2018 Poster Abstracts

Donard Dwyer
LSU Health Shreveport, Department of Psychiatry

Surprising conservation of schizophrenia risk genes in lower organisms reveals their essential function and an evolutionary basis of genetic predisposition

Genetic variation undoubtedly contributes to the likelihood of developing schizophrenia, although little is known about the evolutionary factors that maintain the polygenic risk for disease or what many of the implicated genes do. To address these gaps in knowledge, we have searched for orthologs or homologous counterparts of schizophrenia risk genes in simple model organisms where genetic interactions, related phenotypes and biological function are better understood. C. elegans and zebrafish have identifiable orthologs/homologs 75% and 88% of the time, respectively, which is much higher than the rate of shared genes in general (45-60% and 70%). Thus, schizophrenia risk genes have been highly conserved during evolution, suggesting many are essential for life and could not be substantially altered by natural selection. Phenotype analysis revealed that disruption of C. elegans risk-gene equivalents causes lethality (42% of the genes) or sterility (34%) much more frequently (2-3-fold higher rate) than found in genome-wide disruption studies. Together, these findings are consistent with polygenic mutation-selection balance (inevitable bad luck) models of schizophrenia heritability. Cross-species synteny was observed, that is, C. elegans and zebrafish counterparts of multiple schizophrenia genes at a single risk locus were also located together on the same chromosome in model organisms. Proteins expressed by these syntenic blocks of genes often collectively shared a common biological purpose. We will discuss a model for schizophrenia susceptibility based on inevitable genetic hits that affect multiple genes whose function is essential at different periods of development and adaptation to the environment.Boaz Barak
Tel Aviv University

Altered neuron-glia interactions as a novel pathophysiological approach in Williams syndrome

Williams syndrome (WS) is a neurodevelopmental disorder caused by a heterozygous microdeletion of about 26 genes from chromosomal region 7q11.23, characterized by hypersociability and unique neurocognitive abnormalities. Of the deleted genes, general transcription factor II-i (GTF2I) has been linked to hypersociability in WS, though the molecular and cellular mechanisms affected by GTF2I deletion are poorly understood. To dissect the neural function of Gtf2i and its relevance to WS, we selectively deleted Gtf2i in forebrain excitatory neurons and found WS-relevant abnormalities, including neuroanatomical defects, fine motor deficits, increased sociability and anxiety. Unexpectedly, we found that in the mutant mouse cortex 70% of the genes with significantly decreased mRNA level were involved in myelination. Furthermore, we found reduced mature oligodendrocyte cell numbers, reduced myelin thickness and impaired axonal conductivity. Normalization of myelination properties rescued behavioral deficits selectively. Importantly, transcriptome analysis of human frontal cortex from WS patients similarly revealed significantly lower mRNA level of myelination-related genes, along with reduced myelin thickness and decreased mature oligodendrocyte cell numbers. Our study provides the first molecular and cellular evidence for myelination deficits in WS linked to the deletion of Gtf2i in neurons. Together these data suggest new paths to explore the neurobiological etiology of WS and potential therapeutic targets for associated social and cognitive abnormalities.

Ruchi Sood
Caribbean Medical University SOM

Improvement of Smell and Taste with Discontinuation of Buprenorphine/Naloxone

Buprenorphine/naloxone, a partial opioid agonist, has been described to induce smell and taste aversion (Lonergan et al., 2011) and impairs chemosensation (Mizera et al., 2016). Discontinuation of buprenorphine resulting in enhancing smell and taste has not heretofore been described. Two such cases are presented.

Case 1: A 36 year old, right-handed married male had a 10-year history of opioid abuse including fentanyl, acetaminophen/oxycodone and heroin. A few days prior to presentation he was on a variety of substances including 8 mg buprenorphine/2 mg naloxone every 12 hours, a fentanyl patch 100 mg every 48 hours, snorting heroin ½ gram each day, smoking marijuana daily and cigarettes one pack per day. He was undergoing withdrawal manifested by insomnia, fatigue, anxiety, and poor appetite.

Case 2: A 34 year old right-handed married female with 10 year history of opioid dependence (with past hospitalization for detoxification) presented using 100 microgram patch of fentanyl every 48 hours, snorting heroin 0.5 grams everyday, and smoked one pack per day of cigarettes, with complaints of impaired memory.

Case 1: Clinical Opiate Withdrawal Scale: 21 including diaphoresis with sweat streaming off face, constant rhinorrhea, lacrimation, vomiting, diarrhea, and frequent adventitious movements. After being placed on buprenorphine/naloxone sublingual 4 mg/1 mg twice a day, he observed a total absence of his ability to smell and taste. Within 2 days of suddenly discontinuing buprenorphine/naloxone, his smells and taste returned to 50% of normal. Within a week of him restarting buprenorphine/naloxone, his ability to smell and taste disappeared again.

Case 2: Clinical Opiate Withdrawal Scale: 8 including rhinorrhea, lacrimation, pupils moderately dilated, increased irritability and anxiousness and mild diffuse discomfort.
After being placed on buprenorphine/naloxone 8 mg/2 mg twice a day, she observed a reduced ability to smell and taste to 70% of normal. Within several days of stopping buprenorphine/naloxone, her smell and taste returned to 95% of normal. After restarting the buprenorphine/naloxone her smell and taste dropped down again to 70% of normal. Within a week of her restarting buprenorphine/naloxone, her ability of smell and taste disappeared again.

The use of opiates has been reported to alter taste (Schiffman, 2015) and reduce smell (Lotsch et al., 2012). Mizera specifically listed buprenorphine/naloxone as an origin for chemosensory loss (Mizera et al., 2016). However, the discontinuation of buprenorphine/naloxone has not previously been described to improve smell and taste. Maybe the reduction in olfactory function was partial and due to a reduction in specific G protein-coupled receptors (GPCRs) with reduced cAMP as the second messenger (Lotsch et al., 2012). Given the above, a trial of buprenorphine/naloxone in those with hyperosmia and hypergeusia may be warranted.”

Naveen Nagarajan
Department of Human Genetics, University of Utah

Corticostriatal circuit defects in Hoxb8 mouse model of OCD behaviors

Hoxb8 mutant mice exhibit repetitive behaviors for the life-time. As, in the mouse brain, the only detectable cells that label with Hoxb8 cell lineage appear to be microglia, we suggested that defective microglia cause the repetitive behaviors in neuropsychiatric conditions. Does the Hoxb8 mutation in microglia lead to neural circuit dysfunctions? We demonstrate that Hoxb8 mutants contain corticostriatal circuit defects. Golgi staining, ultra-structural and electrophysiological studies of mutants reveal excess dendritic spines, pre- and postsynaptic structural defects, long-term potentiation and miniature postsynaptic current defects. Hoxb8 mutants also exhibit hyperanxiety and social behavioral deficits similar to mice with neuronal mutations in Sapap3, Slitrk5 and Shank3, reported models of OCD and autism spectrum disorders (ASDs). Long-term treatment of Hoxb8 mutants with fluoxetine, a serotonin reuptake inhibitor, reduces excessive grooming, hyperanxiety and social behavioral impairments. These studies provide linkage between the neuronal defects induced by defective Hoxb8-microglia and neuronal dysfunctions directly generated by mutations in synaptic components that result in mice, which display similar pathological repetitive grooming, hyperanxiety and social impairment deficits. Current experiments focus on optogenetically manipulating Hoxb8 microglia and directly measure calcium signal within Hoxb8 microglia using novel calcium sensors in awake behaving state in order to test the hypothesis whether Hoxb8 microglia activation leads to grooming behavior and elicit calcium transients. Our results shed light on Hoxb8 microglia-driven circuit-specific defects and therapeutic approaches that will become essential to developing novel therapies for neuropsychiatric diseases such as OCD and ASDs with Hoxb8-microglia being the central target. 1608 2018-05-21 12:34:50 https://mpapmg1.wpenginepowered.com/poster-abstract-submission/ Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.162 Safari/537.36

Xiu Liu
University of Mississippi Medical Center

Changes of nAChRs and mGluRs in amygdala and medial prefrontal cortex associated with cue-induced nicotine seeking behavior in rats

Exposure to smoking-associated environmental cues critically contributes to relapse to tobacco use in abstinent subjects. Our previous work has demonstrated that pharmacological blockade of nicotinic acetylcholine receptors (nAChRs) attenuates cue-induced reinstatement of nicotine-seeking behavior (Liu et al, 2007; Liu 2014). The present study examined expression of the nAChRs and the metabotropic glutamate receptors (mGluRs) in the brain of rats that showed cued nicotine-seeking in the response-reinstatement model of relapse. Male Sprague-Dawley rats were trained in daily 1-h sessions to intravenously self-administer nicotine (0.03 mg/kg/infusion, free base) on an FR5 schedule. A nicotine-conditioned cue was established by associating an auditory/visual stimulus with each nicotine delivery. After lever-press responding was extinguished by withholding nicotine delivery and its cue presentation, the reinstatement tests were performed with response-contingent re-presentation of the cue without nicotine availability. Thirty minutes after the test session, brains were collected and prepared for western blot analysis of expression of the α4β2 and α7 subtypes of the nAChRs and expression of the mGluRs 2/3, 5, and 7. Re-exposure to nicotine cue effectively reinstated extinguished responses on the previously nicotine-reinforced lever, indicating the conditioned incentive properties of the cue. Compared to the rats under extinction condition, the cue-reinstated nicotine-seeking animals showed changes in expression of the α4β2 and α7 nAChRs and the mGluRs 2/3, 5, and 7 in the amygdala and medial prefrontal cortex. These data suggest that the unique change of some subtypes of the nAChRs and the mGluRs may be responsible for cue-induced reinstatement of nicotine-seeking behavior. This work sheds light on fully understanding neurobiological basis of the conditioned incentive motivation by nicotine-associated environmental cues. 1610 2018-05-24 07:53:57 https://mpapmg1.wpenginepowered.com/poster-abstract-submission/ Mozilla/5.0 (Windows NT 6.1; WOW64; Trident/7.0; rv:11.0) like Gecko

Vincent Millischer
Department of Molecular Medicine and Surgery (MMK), Karolinska Institutet, Stockholm, Sweden
Neurogenetics Unit, Center for Molecular Medicine, Karolinska University Hospital, Stockholm, Sweden

SLC1A2 Variant Leading to Elevated Anterior Cingulate Cortex Glutamate Is associated with Rapid Cycling

Glutamate is the principal excitatory neurotransmitter in the central nervous system and glutamatergic dysregulation is implicated in the underlying neurobiology of mood disorders. We have previously reported the association of two SNPs (rs3812778/rs3829280), in perfect linkage disequilibrium (r2=1), in the 3’ untranslated region of the EAAT2 gene SLC1A2 with anterior cingulate cortex glutamate levels measured by 2-dimensional J-resolved averaged PRESS sequence, minor allele carriers having significantly higher levels of glutamate in comparison to common allele homozygotes.

Here we report rs3812778/rs3829280 being associated with rapid cycling in a large heterogeneous group of depressed patients (n=1931). Minor allele carriers were significantly less common among patients with non-rapid-cycling bipolar disorder and patients with unipolar depression than among patients with rapid-cycling bipolar disorder. In silico analysis revealed an association of minor allele carriers with an upregulation of the astrocytic marker CD44 localized right next to SLC1A2 on chromosome 11.

CD44 knock-out mice having markedly reduced glutamatergic signaling. We hypothesize that SNPs in LD with rs3829280/rs2812778 are associated with augmented glutamatergic neurotransmission through an upregulation of CD44, potentially leading to a higher risk of rapid cycling and consequently a more pernicious course of disease with poorer prognosis. Furthermore, our data suggests that biological differences might not be driven by established diagnostic criteria, but rather by clinical subphenotypes like the presence of rapid-cycling.

Marta Vieira

Synaptic network dysfunction in neurological disorders: Understanding the impact of rare genetic variants of NMDARs in autism-spectrum disorders and epilepsy

Neurological disorders, including autism-spectrum disorders (ASDs) and epilepsy (EPI) are highly prevalent, complex syndromes, with genetic and environmental components. Intensive research efforts have been made to identify the causative genes. Interestingly, selective clusters of genes with copy number variations and rare variants have been identified. One such cluster is the synaptic protein network, which suggests that proteins related to synaptic function and activity may be dysfunctional in these pathologies. One of the recurrently affected genes in ASDs is GRIN2B, which encodes for GluN2B, a subunit of N-methyl-D-aspartate receptors (NMDARs). Indeed, many rare variants of GluN2B have been identified in ASD probands. The GluN2A subunit, on the other hand, is commonly identified in patients with EPI. This suggests that NMDAR dysfunction may be relevant for these neurological disorders.
We screened a variety of rare variants within the C-terminal domain (CTD) of GluN2A and 2B subunits and their effects on receptor function, spine density, as well as on the molecular complex associated with the receptors. Because the NMDAR CTDs mediate protein-protein interactions with synaptic scaffolds and signaling molecules, these variants might disrupt coupling NMDAR activity to downstream pathways.
We observed that GluN2B rare variants identified in ASD patients caused dramatic changes on NMDAR and AMPAR surface expression, spine density, and the interaction with post-synaptic proteins. To assess the implications of these defects observed in primary cultures, we generated an animal model of one of these rare variants, S1413L. Our results demonstrate that indeed this rare variant of GluN2B results in defects in the post-synaptic network. We observed that the PSD fraction of the mouse hippocampus had reduced levels of the NMDAR subunits as well as MAGUKs and CaMKII. In the cortical PSD fractions, or the whole brain, we did not detect such differences relative to WT animals. Interestingly, our GluN2A rare variants, that were identified in EPI patients, also showed reduced spine density. This effect implies a common mechanism of dysfunction underlying both ASD syndromes and EPI.
Overall, our work supports a model in which dysfunction of the synaptic protein complex underlies defects seen in ASD and EPI. Elucidating these mechanisms will pave the way to the development of novel therapeutic strategies for these disorders.

