Acnp 61st annual meeting: panels, mini-panels and study groups

You have full access to this article via your institution. Download PDF DECEMBER 4-7, 2022 PHOENIX, ARIZONA SPONSORSHIP STATEMENT: Publication of this supplement is sponsored by the ACNP.

All content was reviewed and selected by the Program Committee, which held full responsibility for the abstract selections. Only disclosures for presenting authors are listed. Individual

contributor disclosures may be found within the abstracts PANEL 1. CHASING THE GREEN DRAGON: CHALLENGES AND PROGRESS IN THE DEVELOPMENT OF RODENT MODELS OF CANNABIS AND CANNABINOID SELF

ADMINISTRATION 1.1 SPONTANEOUS AND PRECIPITATED CANNABINOID WITHDRAWAL IN MALE AND FEMALE RATS FOLLOWING SELF-ADMINISTERED OR EXPERIMENTER ADMINISTERED WIN55, 212-2 DELIVERY SADE SPENCER THE

UNIVERSITY OF MINNESOTA, MINNEAPOLIS, MINNESOTA, UNITED STATES BACKGROUND: Synthetic cannabinoids (SCBs) gained popularity as a designer drug alternative to cannabis in the 2000s.

WIN55,212-2 (WIN) is an aminoalkylindole derivative, and drugs from this class are common in SCB products. SCBs produce significant adverse effects, and users of SCBs report more severe

withdrawal compared to people using plant cannabis. This study assessed withdrawal in male and female rats following two methods of administration of the SCB WIN. METHODS: Adult male and

female Long Evans rats received jugular catheter implantation. In experimenter-administered drug studies, rats were given escalating doses of WIN (0.2, 0.4, 0.6, and 0.8 mg/kg) via twice

daily iv infusions (vehicle injected animals as controls). Withdrawal was precipitated with the CB1R inverse agonist SR 141716A (10 mg/kg, ip) four hours after the final infusion (n = 

8/group). Spontaneous withdrawal was observed at 6, 14, 24, 48, 72, and 96 hours (n = 7-8/group). In the self-administration studies (SA), rats were given access to WIN (6.25 ug/kg for

females; 12.5 ug/kg for males) or vehicle under standard FR1 SA parameters. Withdrawal was precipitated with SR 141716A four hours after the final SA session (n = 3-4/group), and spontaneous

withdrawal was scored at 14, 24, 48, 72, and 96 hours (n = 3-6/group). Somatic withdrawal signs were observed over 30 minutes to generate a global withdrawal score (GWS). RESULTS: There was

a main effect of WIN treatment on GWS (F(1,26) = 6.611, p = 0.0162) with precipitated withdrawal from experimenter administered WIN. Sidak’s multiple comparison testing revealed a

significant WIN treatment effect in males only (p = 0.035). There was a main effect of WIN treatment (F(1,41) = 22.15, p < 0.0001) and a Time x Sex interaction (F(5,119) = 2.502, p = 

0.0342) on GWS with spontaneous withdrawal from experimenter administered WIN injections. Tukey’s multiple comparison testing indicated a significant difference between WIN treated male and

female rats at 48 hours with males showing a higher GWS (p = 0.0415). Equivalent analysis was performed after WIN SA. CONCLUSIONS: There is a robust spontaneous cannabinoid withdrawal

syndrome observed following spontaneous withdrawal from experimenter-administered or self-administered SCB in male and female rats. Cannabinoid withdrawal was more robust in male rats in the

non-contingent model, contrary to our expectations. DISCLOSURE: Nothing to disclose. 1.2 ASSESSING THE LONG-TERM IMPACT OF INTRAVENOUS CANNABINOID SELF-ADMINISTRATION IN ADOLESCENT MALE AND

FEMALE RATS SIERRA STRINGFIELD UNIVERSITY OF PITTSBURGH, PITTSBURGH, PENNSYLVANIA, UNITED STATES BACKGROUND: Cannabis use during adolescence is associated with cognitive deficits and risk

of psychiatric disorders in adulthood. Cannabidiol (CBD) is a component of cannabis that may diminish the aversive properties of THC during self-administration or protect against cognitive

impairments associated with drug. We investigated the impact of CBD on intravenous THC self-administration in adolescent male and female rats and establish if it protected against diminished

performance on a cognitive task. We also investigated the protracted and acute effects of THC on prefrontal cortical activity during the cognitive task. METHODS: Adolescent male and female

rats self-administered escalating doses of THC (3, 10, 30, and 100 µg/kg/infusion) alone or combined with CBD (10:1 ratio). In adulthood, rats were trained on a delayed-match-to-sample

working memory task which requires responding after increasingly difficult delays. A separate group of adolescent rats intravenously self-administered THC or a control vehicle solution

throughout adolescence. Rats expressing CaMKII.GCaMP6f in the prelimbic prefrontal cortex performed the working memory task during single-photon calcium imaging sessions. These animals were

given acute injections of THC (0.5 mg/kg, i.p.) or vehicle 30 min before testing. RESULTS: Male and female adolescent rats acquired THC or THC + CBD self-administration with no effect of

drug solution or sex. No significant difference in working memory performance emerged based on sex or adolescent self-administration condition. In rats that self-administered THC or vehicle

during adolescence that received acute injections of THC, females significantly reduced responding on the working memory task compared to males regardless of prior drug exposure. Neuronal

activity increased across all groups but the proportion of differentially modulated neurons corresponding to task accuracy was altered only in females without adolescent THC exposure.

CONCLUSIONS: These results demonstrate that the presence of CBD did not promote THC self-administration in adolescence or impact performance on a cognitive task in adulthood. Additionally,

acute THC enhanced cortical activity regardless of adolescent exposure yet primarily impacted behavior and the associated neuronal response in adult females that were not exposed to THC

during adolescence. DISCLOSURE: Nothing to disclose. 1.3 A RESPONSE-CONTINGENT CANNABIS VAPOR DELIVERY APPROACH FOR INTERROGATING PROTRACTED EFFECTS OF MATERNAL CANNABIS USE IN DEVELOPING

OFFSPRING RYAN MCLAUGHLIN WASHINGTON STATE UNIVERSITY, PULLMAN, WASHINGTON, UNITED STATES BACKGROUND: Cannabis use during pregnancy is a growing public health concern and its impacts on

developing offspring remain elusive. This is partly due to a lack of animal studies that use volitional drug delivery models in pregnant dams. To this end, we developed a model of maternal

cannabis use that employs response-contingent delivery of vaporized cannabis extracts in rat dams and examined whether prenatal cannabis exposure alters emotionality, vapor

self-administration, and corticostriatal inputs in adulthood. METHODS: Sprague Dawley rat dams self-administered cannabis vapor (69.8% Δ9-THC; 150 mg/mL) (n = 11) or vehicle (4:1 PG:VG) (n =

 13) vapor on a fixed-ratio-1 schedule twice daily in hour-long sessions throughout mating and gestation. Isolation-induced ultrasonic vocalizations (USVs) were measured in early life and

behavior in the elevated plus maze and novelty-suppressed feeding tests were assessed in adulthood. Other offspring self-administered cannabis vapor or vehicle vapor according to an

escalating schedule of reinforcement over 21 days during adulthood. Remaining offspring received injections of fluorescent retrobeads (200 nl/side) into the nucleus accumbens core (NAc)

during adulthood and excitatory postsynaptic currents (sEPSCs) were measured in NAc-projecting medial prefrontal cortex (mPFC) neurons. RESULTS: Cannabis dams showed greater discrimination

for the cannabis-paired operandum (p = .003) and earned more vapor reinforcers than vehicle dams (p = .066). Cannabis-exposed pups weighed less than vehicle- (p < .001) and air-exposed

pups (p = .039) on postnatal day (PND) 1 and emitted more USVs on PND6 (p < .05) but no differences in anxiety-like behavior were observed in adulthood. Female offspring self-administered

significantly more cannabis vapor and had higher break points during a progressive ratio challenge irrespective of prenatal exposure condition (both p’s < .05). Conversely, male

cannabis-exposed offspring showed reduced responding for all vapor in adulthood (p = .032). Cannabis-exposed offspring had higher sEPSC frequencies in NAc-projecting mPFC neurons, indicating

altered excitability within the corticostriatal pathway. CONCLUSIONS: These results support the use of the cannabis vapor self-administration approach to investigate long-term effects of

maternal cannabis use in developing offspring. DISCLOSURE: Nothing to disclose. 1.4 PROGRESS AND HURDLES IN ESTABLISHING VAPOR SELF-ADMINISTRATION IN RODENTS JACQUES NGUYEN BAYLOR

UNIVERSITY, WACO, TEXAS, UNITED STATES BACKGROUND: Cannabis is commonly used via inhalation, i.e., smoking and, more recently vaping, of the plant material and/or cannabis extracts. Due to a

high base rate of exposure, there are more people who meet criteria for cannabis dependence than have used cocaine in the past month, or ever tried heroin. Pre-clinical models of the

self-administration of cannabis, extracts or active constituents have been limited and there are only a few recent attempts at inhalation self-administration in rodents. METHODS: Groups (N =

 8) of male Wistar rats were permitted to respond on nose-poke manipulanda for deliveries of propylene glycol (PG) vapor adulterated with ∆9-tetrahydrocannabinol (THC; 50-100 mg/mL)

delivered into sealed vapor exposure chambers through the use of controllers that trigger commercial e-cigarette tanks. The response requirement was incremented from Fixed Ratio 1 to 5 and

dose substitution (12.5, 25, 100 mg/mL) was assessed in a counterbalanced order. In one group intake was limited to 5 vapor deliveries for 24 sessions before evaluating ad libitum

responding. RESULTS: Rats exposed themselves to 10-20 minutes of THC vapor inhalation, a duration which produces cannabinoid typical effects with non-contingent exposure. Mean vapor

deliveries were not systematically changed across up to 50 sessions of training, even as the FR was incremented. Rats obtained more infusions of the 12.5 mg/mL concentration than when 100

mg/mL was available. Responses on the drug-associated manipulandum were below 80% of responses. Ad libitum intake was increased after release of the limitation to 5 vapor deliveries.

CONCLUSIONS: Rats will make responses for deliveries of THC infused vapor in daily sessions, exposing themselves sufficiently for a physiologically significant effect. Dose-dependent changes

in behavior can be observed as well as defense of exposure against increasing workload. Major limitations to the model include a lack of behavioral tuning on the drug-associated lever,

shallow dose-effect functions and individual differences. Much validation work remains to result in a convincing model of self-administration. DISCLOSURE: Nothing to disclose. STUDY GROUP 2.

