Susanne E. Ahmari - US grants

DESCRIPTION (provided by applicant): The candidate requests support for a 5-year program of training and research to develop a novel translational approach for dissecting the brain mechanisms underlying obsessive-compulsive disorder (OCD). Her long-term goal is to use reliable behavioral indicators in patients to link OCD symptoms to specific brain pathways in humans, and use mouse models to understand how molecular and functional changes in these pathways interact to produce dysfunctional behavior. In the proposed training program, the candidate will build upon her previous experience in basic neuroscience research and clinical treatment of anxiety disorders to perform a multidisciplinary project in the Anxiety Disorders Research Clinic and the Division of Integrative Neuroscience at Columbia University, the environment for this proposal. Her training plan includes 1) developing expertise in the clinical phenotype and neural circuitry of OCD and its applicability to mouse models;2) training in identification of human biomarkers corresponding to different components of OCD neural circuits;3) learning mouse behavioral paradigms related to OCD;4) developing expertise in optogenetic/ fiberoptic technology;and 5) training in responsible conduct of research. Completion of these short-term career goals will be enabled by a strong and supportive institutional environment including clinical and basic neuroscience mentors;strong multidisciplinary consultant relationships at other institutions;and a rigorous training plan including relevant coursework. The candidate's research plan builds on several lines of evidence which indicate that OCD symptoms result from dysfunction in the cortico-striato-thalamo-cortical (CSTC) circuit. To complement her training program, she proposes two parallel projects in humans and mice that will use multilevel investigation to deconstruct behavioral components in OCD, and link them to underlying circuits. The aims of the two projects are: 1) to identify reliable behavioral indicators of OCD in humans that correspond to dysfunctional CSTC circuits and can also be studied in mice;and 2) to modulate activity in CSTC circuits using optogenetic/ fiberoptic technology to test the hypothesis that abnormalities in CSTC circuit function lead to OCD-related behaviors in mice. Completion of Aim 1 will determine which behavioral indicators are clinically-relevant and therefore useful as readouts of OCD-related behaviors in her mouse studies These two projects will provide new data relevant to the neural mechanisms of OCD. Preliminary studies demonstrate feasibility of this approach. By conducting these projects, the candidate will learn state-of-the art methods for assessing the behavioral components of OCD, for constructing tissue-specific transgenic mice, and for modulating neural circuits in awake-behaving animals. At completion of the K08, the candidate will have the experience and skills to become an independent translational investigator. PUBLIC HEALTH RELEVANCE: Obsessive Compulsive Disorder (OCD) is a chronic, disabling disorder with 2-3% lifetime prevalence, and is a leading cause of illness-related disability. A better understanding of how dysfunctional circuits lead to OCD symptoms is needed to guide development of new treatments. The culmination of the candidate's proposal is to develop a bi-directional approach for bridging mouse and human research, so that findings from clinical studies can significantly impact the direction of basic research, and findings from animal studies can lead to significant improvements in treatment.

DESCRIPTION (provided by applicant): Support is requested for a 2-year collaborative grant between scientists at the New York State Psychiatric Institute and Vanderbilt University to investigate the pathophysiology underlying OCD. The proposal bridges basic and clinical OCD research by integrating findings from the research team's ongoing human genetic studies into the proposed mouse experiments. The research plan thus capitalizes on the expertise of the team in 1) human OCD genetic studies, 2) development of transgenic mice, 3) biochemical assays, and 4) mouse behavioral analysis. Current understanding of the molecular and cellular abnormalities underlying OCD is limited, in part because post-mortem studies in humans have not been performed. In addition, mouse studies have not yet been convincingly linked to the clinical phenotype and genetic abnormalities seen in OCD patients. To date, the only gene which has been consistently linked to OCD in human genetic studies is SLC1A1, which codes for a protein that transports the neurotransmitter glutamate. In addition, there is evidence from human studies that abnormal regulation of glutamate transmission in striatum is correlated with OCD symptoms. This has led to the hypothesis that abnormal levels of the SLC1A1 glutamate transporter in striatum lead to 1) abnormalities in the glutamate system, 2) changes in brain structure, and 3) OCD symptoms. The proposed R21 will test this hypothesis using novel knock-in mouse technology. In the first aim, the researchers will use an efficient system they have previously developed for manipulating gene expression in mice called the FAST system (Flexible Accelerated STOP TetO-knockin). This will allow them to develop a novel knock-in mouse line called tetO-Slc1a1, which will permit precise regulation of SLC1A1 expression levels in brain regions implicated in OCD. They will then use this mouse line to generate abnormally high levels of the SLC1A1 glutamate transporter specifically in striatum. This will simulate the effect of the version of the gene found most commonly in OCD patients. In the second aim, the mice with abnormally high levels of Slc1a1 in striatum will be characterized by: 1) measuring glutamate system functioning~ 2) examining brain structure~ and 3) testing behavior in OCD-relevant paradigms that measure anxiety and repetitive behaviors. This will provide the first direct test of whether OCD-related dysfunction is caused by abnormal expression of the leading human OCD candidate gene. Completion of this grant will lead to an amenable system for 1) further dissection of the molecular, cellular, and electrophysiologic underpinnings of observed changes~ and 2) determination of whether there is a particular time in development during which the brain is more vulnerable to developing OCD. These studies will lead to a better understanding of how dysfunctional circuits lead to OCD symptoms, which is necessary to guide development of new treatments for this severe mental illness.

