The association between macrocephaly and autism spectrum disorder (ASD) suggests that the mechanisms underlying excessive neural growth could contribute to the pathogenesis of the disorder. Neural progenitor cells (NPCs) derived from induced pluripotent stem cells of ASD individuals with early developmental brain enlargement are inherently more proliferative than control NPCs. Here, we demonstrate that NPCs derived from ASD patients with macrocephaly display an altered DNA replication program and increased DNA damage. When compared to the control NPCs, high throughput genome-wide translocation sequencing demonstrates that ASD-derived NPCs harbored elevated DNA double-strand breaks in replication stress-susceptible genes, some of which are associated with the pathogenesis of ASD. Our results identify a mechanism linking hyperproliferation of NPCs with the pathogenesis of ASD by disrupting long neural genes involved in cell-cell adhesion and migration.

Bipolar disorder (BD) is a neuropsychiatric disease characterized by intermittent episodes of mania and depression that detrimentally affects a person’s ability to carry out day-to-day tasks. Life expectancy is reduced by 10 to 15 years along with an increased suicide rate and comorbidities such as cardiovascular disease and diabetes. Studies have shown that synaptic plasticity and inflammation play a major role in the pathophysiology of bipolar disorder, but how they contribute is still unknown. Astrocytes, an important glial cell type in the brain, are significant in both synaptic plasticity and inflammation but are currently understudied in bipolar disorder. Therefore, we wanted to know the effect that astrocytes derived from induced pluripotent stem cells (iPSCs) from bipolar patients have on the synaptic activity of control neurons measured via multiple electrode array during inflammatory conditions. BD astrocytes showed increased secretion of IL-6, an inflammatory cytokine, when activated with IL-1b, a pro-inflammatory cytokine. Both co-culture with BD astrocytes and addition of conditioned media from BD astrocytes decreased action potentials in control neurons, which was rescued by adding an IL-6 blocking antibody. Our results suggest that increased IL-6 secretion in response to inflammatory stimuli in BD astrocytes may lead to altered neuronal activity in BD patients that could contribute to the bipolar phenotype. These results may further elucidate the etiology of BD and help scientists understand the role of astrocytes and inflammatory proteins in BD.

The individual unit of computation in the human brain is the neuron. To advance our understanding of the human brain, we need to generate relevant cellular models of human neurons in vitro. Direct conversion of human somatic fibroblasts into induced neurons (iNs) allows for the generation of functional neurons while bypassing any stem cell intermediary stages. Although iN technology has an enormous potential for modeling age-related diseases, as well as therapeutic approaches, the technology faces limitations due to variable conversion efficiencies and a lack of thorough understanding of the signaling pathways directing iN conversion. Here, I introduce a new all-in-one inducible lentiviral system that simplifies fibroblast transgenesis for the two pioneer transcription factors, Ngn2 and Ascl1, and a small molecule cocktail that markedly improves iN yields and sheds new insight into the mechanisms influencing direct iN conversion. Next, I will apply the iN protocol on a cohort of Alzheimer’s disease (AD) and healthy age matched control donor samples to model age-related molecular phenotypes in AD neurons. I previously found that AD brains have significantly higher proportions of neurons that express senescence markers, and their distribution indicates bystander effects. AD patient-derived iNs exhibit strong transcriptomic, epigenetic, and molecular biomarker signatures that illuminate a specific human neuronal senescence-like state. AD iN single-cell transcriptomics revealed that senescent neurons face oncogenic challenges, and metabolic dysfunction, and they display a proinflammatory signature. Integrative profiling of the inflammatory secretome of AD iNs and patient cerebral spinal fluid revealed a neuronal senescence-associated-secretory-phenotype, that can trigger astrogliosis in human astrocytes. Finally, I show that targeting senescence-like neurons with senotherapeutics could be a novel strategy for preventing or treating AD.

Autism spectrum disorder (ASD) is a multifactorial disease, suspected to originate during early fetal cortical development. Due to its symptomology and complex genetic etiology, developmental studies of autism largely rely on analysis of postmortem tissue. Recently, cortical organoids have impacted ASD research by providing highly organized tissue models that recapitulate early fetal human development. Here, forebrain cortical organoids (FBOs) derived from human induced pluripotent stem cells (iPSCs) are employed in a novel assembloid technique to model the basic organization principles of the cortical connectome. In vitro fusion of two FBOs results in a single assembloid unit that demonstrates key developmental processes: axon fasciculation and pathfinding. Characterization of the assembloid system was performed using immunohistochemistry and identified fascicules (“tracts”) which projected into the adjacent cortical unit and strongly correlated with an axonal identity. Additionally, electrophysiological recordings and correlative electron microscopy provide evidence of functional activity and reciprocal connectivity between assembloid “hemispheres”. Using the assembloid model, knock down of two high confidence ASD risk genes was assessed. Live cell imaging was used to capture short-term (seven days) and long-term changes (six weeks) in axon tracts. Findings from the project demonstrated that disruption of Contactin-associated protein-like 2 (CNTNAP2, Caspr2) or immunoglobulin superfamily containing leucine-rich repeat 2 (ISLR2, Linx) impact axon pathfinding and fasciculation. Knock down of CNTNAP2 resulted in absence of axon tracts during the early developmental timepoint, while knock down of ISLR2 resulted in a loss of fasciculation but not outgrowth, at both time points assayed. These findings elucidate the role of two key ASD risk genes during development and their putative impact on human cortical connectivity.