Annie Kathuria
Massachusetts General Hospital, Harvard Medical School

The role of interneurons in Schizophrenia: a stem cell approach

Schizophrenia (SCZD) is a crippling neurological disorder with a world-wide prevalence of 1%. Cognitive impairments is the most important predictor of functional outcomes in patients with schizophrenia. However, efficacious treatment of cognitive deficits in psychotic disorders remains a significant challenge in clinical practice. Even though, antipsychotic medications provide symptom relief by reducing hallucinations, they do not improve the cognitive deficits that is the core feature in schizophrenia. There is an urgent need for new therapeutic approaches that target the neurobiology of cognitive impairments. Our research focuses on developing stem cell-based models to study the molecular and cellular basis of schizophrenia using iPSCs (Induced pluripotent stem cells) generated from patients. We are investigating the functional connection between excitatory and inhibitory cortical neurons derived from patient and healthy iPSCs. Currently, we have identified decrease in synaptic co-localisation and spine density in the patient lines. Therefore, focusing our research on high-throughput screens of small molecule libraries to discover compounds that can normalize/modulate cellular disease signatures.

Ghanshyam Pandey
University of Illinois at Chicago

Innate immunity in the postmortem brain of depressed and suicide subjects: role of Toll-like receptors

Abnormalities of Toll-like receptors (TLRs) have been implicated in the pathophysiology of depression and suicide. Interactions of TLRs with pathogen-associated molecular patterns (PAMP) and damage-associated molecular patterns (DAMP) initiate signaling through myeloid differentiation primary response-88 (MyD88) and produce cytokines through the activation of the transcription factor nuclear factor kappa beta (NF-kB). We have earlier shown an increase in the protein and mRNA expression of TLR3 and TLR4 in the prefrontal cortex (PFC) of depressed suicide (DS) subjects compared with normal control (NC) subjects. To examine if other TLRs are altered in postmortem brain, we have now determined the protein and mRNA expression of other TLRs (TLR1, TLR2, TLR5, TLR6, TLR7, TLR8, TLR9 and TLR10) in the PFC of DS, depressed non-suicide (DNS), non-depressed suicide (NDS) and NC subjects. We determined the protein expression by Western blot and mRNA expression levels by real-time PCR (qPCR) in the PFC of 24 NC, 24 DS, 12 DNS and 11 NDS subjects. Combined with our previous study of TLR3 and TLR4, we found that the protein expression of TLR2, TLR3, TLR4, TLR6 and TLR10, and mRNA expression of TLR2 and TLR3 was significantly increased in the DS group compared with NC group. This study demonstrated that certain specific TLRs are altered in DS subjects, and hence those TLRs may be appropriate targets for the development of therapeutic agents for the treatment of suicidal behavior.

Supported by: NIMH RO1 MH098554

Naveen Nagarajan
University of Utah

Corticostriatal circuit defects in Hoxb8 mouse model of repetitive behaviors

Hoxb8 mutant mice exhibit repetitive behaviors for the life-time. As, in the mouse brain, the only detectable cells that label with Hoxb8 cell lineage appear to be microglia, we suggested that defective microglia cause the repetitive behaviors in neuropsychiatric conditions. Does the Hoxb8 mutation in microglia lead to neural circuit dysfunctions? We demonstrate that Hoxb8 mutants contain corticostriatal circuit defects. Golgi staining, ultra-structural and electrophysiological studies of mutants reveal excess dendritic spines, pre- and postsynaptic structural alterations, miniature postsynaptic current, synaptic transmission and long-term potentiation defects. As a manifestation of synaptic structural and functional defects, Hoxb8 mutants also exhibit hyperanxiety and social behavioral deficits similar to mice with neuronal mutations in Sapap3, Slitrk5 and Shank3, reported models of OCD and autism spectrum disorders (ASDs). Long-term treatment of Hoxb8 mutants with fluoxetine, a serotonin reuptake inhibitor, reduces excessive grooming, hyperanxiety and social behavioral impairments. These studies provide linkage between the synaptic defects induced by defective Hoxb8-microglia and neuronal dysfunctions directly generated by mutations in synaptic components that result in mice, which display similar pathological repetitive grooming, hyperanxiety and social impairment deficits. Current experiments focus on optogenetically manipulating Hoxb8 microglia and directly measure calcium signal within Hoxb8 microglia using novel calcium sensors in awake behaving state in order to test the hypothesis whether Hoxb8 microglia activation leads to grooming behavior and on how Hoxb8 microglia scans synaptic environment to induce distinct patterns of calcium transients. Our results shed light on Hoxb8 microglia-driven circuit and synapse-specific defects and therapeutic approaches that will become essential to developing novel therapies for neuropsychiatric diseases such as Autism Spectral disorders (ASDs) with Hoxb8-microglia being the central target.

Shravika Lam
Johns Hopkins University

Upregulation of microRNA-124 Contributes to Overlapping Phenotypes Between Schizophrenia and Bipolar Disorder

Our lab takes a phenotype-centered multimodal approach for mental disorders to discover critical mediators for the pathophysiology. One of the key resources for this approach is biopsied tissues and cells from patients under active psychiatric manifestations compared with those from matched controls. Through a molecular profiling study with olfactory neuronal cells, our lab discovered significant upregulation of a microRNA, miR-124, in patients with schizophrenia (SZ) as well as those with bipolar disorder (BP), when compared to healthy controls. Compilation and examination of polygenic risk scores (PRS) indicated that the PRSs for both SZ and BP significantly contribute to miR-124 upregulation. We examined the functional consequence of miR-124 dysregulation considering common behavioral dimensions that are impaired in both SZ and BP in a mouse model that overexpresses miR-124 in the pyramidal neurons of the medial prefrontal cortex. We observed behavioral deficits in sociability, social novelty recognition, and psychostimulant sensitivity in the model. Furthermore, induced alteration in excitatory synaptic transmission properties, reflected by the increased amplitude of miniature excitatory synaptic currents (mEPSCs), was also observed. Bioinformatics analyses indicate that these deficits may be a result of miR-124-mediated suppression of GluA2 expression, which we experimentally confirmed. Application of the calcium permeable AMPA-R antagonist Naspm rescued the aforementioned behavioral deficits and normalized potentiated excitatory synaptic transmission following miR-124 overexpression. Altogether, the present study shows that polygenic contribution, substantially shared across SZ and BP, converges onto dysregulation of miR-124, which in turn disturbs excitatory synaptic transmissions and causes symptomatic phenotypes common to SZ and BP.

Colleen McClung
University of Pittsburgh School of Medicine

Novel diurnal patterns of gene expression in the schizophrenic brain

Schizophrenia (SCZ) is a debilitating psychiatric disorder that is associated with significant disturbances in circadian rhythms. In the current study, we utilized a time-of-death analysis of RNA sequencing data from the Common Mind Consortium, and selectively from the University of Pittsburgh brain bank, to identify and compare gene expression rhythms in the human dorsolateral prefrontal cortex (dlPFC) of SCZ subjects to comparison control subjects. We discovered that approximately 18% of the transcripts in the dlPFC are rhythmic and that many of these genes are similar to those identified in the previous microarray study from a different cortical region. Interestingly, there was only a small degree of overlap between rhythmic transcripts in control and SCZ subjects. Moreover, transcripts from SCZ subjects displayed a distinct pattern of rhythmicity, with most genes showing a peak in expression during the day and a trough at night, compared to control subjects in which transcripts peaked at various times of the day. Many of the transcripts that are only rhythmic in SCZ subjects are associated with mitochondrial function, with daytime peaks in expression matching the overall expression levels of control subjects and the nighttime trough falling below this level. Moreover, many of the changes in gene expression that have been reported in SCZ vs control subjects are found only in subjects that died at night. These data suggest that gene expression rhythms in the dlPFC of schizophrenia subjects are largely distinct from healthy controls and may result in altered transcript levels, particularly during the night.

Ralda Nehme
Broad Institute of Harvard and MIT

A human stem cell resource for modeling genetic variation

Genomic associations have elucidated a number of loci implicated in many neuropsychiatric diseases. However, the limited number of biological systems available to researchers has made it difficult to transfer these associations into causations. We established the Stanley Center Biobank (SCBB) at the Broad Institute to provide a vast resource for geneticists and biologists to band together to delineate brain function in neuropsychiatric disease. The aim of the SCBB is to derive, expand, bank, and sequence the genome of pluripotent stem cell (PSC) lines from cases and controls across a spectrum of psychiatric conditions. In addition to the various disease backgrounds, an immense effort is underway to obtain samples from major ancestral groups across the world to incorporate population genetic differences. We are using cohorts in our collection to implement “population in a dish” experimental systems that collect data on cells from hundreds of donors. We are also investigating the cis- and trans-effects of gene variants associated with neurodevelopmental disorders, illustrated by a growing collection of lines harboring a deletion of chromosome 22q11.2, which confers a substantial risk for neurodevelopmental brain disorders, including intellectual disability, schizophrenia and autism. The implementation of this resource will establish better validated human cellular models and greatly expand the scalability of translational, quantitative experimental systems.

Judy Van de Water
University of California, Davis

Antigen-driven rat model of maternal autoantibody related autism

Maternal autoantibodies reactive to proteins in the developing have been previously described in a subset of mothers of children with autism spectrum disorder (ASD), but not in mothers of typically developing children. We recently developed an active immunization rat model of maternal autoantibody related (MAR) ASD that mimic those found in the mothers of children with ASD to directly assess the pathologic significance of prenatal exposure to epitope-specific autoantibodies, and to evaluate the impact of maternal autoantibody exposure on complex, reciprocal play behavior. To generate epitope-specific autoantibodies female rats randomly assigned to MAR-ASD treatment received a series of immunizations containing the immunodominant peptide epitopes of the four primary target proteins of MAR-ASD (LDH A and B, CRMP1, and STIP1). Control females were injected with saline/adjuvant only. Autoantibody-positive females were bred, and male and female offspring were tested for autism-relevant behaviors and developmental milestones from early postnatal through adulthood. Offspring prenatally exposed to MAR-ASD antibodies emitted fewer ultrasonic vocalizations as pups at postnatal day 12 (p=<0.05), spent less overall time engaged in social interaction as juvenile and young adults (p=<0.05), and specifically spent less time in reciprocal play behavior as juveniles (p=<0.05) and engaged in more self-grooming behavior as adults (p=<0.05). The developmental trajectory of social impairments and repetitive behaviors observed parallels features of human autism and lends support to prenatal autoantibody exposure as a risk factor for ASD. Furthermore, the MAR-ASD animals had prominent IgG staining in the brains compared to controls suggesting active binding to their CNS targets.

Xiao-Hong Lu
LSU Health Sciences Center

Dosage sensitivity intolerance of Schizophrenia-associated VIPR2 microduplication manifests elevated PKA signaling and disrupted corticostriatal neurodevelopment and cognition in a novel Bacterial Artificial Chromosome transgenic mouse model

Multiple genome-wide association studies (GWAS) have identified copy number variations (CNVs) that have been convincingly shown to significantly contribute to the risk of schizophrenia, autism, and other neuropsychiatric disorders. Intriguingly, even phenotypically healthy carriers of the CNVs manifested cognitive deficits, suggesting the neurodevelopmental CNVs affect cognition and may confer disease risk. In particular, two large-scale GWAS studies pinpointed a CNV at chromosomal locus 7q36.3 in schizophrenia patients with all of the microduplications occurring within a single gene: Vasoactive intestinal peptide receptor 2 (VIPR2). To translate such a genetic vulnerability into a mechanistic and pathophysiologic insight, we have developed a series of conditional VIPR2 Bacterial Artificial Chromosome (BAC) transgenic mouse models of VIPR2 CNV with one extra (microduplication) or four extra copies of human VIPR2, and one fully humanized VIPR2 CNV mice in the murine Vipr2 null background. VIPR2 CNV mouse models recapitulate the gene expression and signaling deficits seen in human CNV carriers. Human VIPR2 microduplication (one extra copy) elicits deficits in spatial working memory, sensory gating, and social recognition memory. The behavioral deficits were preceded by elevated PKA signaling and disruption of early postnatal corticostriatal maturation. These observations highlight the vulnerability of corticostriatal maturation to early postnatal genetic perturbations. They further validate a casual pathogenic role of the dosage sensitivity intolerance of VIPR2 CNV in disruption of the normal developmental trajectory of corticostriatal neurocircuits and manifestation of cognitive deficits. These novel conditional BAC transgenic mouse models of VIPR2 CNV will significantly facilitate genetic dissection of when/where/how the genetic vulnerabilities affect development, structure and function of neural circuits and may serve as a useful tool for therapeutic development.

Andrea Ganna
Broad Institute

Genome-wide association study of school grades informs cognitive genetic architecture of six major psychiatric disorders.

Education attainment (EA) correlates strongly with psychiatric disorders—both phenotypically and genetically. The extent to which the genetic architecture of EA in individuals with psychiatric disorders differs from the genetic architecture of EA in the general population is not known. To address this, we performed a GWAS of school grades in individuals with psychiatric disorders in the iPSYCH cohort, the first of its kind.

We conducted GWAS of ninth level grades given at Danish and mathematics exit-examinations in the municipal schools in Denmark. We studied ~35000 individuals: attention deficit hyperactivity disorder (ADHD, N=4298), autism spectrum disorder (ASD, N=3466), schizophrenia (SCZ, N=1074), major depressive disorder (MDD, N=10,140), bipolar disorder (BPD, N=607), anorexia nervosa (AN, N=1798) and controls (N=11,273, i.e., without any of the above six diagnoses). School grades and psychiatric diagnoses were extracted from Danish national registers. We analyzed ~8M variants using a linear regression adjusted for age, sex, genotyping waves and ten principal components. We analyzed all samples together (with psychiatric diagnoses as a covariate) and each psychiatric group, as well as controls, separately. In addition, we analyzed ~500,000 UK Biobank (UKBB) individuals to (a) replicate genome-wide significant (GWS) loci (b) to measure the variance explained by genome-wide polygenic scores (GPS) trained on our GWASs and (c) to study how GPS for school grades correlate with GPS for psychiatric disorders in the general population. For (a) and (b) the analyzed traits were college completion and verbal and numerical reasoning (VNR) scores.

