ABSTRACT. Fetal alcohol spectrum disorders (FASD) is the leading preventable neurodevelopmental disorder in the Western world. One hallmark of FASD is cell death in central nervous system which has been linked to neuroinflammatory responses after developmental alcohol exposure. Genetics have been shown to have a role in the severity of alcohol’s effect on the developing brain as well as influence neuroinflammatory responses. In the present study, we aim to test whether there is an interaction between genetics and neuroimmune responses following neonatal ethanol exposure using C57Bl/6J (B6), DBA/2J (D2) mice and previously identified BXD recombinant inbred strains that show differential susceptibility to ethanol-induced cell death in the developing hippocampus. Neonatal mice were treated on postnatal day (P) 7 during the brain growth spurt, a time equivalent to the third trimester in humans. Animals received a subcutaneous injection of either 5.0g/kg ethanol in saline solution or isovolumetric saline given in two equal doses two hours apart. Animals were sacrificed 7 hours after initial ethanol exposure. Expression of neuroinflammatory markers were examined in the hippocampus using RT-qPCR, including Il1β, Tnf-α, Ccl2, and Il6. Variables examined were treatment, strain, and sex, and their interactions. Different markers showed differential expression with effects of strain observed in some and strain-by-treatment effect found in others. For example, pro-inflammatory markers were differentially expressed in BXD strains with high ethanol-induced cell death but not differentially expressed in strains with low ethanol-induced cell death. These results demonstrate a complex interaction between genetics, neuroinflammatory markers, and developmental alcohol exposure.

JA Baker1, JW Johnson2, CJM Kane2, KM Hamre1

1Department of Anatomy and Neurobiology, University of Tennessee Health Science Center, Memphis, TN, USA. 2Department of Neurobiology, University of Arkansas for Medical Sciences, Little Rock, AR, USA.

Funding Support: NIAAA (USA) F31AA026498, R01AA018834, R01AA024695, R21AA023723, R01AA023508, P30GM110702, & UTHSC/UAMS CORNET Award

ABSTRACT. Prenatal alcohol exposure disrupts physical, neurological and behavioral development, leading to a range of fetal alcohol spectrum disorders (FASD). Nutritional factors may modify alcohol's teratogenic effects. In fact, using an animal model, we have found that supplementation with the essential nutrient, choline, can reduce the severity of a number of behavioral alterations associated with prenatal alcohol exposure, whereas choline deficiency exacerbates ethanol’s damaging effects. Choline levels during the early development, in particular, influence performance on tasks that depend on the functional integrity of the hippocampus. But how choline affects hippocampal structure and function remain unknown. In the present study we examined the effects of choline on bi-directional hippocampal plasticity. Prenatal alcohol exposure impaired long-term potentiation (LTP) in the dentate gyrus; choline supplementation enhanced LTP in both ethanol-exposed and controls subjects and may also facilitate long-term depression. These data are the first to demonstrate that choline may improve hippocampal function, complementing our previous data showing that choline likely acts via several mechanisms to alter hippocampal development. Importantly, there are also genetic differences in choline metabolism, which could influence risk for FASD, as well as responsiveness to choline intervention. Understanding how genetic and nutritional factors modify prenatal alcohol effects has important implications both for prevention and treatment of FASD, particularly as choline supplementation is currently being investigated in clinical trials.

JD Thomas1, BR Christie2

1Center for Behavioral Teratology, San Diego State University, San Diego, CA, USA; 2Division of Medical Sciences, University of Victoria; Department of Psychology, San Diego State University. Funding Support: AA012446

ABSTRACT. Cannabinoid misuse is an establish risk factor for psychiatric disorders and represents a public health issue. The increase in risk for psychiatric disorders is stronger for individuals at genetic risk or when consumption happens during brain development but the underlying mechanistic role for the association between cannabinoid consumption and psychiatric disorders remains not well understood. Recent advances in genetic manipulation and behavioural research make zebrafish (Danio rerio) a suitable model to study the short and long-lasting effects of developmental exposure to drugs.

The aims of this study are (1) to investigate whether the developing central nervous system is susceptible to the effects of the psychoactive ingredients of synthetic cannabinoids, namely JWH-018, using zebrafish as a model and (2) to test whether the effects of JWH-018 in larvae and adult zebrafish are moderated by loss of function mutations in Disrupted In Schizophrenia 1 (Disc1), a gene previously associated to psychopathology in human association studies. Zebrafish embryos exposed to JWH-018 showed increased locomotion at high doses and reduced response to startle stimuli at lower doses. During adulthood, zebrafish exposed to JWH-018 showed decreased anxiety-like behaviour.

This is the first study looking at the behavioural effects of early developmental exposure to JWH-018, suggesting that exposure to this drug during early-development leads to short-term behavioural changes in zebrafish. Although further studies in human populations are needed to confirm the effects of synthetic cannabinoids during pregnancy, these results add further evidence to the increased risk for psychiatric disorders after exposure to cannabinoids during pregnancy.

Judit García-González1, Bruno De Quadros1, Maroua Akkari1, Caroline H. Brennan1†

1Department of Experimental and Biological Psychology. School of Biological and Chemical Sciences, Queen Mary, University of London.