DEVELOPING NOVEL CLINICAL OUTCOME ASSESSMENTS (COAS) FOR PSYCHIATRIC ILLNESSES LINDA BRADY*, SONYA EREMENC, RICHARD KEEFE, MADHUKAR TRIVEDI, MICHAEL SAND, KENNETH KOBLAN, ELIZABETH

STAFFORD, RACHEL STREIFF, SARAH LISANBY NATIONAL INSTITUTE OF MENTAL HEALTH, ROCKVILLE, MARYLAND, UNITED STATES STUDY GROUP SUMMARY: There are many challenges in the development of drugs for

the treatment of psychiatric disorders. This panel will focus on the development of fit-for-purpose clinical outcome assessments (COAs) to assess treatment benefit in clinical trials. COAs

are instruments that measure or reflect how patients feel, function, or survive. COA measures include patient-reported (PRO), clinician-reported (ClinRO), observer-reported (ObsRO), and

performance (PerfO) outcome measures. Optimizing the match between trial population, instrument, and endpoint will increase the validity and power of clinical trials, especially for the

development of new treatments. Ongoing and recent efforts as well as innovative approaches will be considered. Examples of new COAs for MDD to be discussed include the Symptoms of Major

Depressive Disorder Scale (SMDDS), which is a PRO measure with a 7-day recall period that was developed by the Critical Path Institute (C-Path) PRO Consortium Depression Working Group (WG).

The SMDDS was qualified by FDA as a PRO measure within a limited context of use (COU) in clinical trials to assess self-reported depression symptom severity in adults with MDD. The

Depression WG2.0 is developing two new COAs for MDD with the ability to assess the effectiveness of rapid acting antidepressants: 24 hour-recall and momentary assessment measures. Letters of

Intent (LOIs) for each measure were submitted to FDA’s COA Qualification Program (COAQP) and both measures were accepted into the qualification program. FDA operates the COAQP under the

21st Century Cures Act as part of its Drug Development Tool efforts. Qualification of a COA is a regulatory conclusion that the FDA finds the COA to be a well-defined and reliable assessment

of patients’ symptoms, overall mental state, or functioning within a specific context of use. FDA can also review COAs under a sponsor’s drug development program. Another example of a new

COA to be discussed is a PerfO measure for schizophrenia, the Virtual Reality Functional Capacity Assessment Tool (VRFCAT), a computerized tool developed as a measure of functional capacity

with the potential to translate to real-world functional improvements associated with cognitive change. The VRFCAT LOI was accepted into the COAQP as a proposed co-primary outcome measure in

schizophrenia treatment trials. It is currently being used as a secondary endpoint measure in sponsor drug trials in schizophrenia. The psychometric properties of VRFCAT and ongoing

development efforts will be discussed. With these case examples in mind, study group participants will discuss the COAQP process and lessons learned in the development of fit-for-purpose

COAs for use in clinical trials. Participants will also explore areas of unmet need for new COAs to enable innovation in treatment development. The participants include academic and

pharmaceutical investigators and clinical trialists, C-Path, advocacy organizations, an individual with lived experience, NIMH, and FDA. Collaboration is key for successful development of a

new COA. Qualitative research with patients, caretakers, and healthcare providers offers the foundational insight for instrument development and selection of the appropriate tool to fit the

research need. This study group offers a venue for Q and A with multiple stakeholders to consider the challenges and opportunities for advancing the development of new COAs for psychiatric

disorders. DISCLOSURE: Nothing to disclose. PANEL 3. NOVEL SYNERGISTIC TREATMENT COMBINATIONS TO LEVERAGE PLASTICITY INDUCED BY DRUGS ACTING AT NMDA RECEPTORS 3.1 A TALE OF TWO RECEPTORS:

SIMULTANEOUS, PROPHYLACTIC TARGETING OF NMDARS AND 5-HT4RS EXERTS ADDITIVE EFFECTS IN PROTECTING AGAINST STRESS-INDUCED ANXIETY-LIKE BEHAVIOR CHRISTINE ANN DENNY COLUMBIA UNIVERSITY IRVING

MEDICAL CENTER, NEW YORK, NEW YORK, UNITED STATES BACKGROUND: Serotonin (5-HT) receptors and N-methyl-D-aspartate receptors (NMDARs) have both been implicated in the pathophysiology of

stress and depression. Here, we evaluated whether combined dosing of (R,S)-ketamine, an NMDAR antagonist, and prucalopride, a 5-HT type 4 receptor (5-HT4R) agonist, would have additive

effects, resulting in improvements in stress-induced maladaptive behavior. METHODS: A single injection of saline, (R,S)-ketamine, prucalopride, or a combined dose of (R,S)-ketamine +

prucalopride was administered 1 week before a 3-shock contextual fear conditioning (CFC) paradigm in 129S6/SvEv mice of both sexes. Drug efficacy was assayed using a variety of behavioral

tests, including the forced swim test (FST), elevated plus maze (EPM), open field (OF), marble burying (MB), and novelty-suppressed feeding (NSF). Neural activity (e.g., c-fos

immunoreactivity) was quantified throughout the hippocampus. RESULTS: Prophylactic (R,S)-ketamine + prucalopride administration attenuated learned fear in male mice (n = 6-15 mice per group;

ANOVA, p = 0.0035) and decreased behavioral despair in both male and female mice (n = 6-15 mice per group; RMANOVAs, p = 0.0202 and p = 0.0133, respectively). Prophylactic (R,S)-ketamine +

prucalopride administration had an additive effect of decreasing stress-induced hyponeophagia in male and female mice (n = 5-15 mice per group; Mantel-Cox, p < 0.0001 and p = 0.0007,

respectively). Notably, prophylactic (R,S)-ketamine + prucalopride administration at lower doses was effective in the NSF, but not when administered separately at those doses. Combined

(R,S)-ketamine + prucalopride administration, but not either drug alone, significantly increased c-fos expression throughout the hippocampus (n = 5-9 mice per group; ANOVAs, p’s <

0.0001). CONCLUSIONS: Together, these results indicate that combined (R,S)-ketamine + prucalopride prophylactic administration exerts additive neural and behavioral effects compared to

administration of either drug alone. Our results suggest that combined administration of an NMDAR antagonist and 5-HT4R agonist has additive benefits for stress-induced pathophysiology,

providing preliminary evidence that future clinical studies using this combined treatment may prove advantageous. DISCLOSURES: Silo Pharma: Contracted Research (Self), Janssen, McLean:

Honoraria (talk given) (Self), Columbia University (numerous patents and provisional patents): Patent (Self), NIH Study Section: Honoraria (Self), Transformative Award, NINDS R21, NIA R21,

Columbia Trx Grant: Grant (Self), For the Love of Travis (donation to lab): Other Financial or Material Support (Self) 3.2 EFFICACY OF ADJUNCTIVE D-CYCLOSERINE TO INTERMITTENT THETA BURST

STIMULATION FOR MAJOR DEPRESSIVE DISORDER: A RANDOMIZED CLINICAL TRIAL ALEXANDER MCGIRR HOTCHKISS BRAIN INSTITUTE, UNIVERSITY OF CALGARY, CALGARY, CANADA BACKGROUND: Transcranial magnetic

stimulation is a non-invasive treatment for major depressive disorder (MDD) that is believed to depend on synaptic plasticity in targeted circuits. Protocols such as theta-burst stimulation

(TBS) are N-methyl-D-aspartate (NMDA) receptor dependent, suggesting that targeting the NMDA receptor could be an effective means of enhancing treatment outcomes. Here, we test whether low

doses of the NMDA receptor partial-agonist, D-Cycloserine, can enhance intermittent TBS (iTBS) treatment outcomes in MDD. METHODS: In this single site 4-week randomized double-blind

placebo-controlled trial, fifty participants with treatment resistant MDD were randomized 1:1 to iTBS+Placebo or iTBS+D-Cycloserine (100mg). Participants were asked to take the adjunct or

placebo at least 60 minutes prior to daily iTBS treatments for the first two weeks, and iTBS continued without an adjunct for weeks three and four. The primary outcome was change in

depressive symptoms as measured by the Montgomery Asberg Depression Rating Scale (MADRS) at the conclusion of treatment. Secondary outcomes included clinical response, clinical remission,

and clinical global impression (CGI). RESULTS: The iTBS+D-Cycloserine group had greater improvements in MADRS scores compared to the iTBS+Placebo group (mean difference -6.15, 95%CI -2.43 to

-9.88; Hedges’ g = 0.99, 95%CI 0.34-1.62). Rates of clinical response were higher in the iTBS-D-cycloserine than in the iTBS+Placebo group (73.9% vs 29.3%), as were rates of clinical

remission (39.1% vs. 4.2%). This was reflected in lower CGI severity ratings, and greater CGI-improvement ratings. No serious adverse events occurred. CONCLUSIONS: Adjunctive D-Cycloserine

is a promising strategy for enhancing transcranial magnetic stimulation treatment outcomes in MDD using iTBS. DISCLOSURE: Provisional Patent: Patent (Self) 3.3 A BRIEF, FULLY AUTOMATED

NEUROCOGNITIVE TRAINING INTERVENTION EXTENDS THE ANTIDEPRESSANT EFFECT OF A SINGLE KETAMINE INFUSION: CLINICAL AND NEUROIMAGING FINDINGS JARED KOPELMAN UC SAN DIEGO, SAN DIEGO, CALIFORNIA,

UNITED STATES BACKGROUND: Intravenous ketamine (ket) is posited to rapidly reverse depression by enhancing neuroplasticity. We posited that ketamine would provide a ‘window of opportunity’

in which to introduce automated neurocognitive training techniques to consolidate adaptive forms of cognitive processing while neuroplasticity remains high. METHODS: 154 unipolar depressed

adults who failed at least one antidepressant medication were randomized to one of three arms: (1) a single infusion of ket (0.5mg/kg over 40min) followed by 4 days (~2.5 hrs total) of

active ‘automated self-association training’ (ASAT), targeting self-worth through ‘evaluative conditioning,’ (ket+ASAT;n = 53); (2) ket infusion followed by 4 days of sham training

(ket+Sham;n = 50); or (3) saline (sal) infusion followed by 4 days of training (sal+ASAT;n = 51). The Montgomery-Asberg Depression Rating Scale quantified depression severity over a 30 day

period. In a subset of patients (67 ket, 31 sal) whole-brain Diffusion Tensor Imaging (DTI) data were collected prior to and 24 hrs post-infusion. DTI mean diffusivity (DTI-MD), a putative

marker of neuroplasticity, was calculated for 7 ROIs (left and right BA10, left and right amygdala, and left and right hippocampus, and vACC). RESULTS: Ket rapidly reduced depression scores

at 24 hrs post-infusion (β*=-1.30;t150 = -4.29;p < .0001; 52% responders vs. 25% in sal arm). Over the 30-day acute phase, ket+ASAT produced a stable enduring decrease in depression

relative to saline+ASAT (β*=-0.61;t148 = -3.62;p = .0004). By contrast, depression scores following ket+Sham followed an increasing linear trajectory from 24 hrs to 30 days, as the post-ket

effect waned and depression approached sal+ASAT levels (group*time relative to saline+ASAT: β*=0.015;t568 = 2.35;p = .019). In follow-up, the ket+ASAT arm showed continued benefits on

self-reported depression 3 mo post-infusion (d = .57;t83 = 2.6;p = .01). Our neuroimaging results indicated that individual differences in DTI-MD in medial prefrontal and limbic ROIs were

associated with improvement in depression scores (p’s < .05), primarily in the ket group. CONCLUSIONS: In the present study, ASAT extended the rapid antidepressant effect of a single dose

of ket to at least 30 days. In addition, our DTI data indicate the acute effects of ket on depression may be mediated, at least in part, by acute changes in plasticity. DISCLOSURE: Nothing

to disclose. PANEL 4. CILIA GPCRS IN NEUROPSYCHIATRIC DISORDERS 4.1 PRIMARY CILIA SIGNALING SHAPES EXCITATORY SYNAPTIC CONNECTIVITY JIAMI GUO HOTCHKISS BRAIN INSTITUTE, UNIVERSITY OF

