Mutations in DDX3X are associated with autism spectrum disorder, brain malformations, and epilepsy, and account for up to 3% of cases of females with unexplained intellectual disability. However, little is currently known about the molecular mechanism linking DDX3X mutations to neurodevelopmental disease. DDX3X encodes an RNA helicase of the DEAD-box protein family and has been implicated in many aspects of RNA metabolism, yet we still lack a mechanistic understanding of DDX3X’s role in these processes, as well as how patient mutations in DDX3X affect RNA metabolism.

To address this gap in knowledge of understanding of how DDX3X mutations perturb cellular function and contribute to neurodevelopmental disease, this study investigated the mechanism of how DDX3X regulates its target transcripts at the level of mRNA nuclear export, and how pathogenic mutations in DDX3X alter these processes. In addition to DDX3X’s role in translation, DDX3X has been implicated in nuclear mRNA processing, and our preliminary data show that the genes in the mRNA export pathway are genetic interaction partners of DDX3X. However, we still do not understand what the function of DDX3X is in the nucleus or mRNA export. This study aimed to gain a greater understanding the precise mechanism of how DDX3X regulates RNA metabolism both at the level of translation initiation as well as mRNA export. This knowledge is critical for advancing our knowledge of RNA metabolism, as well as understanding and developing treatments for patients with DDX3X syndrome.

Agenesis of the corpus callosum (ACC) is one of the of the most common defects of the central nervous system with an incidence of 1/20000-30000 live births, putting it behind only spinal cord defects in terms of prevalence. While many candidate genes have been identified in patients with ACC, only 30% of all cases of ACC with a suspected genetic cause have an identified candidate gene. Patients with ACC rarely have ACC in isolation and usually have other neurologic (epilepsy, intellectual disability, cortical malformations), and neurodevelopmental (autism spectrum disorder, mood disorders) problems as well. Genes implicated in agenesis of the corpus callosum vary wildly from genes implicated in neurogenesis(PAX6, DISC1), axonal targeting and interaction(L1CAM), neuronal specification(SATB2), and neuronal survival. Identifying and characterizing potential genes in ACC gives us insight into neurodevelopment as a whole. My thesis work focused on two genes implicated in agenesis of the corpus callosum whose role in brain development was heretofore unknown, DEAD-box Helicase 3 X-linked (DDX3X), an RNA helicase, and C12ORF57 which we have tentatively named Camkinin a gene which had no previously known function. My work has shown that a subset of missense mutations in DDX3X completely inhibit its RNA helicase activity, which correlates with striking brain anatomy changes and more severe clinical outcomes. I have also reported the first evidence that demonstrates a likely function for the novel protein Camkinin, suggesting that it regulates the kinase CamKIV and through that controls synaptic scaling in excitatory neurons. Thus, through the two arms of my research we were able to further elucidate the brain specific function of these proteins and discover potential therapeutic avenues for patients.

Parkinson’s Disease (PD) is a neurodegenerative movement disorder affecting approximately 1% of the population over 60, with no existing treatments that slow, stop, prevent, or reverse neuron loss. Missense mutations in Leucine-Rich Repeat Kinase 2 (LRRK2) are the most common identified cause of PD; however, the disease-driving and normal cellular functions of LRRK2 are not fully defined. We used two screening approaches to gain insight into LRRK2 function and regulation: a proteomic screen to identify novel LRRK2 interactors, and a CRISPRi screen to identify modifiers of LRRK2 turnover. Our proteomic screen generated a quantitative global interactome of full-length LRRK2 that identified 48 novel interacting proteins, including Tripartite Motif-Containing Protein 1 (TRIM1), a microtubule-associated E3 ubiquitin ligase. We find that TRIM1 binds an unstructured regulatory domain on LRRK2 to recruit it to the microtubule. Furthermore, TRIM1 ubiquitinates LRRK2, leading to its proteasomal degradation. Expression of TRIM1 also modulates LRRK2 kinase activity and cytotoxicity: TRIM1 inhibits Rab29-mediated LRRK2 kinase activation and rescues the neurite outgrowth deficit caused by the most common PD-driving mutation, LRRK2 G2019S. In our second screen, we sought to identify additional novel regulators of LRRK2 turnover, focusing on proteasomal degradation caused by pharmacological LRRK2 kinase inhibition. LRRK2 kinase inhibitors are a promising therapeutic strategy for PD, but also lead to ubiquitination and degradation of LRRK2 by unknown ubiquitin ligases, which appears to cause pulmonary toxicity in non-human primates. We identified two additional E3 ubiquitin ligases that target LRRK2 for degradation downstream of kinase inhibition. Together, these findings identify LRRK2 localization, ubiquitination, and degradation as important regulatory events with relevance for kinase activity, cytotoxicity, and pharmacological intervention in PD.

