In this thesis, we study the neurocomputation by implementing two different neuron models. One is a semi magnetic micro p-n wire that emulates nerve fibres and supports the electrical propagation and regeneration. The other is a silicon neuron based on Hodgkin-Huxley conductance model that can generate spatiotemporal spiking patterns. The former model focuses on the spatial propagation of electrical pulses along a transmission line and presents the thesis that action potentials may be represented by solitary waves. The later model focuses on the dynamical properties such as how the output patterns of the active networks adapt to external stimulus. To demonstrate the dynamical properties of spiking networks, we present a central pattern generator (CPG) network with winnerless competition architecture. The CPG consists of three silicon neurons which are connected via reciprocally inhibitory synapses. The network of three neurons was stimulated with current steps possessing different time delays and that the voltage oscillations of the three neurons were recorded as a function of the strengths of inhibitory synaptic interconnections and internal parameters of neurons, such as voltage thresholds, time delays, etc. The architecture of the network is robust and sensitively depends on the stimulus. Stimulus dependent rhythms can be generated by the CPG network. The stimulus-dependent sequential switching between collective modes of oscillations in the network can explain the fundamental contradiction between sensitivity and robustness to external stimulus and the mechanism of pattern memorization. We successfully apply the CPG in modulating the heart rate of animal models (rats). The CPG was stimulated with respiratory signals and generated tri-phasic patterns corresponding to the respiratory cycles. The tri-phasic stimulus from the CPG was used to synchronize the heart rate with respiration. In this way, we artificially induce the respiratory sinus arrhythmia (RSA), which refers to the heart rate fluctuation in synchrony with respiration. RSA is lost in heart failure. Our CPG paves to way to novel medical devices that can provide a therapy for heart failure.

Both neuroinflammation, and an increase in microglial cells, have been associated with Autism Spectrum Disorder (ASD) through observation in human subjects as well as in mouse models. A mother having an infection early in pregnancy increases the chances for autism in her child. (Atladottir et al., 2012). This process is known as Maternal Immune Activation (MIA), and the proposed mechanism is that inflammatory signals cross from the mother to child; then in response to increased pro-inflammatory cytokines, microglia within the brain are activated to combat the infection. Microglia are essential to healthy synaptogenesis and neuronal growth, and a change in their signaling early in development has been shown to alter behavior in mouse models that replicate MIA. We use microglial depletion as a therapy to counteract the potentially harmful pro-inflammatory response in the developing mouse brain. Four experimental groups - control, MIA, microglial depleted, and a therapy group (MIA plus microglial depletion)- were run through a comprehensive series of behavioral and electrophysiological assessments. Layer 5 pyramidal cells (L5PNs) were targeted for recording in medial frontal cortex – a mouse cortical area important for cognition and social behavior. L5PNs are a heterogeneous population with cortical and subcortical targeting. Subcortical targeting neurons are thick tufted morphologically, and have an intrinsically bursting spike pattern. Analysis of the intrinsically bursting neurons revealed significant differences between the maternal inflammation and the microglial depletion groups across multiple physiological properties. Therefore, the therapy group had electrophysiological characteristics more consistent with the microglial depleted model than the autism model.

Thesis. 1976. Ph.D.--Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science. / Microfiche copy available in Archives and Engineering. / Bibliography: leaves 377-385. / Ph.D.

