Mark J. Zylka - US grants

There are a number of human neurobiological disorders that are linked to defects in the circadian timing system. Many of these disorders have been treated effectively with the neuroendocrine hormone melatonin. Circadian effects of melatonin are believed to be mediated by a melatonin receptor subtype (Mel-1a) that is found in the human suprachiasmatic nucleus (SON). In this research proposal, we will use homologous recombination with the mouse Mel-1a gene locus to target nuclear-localized beta-galactosidase (nlacZ) expression to cells within the SON. Unlike Mel-1a mRNA, the nlacZ marker can readily be detected at a cellular level. This will facilitate sensitive double-immunocytochemical labelling to determine if cells that express Mel-1a are neurons and/or glia and will be used to determine what neurotransmitters and neuropeptides colocalize in Mel-1a cells. Most importantly, this knock-in strategy will provide us with the ability to determine, at the level of single cells, if the Mel-1a receptor is expressed in SON neurons that possess an autonomous circadian oscillator. These experiments have the potential of defining and fully characterizing a specific cell-type within the mammalian SON. Additionally, these studies could provide the first molecular tools for accessing the biological clock.

[unreadable] DESCRIPTION (provided by applicant): Neuropathic pain is a chronic and debilitating disease that is difficult to treat with existing analgesics. Recent studies reveal a critical role for lysophosphatidic acid (LPA) receptor activation in causing neuropathic pain. We found that many small diameter, presumably nociceptive, dorsal root ganglia (DRG) and trigeminal neurons express a transmembrane-localized phosphatase that dephosphorylates LPA. Based on these observations, our long term research objectives are to demonstrate that this phosphatase modulates LPA receptor signaling in vitro, using cultured sensory neurons, and in vivo, using mouse neuropathic pain models. To accomplish this, we will first use immunohistochemistry to characterize the sensory neurons that express this phosphatase and identify the peripheral tissues that are innervated by phosphatase-containing afferents. Next, we will test the hypothesis that this phosphatase modulates LPA receptor signaling in cell lines and in dissociated mouse DRG neurons using calcium imaging and luciferase reporter gene assays. We will use genetically modified mice expressing Green Fluorescent Protein to identify the DRG neurons that express this phosphatase. Finally, we will test the hypothesis that this phosphatase modulates neuropathic pain behaviors in wild-type and phosphatase knockout mice. [unreadable] [unreadable] [unreadable]

Use of genetically engineered mice permits investigators to study the development and function of the nervous system in vivo and associate gene function with disease. To generate these mice, investigators must first design and construct complex knock-in, knock-out, reporter gene or transgenic constructs. This represents a major obstacle for many labs, particularly because these large DNA constructs are difficult to engineer using classic molecular cloning techniques. To overcome this obstacle, we established the BAG Engineering Technology Core (Core 3) at the UNC Neuroscience Center. BAG Engineering (also called Recombineering) is based upon a highly efficient phage-derived ¿. Co// homologous recombination system to efficiently and precisely engineer large DMA constructs. Our core specifically uses the Lambda-RED cloning system (Copeland et al., 2001). In the 3 and !4 years since the core was established, we used the Lambda- RED system to generate a diverse array of DNA constructs for NINDS-funded investigators, including gene targeting constructs (knock-in / knock-out and Cre / LoxP based) and BAG transgenes (to epitope tag proteins in vivo, to direct the expression of GFP and CRE proteins in a tissue specific manner). The NIH-funded GENSAT project highlights the utility of using BAG transgenic mice to study nervous system function (Gong et al., 2003; Heintz, 2004).