ABSTRACT. Adolescence is a developmental period marked by numerous behavioral changes, including an increase in recreational drug use. Cannabis use is often initiated during this period, and its daily use is more prevalent than that of alcohol or tobacco among U.S. high school students. In addition, adolescents typically find cannabis’s primary psychoactive component, ∆9-tetrahydrocannabinol (THC), to be less aversive than adults. Nevertheless, it has yet to be demonstrated that adolescent rodents find THC rewarding using place conditioning. Using an edible THC self-administration procedure in an inbred mouse strain (C57BL/6J (B6)) with a propensity for consuming psychoactive substances, the conditioned rewarding properties of various edible THC doses in adolescents and adults were assessed. Results will be compared to those following experimenter-administered (injected) THC. Adolescents differed from adults on a number of edible THC-related measures, including consumption, locomotor response, and place conditioning. Furthermore, preliminary evidence suggests that subsequent compulsive-like self-administration of alcohol is impacted following edible THC consumption, an effect which might be age-dependent. Taken together, these data suggest that adolescents differ from adults in their use of and response to edible THC, which could have a protracted impact on subsequent substance use. Modeling self-administration of THC during the developmental period in which most humans initiate its use will aid in understanding the particular, concurrent and protracted behavioral and neurobiological consequences resulting therefrom.

MP Smoker1,2, SL Boehm1,2 1Department of Psychology, 2Indiana Alcohol Research Center, Indiana University-Purdue University Indianapolis, Indianapolis, IN, USA: Indiana Clinical and Translational Sciences Institute, funded in part by NIH/National Center for advancing Translational Sciences (UL1 TR001108); and NIH/NIAAA (AA00761).

ABSTRACT. Using a non-invasive sleep surrogate as a high-throughput phenotyping tool (immobility-defined sleep using video tracking software), we screened for primary recessive sleep phenotypes in ENU G3 pedigrees. Among the pedigrees screened, we focused on one where multiple individuals expressed reduced immobility-defined sleep relative to the entire population screened. To validate these findings, EEG recordings in affected individuals revealed striking differences during NREM sleep while, surprisingly, time spent in REM sleep was dramatically reduced. Mapping and whole genome sequencing of an affected individual revealed a single mis-sense mutation in the transmembrane region of the vesicular SNARE protein, VAMP2. Investigations into the nature of the mutation were driven by observations that this allele was quite distinct from either heterozygous or homozygous Vamp2 null mutant mice. The mutant protein is less stable than wildtype, steady state levels being 25-65% of wild-type levels depending on brain region. Furthermore, discrepancies in phenotype suggested that remaining mutant protein may interfere with SNARE protein zippering, membrane pore opening and/or vesicular release. Consequently, electron microscopy and electrophysiological recordings in hippocampal slices confirmed a deficit in vesicular release in mutants. Further investigations using a fluorescence vesicular release reporter in primary hippocampal neurons indicated that the vesicular release probability pv was profoundly decreased in homozygous neurons. The study confirms that rapid hierarchical screening tools remain a reliable approach towards high-throughput gene discovery while it highlights the relevance of synaptic mechanisms in complex behavioural processes such as sleep regulation.

Gareth Banks1, Mathilde Guillaumin2, Petrina Lau1, Erica Tagliatti3, Vladyslav Vyazovskiy2, Kirill Volynski3, Stuart Peirson2, Patrick M Nolan1

1MRC Harwell Institute, Mammalian Genetics Unit, Harwell Campus, Oxfordshire, OX11 0RD, UK. 2Sleep and Circadian Neuroscience Institute (SCNi), Nuffield Department of Clinical Neurosciences, University of Oxford, UK. 3University College London, Institute of Neurology, Queen Square, London, WC1N 3BG, UK. Funding MRC, MC_U142684173.

ABSTRACT. The BXD mouse family is the most deeply phenotyped mammalian model system, with >7000 classical phenotypes in GeneNetwork.org. GeneNetwork allows examination of complex interactions between gene variants, phenotypes from different biological levels, and environmental factors. The family consists of 152 inbred strains, each of which is a unique mosaic of alleles from the C57Bl/6J and DBA2/J parents. We have carried out 40X sequencing of 152 BXD strains using a Chromium linked-read strategy, with mean fragment length of 44kb. We have increased the number of known, segregating, small variants in the BXD to >6 million, and are working on large structural variants. Most variants segregate with an ~50/50 allele frequency, making them ideal for mapping. We have produced a draft ‘infinite marker map’, intended to identify every recombination event in the BXD family. Analysis with this draft map shows an improvement in both mapping power and precision. We also confirmed ~15,000 epoch specific variants, mainly clustered in ~1300 regions of the genome. These variants occurred in the parental strains over the 40 years between the first and last set of BXD strains being produced. Identification of private variants allowed predictions of variants causing strains to be outliers for phenotypes in GeneNetwork. Three genes are candidates for an abnormal memory and anxiety phenotype in BXD74. This family is an excellent resource for testing networks of causal and mechanistic relations among millions of molecular and organismal traits, including, addiction, neurodegeneration, and longevity. Full sequencing of all lines has only increased its usefulness.

David G. Ashbrook1, Danny Arends2, Pjotr Prins1, Megan K. Mulligan1, Suheeta Roy1, Evan G. Williams3, Cathleen Lutz4, Alicia Valenzuela4, Casey Bohl1, Jesse Ingels1, Melinda McCarty1, Arthur Centeno1, Johan Auwerx6, Saunak Sen7, Lu Lu1, Kelley Harris8, Abraham Palmer9, Yu-yu Ren9, Jonathan K Pritchard10, Andrew G. Clark11, Robert W. Williams1

1. Department of Genetics, Genomics and Informatics, University of Tennessee Health Science Center, Memphis, TN, USA 2. Lebenswissenschaftliche Fakultät, Albrecht Daniel Thaer-Institut, Humboldt-Universität zu Berlin, Invalidenstraße 42, Berlin, Germany 3. Department of Biology, Institute of Molecular Systems Biology, ETH Zurich, Zurich, Switzerland 4. Mouse Repository and the Rare and Orphan Disease Center, The Jackson Laboratory, Bar Harbor, ME USA 5. Division of Evolution & Genomic Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Oxford Road, Manchester, UK 6. Laboratory of Integrative and Systems Physiology, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland 7. Department of Preventive Medicine, University of Tennessee Health Science Center, Memphis, TN, USA 8. Department of Genome Sciences, School of Medicine, University of Washington, Seattle, WA, USA 9. Institute for Genomic Medicine, Department of Psychiatry, University of California San Diego, La Jolla, CA, USA 10. Department of Genetics, Stanford University, Stanford, CA, USA, 11. Department of Molecular Biology & Genetics, Cornell University, Ithaca, NY.