Dementia with Lewy bodies (DLB) is a neurodegenerative disease that shares clinical features with Alzheimer’s disease (dementia) and Parkinson’s disease (movement disorder). Diagnosis is confirmed at autopsy by the presence of Lewy bodies in the brain. These intracellular protein aggregates containing α-synuclein are thought to impair movement when in the brainstem, yet it remains unclear whether they impair memory when in the hippocampus. Genetics have also been implicated in DLB in that certain mutations increase the risk of developing the disease. However, the links between these genetic alterations and neuronal dysfunction are not fully established.

To address these issues, we pursued two complementary approaches: 1) we characterized hippocampal Lewy pathology burden and distribution in 95 autopsy-confirmed DLB cases, correlating these pathology data to results from memory testing; 2) we reprogrammed fibroblasts from clinically diagnosed DLB patients and relevant controls into neurons with a direct conversion method, then used them to investigate differential gene expression and resulting functional differences.

Lewy pathology in our postmortem DLB cases was predominant in two hippocampal-related subregions: the CA2 subfield and the entorhinal cortex (EC). Clinicopathological correlations with measures of verbal and visual memory supported a role for EC Lewy pathology, but not CA2, in causing memory dysfunction. Lewy pathology in CA1—the main output region for CA2—correlated best with results from memory testing despite a milder pathology, suggesting that CA1 may be more functionally relevant than CA2 in the context of memory impairment in DLB.

To probe the mechanisms leading to neuronal dysfunction in a dynamic system, we reprogrammed DLB patient fibroblasts to induced neurons (iNs). These iNs displayed multiple neuronal features, including the expression of specific markers. We conducted RNA-sequencing to compare patient and control lines, and found genes relating to synaptic function (including the gene for α-synuclein) and oxidative stress upregulated in the iNs, as opposed to the parent fibroblast lines which had very few differentially expressed genes. We confirmed the increase in α-synuclein at the protein level via staining, and mapped the “DLB signature” we identified to a pro-inflammatory state, offering potential new avenues for screening and intervention in the disease.

Alzheimer’s disease (AD) is a progressive neurodegenerative disease characterized by loss of neurons and synapses throughout the brain. Increased neuroinflammation has been observed in the brains of AD patients and is hypothesized to play a role in the pathology of the disease. Inflammation in the brain is thought to be primarily initiated by the non-neuronal cells of the brain, glia cells. One type of glia, astrocytes, are suspected to play a role in the onset and progression of AD through inflammation. Recent characterization of reactive astrocytes has identified two activated astrocyte states: a harmful, pro-inflammatory A1 state or a helpful, anti-inflammatory A2 state when exposed to CNS injury. In this thesis, I asked if A1 or A2 astrocyte states were different between AD and control astrocytes. Here, I analyzed RNA-sequencing data from AD astrocytes generated from patient-derived induced pluripotent stem cells (iPSCs) along with age-matched controls to identify the relative proportions of A1 and A2 states. Surprisingly, I found that AD astrocytes have an increase in A2-type, “helpful” astrocytes relative to age-matched cognitively normal controls. These results reflect the complexity of AD and suggest new ways of thinking about inflammation in AD.

The elegantly delineated anatomy of the hippocampus has long served as a substrate for theories of memory formation and recall. Principally, the Dentate Gyrus (DG) has been proposed to be important in the encoding and formation of new memories through a process known as pattern separation, while downstream CA3 has been proposed to play a role in memory recall through a process known as pattern completion. In vivo optical methods have provided access to study computation within the DG and the activity of dentate granule cells (DGCs) has been associated with hippocampus-dependent behavioral pattern separation and pattern completion. Increasing evidence supports the idea that a rare population of immature adult-born DGCs (abDGCs) 4-6 weeks of age support pattern-separation in the DG by enhancing feedback inhibition onto mature DGCs. These studies rely almost exclusively on recominant Adeno-associated Virus (rAAV) for transgene delivery, however the toxic effects of rAAV on this circuit were not been assessed.

