ABSTRACT. Alisha Chunduri1, David G. Ashbrook2

Understanding gene-by-environment interactions is important across biology, particularly behaviour. Families of isogenic strains are excellently placed, as the same genome can be tested in multiple environments. The BXD's recent expansion to 140 strains makes them the largest family of murine isogenic genomes, and therefore give great power to detect QTL. Indefinite reproducible genometypes can be leveraged; old data can be reanalysed with emerging tools to produce novel biological insights. To highlight the importance of reanalyses, we obtained drug- and behavioural-phenotypes from Philip et al. 2010, and reanalysed their data with new genotypes from sequencing, and new models (GEMMA and R/qtl2). We discover QTL on chromosomes 3, 5, 9, 11, and 14, not found in the original study. We narrowed down the candidate genes based on their ability to alter gene expression and/or protein function, using cis-eQTL analysis, and variants predicted to be deleterious. Co-expression analysis (‘gene friends’) and human PheWAS were used to further narrow candidates. Prominent candidate genes include: Slitrk6 in a Chr 14 QTL for locomotion in the center of an open field, we show to be part of a coexpression network involved in voluntary movement, and association with neuropsychiatric phenotypes in PheWAS; and Cdk14, one of only 3 genes in a Chr 5 QTL for handling induced convulsions after ethanol treatment, that is regulated by the anticonvulsant drug valproic acid. By using families of isogenic strains, we can reuse and reanalyse data to discover novel and highly plausible candidate genes involved in response to environment.

1Department of Biotechnology, Chaitanya Bharathi Institute of Technology, Hyderabad 500075, India 2Department of Genetics, Genomics and Informatics, University of Tennessee Health Science Center, Memphis, TN 38163, USA

ABSTRACT. AR Dunn1, M Dai1, J-G Zhang1, VM Philip1, D Bridges2, KMS O’Connell1, CC Kaczorowski1

Alzheimer’s disease (AD) is highly complex, with individual genetic and environmental factors mediating disease risk and progression. There are also significant sex differences in aging and AD risk, and several common comorbidities (e.g. metabolic dysfunction) further add challenges to understanding disease mechanisms and targeting treatment strategies. We recently developed a population of genetically diverse AD mice (AD-BXDs; Neuner, et al. 2019). Because they model human genetic heterogeneity, AD-BXDs are ideally suited to investigate translationally-relevant genetic, environmental, and sex factors contributing to AD. Here, we used AD-BXDs to determine how genetics, diet, and sex interact to modify cognitive and metabolic symptoms of AD. We fed either normal chow diet or chronic high-fat diet (HFD) to 39 strains of AD-BXDs and nontransgenic littermates. We measured metabolic and cognitive traits across their lifespan. Metabolic and cognitive response to aging, diet, and the 5XFAD transgene varied across the AD-BXD population. In general, female mice were more susceptible to AD-related cognitive decline and weight loss, and response to HFD differed by sex and AD genotype. Female strains that were able to maintain body weight on HFD were less susceptible to cognitive decline, suggesting that HFD may be protective in some individual females. We identified a locus on Chr. 18 associated with susceptibility to diet-sensitized AD-risk that was specific to female mice. Within this locus, we nominate Myo5b as a novel candidate contributing to AD pathogenesis sensitized by sex and HFD that may be exploited to delay, prevent or treat cognitive and metabolic symptoms of AD.

The Jackson Laboratory, Bar Harbor, ME, USA 04609 University of Michigan School of Public Health, Ann Arbor, MI, USA 48109

ABSTRACT. Samantha Zanineli1, Maureen Tanner1, Camila Oliva1, Lina Crisostomo1, Alexys Barret

Adaptation to an environment often involves the evolution of multiple morphological, physiological and behavioral traits. For most organisms little is known about the genetic and functional relationship between adaptive traits. The Mexican tetra, Astyanax mexicanus, consists of fish that perpetually inhabit at least 30 caves in Northeast Mexico, and ancestral-like surface fish, which inhabit the rivers of Mexico and Southern Texas. The recent application of genomic sequencing and transgenesis approaches have established A. mexicanus as a model for studying the evolution of complex traits. Cave populations of A. mexicanus are interfertile with surface populations and have evolved numerous traits including eye degeneration, insomnia, albinism and enhances mechanosensory function. The interfertility of different populations from the same species presents the unique opportunity to defined the genetically map evolved traits. To define the relationships between these traits, we are assaying individual F2 hybrid fish for numerous morphological and behavioral traits. To measure the relationship between traits in genetically variable animals, individual F2 hybrid fish are characterized for locomotor behavior, prey-capture behavior, startle reflex and morphological attributes. To determine whether traits are regulated by shared genetic architecture we are currently determining the degree of trait co-segregation in hybrid fish. Following completion of phenotyping, individual fish will be processed for sequencing and we will identify quantitative trait loci that for different traits. Taken together, this approach provides a novel system to identify genes contributing to complex trait evolution in an emerging model system.

Thompson1, Julianna Booth1, Nicole Hinz2, Wayne Robinson1, Maureen Tanner1, Samantha Zanineli1, Evan Lloyd3, Morgan O’Gorman3, Brittnee McDole3, Alexandra Paz3, Rob Kozol 3, Johanna E. Kowalko3, Yaouen Fily 3, Erik R. Duboue3, Alex C. Keene3 1. NIH U-RISE Program, Florida Atlantic University, Jupiter, FL 33401 2. Department of Biology, Barnard College, New York, NY, 10027 3. Jupiter Life Science Initiative, Florida Atlantic University, Jupiter, FL 33458

ABSTRACT. Laboratory mice (inbred and outbred) are unable ability to solve the extrapolation task (the food bait moves out of animal view and it could be found if move in the same direction). The small proportions of mice, nevertheless, were able to solve this elementary logic task with the performance scores significantly higher in comparison with the chance level. This served as the prerequisite to perform the selection for high scores of extrapolation task solutions. Mice of the first selection generations (F1-F10) of the new EX strain solved the task successfully in the proportions, significantly above the 50% chance level. In further generations the level of successful solutions by mice of the selected strain became unstable. At the same time mice of these generations were significantly superior in comparison with non-selected mice in the success of puzzle-box test solutions. On these grounds the selection of mice for high and low performance in puzzle-box solution scores had been started and mice of three selection generations (in "plus" and “minus” directions) were tested using the simplified puzzle-box test protocol (1 day testing). The puzzle-box test contained 4 stages with the underpass into the dark, safe compartment being unobstructed, masked by wood-shavings and masked by the light plug (presented twice) Mice of the “plus” substrain demonstrated significantly higher plug-stage task solutions than mice, selected for “minus” performance. The detailed analysis revealed that the “plus” substrain mice were more efficient in the realization of solution, while mice of “minus” groups made desperate and unsuccessful attempts to remove the plug (performing the manipulation movements). Thus, mice of both groups understood the object permanence rule, the higher percentages of real task solutions by “plus” group mean their higher executive function performance.