Specific regions within the adult mammalian brain maintain the ability to generate neurons. The largest of these, the ventricular-subventricular zone (V-SVZ), comprises the entire lateral wall of the lateral ventricles. Here, a subset of glial fibrillary acid protein (GFAP)-positive astrocytes (B cells) gives rise to neurons and oligodendrocytes throughout life. This process of neurogenesis involves quiescent B cells becoming proliferative (epidermal growth factor receptor (EGFR)-positive) and giving rise to neuroblasts via transit amplifying precursors. The neuroblasts then migrate through the rostral migratory stream (RMS) to the olfactory bulbs (OBs), where they mature into neurons. Studying the stem cells in the V-SVZ has been hindered by the shortage of molecular markers to selectively target them. Using microarray and qPCR analysis of putative quiescent neural stem cells we determined that they were enriched for PDGFRβ mRNA. We used immunostaining to determine the in vivo identity of PDGFRβ+ cells, and discovered that only GFAP+ cells within the V-SVZ stem cell lineage express PDGFRβ. Moreover, these PDGFRβ+ B cells contact the ventricle at the center of ependymal pinwheel structures and the vast majority of them are EGFR-. Importantly, the V-SVZ/RMS/OBcore axis was highly enriched for PDGFRβ expression compared with other brain regions. Detailed morphological analyses of PDGFRβ+ B cells revealed primary cilia at their apical process in contact with the ventricle and long radial processes contacting blood vessels deep within the V-SVZ, both of which are characteristics of adult neural stem cells. When PDGFRβ+ cells were lineage traced in vivo they formed olfactory bulb neurons. Using fluorescence-activated cell sorting (FACS) to purify PDGFRβ+ astrocytes we discovered this receptor is expressed by all adult V-SVZ neural stem cells, including a novel population of EGFR+ PDGFRβ+ cells which correspond to the activated neural stem cells. RNA-sequencing analysis of the purified populations revealed that PDGFRβ+ EGFR+ cells possess a transcriptional profile intermediate between quiescent neural stem cells and actively proliferating GFAP- progenitor cells. Finally, when PDGFRβ is deleted in adult GFAP+ NSCs we observe a decrease in EGFR+ and Dcx+ progenitor cells, together with an increase in quiescent GFAP+ astrocytes. A larger proportion of these mutant cells come in contact with the ventricular lumen, suggesting that PDGFRβ is required for V-SVZ astrocytes to act as stem cells, possibly by mediating interactions with their niche. Taken together, these data identify PDGFRβ as a novel marker for adult V-SVZ neural stem cells that is an important regulator of their stem cell capabilities.

A central question of developmental neurobiology is how the extraordinary variety of cell types in the nervous system is generated. A large body of evidence suggests that transcription factors acting as terminal selectors control cell fate determination by directly activating cell type-specific gene regulatory programs during neurogenesis. Neurons within the same class often further differentiate into subtypes that have distinct cellular morphology, axon projections, synaptic connections, and neuronal functions. The molecular mechanism that controls the subtype diversification of neurons sharing the same general fate is poorly understood, and only a few studies have addressed this question, notably the motor neuron subtype specification in developing vertebrate spinal cord and the segment-specific neuronal subtype specification of the peptidergic neurons in Drosophila embryonic ventral nerve cord. In this dissertation, I investigate the genetic basis of neuronal subtype specification using the Touch Receptor Neurons (TRNs) of Caenorhabditis elegans. The six TRNs are mechanosensory neurons that can be divided into four subtypes, which are located at various positions along the anterior-posterior (A-P) axis. All six neurons share the same TRN fate by expressing the POU-domain transcription factor UNC-86 and the LIM domain transcription factor MEC-3, the terminal selectors that activate a battery of genes (referred as TRN terminal differentiation genes) required for TRN functions. TRNs also have well-defined morphologies and synaptic connections, and therefore serve as a great model to study neuronal differentiation and subtype diversification at a single-cell resolution. This study primarily focuses on the two embryonically derived TRN subtypes, the anterior ALM and the posterior PLM neurons; each contains a pair of bilaterally symmetric cells. Both ALM and PLM neurons have a long anteriorly-directed neurite that branches at the distal end; the PLM, but not the ALM, neurons are bipolar, having also a posteriorly-directed