Gender and psychiatric diagnoses were strongly associated with grades: females performed better; AN scored the highest; ADHD scored the lowest. First, we performed a GWAS of Danish and math (DM), totaled to a single score per individual. We identified three GWS loci (3p21.31, 5p13.2 and 6q16.1), two were GWS in the latest EA-GWAS from SSGAC consortium. The SNP-heritability (h2) was 0.26 (SE=0.01). Genetic correlation with the previous EA-GWAS was 0.91 (SE=0.02). GPS for EA explained up to 7.5% variance (adjusted R2) in DM. In the analyses stratified by psychiatric disorders, we found one MDD-specific GWS locus (7p15.2). Maximum variance explained by EA-GPS differed between groups: controls and BPD showed >10% variance. Second, we performed a GWAS of Danish and math separately. Since Danish and math are correlated (70%), we aimed to isolate the genetic signals specific to Danish and math. Hence, we did Danish-GWAS with math as a covariate (DadjM) and math-GWAS with Danish as a covariate (MadjD). The h2 was ~0.18 in all four. We observed, however, differences in (a) GWS loci and (b) genetic correlations with psychiatric disorders. These differences were stronger in DadjM and MadjD. We identified two MadjD-specific GWS loci (6p22.1 and 11q23.2). In the UKBB individuals, we found a strong negative correlation of GPS for all psychiatric disorders with GPS for math, MadjD and IQ. On the other hand, we found a strong positive correlation (except ADHD) with GPS for EA, DM, Danish and DadjM. Phenotypes followed the same trend: college completion correlated with Danish-GPSs and VNR correlated with math-GPSs.

Our findings showed that a substantial portion of the genetic variants influencing EA phenotypes (EA, DM, Danish and DadjM) shared with psychiatric disorders, are not cognitive—but, rather, we speculate, behavioral.

Frank Middleton
SUNY Upstate Medical University

Validation of a polyomic salivary RNA test for childhood autism spectrum disorder

Background: Efforts to define molecular biomarkers of ASD have not yet resulted in an objective and reliable test. Many of these studies have been limited by a focus on single molecule types, small sample sizes, and lack of non-autistic developmental delay controls. We hypothesized that a saliva-based polyomic RNA panel might achieve enhanced performance.

Methods: 442 children were enrolled, ages 18-83 months, who were either neurotypical (n=134), or had diagnoses of ASD (n=224), or non-autistic developmental delay (DD, n=84). Comprehensive human and microbial RNA abundance was measured in saliva samples using unbiased next generation sequencing. Samples were divided into a training set (84%) and a hold-out validation set (16%), balanced for ASD and gender. The training set was used for machine learning development of a polyomic RNA-based algorithm that best distinguished ASD and non-ASD children, accounting for demographic features and medical co-morbidities. The validation set was not used in model development.

Results: In the training set (n=372; mean age 51 months; 75% male; 50% ASD), a set of 32 RNA features classified ASD with an area under the curve (AUC) of 0.87. In the hold-out validation set (n=70; mean age 50 months; 83% male; 51% ASD), the AUC was 0.88 (with 83% sensitivity and 85% specificity). We also tested the algorithm in another distinct replication cohort, (n=15 ASD children), and achieved 80% accuracy.

Conclusions: Salivary polyomic measurement of human and microbial RNA represents a novel approach that can accurately identify ASD in children and add objective support for ASD diagnoses.

Sreeja Kalapurakkel
Stanley Center for Psychiatric Genetics at Broad Institute; Harvard University

Cognitive Variation and Risk for Psychopathology in the Philadelphia Neurodevelopmental Cohort

Although recent studies using population samples have clearly demonstrated a significant role for common genetic variants in intelligence across the life span, there have been only few molecular genetic investigations of individual cognitive domains and their independent associations with risk for psychopathology. We utilized genotype and phenotype data from the Philadelphia NeuroDevelopmental Cohort (PNC), a population-based sample of over 9500 individuals from the greater Philadelphia area, ages 8-21, who received medical care at the CHOP (Children’s Hospital of Philadelphia) network, to investigate potential associations between the genetic architecture of diverse cognitive abilities in children and polygenic risk for various forms of psychopathology. Cognition data was collected using the Computerized Neurocognitive Battery, a toolkit consisting of fourteen standard neurobehavioral tasks measured for both speed and accuracy. Based on previous literature and preliminary analysis, we determined certain cognitive domains to show significant univariate heritability, including spatial reasoning, nonverbal reasoning, emotion identification, and language reasoning. We chose to further examine these associations, along with attention and verbal memory (chosen for their moderate association to executive functioning and memory respectively), for associations with risk for particular forms of psychopathology. Our analysis would go on to find significance in the association between polygenic risk for ADHD and both attention and language reasoning (r=-0.06, p= 2.99e-05 for attention; r=-0.05, p = 0.0034 for language reasoning). These results suggest that risk for ADHD is linked to reasoning and executive functioning across development.

The recent advancement of genomics into clinical neuroscience has generated a need for behavioral assessments of domains that can serve as biomarkers of psychopathology. Rapid expansion of genetics research into the global setting has sparked a need for assessments that not only capture information on heritable traits associated with psychopathology, but do so in a short time frame, can be translated across languages without losing meaningful genetic signal, and can be easily administered even in non-clinically-ascertained populations. Our analysis with the Penn Computerized Neurocognitive Battery suggests that it is possible to construct tools that meet all these criteria.

Anna Victoria Molofsky
University of California-San Francisco

Glial-encoded innate immune mechanisms of synapse remodeling during brain development

Neuronal synapse formation and remodeling is essential to central nervous system (CNS) development and is dysfunctional in neurodevelopmental diseases. Defects in synaptic pruning have been linked to psychiatric diseases including autism and schizophrenia, particularly via the function of microglia, the resident immune cells of the brain. Astrocytes, the support glia of the brain, have also been shown to drive synapse formation and remodeling during brain development. However, the molecular mechanisms that regulate communication between these two cell types during synapse formation are unknown. Here we show that the IL-1 family cytokine Interleukin-33 (IL-33) is produced by developing astrocytes and is developmentally required for normal synapse numbers and neural circuit function in the thalamus. We find that IL-33 signals primarily to microglia under physiologic conditions, that it promotes microglial synapse engulfment, and that it can drive microglial-dependent synapse depletion in vivo. In the absence of this signaling pathway, excitatory/inhibitory synaptic balance is altered and thalamic circuits become hyperexcitable. These data reveal a cytokine-mediated mechanism required to maintain synapse homeostasis during CNS development. They also emphasize the physiologic roles of immune signaling in the developing brain, with implications for understanding links between immune dysfunction and psychiatric diseases.

Gianluca Ursini
Lieber Institute for Brain Development

Convergence of placenta biology and genetic risk for schizophrenia

Early life events associated with placental pathophysiology influence later susceptibility to many adult diseases and may contribute to define the environmental context in which genes enhance risk for complex disorders like schizophrenia. Here we analyze the role of intrauterine and perinatal complications (Early Life Complications, ELCs) and placental gene expression in modulating the association of schizophrenia with genomic risk, as measured with polygenic risk scores (PRS) based on GWAS significant variants. We found that PRS interacts with ELCs on case-control status, in three independent samples from USA, Italy and Germany (n= 1693, p= 6e-05); in each sample the variance of schizophrenia explained by PRS is multiplicatively higher in the presence of a history of ELCs compared with the absence of ELCs. The relationship between PRS and ELCs is replicated in two independent samples of only cases from Germany and Japan (n=2038, p=1e-04). The gene-set based on the schizophrenia loci interacting with ELCs is highly expressed in multiple placental tissues (p<0.001) and dynamically regulated in placental samples from complicated, in comparison with normal, pregnancies (p<0.05). These differences are significantly greater in placentae from male compared with female offspring (p<10-8). GWAS SNPs marking the loci containing genes highly expressed and dynamically modulated in placenta (PlacPRS genes) drive the interaction between PRS and ELCs (p=0.002), while PRS constructed from the remaining loci do not interact with ELCs (NonPlacPRS, p=0.60). Pathways and biological functions associated with NonPlacPRS genes are reminiscent of previous analyses about schizophrenia risk-genes, while PlacPRS genes implicate an orthogonal biology, with roots in the fetal/placental response to hypoxic stress. Our data suggest that the most significant schizophrenia GWAS variants contribute to risk by converging on a developmental trajectory sensitive to ELCs and altered placental gene expression, which may underlie the male preponderance of schizophrenia and offer new insights into primary prevention.

Giovanna Punzi
Lieber Institute for Brain Development

Genome-wide expression profiling of suicide

Suicide is the 10th leading cause of death for all age group combined and is on the rise across the U.S., according to the CDC. Recent research has turned to the detection of biological markers for the development of suicide prevention approaches. The biology underling the choice of a violent method may represents a more precise feature to target in order to detect genetic signatures for the behavior at large.
RNA sequencing data from post mortem human brain (228 Caucasian patients; adults) were examined to validate, at genome-wide level of significance, previous association specifically with suicide by violent means, and to detect further candidates potentially related to the same signal.
A differential expression analysis was conducted on all gene features, correcting for diagnosis, sex, age, and qSVs, a measure of RNA integrity.
At PFDR-corr.≤0.05, minimal signal (i.e. ~10 expressed regions) arose when comparing non-suicide with suicide (all kinds of method); differentially expressed features further decreased when looking only at suicide by non-violent means. However, comparison between non-suicides and suicides specifically by violent means produced a remarkably greater list of features (i.e. over 1400 expressed regions). These results confirm our previous findings and suggest the engagement of specific signaling. Finally, gene-set analysis testing a catalog of immune genes on the genes differentially expressed in suicide by violent means points to the microglia, and complement, as the most enriched features (p=6.72e-05 and p=2.67e-05, respectively). These results confirm that classifying suicide by method is key in revealing the underlying different biology.

John Enwright
Department of Psychiatry, Translational Neuroscience Program, University of Pittsburgh Medical Center, Pittsburgh, PA

Development trajectories of PV neurons and their perineuronal nets in layer 3 of the primate dorsolateral prefrontal cortex

Multiple studies have shown that parvalbumin (PV) interneurons and perineuronal nets (PNNs), a condensed form of extracellular matrix that surrounds PV cells, are altered in the dorsolateral prefrontal cortex (DLPFC) of schizophrenia subjects. These findings include lower PV mRNA and protein levels as well as reduced labeling of PNN components. Additionally, PV neurons from schizophrenia subjects show reduced expression of several markers associated with the closure of critical periods, including reticulon 4 (RTN4) and LYNX1. One interpretation of these findings is that in schizophrenia PV neurons never fully mature, remaining in a hyperplastic state. To gain further insight into PV cell and PNN development, quantitative immunohistochemistry was performed for PV and two PNN markers (aggrecan, a core PNN glycoprotein, and WFA, a label for the glycosylation state of the PNN) in a cohort of 28 monkeys from neonates to adults. This developmental analysis shows PV neurons (as assessed by levels of PV protein) mature early, while PNN development in the DLPFC is protracted. Though PNNs are present around PV cells early in childhood; labeling of PNNs increases greatly during adolescence and reaches adult levels at the end of adolescence. Together these findings suggest that reduced PNN labeling in schizophrenia may be the result of an altered developmental trajectory during adolescence.

Zachary Freyberg
University of Pittsburgh

Neuronal depolarization drives increased dopamine synaptic vesicle loading via VGLUT

The ability of presynaptic dopamine terminals to tune neurotransmitter release to meet the demands of neuronal activity is critical to neurotransmission. Although vesicle content has been assumed to be static, in vitro data increasingly suggest that cell activity modulates vesicle content. Here we use a coordinated genetic, pharmacological and imaging approach in Drosophila to study the presynaptic machinery responsible for these vesicular processes in vivo. We show that cell depolarization increases synaptic vesicle dopamine content prior to release via vesicular hyperacidification. This depolarization-induced hyperacidification is mediated by the vesicular glutamate transporter (VGLUT). Remarkably, both depolarization-induced dopamine vesicle hyperacidification and its dependence on VGLUT2 are seen in ventral midbrain dopamine neurons in the mouse. Together, these data suggest that in response to depolarization, dopamine vesicles utilize a cascade of vesicular transporters to dynamically increase the vesicular pH gradient, thereby increasing dopamine vesicle content.

David Lloyd
Neuroscience Research Australia

Wild-type like behavioural effects of acute ∆-9 tetrahydrocannabinol (D9-THC) in Type III neuregulin 1 knockout and type III neuregulin 1 overexpressing mouse models

Introduction: The schizophrenia risk gene NRG1/Nrg1 modifies the neuro-behavioural response to psychoactive ∆-9 tetrahydrocannabinol (D9-THC). Nrg1 contains 30+ Nrg1 functionally-distinct isoforms and it was the aim of this study to determine the role of type III Nrg1 in Nrg1 X D9-THC interactions.

Method: Adult male (28-36wks) Nrg1 type III heterozygous knockout (type III Nrg1 +/-) or Nrg1 type III overexpressing (Nrg1 III tg) compared with matched wild-type (WT) (minimum n≥10/group). Mice received vehicle or 3 or 10mg/kg D9-THC and tested 30 min later for locomotion, exploration, anxiety, social interaction, social novelty preference (Nrg1 III tg only) and sensorimotor gating.