Herein, we demonstrate that neural progenitor cells (NPCs) and immature DGCs within the adult murine hippocampus are particularly sensitive to rAAV induced cell-death. Cell loss is dose-dependent and nearly complete at experimentally relevant viral titers. rAAV induced cell-death is rapid and persistent, with loss of BrdU labeled cells within 18 hours post-injection and no evidence of recovery of adult neurogenesis when assessed at 3 months post-injection. The remaining mature DGCs appear hyperactive 4 weeks post-injection based on immediate early gene expression, consistent with previous studies investigating the effects of attenuating adult neurogenesis. This rAAV-induced toxicity is intrinsic to the rAAV viral vector. Finally, efficient transduction of the DG is achieved by injection of rAAV2-retro serotyped virus into CA3 and permits in vivo 2-photon calcium imaging of dentate activity while leaving adult neurogenesis intact. Using this method we demonstrate functional changes to mature DGC activity 2-weeks following abDGC ablation using rAAV. These findings expand on recent reports implicating rAAV linked toxicity in stem cells and other cell types and suggests future work using rAAV in the DG should be carefully evaluated.

Astrocytes are one of the primary cells in the central nervous system that regulate the inflammatory response. However, master regulatory proteins that mediate the astrocyte inflammatory response remain to be elucidated. To identify these regulator genes, we made two embryonic stem cell lines (ES) with a doxycycline inducible Cas9 through transduction into the AAVS1 (Adeno-Associated Virus Integration Site 1) safe harbor locus. We also developed an 8-day protocol for a CRISPR-cas9 Kinome-wide, pooled screen to identify genes that when knocked out differentially modulate the inflammatory response. Many inflammatory response pathways such as the Tumor Necrosis Factor Receptor, Interleukin 1 Receptor type 1, and Toll Like Receptor 4 converge on the Nuclear Factor Kappa B (NF-κB) transcription factor; thus, we use this gene to identify genes which enhance or activate or repress the inflammatory response. The inflammatory cytokine, Interleukin 1-beta (IL1β) was used to stimulate the inflammatory response in our astrocytes. However, in the process of screen construction, we observed unstable Cas9 protein expression within our cells and explored why this would occur.

Autism spectrum disorder (ASD) is a neurodevelopmental condition that heterogeneously impacts core domains of human function, such as communication, socialization, and cognition. Underlying these symptoms are complex genetic changes in a diverse array of genes and pathways. Developing a mechanistic understanding of ASD remains a major challenge that, left unanswered, will continue to hinder progress in the clinical management of this increasingly prevalent disorder. New approaches are needed to improve our understanding of the neurobiology of ASD, as well as current clinical paradigms.

In this dissertation, I discuss how techniques from the quantitative fields of machine learning and dynamical systems theory may be applied to decipher the cellular and molecular complexity of ASD. In Chapter 2, I address how whole genome sequencing data can be analyzed through machine learning to predict ASD phenotypes. The machine learning analysis describes novel methodology to vectorize genomics data and build interpretable classification models that relate to well-supported findings in the literature. In Chapter 3, I show that neuronal activity can be modeled as a nonlinear dynamical system to yield novel measures of neuronal state and dysfunction. I provide evidence for the minimum embedding dimension (MED) as a marker of diminished dynamical complexity of electrical activity recorded from ASD patient-derived neurons. I also probe the clinical and gene expression correlates of MED, which overlap with known ASD risk genes and pathways related to neurodevelopment in cortical and deep brain structures.

The approaches outlined in this dissertation seek to transfer tools from different quantitative fields to identify convergent cellular and molecular findings that characterize ASD. Genome classification and dynamical analysis improve on existing methods for the identification of disease signatures and reveal a set of common biological pathways and brain regions in ASD.

Alzheimer’s Disease (AD) is a progressive neurodegenerative disease regarded as a multifactorial disorder, a condition caused by multiple factors. With the majority of AD patients aged 65 and older, age is one of the strongest risk factors for AD, but its causes are not fully understood. Previous transcriptomic profiling of AD neurons identified a list of differentially expressed genes unique to AD; YY1, RUNX1, NFIB, and PPARγ emerged as some of the most upregulated transcription factors (TFs) reportedly associated with biological dysfunctions phenotypic of AD in various cell types. However, their function in aged neurons remains unknown. To investigate the possible contributions of YY1, RUNX1, NFIB, and PPARγ to AD pathology, we questioned their ability to reproduce AD-related phenotypes in aged neurons by characterizing transcriptomic changes caused by individual overexpression of each TF. We utilized induced neurons (iNs) produced from the reprogramming of old, healthy, donor-derived fibroblasts directly into neurons as a human-relevant means of capturing the age. All four TFs retain aspects of their previously described functions in old iNs with YY1 regulating the metabolic program, RUNX1 restricting neuronal identity, PPARγ regulating lipid metabolism and inflammation, and NFIB promoting the expression of developmental genes, possibly facilitating oncogenic transformations and contributing to the vulnerable hypo-mature state seen in AD neurons. Further investigation confirmed that YY1 and RUNX1’s transcriptomic effects are functionally translatable, identifying them as potential targets for medical intervention.