Supported by RFBR (grant # 20-015-00287 and State Programm N AAA-A16-116021660055-1. 1Biology Department, Lomonossov Moscow State University, Vorobievy Gory, build.12, 119234, Moscow Russia

ABSTRACT. P. Cronell, S. Shams, L. Westberg

Rett syndrome, an X-linked neurodevelopmental disorder, exclusively affects females and is characterized by slowed development, loss of purposeful use of the hands, slowed brain and head growth, seizures and intellectual disability. Another feature in children with Rett syndrome are autistic-like behaviours such as loss of social interaction and communication. Deficits in the X-linked gene methyl-CpG binding protein 2 (MECP2) has been identified as the genetic cause of Rett syndrome. As an in vivo model that is easily modified genetically and allows generation of hundreds of test subjects, the vertebrate species zebrafish (Danio rerio) holds many advantages for studying MECP2 mutations. Here we investigated social behaviour, locomotor activity, and propensity to seizures in zebrafish lacking MECP2, compared to wildtype siblings (n ≤ 20). Social preference test was investigated in 21-days-old siblings to determine phenotypes for fish lacking one or both copies of the mecp2 gene. To determine the locomotor activity in response to a stimulating convulsant, wild-type fish and heterozygous and homozygous mutant fish (n ≤ 80) were treated with pentylenetetrazol and general locomotor activity was measured at 5-days of age. Whereas the lack of MECP2 did not change social preference for visual social stimulus in mecp2 mutants at 21-days of age lower locomotor activity was seen in mecp2 mutant fish at 5-days of age compared to sibling controls. Our results suggest that MECP2 regulates locomotion but not sociality in young zebrafish.

Department of Pharmacology, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Sweden Funding support: Swedish Research Council

ABSTRACT. HHA Thorpe1, Y Al-Sabagh1, S Hamidullah1, BW Jenkins1, MA Talhat1, C Suggett, CJ Reitz2, TA Martino2, JY Khokhar1

Alcohol use is a contributor in the premature deaths of approximately 3 million people annually. Among the risk factors for alcohol misuse is circadian rhythm disruption, including by genetic variation. Genetic manipulation and pharmacological inhibition of the circadian nuclear receptor REV-ERBα are known to disrupt molecular feedback loops integral to daily oscillations and impact diurnal fluctuations in the expression of proteins required for midbrain dopamine neurotransmission. However, the role of REV-ERBα in addiction neurobiology is unknown. Herein, we show that genetic deletion of REV-ERBα alters alcohol preference in male and female mice. REV-ERBα null mice showed the lowest alcohol preference in a two-bottle choice test across all genotypes, whereas there was no alcohol preference difference between heterozygotes and wildtypes. In a separate experiment, alcohol-consuming wildtype C57Bl/6N mice were administered the REV-ERBα/β inhibitor SR 8278 (25mg/kg or 50mg/kg) for seven days and alcohol preference was evaluated daily. No differences in alcohol preference were observed between treatment and vehicle groups. Regardless of genotype or treatment, female mice demonstrated higher alcohol preference than males. Our data supports that genetic variation in REV-ERBα may be a factor in the risk for problematic alcohol use, although pharmacological evidence could suggests a role of REV-ERBβ in alcohol consumption patterns or that temporal expression of REV-ERBα through neurodevelopment may impact adult alcohol consuming behaviours.

1Department of Biomedical Sciences, University of Guelph, Guelph, Ontario, Canada Funding Support: Discovery Grant from the Natural Sciences and Engineering Research Council award (RGPIN-2019-05121) to JYK 2Centre for Cardiovascular Investigations, Department of Biomedical Sciences, University of Guelph, Guelph, ON N1G 2W1, Canada. Funding Support: Canadian Institutes of Health Research (CIHR) and the Heart and Stroke Foundation to TAM

ABSTRACT. KR Nygaard1,2, RG Swift1,2, SE Maloney2,3, JD Dougherty1,2,3

Williams Syndrome is a neurodevelopmental disorder caused by a microdeletion on chromosome 7, and characterized by hypersociability, generalized anxiety, specific phobias, and learning disabilities. Duplication of the same region produces opposite social features, indicative of a copy number effect. The genetic landscape of Williams Syndrome provides a unique opportunity to study the underlying mechanisms of these behavioral features, but as a rare disorder, animal models are essential. The region is conserved in mice, and a heterozygous Complete Deletion mouse model mirrors the human deletion. This study deeply phenotyped behaviors of the Complete Deletion mouse model and simultaneously investigated the potential copy number effect of a single gene of interest in the region, Gtf2ird1, which has been linked to social behavior. Complete Deletion (CD) mice were bred to transgenic mice overexpressing Gtf2ird1 (OE). At least 20 mice of each genotype (WT, CD, OE, and CD/OE), were used to assess social and anxiety behaviors, in addition to other relevant traits such as fear, in a suite of tasks including Open Field, Social Approach, Elevated Plus Maze, Novelty Avoidance, Social Operant, and Conditioned Fear. The Complete Deletion mice show deficits in many tasks (e.g., decreased time spent in the center during Open Field) that are not attenuated with Gtf2ird1 overexpression. A second, independent cohort further assessed sociability and anxiety utilizing additional tasks. The comprehensive results of this study will reveal whether the Complete Deletion model is a useful tool for investigating the underlying mechanisms influencing social, anxiety, and fear behaviors in Williams Syndrome.

1Department of Genetics, 2Department of Psychiatry, 3Intellectual and Developmental Disabilities Research Center, Washington University in St. Louis School of Medicine, St. Louis, MO, USA Funding Support: NSF-DGE-1745038 (KRN) and 5R01MH107515-05 (JDD)

ABSTRACT. IK Green1, H Lee1, KJ Clark1

Dysregulation in the hypothalamic-pituitary-adrenal (HPA) axis signaling pathway has been documented as a major contributing factor to susceptibility towards psychiatric and other chronic illnesses. The role of three genes coding for receptors involved in this pathway, nr3c1, nr3c2, and mc2r, which code for glucocorticoid receptors (GR), mineralocorticoid receptors (MR), and adrenocorticotropic hormone (ACTH) receptors, respectively, have been established. In our previous research, we established a hyperosmotic shock and a light-based assay to illicit an acute stress response in larval zebrafish. Using nr3c1, nr3c2, and mc2r knockouts, we demonstrated that the locomotor response was dependent on HPA signaling through GR and ACTH receptors. We are transitioning from lab developed methodologies to commercial equipment. These systems simplify our light-based assay by automating the process. Commercial units operate off a script that allows investigators to write programs that consistently carry out their experiments while providing automated locomotor activity tracking. Unfortunately, the hyperosmotic challenge that we previously developed is more inconsistent than the light assay and is not amenable to the commercial system. Here, we propose a new behavioral paradigm utilizing vibration as an alternate acute stressor in larval zebrafish. Comparisons between the various assay systems are evaluated. Ultimately, we expect the vibration-based assay will be a viable secondary assay that we can use alongside our light assay to discover new genes that play a role in rapid responses to acute stress. Broader applications of this research are relevant to investigation of therapeutic treatments for a variety of illnesses related to HPA-axis signaling dysregulation.