Results: Type III Nrg1 +/- (reduced Nrg1) mice showed deficits in following time in the social interaction test [F(1,70) = 4.22, p < 0.05)] and reduced PPI [F(1,69) = 4.12, p < 0.05] vs WT. Alternatively, Nrg1 III tg (increased Nrg1) mice displayed reduced social interaction time vs WT during the conditioning stage of the social novelty test [F(1,40) = 6.8, p = 0.01]. Although following was reduced by D9-THC in WT mice, this effect was less pronounced in type III Nrg1 +/- mice [F(2,70) = 2.92, p = 0.06]. No differential effects of THC were observed for Nrg1 III tg mice vs WT.

Conclusions: Neither type III Nrg1 +/- and Nrg1 III tg had strong effects on the behavioural response to acute THC. This suggests that Nrg1 isoforms other than type III may be responsible for the previously observed Nrg1 x acute D9-THC interactions.

Suma Jacob
University of Minnesota

Novel computational approach to identify neurodevelopmental anomalies related to variation in autism phenotype and potential etiologies

Novel computational approaches may help us identify phenotypic outliers and subgroups with specific developmental biomarker profiles. We utilized unsupervised anomaly detection to identify risk features that may predict autism or related disorders in early development. The data is from a community sample of 1570 children between 17- 25 months of age. Caregivers filled out Video-Referenced Rating of Reciprocal Social Behavior, the Repetitive Behavior Scale for Early Childhood (RBS-EC), and the MacArthur-Bates Communicative Development Inventories (MCDI), and demographic information online. We then were able to determine that 80 children had local outlier factor (LOF) anomaly scores > 1.32 out of the overall sample of 1570 toddlers. This represents about 5% of the sample and fits between prevalence data of 13% for developmental disabilities and the 1.5-2 % estimates for prevalence data of ASD. The longitudinal follow up subsample allowed us to determine if a subset of children (N=107) with higher LOF scores maintained higher risk symptoms when they were reassessed around 28 months of age. At follow-up, an independent assessment (ITSEA) was given to caregivers in order to assess their child’s social-emotional score. At this later time point, the subgroup with LOF >1.32 scored significantly lower (M=15.1, SD=4.1) than the larger group with LOF <1.32 (M=17.6, SD=2.0) at p=0.005 on this social-emotional measure. This indicated that the subgroup with greater LOF scores continued to show lower social impairment. Anomaly detection allowed us to detect outlier patterns and this may help us identify syndromic features associated with molecular etiologies.

Chen Tian
University of Southern California

ASD/ID related de novo mutations in the TRIO-Rac1 pathway alter synaptic function

The Rho guanine nucleotide exchange factor (RhoGEF) Trio promotes actin polymerization by directly activating the small GTPase Rac1. Growing evidence suggests that disruption of the Rac1 signaling pathway at glutamatergic synapses contributes to Autism Spectrum Disorder/Intellectual disability (ASD/ID)-related behaviors seen in animal models of ASD/ID. Here, in humans, we discover a large cluster of ASD-related de novo mutations in Trio’s Rac1 activating domain, GEF1. In accordance with pathological increases or decreases in glutamatergic neurotransmission observed in animal models of ASD/ID, we find that these mutations result in either reduced synaptic AMPA receptor expression or enhanced glutamatergic synaptogenesis. Rac1, activated by Trio’s GEF1 domain, has recently been identified as an ID risk gene. Similar to hypofunctional ASD-related mutations in Trio, an ID-related de novo missense mutation in Rac1 results in a reduction in synaptic AMPAR expression. Additionally, this mutation prevents the induction of LTP, the cellular mechanism underlying learning and memory formation. Together, our findings implicate both excessive and reduced Trio/Rac1 activity and the resulting synaptic dysfunction in ASD/ID-related pathogenesis, and point to the Trio-Rac1 pathway at glutamatergic synapses as a possible key point of convergence of many ASD/ID-related genes.

Yabin Wei

Pool-sequencing of HTR7 gene identifies common variants influencing response to antidepressants

Predicting antidepressant response has been a clinical challenge in both bipolar disorder (BD) and unipolar depression. Although several genome-wide association studies have suggested a number of genetic variations to be associated with antidepressant response, however the results are difficult to replicate. HTR7 gene is involved in the pathophysiology of depression and has been shown to interact with the action of antidepressants, providing a good candidate for the study of antidepressant treatment response. In this report, we first used a cost-effective pool-sequencing strategy to sequence the whole HTR7 gene and its regulator regions to investigate the association of common variations in HTR7 and response to four selective serotonin reuptake inhibitors (SSRIs, including citalopram, paroxetine, fluoxetine and sertraline) and lithium in a retrospective cohort mainly consist of BD. We found 80 SNPs with FDR<0.05 associated with response to paroxetine. One SNP which is located at the promoter region also showed nominal P<0.05 in fluoxetine group. Taqman assay validated the sequencing result of this SNP. Logistic regression demonstrated genetic variation of this SNP and gender were associated with response to two SSRIs (paroxetine and fluoxetine). In the second stage, we investigated if this SNP could predict antidepressant response in two prospective cohort consist of unipolar depression. We replicated the results observed in the BD cohort. Functional study showed only the higher activity allele of this SNP interacted with CEBPB transcription factor (TF) while low activity allele did not show any interaction with TF. Our results provided novel pharmacogenomic evidence to support a role of HTR7 in predicting antidepressant response.

Fatma Özlem Hökelekli
Hacettepe University, Institute of Neurological Sciences and Psychiatry

The Effect of FGF2-antisense Overexpression on Contextual Memory

Background: Antisense transcription occurs pervasively in mammalian genome. FGF2-AS is a natural antisense transcript that is transcribed from the opposite strand of fibroblast growth factor (FGF2) and thought to regulate its expression. Although FGF2’s role in conditioned fear expression and fear extinction memory is well-established, role of FGF2-AS in learning and memory is addressed in only one study. In this study FGF2-AS was found to be expressed more in prefrontal cortex(PFC) of fast learning mice. In this study, we assessed the effects of FGF2-AS overexpression in medial PFC on contextual fear memory.
Methods: 8-10 week old male Sprague-Dawley rats were used for the experiments. Either a blank adeno-associated virus(AAV) (n=13) or AAV expressing FGF2-AS (n=16) was injected bilaterally into the prelimbic and infralimbic cortices. The effects of FGF2-AS overexpression in medial PFC on contextual memory was evaluated using passive avoidance task. Latency to enter the dark compartment(LE) was assessed as a measure of contextual fear memory. Results were analyzed by Kaplan Meier survival analysis, followed by log-rank comparison. Data are mean(±SE).
Results: Control group’s mean LE was 83(±28) seconds, while it was 56(±16) seconds in the FGF2-AS overexpression 6-hours after the training(p=0.397). At 24-hours post-training, mean LE was 158(±36) seconds in the control group, whereas it was 134(±32) seconds in the FGF2-AS overexpression group(p=0.672).
Conclusion: FGF2-AS overexpression in medial PFC did not change contextual memory in passive avoidance task.

Christopher van Dyck
Yale School of Medicine, Department of Psychiatry and the Alzheimer’s Disease Research Unit

Pet Imaging Of Synaptic Density (Synaptic Vesicle Glycoprotein 2a, Sv2a) In Alzheimer’s Disease: Initial Experience

Objectives: Positron Emission Tomography (PET) imaging of glucose metabolism with 18F-fluorodeoxyglucose (18F-FDG) has been widely used to measure synaptic activity and track Alzheimer’s disease (AD) progression. However, tracers for new molecular targets are needed to directly monitor synaptic density.

Methods: Here we compared synaptic vesicle glycoprotein 2A (SV2A) density as measured by 11C-UCB-J in 10 AD (72.9±8.7 years, 11C-PiB+) and 11 cognitively normal (CN, 72.7±6.3 years, 11C-PiB–) individuals. Participants were scanned on the HRRT after bolus injection of 11C-UCB-J. The arterial input function was analyzed with the 1-tissue (1T) compartment model in 9 AD and 8 CN participants to estimate volume of distribution (VT). Using the centrum semiovale as reference region, regional BPND values (VT ROI/ VT centrum semiovale–1) were also estimated for the full sample. 9 AD and 11 CN participants were scanned with 18F-FDG to explore the relationship between SV2A and regional metabolic activity (Ki ratio).

Results: AD participants demonstrated significant reductions in hippocampal SV2A BPND (p=0.005, t-test) and VT (p=0.011). These reductions persisted after partial volume correction (BPND: p=0.020, VT: p=0.056). Exploratory analyses of other brain regions and statistical parametric mapping (SPM) also revealed reductions in entorhinal cortex. The 18F-FDG Ki ratio was significantly associated with 11C-UCB-J BPND in hippocampus (r=0.84, p=0.001).

Conclusions: 11C-UCB-J PET may provide a direct measure of synaptic density in AD. SV2A PET imaging with 11C-UCB-J holds promise as an in vivo biomarker for trials of disease-modifying therapies—particularly those that target the preservation and restoration of synapses in AD.

I wish to be considered for a brief talk in an oral symposium prior to the poster session.”

Jungyoon Kim
Ewha Brain Institute & Department of Brain and Cognitive Sciences, Ewha W. University, Seoul, South Korea

Prefrontal GABA concentration in association with cerebral perfusion in trauma-exposed individuals

The clinical heterogeneity observed in posttraumatic stress disorder (PTSD), such as resilience to the progression of PTSD and recovery from symptoms of posttrauma, is suggested to be mediated in part by the medial prefrontal cortex (mPFC). Considering the dynamic role of gamma-aminobutyric acid (GABA) in the regulation of neuronal activation, the current study investigated the neurochemical modulation in the prefrontal areas of the brain that may attribute to the clinical heterogeneity of traumatic stress. Using multimodal imaging including magnetic resonance (MR) spectroscopy and arterial spin labeling perfusion MR imaging, the current study examined 50 individuals who were recently exposed to an index trauma (25 diagnosed with PTSD [PTSD group] and 25 without [resilience-recovery group]) as well as 50 age- and sex-matched individuals who were not exposed to any trauma (control group). GABA concentration and resting cerebral blood flow (rCBF) within the mPFC were measured and compared between groups, and the association between GABA concentration and rCBF within the mPFC were examined in each trauma-exposed group. In reference to the control group, prefrontal GABA concentration was higher in the PTSD group but not in the resilience-recovery group. The GABA concentration within the mPFC were significantly associated with hypo-perfusion in the PTSD group, while the association was opposite in direction in the resilience-recovery group. The current findings suggest a distinct GABAergic modulation in relation to cerebral perfusion in the prefrontal areas, which may contribute to the underlying mechanism in the clinical heterogeneity of posttraumatic stress.

M. Mercedes Perez-Rodriguez
Icahn School of Medicine at Mount Sinai

Pharmacological Enhancement of Social Cognition in the Psychosis Spectrum

Background: Abnormalities in social cognition represent a core feature of schizophrenia spectrum disorders. Schizotypal personality disorder (SPD) is a milder disorder within the schizophrenia spectrum, characterized by attenuated, schizophrenia-like traits without overt psychosis.
Oxytocin modulates social cognition. However, oxytocin’s effect on social cognitive errors in SPD remains unexplored. We aimed to: 1) characterize social cognitive (mentalizing) errors in SPD and test their relationship with positive and negative symptoms; 2) test the effect of intranasal oxytocin on mentalizing errors.
Methods: Subjects: 15 SPD patients, 15 healthy controls [HC], and 15 psychiatric controls (PC). Intervention: intranasal oxytocin 24/40IU/placebo. Measures: Movie for the Assessment of Social Cognition (MASC), a naturalistic video task measuring mentalizing accuracy, “no mentalizing”, “hypomentalizing” and “hypermentalizing” errors. The “hyper-hypomentalizing score” was computed to capture the predominant mentalizing tendency; PANSS; Schizotypal Personality Questionnaire, SPQ. Mentalizing measures were compared across groups and treatments (oxytocin 24IU/40IU vs placebo) using ANOVA. Pearson correlations assessed the relationship between social cognition and symptoms.
Results: SPD patients had lower mentalizing accuracy (F=10.11;df=1;p=0.003), made more “No mentalizing” or “hypomentalizing” errors (F=12.92;df=1;p=0.001), and had lower hyper-hypomentalizing scores than HCs. Single-dose intranasal oxytocin significantly increased the hyper-hypomentalizing score (F=6.84,df=1,p=0.019).
“No mentalizing” and “hypomentalizing” errors were significantly correlated with negative symptoms. “Hypermentalizing” errors were significantly correlated with positive symptoms and the “ideas of reference” and “suspiciousness” SPQ subscales.
Conclusion: Our results support the role of social cognitive impairments as an underlying factor of positive and negative symptoms of psychosis. Intranasal oxytocin may ameliorate social cognitive errors in the psychosis spectrum.”

Summer Thyme
Harvard University

Phenotypic landscape of schizophrenia-associated genes defines candidates and their shared functions

Genomic studies have identified hundreds of candidate genes near loci associated with risk for schizophrenia. To define candidates and their functions, we mutated zebrafish orthologues of 132 human schizophrenia-associated genes and created a phenotype atlas consisting of whole-brain activity maps, brain structural differences, and profiles of behavioral abnormalities. Phenotypes were diverse but specific, including altered forebrain development and decreased prepulse inhibition. Exploration of these datasets identified promising candidates in more than 10 gene-rich regions, including the magnesium transporter cnnm2 and the translational repressor gigyf2, and revealed shared anatomical sites of activity differences, including the pallium, hypothalamus or tectum. Single-cell RNA sequencing uncovered an essential role for the understudied transcription factor znf536 in the development of forebrain neurons implicated in social behavior and stress. This phenotypic landscape of schizophrenia-associated genes prioritizes more than 30 candidates for further study and provides hypotheses to bridge the divide between genetic association and biological mechanism.