CALGARY, CALGARY, CANADA BACKGROUND: In the mouse cerebral cortex, pyramidal neurons, the main class of excitatory neurons in the cortex, elaborate a tree-like structure called a dendrite,

which contains thousands of small protrusions called dendritic spines to receive synaptic inputs from other neurons. The appropriate size and shape of dendritic arbors and spines is a key

determinant of how neurons receive, integrate, and encode circuit information. Knowing how dendrites and dendritic spines acquire their form during development is thus essential to

understanding the emergence of functional neural circuits. It is clear that environmental signals, such as diffusible morphogen cues, have major influences on dendrite morphogenesis12. We

recently found that primary cilia serve as centralized signaling hubs regulating dendrite morphogenesis. This discovery challenges the existing view that the extracellular environmental cues

primarily act directly on the dendritic membrane to control dendritic shape. Instead, it identifies an additional, undefined, centralized signaling mechanism that stems from the primary

cilia positioned at the neuronal soma. METHODS: We generated mice with ciliary gene deletion specifically in glutamatergic excitatory neurons. Dendrite and spine morphology was analyzed by

confocal microscopy. Synaptosome was analyzed using proteomics. Behavioral changes were analyzed by behavioral assays. AAV viruses were injected into the prefrontal cortex in the adulescent

stage to induce deletion of ciliary genes and changes in social dominance were examined. RESULTS: Dendritic growth is reduced, and spine morphology is changed in ciliary mutant neurons.

Synaptosome shows changes in protein composition in ciliary mutants. Mutant mice display changes in motor and cognitive behaviors. Induced ciliary gene deletion in the adulescent stage leads

to changes in spine dynamics and social dominance. CONCLUSIONS: Upon completion of this project, we will have gained transformative insights into how neuronal primary cilia function to

regulate dendrite morphogenesis. These efforts will help uncover hitherto undefined cell biological mechanisms fundamental for neurodevelopment and functional wiring of the brain.

DISCLOSURE: Nothing to disclose. 4.3 PRIMARY NEURONAL CILIA LOSS IMPACTS BEHAVIORAL RESPONSES TO PSYCHOSTIMULANTS IN CELL-SPECIFIC MANNERS JEREMY MCINTYRE UNIVERSITY OF FLORIDA, COLLEGE OF

MEDICINE, GAINESVILLE, FLORIDA, UNITED STATES BACKGROUND: In recent years, neuronal primary cilia have garnered much attention for their role in a variety of neurobehavioral contexts

including drug responses, thus offering a new landscape to explore. Primary cilia are microtubule-based organelles that project from the surface of nearly all mammalian cells, including

neurons. They function as signaling hubs and are enriched with a diverse array of GPCRs, including several known to be associated with motivation and drug-related behaviors; however, our

understanding of how cilia regulate neuronal function and behavior is still limited. We previously showed that neuronal cilia loss alters locomotor responses to amphetamine in a cell-type

specific manner. The objective of the current study was to investigate the contributions of primary cilia on specific neuronal populations to behavioral responses to cocaine. METHODS: To

test the consequences of cilia loss on cocaine-induced locomotion and reward-related behavior, we selectively ablated cilia from dopaminergic or GAD2-GABAergic neurons in male and female

mice, using DAT:CRE or GAD2:Cre respectively. Locomotor effects were tested to both acute and repeated administration of cocaine. Conditioned-Place preference was used to analyze differences

in the rewarding effects of cocaine in mice lacking cilia. Data were analyzed use two-factor ANOVAs. RESULTS: Cilia ablation on either dopaminergic or GAD2-GABAergic neurons failed to

significantly alter acute responses to cocaine at a range of doses (3, 10, and 30mg/kg). With repeated administration, mice lacking cilia on GAD2-GABAergic neurons exhibited enhanced

locomotor sensitization to cocaine at 3mg/kg compared to wild-type littermates ((F(1,21) = 6.448, p = .019)), whereas mice lacking cilia on dopaminergic neurons exhibited reduced locomotor

sensitization to cocaine at 10 and 30mg/kg (F(1,17) = 6.458, p = .021)). To test the rewarding effects of cocaine we used a conditioned place preference test. Mice lacking cilia on

GAD2-GABAergic neurons show no difference in cocaine CPP. However, mice lacking cilia on dopaminergic neurons exhibit reduced CPP compared to wild-type littermates. CONCLUSIONS: Combined,

our results show that behavioral effects of cilia ablation are cell- and drug type-specific, and that neuronal cilia regulate both the locomotor-inducing and rewarding properties of cocaine.

DISCLOSURE: Nothing to disclose. PANEL 5. STRESS AND SOCIAL DETERMINANTS AFFECT PREGNANCY AND EARLY DEVELOPMENT: A CRITICAL WINDOW FOR GENERATIONAL STUDIES 5.2 UNEQUAL BEGINNINGS: MATERNAL

LIFE STRESS AND HEALTH INEQUITIES AFFECT THE NEXT GENERATION CATHERINE MONK COLUMBIA UNIVERSITY VAGELOS COLLEGE OF PHYSICIANS AND SURGEONS, NEW YORK, NEW YORK, UNITED STATES BACKGROUND:

Preterm birth (PTB) affects 1 in 10 women, is the leading cause of neonatal mortality and morbidity, and is associated with future risk for poor mental (ADHD) health. In the U.S., racial and

ethnic disparities in PTB exist independent of socio-economic status: overall, 11.37-9.38% for Hispanic versus 9.10% for non-Hispanic White women. Psychosocial stress and childhood trauma

are associated with risk for PTB. PTB has an intergenerational impact: mothers born preterm are more likely to give birth preterm. Biomarkers to predict PTB have proven unsuccessful, and do

not account for the recognition of intergenerational transmission via maternal heritage. Mitochondria, which contain their own genome, the mitochondria DNA, are inherited from the mother and

represent a potential intersection point between psychosocial experiences and their biological embedding in PTB. METHODS: In early 2nd trimester in n = 187, 27 data points are obtained from

questionnaires, ambulatory diaries, and physical assessments to phenotype stress groups. In a subsample of women, placental RNA-seq data is currently available allowing for summary scores

of cellular energetic pathways to determine whether maternal stress phenotypes differ in mitochondrial-related gene expression in the placenta (n = 37). RESULTS: Three latent profiles of

maternal stress were identified: 66.8% Healthy; 17.1% Psychologically Stressed, 16% Physically Stressed. Hispanic vs Non-Hispanic PTB was 16% vs 9%. For Hispanic women but not non-Hispanic

White women, stress group was associated with risk of PTB (OR = 5, 95% CI 1.26-19.9; OR = 1.33, 95% CI: 0.13-14.2). Psychologically stressed mothers showed higher transcript levels for

mitochondrial biogenesis and respiratory chain transcript accompanied by lower metabolic sensing and lower glycolysis transcript expression, suggesting they might rely more on placental

mitochondrial energy production than healthy pregnant women (Fig. 1). CONCLUSIONS: These findings from RNA-seq show that psychologically stressed mothers have higher expression of a gene

involved in making new mitochondria and mitochondrial respiratory chain proteins, suggesting that placenta from these mothers are making more ATP; we plan to look directly at the enzymatic

activity of the respiratory chain and test that hypothesis. DISCLOSURE: Curio: Advisory Board (Self) 5.3 THE IMPACT OF SOCIAL DETERMINANTS OF HEALTH ON INFANT BRAIN DEVELOPMENT CYNTHIA

ROGERS WASHINGTON UNIVERSITY SCHOOL OF MEDICINE, ST. LOUIS, MISSOURI, UNITED STATES BACKGROUND: Social determinants of health like exposure to poverty, neighborhood crime, and racial

discrimination are increasingly recognized as having deleterious effects on development. What is less well known is how exposure to these social determinants during the prenatal period may

impact the developing brain. This presentation will review data from ongoing longitudinal studies of prenatal exposure to social determinants and neonatal and infant brain development.

METHODS: These studies include approximately 400 caregiver-infant dyads that were predominantly Black and enriched for exposure to adversity. Participants were recruited during pregnancy

with assessments of income to needs, neighborhood poverty, neighborhood crime exposure, racial discrimination, depression, perceived stress, and stressful life events. Infants underwent MRI

scans during the neonatal period. Image analyses included resting state functional connectivity, diffusion tractography of white matter tracts and structural analyses of volume and surface

area with a focus on brain regions related to emotion regulation and emotion processing. Latent factors encompassing social disadvantage variables and psychosocial stress variables were

created and related to neonatal brain measures. RESULTS: Prenatal social disadvantage was significantly related to the functional connectivity of cortical networks (R2 = 0.29, p < .05)

but particularly with frontoparietal, default-mode, and ventral attention networks. Similarly, prenatal social disadvantage was related to neonatal diffusion measures of the cingulum bundle

(β = -.24, p < .001) the uncinate fasciculus (β = -.16, p < .01) and the fornix (β = -.11, p < .05). Social disadvantage was also related to hippocampal (R2 0.22, p = .009) and

amygdala (R2 0.41, p = .009) volumes. Psychosocial stress was not significantly related these measures when controlling for social disadvantage. CONCLUSIONS: Prenatal exposure to social

determinant of health factors particularly those that index social disadvantage and poverty were related to multiple measures of neonatal brain development including cortical networks, white

matter tracts and subcortical regions that are related to emotion regulation and emotion processing. Future work will relate these findings to early childhood developmental outcomes.