1Biochemistry & Molecular Biology, Mayo Clinic, Rochester, Minnesota, USA Funding Support: NIGMS R01 Diversity Supplement, GM134732-02S1

ABSTRACT. Neha Rajput1, Stiven Roytman1, Justin W. Kenney1

Every individual behaves in their own way under the combined influence of environmental and biological factors. To better understand individual differences in behavior and how they are in-fluenced by biology, we analyzed zebrafish swim behavior during exploration of a novel tank in fish from different strains and both sexes. When exposed to a novel environment, zebrafish ex-hibit a variety of exploratory behaviors. To assess this behavior, we captured three-dimensional swim trajectories using depth-sensing cameras. To determine how sex and genetics influences behavior we recorded swim trajectories from 400 animals across four inbred strains (AB, Tu, TL, and WIK) and both sexes. Markerless tracking of the fish was done using a deep learning ap-proach (DeepLabCut) to identify the relative positions of different body parts (Head, Trunk, and Tail). We found that fish behavior stratifies into multiple distinct clusters and that cluster mem-bership was influenced by strain and sex. Additionally, we found consistent individual differ-ences in behavior over multiple days.

1Department of Biological Sciences, Wayne State University, Detroit, MI 48202 Funding Support: Wayne State University start-up funds

ABSTRACT. Ferraguto C.1, Peineau T.2, Bouleau Y.2, Piquemal M.1, Dulon* D.2, Pietropaolo* S.1

In our lives, at any given moments, we are surrounded by sounds and our brain must process all this sensory information, deciding what is important and choosing how to react. Abnormal auditory sensory processing leads to a wide range of behavioral symptoms, such as avoidance to acoustic stimuli and reduced sound tolerance. This phenomenon is known as hyperacousis, which affects 6- 10% of general population with a debilitating impact on daily life. Hyperacousis has been also widely reported in people with neurodevelopmental disorders (NDDs), in which hypersensivity to sound exacerbates social interaction and communication problems, typical of this condition. It has been suggested that hyperexcitability in the auditory system leads to hyperacusis. However, the neural mechanisms underlying this phenomenon are still unknown. One of the main hypothesis is that dysregulation of ion channels expressed in auditory neurons could trigger hyperexcitability, causing hyperacousis. Reduced expression/functionality of big conductance calcium-activated potassium channels (BKCa channels) has been linked to hyperacousis and has been also described in patients of two NDDs presenting hyperacousis, i.e., Fragile X syndrome (FXS) and William-Beuren syndrome (WBS). Therefore, BKCa channels represent a potential therapeutic target for hyperacousis. The aim of this project is to test the efficacy of a novel BKCa channel agonist in hyperacousis-induced wild-type mice, as well as in Fmr1-KO and CD mice, i.e. the mouse models of FXS and WBS. The results obtained will contribute to advancing our overall understanding of auditory processing, and to promote the designing of novel pharmacological interventions for acoustic pathologies.

1 University of Bordeaux, INCIA, Pessac Cedex, France; CNRS, INCIA, UMR 5287, Bordeaux, France. 2 University of Bordeaux, Inserm U1120, Bordeaux, France.

ABSTRACT. Myra E. Bower, Laura Saba, Matthew C. Keller, Jerry A. Stitzel, Charles Hoeffer, John K. Hewitt, Marissa A. Ehringer

Over the past 10 years, a variety of ‘omics resources have been applied to identify genes and pathways contributing to risk for complex disorders. However, it remains challenging to integrate these results and to disseminate the information in a manner that makes the most important concepts accessible to non-genetic researchers who might be in the best position to pursue functional studies investigating biological mechanism. We are developing a website of Gene Report Cards that houses information about genes involved in smoking-related behaviors. This website includes a single “landing” page for each gene, with links to relevant publications, bioinformatics websites, databases, and other information on each individual gene. Providing an easily accessible resource to individual gene information will allow researchers to expand on the knowledge basis of single gene mechanisms by easily referencing available information and recognizing missing data fields. The Gene Report Card website will summarize evidence from human studies, including GWAS across multiple phenotypes, TWAS, eQTL, and epigenetic results. It also includes a summary of evidence from mouse QTL, TWAS, and other evidence from mouse models. Finally, the latter third of the Gene Report Card provides information about other characteristics relevant to the gene/protein such as target drugs, conservation across species, and gene similarity clustering. We expect these Gene Reports Cards to serve as a useful tool for basic researchers interested in understanding the current status of ‘omics results from smoking-related studies.

Institute for Behavioral Genetics, University of Colorado, Boulder, CO, USA Supported by the University of Colorado internal funds

ABSTRACT. MK Lara1, JL Brabec1, AL Tyler2, JM Mahoney1,2

Generalized epilepsies, including juvenile myoclonic epilepsy (JME), juvenile absence epilepsy (JAE), and childhood absence epilepsy (CAE), are complex disorders with genetic etiology, and multiple significant loci associated with genetic generalized epilepsies have been identified through genome wide association studies (GWAS). This group of epilepsies are characterized by generalized seizures and present with spike-wave discharges (SWD), an electroencephalogram finding that accompanies absence seizures. Recently, Tyler et al. published a genetic mapping analysis of a complex mouse population containing multiple backcrosses and intercrosses between C57BL/6J and C3HeB/FeJ mice in which one of three absence epilepsy causing gene mutations was fixed (Gabrg2, Scn8a, or Gria4). In their epistasis analysis of the frequency and duration of SWD, they detected multiple novel loci, and in particular, mapped a suppressing interaction influencing SWD phenotypes between loci on chromosomes 2 and 7. While the mapped loci were large, we hypothesized that we could narrow candidate genes within these loci to genes that interact in cerebral cortex- and thalamus-specific epilepsy gene networks, as these regions support absence seizures. We applied a network-based functional prediction (NBFP) strategy using machine learning to identify gene networks among JME, CAE, and JAE GWAS genes and functionally scored positional candidate genes within the chr 2 and 7 loci in mice. Leading candidates identified by our approach have strong functional associations to genes involved in human epilepsy risk, suggesting their possible interaction influencing SWD phenotypes across multiple mouse models.