Project Summary / Abstract More Americans suffer from chronic pain than heart disease, diabetes and cancer combined. Unfortunately, existing analgesics are not completely effective for all pain conditions and have serious side effects. These facts highlight what is undoubtedly a critical challenge for modern biomedical research-the need to provide pain relief without serious side effects. We will directly address this challenge by harnessing ectonucleotidases that are found endogenously in nociceptive (pain-sensing) circuits in dorsal root ganglia (DRG) and spinal cord. Ectonucleotidases degrade purine nucleotides (like ATP and ADP) that cause pain into adenosine-a compound that has analgesic properties in rodents and humans. Adenosine suppresses pain by acting through A1-adenosine receptors. To fully harness ectonucleotidases for the treatment of pain, we will identify all of the ectonucleotidases that metabolize nucleotides to adenosine in nociceptive circuits and then determine if these enzymes can be used alone or in combination to treat acute and chronic pain. We will utilize genetically modified mice that are missing these enzymes, recombinant ectonucleotidase proteins, behavioral assays and patch clamp electrophysiology for these experiments. In addition, we will use medicinal chemistry to synthesize adenosine prodrugs that can be converted into potent A1-adenosine receptor agonists by ectonucleotidases. We will measure the stability of these prodrugs in serum and use behavioral and physiological assays to assess analgesic efficacy and side effects. We will also test ectonucleotidases and prodrugs for efficacy in animal models of chronic inflammatory and neuropathic pain.

DESCRIPTION (provided by applicant): Angelman syndrome (AS) is a genetic disorder characterized by developmental delay, absent speech, intellectual disability, severe epilepsy, ataxia, and abnormal sleep. AS is caused by mutations in or deletion of Ube3a, an E3 ubiquitin ligase that is expressed biallelically in most tissues but is monoallelically expressed in the brain. Maternal-specific expression of Ube3a in the brain is thought to be due to production of an antisense transcript that overruns the paternal copy of Ube3a in mice and humans. Mice with maternal-specific deletions of Ube3a model many of the neurodevelopmental symptoms associated with AS, including epilepsy, learning deficits, and motor abnormalities. Using a high-throughput, unbiased screen with neurons from a Ube3a- YFP knockin mouse, we identified several small molecules that unsilence the paternal Ube3a allele at nanomolar concentrations. We hypothesize that the physiological and behavioral dysfunctions associated with Angelman syndrome can be treated by unsilencing the paternal Ube3a allele in vivo with one of these drugs. In this proposal we will: (1) Test the hypothesis that our lead compound upregulates paternal Ube3a in vivo; (2) Test the hypothesis that our lead compound can rescue physiological and behavioral deficits in Angelman syndrome model mice; (3) Test the hypothesis that genetic knockdown/out of the molecular target of our lead compound unsilences paternal Ube3a; (4) Test the hypothesis that the expression of the Ube3a-sense and Ube3a-antisense transcript levels can be used as biomarkers of drug efficacy (i.e. Ube3a unsilencing). Our research could lead to the first pharmacological treatment for Angelman syndrome (an autism spectrum disorder), and indeed for any disorder caused by mutation of an imprinted gene.

PROJECT SUMMARY New approaches for treating chronic pain are needed, particularly since existing analgesics have serious side effects and are not always effective at treating inflammatory pain and neuropathic pain-the two most common forms of chronic pain in humans. Inflammation and nerve injury lead to the release of a complex mix of chemicals that signal through molecularly diverse pronociceptive (pain-producing) receptors. Activation of these receptors increases the excitability and sensitivity of nociceptive dorsal root ganglia (DRG) and trigeminal ganglia neurons. Unfortunately, efforts to block individual pronociceptive receptors have so far failed to produce effective treatments for chronic pain. Here, we propose an innovative approach to reduce pain hypersensitivity that bypasses this long-standing problem associated with receptor diversity. Our approach is based on selectively reducing the level of the lipid second messenger phosphatidylinositol 4,5- bisphosphate (PIP2) in DRG neurons. Most pronociceptive receptors require PIP2 to initiate downstream signaling. Moreover, many TRP channels that detect noxious stimuli and ion channels that regulate membrane excitability require PIP2 for activity. PIP2 thus sits at a key convergence point for diverse receptors, ion channels and signaling pathways that promote and maintain chronic pain. In preliminary studies with mice, we identified a lipid kinase that generates at least 50% of all PIP2 in DRG neurons. Moreover, inactivation of this lipid kinase profoundly reduced nociceptive sensitization in response to an inflammatory agent. Based on our preliminary data, we hypothesize that this lipid kinase acts through PIP2 dependent mechanisms to regulate pronociceptive receptor signaling in DRG neurons and pain sensitization in vivo. To test this hypothesis we will: 1. Evaluate the extent to which this lipid kinase regulates nociceptive sensitization in vivo, using mouse models of acute, chronic and spontaneous pain, including models of inflammatory pain and neuropathic pain. We will use an innovative genetic approach to knock-down kinase activity. This approach selectively reduces PIP2 concentration in DRG but does not affect PIP2 concentration in other tissues that process pain signals. We will further evaluate PIP2-dependence using biochemical rescue experiments. 2. Evaluate the extent to which this kinase regulates signaling through diverse pronociceptive receptors, including G protein-coupled receptors, a tyrosine kinase receptor, and TRP channels that detect noxious stimuli. 3. Utilize a new conditional knockout mouse to inducibly delete this kinase only in sensory neurons of adults and to evaluate the extent to which this kinase regulates initiation and maintenance of inflammatory pain and neuropathic pain. We will be the first to rigorously study the importance of this kinase in the setting of chronic pain. Our preliminary data suggest this lipid kinase is a master regulator of pain signaling and sensitization.