Leon French
Centre for Addiction and Mental Health

Transcriptomic characterization of the human habenula highlights drug metabolism and the neuroimmune system

Due to size and accessibility, most information about the habenula is derived from rodent studies. To better understand the molecular signature of the habenula we characterized the genes that have high expression in the habenula. We compared anatomical expression profiles of three normal adult human brains and four fetal brains. We used gene set enrichment analyses to determine if genes annotated to specific molecular functions, cellular components, and biological processes are enriched in the habenula. We also tested gene sets related to depression and addiction to determine if they uniquely involve the habenula. As expected, we observed high habenular expression of GPR151, nicotinic cholinergic receptors, and cilia-associated genes (medial division). Genes identified in genetic studies of smoking and associated with nicotine response are enriched in the habenula. Cannabis receptor 1 and genes associated with cannabis use are enriched in the lateral division. Genes associated with major depressive disorder do not have enriched expression in the habenula but genes negatively correlated with hedonic well-being are, providing a link to anhedonia. We observed enrichment of genes associated with diseases that are comorbid with addictions (hematopoiesis, thrombosis, liver cirrhosis, pneumonia, and pulmonary fibrosis) and depression (rheumatoid arthritis, multiple sclerosis, and kidney disease). These inflammatory diseases mark a neuroimmune signature that is supported by genes associated with mast cells, acute inflammatory response, and leukocyte migration. We also found enrichment of cytochrome p450 genes suggesting the habenula is uniquely sensitive to endogenous and xenobiotic compounds. Our results suggest the habenula receives negative reward signals from immune and drug processing molecules. This is consistent with the habenular role in the “anti-reward” system and suggests it may be a key bridge between autoimmune disorders, drug use, and psychiatric diseases.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 34″][vc_column_text]Cameron Pernia
Sanford Burnham Prebys Medical Discovery Institute

Hyper-excitable neurons produce hypo-functional neuronal networks: Evidence from modeling bipolar disorder with mouse models and hiPSCs

Bipolar disorder (BD) is a neuropsychiatric disease that impacts 2.6% of the adult population, and is characterized by oscillations in depressive and manic behavior. BD is the most fatal psychiatric disease due to a high suicide rate, and little is known regarding its underlying pathology. Currently there is no therapy that is both safe and efficacious for treating BD, which is a critical unmet need. Recent discoveries utilizing transgenic mouse models have demonstrated collapsin response mediator protein-2 (CRMP2) plays an integral role in BD’s molecular pathology, but how CRMP2 mediates BD has yet to be elucidated. Employing CRMP2 transgenic mice as models for BD, we have discovered CRMP2 activity impacts neuronal electrophysiology, structure, and proteomics. Interestingly, many of the aberrations found in the transgenic CRMP2 neurons superficially appear counter-intuitive, but under further examination expose the complexity of how neuronal circuits function. Specifically, BD-like transgenic CRMP2 neurons appear to have hyperactive calcium activity, while having less neuronal-network signaling. Collectively, these works begin to illuminate long sought-after insights in BD pathology, and offer new targets for future BD therapeutics to be designed for.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 35″][vc_column_text]Paulo Lizano
Beth Israel Deaconess Medical Center and Massachusetts General Hospital

A role for retinal imaging and stem cell derived brain endothelial cells for evaluation microvascular dysfunction in psychosis

The neurobiology of schizophrenia (SZ) comprises excessive and progressive reduction in cortical thickness (CT) and cerebral blood flow (CBF) over widespread areas of the brain. Since neurogenesis and vasculogenesis occur in a coordinated fashion, a defect in neurovascular development could result in poor functionality of neurons in SZ. To overcome the limitations in brain imaging (i.e. poor neuronal and vascular resolution) and the lack of proper cellular models (i.e. lack of patient derived stem cells), we are using non-invasive retinal imaging methods and human induced pluripotent stem cells (iPSC) to study these properties. Optical coherence tomography (OCT, for retinal layers) and OCT angiography (OCTA, for vessel density, tortuosity, and permeability) represent significant improvements on brain imaging techniques and can enhance our understanding of brain pathology. Additionally, the recent development of efficient protocols for the generation of brain microvascular endothelial cells (BMEC) from iPSCs provide a new opportunity to investigate the primary effects of BMECs on psychosis pathogenesis by examining morphological (tight junction fidelity), physiological (barrier function, transporter activity, and migration), molecular (transcriptomic) and the effects of medications on these phenotypes. This proposal aims to investigate the utility of OCTA and BMEC in studying the neurobiology of psychosis.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 36″][vc_column_text]Alfredo Oliveros
Mayo Clinic

Forebrain-specific expression of secreted Frizzled-related Protein 3 Impairs Neurogenic Regulated Memory Function

Wnt signaling is implicated in a number of neurodevelopmental and neuropsychiatric disorders including depression, schizophrenia and autism. Our prior study identified the endogenous Wnt signaling inhibitor Secreted Frizzled Receptor Protein subtype 3 (sFRP3), as a regulatory gene that is significantly reduced by antidepressant treatment (i.e. fluoxetine, electroconvulsive therapy). We found that genetic deletion of sFRP3 promotes antidepressant action in mice and humans, potentially increasing adult hippocampal neurogenesis. To determine whether sFRP3 functionally affects adult hippocampal neurogenesis and cognitive function in our current study, we utilized a forebrain-specific sFRP3 tetracycline inducible overexpression mouse model (sFRP3 OX) and found significant decreases in hippocampal dentate gyrus (DG) size, suggestive of impaired DG formation. Confirming these results, sFRP3 OX mice exhibited decreased neural progenitor proliferation and neuronal differentiation without exhibiting neuronal cell death in the DG, indicating impaired development induced reductions in DG formation. Moreover, we also demonstrate that sFRP3 OX impairs dendrite development of newborn neurons. Functionally, sFRP3 OX mice exhibited dysfunctional performance during the Y-maze and T-maze assays, suggestive of cognitive impairment. Taken together, our results reveal that the Wnt-signaling inhibitor sFRP3 is a critical regulator of this pathway, exerting significant control over neurogenic maturation and memory function. Our findings provide key evidence for development of effective therapeutics targeting neurological or neuropsychiatric conditions typifi[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 37″][vc_column_text]Ozge Oztan
Stanford University

Cerebrospinal fluid vasopressin concentration differentiates cases and controls and predicts symptom severity in children with autism

Autism is a brain disorder characterized by social impairments and the presence of repetitive behaviors. Progress in understanding and treating autism has been hindered, in part, by the extraordinary difficulty in obtaining brain-relevant tissues [e.g., cerebrospinal fluid (CSF)] by which to study disease biology more directly. Here we bridge this barrier to progress by testing whether CSF concentrations of the “social” neuropeptides arginine vasopressin (AVP) and oxytocin (OXT) differentiate cases and controls and predict symptom severity in the largest pediatric cohort studied for this purpose to date. Children with autism were diagnosed based on clinical criteria which was confirmed with research diagnostic methods. Cases and controls were matched 1:1 on sex and within a 1-year band on age (combined cohort: N=72; N=48 males, N=24 females, aged 1.5 to 9 years). CSF AVP concentration was lower in the autism vs. control group and differentiated individual cases from controls. Lower CSF AVP concentration also predicted greater severity of social, but not repetitive, behavior symptoms in children with autism. Finally, these findings were more pronounced in males vs. females with autism, and were specific to AVP, as no evidence implicating the structurally related neuropeptide OXT was found. As AVP preferentially regulates social behavior in male mammals, the present findings suggest that brain-related AVP signaling deficits may be relevant to better understanding the male-biased risk for, and symptom presentation in, autism. These findings also suggest that AVP may be a promising CSF marker of, and potential therapeutic agent for, autism’s social deficits.

I would like to be considered for a brief talk in an oral symposium.”[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 38″][vc_column_text]Songhee Jeon
Department of Biomedical Sciences, BK21 PLUS Center for Creative Biomedical Scientists at Chonnam National University

Cognition impairment and depressive behavior in the X-linked moesin knock out mice

Moesin is a cytoskeletal adaptor protein that plays an important role in modification of the actin cytoskeleton. In neurons, moesin knockdown is involved in the morphogenesis of mushroom body axons and long-term memory in Drosophila. However, moesin knockout (KO) mice have not been evaluated for behavioral phenotypes or brain development. Here, we made moesin KO mice by using CRISPR/CAS9 system and examined it. Moesin KO mice showed depressive behaviors in the open filed test and cognition impairment in the passive avoidance test compared to that of wild type. Moreover, Golgi impregnation revealed abnormal neuron structures in cortex and hippocampus of moesin lacking mice. Western blot analysis confirmed the absence of moesin and revealed up or down regulation of various synaptic proteins such as SNPA25, Synataxin1A, synaptophysin, Munc18, amphiphysin in the cortex and hippocampus of moesin KO mice. Thus, these suggest that moesin plays an essential role in synaptic structure and function, affecting a subset of cognitive processes.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 39″][vc_column_text]Tasnim Rahman
Neuroscience Research Australia, Australia
School of Psychiatry, University of New South Wales, Australia.
Schizophrenia Research Institute, Australia

Deficits In Bdnf Promoter Mrnas Within Orbitofrontal Cortex In Schizophrenia

People with schizophrenia with high gene expression of inflammatory markers – classified as “high inflammation” – have altered inhibitory interneuron and promoter-specific BDNF transcripts compared to those with low gene expression of inflammatory markers in dorsolateral prefrontal cortex (dlPFC). There are similar alterations in inflammatory markers in orbitofrontal cortex (OFC) in people with schizophrenia. We investigated if: (1) deficits in BDNF promoter mRNAs extend to OFC; and (2) if these deficits vary with the gene expression of inflammatory markers.
BDNF promoter I-IX, II-IX, IV-IX, and VI-IX mRNAs were measured by qPCR, normalised, and compared using ANCOVAs (covariates identified via stepwise regression: PMI, age, RIN) by diagnosis (38/38) or subgroups clustered by inflammatory marker gene expression (“low inflammation” controls: 33; “low inflammation” schizophrenia cases: 27; and “high inflammation” schizophrenia cases: 11).
Diagnostically, BDNF I-IX (F(1,72)=0.574) and VI-IX (F(1,70)=0.339) mRNAs were unchanged (p>0.05), whereas BDNF II-IX (F(1,71)=10.647) and IV-IX (1,71)=11.443) mRNAs were reduced in schizophrenia cases (p<0.01). By inflammatory subgroups, BDNF VI-IX mRNA was unchanged (F(2,64)=0.630, p>0.05), whereas I-IX mRNA was reduced in “high inflammation” schizophrenia cases (F(2,66)=4.385) versus both “low inflammation” controls and “low inflammation” schizophrenia cases (p<0.05). BDNF II-IX (F(2,65)=10.103) and IV-IX (F(2,65)=13.154) mRNAs were reduced in both schizophrenia subgroups versus “low inflammation” controls, with exacerbated reductions in “high inflammation” schizophrenia cases versus “low inflammation” schizophrenia cases (all p<0.05).
This indicates that people with schizophrenia have reduced neurotrophic support that is exacerbated with elevated cytokines in dlPFC and OFC. This dysregulation may have detrimental effects on inhibitory neurons in schizophrenia.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 40″][vc_column_text]Erin Flaherty
Icahn School of Medicine

Functional evaluation of the neuronal impact of patient-specific aberrant NRXN1α splicing

Heterozygous deletions in NRXN1, a presynaptic cell-adhesion protein essential for synaptic function, are strongly associated with multiple neuropsychiatric and neurodevelopmental disorders. Although highly alternatively spliced in mice, the diversity of the NRXN1 splice repertoire has not been catalogued in human brains, and the extent to which this splicing pattern is recapitulated in vitro by human induced pluripotent stem cell (hiPSC)-derived neurons is unclear. By integrating targeted single-molecule long-read and short-read sequencing approaches, we catalogued 202 translatable human NRXN1α isoforms. We further demonstrated that hiPSC-neurons expressed at least 103 NRXN1α isoforms, recapitulating 63% of the NRXN1α transcriptional repertoire we detected in the human brain. Using a cohort of hiPSC-neurons from psychosis cases with rare heterozygous intragenic NRXN1 deletions, we observed reduced expression of 49 wildtype NRXN1α isoforms and identified 41 novel isoforms produced from the mutant allele. NRXN1+/- hiPSC-neurons exhibited decreased neuronal activity, which could be partially ameliorated by increasing the expression of a single control NRXN1α isoform, and partially recapitulated in controls by overexpression of a single mutant NRXN1α isoform. We directly link aberrant neuronal NRXN1α alternative splicing in psychosis cases to perturbations in neuronal function.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 41″][vc_column_text]Shanshan Zhu
Johns Hopkins University/Department of Psychiatry

Interplay of Ankyrin-G with L-type voltage gated calcium channel (Cav1.2): potential pathogenic pathway for psychosis

Genome Wide Association Studies have identified loci at ANK3 (Ankyrin-G) and CACNA1C (Cav1.2) as risk factors for schizophrenia and bipolar disorder. Understanding functional interactions of psychiatric risk factors may help clarify pathways relevant to psychosis. Ankyrin-G is a cytoskeletal protein that organizes axon initial segment (AIS) in neurons. We have previously developed a conditional mouse knockout of Ankyrin-G in the forebrain, with behavior reminiscent of human bipolar disorder. Cav1.2 is the alpha subunit of the L-type voltage-gated calcium channel. Using immunoprecipitation and immunofluorescence, we detected interaction and co-localization between Ankyrin-G and Cav1.2 calcium channel in vitro and during development in vivo. Ankyrin-G has been previously shown to be degraded by calpain. We hypothesized a pathway: Activation of Cav1.2 calcium channels increases calcium entry, which activates calpain, leading to degradation of Ankyrin-G. We tested this hypothesis by applying positive and negative modulators on primary cultured neurons followed by quantifying AIS Ankyrin-G immunolabeling. The calcium channel agonist (Bay K8644) decreased Ankyrin-G levels. The effect could be blocked by a calcium channel inhibitor (Nimodipine), a calpain inhibitor (MDL 28170), or a calcium ion chelator (BAPTA-AM). Primary neurons from a mouse model expressing a GFP-ankyrin-G fusion protein, which labels the AISs yielded similar results. These data provide initial evidence for a pathway involving L-type calcium channels, calpain and Ankyrin-G. Our animal and cell models may be useful tools to study cellular interactions of two gene products related to psychosis, explore mechanisms of known psychiatric drugs, and potentially to develop new therapeutic strategies.