1 Department of Neurological Sciences, University of Vermont, Burlington, VT, United States 2 The Jackson Laboratory, Bar Harbor, ME, United States Funding Support: National Library of Medicine (5R21M012615-02) and the National Institute of General Medical Sciences (5P20GM130454-02)

ABSTRACT. Christiann H. Gaines1,2, Sarah H. Schoenrock1, Ian M. Carroll3, Fernando Pardo Manuel de Villena1, Martin T. Ferris1, Lisa M. Tarantino1,4

Substance use disorders (SUDs) are highly prevalent and impose a significant burden on society. Despite the high prevalence, there exist very few effective SUD treatments. The lack of treatments is due, in part, to gaps in our understanding of the etiology underlying these devastating disorders. SUD risk is influenced by genetic and environmental factors. Identifying mechanisms that increase SUD risk provides a potential avenue for novel treatments. I performed genetic mapping in a reduced complexity cross between two closely related inbred mouse substrains that differ in cocaine-induced locomotor activation, C3H/HeJ (HeJ) and C3H/HeNTac (NTac). Mapping identified only a suggestive locus on Chr 19 indicating that behavioral differences in C3H substrains may be influenced by other non-genetic factors. I identified prominent differences in the composition of the gut microbiota in HeJ and NTac and hypothesized that these differences may play a role in locomotor sensitivity to cocaine. I successfully shifted the composition of gut microbiota by cross fostering pups of each substrain to dams of the alternate substrain. Cross fostered groups of HeJ and NTac pups shared the same gut microbiota but maintained divergent locomotor response to cocaine. These data indicate that the composition of the gut microbiome at the time of testing is not the cause of cocaine-induced locomotor activation differences in these C3H substrains. We are currently testing candidate genes in the Chr 19 interval and investigating alternative hypotheses to explain the observed behavioral differences.

1Department of Genetics, 2Neuroscience Curriculum, School of Medicine, 3Department of Nutrition, Gillings School of Public Health and 4Division of Pharmacotherapy and Experimental Therapeutics, Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, NC Funding Support: Pilot Funding from P50 DA039841 (Center for Systems Neurogenetics of Addiction, Elissa Chesler, PI), R21 DA052171, T32DA007244-31.

ABSTRACT. Aleksandra M. Mech1, Saeedeh Husseinian1, Munise Merteroglu2, William Havelange1, Adele Leggieri1, Elisabeth M. Busch-Nentwich2, Caroline Brennan1 

Addiction, including nicotine addiction, is one of the major mental health disorders and a leading cause of death in the world. Previous studies have shown the relevance of genetic factors responsible for a progression to abusive usage, with heritability for smoking estimated to stand at 0.5. Increased understanding of the genetics of smoking is necessary to identify novel drug targets and improve treatment. Here, we used a novel self-administration assay for juvenile fish to screen ENU-mutagenized zebrafish lines for nicotine seeking. The assay consists of 3 asymmetrically connected chambers with nicotine being administered to one of the chambers by diffusion from a point source thereby setting up a concentration gradient across the 3 chambers. Increased time spent in proximity to the nicotine source is taken as evidence of nicotine seeking behavior. We screened 54 families covering 3318 dominant and 1037 recessive alleles for nicotine seeking. Wild-type fish showed limited tendency to approach or avoid the nicotine source. However, 10 families of ENU mutagenized fish showed preference for, or aversion to, the nicotine chamber. Two of these families are predicted to house a dominant mutation affecting the behavior and 8 recessive mutations. So far, we have shown the heritability of the observed phenotype for two of the lines. Future work will assess the heritability for the remaining families and identify the pathways affected.

1School of Biological and Chemical Science, Queen Mary University of London; 2Cambridge Institute of Therapeutic Immunology & Infectious Disease, University of Cambridge

Funding support: NIH R01 grant number U01 DA044400-01 to CHB and EMB

ABSTRACT. G Sala1, H Shoji1, T Miyakawa1

Numerous tests of mouse behaviour are conducted over multiple time points (i.e., time series). The statistical methods commonly employed to analyse time series are linear regression techniques (e.g., repeated-measures ANOVA). However, in most cases, such approaches are not powerful enough to handle time-series data’s complexity. More specifically, these methods make unrealistic assumptions (e.g., linear relationships between variables and normality of response variables). The consequences of violating these statistical assumptions are twofold: first, the model’s estimates (e.g., p-values) may be miscalculated; second, phenotypes of interest may be overlooked. Here, we show how to resolve these issues with Generalized Additive Models for Location Scale and Shape (GAMLSS). GAMLSS is a supervised machine-learning technique that relaxes the abovementioned statistical assumptions (i.e., linearity and normality). Furthermore, the assumption of linearity can be relaxed even for the model’s random structure. As a demonstrative example, we analyse longitudinal (i.e., recorded over seven days) home-cage locomotor activity in wild-type and genetically engineered subjects from 94 mouse strains (Mouse Phenotype Database). The results show that the statistical assumptions, which determine the model’s reliability, are always met. Most notable, this modelling approach can easily discriminate between different mutation-induced phenotypes. Some mutants exhibit an alteration of their locomotor activity over the whole experimental period. By contrast, in other strains, only the subjects’ circadian cycle is disrupted (e.g., enhanced activity during specific hours in mutants). Given its superior accuracy and pattern-detection capabilities over commonly employed methods, we recommend using GAMLSS for any analysis involving time-series data of mouse behaviour.

1 Institute for Comprehensive Medical Science (ICMS), Fujita Health University. Funding: This work was supported by the Japan Society for the Promotion of Science [20K14267 granted to GS].

ABSTRACT. S. Shams and L. Westberg

Oxytocin is a neuropeptide that regulates a wide range of mammalian social and non-social behaviors, such as bonding, social recognition, and anxiety. Zebrafish (Danio rerio) is a highly social vertebrate with phylogenetic conservation in oxytocin and relevant neurotransmitter systems, and relies heavily on visual social information (not olfactory). Since the involvement of the oxytocin system in zebrafish social behavior is not well-established we used adult CRISPR-Cas9-mutants lacking either of the two zebrafish oxytocin receptor genes, oxtr and oxtrl, and studied them in a social behavioral task. We observed socialization between four fish (same-genotype, n ≥ 20 for each group) in a large open-field (40x body length) and measured shoaling, schooling, group excursions, general activity, and specific motor patterns of mutant and sibling control fish. Following automated social stimulation, we also measured the levels of neurotransmitters and amino acids (dopamine, serotonin, norepinephrine, glutamate, GABA, and glycine) in multiple brain regions using HPLC for these mutant fish and their wild-type control fish. Our data show that in the large ethologically-relevant open-field, zebrafish lacking oxytocin receptors display subtle differences in locomotion and anxiety-related behaviors, and significant deficits in social behavior. We also found differences in neurotransmitter levels in the measured brain regions, indicating a role of oxytocin in regulation of neurotransmitters associated with social and emotional behavior. These results suggest that the two oxytocin receptors may play important roles in zebrafish social behavior. Further, these findings advance our understanding of neural mechanisms underlying oxytocin-regulated social interaction in zebrafish.