DESCRIPTION (provided by applicant): Autism spectrum disorders (ASD) now affect an astounding 1 in every 88 children in America (CDC Report, 2012). Individuals with ASD show symptoms that include deficits in communication and social interactions as well as repetitive behaviors. A number of genes associated with synapse formation and function are mutated in patients with ASD, suggesting autism is a disorder of the synapse. ASD is a debilitating lifelong medical condition for patients and their caregivers, so finding new ways to diagnose, treat and prevent ASD remains one of the biggest challenges in neuroscience. In unpublished studies, we made a revolutionary discovery that directly addresses this challenge. Specifically, we found that a single molecular mechanism-transcription elongation-can be directly linked to expression of a large number (>34) of synapse-associated ASD candidate genes. Elongation is the process where RNA polymerase II, in coordination with numerous elongation factors, traverses DNA to generate a gene transcript. This process has never been studied in the context of any brain disease. Here, we will test the novel hypothesis that deficits in transcription elongation can reduce expression of ASD candidate genes and impair synapse function. This hypothesis is strongly supported by clinical data from ASD patients and by our unpublished experiments with cultured cortical neurons. To test this hypothesis, we will determine the extent to which genes associated with the elongation machinery regulate a) expression of numerous ASD candidate genes, b) synapse function and c) ASD-like behaviors in mouse models. We noticed that at least 10 genes associated with transcription elongation are mutated in patients with ASD. We will determine the extent to which these 10 mutations impair transcription elongation and expression of known ASD genes in neurons. Lastly, we describe a novel approach that will allow us to identify chemicals commonly found in the environment that impair

PROJECT SUMMARY Heritability studies indicate that genetic and environmental factors contribute to autism risk. While new sequencing technologies were used to identify hundreds of de novo gene mutations linked to autism, only a small number of environmental risks for autism have been identified to date. Moreover, these environmental risks were identified retrospectively, after a large number of people were exposed. There is thus a significant public health need to identify environmental risks for autism prospectively, before these risks contribute to disease. Brain transcriptional changes differentiate individuals with autism from neurotypical controls. This transcriptional signature of autism is defined by reduced expression of synaptic transmission genes and elevated expression of neuroimmune/microglial genes. Here, we hypothesize that candidate environmental risks for autism can be prospectively identified using the transcriptional signature of autism as a guide. We recently found that strobilurin fungicides reproducibly produce this transcriptional signature in embryonic cortical neuron cultures, making these fungicides ideal chemicals to test this hypothesis. Strobilurin fungicides poison mitochondrial complex III and, as we found, generate reactive oxygen species (ROS) and destabilize microtubules in neurons. Usage of these fungicides is surging on a diversity of food crops and one strobilurin is now being used in wallboards, posing a potential source for chronic exposure. Here we will comprehensively evaluate the extent to which prenatal fungicide exposure produces autism-related phenotypes in wild-type mice and exacerbates pathology in a new mouse line that models a human de novo autism-linked mutation. We will use a low dose that approximates human exposures and a higher dose that effects physiology and behavior when administered orally. To greatly accelerate the pace at which additional environmental risks for autism are identified, we will transcriptionally profile thousands of environmental-use chemicals on primary neuron cultures using an innovative targeted sequencing approach. We found that primary neuron cultures model the molecular and cellular diversity of the intact brain. Our preliminary data indicate this targeted sequencing approach can be performed robotically in 384-well dishes with cultured primary neurons and can identify chemicals that produce the transcriptional signature of autism. This targeted sequencing approach can also identify chemicals that produce transcriptional changes associated with other brain disorders.

