J Minerva1, JA Krawitz1, N Rashkover1, N Ma1, RA Jain1

When experiencing sudden acoustic stimuli, animals must integrate stimulus characteristics, context, and history to select if and how they respond. One aspect of this decision-making is habituation learning, where individuals diminish their responses to repeated innocuous stimuli, allowing them to attend to more relevant stimuli. A key gene regulating this habituation learning is ap2s1, a subunit of the AP2 complex which regulates clathrin-mediated endocytosis. Human AP2S1 alleles are associated with autism spectrum disorder and learning deficits, though the mechanisms of how this gene regulates learning are unknown.

To identify neural mechanisms through which ap2s1 regulates learning, we used a fluorescent calcium indicator (GCaMP5G) to quantify neuronal activity and pre- and postsynaptic calcium transients in critical circuit components for acoustically-evoked behavior in zebrafish larvae. We specifically examined the reticulospinal Mauthner command neurons initiating rapid escape responses, and their major sources of excitatory drive: the auditory nerve exciting the Mauthner lateral dendrites, and spiral fiber neurons synapsing on the Mauthner axon initial segments (AIS). ap2s1 mutants bias behavioral selection toward Mauthner-driven responses, and we observe increased Mauthner firing underlying their impaired habituation behavior. While auditory nerve driven dendritic activation is unaffected in ap2s1 mutants, we find elevated calcium transients at spiral fiber axon termini presynaptic to Mauthner AIS in naïve ap2s1 mutants. Presynaptic AIS activation attenuates with repeated stimuli, but remains elevated in mutants. These data suggest ap2s1 regulates excitatory synaptic strength to modulate habituation learning and behavioral choice through the spiral fiber pathway, suggesting AP2S1 similarly regulates excitatory drive in humans.

1Department of Biology, Haverford College, Haverford, PA, USA

Funding Support: NIH R15EY031539

BK Leung1, S Merlin5, AK Walker2,6, V Eapen2,7, R Clarke2,3, BW Balleine1, TM Furlong4,6

1Decision Neuroscience Laboratory, School of Psychology, 2Discipline of Psychiatry and Mental Health, 3Ingham Institute, 4School of Biomedical Sciences, University of New South Wales, Sydney, NSW, Australia. 5Medical Science, School of Science, Western Sydney University, Campbelltown, Sydney, NSW, Australia. 6Neuroscience Research Australia, Randwick, NSW, Australia. 7Mental Health Research Unit, South Western Sydney Local Health District, Liverpool, Australia.

Symptoms of autism spectrum disorder (ASD) and Gilles de la Tourette syndrome (GTS) may emerge because of imbalances in cortico-striatal neurocircuits. We have recently developed a mouse-model of ASD and GTS by knocking down Immp2l, a mitochondrial gene associated with these disorders. Immp2l knock-out (KD) mice and wild type-like mice (WT) were trained on Pavlovian and instrumental learning procedures where auditory cues predicted food delivery and lever-press responses earned a food outcome. It was demonstrated that goal-directed learning was not changed for Immp2l KD mice, as lever-press responses were sensitive to changes in outcome value, and to contingency reversal and degradation. However, Immp2l KD mice were more responsive to auditory stimuli paired with food as indicated by a non-specific increase in lever response rates during Pavlovian-to-instrumental transfer. Finally, there were no alterations to neuron density in striatum or any prefrontal cortex or limbic brain structures examined. The current study suggests that the Immp2l gene is not necessary for learned maladaptive goal driven behaviours in ASD or GTS, but that it may contribute to increased capacity for external stimuli to drive behaviour. Such alterations could potentially influence the expression of tics and repetitive behaviours. Given that this is the first application of this battery of instrumental learning procedures to a mouse model of ASD or GTS, it is an important initial step in determining the contribution of known risk-genes to goal-directed behaviours, which should be more broadly applied to other rodent models of ASD and GTS.

Funding Support: Strategic Initiative Project from the Australian Research Council (CE140100007), Sydney Partnership for Health Education Research and Enterprise (SPHERE) Grant, and Ingham Institute Mental Health Research Award.

