gene mutations that cause familial versions of the disease, there are several contributing genes with the C9orf72 mutation accounting for the majority. Ultimately, in both diseases, numerous cellular functions are impacted including protein homeostasis, transcriptional regulation, cellular signaling, and bioenergetics; causing dysfunction and cell death.

A shared pathogenic feature which seems, at least in part, to be causative is the dysregulation of gene expression. In order to study this and other pathogenic mechanisms, we utilized patient-derived induced pluripotent stem cells (iPSCs) from both HD and ALS subjects and differentiated them into relevant cell types. For HD, mixed neural cells directed towards a striatal fate and brain endothelial cells (BECs) were generated, and motor neurons generated from ALS subjects. These cells were compared to lines generated from unaffected, healthy subjects for unbiased “omics” analysis to identify potential mechanisms underlying neurodegeneration.

My data suggest that mutant Huntingtin impairs neurodevelopmental pathways that could disrupt synaptic homeostasis and increase vulnerability to the expanded polyglutamine repeats over time. Additionally, I provide the first iPSC-derived barrier-forming EC model for a neurodegenerative disease to date, demonstrating specific barrier defects that may underlie crucial aspects of HD pathology.

RNA-seq analysis was also performed for iPSC-derived ALS motor neurons as a component of a large consortium effort to identify cell-based signatures. Alternative splicing and mislocalization of RNA binding proteins (RBPs) appears to also contribute to ALS pathogenesis, and here we identify RBPs that could contribute to this aberrant alternative splicing and dysregulated RNA biology.

Disruption of protein homeostasis, leading to accumulation of insoluble high molecular weight protein complexes containing the Huntingtin (HTT) protein and SUMOylated proteins, and transcriptional dysregulation are key features in Huntington’s disease (HD). Genetic modifiers contributing to HD age of onset have recently been identified and have critical roles in DNA damage repair (DDR) pathways. The mechanisms involved in DDR rely strongly on signaling cascades and post-translational modifications such as SUMO to maintain genomic integrity. Further, the Huntingtin (HTT) protein itself scaffolds DDR proteins. We previously showed that striatal reduction of the E3 SUMO ligase PIAS1 was neuroprotective and modulated disease associated pathologies including accumulation of mutant HTT in a mouse model of HD. However, the exact mechanistic contributions of PIAS1 towards HD pathogenesis have not yet been fully elucidated. To further evaluate PIAS1 function in the context of HD, knock-down was investigated in human patient medium spiny neurons differentiated from induced pluripotent stems cells and two disease mouse models. My findings suggest that PIAS1 functions as a key regulator of post-translational modification and protein homeostasis in HD neurons and mediates the functional activity of the transcription-coupled DNA damage repair complex in the striatum. Reduction of PIAS1 facilitated DNA repair, normalized aberrant transcriptional profiles related to synaptic function, and may stabilize the CAG-repeat within HTT. The results of this research provide the first mechanistic link between SUMOylation and DNA damage repair in the central nervous system. Specifically, they provide insight into how DNA damage repair pathways and post-translational modifications might contribute towards HD, and overall for targeting pathway mediators to restore homeostatic balance, with broad implications for HD and other neurodegenerative diseases.

Stress granules (SGs) are membraneless organelles that form in response to cellular stress. Chronic cellular stress associated with neurodegenerative disease results in the persistence of SG structures. Chronic expression of mutant huntingtin generates various forms of cellular stress, including activation of the unfolded protein response and oxidative stress. However, it has yet to be determined whether SGs are a feature of Huntington’s disease (HD) neuropathology. For my dissertation, I investigated localization and levels of the SG nucleating protein G3BP1 and found a significant increase in the density of G3BP1-positive SGs in the cortex and hippocampus of R6/2 transgenic mice, as well as in the superior frontal cortex of HD patient brains. I also examined the miRNA composition of extracellular vesicles (EVs) present in the cerebrospinal fluid (CSF) of HD patients and show that a subset of their target mRNAs is differentially expressed in the prefrontal cortex of HD patients. Of these targets, there is an enrichment of SG components, including G3BP1. These findings suggest that SG dynamics might play a role in the pathophysiology of HD. Lastly, because the majority of the CSF is produced by choroid plexus epithelial cells, and a significant portion of CSF EVs likely originate from CPECs, I initiated the generation induced pluripotent stem cell (iPSC)-derived CPECs from HD and control lines that will allow for future mechanistic studies of EV secretion. Together, my work provides evidence of a potential role for CSF EV miRNAs in the regulation of SGs, identifies specific miRNAs that modulate the SG component G3BP1, and demonstrates feasibility for the generation of an iPSC-derived CPEC model.

Loss of proteostasis and failure to effectively degrade cellular waste through the lysosome are common features among neurodegenerative diseases, including Alzheimer’s disease (AD) and Huntington’s disease (HD). Dysregulation of autophagy, the process of trafficking waste proteins to the lysosome for degradation, may lead to accumulation of toxic misfolded waste proteins and ultimately contribute to cell death. A single genetic mutation, a CAG repeat expansion within the HD gene, causes HD. Late-onset AD is sporadic and polygenic, but the APOE4 allele has emerged as the strongest genetic risk factor. Understanding what role wild-type HTT and APOE may play in autophagy pathways and how autophagy in turn contributes to clearance of these proteins may lead to a clearer understanding of disease pathogenesis and possibly opportunities for therapeutic intervention. This thesis describes my findings that APOE is degraded by autophagy, and that autophagic trafficking of APOE4 is increased in cellular models, causing lysosomal swelling and reduced overall autophagic flux. Knockout or knockdown of HTT modulates the contents of lysosomes, stimulating the lysosomal degradation of some autophagic substrates including APOE in hepatic cells. I investigated whether the combination of aberrant lysosomal trafficking of APOE4 with lysosomal trafficking induced by reduced HTT levels may cause toxicity in vivo. Preliminary studies of peripheral knockdown of HTT in APOE4-expressing mice have revealed toxicity in liver tissue, a major source of APOE expression. Ongoing and future studies of the interaction between these two neurodegenerative disease-associated proteins, and their function in and degradation by autophagy, may have implications for possible therapeutic strategies for HD involving peripheral total-HTT lowering, and for development of disease-modifying strategies for AD.