ABSTRACT. Although behavioural studies in mice are generally performed on individual animals in an isolated out-of-cage setting, their housing status prior to this is never systematically recorded. At present it is poorly understood how individual animals adapt to a group housed environment and how challenges arising from such interactions may influence downstream phenotyping. We have used the Actual Analytics HCA system (Bains et al 2016) to analyse how co-housing two mouse strains with strikingly divergent daily activity patterns (C57BL/6J and FVB/NCrlBRH) affects the individual animals within the cage. We demonstrate a variety of strain and sex specific differences in the extent to which an individual adapts to the activity rhythms of its cage-mates. For example, while male C57BL/6J animals are largely unaffected by the presence of others in their cage, male FVB/NCrlBRH animals more readily synchronise to the activity patterns of their cage-mates. We also demonstrate that these changes are dependent upon the age at which co-housed individuals are introduced into a cage and that some changes persist when the animal is removed from the group housed environment and placed into single housing. In conclusion, this work highlights the complexity of home cage synchronicity upon individual animals. Furthermore, this data highlights that the number and genotypes of animals in a cage may have a profound effect on downstream phenotyping performed on mouse models.

Gareth Banks1, Rasneer S. Bains1, Rowland R. Sillito2, Douglas Armstrong3, Sara Wells1 and Patrick M. Nolan1.

1. MRC Harwell Institute, Harwell, Oxfordshire, UK 2. ActualAnalytics, Wilkie Building, Teviot Row, Edinburgh, Scotland 3. Institute for Adaptive and Neural Computation, University of Edinburgh, Scotland

ABSTRACT. Neurobehavioral conditions, such as autism spectrum disorder, are phenotypically and genotypically complex. Not only do autistic children present with deficits in multiple behavioral domains, but monogenic causes revealed by traditional GWAS cannot account for non-linear genetic interactions contributing to autism-related traits. In this work, we first mined autism diagnostic tests for behavioral processes and phenotypes, incorporating these into the Neuro Behavior Ontology. Next, we performed genome-wide association testing for SNPs and copy number variation to annotate, for the first time, 30 novel behavioral traits in children with autism from the Simons Simplex Collection. We then used machine learning (Iterative Random Forest) to detect stable epigenetic interactions among novel behavioral traits and SNPs, creating networks of genomic loci to explain partial heritability for each trait. Many traits were associated with known autism-related loci. Amongst even highly-related behavioral traits, groups of associated loci had inverse effect sizes, highlighting the diversity of gene/phenotype associations in one autism cohort. Additionally, we created a network of autism-related phenotypic traits and associated genes to interrogate how genomic loci differentially affect behavioral traits in this population, expanding the power of the network by including links between genes co-expressed in the frontal cortex in humans. Altogether, this network approach prioritizes communities of genes associated with novel autism-related traits for patient stratification and, in the future, personalized medicine approaches.

JA Williams1,2,3, DC Russ1,2, LT Slater1,2, PM Nolan3, MM Simon3, A-M Mallon3, GV Gkoutos1,2 1Institute of Cancer and Genomic Sciences, Centre for Computational Biology, University of Birmingham, Birmingham UK 2Institute of Translational Medicine, University Hospitals NHS Foundation Trust, Birmingham UK 3Mammalian Genetics Unit, MRC Harwell Institute, Harwell Campus, UK

ABSTRACT. By the end of the 19th Century, the concept that all organisms, including humans, use a behavioral repertoire of innate and learned behavior was well established. At the same time Cajal and others offered a mechanistic explanation in which each behavior was to be found in a connected set of neurons and that learning occurred by increasing the stable strength of communication between neurons. This connectionist mechanism remains as the dominant theory in neuroscience to this day. I will describe the unexpected twists and turns in a journey that started 30 years ago when molecular biology was first applied to the study of synapses and behavior. Our findings have taken us to a new theory called the Synaptomic theory which revises and extends the connectionist theory. The Synaptomic theory explains how the repertoire of innate and learned behaviors can be stored and recalled using the remarkable complexity of the synapse proteome and the diversity of synapses in the synaptome, and how the more than 100 brain diseases disrupting postsynaptic genes cause their behavioral phenotypes.