DISCLOSURE: Nothing to disclose. 5.4 TRANSLATIONAL APPROACHES TO THE IMPACT OF STRESS ON MATERNAL MICROBES AND THE NEXT GENERATION TAMAR GUR OHIO STATE UNIVERSITY COLLEGE OF MEDICINE,

COLUMBUS, OHIO, UNITED STATES BACKGROUND: Previous studies show that exposure to prenatal stress has long term consequences on offspring. Work from our lab pinpoints the maternal microbiome

and the chemokine CCL2 as key mediators of long term neuroinflammation and reductions in social behavior in rodents. Here we address the central role of fetal CCL2 and extend our preclinical

findings into a prospective cohort clinical study. METHODS: Preclinical Study: Intra-amniotic injections of recombinant CCL2 or saline were performed on E16.5. Tissues were collected 24

hours after the procedure to measure concentration of CCL2 in maternal and fetal tissues by ELISA. A second cohort went through parturition and social behavior was measured in adulthood in

the 3 chamber social approach paradigm. Clinical Study: A prospective longitudinal study (N = 38) was performed in the peripartum period with maternal rectal and vaginal swabs and cord blood

collected. Assessment included measures of perceived stress (PSS), anxiety, depression (CESD). PacBio full-length 16S rRNA was used to identify microbial communities. RESULTS: Preclinical:

Following injection of CCL2 into the amniotic sac, CCL2 was increased in the fetal plasma, fetal liver, and fetal brain (n = 6-7; p < 0.001.) CCL2 concentrations were not impacted in the

maternal plasma or placenta. Female offspring exposed to CCL2 demonstrated increased neuroinflammation in prefrontal cortex (CCL2, TNFα; n = 6-7; p < 0.05) and did not exhibit social

preference (n = 6-7; p = 0.134), in contrast to the control group (n = 6-7; p = 0.00029). Clinical: Fecal diversity metrics in 2nd trimester were associated with increased stress (PSS) and

depression (CESD) scores. In the 2nd Trimester, Fecal Prevotella was increased in mothers that reported higher PSS. PSS metrics at delivery were associated with decreased CCL2 levels in the

cord blood. CONCLUSIONS: In a preclinical model we show amniotic CCL2 injection is sufficient to induce neuroinflammation and long-term reduction in social behavior. In a clinical study we

found maternal perceived stress and depressive symptoms are associated with alterations in maternal microbiota and cord blood CCL2 levels. Leveraging preclinical and clinical research can

expedite our mechanistic understanding of how prenatal stress is transmitted to the next generation. DISCLOSURE: Nothing to disclose. STUDY GROUP 6. NAVIGATING THE MAZE: HOW YOU AND YOUR

UNIVERSITY CAN PREPARE AND PROTECT YOUR LAB AGAINST THE RISKS ASSOCIATED WITH ANIMAL RESEARCH GEORGIA HODES*, MARGAUX KENWOOD, ANDRE DER-AVAKIAN, ERIC NESTLER, WILLIAM CARLEZON, JAMES

JENTSCH, MAR SANCHEZ, ZOE DONALDSON, EDYTHE LONDON, JILL TURNER VIRGINIA POLYTECHNIC INSTITUTE AND STATE UNIVERSITY, BLACKSBURG, VIRGINIA, UNITED STATES STUDY GROUP SUMMARY: Animal research

has produced some of the most important scientific developments of the past century, including the recent development of a new class of mRNA vaccines that has saved countless human lives.

Calls to protect research animals over the past 40 years have produced important regulations that aid in the welfare of animals and humans alike. However, there is a strong political voice

calling for the end of all animal research. As a result, there is a growing danger in the form of threats and harassment to scientists, their laboratories, and institutions. The members of

this study group will discuss their personal experiences with being targeted by animal rights activists. The participants have a wide range of experiences with different forms of animal

research and will include information on how institutional support, or lack thereof, impacted their response to threats and harassment. We will use this as a starting point for a discussion

with the audience about the following: * 1. Key measures to take before a lab is targeted, including discussions to have with your lab members and institution, so they are prepared and know

how to respond. * 2. What to do if an anti-science group targets you, or your laboratory. * 3. How to talk to the public about animal research so they can understand and appreciate the

importance of this form of science and the vast regulations already protecting research animals. Our study group discussions will help basic and clinical scientists to better understand the

issues faced by animal researchers, along with the regulations and protections already instituted to protect research animals. By engaging in this discussion, we hope to help scientists

safely navigate the growing complexity surrounding this area of science, which is critical to treating physical and mental health. DISCLOSURE: Nothing to disclose. PANEL 7. IS THE

MODIFICATION OF WHITE MATTER MICROSTRUCTURE A VIABLE TREATMENT STRATEGY FOR PSYCHOPATHOLOGY? 7.1 MYELINATION REGULATES EXPERIENCE-DEPENDENT NEURONAL PLASTICITY IN THE MAMMALIAN CORTEX WENDY

XIN UCSF, SAN FRANCISCO, CALIFORNIA, UNITED STATES BACKGROUND: Developmental myelination is a protracted process in the mammalian brain, occurring postnatally in mice and continuing through

the first three decades of life in humans. One theory for why oligodendrocytes mature so slowly posits that myelination may stabilize neuronal circuits and temper neuronal plasticity as

animals age. We tested this hypothesis in the visual cortex, which has a well-defined critical period for experience-dependent neuronal plasticity. METHODS: To prevent myelin progression, we

conditionally deleted Myrf, a transcription factor necessary for oligodendrocyte maturation, from oligodendrocyte precursor cells (Myrf cKO) in adolescent mice. To induce plasticity, adult

control and Myrf cKO mice were monocularly deprived by eyelid suture. Functional and structural neuronal plasticity in visual cortex were assessed by in vivo optical intrinsic imaging (n = 8

control, 9 cKO) and in vivo longitudinal two photon imaging of dendritic spines (n = 5 CTL, 4 cKO), respectively. Both male and female mice were included in histological and in vivo imaging

experiments. RESULTS: The density of mature oligodendrocytes and myelin sheaths progressively increased from P28 to P60 in control mice but plateaued by P28 in Myrf cKO mice. Following

monocular deprivation, visual cortex activity in response to visual stimulation of the deprived eye remained stable in adult control mice, as was expected based on previous experiments in

wildtype mice. By contrast, visual cortex responses to the deprived eye decreased significantly following monocular deprivation in Myrf cKO mice, reminiscent of the plasticity observed in

adolescent mice (mean difference in visual cortex responses 4 days post-monocular deprivation: control -0.33% vs cKO -17.78%; p = 0.0003). Furthermore, the density of dendritic spines in

visual cortex was reduced following monocular deprivation in Myrf cKO, but not control, mice. CONCLUSIONS: These results support the concept of myelin acting as a brake on neuronal

plasticity during development and suggest that oligodendroglia and myelination play a critical role in shaping the maturation and stabilization of cortical circuits. DISCLOSURE: Nothing to

disclose. 7.2 EARLY LIFE STRESS IMPAIRS PERFORANT PATHWAY (PP) DEVELOPMENT ARIE KAFFMAN YALE UNIVERSITY SCHOOL OF MEDICINE, NEW HAVEN, CONNECTICUT, UNITED STATES BACKGROUND: Early life

stress (ELS) impairs normal myelination across diverse mammalian species, but the mechanisms responsible for this developmental abnormality are yet to be clarified. Here we show that pups

raised with limited bedding (LB), a mouse model of ELS, have significant myelination deficits in the developing perforant pathway (PP) in the hippocampus that are due to sparse axonal

innervation from the entorhinal cortex. METHODS: Balb/CByj mice were exposed to control condition (500 cc bedding, 5 X 5 nestlet) or LB (125cc, no nestlet) from postnatal day 0 (P0) to P25.

RNA was harvested from the hippocampus at P17 and sequenced at a depth of 80MR/sample (n = 7-9 mice per rearing and sex group). Expressed transcripts (TPM > 3, total of 13,245 genes) were

analyzed using DESeq2 script in R to identify differentially regulated genes with false discovery rate (FDR) < 0.05, and to map major pathways affected by LB using MetaCore™ (Clarivate

Analytics). Confocal microscopy confirmed abnormal myelination and revealed reduced axonal innervation in the PP (n = 4-5, P17 pups per rearing and sex group). Additional tissue was

collected after contextual fear conditioning at P33 for high resolution dMRI and retrograde tracing (n = 6-8 mice per rearing and sex group). RESULTS: RNA-seq identified 747 differentially

regulated genes (FDR < 0.05) with abnormal myelination as the most significant pathway impaired by LB (FDR = 3.83e-11). LB reduced the expression of genes necessary for oligodendrocyte

differentiation but not oligodendrocyte progenitor cells proliferation. Follow-up Immunohistochemistry studies showed that the PP is one of the most myelinated structures in the developing

hippocampus and confirmed reduced oligodendrocyte maturation in the PP of LB mice (F (1, 16) = 35.56 P < 0.0001, np2 = 0.69). Abnormal myelination was associated with reduced axonal

markers (F (1, 16) = 9.88, P < 0.0063, np2 = 0.38) suggesting that LB impaired axonal pathfinding and connectivity between the entorhinal cortex and the dorsal hippocampus. This assertion

was further confirmed using diffusion MRI tractography and retrograde tracing and was correlated with abnormal contextual freezing in P33 juvenile mice. CONCLUSIONS: Sparse PP connectivity

between the entorhinal cortex and the dorsal hippocampus contributes to hippocampal-dependent deficits in juvenile LB mice. DISCLOSURE: Nothing to disclose. 7.3 ASSESSING THE EFFECTS OF

SOLIFENACIN, AN ANTIMUSCARINIC AGENT, ON ANXIOUS BEHAVIORS AND WHITE MATTER MICROSTRUCTURE IN NON-HUMAN PRIMATES NAKUL AGGARWAL UNIVERSITY OF WISCONSIN, MADISON, WISCONSIN, UNITED STATES

BACKGROUND: Myelination subserves efficient neuronal communication, and alterations in white matter (WM) microstructure have been implicated in numerous psychopathologies, including

pathological anxiety. Recent work in rodents suggests that muscarinic antagonists may enhance myelination with behavioral benefits. Here, we present data from a first-in-primate study

exploring the effects of solifenacin on anxious behaviors and WM microstructure in non-human primates (NHP), as a potentially translatable therapeutic strategy for human pathological

anxiety. METHODS: 12 preadolescent rhesus macaques (6 control, 6 experimental; 8 F, 4 M) were included in a pre-test/post-test control group study design. The experimental group received 3

months of daily IM injections of solifenacin succinate (0.5 mg/kg); controls received vehicle. Subjects underwent pre- and post-assessments of: 1) anxious temperament (AT)-related behaviors

in the potentially threatening no-eye-contact (NEC) paradigm (30-min); and 2) WM imaging metrics, including diffusion tensor imaging (DTI) and quantitative relaxometry (QR). RESULTS: We

found significant Time (pre vs. post) by Group (control vs. experimental) interactions with respect to both freezing and cooing during the NEC. Specifically, experimental subjects, compared

to controls, exhibited effects consistent with a reduction in anxiety: reduced freezing (p = 0.01) and increased cooing (p = 0.004) post-treatment. Whole-brain voxelwise analyses of

post-minus-pre differences in DTI and QR metrics suggested increases in WM specific to the experimental group (all p’s < 0.005, uncorrected), including: increased fractional anisotropy in

left stria terminalis and superior cingulum; decreased radial diffusivity (low values~greater myelination) in left sagittal striatum and stria terminalis; increased axial diffusivity (high

values~greater axonal integrity) in right uncinate fasciculus, right external capsule, left dorsal prefrontal WM, and left superior corona radiata; and increased longitudinal relaxation

rates (high values~greater myelination) in bilateral inferior frontal gyrus. CONCLUSIONS: Our findings from this first-in-primate study support further investigation of the utility of

antimuscarinic agents in targeting WM microstructure as a means to reduce levels of pathological anxiety. DISCLOSURE: Nothing to disclose. 7.4 SPECIFICITY AND CAUSALITY OF WHITE MATTER

ABERRANCIES IN YOUTH PSYCHOPATHOLOGY JULIA LINKE NATIONAL INSTITUTES OF HEALTH, NIMH, BETHESDA, MARYLAND, UNITED STATES BACKGROUND: Aberrant maturation of white matter (WM) tracts and

changes in WM microstructure in reaction to specific experiences (e.g., early adversity, psychopathology) have been proposed as pathophysiological mechanisms across mental disorders.