Rebecca Mok
University of Toronto

Altered connectivity in Rett syndrome stem cell-derived cortical neural networks

Induced pluripotent stem cells (iPSCs) provide easily accessible and renewable sources of human cells for in vitro disease modeling applications. We use iPSCs to generate neurons for morphological and electrophysiological study from individuals with the neurodevelopmental disorder Rett syndrome (RTT). RTT is caused by heterozygous mutations in the X-linked gene MECP2, a transcriptional and translational regulator in neurons. We previously reported that RTT neurons have reduced soma area and dendrite length, along with impaired action potential firing. Based on these findings in individual cells, we hypothesize that RTT neurons have compromised ability to form connections with neighbouring neurons, leading to altered connectivity on the network level.

To examine neural circuit formation and network synchronicity, we used micro-electrode arrays (MEAs) to record extracellular voltage changes in monolayer cultures twice per week. We compared two isogenic pairs of control and MECP2-null excitatory cortical neurons differentiated by inducible Ngn2 expression. We showed that RTT neurons have reduced network burst frequency compared to controls (6 weeks, 4 replicates), which was diminished by AMPAR antagonist treatment, indicating synaptic transmission was responsible for bursting. We developed spatial-temporal analysis to evaluate network formation kinetics and map connectivity nodes. We computed spike count correlations at given electrode positions, and preliminary results reveal that correlations decay faster as a function of distance in RTT relative to control networks. Our results identify alterations in RTT neural network development using computational MEA approaches, which will form the basis of a drug testing platform aimed at rescuing the observed functional phenotypes.

Alannah Miranda
University of California, San Diego

Rare NTRK1 variant associated with mood disorders affects neural stem cell development

Bipolar disorder (BP) is a neuropsychiatric disorder that is characterized by a fluctuation between depressive and manic phases. In previous studies, genetic data was collected from three generations of a family in which mood disorders, including BP, cosegregated with a rare kidney disease, medullary cystic kidney disease (MCKD). Whole genome sequencing and Sanger sequencing identified a mutation in NTRK1 which cosegregated with the causative mutation for MCKD. The variant identified, rs144901788, causes an amino acid change from glutamate to lysine at position 492. This is located in close proximity to the SHC binding site at tyrosine 490 (Y490). TrkA Y490 autophosphorylation occurs in response to nerve growth factor (NGF) binding, which has been shown to play a role in cell survival, neurite outgrowth and neural development. This study uses neural stem cells (NSCs) derived from induced pluripotent stem cells (iPSCs), reprogrammed from lymphoblasts. We hypothesized that the mutant NSCs would show deficits in neural development when compared to control NSCs. qPCR analysis showed that mutant NSCs had a decrease in NTRK1 gene expression, relative to control NSCs. Cell growth assays demonstrated that mutant NSCs grew at a slower rate when compared to control NSCs. This study demonstrates a distinct difference in the development of the mutant NSCs. Future work includes understanding through which mechanism these changes occur. Understanding how NTRK1 mutations may affect NSCs, may allow for establishing distinct sub-forms of illness that operate through different pathways, yet ultimately culminating in a final disease presentation.

Neil Gray
University of Hawaii, John A. Burns School of Medicine

Comparative Gene Expression of Serotonergic, Dopaminergic, and Noradrenergic Neurons

The monoamine neurotransmitter systems play a variety of roles in the central nervous system and are the targets of many neuropsychiatric medications. In this study, we used manual sorting of fluorescently-labeled neurons to obtain purified monoaminergic neurons from the dorsal raphe nuclei, the substantia nigra pars compacta, and locus coeruleus. We then used mRNA amplification and Affymetrix Exon 2.0 microarrays to quantify gene expression, using whole brain homogenate as a reference to find highly enriched genes. As a positive control, we first interrogated the data for transcripts with well-known expression patterns: for example, TPH2 and 5-HTT were highly enriched in serotonergic (but not noradrenergic or dopaminergic) cells, DBH only in noradrenergic neurons, and VMAT2 was enriched in all three populations. Also as expected, parvalbumin and GFAP were barely detected in all three populations, at ~50-fold lower levels than in whole brain homogenate. In exploratory analysis, we identify a number of transcripts that were found to be enriched in one or more of the monoaminergic populations. For example, the peptide transmitter urotensin IIB was enriched specifically in noradrenergic neurons, while crystallin-beta A2 was enriched specifically in serotonergic neurons. In total, we found 180 transcripts that were enriched at least 8-fold in one or more of the monoaminergic populations, relative to whole brain homogenate. We believe our data illustrate the feasibility of ‘transcriptome neuroanatomy’ and provide a valuable database for generating hypotheses and possible drug targets related to the monoaminergic systems.

Snow Naing
University of California, San Francisco

Quantitative Proteomics in Xenopus tropicalis brains to study the functions of Autism Spectrum Disorder (ASD) risk genes

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder with a complex genetic architecture. Advances in genomics technology have led to the identification of over 65 ASD risk genes, and yet, an understanding of the underlying pathways is lacking. Studying the brain proteome of an in vivo model with disrupted ASD risk genes can shed light on the molecular networks of ASD neuropathology. To do so, we leverage the diploid vertebrate tetrapod, Xenopus tropicalis (frog) because their brain development is comparable to that of humans and generating many genetic mutants in parallel is routine. However, it is not known how much of the X. tropicalis tadpole brain proteome overlaps with that of humans. Mass spectroscopy (MS) based proteomics is an excellent method to map and compare complex proteomes. One limitation for X. tropicalis tadpole brain is the absence of a high quality protein reference database, a prerequisite for MS-based proteomics. To address this issue, we assembled a custom reference proteome from RNA sequencing data of X. tropicalis tadpole whole brain and then assigned protein names from a curated human reference database using reciprocal BLAST methodology. The generated database consists of 32,068 proteins, out of which 24,524 proteins (76.5%) were assigned human gene symbols. Using this custom reference protein database for MS-based proteomics, we identified over 2,500 proteins in X. tropicalis brain samples. With this custom protein database, we are now poised to study the proteome changes in X. tropicalis brains following disruption of many ASD risk genes in parallel.

Helen Willsey

Leveraging Xenopus tropicalis and genome editing to reveal molecular mechanisms underlying psychiatric disorders

Advances in genetics have led to the identification of a reliable list of high-confidence risk genes for psychiatric disorders such as autism spectrum disorder (ASD) and Tourette Disorder (TD). Here we present a screening strategy leveraging CRISPR/Cas9 genome editing in Xenopus tropicalis (diploid frogs) to determine the neurodevelopmental phenotype(s) of these risk genes. We aim to identify ‘convergent phenotypes,’ which are phenotypes that occur in multiple genes of diverse cellular function within a disorder, with the hypothesis that these phenotypes are more likely to be relevant to pathobiology. We will assay the effects of gene loss by longitudinal microscopy of the brain throughout tadpole development and by RNA-Seq. By combining the CRISPR/Cas9 system, a tractable diploid vertebrate model organism amenable to high-throughput gene editing and phenotyping, and a reliably-associated set of risk genes, this study aims to understand the pathobiology of ASD and TD.

Alex Nord
UC Davis

Matenrnal immune activation alters development of fetal cortex in mouse

Maternal immune activation (MIA) has emerged as risk factor for neurodevelopmental disorders (NDDs), including autism and schizophrenia. MIA is among the strongest environmental risk factors and investigating the molecular mechanisms underlying pathology in offspring following MIA is crucial for understanding the gene-environment interactions associated with neurodevelopmental disorders. Here, we assayed gene expression in fetal cortex following mid-gestational MIA via maternal poly(I:C) injection at E12.5. We identified strong transient transcriptional signatures and developmental changes in fetal cortex of embryos exposed to maternal immune challenge. Changes included an initial acute signature that correlates with activation of stress response pathways in the fetal brain, followed by alterations in proliferative cell populations, neuronal differentiation, astrogenesis, and cortical lamination that emerged at E14.5 and peaked at E17.5. We also looked at epigenetic signatures at E14.5 by ChIP-seq analysis and found changes in histone modifications linking transcriptomic and epigenomic changes in fetal brain. To validate changes in gene expression signatures, we examined candidates from our transcriptomic data, finding anatomical and protein expression changes in E17.5 cerebral cortex concordant with signatures present in the RNA data, including reduced proliferative populations (i.e. Ki67, Pax6), altered cortical lamination patterns (Tbr1, Ctip2) and precocious astrogenesis (GFAP), with no evidence of previously reported cortical focal dysplasia (Satb2, Cux1). In summary, this systems-level characterization of MIA-associated neuropathology, validated at the protein level, provides novel insight into the molecular and developmental brain pathologies linking MIA and neurodevelopmental sequelae.

Hanna Ollila
Stanford University and Harvard Medical School

Two models for Seep Fragmentation

Narcolepsy is a sleep disorder characterized by excessive daytime sleepiness and fragmented sleep. Known triggers for narcolepsy are influenza-A infection and 2009 H1N1 influenza Pandemrix vaccine. Phenotypically similar fragmentation of sleep can occur in depression and insomnia. We built an international consortium to dissect disease mechanisms in narcolepsy (5,339 cases and 20,518 controls). In addition, we examined phenocopy of narcolepsy using 435,000 individuals in UK Biobank with reports of sleep problems and excessive daytime sleepiness. GWAS in narcolepsy found significant associations with HLA and 11 other loci (TRA, TRB, CTSH, IFNAR1, ZNF365, P2RY11, PRF1, CD207, SIRPG, IL27 and ZFAND2A). A partitioned heritability analysis indicated specific enrichment of functional elements active in cytotoxic and helper T cells. In addition, functional analysis showed the genetic variants in TRA and TRB loci act as remarkably strong chain usage QTLs for TRAJ*24 (p-value = 0.0017), TRAJ*28 (p-value = 1.36*10-10) and TRBV*4-2 (p-value = 3.71*10-117). Phenocopy analysis of UK Biobank traits revealed a different disease mechanism that led to similar symptoms. We saw that 20% of heritability in sleep complaints was explained by binding of USF1 transcription factor. USF1 bound at GWAS significant loci for sleep traits showing allele specific binding among the variants, including PER1 –a key sleep regulator. The findings were validated in USF1 knockout animals, which showed fragmented sleep. The findings show that narcolepsy is mediated through high-risk genetic factors, especially T cell receptor chains, whereas phenocopies in normal population are created through misregulation by USF1 and its target genes.

Chia-Chien Chen
University of California Santa Cruz

Selective activation of parvalbumin interneurons prevents stress-induced synapse loss and perceptual defects

Stress, a prevalent experience in modern society, is a major risk factor for many psychiatric disorders. Although sensorimotor abnormalities are often present in these disorders, little is known about how stress affects the sensory cortex. Combining behavioral analyses with in vivo synaptic imaging, we show that stressful experiences lead to accelerated loss of dendritic spines along the apical dendrites of layer (L) 5 pyramidal neurons (PNs) in the mouse barrel cortex, and such spine loss closely associates with deteriorated performance in a whisker-dependent texture discrimination task. Furthermore, the activity of parvalbumin-expressing inhibitory interneurons (PV+ INs) decreases in the stressed mouse due to reduced excitability of these neurons. Importantly, both behavioral defects and structural changes of L5 PNs are prevented by selective pharmacogenetic activation of PV+INs in the barrel cortex during stress. Finally, stressed mice raised under environmental enrichment (EE) maintain normal activation of PV+ INs, normal texture discrimination, and L5 PN spine dynamics similar to unstressed EE mice. Our findings suggest that the PV+ inhibitory circuit is crucial for normal synaptic dynamics in the mouse barrel cortex and sensory function. Pharmacological, pharmacogenetic and environmental approaches to prevent stress-induced maladaptive behaviors and synaptic malfunctions converge on the regulation of PV+ IN activity, pointing to a potential therapeutic target for stress-related disorders.

Dibyadeep Datta
Yale University, Department of Neuroscience

Molecular Insults to Higher-Order Dorsolateral Prefrontal Cortical Circuits: Implications for the Pathogenesis of Schizophrenia

The newly evolved prefrontal cortex mediates highest-order cognition and supports the neural basis of mental representation, the fundamental source of abstract thought. Particularly, the dorsolateral prefrontal cortex (dlPFC) is critical for spatial working memory (WM). Alterations in dlPFC circuitry appear to contribute to the pathophysiology of the cognitive deficits in schizophrenia. Within the PFC, ensembles of glutamatergic pyramidal cells in layer III microcircuits engage in recurrent excitation to sustain persistent neural firing, required to maintain WM. Our research has elucidated that dlPFC microcircuits are regulated at the molecular level by feedforward, cAMP-PKA-calcium pathways that renders them especially susceptible to atrophy; a phenomenon described as Dynamic Network Connectivity. A crucial molecular hub that gates feedforward, cAMP-PKA-calcium signaling is the phosphodiesterase PDE4D. We used immuno-electron microscopy (immunoEM) to localize PDE4D in dlPFC layer III and identify spatial interactions with subcellular organelles in adult rhesus macaques. ImmunoEM suggests PDE4D is positioned to regulate calcium signaling near glutamate-like axospinous synapses and essential organelles, such as spine apparatus in dendritic spines, and in association with microtubules, smooth endoplasmic reticulum and mitochondria in dendritic shafts. Inhibition of PDE4 activity using etazolate in in-vivo electrophysiology experiments in cognitively engaged monkeys reduced task related dlPFC firing. Therefore, PDE4D is critically positioned in dlPFC glutamatergic microcircuits to modulate cAMP regulation of calcium signaling. Impairments in PDE4D activity in schizophrenia patients suggested by genetic studies and inhibition through elevated inflammatory-signaling in the illness may render dlPFC microcircuits particularly susceptible to cAMP dysregulation of internal calcium release and WM deficits.