ABSTRACT. Binge drinking is a major predictor of alcohol use disorder (AUD). Glucocorticoid receptor (GR) has been implicated in AUD, but GR dysregulation has not yet been identified as a genetic risk factor in non-dependent animals. Here, we examine the effects of compounds that modulate GR activity, mifepristone and tacrolimus, in a binge-like drinking task (DID) in the High Drinking in the Dark (HDID-1) mice. HDID-1 mice of both sexes were tested in a 2-day DID procedure, with drug/vehicle administered prior to 20% ethanol access on Day 2. Intake was evaluated after 2-hours of access on both days and 4-hour access on Day 2, after which blood was collected for blood ethanol concentration (BEC) analysis. In subsequent weeks, water and saccharin intake were also evaluated. Mifepristone and tacrolimus both reduced ethanol intake (g/kg) and BEC levels relative to vehicle, with no sex differences emerging. Neither drug reduced water intake, but tacrolimus reduced saccharin intake (ml/kg). In summary, HDID-1 mice exhibit a rapid behavioral response to mifepristone and tacrolimus, suggesting GR is dysregulated in these genetically at-risk mice. Our lab is currently examining gene expression of GR and its regulator proteins (FKBP51, FKBP52) in HDID-1 mice and their low-drinking founder line, HS/Npt, to determine whether selection has produced a sensitized GR system, and we are testing specific antagonists to GR (CORT113176) and FKBP51 (SAFit2) in HDID-1 mice. We are also exploring whether mifepristone reduces drinking by increasing the aversive properties of ethanol in an ethanol conditioned taste aversion task.

A. Savarese, P. Metten, S. Spence, J. Schlumbohm, W. Hack, A.W. Clark, & J. C. Crabbe

Behavioral Neuroscience Department, Oregon Health & Science University and VA Portland Health Care System, Portland, OR, 97239, USA

Support: VA Grant 101 BX000313; by NIAAA grants AA020245, AA013519, AA007468, and AA010760; and by the family of John R. Andrews

ABSTRACT. Reading and language abilities are critical factors for educational achievement and success in adulthood. These traits are highly heritable, but their underlying genetic architecture is largely undetermined. Genetic studies of reading and language traditionally focus on children with developmental disorders, however unselected adult samples would provide larger sample sizes thereby increasing power to identify genetic factors of small effect size. Here, we introduce an Australian adult population cohort (42 – 73 years of age, N = 1,195) with validated measures of reading and language ability, including non-word reading to assess phonological processing: a core component of reading skill. Genome-wide association was performed for a reading and spelling composite score, non-word reading, phonetic spelling, non-word repetition (a marker of language ability), and self-reported reading impairment. Here we focus only on replicating previous single nucleotide polymorphisms (SNPs) and genes associated with dyslexia and specific language impairments (SLI), for which we are sufficiently powered.

In gene-based tests, FOXP2, a well-established language gene, was identified in the top three most significant genes for non-word repetition (p = 1.21 x 10-4). The dyslexia candidates MRPL19and S100B were significantly associated with phonetic spelling (p = 3.11 x 10-2) and non-word repetition (p =4.58 x 10-2) respectively. For the reading and spelling composite score, SNPs rs9722 and rs9467075, previously associated with dyslexia, were significant (p = 1.67 x 10-2; 2.27 x 10-2 respectively) and seven further previously-reported SNPs (rs17236239, rs2710102, rs759178, rs600753, rs2538976, rs2538991 and rs807701) were significantly associated with non-word reading. Gene-set analyses of 14 candidate dyslexia genes and five SLI genes were not significant, but the neuron migration pathway was significantly associated with the composite reading and spelling score (p = 3.68 x 10-2). This research contributes to the identification and replication of genetic factors in reading and language disorders, crucial for understanding their aetiology and informing intervention strategies.

Presented by Catherine Doust (PhD Candidate, University of Edinburgh) Authors: Scott D Gordon, Nicholas G Martin, Simon E Fisher, Timothy C Bates, Michelle Luciano

ABSTRACT. Methylmercury (MeHg), an organic form of the heavy metal mercury, is one of the most neurotoxic environmental pollutants, demonstrated to have high potency for causing developmental neurotoxicity even at low levels. Yet, the role of prenatal MeHg exposure in the aetiology of neurodevelopmental disorders is controversial. In this study, we investigated the life-long behavioural effects of developmental exposure to low-levels of MeHg, below those causing overt toxicity, using the zebrafish model. Embryos were exposed to non-dosed water or water containing 5-30 nM MeHg from 6 hours post fertilization (hpf) for 96 h, then washed and transferred to non-dosed water for the remainder of the experiment. At 6 days pf (dpf), larvae were tested for locomotor activity in response to alternating light and dark conditions. To determine the long-term effects, developmentally exposed adult zebrafish were tested in a behavioral battery including assays for anxiety-related behavior, social affiliation, and sensorimotor response and habituation. Our results indicate that developmental exposure to MeHg caused a long-term increase in anxiety-related response in both the larvae and the adult fish. In addition, there is indication for motor deficits at higher levels of exposure. These results and their possible mechanisms of action will be discussed, as well as translational importance and future directions.

Supported by Marie Curie Individual Fellowship

L Glazer, CH Brennan. School of Biological and Chemical Sciences, Queen Mary University of London, London, UK, E1 4NS

ABSTRACT. Zinc finger homeobox 3 (Zfhx3) is a brain-region enriched transcription factor that binds to AT motifs in promoter and enhancer regions. In humans, coding sequence mutations have been linked to a range of diseases, including schizophrenia. Through the investigation of missense and null mutations by our group, Zfhx3 has been shown to have an important role in transcriptional regulation of neuropeptides and their receptors within the suprachiasmatic nucleus (SCN), allowing for the maintenance of typical circadian rhythms. In addition to the SCN, data from the Allen Brain Atlas show Zfhx3 expression within other regions of the mouse brain, including dopaminergic regions such as the ventral tegmental area and the substantia nigra. Zfhx3 is also upregulated in the cortex of schizophrenic patients. Given the dopaminergic hypothesis of schizophrenia, the upregulation seen in the cortex, and the genetic associations, we wish to further investigate the contribution of Zfhx3 to schizophrenia-related behavioural changes in mice. We are investigating the distribution of Zfhx3 expression within dopaminergic regions and identifying specific cell types using double immunofluorescence labelling. So far we have identified that while Zfhx3 is expressed in most dopaminergic neurons, it is also expressed in other cell types. We have also established a DAT-Cre driven Zfhx3 knockout mouse – a conditional knockout mouse model which deletes Zfhx3 only in dopaminergic cells. These mice have been subjected to a behavioural phenotype pipeline, focused on schizophrenia-related endophenotypes. We discuss the biological relevance of these findings, particularly in how they relate to dopaminergic function in physiology and disease.