DESCRIPTION (provided by applicant): Stereotyped and compulsive behaviors are prominent, disabling, and notoriously treatment-resistant symptoms in multiple severe psychiatric disorders, including autism, Obsessive Compulsive Disorder (OCD), and schizophrenia. It is thought that these perseverative behaviors result from disruptions in multiple neurocognitive domains, including response inhibition and habit formation. However, the pathologic mechanisms leading to abnormal repetitive behaviors are still unknown. Though perseverative behaviors are associated with abnormal activity in cortico-striatal-thalamic (CSTC) circuits, it is still unclear how aberrant neural activity triggers these maladaptive actions. In addition, we have very limited insight into factors that lead to persistence of abnormal repetitive behaviors. Obtaining this knowledge is a critical step towards developing novel strategies for interrupting perseverative behaviors before they become hard-wired, or even completely preventing their onset. In this project, we will combine cutting-edge neuroscience technologies and translatable neurocognitive probes to identify molecular and circuit changes linked to onset, persistence, and successful treatment of perseverative behaviors, with a goal of identifying novel treatment strategies that cut across diagnostic boundaries. Though this project will focus on OCD since convergent functional imaging findings from OCD patients provide a strong clinical foundation for circuit-based translational studies, knowledge gained will be broadly applicable to other severe psychiatric disorders with perseverative thought patterns and actions. In previous work, we showed that repeated hyperstimulation of projections from orbitofrontal cortex (OFC) to ventromedial striatum (VMS) led to a progressive and stimulation-independent increase in perseverative grooming, a mouse behavior relevant to OCD in humans. This was accompanied by increased evoked activity at OFC-striatal synapses. Behavioral abnormalities persisted even in the absence of stimulation, suggesting that brief but repeated hyperstimulation of OFC-VMS projections is sufficient to yield long-lasting pathologic changes in circuit function. In this project, we will test the central hypothesis that plasticity changes at key CSTC circuit nodes underlie the induction of abnormal repetitive behaviors, and may serve as an avenue for treatment. In Aim 1, we will map the extended neural network associated with onset of abnormal repetitive behaviors, by determining the extent of plasticity changes using 1) in vivo physiology and 2) measurement of brain-wide levels of a neural plasticity marker, ?fosB. In Aim 2, we will determine optimal sites for treatment of compulsive/ stereotyped behaviors via blockade of abnormal activity in linked circuit nodes in transgenic OCD models; this will inform targeting strategies for brain-stimulation based treatments. In Aim 3, we will begin to identify molecular and environmental factors associated with persistence and relapse of perseverative behaviors, with a goal of finding new targets to prevent symptom entrenchment. The ultimate goal of these studies is to identify new treatment options for disabling perseverative and compulsive behaviors.