PROJECT SUMMARY Neuropathic pain originates from nerve injury and impairs the quality of life for a substantial number of Americans. Nerve injury causes cellular and molecular changes in the spinal cord, but the precise effects on pain-related spinal circuits and cell types has not fully been resolved. Molecularly distinct cell types in a complex tissue can now be comprehensively identified and studied using single cell RNA sequencing (scRNAseq) technologies. In preliminary studies, we found that the scRNAseq approach known as Drop-seq can be used to identify the principle cell types of the mouse spinal cord, including neurons, astrocytes, oligodendrocytes, microglia, and endothelial cells. In this proposal, we hypothesize that single cell sequencing can be used to comprehensively identify cellular and molecular changes that take place in the spinal cord following peripheral nerve injury. To accomplish this, we will complete the following aims: (1) Test the hypothesis that Drop-seq can identify changes in spinal cell populations in a mouse neuropathic pain model, which features prolonged (>21 d) mechanical allodynia?a core behavioral phenotype of neuropathic pain. (2) Test the hypothesis that scRNAseq at varying timepoints can identify cell-specific transitions from acute to chronic pain. Our research will provide the first comprehensive understanding of how spinal cell types, and the neural circuits they make up, change in response to nerve injury?an insult that produces stable behavioral changes indicative of neuropathic pain.

The purpose of ?UNC Neuroscience Center Research Cores? (P30 NS045892-14) is to provide NINDS-funded and other NINDS-priority investigators with research capabilities that cannot be supported or sustained by individual laboratories. For the past 13 years, the Cores funded by NS045892 supported almost all NINDS- funded research at the UNC?Chapel Hill School of Medicine. During this time, UNC neuroscientists utilized these Cores to facilitate breakthrough discoveries in NINDS priority areas of neurogenetics, pain, neurodegeneration, and systems neuroscience. For our competitive renewal, we propose two Cores: Microscopy and Bioinformatics. We regularly upgraded our Microscopy Core to take advantage of increased sensitivity, scanning speeds, tiling capabilities, and now super-resolution capabilities. In addition to a Zeiss LSM 780 and other scopes, our Microscopy Core will house a new Zeiss LSM 880 with Airyscan and Fast module. The Core will also provide IT infrastructure and a pipeline for computationally intensive image processing and analysis of light sheet microscopy data. For the Bioinformatics Core, we will fund two bioinformaticians to analyze high-throughput sequencing (HTS) data, including exome and genome sequencing, bulk and single-cell RNA-seq, bulk and single-cell ATAC-seq, Hi-C, and ChIP-seq data. Access to advanced microscopic imaging and bioinformatics support is essential to the performance of cutting- edge research in NINDS-priority areas. With continued support, our NINDS-funded and other NINDS-priority investigators will have their transformative ideas enabled via convenient and affordable access to these key technologies.

PROJECT SUMMARY- ADMINISTRATIVE CORE The Administrative Core provides management for the Center grant Core services. The Administrative Core will ensure that the Microscopy Core and the Bioinformatics Core operate smoothly to support projects of Qualifying Investigators, other investigators in NINDS-priority areas, and young investigators. The Administrative Core also assesses work flow of the two Cores and monitors the prioritization plan. Finally, the Administrative Core monitors the activity reports prepared by the Core Directors and oversees the recharge arrangements in place for each Core.

PROJECT SUMMARY-BIOINFORMATICS CORE The Bioinformatics Core analyzes high-throughput sequencing (HTS) datasets for NINDS-funded and other NINDS-priority investigators at UNC. The Bioinformatics Core leverages Center grant support from NINDS, an Intellectual Disabilities Research Center (IDDRC) grant from NICHD (awarded to the Carolina Institute for Developmental Disabilities; CIDD), and Institutional support from the Lineberger Comprehensive Cancer Center (LCCC). This core serves NINDS-funded and priority faculty, CIDD faculty who study neurodevelopmental disorders, and LCCC faculty who study brain cancers. Support for the Core Director is split 50:50 between NINDS/CIDD and LCCC, while two additional bioinformaticians (TBN) will be supported, one by NINDS/CIDD (50:50) and the other by LCCC (100%).