Paige Chandler[1][2], Ashleigh Wilcox[1], Gareth Banks[1], Pat Nolan[1]

[1] – Neurobehavioural Genetics, Mammalian Genetics Unit, MRC Harwell, Oxford, Oxfordshire, UK [2] – Department of Biochemistry, University of Oxford, 3 South Parks Road, Oxford, Oxfordshire, UK

ABSTRACT. Decreased pH has been observed in the postmortem brain of patients with particular psychiatric disorders; however, it remains controversial whether this phenomenon is a primary feature of these diseases or a result of confounding factors. In our recent studies involving mouse models of schizophrenia, bipolar disorder, and autism spectrum disorders, we found that a decrease in brain pH, along with an increase in lactate levels, may be the underlying pathophysiology of these psychiatric disorders, rather than a mere artifact. Thus, we launched the Brain pH project, with the aim of improving our understanding of the prevalence of brain pH changes, particularly those that occur in psychiatric disorders, using animal models. Thus far, we have analyzed brain pH and lactate levels within 37 models of animals, including genetically modified mice, drug-treated mice, and mice that have undergone other experimental manipulations. Among the animal models examined, 12 showed changes in either or both brain pH and lactate levels. Importantly, meta-analysis of those data revealed a highly significant negative correlation between brain pH and lactate levels, suggesting that the increased lactate levels may lead to reduction of brain pH, whereas reduced lactate levels may lead to increased brain pH. We combined the biochemical data with behavioral data (e.g., data available from the Mouse Phenotype Database: http://www.mouse-phenotype.org/) to investigate the relationships of brain pH and lactate levels with behavioral outcomes, such as locomotor activity, anxiety-like behaviors, and working memory performance. We will discuss our interpretations of the relationships demonstrated in this analysis.

Hideo Hagihara1, Hirotaka Shoji1, Takao Kohno2, Atsuko Hayata-Tkano3,4, Kota Tamada5, Kei Hori6, Tetsuya Tatsukawa7, Matthieu Raveau7, Mihiro Shibutani8, Shuji Wakatsuki9, Yoko Hagino10, Takaoki Kasahara11, Tadahiro Numakawa12, Hikari Otabi13, Ikuo Nobuhisa14, Yoshio Hoshiba15, Haruko Nakamura16, Shota Katori17, Kyosuke Yamanishi18, Yoshihiro Takamiya1, Mika Tanaka1, Ipek Yalcin19, Masayuki Matsushita20, Mitsuharu Hattori2, Hitoshi Hashimoto3,4, Toru Takumi5, Mikio Hoshino6, Katsuhiko Tabuchi21, Kazuhiro Yamakawa7, Izuho Hatada8, Toshiyuki Araki9, Kazutaka Ikeda10, Tadafumi Kato11, Hiroshi Kunugi12, Atsushi Toyoda13, Johji Inazawa22, Tetsuya Taga14, Akiko Hayashi-Takagi15, Yoshio Goshima16, Takuji Iwasato17, Shigeo Okabe23, Herbert Y Meltzer24, Tsuyoshi Miyakawa1

1. Division of Systems Medical Science, Institute for Comprehensive Medical Science, Fujita Health University, Aichi, Japan 2. Department of Biomedical Science, Graduate School of Pharmaceutical Sciences, Nagoya City University, Aichi, Japan 3. Laboratory of Molecular Neuropharmacology, Graduate School of Pharmaceutical Sciences,

ABSTRACT. The endocannabinoid system (ECS) is an important modulator of neuronal functions in the mammalian brain, critically regulating the expression of several behaviors. Recent lines of evidence have suggested a role of the ECS in the control of social behaviors and their dysfunction in rodents and humans, although the precise relevance of the main cannabinoid receptor (CB1) for social intraspecific communication has not been investigated yet. Our study focused on ultrasonic (US) communication in CB1-KO mice, taking into account the relevance of age and sex differences. US emission in response to maternal separation was examined in male and female CB1-KO pups and their wild type (WT) littermates at post-natal days (PND) 4, 6, 8, and 10. US communication was then reanalyzed at adulthood, together with social interaction towards an adult female conspecific. The results clearly show a selective role of CB1 in modulating US communication, with an effect that was evident both at development and adulthood in both sexes. These data have therefore strong implications for social and communication pathologies, such as Autism Spectrum Disorders, suggesting a potential key-role of the ECS in these developmental diseases.