neurite. ALM neurons form excitatory gap junctions with interneurons that control backward movement and inhibitory chemical synapses with interneurons that control forward movement, whereas PLM neurons do the reverse. Therefore, the clear differences between ALM and PLM neurons offer the opportunity to identify the mechanisms controlling subtype specification. Using the TRN subtypes along the A-P axis, I first found that the evolutionarily conserved Hox genes regulate TRN differentiation by both promoting the convergence of ALM and PLM neurons to the common TRN fate (Chapter II) and inducing posterior subtype differentiation that distinguishes PLM from the ALM neurons (Chapter III). First, distinct Hox proteins CEH-13/lab/Hox1 and EGL-5/Abd-B/Hox9-13, acting in ALM and PLM neurons respectively, promote the expression of the common TRN fate by facilitating the transcriptional activation of TRN terminal selector gene mec-3 by UNC-86. Hox proteins regulate mec-3 expression through a binary mechanism, and mutations in ceh-13 and egl-5 resulted in an “all or none” phenotype: ~35% of cells lost the TRN cell fate completely, whereas the rest ~65% of cells express the TRN markers at the wild-type level. Therefore, Hox proteins contribute to cell fate decisions during terminal neuronal differentiation by acting as reinforcing transcription factors to increase the probability of successful transcriptional activation. Second, Hox genes also control TRN subtype diversification through a “posterior induction” mechanism. The posterior Hox gene egl-5 induces morphological and transcriptional specification in the posterior PLM neurons, which distinguish them from the ALM. This subtype diversification requires EGL-5-induced repression of TALE cofactors, which antagonize EGL-5 functions, and the activation of rfip-1, a component of recycling endosomes, which mediates Hox activities by promoting subtype-specific neurite outgrowth. Thus, these results suggest that neuronal subtype diversification along the A-P axis is mainly driven by the posterior Hox genes, which induces the divergence of posterior subtypes away from the common state of the neuron type. I have also performed an RNAi screen to identify novel regulators of the TRN fate and identified the LIM domain-binding protein LDB-1 and the Zinc finger homeodomain transcription factor ZAG-1 as part of the regulatory network that determines TRN fate (Chapter IV). LDB-1 binds to and stabilizes MEC-3 and is also required for the activation of TRN terminal differentiation genes by MEC-3. ZAG-1 promotes TRN fate by preventing the expression other transcription factors EGL-44 and EGL-46, which inhibits the expression of TRN fate by competing for the cis-regulatory elements normally bound by the TRN fate selectors UNC-86/MEC-3. The mutual inhibition between ZAG-1 and EGL-44 establishes a bistable switch that regulates cell fate choice between TRNs and FLP neurons. I also investigated the genetic basis of neuronal morphogenesis using TRNs. By conducting a forward genetic screen searching for mutants with TRN neurite outgrowth defects, I identified a series of genes required for axonal outgrowth and guidance in TRNs. Following a few genes identified from the screen, genetic studies have revealed two novel mechanisms for neuritogenesis. First, Dishevelled protein DSH-1 attenuates the strength of Wnt signaling to allow the PLM posterior neurite to grow against the gradient of repulsive Wnt proteins, which are enriched at the posterior side of PLM cell body and normally repel the axons toward the anterior (Chapter V). Second, guanine nucleotide exchange factor UNC-73 and TIAM-1 promotes anteriorly and posteriorly directed neurite outgrowth, respectively; and outgrowth in different directions can suppress each other by competing for the limited neurite extension capacity (Chapter VI). As side projects, I performed mRNA expression profiling using isolated and separated populations of in vitro cultured ALM and PLM neurons and identified hundreds of genes differentially expressed between the two subtypes (Appendix I). I have also studied subtype differentiation of the VC motor neurons in the ventral nerve cord of C. elegans and discovered a mechanism by which histone modification patterns the expression of subtype-specific genes during terminal neuronal differentiation (Appendix II). In summary, my doctoral research established a framework for the study of neuronal subtype specification using the C. elegans TRNs and uncovered the genetic mechanisms for a variety of aspects of terminal neuronal differentiation. By investigating the generation of neuron type and subtype diversity in a well-defined model organism, my study provides novel insights for understanding the development of the nervous system.