However, tracts implied in different disorders largely overlap. Further, causality within these WM-behavior relationships still needs to be established. Data presented here addresses

questions regarding specificity and causality. METHODS: We obtained diffusion tensor imaging (DTI) data in two independent samples. Specificity questions were addressed in 144 youth (mean

age 12.9 years (8-18), 68 girls, with anxiety disorders, attention-deficit/hyperactivity disorder, disruptive mood dysregulation disorder, or without a diagnosis). Youth and parents provided

anxiety, irritability, inattention, and hyperactivity ratings that were subjected to a bi-factor model, thereby parsing unique and shared aspects of these co-occurring symptoms. Preliminary

analyses regarding causality were conducted in the context of an ongoing trial, where DTI data are acquired before and after an exposure-based cognitive behavioral therapy (CBT; 23 anxious

youth: 12.3 years (8-15), 13 girls; 20 age and gender-matched controls). DTI data was processed with TBSS. RESULTS: In our cross-sectional analyses the shared factor was associated with

widespread alterations in interhemispheric and fronto-limbic connections (x = 12, y = -9, z = 31, k = 17688, pFWER = .007). The anxiety-specific factor was associated with alterations in the

splenium of the corpus callosum (x = -17, y = -52, z = 23, k = 435, pFWER = .008), while the irritability-specific factor was linked to the corticospinal tract (x = -25, y = -19, z = 34, k 

= 401, pFWER = .009). There were no findings for the ADHD-specific factors. In the treatment study, lower fractional anisotropy in the clusters associated with the shared (ab path = .031)

and anxiety-specific factor (ab path = .031) in the cross-sectional analysis, partially mediated reduction in anxiety symptoms. CONCLUSIONS: Findings implicate alterations in fronto-limbic

WM circuitry as a transdiagnostic mechanism of youth psychopathology. Furthermore, they suggest changes in WM microstructure as a mediator of CBT-response. DISCLOSURE: Nothing to disclose.

MINI PANEL 8. FACE EXPRESSION AND GAZE DYNAMICS IN SCHIZOPHRENIA AND ITS RISK STATES 8.1 GAZE FOLLOWING IN THIRD-PARTY OBSERVERS OF SIMULATED SOCIAL CONFLICT SHOWS FEW REFLEXIVE AND MANY

MENTALIZING FEATURES KATALIN GOTHARD THE UNIVERSITY OF ARIZONA, TUCSON, ARIZONA, UNITED STATES BACKGROUND: Most primates exchange gaze-mediated social signals in the form of gaze-following

and joint-attention saccades. Gaze following saccades start from the eye/face of the social partner and follow their gaze direction. Joint-attention saccades land on the specific object that

is explored by the social partner. METHODS: We quantified the scan paths of 3 macaques who watched videos of simulated social conflict between conspecifics. Conflict was simulated by

juxtaposing two videos depicting a threatening and an appeasing individual, oriented toward each other, with the timing of the facial and bodily displays adjusted to mimic an exchange of

social signals. Motion and the presence of facial expressions were registered in ethograms. Gaze-following and joint-attention saccades (henceforth JAGF) were hand-scored based on

pre-established criteria. RESULTS: From the 3 viewers we have analyzed 3,743 trials and identified 14,686 saccades that meet the criteria for JAGF. In two viewers, JAGF saccades occurred

during frames that had significantly less motion than average (chi-square p < 0.001, N = 306); the third viewer showed the opposite effect (chi-square p < 0.001, N = 86). All 3 viewers

produced more JAGF saccades in response to facial expressions (chi-square p < 0.001, N = 14,185). Viewers followed preferentially the gaze of the threatened monkeys, who elicited more

joint attention than gaze-following saccades. This way the viewer explored vicariously the source of threat perceived by the appeasing individual. The temporal distribution of JAGF saccades

showed clusters aligned to specific video segments. Some clusters showed significant increases in JAGF saccades across all 3 viewers, others were viewer-specific. Clusters often contained

sequences of JAGF saccades, with the attention of the viewer rapidly alternating between the eyes and face of the interacting individuals. CONCLUSIONS: The clustering of JAGF saccades and

their enhancement by facial expressions argue for reflexive responses to visual cues. The low probability of JAGF saccades (>20%) however, argues against simple reflexivity. The biased

attention to the threatened animal, joint attention with animals who point with their gaze to a source of danger, and the large individual differences, suggest that the scan paths of

observers of conflict are informed by cognitive processes. DISCLOSURE: Nothing to disclose. 8.2 VISUAL SCANNING AND BRAIN DYNAMICS OF NATURALISTIC SOCIAL PERCEPTION IN SCHIZOPHRENIA GAURAV

PATEL COLUMBIA UNIVERSITY, NEW YORK, NEW YORK, UNITED STATES BACKGROUND: Visual scanning of naturalistic social scenes requires the dynamic interaction of cortical systems underlying visual

processing, face-emotion recognition, attention, and theory-of-mind (ToM) operations. Recently we found that these systems converge in the temporoparietal junction/posterior superior

temporal sulcus (TPJ-pSTS) and that TPJ-pSTS abnormalities affect social perception in schizophrenia participants (SzP). To expand upon this work, first we detail similarities in visual

scanning deviations between SzP and those at clinical high-risk for schizophrenia (CHR). Second we examine differences in movie-evoked dynamic brain states between SzP and healthy controls

(HC) and their relationship to social perception. METHODS: In Experiment 1, 46 HC, 43 SzP, and 27 CHR performed The Awareness of Social Inference Test (TASIT) with eye-tracking. For each

TASIT video frame, eye-positions were scored as a z-transformed distance from the mean HC eye-position (distance/1 SD distance, HCs compared to a leave-self-out mean) and compared between

intervals where SzP were divergent (z-transformed distance>2) or not. In Experiment 2, 27 SzP and 21 HC watched a 15-minute video clip while BOLD-fMRI data was collected. K-means

clustering was used to cluster the BOLD time-courses into co-activation pattern (CAP) states. RESULTS: In Experiment 1, the CHR eye-positions deviated more during the SzP divergent vs.

non-divergent periods (t(26) = 5.6, p < 10-5). In Experiment 2, 5 CAP states were isolated, including one where all TPJ-pSTS areas were activated. Fractional occupancy of this state was

lower in SzP (t(46) = 3.1, p = 0.003), correlated with TASIT accuracy (r = 0.51, p = 0.008), and the degree of visual motion in the movie (p < 10-13). We also found a mode consisting of

the TPJ-pSTS, ToM and frontoparietal CAP states that was evoked only when viewing social interactions. CONCLUSIONS: In Experiment 1, the shared gaze divergence in CHR and SzP suggests a

common underlying mechanism. In Experiment 2, the involvement of the TPJ-pSTS state in viewing and understanding of social interactions highlights the importance of coordinated activation of

underlying areas for processing of those scenes. The results demonstrate the utility of naturalistic stimuli in identifying moments of abnormalities, which may aid in refining the timing of

neurostimulation delivery (e.g., TMS) to enhance treatment effects. DISCLOSURE: Pfizer, Inc, Employee, Spouse 8.3 CORRELATION OF INTER-SPEAKER FACIAL EMOTION ESTIMATES DURING DYADIC

INTERACTIONS IN PEOPLE WITH SCHIZOPHRENIA AND ITS RISK STATES STEPHEN HEISIG ICAHN SCHOOL OF MEDICINE AT MOUNT SINAI, NEW YORK, NEW YORK, UNITED STATES BACKGROUND: Abnormal face expression

is a sign of schizophrenia typically assessed by clinical scales. We characterize individual facial expressions in 10 healthy controls (HC), 7 schizophrenia (SZ), and 20 clinical high risk

(CHR) patients and their correlations with interviewer expressions in the context of open-ended interviews using software models. We hypothesize group differences in individual expression

and dyadic correlation. METHODS: Open-ended 30-minute interviews were recorded using Zoom. We used the OpenFace facial analysis toolkit 2.0 (Baltrušaitis et. al. IEEE 2018) to estimate face

action units. 48 emotional intensity estimates of facial expression per video frame were generated using HUME AI (Cowan et.al. Nature 2020) software. We calculated the overall mean and

percent present for each of the 18 action units captured by OpenFace, focusing on AU07, (orbicularis oculi), which is key in social signaling. From the time-locked interviewer and

participant video frames, we generated the dyadic correlation of each emotion estimate throughout the interview. The action unit and emotional estimate model labels are treated as objective,

mechanistic features to describe the faces. RESULTS: We replicated prior findings in SZ of decreased overall AU means (p = 0.001) and in particular AU07, (orbicularis oculi) (p = 0.018)

extending these findings to CHR patients. Using HUME AI features no group differences were found in emotion estimates. However, we did find significant differences in patient/interviewer

dyadic correlations (as compared to healthy/interviewer dyads) for 20 of the 48 emotion estimates (Doubt, Joy, Confusion, Contempt, Realization, Amusement, Determination, Anxiety, Anger,

Empathic Pain, Tiredness, Contemplation, Entrancement, Awkwardness, Distress, Envy, Concentration, Disappointment, Excitement, Boredom), (all p < 0.03.) CONCLUSIONS: Employing a novel

computational technique, we found decreased and disordered facial synchrony during naturalistic conversation in patients with SZ and individuals at CHR for SZ. Decreased facial synchrony in

patients may be related to decreased face motor activity, as well as deficits in visual and/or face emotion processing, or disturbance in gaze; this will be the focus of future study.