Krishna Vadodaria Vadodaria
Salk Institute for Biological Studies

Studiying SSRI resistance using MDD-patient derived neurons in vitro

Major depressive disorder (MDD) is currently treated with selective serotonin reuptake inhibitors (SSRIs), despite a significant percent of patients not responding to SSRIs. However, neural mechanisms contributing to SSRI resistance remain poorly understood. Human induced pluripotent stem cell (iPSC) and neuron-based disease modeling approaches have given novel insight into pathology of neuropsychiatric disorders as well as hope for the development of in vitro assay platforms. In our study, from a larger cohort of well-characterized MDD patients, we have generated iPSCs from extreme SSRI-responders and SSRI-nonresponders and derived forebrain neurons. We focused on studying serotonergic neurotransmission using patient derived neurons and found alterations in serotonin receptors, suggesting that these differences may contribute to SSRI resistance.

Yeni Kim
National Center for Mental Health

Nonlinear analysis reveals network differences between ASD and control neuronal lines derived from iPSCs

Nonlinear network analysis of multielectrode array data from iPSCs -induced neuronal lines reveal a consistent and statistically significant separation between ASD and control neurons, which traditional spiking variables failed to show. The nonlinear network variable also captured intersubject variance and was thus used as a covariate in differential gene expression analysis. Compared to ASD disease status, the nonlinear network variable yielded robust expression changes in biologically relevant pathways (synaptic and developmental genes). Further gene expression analysis of neuronal lines following risperidone treatment revealed alteration of developmental, metabolic, synaptic, and transcriptional genes, suggesting potential mechanisms of drug efficacy in autism.

Michael Gandal
UCLA-Semel Institute, Department of Psychiatry and Biobehavioral Sciences

Transcriptomic evidence for dysregulation of immune signaling and retrotransposition in a non-human primate model of maternal immune activation

Maternal immune activation (MIA) is an established risk factor for multiple neurodevelopmental and psychiatric disorders including schizophrenia. However, the molecular and neurobiological mechanisms through which MIA imparts risk for these disorders remain poorly understood. A recently developed nonhuman primate model of exposure to the viral mimic poly:ICLC during pregnancy shows abnormal social and repetitive behaviors, providing an unprecedented opportunity for mechanistic dissection. Here, we performed RNA-sequencing on homogenate brain tissue across four brain regions from 4-year old MIA-exposed (n=9) or saline-exposed (n=4) offspring. We used weighted gene co-expression network analysis (WGCNA) to identify modules of co-expressed genes, which we further characterize by gene ontology and cell-type specific marker enrichments. We identify 144 genes differentially expressed (FDR<0.1) across all brain regions (prefrontal cortex, anterior cingulate, hippocampus, and primary visual cortex) in MIA-exposed offspring. Differential expression was observed for multiple major histocompatibility complex genes as well as known regulators of retrotransposition, and we further observe upregulation of the ERV1 class of endogenous retroviruses. Gene-set enrichment analyses highlight significant upregulation of oxidative phosphorylation pathways and downregulation of microglial marker genes. Finally, co-expression networks identify region-specific dysregulation of oligodendrocyte and BDNF signaling. Together, these results begin to elucidate the brain-level molecular mechanisms through which maternal immune activation imparts risk for psychiatric disease.

Tierney Baum
Institute for Neurodegenerative Diseases, UCSF
Weill Institute for Neurosciences, University of California San Francisco

Psy-Clone: A Context-Specific Automated cDNA Selection Tool

Whole exome sequencing has yielded a large number of risk genes for several neuropsychiatric disorders, including autism spectrum disorders (ASD), intellectual disability, epileptic encephalopathies, and Tourette disorder. Critically, these efforts rely on the selection and acquisition of cDNA constructs for genes of interest. While there exist both academic and commercial sources for individual constructs or entire libraries, the choice of cDNA for these experiments is often unclear, particularly with respect to the choice of isoform, and as is whether or not non-reference variants are acceptable. Our lab has developed a web based platform that integrates several bioinformatics tools to facilitate efficient screening and annotation of cDNA construct sequences for use in molecular biological experiments. This tool automatically translates the nucleotide sequence of the construct into an amino acid sequence, aligns the predicted sequence to a reference database of protein sequences and identifies the best protein and isoform match. This tool can annotate any variants present in the construct and incorporates disease-associated mutations and transcriptomic data to help the user assess suitability for a particular biological context (e.g., isoform affected by disease-associated variant, isoform expressed in tissue of interest).

Dalton Hughes
Duke University, Department of Neurobiology

Network Dynamics of Negative and Positive Valence Systems in Decision Making

Evaluating risk and reward potential in the execution of motivated behaviors is important in decision-making and requires the activity of positive and negative valence systems. Critically, the imbalance of positive and negative valence systems may underlie many core symptoms in Major Depressive Disorder (MDD). Electrical processing within key brain regions of the mesocorticolimbic pathway has been well-established in mediating valence systems and a network-level representation of these computations would provide important insight into the behavioral alterations displayed in MDD. To better understand the balance of positive and negative valence systems we developed a behavioral task, modeled after the classic elevated plus maze and sucrose preference tasks, which directly quantifies the impact of anxiogenic stimuli on reward-motivated behavior. In vivo recordings of electrical activity across multiple brain regions were recorded as mice performed this task in order to capture activity patterns that underlie reward approach and anxiety-related behaviors. Electrophysiological data will be analyzed using machine-learning techniques in order to generate neural models that reflect the valence networks engaged. Network models of anxiogenic drug administration and reward-motivated behaviors across multiple behavioral tasks serve as positive controls in our machine learning approach. The framework discovered through this study has the potential to facilitate the development of new revolutionary approaches for diagnosis and treatment of MDD.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 56″][vc_column_text]Yefim Zaltsman
Institute for Neurodegenerative Diseases, UCSF
Weill Institute for Neurosciences, University of California San Francisco

Mapping the protein-protein interaction networks of autism spectrum disorder genes

Recent advances in whole exome sequencing and de novo variant detection have enabled discovery of genes implicated in several neuropsychiatric disorders. In particular, a substantial number of risk genes of large effect have been identified for autism spectrum disorder (ASD). Translating these findings in gene discovery to an understanding of the underlying pathobiology, however, has been challenging. This is in part due to a lack of understanding of the biological pathways ASD risk genes are central to in the developing human brain. Analyzing a map of protein-protein interactions (PPIs), centered around the proteins encoded by ASD risk genes, is one path forward for identifying key convergent pathways. However, there is a dearth of data on the physical interaction partners of proteins encoded by these genes, and how risk-associated mutations affect these interactions. To address this deficiency, and as part of a broader Psychiatric Cell Map Initiative (PCMI, pcmi.ucsf.edu), we are comprehensively mapping the PPIs for ASD risk genes in multiple cell types, starting with HEK293T cells and expanding to iPSC-derived neurons. We have generated tagged cDNA constructs for each ASD risk gene and are using affinity purification-mass spectrometry to characterize the PPIs. We anticipate that these PPI networks, along with parallel efforts in the PCMI to analyze protein-DNA and genetic interaction networks, will converge on specific biological pathways thus helping to elucidate the underlying pathobiology of ASD. As many of the risk genes for ASD also overlap with genes implicated in intellectual disability and schizophrenia, these findings can potentially carry relevance to other neuropsychiatric disorders.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 57″][vc_column_text]Kristina Yim
Yale School of Medicine Department of Genetics

Single-cell transcriptome analysis of embryonic mouse cortex reveals developmental trajectories of ASD risk gene expression

Genome-wide surveys in autism spectrum disorder (ASD) cohorts have identified genes that are significantly associated with ASD risk. There is a critical need to better understand how protein-disrupting genetic changes in ASD risk genes are linked to common neurobiological outcomes, to guide treatment and spark investigations of novel therapeutics. Notably, many ASD risk genes play a role in regulating the expression of other genes. We have previously shown that the ASD risk gene CHD8 regulates other ASD risk genes during neurodevelopment. These findings suggest that regulatory genes associated with ASD risk target common expression networks in the developing brain.

We performed single-cell transcriptome analysis of embryonic mouse cortex sampled at multiple time points spanning corticogenesis. Our analyses reveal a maturational trajectory that recapitulates cortical neurogenesis at cellular resolution. We use this trajectory to identify subsets of ASD risk genes that show highly correlated expression across multiple cortical regions and cell types. We also see ASD risk genes co-expressed within specific cell types, such as Chd8 and Tbr1 in the early cortical plate. Immunohistochemistry of selected ASD risk genes in sectioned mouse cortices confirms the patterns of expression found by our single-cell analysis. By deciphering how the individual ASD risk genes impact brain development, we can better understand how risk genes may be functionally connected and contribute to risk for neurodevelopmental disorders.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 58″][vc_column_text]Ruth Huttenhain
Department of Molecular & Cellular Pharmacology, University of California, San Francisco

Identification of spatiotemporally resolved GPCR protein interaction networks regulating receptor function

G protein-coupled receptors (GPCRs) represent the largest family of signaling receptors and drug targets. Following ligand-induced activation of GPCRs signal transduction is mediated by protein interaction networks operating on short timescales and across multiple cellular locations. While temporal dynamics of protein interactions have been previously characterized, a major challenge remains largely unmet: how to interrogate the protein interaction networks engaged by GPCRs while capturing both their spatial and temporal context.
Here, we developed a novel analytical approach combining APEX-based proximity labeling with quantitative proteomics and a system of spatial references, which delivers, with sub-minute temporal resolution, protein interaction networks and subcellular location for a target protein of interest.
We applied this approach to interrogate how protein interaction networks engaged by GPCRs respond to ligand-induced activation. We not only validated capture of proteins known to interact with the receptors, including those with transient or low affinity interactions, but demonstrated that our pipeline can be used to discover new network components regulating receptor function. Specifically, we identified TOM1 and WWP2 as members of a previously unrecognized ubiquitin network that controls homeostatic down-regulation of the delta-type opioid receptors. Currently, we extend this approach to examine the protein interaction networks engaged by the mu-type opioid receptor after stimulation with full, partial, and G protein biased agonists.
In summary, we described and validated a methodology that extends the utility of APEX-mediated proximity labeling to achieve spatiotemporally resolved protein network interrogation in intact cells.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 59″][vc_column_text]Nia Teerikorpi
Institute for Neurodegenerative Diseases, UCSF Weill Institute for Neurosciences, University of California San Francisco
Tetrad Graduate Program, University of California, San Francisco

Investigating the role of ASD-risk gene CUL3 in neurodevelopment using iPSC-derived neural cells

Exome sequencing of patients with autism spectrum disorder (ASD) have identified multiple de novo mutations putatively disrupting the gene CUL3, which encodes the E3 ubiquitin ligase cullin-3. CUL3 has many functions, including cell-cycle regulation, cell-fate determination, vesicular transport, and cell migration . Gene co-expression network analysis through brain development has placed CUL3 as a central “hub” gene, based on its strong correlation with a large subset of high confidence ASD (hcASD) genes. However, the mechanism(s) underlying pathobiology when this gene is disrupted are unclear, due in part to its pleiotropic functions. Therefore, we aim to de-convolute the complex role of CUL3 in relation to other hcASD genes through neural development. Using iPSC-derived neural cells as a model system for neural development, we have performed CRISPR inhibition (CRISPRi) of CUL3 to investigate changes in proliferation, developmental trajectory, and gene expression profile through differentiation of neural progenitor cells to immature excitatory neurons. At the same time, and as part of the Psychiatric Cell Map Initiative (pcmi.ucsf.edu), we are also mapping the protein-protein interaction networks of CUL3 in these neural cells using mass spectrometry. By leveraging multiple approaches spanning genetics and proteomics to investigate CUL3, we aim to better understand its role in early development and ASD pathophysiology.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 60″][vc_column_text]Aaron Gordon
University of California, Los Angeles

Long term maturation of human cortical forebrain spheroids models post-natal brain development

In vitro cultures derived from human induced pluripotent stem cells, are quickly becoming invaluable tools for studying early brain development. These systems provide a means to model both normal human brain development as well as to study brain development in neurodevelopmental disorders. One current limitation of many of these systems is that they remain relatively immature, mostly modeling the early to mid-fetal stages of brain development (Stein, 2014). Human cortical spheroids (hCS) (Pasca, 2015) are 3-dimensional cultures that have the unique ability to be cultured for long periods of time in-vitro. Here, we leveraged this unique ability to show that long-term cultures of hCS can achieve postnatal maturity based on two distinct genome-wide measurements: a) RNA sequencing based network analysis and transition mapping, and b) DNA methylation array based age prediction, which has shown to be the most accurate predictor of age (Horvath, 2013). To our knowledge this is the first time in-vitro brain models have been shown to develop beyond pre-natal stages. Strategies to further accelerate this process will likely be very valuable. We show that hCS are a viable model for brain development as they closely mimic in-vivo cortical development based on two independent, unbiased, whole genome analyses.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 61″][vc_column_text]Amy Lin

22q11.2 deletion or duplication carriers display convergent and divergent cognitive/behavioral phenotypes relevant to Schizophrenia and Autism Spectrum Disorder