Amyotrophic Lateral Sclerosis (ALS) is a neurological disease characterized by the degeneration of upper and lower motor neurons. Genetic studies have revealed that many ALS-associated genes are involved in autophagy, but the role of this pathway in motor neurons remains poorly understood. Here, we use the SOD1G93A mouse model to investigate the role of autophagy in ALS. We find neuronal subtype-specific regulation of autophagy over the course of disease progression. Vulnerable motor neurons form large GABARAPL1-positive autophagosomes that engulf ubiquitinated cargo recognized by the selective autophagy receptor p62. Other motor neurons and interneurons do not engulf cargo within GABARAPL1-positive autophagosomes and instead accumulate somatodendritic aggregates. To investigate whether motor neuron autophagy is protective or detrimental, we generated mice in which the critical autophagy gene Atg7 is specifically disrupted in motor neurons. Phenotypic analysis of these mice revealed that autophagy is dispensable for motor neuron survival but plays a key role in regulating presynaptic structure and function. By crossing these mice to the SOD1G93A mouse model, we find that autophagy inhibition accelerates early neuromuscular denervation and neurological dysfunction. However, loss of autophagy in motor neurons eventually leads to an extension of lifespan, and this is associated with reduced pathology in interneurons and glial cells. These data suggest that vulnerable motor neurons rely on autophagy to maintain neuromuscular innervation early in disease. However, autophagy eventually acts in a non-cell autonomous manner to promote disease spread and neuroinflammation. Our results reveal counteracting roles for motor neuron autophagy early and late in ALS disease progression.

Neurons rely on precise spatial and temporal control of neurotransmitter release to ensure proper communication. Neurotransmission occurs when synaptic vesicles in the presynaptic compartment fuse with the plasma membrane and release their contents into the synaptic cleft, where neurotransmitters bind to receptors on the postsynaptic neuron. Synaptic vesicle pools must maintain a functional repertoire of proteins in order to efficiently release neurotransmitter. Indeed, the accumulation of old or damaged proteins on synaptic vesicle membranes is linked to synaptic dysfunction and neurodegeneration. Despite the importance of synaptic vesicle protein turnover for neuronal health, the molecular mechanisms underlying this process are unknown. In this thesis, we present work that uncovers key components that regulate synaptic vesicle degradation. Specifically, we identify a pathway that mediates the activity-dependent turnover of a subset of synaptic vesicle membrane proteins in mammalian neurons. This pathway requires the synaptic vesicle-associated GTPase Rab35, the ESCRT machinery, and synaptic vesicle protein ubiquitination. We further demonstrate that neuronal activity stimulates synaptic vesicle protein turnover by inducing Rab35 activation and binding to the ESCRT-0 component Hrs, which we have identified as a novel Rab35 effector. These actions recruit the downstream ESCRT machinery to synaptic vesicle pools, thereby initiating synaptic vesicle protein degradation via the ESCRT pathway. Interestingly, we find that not all synaptic vesicle proteins are degraded by this mechanism, suggesting that synaptic vesicles are not degraded as units, but rather that SV proteins are degraded individually or in subsets. Moreover, we find that lysine-63 ubiquitination of VAMP2 is required for its degradation, and we identify an E3 ubiquitin ligase, RNF167, that is responsible for this activity. Our findings show that RNF167 and the Rab35/ESCRT pathway facilitate the removal of specific proteins from synaptic vesicle pools, thereby maintaining presynaptic protein homeostasis. Overall, our studies provide novel mechanistic insight into the coupling of neuronal activity with synaptic vesicle protein degradation, and implicate ubiquitination as a major regulator in maintaining functional synaptic vesicle pools. These findings will facilitate future studies determining the effects of perturbations to synaptic homeostasis in neuronal dysfunction and degeneration.