W Fyke*,1,2, M Premoli*,1,3, V Lemaire-Mayo1, S Middei4, G Marsicano5, WE Crusio1, S Pietropaolo1

1CNRS and University of Bordeaux, Institut de Neurosciences Cognitives et Intégratives d'Aquitaine, UMR 5287, Pessac, France; 2SUNY Downstate Medical Center -Neural and Behavioral Science Program, Brooklyn NY, USA; 3University of Brescia, Department of Molecular and Translational Medicine, Brescia, Italy; 4Italian National Research Council CNR Institue of Cellular Biology and Neurobiology IBCN, Rome Italy; 5Endocannabinoids and Neuroadaptation Group, NeuroCentre Magendie, U862 INSERM, 33077 Bordeaux, France. *Equal contributions

ABSTRACT. Autism spectrum disorder (ASD) is a developmental disorder characterized by numerous behavioural symptoms, including repetitive behaviours, inhibited communication, and reduced sociability. The MAM domain-containing glycosylphosphatidylinositol anchor 2 (MDGA2) gene has been linked to ASD in humans and haploinsufficency of MDGA2 in a novel mouse model (MDGA2+/-) resulted in behavioural phenotypes seen in humans with ASD (Connor et al., 2016, Neuron. 91:1052-68). To examine the effects MDGA2 on behaviour throughout the lifespan, we tested MDGA2+/- mice (F 10, M 7) and wildtype littermate controls (F 8, M 11) on a neurodevelopmental test battery (P1 – P24). Neurodevelopmental behaviours were assessed through milestone emergence, sensorimotor, and cognitive tests. Subsequently, a behavioural battery was performed at 2, 9, and 15 months of age to examine behaviours linked to ASD. Social impairment was determined with the social affiliation test, the three-chamber test of social interaction, and social transmission of food preference. The marble burying test was used to examine motor stereotypy, while the rotarod test assessed gross motor coordination and learning. Visuospatial learning and memory was assessed with the Morris Water Maze. Results indicate some genotype, sex and age differences, which will be described in detail. We will be examining if the differences found during the neurodevelopmental stage persist throughout the lifespan. Because ASD is a behaviorally-defined disorder, it is essential to use behavioral bioassays that are sensitive to associated symptoms in any prospective mouse model of ASD. These results will help determine the value of the MDGA2+/- mouse as a translational model of ASD.

Thalia Garvock-de Montbrun1, Michaela K. Purdon1, Emre Fertan1, Richard E. Brown1

1Department of Psychology & Neuroscience, Dalhousie University Halifax, Nova Scotia, B3H 4R2 Funding support: Brain Canada (MIRI 2014)

ABSTRACT. Apathy is defined as a quantitative reduction of goal-directed behavior (GDB). Although multidimensional model of apathy has been established, its underlying mechanisms and neural basis are poorly known. We hypothesize that at least one of the three key systems that are necessary to produce and control GDB is disrupted in apathy: motivation to act, executive abilities, and auto-activation. Each of these systems relies on a specific neuroanatomic circuitry, involving prefrontal cortex. Building on these observations we propose a new original computerized behavioral tool for exploring apathy along its three subcomponents. ICM-APATHY-TASKS (ICMAT), coding in Python, tests in interaction executive abilities (easy and difficult subtasks) and motivation (each subtask performed twice with high or low gain). Modified ICMAT includes testing of auto-activation by pursuing the task in a strongly hetero-guided, and an unguided condition. Twenty subjects in the study ECOCAPTURE 2 (Clinicaltrials.gov: NCT03272230, ends in 2020) completed the ICMAT, 10 apathetic patients with behavioral variant frontotemporal dementia (bvFTD) and 10 matched healthy controls. Twelve additional subjects completed the modified ICMAT: 7 healthy controls and 5 patients with prefrontal cortex lesion. Preliminary results in healthy subjects show a significant effect of difficulty but no effect of reward on performance. Similar results are observed in bvFTD patients. Healthy subjects perform significantly poorer in auto-initiation condition but tend to produce more activity as compared to hetero-guided condition. These results suggest that healthy subject tend to show an exploratory behavior in auto-initiation condition, i.e. activity less efficient but quantitatively higher than in hetero-guided condition. 1 FrontLAB, Inserm U 1127, CNRS UMR 7225, UPMC Paris 06 University, UMR S1127, Sorbonne University, Brain and Spinal Cord Institute (ICM), Hôpital Pitié-Salpêtrière, Paris, France.

A.Boutelier1, B. Batrancourt1, Johan Ferrand-Verdejo1, Richard Levy1,2,3 , Emmanuel Mandonnet1,4

2 Department of Neurology, Institute of Memory and Alzheimer's Disease (IM2A), Pitié-Salpêtrière Hospital, Public Hospitals of Paris (AP-HP), Paris, France.

3 Behavioural Neuropsychiatry, Pitié-Salpêtrière Hospital, Public Hospitals of Paris (AP-HP), Paris, France.

4 Department of Neurosurgery, Hôpital Lariboisière, Paris, France

ABSTRACT. Krushinsky-Molodkina (KM) rat strain was the first one in the range of similar genetic models, created in the late 1940s. When exposed to loud sound develop clonic (in 2-3 sec) and tonic (in 7-10 sec) convulsions. The daily sound exposure results in 15-12 days) with the development of myoclonic seizures, starting with face muscle jerks spreading to neck and body musculature. This phenomenon was described in other audiogenic prone rat strains and represents the typical seizure kindling. As the TLE symptoms in humans include also numerous rather serious psychiatric disorders (apart from myoclonus fits) the KM rat strain could represent the model of complicated changes development in brain structures similar and parallel to those in humans (when the long history of generalized seizures invokes the TLE symptoms). The behavioral peculiarities defects in KM rats could be noted in locomotion, anxiety signs and tendency to develop the catalepsy – both spontaneous and post-ictal. The short seizure development latencies and the possibility to reproduce seizures in the same animals made these animals the valuable tool for study the seizure modulation factors – e.g. strong olfactory stimulation and antiepileptic drugs. The developmental aspects of audiogenic epilepsy is also promising when this strain is used as the genetic model of seizure propensity.