Project Summary/Abstract Marijuana (Cannabis sativa) is the most widely used illicit drug in the U.S. and use in adolescence is common. Onset of cannabis use in early adolescence is associated with temporary cognitive impairments, but the long- term neurobiological consequences are not well understood. In addition, medicinal use of cannabinoids in children and adolescents may be warranted for treating certain seizure-related disorders; therefore, an increased understanding of the costs vs. benefits of adolescent cannabinoid exposure are warranted to inform the ongoing development of policies regulating legal marijuana use. Rodent models provide the opportunity to perform controlled experiments on the timing, duration, and amount of cannabinoid exposure and eliminate potential confounds of human studies, such as pre-existing conditions and exposure to other drugs of abuse. In addition to determining the long-term behavioral consequences of cannabis exposure, it is also important to determine the effects on neural circuit activity that may produce more subtle effects on functional outcomes. Current technologies limit our ability to perform longitudinal recordings of neural activity in a cell-type specific manner across adolescent development, particularly in animals self-administering intravenous drugs of abuse and performing complex cognitive tasks. One solution is to perform in vivo imaging of ensemble neural activity using genetically-encoded calcium indicators (GECIs). In vivo Ca2+ imaging can resolve cellular activity in deep brain structures in awake behaving animals using integrated, head-mounted gradient refraction index (GRIN) lenses and miniaturized epifluorescent microscopes (microendoscopes). Currently, this technique has been primarily restricted to studies in adult mice. However, development of a transgenic rat expressing the latest generation GECI, GCaMP6f, will permit imaging of cortical activity over the course of adolescent development and make it feasible to compare neural circuit development in rodents that self-administer intravenous drugs of abuse, like cannabinoids, relative to controls. Thus, in this two-year project we propose to develop transgenic rats expressing GCaMP6f under the control of a neuronal promoter. We will assess neural activity patterns in prefrontal cortical regions responsible for working memory and cognitive performance, in rats that self-administer cannabinoids in adolescence. Adolescent self-administration groups will be compared to similarly trained adults and to food self-administering controls. We will assess working memory performance and associated cortical activity in these rats across development and in adulthood after a period of abstinence. The results of these studies will clarify the acute vs. long-term consequences of adolescent cannabinoid self- administration on working memory performance, and determine if the development of task appropriate neural activity patterns are affected by ongoing or prior cannabinoid self-administration. In addition, the GCaMP6f transgenic rat will be publicly available for broad use by the neuroscience community.

PROJECT SUMMARY Obsessive-compulsive disorder (OCD) is one of the most disabling, chronic psychiatric disorders, with a lifetime prevalence of 2-3%. Emerging findings point to a significant role for basal ganglia circuits in OCD. Despite this, our understanding of the molecular pathophysiology of OCD remains inadequate, and our treatment options leave most patients with continued impairment. The best-replicated genetic finding in OCD is association with SLC1A1, encoding the neuronal glutamate, aspartate, and cysteine transporter EAAT3/EAAC1. However, the impact of this gene on the normal and abnormal functioning of OCD-related circuits is unknown. To fill this knowledge gap, we developed a STOP-TetO knock-in mouse line that allows us to flexibly manipulate Slc1a1 expression. Using dopamine agonists as a probe, we found that EAAT3 loss decreases basal ganglia-mediated repetitive, stereotyped behavior. Our convergent data support the hypothesis that increased EAAT3 function plays a role in OCD pathology and that decreasing EAAT3 activity may serve as a novel treatment option. Little is known, however, about EAAT3's molecular and functional impact in the basal ganglia. Elsewhere in the brain, EAAT3-mediated transport decreases neurotransmission at perisynaptic glutamate receptors and provides substrate for GABA and glutathione synthesis, but it is unclear which of these functions is important in basal ganglia circuits, and whether EAAT3's impact on dopaminergic neurotransmission is pre- or post-synaptic. Using our flexible mouse model and previously established OCD optogenetic and transgenic mouse models, this R01 will 1) examine effects of EAAT3 ablation and targeted rescue on basal ganglia function and repetitive behavior, and 2) determine if EAAT3 ablation leads to symptom resolution in phenotypically-similar but etiologically-independent mouse models of OCD with abnormal basal ganglia signaling. These data could be leveraged to demonstrate a clear treatment target that motivates development of promising EAAT3 inhibitor lead compounds.