A common organizational feature of any nervous system is the existence of groups of neurons that share a set of common traits but that can be further divided into individual neuron types and subtypes. Understanding the mechanistic basis of neuron type and subtype diversification processes will constitute a major step toward understanding brain development and evolution. In this dissertation, I have explored the mechanistic basis for the specification of motor neuron classes in the nematode C. elegans which serves as a paradigm for neuron diversification processes. Cholinergic motor neurons in the C. elegans ventral nerve cord share common traits, but are also comprised of many distinct classes, each characterized by unique patterns of effector gene expression (e.g. motor neuron class-specific ion channels, signaling molecules, and neurotransmitter receptors). Both the common as well as class-specific traits are directly activated by the terminal selector of cholinergic motor neuron identity, the EBF/COE-like transcription factor UNC-3. Via forward genetic screens to identify mutants that are defective in class specification, I have discovered that the diversification of UNC-3/EBF-dependent cholinergic motor neurons is controlled by distinct sets of phylogenetically conserved, motor neuron class-specific transcriptional repressors. One such repressor is in fact a novel gene previously uncharacterized in C. elegans or any nervous systems and is now named bnc-1. By molecularly dissecting the cis-regulatory region of effector genes, I found that the repressor proteins prevent UNC-3/EBF from activating class-specific effector genes in specific motor neuron subsets via discrete binding sites that are adjacent to those of UNC-3/EBF. And by using CRISPR/Cas9-mediated genome engineering to tag repressor proteins with inducible degrons, I demonstrate that these repressors share the important feature of being continuously required throughout the life of the animal to counteract, in a class-specific manner, the function of the UNC-3/EBF terminal selector that is active in all motor neuron classes. I propose that the strategy of antagonizing the activity of broadly acting terminal selectors of neuron identity in a neuron subtype-specific manner may constitute a general principle of neuron subtype diversification.

Gliomas are brain tumors that present with neurological symptoms including seizures and cognitive deficits. Starting at early stages of tumor development glioma cells diffusely infiltrate brain tissue where they interact with non-neoplastic cells including neurons and can perturb normal brain function. While the clinical consequences of glioma induced cortical dysfunction are well established, the neuronal alterations that underlie cortical dysfunction in glioma are unknown. We hypothesize that glioma cells infiltrate surrounding brain tissue and induce alterations in neurons that may contribute to the neurological symptoms associated with gliomas. Due to intermingling of glioma cells and neurons it has been challenging to isolate and characterize neurons from glioma brain tissue while preserving complex neuronal morphology. To address this issue we developed a new mouse glioma model that allowed us to obtain a neuron specific gene expression profile, otherwise obscured by the predominantly large population of glioma cells within the tumor. In this thesis I use this model to test the hypothesis that infiltrating glioma cells induce phenotypic alterations in neurons that contribute to the neurological symptoms associated with glioma. The Camk2a-Ribotag mouse glioma model enabled us to isolate neuron specific transcripts from glioma brain tissue. The Ribotag mouse has a conditional HA-tagged ribosomal protein (Rpl22) that can be expressed upon Cre-recombination. Camk2a is specifically expressed in excitatory neurons, the Camk2a-Cre mouse induces Cre-recombination in the Ribotag mouse so that Camk2a+ neurons selectively express HA-tagged Rpl22. We used the Camk2a-Ribotag glioma model to isolate neuron specific ribosome bound transcripts to characterize neuronal alterations in glioma. In chapter 2 of this thesis I describe how we developed and characterized the Camk2a-Ribotag mouse glioma model. We first obtained mouse glioma cells that have p53 deletion and overexpress PDGFRa, then we injected these cells in the Camk2a-Ribotag mouse and use this as our glioma model to extract neuron specific ribosome bound transcripts. This method is referred to as translating ribosome affinity purification (TRAP) which is used to obtain cell type specific translational profiles. Using this approach we identified alterations in neuronal gene expression, specifically we show that there is an upregulation of actin binding genes associated with dendritic spine morphology and a downregulation of synaptic genes associated synaptic regulation. We demonstrate that drebrin, an actin binding protein in dendritic spines, is upregulated in tumor brain synaptosomes, we also show a downregulation of dendritic spine density in HA-tagged neurons which suggests that these neuronal alterations contribute to synaptic dysfunction in our glioma model. Dendritic