AR Dearman¹, P Vrtička², J Moore¹, M Kumari¹, L Schalkwyk³

Commonly-used genetic predictors of poor mental health (polygenic indices; PGIs) are based on genome-wide association studies (GWAS) of neuropsychiatric conditions (NCs). Intuitively, such PGIs should capture the direct effects of a person's genetic make-up on their risk of poor mental health. However, typical GWAS do not account for within-family effects, resulting in PGIs which capture risk factors from the rearing environment. Furthermore, endophenotypes (such as brain structure and function) may be better suited to genetic characterisation than NC diagnoses, which are considered behavioural phenotypes. We use sex-stratified path models to estimate the direct and environmentally-mediated effects of PGIs on mental health in four UK population datasets. We include eleven PGIs based on NCs and 30 PGIs based on endophenotypes. Each model simultaneously regresses a PGI onto four outcomes: psychological distress, and three measures of the rearing environment (parental separation, father's employment and mother's education), along with paths between environments and psychological distress. We also run regressions to examine bivariate associations. Bivariate and path model coefficients are compared to estimate the extent to which PGI effects are environmentally mediated. Eleven PGIs are significantly associated with psychological distress. The only PGI consistently associated with psychological distress is depression – all other significant associations are sex- and dataset-specific, including six other NCs and four endophenotypes. The only PGIs without environmental mediation are two endophenotypes. In conclusion, the depression PGI is the most consistent predictor of poor mental health, however its effects may be partly mediated by the rearing environment.

¹Institute for Social and Economic Research, ²Department of Psychology, ³School of Life Sciences;University of Essex, Wivenhoe Park, Colchester, Essex, CO4 3SQ, UK

Funding Support: Economic and Social Research Council & Biotechnology & Biological Sciences Research Council (UK), ES/T00200X/1

Jamie Peters^1,2, Giuseppe Giannotti*^1,3, Jasper A. Heinsbroek^1,2, Robin Vareed^1,4, Nathaniel P. Kregar^1,2, Francesca Mottarlini⁵, Christopher M. Driskill², Kasey L. Brida², and Jeremy J. Day²

Addiction to opioids is characterized by pervasive opioid seeking that results from dysfunction in brain circuits after chronic opioid use. The prefrontal cortex is central to these dysfunctional circuits, but little is known about the microcircuitry and cell types within the prefrontal cortex that regulate opioid addiction and relapse. Using a preclinical model of opioid addiction, we trained rats to self-administer intravenous heroin over a period of weeks. We implemented in vivo fiber photometry and optogenetics combined with a neuroanatomical viral approach to examine the neuronal activity and function of two distinct prefrontal cortex ensembles. One ensemble projects to the nucleus accumbens shell and the other to the lateral hypothalamus, with only a small percentage of collateralization between these targets observed (~5%). These anatomically distinct neuronal ensembles were found to be activated during heroin-associated cues (tone+light paired with heroin infusions) in a similar fashion based on calcium transients recorded at the cell bodies. However, distinct functions for each ensemble were revealed by optogenetic manipulation during the cue-reactivity window, with the prefrontal ensemble projecting to the nucleus accumbens shell functioning as a limiter of heroin seeking, and the lateral hypothalamus-projecting ensemble functioning as a driver of heroin seeking. These prefrontal ensembles also arise from distinct layers of the cortex, and single nucleus RNA sequencing is currently underway to determine whether they also exhibit unique transcriptomes. Importantly, targeting both cue-reactive ensembles bidirectionally was necessary to reduce heroin relapse, underscoring the importance of cell-type and circuit-specific intervention strategies.