Department of Pharmacology, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Sweden Funding support: Swedish Research Council

ABSTRACT. Željko Agić1, Eva Mihelec 1, Valentina Dukić1, Ana Filošević Vujnović1 and Rozi Andretić Waldowski1

Circadian genes regulate many aspects of addiction to drugs in Drosophila and mammals. Cells use redox regulation to regulate their functioning and neuronal plasticity. However, role of redox changes in the addictive drug-induced neuronal plasticity is just beginning to be studied. To determine if flies with mutations in circadian genes have disturbed redox balance, as a potential explanation for their unique behavioral phenotype, we measured activity of antioxidant enzymes catalase (CAT) and Superoxid Dismutase (SOD) and the amount of hydrogen peroxide (H2O2) in period, timeless, Clock and cycle Drosphila male mutant flies. Amount of H2O2 changes as a function of the time of day, it is higher during the day then at night, while in constant darkness H2O2 is on average higher at all time points. Normalized for the amount of tissue the activity of CAT is significantly higher in heads than bodies in all genotypes. SOD activity and the amount of H2O2 is higher in the heads of wild type flies, but in other mutants is genotype dependent. Similarly, CAT is several folds higher early in the day in all genotypes, while SOD activity and H2O2 are genotype dependent. Since each circadian mutant showed specific deficits in the activity of antioxidant enzymes and H2O2 concentration we propose a gene specific role in the regulation of the antioxidant defense. Many neuropsychiatric conditions, including addiction, are influenced by irregularities in the circadian rhythm and have disrupted redox balance, thus our findings contribute to the understanding of biological mechanisms of these diseases.

1 Department of Biotechnology, University of Rijeka, Rijeka, CROATIA Funding support: Croatian Science Foundation (HRRZ) # 2794 and University of Rijeka Research Support, # 13.11.1.4.06 141

ABSTRACT. Saeedeh Hosseinian1, Aleksandra Mech1, Adele Leggieri1, William Havelange1, Munise Merteroglu2, Ian Sealy2, Elisabeth Busch-Nentwich2, Caroline Brennan1

Impulsivity is defined as acting on a momentary basis without consideration of outcomes. Two forms of impulsivity are impulsive action involving deficiency in response inhibition and impulsive choice referring to impairment in delayed gratification. Both forms of impulsivity are associated with many psychiatric disorders such as addiction, ADHD and bipolar disorder. While impulse control disorders show moderate to high degrees of heritability, the genetics of impulsivity is not widely studied. We aimed to identify genes and pathways underlying impulsivity using a forward genetic screen of ENU-mutagenised zebrafish. We screened 64 families estimated to cover 5072 dominant and 1327 recessive alleles. We assessed impulsive action using a zebrafish version of 5-choice serial reaction time task. In this task, after the animal learns the association between a stimulus and a reward, a pre-stimulus interval is applied during which a premature response is recorded as a measure of impulsive action. We identified seven candidate families of which one has been further analysed revealing a heritable deficit in impulse control. The exome sequencing of the founder of the line identified 29 candidate mutations of which three are associated with ADHD including one associated with four other psychiatric disorders as well in GWASs. Future work should identify genes and pathways underlying this phenotype. We demonstrated for the first time that a forward genetic screen of zebrafish for impulsivity could identify lines with potential translational relevance to human.

1School of Biological and Chemical Sciences, Queen Mary University of London, London, UK 2Department of Medicine, University of Cambridge, Cambridge, UK Funding Support: NIH R01 grant number U01 DA044400-01 to CHB and EMB

ABSTRACT. The research in my group is focused on understanding both the ‘causes’ and ‘consequences’ of genomic variation in the brain, and the role this plays in neuropsychiatric and neurodegenerative disease. Despite major advances in understanding the risk factors (both genetic and environmental) for these diseases, the mechanisms involved in the onset and progression of pathology are not fully understood and long-term treatments to reverse cellular disease processes in the brain remain elusive. Although genetic studies have been highly successful in identifying variants associated with brain disorders, there remains uncertainty about the specific causal genes involved and how their function is dysregulated during the progression of neuropathology. Increased understanding about the functional complexity of the genome has led to recognition about the role of non-coding regulatory variation in health and disease. Our work aims to characterise the regulatory regions, epigenetic modifications and transcriptional patterns defining the different brain regions and cell-types in the human central nervous system, and assess their role in neurodevelopment, ageing and disease. In this talk I will present on-going work aimed at identifying regulatory genomic variation and transcriptional diversity associated with a diverse range of brain phenotypes. I will describe the dynamic nature of DNA modifications across brain development and ageing and describe the impact of genetic variation on the epigenome during the life-course. Novel tools mean that it is now feasible to examine epigenetic variation across the genome in large numbers of samples, and I will give an overview of our recent analyses of brain disorders including schizophrenia and autism. Finally, I will outline some of the issues related to regulatory genomic studies of neuropsychiatric disease and explore the feasibility of identifying peripheral biomarkers of disease phenotypes manifest in inaccessible tissues such as the brain.

1University of Exeter Medical School, University of Exeter, UK.

ABSTRACT. Y Ben-Shahar1

Over 50 years ago, the pioneering ethologist Niko Tinbergen published the four fundamental questions biologists must ask to fully understand behavior (paraphrased): 1) What is the impact of the studied behavior on fitness? 2) How does it develop (trait ontogeny)? 3) How did it evolve over the history of the species (trait phylogeny)? 4) How does it work (underlying mechanism)? Yet, despite many exciting advances both in studies of brain functions and in the roles genetic and environmental factors play in regulating brain functions and behavior, almost all we currently know about neurons, brains, and behavior at the ultimate and proximate levels comes from studying just a few genetically-tractable animal models including the mouse, the fruit fly, zebrafish, and C. elegans. This narrow view stands in stark contrast to the early days of neuroscience research, which relied on studies in many different animal species. Prominent examples include the discovery of the neuronal action potential in the squid giant axon, long term memory in Aplysia, and many others. However, the development of genetically-encoded tools for manipulating and observing neuronal activity, and the emergence of species-independent approaches to genome editing, indicate that realizing Tinbergen’s vision is now possible. Here I will discuss strategies for adopting and applying modern neurogenetic tools for observing and manipulating neuronal activity, and genome editing in diverse insect species that are not genetically-tractable. Our current molecular and genetic approaches will increase the diversity of animal species that can be used to advance behavioral genetics research at the physiological, developmental, and evolutionary timescales.