PROJECT SUMMARY- MICROSCOPY CORE The Microscopy Core provides state-of-the-art microscopes, image analysis, expert consultation, and advanced training in support of Qualifying Investigators and other NINDS-priority investigators at UNC. The Core consists of a microscopy expert, a suite of microscope systems, image analysis tools, and associated resources for advanced microscopy. The Microscopy Core leverages support from the NINDS Center grant with support from an Intellectual and Developmental Disabilities Research Center (IDDRC) from NICHD (awarded to the Carolina Institute for Developmental Disabilities; CIDD) to enhance the research programs of neuroscience faculty working in NINDS priority research areas.

PROJECT SUMMARY Angelman syndrome (AS) is a severe neurodevelopmental disorder caused by deletion or mutation of the maternal allele of UBE3A. UBE3A is biallelically expressed in nearly all cells of the body except in mature neurons, where the paternal allele is silenced by an extremely long non-coding RNA called UBE3A-ATS. In light of this biology, the most direct way to treat behavioral dysfunctions associated with AS is to unsilence the intact paternal UBE3A allele. CRISPR/Cas9 technology can be used to target specific regions of the mammalian genome for mutagenesis or transcriptional repression. In unpublished studies, we generated hundreds of S. pyogenes (Sp)Cas9 guide RNAs (gRNAs) that target regions throughout UBE3A-ATS. Several of these gRNAs, when transfected along with SpCas9, potently unsilenced paternal Ube3a in cultured mouse cortical neurons. Some of our most effective gRNAs targeted a region of Ube3a-ATS that is conserved between mice and humans, making it possible to translate our findings to human neurons. Here, we will test the central hypothesis that CNS-directed delivery of Cas9 and a gRNA that targets Ube3a-ATS can enduringly unsilence paternal UBE3A and treat behavioral phenotypes associated with Angelman syndrome. We will use adeno-associated virus (AAV) for delivery because it can drive gene expression for years in the brain. Pilot studies with S. aureus (Sa)Cas9, a smaller Cas9 variant, suggest that our gene therapy approach can be used to unsilence paternal Ube3a in mice for at least three months. To advance this innovative gene therapy towards the clinic, we will evaluate efficacy, on- and off-target effects, and mechanism of action of candidate therapeutic SaCas9 gRNAs that target Ube3a-ATS. We will use cultured neurons from AS model mice and AS-derived human neurons. We will package SaCas9 and an optimized gRNA into a single AAV vector, and then evaluate unsilencing efficacy and longevity for up to two years in mice, as well as biodistribution and toxicity. Lastly, we will evaluate the extent to which AAV-mediated delivery of this CRISPR/Cas9-based gene therapy treats behavioral phenotypes in AS model mice.