¹Department of Anesthesiology, University of Colorado Denver, Anschutz Medical Campus, Aurora, CO 80045, USA

²Present affiliation: Department of Neurobiology, University of Alabama at Birmingham, Birmingham, AL 35294, USA

³Present affiliation: Department of Integrative Physiology and Neuroscience, Washington State University, Pullman, WA 99164, USA

⁴Biomedical Sciences Graduate Program, University of Colorado Denver, Anschutz Medical Campus, Aurora, CO 80045, USA

⁵Department of Pharmacological and Biomolecular Sciences, University of Milan, 20133 Milan, Italy

Funding support: NIDA/NIH R01 DA045836 (J.P.), F31 DA059203 (R.V.), DP5 OD026407 (J.A.H.), and K99 DA048974 (G.G).

KJ Clemens1, S Hesam-Shariati1, K Zimmermann1, R Ceccin2, PD Waters3, NA Habib4, KV Morris5

Recent research highlights dysregulation in Brain-Derived Neurotrophic Factor (BDNF) expression may be linked to a range of neuropsychiatric and neurodegenerative disorders. Targeting BDNF-related pathways has emerged as a potential treatment strategy, specifically through manipulation of long non-coding RNA (BDNF-antisense), although safe therapeutic delivery remains a problem. Here, we targeted Bdnf-AS using short hairpin RNA (shRNA) encapsulated in Integrating Lipid Nanoparticles (iLNPs) injected intravenously into rats. The iLNPs deliver and integrate Bdnf-AS shRNA and exosome packaging machinery to liver cells, which are converted into stable producers expressing Extracellular Vesicle (EV) loaded with the shRNAs. The EVs are hypothesised to cross the blood-brain barrier, regulating Bdnf-AS expression in the brain. Rats receiving shRNA against Bdnf-AS exhibited influenced working memory retention compared to controls. Preliminary molecular analyses confirmed the presence of the vector in the liver gDNA in all treated rats and of the shRNA in the brain. Further analysis of brain, blood, spleen and liver is ongoing. The use of iLNPs for shRNA delivery represents a promising innovative strategy, although further characterization of molecular mechanisms and efficacy of this approach is needed.

1School of Psychology, University of New South Wales, Sydney, NSW, Australia, 2Menzies Health Institute Queensland and School of Pharmacy and Medical Science, Griffith University, Gold Coast, QLD, Australia, 3School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, NSW, Australia, 4Imperial College London, London, United Kingdom, 5Centre for Genomics and Personalised Health, School of Biomedical Sciences, Queensland University of Technology, Kelvin Grove, QLD, Australia

Funding Support: National Health and Medical Research Council Ideas Grant APP2027730

S Gottlieb1,3 , SM Scarboro1,2,3, MF Lopez,1,3 , and HC Becker,1,2,3,4

Oxytocin (OT) has gained interest in the field of alcohol use disorder (AUD), having been shown to reduce alcohol consumption in multiple drinking models when given exogenously. Here we chemogenetically activated OT neurons in the paraventricular nucleus (PVN) of the hypothalamus to investigate the neural mechanism of OT effects on alcohol consumption. Oxt-IRES-Cre mice underwent viral stereotaxic injections to express an excitatory DREADD or control virus in the PVN (n=4-7/sex/virus). After recovery, mice were trained to respond (FR4) for 12% (v/v) alcohol in daily 20- min sessions. After stable baseline responding/intake was established, testing was conducted over several sessions. Mice were tested 30-min following injection (IP) of vehicle, the DREADD-activator CNO (3mg/kg), or OT (0.5mg/kg) in a within-subjects. Lever responses and alcohol intake (g/kg) were measured. Systemic administration of OT significantly reduced lever responses and alcohol intake in both sexes regardless of which virus was injected (p<0.05). Animals who received control virus showed no effect of CNO administration on either behavioral outcome. In animals who received experimental virus, males significantly reduced lever pressing and alcohol intake (p<0.05), and females trended toward decrease (p=0.08). Systemic oxytocin shows promise as a potential therapeutic in AUD. Here we show targeted PVN neural activation of OT decreases alcohol self-administration in a way similar to systemic, elucidating a key brain region in this pathway. Larger sample sizes plan to be incorporated, and further studies aim to investigate the contribution of vasopressin in this oxytocin-induced reduction of alcohol self-administration.