DISCLOSURE: Nothing to disclose. MINI PANEL 9. HIGHLIGHTING PSYCHOSIS NEUROIMAGING RESEARCH IN LATIN AMERICA: INFLAMMATION, ACCELERATED AGING, AND ENVIRONMENT 9.1 SYSTEMIC INFLAMMATION AND

CORTICAL NEUROCHEMISTRY IN NEVER-MEDICATED INDIVIDUALS WITH PSYCHOSIS CAMILO DE LA FUENTE-SANDOVAL INSTITUTO NACIONAL DE NEUROLOGIA Y NEUROCIRUGIA, MEXICO CITY, MEXICO BACKGROUND: Studies on

cellular and cytokines profiles have contributed to the inflammation hypothesis in schizophrenia; however, precise inflammatory dysfunction markers remain elusive. Proton magnetic resonance

spectroscopy (1H-MRS) studies have shown higher levels of myo-inositol (mI) and choline-containing compounds (Cho) in patients with first-episode psychosis (FEP), suggesting

neuroinflammation. Here, we present inflammatory profiles in antipsychotic-naive FEP patients and age-and-sex matched healthy controls (HC), as well as cortical mI and Cho levels using

1H-MRS. METHODS: Inflammatory profiles were analyzed using cytokine spontaneous production of peripheral blood mononuclear cells (PBMCs), and upon mitogenic stimulation, in 48 FEP patients

and 23 HC. 1H-MRS was performed on a 3T scanner, centered on the medial prefrontal cortex and analyzed with LCModel (FEP = 27, HC = 17). Inflammatory markers were compared between groups and

correlations with 1H-MRS findings were also explored. RESULTS: FEP patients (duration of untreated psychosis 232 (±415) weeks (range 1-1720) showed higher proportion of proinflammatory

Th1/Th17 subset (median .265) compared with HC (median .11; two tailed Mann Whitney U = 325, p = .03). Also, FEP patients revealed an increased spontaneous production of IL-6 (median 2340,

HC median 2.695; U = 193, p = .02), IL-2 (median 11.61, HC median 3.86; U = 196, p = .03), and IL-4 (median 3.785, HC median 2.05, U = 144 p < .01). Similarly, FEP subjects showed higher

Cho levels (mean 1.69 ± 0.22, HC mean 1.37 ± 0.56, t = 2.72, df = 42, p = .01). No differences in mI levels or other neurometabolites were found. Lastly, Cho levels correlated with T

regulatory cells (r26 = .44, p = .02), and with classic monocytes (r26 = -.53, p = .004). CONCLUSIONS: FEP subjects were characterized by immune dysregulation, affecting both the innate and

adaptive immune response, with a predominantly Th2 signature. Furthermore, increased Cho levels were also found and correlated with T-regulatory cells. The intense Th2 signature and

proinflammatory response found in subjects, along with the 1H-MRS findings, suggest changes that can be associated with both systemic and central inflammation processes in schizophrenia.

DISCLOSURE: Janssen: Contracted Research (Self) PANEL 10. LET’S TALK: PSYCHOACTIVE DRUGS ALTER THE STRUCTURE AND CONTENT OF SOCIAL SPEECH 10.1 DOSE-DEPENDENT EFFECTS OF PSILOCYBIN ON

LANGUAGE PRODUCED UNDER NATURAL CONDITIONS ENZO TAGLIAZUCCHI UNIVERSIDAD ADOLFO IBAÑEZ, BUENOS AIRES, ARGENTINA BACKGROUND: While the effects of low and high doses of psilocybin have been

investigated in laboratory settings using standardized tasks and questionnaires, the validity of this approach can be questioned due to the influence of contextual factors. We report the

results of two studies conducted under natural conditions based on the analysis of free unconstrained speech: a study of language produced during the effects of a psilocybin microdose (study

1) and a study of verbal interactions between a group of participants who consumed psilocybin during a retreat (study 2). METHODS: Study 1 followed a double-blind placebo-controlled design

(N = 35, 0.5 g of dried psilocybin mushrooms or an inert placebo). Natural language was recorded during the acute effects, when participants answered a series of questions about different

topics. Study 2 was a single blind randomized study (N = 100 participants divided into two matched groups, receiving 1 g or 3 g of dried psilocybin mushrooms, administered by a facilitator

in the context of a retreat). Natural language was recorded during group interactions after the acute effects. Both studies included male and female participants. Language samples were

analyzed in terms of verbosity, sentiment analysis, semantic coherence, and topic detection. Pairwise comparisons were conducted using Wilcoxon (Study 1) or Mann-Whitney (Study 2)

non-parametric tests. Mushroom samples were analyzed using liquid chromatography-mass spectrometry. RESULTS: In Study 1 we found that a low dose of psilocybin (close to 1 mg of psilocybin)

resulted in enhanced positive sentiment and higher verbosity. In Study 2 we found that a higher dose of psilocybin (in the range of 10 mg) resulted in both positive sentiment and higher

verbosity (with longer turns taken by participants in the conversation), and decreased semantic coherence, but with smaller effect sizes than those reported in a previous study of LSD. The

prevalence of topics related to visual perception and ego dissolution was significantly increased in the high vs. low dose condition. CONCLUSIONS: We conclude that natural language is

sensitive to the reported effects of low and high doses of psilocybin consumed in different contexts. Natural language analysis should be further explored as a potentially valuable tool to

quantify the contents of unconstrained reports after psychedelic experiences. DISCLOSURE: Nothing to disclose. 10.2 EFFECTS OF MDMA AND AMPHETAMINE ON SPEECH AND SOCIAL CONNECTION DURING

DYADIC CONVERSATIONS HANNA MOLLA UNIVERSITY OF CHICAGO, CHICAGO, ILLINOIS, UNITED STATES BACKGROUND: Although MDMA and amphetamine are drugs that increase social behavior, MDMA has been

characterized as a unique ‘empathogenic’ drug that induces pro-social effects, which may be observed through quantitative linguistic analysis. METHODS: We conducted a double-blind,

placebo-controlled study in which male and female participants (N = 36) engaged in a dyadic conversation with partners of the same-sex after MDMA or amphetamine. RESULTS: The study is still

ongoing and the results we present will be new data. CONCLUSIONS: The study is still ongoing and the results we present will be new data. DISCLOSURE: Nothing to disclose. 10.3 NATURAL

LANGUAGE PROCESSING OF MOVIE RECALL AND DRUG FLUENCY, AND ASSOCIATED BRAIN FUNCTION, IN COCAINE AND HEROIN ADDICTION RITA GOLDSTEIN ICAHN SCHOOL OF MEDICINE AT MOUNT SINAI, NEW YORK, NEW

YORK, UNITED STATES BACKGROUND: Individuals with addiction attribute excessive salience to drug/drug-related cues, which could enhance performance on neuropsychological probes. For example,

individuals with cocaine addiction outperform healthy controls (HC) when prompted to generate drug-related words. Here we tested this effect in individuals with heroin use disorder (iHUD),

in addition testing for verbal recall following watching of a heroin-related movie, before and after 14-weeks of treatment. We aimed to test for group differences and the potential reduction

in drug-related verbal fluency and recall biases following inpatient treatment. METHODS: The drug fluency task was performed by 32 inpatients with iHUD (age=40.8 ± 9.2, 6 women) and 16 HC

(age=43.8 ± 10.3, 6 women). Participants watched the first 17 min of the movie Trainspotting, containing drug and nondrug scenes. After watching the movie, participants verbally recalled it.

Fourteen iHUD performed the tasks again following 14 weeks of treatment, when adherence was tracked by daily EMAs. RESULTS: The iHUD named more drug-related words than HC on the verbal

fluency task (18.4 ± 5.8 vs. 15.6 ± 3.3 p = .02). In the recall task, iHUD proportionally recalled more drug>nondrug scenes (recall frequency: .7 ± .2 vs. .6 ± .2, p = .045) with a

similar pattern for recall duration (p = .024). While fluency, recall frequency or duration did not change with treatment (ps > .60), there was a trend for less decrease in drug over

nondrug recall frequency between the sessions to be associated with longer lifetime heroin use (r = .5, p = .051). Also, longer drug over nondrug recall duration at the post-treatment

session was associated with worse treatment adherence in the iHUD (r = .6, p = .018). CONCLUSIONS: Here we identify simple and naturalistic, ecologically-valid, speech-based

neuropsychological measures that tap language, memory/retrieval and higher order executive functions, revealing a drug bias in drug addiction. Associations with addiction severity and

treatment adherence suggest these contextually-biased measures may be biomarkers of clinical endpoints. Current efforts are geared towards the use of machine learning (e.g., natural language

processing) for the exploration of these speech-based biomarkers vis-à-vis brain function. DISCLOSURE: Nothing to disclose. 10.4 ARTIFICIAL INTELLIGENCE APPROACHES TO STUDY SPEECH

PRODUCTION RELATED TO DRUG USE GUILLERMO CECCHI IBM RESEARCH, YORKTOWN HEIGHTS, NEW YORK, UNITED STATES BACKGROUND: The assessment of psychological states prior, during and after

psychoactive drug use traditionally relies on a combination of subjective reports and clinicians’ evaluation. The development of automated language analytics allows for the application of a

limited but consistent metric across arbitrarily large drug-related speech datasets, including samples produced in different media and in naturalistic settings. METHODS: We studied 60

regular users of cocaine engaged in an abstinence trail, analyzing baseline open-ended speech samples in which they describe the positive and negative consequences of withdrawal. Applying

Natural Language Processing (NLP) techniques, we measured the semantic similarity of the samples to concepts related to drug use and quality of life, in order to predict outcomes at 12

months after baseline. The predictions were validated using 10-fold subject-based cross-validation. RESULTS: Several of the outcome variables at 12 months, including Cocaine Selective

Severity Assessment and Cravings, were predicted from baseline speech samples with significant accuracy, whereas baseline clinical and demographic variables including age, gender, CSSA and

years of use did not provide any predictive value. CONCLUSIONS: These preliminary results suggest that psychological processes present at baseline and expressed in open-ended speech can be

captured by NLP and used to significantly constrain the expected range long-term outcomes of abstinence treatment. DISCLOSURE: IBM Corporation: Employee (Self) PANEL 11. NEW TRICKS FOR OLD

CELLS: PLASTICITY AND NOVEL ROLES OF HYPOTHALAMIC CRH NEURONS IN MENTAL HEALTH AND DISEASE 11.3 AN UNEXPECTED ROLE FOR CRHPVN NEURONS IN THE SOCIAL BUFFERING OF STRESS JAIDEEP BAINS

HOTCHKISS BRAIN INSTITUTE, UNIVERSITY OF CALGARY, CALGARY, CANADA BACKGROUND: The presence of others, either during or soon after a stressful experience can mitigate the harmful effects of

stress. This phenomenon, known as social buffering, has been described extensively, but the mechanistic underpinnings of social buffering remain elusive. The CRH neurons of the

paraventricular nucleus of the hypothalamus (CRHPVN) are necessary for the social transmission of stress, but their potential role in social buffering of stress is not known. METHODS: Here,

in mice, we use in vivo imaging, optogenetics, behavioral evaluations and ex vivo electrophysiology to evaluate the role of CRH neurons in the paraventricular nucleus in social buffering

after acute stress. RESULTS: We used an established novel object recognition (NOR) task to test for the immediate effects of acute stress on memory. After one day of training in the NOR

chamber, we footshocked mice and re-exposed them to the chamber. We calculated a discrimination index (DI) to evaluate memory (DI = (Timenovel − Timefamiliar)/ (Tnovel + Tfamiliar). Acute

stress (footshock) decreased the discrimination index in male and female mice, consistent with a memory impairment. In female mice, the presence of a naive conspecific in the homecage after

stress prevented stress-induced impairments in memory. By contrast, in males, the presence of a conspecific had no effect on stress-induced memory impairments. In females, social

interactions after stress decrease synaptic metaplasticity that relies on increased CRHPVN activity (Sterley et al, 2018). To determine whether CRHPVN neurons contribute to memory

impairments after stress, we optogenetically silenced these cells in single FS females immediately after stress. This was sufficient to restore novel object discrimination. CONCLUSIONS:

Social buffering, which occurs when a stressed individual interacts with an unstressed conspecific, helps relieve negative effects of stress specifically in females. This includes a decrease

in stress-induced memory impairments specifically. This effect of social interaction on memory is mimicked in single-housed stressed mice by photoinhibition of CRHPVN neurons. This direct

inhibition also results in a reduction in stress-induced short-term potentiation of glutamate synapses on these cells, implicating CRHPVN neurons as key mediators of the stress alleviating

effects of social contact. DISCLOSURE: Enveric Biosciences: Consultant (Self) STUDY GROUP 12. DEVELOPMENTAL COHORT STUDIES PAST, PRESENT, AND FUTURE: LESSONS LEARNED AND DESIGN FOR CLINICAL

IMPACT ARISTOTLE VOINESKOS*, DEANNA BARCH, DAMIEN FAIR, THEODORE SATTERTHWAITE, MICHAEL MILHAM, STEPHANIE AMEIS, JUDITH FORD UNIVERSITY OF TORONTO, TORONTO, ONTARIO, CANADA STUDY GROUP

SUMMARY: The present study group application will highlight efforts to build developmental cohorts to better understand risk trajectories of mental illness, in the context of biological,

psychological, and social factors. There is growing appreciation for the developmental onset of mental illness during childhood or adolescence. Nearly 75% of cases of mental illness onset

before the age of 25, and mental illnesses such as depression and schizophrenia are the top causes of disability in youth. There is even evidence showing that disorders that onset early in

the lifespan, such as autism or ADHD, can themselves be risk factors for disorders that onset in later adolescence such as psychotic disorders. Large cohort studies across the developmental

phase of life offer the opportunity to understand different trajectories of illness and brain development that children and youth experience. Through data sharing and re-use, hundreds if not

thousands of investigators around the world are now asking new questions accelerating the pace of scientific discovery. Cohorts are often population, or catchment-area based, but more

recently new studies are recruiting help-seeking young people (and families). Studies have also evolved from cross-sectional to longitudinal. The advent of digital tools and technologies

mean that additional real-world data can be collected that might index physiological functions and offer new insights regarding symptoms that can be difficult to characterize, such as

insomnia. Finally, there is growing awareness of structural racism and other social disadvantages that minority populations face. Not only are such assessments now included, but active

equity-based approaches to ensure inclusion and retention of such populations is now expected. The title of the study group, reflecting past, present, and future, aims to take lessons

learned, successes, and challenges of recently completed, and in progress cohort studies, and help the field in optimizing design of newly initiated, or to-be initiated studies. The

Philadelphia Neurodevelopmental Cohort, a study of ~ 1,600 children and youth from the greater Philadelphia area (population-based) will serve as an exemplar of a cohort study recently

completed. The Adolescent Brain and Cognitive Development (ABCD) study that is now underway having completed its first wave on ~10,000 9 and 10 year olds will serve to provide early lessons

learned, as well as possibilities for adjustments in real time. Similarly the Human Brain Network (HBN) study, studying children 5-21 years of age (total sample anticipated n~10,000) based

on ascertainment in the greater New York area via parents concerned about potential symptoms or distress in their children will provide an example of how to pivot following a pilot phase

analysis. Finally, the newly launched Toronto Adolescent and Youth (TAY) Cohort Study, aims to recruit ~3,000 children and youth from 11-24 years of age, who are referred to an academic

mental health hospital (in Toronto) to child and youth psychiatric services. The objectives (for this study group) relate to discussing challenges and opportunities in the following

categories: * 1. Population-based vs. clinical cohorts * 2. Data Sharing and Reuse * 3. Engagement and Recruitment of marginalized populations * 4. Data collection strategies (focus on tech

and wearables) * 5. Clinical Impact/Practice Change that could emerge from results DISCLOSURE: Nothing to disclose. STUDY GROUP 13. FOSTERING TRANSLATIONAL RESEARCH AND ACADEMIC/INDUSTRY

COLLABORATION TO ACCELERATE DEVELOPMENT OF NEW MENTAL HEALTH TREATMENTS KIMBERLY VANOVER*, ALIK WIDGE, ANDRE FENTON, DANIELLE GRAHAM, BITA MOGHADDAM, KERRY RESSLER, MARK SCHMIDT, SADE

SPENCER, STEPHEN STRAKOWSKI, JARED YOUNG ENGRAIL THERAPEUTICS, INC., SAN DIEGO, CALIFORNIA, UNITED STATES STUDY GROUP SUMMARY: The current biomedical basic research model and current

organization of funding opportunities are not designed to facilitate the translation of basic research to clinical practice in Psychiatry. Cost and complexity are among the top challenges

for translational research projects. Furthermore, academic / industry collaborations remain infrequent or are not incentivized for discovery and translation. However, the focus on

translational research is accelerating and impacts academic as well as industry research and is the focus of many governmental initiatives. This study group will emphasize the importance of

T1 translational research, bringing findings from basic research into early clinical testing in humans, as well as T2 translational research, establishing effectiveness in humans and

establishing clinical guidelines. We will discuss what needs to change to facilitate more translational research to accelerate the development of new treatments for disorders of the central

nervous system as well as what is at risk to be lost with a change in focus and funding. Specifically, better tools and broader funding opportunities are needed to support translational

models for the translational researcher. This Study Group brings together representatives from academia and industry, across different stages of career, and inclusive of diversity. Our

objective is to bring together leaders to discuss translation of basic research to clinical practice and the opportunities and challenges for academic / industry collaboration. Questions for

discussion will include, but are not limited to: * What are the major gaps facing translational neuroscience in psychiatry? * How can academic/industry collaborations help fill the gaps? *

How do we better de-risk target pipelines at the preclinical level to achieve better prediction of translation in humans? * How do we enhance discovery pipelines for translationally robust

biomarkers for precision medicine in Psychiatry? * Are standards and rigor in animal research sufficient? How do we keep up with the focus on reproducibility and larger sample sizes now

required in human work? * Do the current funding structures favor basic research and/or should more funding be invested in translational work and appropriately powered clinical trials? * Do

many journals favor basic research or clinical studies, but not translational findings? * Is there a gap in expertise, such that preclinical data may not be well understood by clinical

reviewers and vice versa, diminishing competitiveness of papers with both animal and human data? DISCLOSURE: Engrail Therapeutics: Employee (Self), Intra-Cellular Therapies: Stock / Equity

(Self), Evolution Research Group: Consultant (Self) PANEL 14. RACIAL DISPARITIES IN MENTAL HEALTH DURING THE COVID-19 PANDEMIC 14.1 RACE, CYTOKINES, AND DEPRESSION: THE ROLE OF THE EXPOSOME

OLUSOLA AJILORE UNIVERSITY OF ILLINOIS AT CHICAGO, CHICAGO, ILLINOIS, UNITED STATES BACKGROUND: Previous studies have demonstrated that racial discrimination and the perceived stress of

racism have been associated with elevations in pro-inflammatory cytokines. Given the well-documented relationship between pro-inflammatory cytokines and major depression, the goal of the

present study was to examine whether there were racial differences in the relationship between inflammation and major depression using a secondary analysis of a ethnically diverse dataset.

METHODS: 121 subjects were recruited as part of a larger program of research investigating major depressive disorder (MDD) at the University of Illinois at Chicago (UIC). 46 were Black (21

healthy comparison subjects, 25 with MDD) and 75 were White (34 healthy comparison subjects, 41 with MDD). The final inclusion criterion for depressed subjects was a score of ≥15 on the

Hamilton Rating Scale for Depression. Severity of depression was also assessed using the Center for Epidemiological Studies – Depression scale. In addition to C-reactive protein (CRP),

levels of pro-inflammatory cytokines were determined in plasma/serum aliquots by enzyme-linked immunosorbent assay (ELISA) using commercially available Quantakine® kits (R and D Systems,

Inc., Minneapolis, MN) for human IL-1β, IL-6, and TNF-α. Socioeconomic factors were obtained from the Chicago Health Atlas and linked to each participant by the neighborhood where they

resided at the time of the study. RESULTS: In the non-depressed sample, Black participants had significantly higher levels of CRP and IL-6 than White participants. In the context of major

depression, there was no difference between groups. After controlling for BMI, CRP was no longer significant in the non-depressed group, but IL-6 remained significantly different between

groups. After controlling for socioeconomic factors, IL-6 levels were no longer significant between the two groups. Additionally, IL-6 was not correlated with depression severity in Black

participants but was in White participants. CONCLUSIONS: The well-known relationship between pro-inflammatory cytokines and major depression may be moderated by race due to high levels of

cytokines at baseline in Black participants. These elevations may be related to the stressful life events due to discriminatory experiences as indexed by socioeconomic factors and

environmental stressors. DISCLOSURES: KeyWise AI: Founder (Self), Embodied Labs, Blueprint, Milken Institute, Sage Therapeutics: Advisory Board (Self) 14.2 RACIAL DISPARITIES IN COVID-19

STRESS: THE ROLE OF INFLAMMATION APRIL THAMES UNIVERSITY OF CALIFORNIA, LOS ANGELES, CALIFORNIA, UNITED STATES BACKGROUND: The COVID-19 pandemic has disrupted the lives of billions, raising

concerns about overall mental health. Further, data from the Centers for Disease Control demonstrate significant racial disparities in COVID-19 outcomes for African Americans. The current

study examined the predictive association between baseline symptoms of depression and inflammation on COVID-19 related stress outcomes (as measured by the Pandemic Stress Index [PSI];

Harkness, 2020). African American (n = 66) and non-Hispanic White participants (n = 50) aged 30-60 (Mage = 45.4; SD = 9.0) were recruited from a larger study on racial disparities in

cognitive outcomes. METHODS: African American (n = 66) and non-Hispanic White participants (n = 50) aged 30-60 (Mage = 45.4; SD = 9.0) were recruited from a larger study on racial

disparities in cognitive outcomes. For circulating (plasma) levels of inflammation-related biomarkers, we will assess both pro- and anti-inflammatory cytokines using a multiplex assay (IL-6,

IL-10, TNF-a) (R and D Systems Luminex Performance Human High Sensitivity Cytokine Panel). Additional biomarkers (sCD14, CRP, MCP-1/CCL2) will be determined by ELISA (R and D Systems Human