PROJECT SUMMARY While significant progress has been made in identifying de novo gene mutations linked to autism risk, much less attention has been paid to environmental risks and the extent to which these risks cause autism pathology in susceptible individuals. Environmental factors, including gestational exposure to pyrethroid pesticides and valproic acid, are implicated in risk for autism. Prenatal exposure to pyrethroids is also linked to risk for developmental delay and attention deficit hyperactivity disorder (ADHD)?one of the most common neurodevelopmental disorders. However, these environmental risks were identified retrospectively, after a large number of people were exposed. Thousands of chemicals are registered for use in the environment, and humans are potentially exposed to many of these chemicals to varying degrees, including chemicals in plastics and building materials. We currently lack a way to systematically evaluate which environmental-use chemicals have the greatest potential to harm the developing brain. The inability to identify environmental threats to the brain early?before they cause disease?represents one of the major public health challenges of our time. This challenge is particularly relevant to autism, which now affects 1 in 59 individuals in America, and where heritability studies indicate that genetic and environmental factors contribute to autism risk. Our research program is guided by the hypothesis that ?candidate? environmental risks for autism and other neurodevelopmental disorders can be identified rationally, by identifying chemicals and mixtures that target molecular pathways implicated in these disorders. Our long term goals are to 1) identify environmental-use chemicals and mixtures that target molecular pathways implicated in neurodevelopmental disorders. These studies will utilize primary human neural progenitor cells (phNPCs), primary neurons, and endpoints that are compatible with high-throughput screening. 2) Assess real world exposure to these chemicals/mixtures. If environmental sampling and biomonitoring data are not available for these chemicals/mixtures, we will work with a network of Environmental Health Science (EHS) researchers to collect these data. 3) Evaluate exposure risk in vivo using wild-type and CRISPR/Cas9-engineered mice that model human de novo autism-linked mutations. We will prioritize chemicals/mixtures that a) impact one or more phNPC/neuron assay endpoints, b) are verified exposure risks to humans, and c) enter the placenta and/or developing brain following maternal exposure. While the specific projects will evolve over time, we plan to initially focus on individual and joint exposures to pyrethroids and strobilurins?a new class of fungicides that inhibits mitochondria. Both chemical classes impair neuronal functions and co-occur in the home environment. We will evaluate the extent to which prenatal exposure to these and other prioritized chemicals and mixtures exacerbate brain and behavioral phenotypes associated with autism and other neurodevelopmental disorders across the lifespan.

Abstract: The Neuroscience Curriculum (NBIO) at UNC-Chapel Hill (UNC-CH) is one of the oldest neuroscience graduate training programs in the US having granted PhDs continuously for 52 years. NBIO has a reputation for excellence and rigor, and there has been constant recruitment of new faculty which keeps the program innovative and fresh. In support of NBIO, we request renewed funding for T32 NS007431, ?Neuroscience Predoctoral Training at UNC- Chapel Hill?. This T32 supports 10 stipends and enrichment funds for NBIO students in the 1st and 2nd year students in training. The T32 importantly impacts neuroscience graduate training at UNC-Chapel Hill (UNC-CH) in several ways: 1) T32 support for travel provides access to career-enhancing enrichment activities for the most promising students. 2) T32 support increases student access to the best training labs. 3) T32 support enhances our recruitment and mentoring of UR trainees. 4) T32 support leverages resources provided by the UNC School of Medicine and thereby helps support the seminar series and the annual UNC Neuroscience Symposium and Retreat. NBIO is a comprehensive neuroscience graduate training program that has strong leadership and standing committees to guide the program, a thoughtfully constructed and rigorous Core course, a required course on programming and statistics for Neuroscience, and a diverse group of electives. There are numerous, well-attended community activities including weekly seminars by both students and visiting faculty, an annual symposium, presentation of the internationally recognized Perl/UNC Neuroscience Prize, and an annual student organized Neuroscience Retreat. Mentoring in the program is continuous and there are well-designed activities to educate students about academic and non-academic careers, as well as mechanisms to enhance faculty mentoring of students. Faculty and students have outstanding publication records. Multiple previous trainees received individual NRSAs or other prestigious awards during the last funding period. Almost all trainees who received PhDs during the past 10 years are pursuing scientific careers including academic positions, positions in pharma/biotech, other scientific careers, completion of MD/PhD training, and postdoctoral fellowships. There is an outstanding record of recruiting and retaining UR trainees. Renewal would allow UNC?CH to continue training the next generation of neuroscientists including a strong contingent from UR groups.