1Department of Psychiatry and Behavioral Sciences, 2Department of Neuroscience, 3Charleston Alcohol Research Center, Medical University of South Carolina, 4RHJ VA Medical Center, Charleston, SC 29401-USA. Funding Support: NIAAA grants R01 AA026536, P50 AA10761, U01 AA014095, U24 AA29968, T32 AA007474, F31 AA026483 & VA Medical Research (BX000813)

Brad Balderson1 , Narayan Pokhrel2 , Yanning Zuo2 , Benjamin Johnson2 , Olivier George2, Abraham A. Palmer2,3, Graham McVicker1 , Francesca Telese2

The nucleus accumbens (NAc) is a critical brain region for reward processing and addiction. However, the impact that genetic variants associated with addiction have on the gene regulatory networks of specific NAc cell types remains unclear. To investigate this, we utilized an outbred population of rats, which were genotyped and assessed for drug-taking behavior, using an oxycodone intravenous self-administration model. We profiled gene expression and chromatin accessibility in ~450,000 single nuclei from NAc tissue collected from 85 rats following five weeks of oxycodone abstinence. Our analysis identified 15 major cell types within the NAc, including a rare population of D3 medium spiny neurons (MSNs), which express the dopamine receptor Drd3 and comprise ~2% of NAc cells. Differential abundance analysis revealed that higher D3 MSN abundance was associated with increased oxycodone intake. Additionally, we conducted a latent factor analysis to uncover cell type-specific gene expression and chromatin accessibility patterns across individuals. This analysis identified a distinct molecular signature linked to increased drug intake, composed of genes and open chromatin regions enriched in specific cell types, including D3 MSNs and opioid receptor Oprm1-expressing MSNs. Finally, we leveraged the rat genotypes to identify genetic variants associated with the gene expression and chromatin accessibility patterns linked to oxycodone consumption. This analysis yielded significant associations, suggesting a genetic basis for variability in gene expression and chromatin accessibility associated with drugtaking behavior in HS rats. Overall, our single-cell multiomic approach identified genetically encoded regulatory networks that are linked to oxycodone addiction-related behaviors in rats.

1 Salk Institute for Biological Studies, Integrative Biology Laboratory, La Jolla, CA 2 Department of Psychiatry, University of California San Diego, La Jolla, CA, USA 3 Institute for Genomic Medicine, University of California, San Diego, La Jolla, CA, USA Funding Support: NIDA U01DA050239

Lewis M Sherer¹, Vivian M Chan¹, R Steven Stowers², Sarah J Certel³, Karla R Kaun¹

Co-transmission, or the ability of a neuron to release more than one neurotransmitter, neuromodulator, or neuropeptide, has been increasingly recognized as a critical mechanism by which individual neurons can more precisely control their output to downstream circuits. Despite its importance to the nervous system, however, the functional effects of co-transmission are not fully understood. We use the well-characterized neural circuitry and powerful genetic toolkit of Drosophila melanogaster to examine properties of two octopamine/glutamate co-transmitting neurons with known roles in feeding and social behavior. VPM1 is a sexually dimorphic neuron that responds to male pheromones, and VPM4 is not dimorphic and responds to food. We characterize the stimulus-evoked release dynamics of octopamine and glutamate from these neurons using volumetric two-photon imaging of calcium and neurotransmitter-specific fluorescent sensors. Finally, to examine how housing experience and associated internal state impacts co-transmission, we examine changes in activity and transmitter release from VPM1 and VPM4 that occur as a consequence of socialization. These results are critical for a complete understanding of how neurons modulate their outputs to downstream circuitry and lay the foundation for the development of novel therapeutic strategies that target specific neurotransmitter systems.