IB Fedotova, II Poletaeva

Biology Department, Lomonossov Moscow State University Funding Support: RFBI, grant № 18-015-00173 and the State program «Neurobiological basis of animal behavior” № NIOKTR АААА-А16-116021660055-1»

ABSTRACT. Opioid misuse is a critical public health crisis in the United States. Epidemiological research has demonstrated that exposure to stress during adolescence leads to increased substance use. Further, the most frequently reported stressors are social. Through modeling exposure to adolescent social stress in animals we can gain a greater understanding of the long-term behavioral and physiological consequences. Here we set out to determine if exposure to social stress in adolescence alters morphine responses in adulthood in C57BL/6J mice. To determine biological mechanisms that could underlie observed behavioral changes, we examined changes in gene expression as a result of adolescent stress. Male and female C57BL/6J mice exposed to adolescent social stress displayed attenuated morphine sensitization compared to control animals. In contrast, there were no differences in baseline locomotor activity between the groups or in response to an acute injection of morphine. We observed differential expression of 65 genes in the nucleus accubmens following adolescent social stress. Pathway analysis suggest long-term alterations in stress hormone signaling (corticotropin releasing hormone signaling), growth factors (VEGF signaling), and intracellular signaling pathways (CREB signaling, ERK/MAPK signaling). These results suggest that exposure to adolescent social stress changes the response to chronic opioid administration. One mechanism by which social stress may influence this behavior is by altering the normal development of the nucleus accumbens. These data provide important insight into one environmental exposure that may heighten opioid misuse.

H.M. Kamens, C.N. Miller, W.J. Horton, M.J. Caruso, & S.A. Cavigelli

Department of Biobehavioral Health, The Pennsylvania State University, University Park, PA, 16802, USA

ABSTRACT. Under conditions of repeated intermittent exposure to ethanol, a sensitized locomotor stimulant response often develops in mice. The sensitized response may be what remains after tolerance has developed to sedative effects of ethanol. Conversely, sensitization and tolerance may be unrelated. An initial study in C57BL/6J x DBA/2J recombinant inbred strains concluded that the two phenomena are not genetically related, and thus, perhaps mechanistically distinct. Here, we sampled a more genetically diverse panel of 15 standard inbred mouse strains. Changes in activity counts and foot slip errors (a measure of ataxia in a grid balance test) across ethanol treatments indexed sensitization and tolerance, respectively. Results were strain-dependent for change in activity (F[14,132]=8.4, p<0.001) and in number of foot slip errors (F[14,132]=1.8, p<0.05). However, more active mice have greater opportunity to commit foot slips, and these traits were positively correlated (r=0.57, p<0.05). Therefore, we also examined the ratio of errors/activity counts. There was no significant effect of strain for change in this ataxia ratio, but ataxia did decrease across ethanol treatment days, indicating tolerance development. Significant differences among strains in the amount of sensitization developed in the absence of significant strain differences in tolerance development in ataxia ratio indicates that these two consequences of repeated ethanol exposure are not genetically or mechanistically related. A similar conclusion was reached when individual values for each of these traits were obtained in a heterogeneous mouse stock and level of sensitization and tolerance to ethanol were not found to correspond with each other.

Cheryl Reed 1 and Tamara J. Phillips1,2

1Oregon Health & Science University and 2Veterans Affairs Portland Health Care System Portland, OR, 97239 USA Supported by grants from the Department of Veterans Affairs, VA Research Career Scientist Program, NIH NIAAA P60AA010760 and R24AA020245.

ABSTRACT. The differences in the brain weight - the trait, connected to the level of animal cognition,- is usually evaluated comparing the respective scores in different taxa. The selection of laboratory mice for large and small relative brain weight is one of the techniques which permit to evaluate the role of this variable at the intraspecies level. Three selection experiments were performed in our laboratory, their results being more or less similar- with higher cognitive capacities in Large Brain line and higher indices of anxiety and stress reactivity in Small Brain mice. The selection for contrast values of this trait had been stopped at the level of F22 in the third experiment, and the two lines were randomly bred, now for 15 generations. The relative brain weight and behavioral differences stayed during breeding without selection. Large Brain mice were higher in cognitive task solution than Small Brain animals. The “puzzle-box” cognitive task was used, which requires the understanding of “object permanence rule”. At the same time the selection for high scores of cognitive task solution resulted in higher scores of the brain weight in the selected strain. This means, that in laboratory mice (very popular among neurobiologists) the brain weight values could indicate the cognitive capacities level.

NA Ogienko1, OV Perepelkina1, IG Lilp1, II Poletaeva1

1Biology Department, Lomonossov Moscow State University. Funding: RFBI, N 16-04-01169, the State program «Neurobiological basis of animal behavior” № NIOKTR АААА-А16-116021660055-1»

ABSTRACT. The APOE gene encodes apolipoprotein ε (ApoE), a protein that associates with lipids to form lipoproteins that package and traffic cholesterol and lipids through the bloodstream. There are at least three different alleles of the APOE gene: APOE2, APOE3, and APOE4. The APOE4 allele increases an individual's risk for developing late-onset Alzheimer Disease (AD) in a dose-dependent manner. Sex differences have been reported for AD susceptibility, age of onset, and cognitive-related symptom progression, with females being more affected than males. In this study, we use a systems biology approach to examine gene expression patterns in the brains of aged female and male individuals who are positive for the APOE4 allele in order to identify possible sex-related differences that may be relevant to AD. Based on correlation analysis, we identified a large number of genes with an expression pattern similar to that of APOE in APOE4-positive individuals. The number of these genes was much higher in APOE4-positive females than in APOE4 positive males, who in turn had more of such genes than APOE4-negative control groups. Profiling of these genes using Gene Ontology (GO) term classification, pathway enrichment, and differential expression analysis supports the idea of a transcriptional role of APOE with respect to sex differences and AD.