spines are dynamic structures that regulate synaptic function in response to diverse stimuli. mTOR signaling can regulate brain specific functions such as synaptic plasticity. Alterations in mTOR signaling can result in cognitive deficits, epilepsy and brain abnormalities that are associated with neurological disease. We hypothesized that mTOR regulates the neuronal alterations we identified in our glioma model. In chapter 3 of this thesis I describe how we tested this hypothesis by acutely inhibiting mTOR signaling with the ATP competitive inhibitor AZD8055 in the Camk2a-Ribotag mouse glioma model. Using TRAP we show that acute mTOR inhibition reverses many neuron specific alterations that occurs in the glioma infiltrated cortex, actin binding genes that were upregulated in tumor brains were downregulated after mTOR inhibition and synaptic genes that were downregulated in tumor brains were upregulated after mTOR inhibition. These results suggest that key neuron specific alterations are regulated by mTOR signaling in our glioma model. In chapter 4 of this thesis I describe how we used ribosome profiling to identify translational alterations in our Camk2a-Ribotag mouse glioma model. Ribosome profiling in an RNA sequencing based method that is used to measure translation efficiency by calculating the number of ribosomes per transcript. Using this approach we identified an upregulation in the translation of DNA methylation and demethylation gene ontologies. These results suggest that alterations in specific DNA methylation and demethylation gene ontologies are regulated at the level of translation and warrant further analysis of cell type specific translational alterations using ribosome profiling. The work described in this thesis demonstrates 1) use of the Camk2a-Ribotag mouse glioma model for the identification of neuron specific alterations, 2) neuron specific alterations include the upregulation of dendritic spine genes, downregulation of synaptic genes and downregulation of dendritic spine density, 3) acute mTOR inhibition reverses many of these neuronal alterations, 4) ribosome profiling revealed the translational upregulation of epigenetic genes in our mouse glioma model. The findings described in this thesis provide the first characterization of neuron specific transcriptional and translational alterations in glioma infiltrated cortex that and provide new insights into the mechanisms that underlie the devastating neurological symptoms in glioma patients.

Organisms possess internal sensory systems to detect changes in physiological state. Despite the importance of these sensory systems for maintaining homeostasis, their development, sensory mechanisms, and circuitry are relatively poorly understood. To help address these gaps in knowledge, I used the tracheal dendrite (td) sensory neurons of Drosophila larvae as a model to gain insights into the cellular and molecular organization, developmental regulators, sensory functions and mechanisms, and downstream neural circuitry of internal sensory systems. In this thesis, I present data to show that td neurons comprise defined classes with distinct gene expression and axon projections to the CNS. The axons of one class project to the subesophageal zone (SEZ) in the brain, whereas the other terminates in the ventral nerve cord (VNC). This work identifies expression and a developmental role of the transcription factor Pdm3 in regulating the axon projections of SEZ-targeting td neurons. I find that ectopic expression of Pdm3 alone is sufficient to switch VNC-targeting td neurons to SEZ targets, and to induce the formation of putative synapses in these ectopic target regions. These results define distinct classes of td neurons and identity a molecular factor that contributes to diversification of central axon targeting. I present data to show that td neurons express chemosensory receptor genes and have chemosensory functions. Specifically, I show that td neurons express gustatory and ionotropic receptors and that overlapping subsets of td neurons are activated by decrease in O2 or increase in CO2 levels. I show that respiratory gas-sensitive td neurons are also activated when animals are submerged for a prolonged duration, demonstrating a natural-like condition in which td neurons are activated. I assessed the roles of chemosensory receptor genes in mediating the response of td neurons to O2 and CO2. As a result, I identify Gr28b as a mediator of td responses to CO2. Deletion of Gr28 genes or RNAi knockdown of Gr28b transcripts reduce the response of td neurons to CO2. Thus, these data identify two stimuli that are detected by td neurons, and establish a putative role for Gr28b in internal chemosensation in Drosophila larvae. Finally, I present data to elucidate the neural circuitry downstream of td sensory neurons. I show that td neurons synapse directly and via relays onto neurohormone populations in the central nervous system, providing neuroanatomical basis for internal sensory neuron regulation of hormonal physiology in Drosophila. These results pave the way for future work to functionally dissect the td circuitry to understand its function in physiology and behavior.