1 - Department of Neuroscience, Brown University, Providence, RI, USA

2 - Department of Microbiology and Cell Biology, Montana State University, Bozeman, MT, USA

3 - Division of Biological Sciences, Center for Structural and Functional Neuroscience, The University of Montana, Missoula, MT, USA

Funding Support: Brown University Training Program in the Molecular Biology of Aging (5T32AG041688)

B. Martin1, E. Zeitoun1, G. Dieuset1, N. Costet2, V. Latapie3, S. Jamain3, A. Biraben1
The audiogenic seizure (AGS) is a widely used model for preclinical studies of death in epilepsy. Immediately after exposure to the sound, the mouse manifests a stereotyped behavior, which can be successively characterized by a wild running, clonic seizures, a tonic-clonic seizure in which the mouse falls onto its flanks, and a tonic seizure with an extension of the limbs towards the tail, followed or not by death. Our laboratory has recently developed two genetic mouse lines, LAGS+ and LAGS-, based on a bidirectional genetic selection on phenotypic criteria, initiated by a four-way cross derived from DBA/1J, DBA/2J, BALB/cJ and 129/SvTer. Now, both LAGS+ and LAGS- present 100% of tonic seizures when stimulated with a 110 dB white noise. However, they diverge on pronostic. Effectively, after 15 generations of selection, LAGS+ (for lethal audiogenic seizure), presents &gt;99% of lethal AGS while LAGS- (for non-lethal audiogenic seizure) presents &lt;2% of lethal AGS. With the aim of identifying chromosomal regions contributing to AGS-related mortality, we have genotyped LAGS+ and LAGS- mice and compared their allelic frequencies. Directional selection was. At each generation, 10 couples were constituted. At generation 11 for LAGS+ and generation 10 for LAGS-, we genotyped the 10 couples of each line using Illumina GGP GIGA-MUGA Arrays allowing the genotyping of more than 143,000 SNPs. We then performed a homozygosity mapping to identify shared genomic regions between animals with the same phenotype. The analyses identified a multitude of chromosomal regions as potentially involved in mortality in this model of epilepsy.

1Univ Rennes, CHU Rennes, INSERM, Laboratoire de Traitement de Signal et d’Image - UMR 1099, 35000 Rennes, France
2Univ Rennes, Inserm, EHESP, Irset (Institut de Recherche en Santé, Environnement et Travail), UMR_S 1085, F-35000 Rennes, France
3Univ Paris Est Créteil, INSERM, IMRB, Neuropsychiatrie Translationnelle, 94010 Créteil, France

K Huang^1,2, K Conn^1,2, E Greaves^1,2, LK Milton^1,2& CJ Foldi^1,2

Excessive exercise has been reported in up to 80% of patients with anorexia nervosa (AN) and is an essential component of the activity-based anorexia (ABA) paradigm. Exercise in the ABA model is often called “compulsive” and may share similar underlying biological mechanisms with obsessive-compulsive disorder (OCD), including disrupted dorsal striatum (DStr) activity. However, specific effects on compulsive behaviours or differentiated actions of D2R in dorsomedial striatum (DMS) from those in dorsolateral striatum (DLS) remain unknown. In order to determine the involvement of DStr D2R neurons in compulsive exercise and operant behaviours in ABA rats, we injected D2R-Cre driven inhibitory (Gi) DREADDs into the DMS, and operant-paired fibre photometry (D2-Cre driven GCaMP). Female Sprague-Dawley rats were either exposed to ABA conditions (n=15) or underwent the outcome-specific devaluation task (ODT; n=15). CNO was injected during ABA exposure or the instrumental training phase of ODT. We also monitored D2 dynamics in DMS (n=2) or DLS (n=6) during ODT. We showed that the suppression of D2R in DMS specifically increased running wheel activity in ABA (p=0.012) but did not affect compulsive operant responding in ODT. Moreover, changes in calcium release that correlated with operant actions during instrumental learning showed distinct patterns in DMS and DLS that changed throughout training. The results demonstrate the specific involvement of D2R in DStr in compulsive exercise in ABA. This knowledge will aid in the development of targeted treatments for patients with AN that exercise excessively and potentially a reframing of AN definition based on clinical subtypes.

¹Department of Physiology, Monash University, Clayton VIC, Australia; ²Monash Biomedicine Discovery Institute, Clayton VIC, Australia