1Depratment of Biology, Washington University in St. Louis, Saint Louis, Missouri, USA. Funding Support: National Science Foundation (USA) NSF-IOS 1754264 and NSF-DBI 1707221.

ABSTRACT. Laura B. Duvall1

Female Aedes aegypti mosquitoes bite humans to obtain a blood meal to develop their eggs. Remarkably, attraction to humans is completely suppressed for days after the blood meal while the female uses the blood protein to develop her eggs. We investigated neuropeptide Y (NPY)-related signaling in this long-term behavioral suppression, and discovered that drugs targeting human NPY Y2/4 receptors modulate mosquito host-seeking. In a screen of all predicted Ae. aegypti peptide receptors, we identified NPY-like receptor 7 (NPYLR7) as the sole target of these drugs. To obtain NPYLR7-selective small molecule agonists, we carried out a high-throughput screen and isolated highly selective NPYLR7 agonists that inhibit mosquito attraction to humans and blood-feeding from a live host. NPYLR7 mutants are defective in behavioral suppression, and resistant to these drugs. Current work focuses on identifying the anatomical circuits underlying host-seeking suppression and determining how NPYLR7 exerts its behavioral effects.

1 Department of Biological Sciences, Columbia University, New York, NY 10027

ABSTRACT. Alex C. Keene, Johanna Kowalko, and Erik Duboue

Animals are highly tuned to sense changes in both internal and external environments and adapt their behavior accordingly. We are currently developing the Mexican cavefish as a model identify novel regulators of sleep and feeding behavior. These fish display exist as eyed-surface populations that inhabit the rivers of northeast Mexico and multiple blind populations of the same species that inhabit nutrient poor cave environments and have converged on sleep loss. We have identified a number of neuromodulators that contribute to the evolution of sleep loss in A. mexicanus cavefish including upregulation of the wake promoting neuropeptide implicated in human narcolepsy Hypocretin/Orexin. In addition, we have developed transgenic and gene-editing methodology in this emergent model system allowing for systematic investigation of the genes and neurons regulating evolved differences in sleep. Systematic analysis has identified convergence on wide-spread neuroanatomical differences between surface fish and cavefish including hypothalamic expansion that is accompanied by increased sleep intensity. Current studies seek to identify how naturally occurring genetic variation contributes do these phenotypes. Investigating the mechanisms of sleep loss in Mexican cavefish has potential to provide insights into the variation in sleep need throughout the animal kingdom, and even between humans. Further, the resources developed to study sleep can be broadly applied to study other cavefish associated traits including obesity, diabetes, and dysregulation of stress response.

Jupiter Life Science Initiative, Florida Atlantic University, Jupiter FL 33458

ABSTRACT. Christine Merlin1,2, Guijun Wan1,3, Ashley N. Hayden1, Samantha E. Iiams1,2

Many animals use the Earth’s geomagnetic field for orientation and navigation. Yet, the molecular and cellular underpinnings of the magnetic sense remain largely unknown. A biophysical model proposed that magnetoreception can be achieved through quantum effects of magnetically-sensitive radical pairs formed by the photoexcitation of CRYPTOCHROME (CRY) proteins. So far, studies in Drosophila are the only ones to have provided compelling genetic evidence for the involvement of the ultraviolet (UV)-A/blue light-sensitive type 1 CRY (CRY1) in animal magnetoreception. Transgenic overexpression studies in Drosophila also extended this discovery to the monarch butterfly (Danaus plexippus) light-sensitive CRY1 as well as to the light-insensitive mammalian-like type 2 CRYs (CRY2s) of both monarchs and humans. However, because these studies were performed under non-physiological magnetic field intensities and without a directional component, a demonstration that CRYs actually function in a geomagnetic compass at Earth-strength magnetic fields has been lacking. Here, we developed a new behavioral assay in the monarch in which wild-caught fall migrants and laboratory-raised wild-type individuals respond to a reversal of the inclination of Earth-strength magnetic field under full-spectrum and UV-A/blue light, but not in darkness or under green/cyan light. Combined with reverse genetics, we show that CRY1, but not CRY2, is necessary for inclination-based magnetic sensing, and demonstrate that both antennae and eyes, which express CRY1, are magnetosensory organs. Our work argues that only light-sensitive CRYs function in animal light-dependent magnetoreception, and establishes the monarch as a new model to further elucidate the molecular and genetic mechanisms underlying this enigmatic sense.

1 Department of Biology and Center for Biological Clocks Research, Texas A&M University, College Station, Texas 77843, USA; 2 Genetics Interdisciplinary Program, Texas A&M University, College Station, Texas 77843, USA; 3 Department of Entomology, College of Plant Protection, Nanjing Agricultural University, Nanjing 210095, China

ABSTRACT. Samantha Zanineli1, Maureen Tanner1, Camila Oliva1, Lina Crisostomo1, Alexys Barret Thompson1, Julianna Booth1, Nicole Hinz2, Wayne Robinson1, Maureen Tanner1, Samantha Zanineli1, Evan Lloyd3, Morgan O’Gorman3, Brittnee McDole3, Alexandra Paz3, Rob Kozol 3, Johanna E. Kowalko3, Yaouen Fily 3, Erik R. Duboue3, Alex C. Keene3

Adaptation to an environment often involves the evolution of multiple morphological, physiological and behavioral traits. For most organisms little is known about the genetic and functional relationship between adaptive traits. The Mexican tetra, Astyanax mexicanus, consists of fish that perpetually inhabit at least 30 caves in Northeast Mexico, and ancestral-like surface fish, which inhabit the rivers of Mexico and Southern Texas. The recent application of genomic sequencing and transgenesis approaches have established A. mexicanus as a model for studying the evolution of complex traits. Cave populations of A. mexicanus are interfertile with surface populations and have evolved numerous traits including eye degeneration, insomnia, albinism and enhances mechanosensory function. The interfertility of different populations from the same species presents the unique opportunity to defined the genetically map evolved traits. To define the relationships between these traits, we are assaying individual F2 hybrid fish for numerous morphological and behavioral traits. To measure the relationship between traits in genetically variable animals, individual F2 hybrid fish are characterized for locomotor behavior, prey-capture behavior, startle reflex and morphological attributes. To determine whether traits are regulated by shared genetic architecture we are currently determining the degree of trait co-segregation in hybrid fish. Following completion of phenotyping, individual fish will be processed for sequencing and we will identify quantitative trait loci that for different traits. Taken together, this approach provides a novel system to identify genes contributing to complex trait evolution in an emerging model system.