22q11.2 copy number variants represent an important model for studying genetic contributions to psychiatric illness because they confer some of the largest known genetic risk for psychiatric disorders and include genes critical for brain/cognitive development. In a cohort of 91 22q11.2 deletion carriers, 34 22q11.2 duplication carriers, and 82 demographically-comparable controls, I found robust differences in cognitive/behavioral measures relevant to ASD and/or SCZ. 22q-del carriers exhibited the lowest verbal and nonverbal IQs, with controls exhibiting the highest IQs, while 22q-dup carriers were intermediate (22q-del<22q-dup<con). Moreover, both CNV groups showed significantly and similarly elevated ASD-related symptoms relative to controls, specifically social reciprocity, sensory sensitivity, emotion recognition and differentiation, theory of mind, and restricted/repetitive behavior (22q-del=22q-dup>con). In contrast, 22q-del carriers showed significantly more severe positive and negative symptoms of psychosis compared to 22q-dup carriers, who were no different than controls (22q-del>22q-dup=con). There were also significant interactions between age and group for measures of sensory sensitivity, processing speed, and working memory. Lastly, cross-correlational analyses revealed that IQ, verbal memory, and emotion recognition were negatively correlated with repetitive/restrictive behavior and impaired social responsivity. Also, processing speed, verbal memory, and working memory were negatively associated with psychotic symptomatology. In contrast, sensory sensitivity did not seem to correlate with any other cognitive or behavioral measure. Hierarchical clustering analysis showed that clusters were broadly preserved between groups, suggesting that relationships between cognitive/behavioral measures are not perturbed by 22q11.2 deletions or duplications. Future directions include integrating these cognitive/behavioral measures with measures of gene expression and regional neuroanatomical structure.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 62″][vc_column_text]Nawei Sun
Institute for Neurodegenerative Diseases, UCSF Weill Institute for Neurosciences, University of California San Francisco

Identifying Convergent Transcriptional Signatures of Autism Spectrum Disorder

The involvement of genetic components in the etiology of autism spectrum disorders (ASD) has long been indicated by family and twin studies. Traditional genetic approaches failed to identify a single genetic factor responsible for the phenotypes but suggested that the genetic architecture of ASD is better described by extreme locus and allelic heterogeneity, and contributions from both common and rare variants. With the emergence of whole exome sequencing technology and statistical approaches focusing on rare and/or de novo variants, substantial progress has been made in the discovery of risk genes, resulting in a large list of strongly associated genes. Based on the notion that this large number of risk genes will ultimately converge on a smaller number of biological pathways, we hypothesize that identifying convergent molecular signatures downstream of risk gene perturbation will highlight core biological pathways underlying ASD. Therefore, as part of the Psychiatric Cell Map Initiative (http://pcmi.ucsf.edu/), we will systematically perturb the top 20 ASD-associated genes (false discovery rate ≤ 0.01) in neuronal cells, evaluate gene expression profiles after perturbation, characterize the extent to which these result inconvergent transcriptional signatures, and relate these findings to higher-order hypotheses about ASD pathobiology. More specifically, we will use the CRISPR interference (CRISPRi) approach to repress the top 20 ASD risk genes individually in induced pluripotent stem cells (iPSCs) from human control individuals. We will then differentiate these ASD gene-perturbed iPSCs into neuronal cells including neural progenitor cells and neurons, and assess the transcriptomes of these cells at various neurodevelopmental stages with time course Ion Torrent AmpliSeq whole transcriptome profiling. Finally, we will perform systems biological analyses aimed at identifying convergent molecular pathway(s) among these ASD risk genes.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 63″][vc_column_text]Sheng Wang
Institute for Neurodegenerative Diseases, UCSF Weill Institute for Neurosciences, University of California San Francisco

Investigating the genetic architecture of Tourette Disorder

Tourette disorder (TD) is an early onset neurodevelopmental disorder characterized by chronic motor and vocal tics. We previously established the contribution of de novo damaging sequence variants through whole exome sequencing (WES) of 511 trios. Here, we sequence an additional 291 TD trios and conduct a combined analysis of all 802 trios. We observe an overrepresentation of de novo damaging sequence variants in simplex but not multiplex families, suggesting differing genetic architectures. Moreover, we observe preliminary evidence for an increased rate of de novo damaging sequence variants in female versus male probands, indicating a female protective effect, consistent with the male sex bias of this condition. Finally, we observe that the genes mutated in TD patients are enriched for those related to cell polarity, suggesting a common pathway underlying pathobiology. We also identify de novo copy number variants (CNVs) in the WES data, and observe, for the first time, a statistically significant excess of de novo CNVs in TD. We confirm this association with microarray genotyping data. Overall, we estimate that 9.7% of cases from TD simplex families carry a de novo damaging sequence variant and 1.5% carry a de novo CNV mediating risk. Additionally, we identify significant overlap of de novo sequence variants between TD and obsessive-compulsive disorder, and de novo CNVs between TD and autism spectrum disorder, consistent with shared genetic risk.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 64″][vc_column_text]Kasey Davis
Department of Psychiatry and Behavioral Sciences, Stanford University, Palo Alto, CA

The Impact of Glucocorticoids on Neuronal Function in MDD

Background: Major depressive disorder (MDD) is one of the most prevalent psychiatric disorders, characterized by depressed mood and lack of interest or pleasure in daily activities. In patients with MDD, impairments in neurogenesis, synaptogenesis and corticogenesis triggered by stress, are thought to cause impaired neuronal morphology and function. Disturbed hypothalamic-pituitary-adrenocortical axis regulation and elevated levels of glucocorticoids have been linked to MDD. Glucocorticoids are involved in neurotransmitter synthesis, neuronal survival and neuronal differentiation, making it an excellent candidate for stress perturbation. Animal studies have shown that excessive corticosterone reduces mitosis rates of hippocampal progenitor cells and the survival of newly formed neurons, while in post-mortem studies of MDD patients, glucocorticoid receptor mRNA is reduced. We hypothesize that stress-induced dysregulation will contribute to abnormalities of the prefrontal cortical neurons derived from iPSCs in patients with MDD compared to controls. To more fully elucidate the impact of glucocorticoids on neuronal function in MDD, we examined pyramidal neurons derived from iPSCs from MDD patients and healthy, age, gender and education match controls to test if neuronal function in MDD patients differs from healthy controls.

Methods: We derived stable induced neurons (siNs) from 3 MDD patients and 3 control iPSC lines using a Tol2 recombinase system . Cell lines were differentiated for 7 days with dox treatment and then treated for 7 more days in the absence or presences of cortisol. The molecular and cellular phenotypes before and following stress exposure were characterized. After 14 days, siNs were harvested for microarray analysis.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column][vc_text_separator title=”Poster Abstract 65″][vc_column_text]Cynthia Schumann
UC Davis MIND Institute, Dept of Psychiatry and Behavioral Sciences

Altered dendritic morphology in dorsolateral prefrontal cortex of nonhuman primates prenatally exposed to maternal immune activation

Maternal infection during pregnancy may increase the risk for neurodevelopmental disorders in offspring. Rhesus monkeys (Macaca mulatta) prenatally exposed to maternal immune activation (MIA) develop atypical behaviors, including increased repetitive behaviors and atypical social interactions. Here we present evidence of underlying brain pathology. Pregnant rhesus monkeys were injected with a modified form of the viral mimic polyI:C (poly ICLC) at the end of the first trimester (n=5) or second trimester (n=4). Brain tissue was collected from MIA-treated (n=9) male offspring or control (n=4) male offspring at 4 years of age. Blocks of dorsolateral prefrontal cortex (DLPFC/BA46) were processed with the Golgi-Cox impregnation method to analyze neuronal dendritic morphology and spine density. The DLPFC was traced utilizing Neurolucida software (MBF Bioscience) and further subdivided into supra- (II/III) and infra- (V/VI) granular cell layers. For each case in each of the two regions, 10 pyramidal cells were traced in their entirety, including all apical, oblique and basal dendrites, and their spines. Somal size and apical dendrite trunk diameter were measured on 30 cells per case per region. Apical dendrite diameter was collected over a 30μm section located 100±10 μm from the soma. Nonparametric statistics were applied to assess group differences in these summary measures. Compared to controls, MIA-treated offspring exhibit a greater number of oblique dendrites (infra- p<.01, supra- p<.01). There were no differences detected in spine density or soma size in either DLPFC region. However, apical dendrites of MIA-treated offspring were smaller in diameter in infragranular layers (p<.01) and trend difference in supragranular layers. These data provide evidence that prenatal exposure to MIA alters dendritic morphology in a nonhuman primate MIA model, which may have profound implications for revealing the underlying neuropathology of neurodevelopmental disorders related to maternal infection.

Nadine Schrode
Icahn School of Medicine at Mount Sinai

CRISPR-based functional evaluation of common SZ risk variants

Schizophrenia (SZ) is a highly heritable neuropsychiatric disorder. Although a large contribution of common variants to disease risk has been found, the functional mechanisms remain unclear.
We established a systematic and scalable strategy to evaluate the growing number of SZ-associated variants and genes in a hiPSC-based neuronal platform.
Through fine-mapping and eQTL analysis of 108 loci previously associated with SZ, we prioritized one putative causal eSNP for CRISPR-editing and three SZ-eQTL genes for CRISPRa/i-based modulation of endogenous gene expression. We found that CRISPR-editing of a single non-coding SNP (rs4702) can change neuronal expression of the cis-gene target (FURIN). Moreover, CRISPRa/i-based modulation of the SZ-eQTL genes SNAP91 and TSNARE1 combined with RNA-seq revealed downstream differentially expressed genes enriched for known SZ risk variants as well as a surprising degree of reciprocal convergence between SNAP91 and TSNARE1. Further phenotypic analyses of synaptic function confirmed known roles of SNAP91 and hypothesized roles of TSNARE1 in the presynaptic neuron, but also showed, for the first time, the impact of SNAP91 on post-synaptic neuronal function. Interestingly, downstream genes whose expression changed synchronously with synaptic activity were predicted to form a complex protein network, containing numerous genes involved in neuropsychiatric disorders.
Overall, our findings demonstrate the convergent effect of SZ-eQTL genes in excitatory neurons, and clarify known and novel roles at the level of synaptic function, expanding our understanding of how subtle genetic variation that impacts expression of SZ-eQTL genes might be sufficient to impact neuron function and increase risk to SZ.

Kellie Tamashiro
Johns Hopkins University School of Medicine

Exercise during gestation mitigates cognitive impairment in rat offspring exposed to maternal high-fat diet

Maternal consumption of high fat (HF) diet during gestation and lactation has negative effects on offspring cognitive function. We hypothesized that maternal exercise during gestation would mitigate offspring’s cognitive deficits associated with maternal HF diet. Pregnant Sprague-Dawley rats were given ad libitum access to standard chow (CH)(n=24) or high fat (HF)(n=24) diet and they remained on their respective diets throughout gestation and lactation. Half of each dietary group half remained sedentary (SED) while the other half had voluntary access to a running wheel (RW) in their home cage during gestation only resulting in 4 groups: CH-SED, CH-RW, HF-SED and HF-RW. On postnatal day (P)21, offspring were weaned onto CH diet. Body weight and plasma leptin levels were elevated at weaning in both HF-SED and HF-RW animals compared to CH control groups. Adult male and female offspring were tested for cognitive performance in the Novel Object Recognition Test and Barnes Maze. Male and female HF-SED offspring displayed cognitive impairment compared to CH control groups. Cognitive deficits were not observed in HF-RW offspring, with both male and female HF-RW offspring performing comparably to CH controls. At weaning, male HF-SED offspring had decreased leptin receptor (Lepr) expression in the hippocampus, whereas Lepr expression in HF-RW offspring was not different compared to CH-SED. Together, our results suggest that gestational exercise has beneficial effects against maternal HF diet associated cognitive impairment in offspring and may do so, in part, by normalizing Lepr expression in hippocampus.

Wun Sin
University of British Columbia

Functional variomics of ASD-associated gene PTEN

Routine sequencing of human genomes has provided a wealth of new gene variant information, resulting in thousands of newly identified missense variants that have not been functionally assessed. To fully utilize their potential in diagnosis or as disease risk factors, we examine whether we can infer the pathogenicity of the variants (even rare ones) based on their functional characteristics. As an initial step, we will examine the functional consequence of protein stability in more than 100 PTEN variants.
PTEN is involved in learning, memory and synaptic development. It is a protein and lipid phosphatase that inhibits the Akt/mTor pathway when active. Here we present two platforms, yeast and human HEK cells for the rapid assessment of missense variant PTEN protein stability. In addition, the yeast model also allows us to identify novel genetic interactions. On the other hand, the human cell-based platform allows us to identify new binding partners via the use of BioID2 technology. Comparison of findings from multiple platforms will serve to facilitate in the interpretation of variant phenotype in ASD.”

Ma-Li Wong
State University of New York Upstate Medical University

Rare genetic variants and antidepressant response

Rare variants can contribute around 30-40% of functional variability in genes related to drug action. Currently, approximately 12% of Americans take antidepressants; thus, the ability to predict antidepressant could have a significant impact in public health. We investigated the role of rare functional genetic variants in antidepressant response in Mexican-American individuals with DSM-VI criteria for major depressive disorder who participated in a prospective randomized study of 8 weeks of double-blind treatment with desipramine or fluoxetine. Hamilton Depression Rating Scale measurement was our primary outcome measure; regression- and permutation-based kernel-based adaptive cluster (KBAC) analyses were obtained using whole exome genotyping data remitters and non-responders. Several genes significantly associated with treatment remission were identified (FDR<0.05), and their network and pathway analysis revealed the involvement of the following processes: sensory transduction, regulation of response to cytokine stimulus, and meiotic cell cycle process. Our results corroborate the involvement of rare variants in antidepressant drug response in major depressive disorders.