PROJECT SUMMARY UBE3A is an E3 ubiquitin ligase that targets itself and other substrates for proteasomal degradation. In the developing brain, neuronal progenitors and immature neurons biallelically express Ube3a, but as neurons mature, Ube3a expression becomes restricted to the maternally-inherited allele. Mutations that elevate maternal or paternal Ube3a are linked to autism risk, but precisely how UBE3A excess impairs neurodevelopment is unclear. Recently, we found that phosphorylation of threonine 485 (T485) inhibits UBE3A ubiquitin ligase activity. UBE3A T485 phosphorylation initiates embryonically and peaks at birth, suggesting that phosphorylation might protectively limit UBE3A activity during early cortical development. Additionally, we found that an autism-linked de novo mutation in UBE3A (T485A) disrupts this phosphorylation site, effectively locking UBE3A always-on. We engineered a mouse that precisely models this human UBE3A T485A mutation, allowing us to evaluate how this novel gain-of-function mutation affects brain and behavioral phenotypes when inherited maternally or paternally. In preliminary studies, we found that cortical thickness and brain weight were significantly increased at birth in all three Ube3a T485A genotypes (paternal, maternal, homozygous). Mutations in other autism-linked genes increase brain weight to a similar extent. These findings suggest a novel and previously unrecognized prenatal function for UBE3A in brain development. All three Ube3a T485A mutant genotypes also had behavioral phenotypes consistent with neurodevelopmental disorders. Since little is known about how UBE3A impairs brain function at any age, we performed unbiased proteomics to identify brain-relevant substrates. Our preliminary proteomics data link UBE3A directly to the proteasome, a structure that can influence the cell cycle and signaling pathways important for brain development. These and other data lead us to hypothesize that UBE3A T485A alters the balance of cell proliferation and differentiation during brain development, in part by impairing proteasome function, and contributes to autism-associated phenotypes later in life. The experiments in this proposal will rigorously demonstrate that (1) UBE3A T485A alters the balance of progenitor proliferation and differentiation in the cerebral cortex, (2) parent-of-origin inheritance of Ube3a T485A influences autism-related brain and behavioral phenotypes, and (3) UBE3A T485A interacts with the proteasome and impairs proteasome function in the brain. Unbiased proteomics experiments will identify brain-relevant substrates of UBE3A, and broaden our understanding of which molecular pathways are affected by gain-of-function mutations that enhance UBE3A activity.

PROJECT SUMMARY Opioid analgesics are commonly used to treat pain but have serious side effects, including addiction, dependence, and death from overdose. While there is a significant need for new non-addictive analgesics, efforts to develop new pain medicines have met with limited success. In part, this failure is due to an overreliance on evoked pain measures in preclinical models. Indeed, most preclinical models do not measure spontaneous pain?the main symptom of chronic pain in humans. To increase translational relevance, the Mouse Grimace Scale (MGS) was developed to quantify characteristic facial expressions associated with spontaneous pain. The MGS is reproducible across labs and was used to evaluate the efficacy of analgesics. However, the MGS has not been widely adopted due to its high resource demands and low throughput. To overcome this limitation, we adapted a machine learning model to classify the presence or absence of pain from mouse facial expressions. We called this model the automated Mouse Grimace Scale (aMGS). After training, this model identified mice in pain with 94% accuracy, comparable to a highly-trained human. However, our original ?aMGS 1.0? is limited in several respects. It is only accurate at detecting facial grimacing in white- coated mice, and produces a binary assessment (?pain? vs. ?no pain?) instead of a graded score. Moreover, aMGS 1.0 cannot dynamically determine pain status from full-motion videos. Additionally, we relied on an older piece of software that does not consistently extract high-quality images of the mouse face. The aMGS 1.0 also has difficulty distinguishing between images of sleeping and grimacing mice. Finally, aMGS 1.0 suffers from a ?black box? problem inherent to most machine learning algorithms, in that we do not know what facial details it uses to produce a pain assessment. Here we propose to overcome all of these limitations by developing a more sophisticated version of our automated pain classifier (aMGS 2.0). To achieve this goal we will: 1) Develop and validate a new open-source platform to classify (frame-by-frame) spontaneous pain intensity from mouse facial expressions, using albino (white) mice and motion information. 2) Enhance the generality of aMGS 2.0 for use with black mice. And, 3) Develop a user-friendly web-based platform that operates on computer-based and mobile devices. We will validate the utility of aMGS with three pain assays that produce grimaces in rodents?inflammatory pain, post-surgical (laparotomy) pain, and neuropathic pain. To increase rigor and reproducibility, two pain assays will be performed and scored with aMGS 2.0 in an independent lab. Numerous investigators in the pain field have expressed interest in using our proposed model. The platform will include a cloud-based data repository and analytic tools to facilitate curation of public data, continuous improvement of the model over time, and integration of new analytic tools. One analytic tool that we plan to develop will identify mouse features that most influence pain classification.