Hsu M1*, Dedhia M1*, Crusio WE1,2,3 and Delprato A1

1BioScience Project, Wakefield, MA. USA 2Institut de Neurosciences Cognitives et Intégratives d'Aquitaine, UMR 5287, University of Bordeaux, Pessac Cedex, France. 3Institut de Neurosciences Cognitives et Intégratives d'Aquitaine, UMR 5287, CNRS, Pessac Cedex, France

*Equal contributions

ABSTRACT. This study is examined a gender-specificity of genetic and environmental factors associated with family constellation in the formation of individual differences in aggression.

Materials and methods The data were collected among 367 representatives in complete biological families with two same-sex sibs, inhabitants of Kharkiv. The aggression level was defined by BDHI. The similarity among relatives was studied by interclass / intraclass correlation coefficients. The heritability indexes was determined by Falconer`s formula.

Results In male siblings physical aggression Н²= 84%, h² = 24%, Gd = 60% (rsb = 0.314, р = 0.045; rpo = 0.343, p = 0.035). The variability of the level of indirect, verbal aggression and irritability depends on birth order: an increase of similarity between fathers and younger sons compared to fathers and older sons was noted. Indirect aggression: r = 0.312, р=0.069 / r = – 0.170, р=0.271. Verbal aggression: r = 0.427, р=0.011 / r = – 0.143, р=0.355. Irritability: r = 0.327, р=0.055 / r = – 0.126, р=0.415. In female siblings physical aggression h² = 35%, indirect aggression h² = 41%, irritability h² = 36%. The impact of birth order is not revealed.

Conclusion The heritability of physical aggression in male may included the shared environment context defined by sibling interpersonal interactions and missed the role of major or candidate genes because of high level of the effects of dominance and epistasis.

Marina Shustikova Institute of Biology, V.N. Karazin Kharkiv National University, Ukraine.

ABSTRACT. While exposure to stress is a major risk factor for mood disorders, stress does not universally cause illness. For example, although women are less likely to experience trauma than men, they are twice as likely to develop depression. We previously reported that a single injection of (R,S)-ketamine before stress protects against depressive-like behavior and attenuates learned fear in male mice. However, the prophylactic efficacy of ketamine and ketamine metabolites in female mice remain largely unknown. To address this question, female mice were administered (R,S)-ketamine, (2R,6R)-HNK, or (2S,6S)-HNK at various doses 1 week before one of a number of stressors. Prophylactic efficacy was validated using the forced swim test (FST). In a separate set of experiments, we examined whether sex hormones influenced the efficacy of prophylactic compounds. We found that (R,S)-ketamine and (2R,6R)-HNK, but not (2S,6S)-HNK, significantly reduced immobility in the FST compared to saline controls. Interestingly, in females, ketamine was prophylactic at a lower dose than previously shown in males. (2R,6R)-HNK was prophylactic at a significantly smaller dose and at a faster rate than its precursor (R,S)-ketamine. Moreover, the prophylactic efficacy of these compounds was mediated by ovarian hormones. Overall, these data indicate that (R,S)- ketamine and (2R,6R)-HNK are effective prophylactics against stress in females and that their sex-specific effects may be modulated by gonadal hormones. Our findings offer insights into the prevention of stress- related impairments in a susceptible population and may further elucidate underlying sex-specific neuropathology contributing to the onset of depression.

Funding support: Coulter Biomedical Accelerator program, Columbia University

Briana K. Chen1, Christina T. LaGamma2, Rebecca A. Xiaoming Xu4,5, Shi-Xian Deng4,5, Donald W. Landry4,5, and Christine A. Denny2,3.

1Doctoral Program in Neurobiology and Behavior, Columbia University, New York, NY 2Division of Systems Neuroscience, Research Foundation for Mental Hygiene, Inc. (RFMH) / New York State Psychiatric Institute (NYSPI), New York, NY, 3Department of Psychiatry, Columbia University, New York, NY, 4Department of Medicine, Columbia University, New York, NY, 5Organic Chemistry Collaborative Center (OCCC), Department of Medicine, Columbia University, New York, NY.

ABSTRACT. When an animal repeatedly encounters a signal coupled with either a punishment or a reward, it eventually learns to expect both to occur together in a process called associative learning. A central goal in neuroscience is to understand how neural circuits integrate conflicting (rewarding and aversive) experiences that need to be behaviourally resolved during learning.

To shed light into this process at the molecular and cellular level, we are dissecting a neural circuit for sexual conditioning in the C. elegans male.

Sexual conditioning is a form of male-specific associative learning by which a rewarding experience with mates overrides an aversive association with starvation, thus switching the males’ behaviour to a stimulus from repulsion to attraction (Sakai et al., 2013). Previously, our lab implicated the Mystery Cells of the male (MCMs) interneurons and the neuropeptide PDF-1 as regulators of the sexually conditioned switch (Sammut et al., 2015).

Here we show a dual role for the neuropeptide PDF-1 in the regulation of both aversive and appetitive learning in C. elegans. By using a Cre-Lox intersectional strategy we find that PDF acts in distinct subsets of neurons to promote reward and to promote aversion. Also the molecular mechanisms underlying aversion and reward seem to be distinct: whereas only PDF-1 is required to signal reward, both PDF-1 and PDF-2 act redundantly to promote aversion. A similar dual role in encoding value has been recently described for dopamine in both mice and flies, suggesting a conserved value-assignment logic in neural circuits across evolution.

Laura Molina-Garcia1, Susana Colinas-Fischer1, Sergio Benavides-Laconcha1, Lucy Lin1 & Arantza Barrios1

1Department of Cell and Developmental Biology. University College London