PROJECT SUMMARY/ABSTRACT Clinical imaging studies in patients with obsessive compulsive disorder (OCD) have 1) highlighted the importance of the prefrontal cortex (PFC) in the pathophysiology and treatment of OCD, and 2) demonstrated marked dissociations in PFC neural activity changes associated with different elements of the disorder. Specifically, PFC hyperactivity is typically observed during symptom provocation, while hypoactivity is observed during cognitive testing. Devising interventional treatment strategies is therefore a challenge because it is unclear if PFC activity should be increased or decreased for therapeutic efficacy. In human studies, it is not possible to dissect the potential neural substrates for these state-dependent differences in PFC activity. Because in vivo calcium imaging in awake behaving mice allows tracking of neural activity in individual neurons across time and across different behavioral paradigms, we can now use this technique in preclinical models to determine whether discrete populations of PFC neurons show neural activity changes associated with symptom provocation versus cognitive impairment in OCD. To date, mouse models have provided substantial mechanistic insight into striatal dysfunction during OCD-relevant compulsive grooming, a parallel of symptom provocation. In contrast, although OCD patients reliably show changes in PFC-dependent neurocognitive domains, there have been no studies testing higher order cognition using translational paradigms in these models. To address this gap, we recently demonstrated that SAPAP3 KOs, the most widely used and well-validated OCD mouse model, show OCD-relevant cognitive impairments in a reversal learning paradigm, in addition to the previously described compulsive grooming phenotype. This preclinical model therefore provides us with the first opportunity to identify the neural activity patterns associated with OCD-relevant compulsive grooming vs. cognitive impairments in individual PFC neurons, providing a parallel to human fMRI studies performed during symptom provocation vs neurocognitive testing. Using miniature microendoscopes to perform in vivo calcium imaging in freely-moving mice, Aim 1 will compare neural activity during reversal learning and compulsive grooming in the lOFC, an area that is critical for reversal learning but which shows hypoactivity in OCD patients during task performance. Aim 2 will compare neural activity across reversal learning and compulsive grooming paradigms in mPFC. Although mPFC is not classically associated with reversal learning in wild-type mice, our preliminary data indicate that SAPAP3 KOs that successfully acquire reversal learning have compensatory activity in mPFC. Completion of these studies will allow us to determine whether non-overlapping ensembles of neurons show hyperactivity during compulsive grooming and hypoactivity during cognitive testing, suggesting discrete inputs or outputs that could be modulated with targeted brain stimulation.

PROJECT SUMMARY/ABSTRACT Despite the fact that abnormal repetitive behaviors are prominent, disabling, and notoriously-treatment resistant symptoms of many severe childhood onset neuropsychiatric disorders such as Obsessive Compulsive Disorder (OCD), Tourette Syndrome (TS), and autism, we still have a quite limited understanding of how they are encoded in the brain. Convergent clinical studies have highlighted the importance of cortico- striatal circuits in the development of abnormal repetitive behaviors, with functional neuroimaging studies consistently demonstrating 1) symptom-associated striatal hyperactivity that is 2) resolved by effective treatment. However, it is unknown how the two major opposing cell-types of the striatum, D1 and D2-spiny projection neurons (SPNs), contribute to striatal hyperactivity during these aberrant behaviors, and how activity in these two cell types is impacted by pharmacologic treatments. Although a prevailing theory suggests that intrinsically-generated abnormal repeated motor patterns might result from either excessive activation of the D1-associated direct pathway or decreased activation of the D2-associated indirect pathway, there is little direct evidence to support this idea. To begin to dissect the contributions of D1 and D2-SPNs to striatal hyperactivity and these maladaptive behaviors, we used an animal model system that displays both hyperactivity in central striatum (CS) and perseverative actions including compulsive grooming and abnormal reversal learning (Manning et al, in prep): SAPAP3-KO mice. Using in vivo microscopy in freely moving animals, we demonstrated that SAPAP3-KOs have increased grooming-associated striatal firing rates, consistent with published work. Surprisingly, when we selectively examined D1-SPNs, contrary to expectations we saw decreased activity compared to WT at initiation of compulsive grooming events, suggesting decreased responsiveness of D1-SPNs to cortical inputs in vivo. This activity pattern was normalized by effective fluoxetine treatment. These data suggest a novel model in which decreased activity in D1-SPNs and excessive activity in D2-SPNs promotes initiation of abnormal repetitive behaviors. In this project we will use state- dependent optogenetics, in vivo microscopy, and in vivo electrophysiology to both directly test this model and determine the impact of effective fluoxetine treatment on striatal D1, D2, and FSI (fast-spiking interneuron) activity patterns. In Aim 1, we will identify D2-activity patterns during abnormal repetitive behaviors using in vivo microscopy and electrophysiology in freely-moving mice. In Aim 2, we will use in vivo microscopy to identify D1- and D2-SPN activity patterns associated with successful fluoxetine treatment, and determine whether silencing D2-SPN activity can recapitulate this normalization. In Aim 3, we will explore the relationship between FSI activity and the fluoxetine treatment response. The ultimate goal of these studies is to help refine neurostimulation-based treatment strategies for disabling perseverative and compulsive behaviors.