1. NIH U-RISE Program, Florida Atlantic University, Jupiter, FL 33401 2. Department of Biology, Barnard College, New York, NY, 10027 3. Jupiter Life Science Initiative, Florida Atlantic University, Jupiter, FL 33458

ABSTRACT. JM Sadino1, CJ Kelly1, XG Bradeen1, ZR Donaldson1,2

The loss of a spouse is often cited as the most traumatic event in a person’s life. However, for most people, the severity and maladaptive effects of grief subside over time via a currently understudied neuromolecular adaptive process. Like humans, socially monogamous prairie voles (Microtus ochrogaster) form opposite-sex pair bonds, and upon partner separation, show behavioral and neuroendocrine stress-phenotypes that diminish over time. Eventually, they can form a new bond, a key indicator that they have adapted to the loss of their partner. Thus, prairie voles provide an ethologically-relevant model for examining potential neuromolecular changes that emerge following partner separation that regulate adapting to loss. Here, we test the hypothesis that extended partner separation diminishes pair bond associated behaviors and causes pair bond transcriptional signatures to erode. Following two weeks of pairing, nucleus accumbens tissue was collected from animals that either remained paired or were separated after 48hrs or 4wks. In a separate cohort, we assessed partner preference and selective aggression at these timepoints. Surprisingly, pair bond behaviors persist throughout separation, although partner preference metrics become more variable with prolonged separation. In addition, we found that the pair bond transcriptional signature is stable over time in animals that remain paired, but erodes following prolonged partner separation. Eroded genes are associated with mitochondrial regulation and gliogenesis, potentially indicating cellular stress. This result parallels recent findings examining the male prairie vole pair bonding transcriptional signature. Our results provide the foundational framework for studying the molecular basis of pair bond loss adaptation.

1Department of Molecular, Cellular, and Developmental Biology, University of Colorado Boulder, 80309, USA, 2Department of Psychology and Neuroscience, University of Colorado Boulder, 80309, USA

Funding Support: NIH DP2OD026143, Whitehall foundation, Dana Foundation

ABSTRACT. Over 250 mouse inbred strains have now been sequenced, including over 200 RI strains (BXD and CC). This allows the potential for a massive diallel cross (DAX) of over 60,000 isogenic F1s. Each one of these is fully “pre-sequenced” because we have sequence for both fully inbred parents. F1 genomes can be replicated any number of times, in any environment, and studied at any age. This not only improves traditional QTL mapping, but also allows us to identify dominance effects (since the DAX includes heterozygotes), parent-of-origin effects (by using reciprocal crosses), and epistatic interactions. Causal models, predictions, and ROC curves can be made in one subset of the hypothetical 60,000 F1s, and tested or refined in another subset. With the large number of available genometypes, it is possible to identify higher-order epistatic interactions among variants. If one includes genetically engineered models into a DAX, then the complex effects of genetic background and GXE can be jointly identified, and modifier alleles can be mapped. Because all animals are isogenic, a consistent polyphenome can be built up systematically, and because all DAX are derived from known parents, this polyphenome (a phenome extended across ages, treatments, sexes, etc) is integratable with decades of previous data. Data become better with age–what we call smart quadratic data.

New technologies and new tools make a DAX an important next step towards the realization of experimental precision medicine - being able to predict gene-by-gene-by-sex-by-environment (age, treatment, etc.) interactions, and to test these predictions. 

Department of Genetics, Genomics, and Informatics, The University of Tennessee Health Science Center, Memphis, TN, USA

ABSTRACT. Background: Bullying is a significant early life stressor linked to internalizing and externalizing behaviors, suicidal tendencies and poor academic performance. However, sex-specific differences within these correlates have not been studied in a large, diverse sample. Methods: Using ABCD 2.01 data repository (11,875 9-10-year-old participants), we evaluated associations between parent-reported bullying experienced by the child, and the child’s behavioral problems assessed using the Child Behavioral Checklist (CBCL); suicidal ideation/intent, with the Kiddie Schedule for Affective Disorders and Schizophrenia (K-SADS); cognition, using the NIH Toolbox® Cognition Battery; and academic performance. Statistical analyses included the generalized additive models, likelihood ratio and chi-square tests, with adjustments for socioeconomic and demographic status.

Results: 11,016 children from 21 ABCD sites were included in the analyses. 1,683 of these participants (15.3%) experienced bullying problems, 1.2 times more prevalent in boys than in girls. Bullying was associated with elevated CBCL scores. Boys had greater behavioral problems than girls, and associations between bullying and several CBCL domain scores were stronger in boys than in girls (P<0.001). Additionally, bullied children were 2.4 times more likely to display suicidal tendencies than non-bullied children. Worsening CBCL scores were associated with greater suicidal tendencies among bullied children compared to non-bullied peers (P<0.05). Cognitive scores varied by sex and were lower among bullied children (ΔR2<0.01, P<0.05). Boys and bullied children had poorer grades (P<0.001), worse in those with suicidal symptoms (P<0.001). Conclusions: These findings highlight the sex-specific negative consequences of bullying. Future longitudinal studies will identify factors that might promote resilience to bullying.

Acknowledgments: Data used in the preparation of this article were obtained from the Adolescent Brain Cognitive DevelopmentSM (ABCD) Study (https://abcdstudy.org), held in the NIMH Data Archive (NDA). This is a multisite, longitudinal study designed to recruit more than 10,000 children age 9-10 and follow them over 10 years into early adulthood. The ABCD Study® is supported by the National Institutes of Health and additional federal partners under award numbers U01DA041048, U01DA050989, U01DA051016, U01DA041022, U01DA051018, U01DA051037, U01DA050987, U01DA041174, U01DA041106, U01DA041117, U01DA041028, U01DA041134, U01DA050988, U01DA051039, U01DA041156, U01DA041025, U01DA041120, U01DA051038, U01DA041148, U01DA041093, U01DA041089, U24DA041123, U24DA041147.