Astrocytes perform critical homeostatic functions in the central nervous system (CNS). In CNS injury or disease, astrocytes respond to pathophysiological perturbations by becoming “reactive”, which is defined as the adoption of context-specific cell states and associated alterations in morphology and function. Frequently, inflammatory processes play an important role in the pathophysiology of CNS injuries and diseases. As an immunocompetent CNS cell type, astrocytes actively participate in inflammatory signaling cascades involving interaction with both microglia as well as infiltrating peripheral immune cells. Proinflammatory cytokines, such as IL-1a, TNF, and C1q, induce inflammatory reactive astrocytes that lose homeostatic functions while concurrently releasing factors that are potentially harmful in specific contexts. Given that inflammatory astrocyte reactivity has been implicated in numerous neurodegenerative and neuroinflammatory diseases, in addition to being associated with normal aging, it is an attractive target for therapeutic development. However, the cellular pathways that control inflammatory astrocyte reactivity are still not fully understood.

In Chapter 1 of this dissertation, we broadly introduce astrocyte reactivity as a concept and the existing literature on inflammatory astrocyte reactivity.

In Chapter 2, we set out to characterize astrocyte reactivity in the context of Alzheimer’s disease (AD) using single-nucleus RNA-sequencing of post-mortem AD brain tissue. In addition to discovering that reactive astrocytes in AD downregulated homeostatic pathways, we also found that a specific subpopulation of excitatory neurons were selective vulnerable in AD, which we were able to validate using immunostaining in an independent cohort of cases and controls.

In Chapter 3, we developed a scalable method to generate hiPSC-derived astrocytes that allowed us to harness the power of pooled CRISPRi screening to systematically identify cellular pathways controlling inflammatory astrocyte reactivity. Following up on the top hits from CRISPRi screens with single-cell transcriptomics, we found that autocrine-paracrine IL-6 and interferon signaling drove two distinct inflammatory reactive states that were promoted by and inhibited by STAT3, respectively. Furthermore, we found that the inflammatory reactive states we identified corresponded to those observed in other experimental contexts, both in vitro and in vivo. Lastly, we also performed immunostaining supporting the existence of these inflammatory reactive states in human brains in the context of Alzheimer’s disease and hypoxic-ischemic encephalopathy.

In Chapter 4, we focused on how autophagic pathways are rewired in inflammatory reactive astrocytes and how this contributes to the neurotoxic activity of inflammatory reactive astrocytes. We found that lysosomes are remodeled and alkalinized in inflammatory reactive astrocytes, and that lysosome exocytosis drives astrocyte-mediated neurotoxicity. Through CRISPRi screening, we uncovered mTOR as a regulator of neurotoxic lysosome exocytosis. These results pinpoint lysosome remodeling and exocytosis in inflammatory reactive astrocytes as a potential therapeutic target.

We believe these studies contribute to our understanding of inflammatory astrocyte reactivity and may inform the development of therapeutics that modulate specific aspects of astrocyte reactivity for the treatment of neurodegenerative and neuroinflammatory diseases.

Calcium homeostasis is essential to cardiac and skeletal muscle physiology, where contractile function requires tight but dynamic control of calcium levels across different cellular compartments. The major calcium storage protein in muscle tissues is calsequestrin, a highly acidic protein responsible for buffering up to 50 % of total sarcoplasmic reticulum (SR) calcium. Mutations in cardiac calsequestrin cause a highly lethal familial arrhythmia,

catecholaminergic polymorphic ventricular tachycardia (CPVT), while mutations in skeletal muscle calsequestrin have been linked to myopathies. Calsequestrin’s high density calcium storage is facilitated by homomultimerization of the protein into filaments, but a compelling atomic-resolution structure of a calsequestrin filament is lacking. This gap in knowledge has limited our understanding of calsequestrin biochemistry, SR calcium storage, and molecular mechanisms of calseqestrin-associated diseases. We report here a crystal structure of a cardiac calsequestrin filament with supporting mutation analysis by an in vitro filamentation assay. We also report and characterize a novel disease-associated mutation, S173I, which localizes to the structure’s filament-forming interface. In addition, we show that a previously reported dominant calsequestrin-associated CPVT mutation, K180R, maps to the same multimerization surface. Both mutations disrupt filamentation in vitro, suggesting a model where dominant disease arises from mutations that disrupt multimer formation. Finally, a ytterbium-derivatized structure pinpoints multiple credible calcium sites at filament-forming interfaces, explaining the atomic basis of calsequestrin filamentation in the presence of calcium. This work advances our basic understanding of calsequestrin biochemistry and also provides a unifying structure-function molecular mechanism by which dominant-acting calsequestrin mutations provoke lethal arrhythmias.

Organisms must learn from their environment and keep these memories for the long term to ensure their survival. If an adverse event is not remembered, the animal may not outlive it a second time. Thus, understanding how an adverse stimulus is encoded is especially important. In this dissertation, we find that C. elegans have a 2-hour critical period for memory consolidation, where sleep and the OSM-9 TRPV-like cation channel are required to successfully form a long-lasting memory. Specifically, OSM-9 may function in integrating old and new information associated to starvation. In addition, we find a new behavioral marker for swimming that is maintained post swim-based repetitive olfactory conditioning. The neural correlates of this proprioceptive memory can be observed in animals up to 2 hours post-conditioning and is formed independently of aversive olfactory memory. We found that animals integrate past and present environmental stimuli to form a new motor plan, which biologically functions to accelerate crawl-to-swim transition. Neural correlates of starvation associated changes in reversal motor pattern also show a change in conditioned animals in comparison to naïve animals, however these changes are not maintained. This specificity in gait selection and procedural memory can be further probed to understand how different types of experiences are organized in the brain.