A full list of supporters is available at https://abcdstudy.org/federal-partners.html. A listing of participating sites and a complete listing of the study investigators can be found at https://abcdstudy.org/consortium_members/. ABCD consortium investigators designed and implemented the study and/or provided data but did not necessarily participate in analysis or writing of this report. This manuscript reflects the views of the authors and may not reflect the opinions or views of the NIH or ABCD consortium investigators

ABSTRACT. Vivek Kohar1, Glen Beane1, Tom Sproule1, Jarek Trapszo1, Brian Q. Geuther1, Gautam Sabnis1,Keith Sheppard, Vivek Kumar1

1The Jackson Laboratory, 600 Main St, Bar Harbor, ME, USA 04609

We present an integrated, machine learning based advanced mouse phenotyping platform for laboratory mice that includes hardware and software for real-time monitoring as well as an open source GUI software - JAX AutoBehavior Analysis (JAX-ABA) to annotate multiple behaviors, train machine learning classifiers using the annotated data, and then predict behavior bouts in unlabeled videos. We have provided 3D designs of hardware and detailed specifications for camera, lightning, noise, and other electronic components. We provide insights into the performance of JAX-ABA for varying amount of annotated training data, various classifiers, and choice of features. We also provide a webapp to enable other researchers to easily browse through and share labelled datasets, classifiers, and run those classifiers on their videos. The platform has been tested with over 160 classical laboratory, wild derived and recombinant (BXD) inbred strains. We demonstrate the utility of our platform through survey of grooming and gait behaviors in the open field in ~1000 BXD mice across 101 strains and previously published ~2500 mice across 62 classical laboratory and wild derived strains. We compute the heritability of phenotypes and perform QTL analysis and GWAS to identify the genetic architecture of these behaviors. The JAX-ABA platform for data collection and analysis, along with the existing data allows researchers to compare their behavioral data with historical strain data. This allows the placement of new strain data in the same behavioral space as other strains and disease models allowing data integration across labs, experiments, and time.

This work was funded by The Jackson Laboratory Directors Innovation Fund, National Institute of Health DA041668 (NIDA), DA048634 (NIDA), and Brain and Behavioral Foundation Young Investigator Award (V. Kumar)

ABSTRACT. Leinani Hession*,1, Gautam Sabnis*,1, Others TBD1, Gary Churchill1, **, and Vivek Kumar1, **

Chronological aging is uniform, but biological aging is heterogeneous. Clinically, this heterogeneity manifests itself in health status and mortality, and distinguishes healthy from unhealthy aging. Clinical frailty indexes (FIs) serve as an important tool in gerontology to capture health status. FIs have been adapted for use in mice and are an effective predictor of mortality risk. To accelerate our understanding of biological aging, high-throughput approaches to pre-clinical studies are necessary. Currently, however, mouse frailty indexing is manual and relies on trained scorers, which imposes limits on scalability and reliability. Here, we introduce a machine learning based visual frailty index (vFI) for mice that operates on video data from an open field assay. We generate one of the largest mouse FI datasets comprising 256 males and 195 females. From video data we use neural networks to extract morphometric, gait, and other behavioral features that correlate with manual FI score and age. We use these features to train a regression model that accurately predicts frailty within 1.03 ±0.08 (3.9% ± 0.3%) of the pre-normalized FI score in terms of median absolute error (MAE). As sex-specific aspects of aging in mice are still poorly characterized, we provide an analysis on sex. In sum, we show that features of biological aging are encoded in open-field video data and can be used to construct a vFI that can complement or replace current manual FI methods. This new method will lead to increased accuracy, reproducibility, and scalability, and potentially enable large scale mechanistic and interventional studies of aging.

*Equal Contribution 1The Jackson Laboratory, 600 Main Street, Bar Harbor, ME 04609 **Corresponding Authors: Gary.Churchill@Jax.org, Vivek.Kumar@Jax.org

ABSTRACT. WC Booher1,2, LA Vanderlinden3,4, LA Hall1, CA Lowry6-10, LM Saba3, and MA Ehringer1,2

High and Low Activity strains of mice were bidirectionally selected for differences in open-field activity (OFA) and subsequently inbred to use as a genetic model for studying anxiety-like behaviors (DeFries et al., 1978). In addition to the extreme differences in OFA, the High and Low Activity mice demonstrate corresponding differences in other anxiety-related behaviors as assessed by light-dark box, elevated plus-maze, and novel object exploration (Booher et al., 2021). Hippocampal RNA-sequencing of the High and Low Activity mice identified 3,901 differentially expressed protein-coding genes, with both sex-dependent and sex-independent effects. Functional enrichment analysis (PANTHER) highlighted 15 GO terms, which allowed us to create a narrow list of 264 top candidate genes. Of the top candidate genes, 46 encoded four Complexes (I, II, IV, and V) and two electron carriers (cytochrome c and ubiquinone) of the mitochondrial oxidative phosphorylation (OXPHOS) process. The most striking results were in the female high anxiety, Low Activity mice, where 39/46 genes relating to OXPHOS were upregulated. In addition, comparison of our top candidate genes with two previously curated High and Low Activity gene lists and a PubMed literature review suggests Ndufa13, Ndufb7, Ndufc2, Ndufs2, Coq7, Arrb1, Crtc1, Dcc, Dlg2, Lrp8, Ntkr3 and Trim32 (the first five are involved in OXPHOS) as genes of interest for anxiety-like behaviors. In addition to the current results, mitochondrial dysfunction has recently appeared as both a cause and effect of anxiety-related disorders and thus should be further explored as a possible novel pharmaceutical treatment for anxiety disorders.

1.Department of Integrative Physiology, University of Colorado, Boulder, CO, USA 2.Institute for Behavioral Genetics, University of Colorado, Boulder, CO, USA 3.Department of Pharmaceutical Sciences of the Skaggs School of Pharmacy and Pharmaceutical Sciences, University of Colorado Anschutz Medical Campus, Aurora, CO, USA 4.Department of Epidemiology in the Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA 5.Department of Biostatistics and Informatics at the Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA 6.Department of Integrative Physiology, Center for Neuroscience, and Center for Microbial Exploration, University of Colorado Boulder, Boulder, CO 80309, USA 7.Departments of Psychiatry, Neurology, and Physical Medicine & Rehabilitation, University of Colorado Anschutz Medical Campus, Aurora, CO 80045, USA 8.Rocky Mountain Mental Illness Research Education and Clinical Center (MIRECC), Rocky Mountain Regional Veterans Affairs Medical Center (RMRVAMC), Aurora, Colorado, 80045, USA 9.Military and Veteran Microbiome: Consortium for Research and Education (MVM-CoRE), Denver, CO 80220, USA 10.Senior Fellow, inVIVO Planetary Health, of the Worldwide Universities Network (WUN), West New York, NJ 07093, USA

Funding 1.National Institute of Mental Health training grant (T32 MH016880) 2.National Institute of Mental Health (grant number 1R21MH116263) 3.The National Center for Complementary and Integrative Health (grant numbers R01AT010005 and R41AT011390) 4.The Colorado Office of Economic Development and International Trade (OEDIT) Advanced Industries Accelerator Program (grant number CTGG1-2020-3064) 5.The Department of the Navy, Office of Naval Research Multidisciplinary University Research Initiative (MURI) Award (grant number N00014-15-1-2809).