Quantikine ELISAs), as they cannot be multiplexed with the cytokines or with each other due to differences in circulating concentrations and assay requirements. RESULTS: Higher PSI scores

were reported among African Americans in comparison to non-Hispanic Whites. African Americans demonstrated higher levels of CRP in comparison to non-Hispanic Whites, F (2, 113) = 7.66, p = 

.02. Linear regression was used to examine predictive associations baseline depression, race, CRP as well as interactions on PSI scores. As expected, the overall model predicted PSI scores,

(r2 = .18, p < .0001). Independent associations were found between depression (b = .57, p < .001), CRP (b = .32, p < .02), race (b = -.27, p = .03) and the CRP X depression

interaction (b = .22, p = .04) on PSI outcomes. The association between depressive symptoms and CRP on PSI scores were in the expected direction. The relationship between depression and PSI

scores were strongest among those with high levels of CRP. CONCLUSIONS: Study results provide preliminary evidence for the role of inflammation in the relationship between depression and

COVID-19 pandemic stress scores. DISCLOSURE: Nothing to disclose. 14.3 ASSOCIATIONS BETWEEN SEGREGATION, CORONAVIRUS RACIAL BIAS, AND COVID-19 PANDEMIC DISTRESS RODMAN TURPIN DEPARTMENT OF

GLOBAL AND COMMUNITY HEALTH, FAIRFAX, VIRGINIA, UNITED STATES BACKGROUND: Racial/ethnic minorities are disproportionately impacted by the COVID-19 pandemic, as they are more likely to

experience structural and interpersonal racial discrimination, and thus social marginalization. Based on this, we tested for associations between pandemic distress outcomes and four

exposures: racial segregation, coronavirus-related racial bias, social status, and social support. METHODS: We collected data as part of a larger longitudinal national study on mental health

during the pandemic (n = 1,301). We tested if county-level segregation, individual-level social status, social support, and coronavirus racial bias were associated with pandemic distress

using cumulative ordinal regression models, both unadjusted and adjusted for covariates (gender, age, education, and income). RESULTS: Both our segregation index (PR = 1.19; 95% CI 1.03,

1.36) and coronavirus racial bias scale (PR = 1.17; 95% CI 1.06, 1.29) were significantly associated with pandemic distress. Estimates were similar after adjustment for both segregation (aPR

 = 1.15; 95% CI 1.01, 1.31) and coronavirus racial bias (PR = 1.12; 95% CI 1.02, 1.24). Social status (aPR = 0.74; 95% CI 0.64, 0.86) and social support (aPR=0.81; 95% CI 0.73, 0.90) were

associated with lower pandemic distress after adjustment. CONCLUSIONS: Segregation and coronavirus racial bias are relevant pandemic stressors, and thus have implications for minority

health. Future research exploring potential mechanisms of this relationship, including specific forms of racial discrimination related to pandemic distress and implications for social

justice efforts, are recommended. DISCLOSURE: Nothing to disclose. 14.4 STRESSORS DURING COVID: RACIAL BIAS AND DRUG USE BRENDA CURTIS NATIONAL INSTITUTE OF HEALTH/NIDA, BALTIMORE, MARYLAND,

UNITED STATES BACKGROUND: Black, Asian, and Latinx people are disproportionately impacted by the COVID-19 pandemic and they are more likely to experience coronavirus-related racial

discrimination. Prior to the COVID-19 pandemic, racial and ethnic minority adults reported similar or lower levels of alcohol and other drug use than their white counterparts. The pandemic

appears to have begun to reverse these trends with racial and ethnic minorities reporting larger increases in substance use during the pandemic. This study examined the association among

coronavirus-related victimization distress, perceptions of pandemic associated increase in societal racial biases, and substance use risk among Asian, Black, Latinx, and non-Hispanic White

adults. METHODS: Participants were part of a larger longitudinal survey study on mental health and substance use during COVID-19. Adults (N = 1336) who self-identified as Asian (8.53%),

Black (10.55%), Latinx (10.93%), and non-Hispanic White (69.99%) completed measures that included demographic and COVID-19 related stressors, the coronavirus victimization distress scale

(CVD), the coronavirus racial bias scale (CRB), and drug use consumption (alcohol, tobacco, cannabis, and other drugs). RESULTS: Across race/ethnicity, binary logistic regression analyses

controlling for demographic variables indicated coronavirus victimization distress was associated with higher odds of tobacco use risk (AOR = 1.36, 95% CI [1.01, 1.81]) and coronavirus

racial bias beliefs were associated with higher odds of drug use including cocaine, hallucinogens, inhalants, and methamphetamine (AOR = 1.31, 95% CI [1.00, 1.71]). Logistic regressions for

each racial/ethnic group found different patterns of relationships between CVD, CRB and substance use disorder (SUD). CONCLUSIONS: Results highlight the significance of examining how the

current pandemic has exacerbated racial/ethnic systemic inequalities through COVID-19 related victimization. The data also suggest that across all racial/ethnic groups perceptions of

pandemic instigated increases in societal racial bias is a risk factor for SUD. The study calls for further empirical research on substance use treatment sensitive to specific needs of

diverse populations during the current and future health crises. DISCLOSURE: Nothing to disclose. PANEL 15. THE FORCES THAT SHAPE US: INTERACTIONS OF EARLY EXPERIENCE, SEX, AND PUBERTY ON

THE DEVELOPMENT OF CIRCUITRY AND BEHAVIOR 15.1 SEX-SPECIFIC INFLUENCES OF EARLY EXPERIENCE AND PUBERTAL TIMING ON THREAT RESPONSIVENESS IN RATS HEATHER BRENHOUSE NORTHEASTERN UNIVERSITY,

BOSTON, MASSACHUSETTS, UNITED STATES BACKGROUND: Individuals who endured traumatic environments in childhood commonly struggle with responding appropriately to threats throughout life.

Genetic sex contributes to the nature and timing of these effects, with evidence that risk is unmasked in females following periods of hormonal activation, including puberty. Pubertal timing

itself appears to be affected by early life adversity, evidenced by earlier puberty initiation in females. I will present data from an animal model illustrating sex-specific changes to

threat responses in rats exposed to maternal separation (MS), using a novel behavioral paradigm. I will also discuss potential mechanisms underlying precocial puberty after MS, as well as

preliminary findings suggesting that heightened estrogen signaling can cause enhanced threat responsiveness in females. METHODS: Males and females (n = 6-12) underwent control rearing or MS

postnatal days (P)2-20. One cohort was tested for baseline acoustic startle response (ASR) on P25, P35, or P55 on Day 1, and on Day 2 ASR was preceded with presentation of a social threat

cue [playback of a 22 Hz ultrasonic vocalization (USV)] in the ASR chamber. Another cohort was sacrificed at P10, P15, or P15 and hypothalamus was collected for qPCR analyses of

puberty-related gene expression (GnRH, RFRP, Kisspeptin). A third cohort received AAV-shRNA against estrogen receptor (ER)alpha into the basolateral amygdala (BLA) at P14, then were tested

for baseline and USV-potentiated ASR at P55. RESULTS: MS enhanced baseline ASR at P35 in females (p < 0.001), which was correlated with earlier age of puberty initiation. MS females had

heightened age-related increases of Kisspeptin mRNA in the anteroventral periventricular nucleus of the hypothalamus compared to controls. While USV presentation potentiated ASR in P55

controls, MS blunted this potentiation, and females overall showed lower USV-potentiation compared to males (p < 0.05). CONCLUSIONS: These data suggest that kisspeptin driven puberty

acceleration is associated with enhanced ASR in MS females. We are currently testing the extent to which ERalpha signaling in the BLA drives this heightened response. I will also discuss a

new translational model demonstrating a blunted ASR potentiation to social threat after early adversity, with potential mechanisms of risk, resilience, and adaptation in males and females.

DISCLOSURE: Nothing to disclose. 15.2 EARLY LIFE ADVERSITY ARRESTS PREFRONTAL CORTICAL MATURATION DURING ADOLESCENCE KUEI TSENG UNIVERSITY OF ILLINOIS, CHICAGO, ILLINOIS, UNITED STATES

BACKGROUND: Early life experiences have been known to impact the development and maturation of corticolimbic connectivity and its regulation of cognitive and affective behaviors. For

instance, individuals with a history of early life trauma are often at higher risk for anxiety-like disorders in adolescence, which are thought to result from a dysregulation of affective

and cognitive processes, and the maturation of the prefrontal cortex (PFC) and associated neural circuits. It is therefore conceivable that early life events contribute to shape the

developmental trajectory of PFC maturation during adolescence. METHODS: To test this idea, we implemented an early-life adversity paradigm (i.e., maternal separation, MS) and assessed to

what extent synaptic activity in the PFC and its maturation from postnatal day (P) 30 to adulthood becomes disrupted in both male and female rats. Both excitatory and inhibitory synaptic

activity were recorded and compared in PFC pyramidal neurons using a combination of ex-vivo whole-cell patch-clamp and DREADD techniques. RESULTS: Data revealed that the normal developmental

gain of PFC GABAergic synaptic activity was not observed in animals exposed to MS, while AMPA-mediated transmission was not affected. As a result of the selective GABAergic disruption in MS

animals, the excitatory-inhibitory (E-I) ratio that normally becomes balanced by P50 remains unbalanced (e.g., >1.0) in adulthood, resembling the level of E-I synaptic activity typically

found in the PFC of P30-40 rats. Moreover, transient chemogenetic inhibition of PFC GABA interneurons during this early adolescent period was sufficient to elicit similar E-I imbalance that

endures through adulthood. CONCLUSIONS: Collectively, our findings indicate that PFC inhibitory synapses are preferentially susceptible to early life adversity, likely through a disruption

of activity dependent mechanisms that are needed to enable the GABAergic maturation during adolescence. In turn, a disinhibited PFC state could increase the risk for psychiatric disorders

during adolescence with abnormal affective and cognitive responses. DISCLOSURE: Nothing to disclose. 15.3 EARLY SCARCITY ALTERS THE BASOLATERAL AMYGDALA TRANSCRIPTOME AND ADDICTION-RELATED

BEHAVIORS AMELIA CUARENTA TEMPLE UNIVERSITY, PHILADELPHIA, PENNSYLVANIA, UNITED STATES BACKGROUND: Adversity is a risk factor for psychiatric disorders, however, stress that is not

overwhelming can promote resilience. We use limited bedding and nesting (LBN) to model mild adversity. Our prior work shows LBN induces neurobiological alterations that reduce some

addiction-related behaviors including reducing morphine taking. These behaviors rely on cues as a driver of behavior and performance. Exposure to cues previously paired with drug taking can

induce craving and drug-seeking behavior following periods of abstinence. We are investigating whether LBN alters incubation of morphine craving, a cue-driven behavior; whether LBN alters

cocaine self-administration; the molecular changes induced by LBN in the basolateral amygdala (BLA), a region important for responses to stress and for the integration of cues. METHODS: Rats

were reared in LBN or control housing from postnatal days 2 - 9. In LBN, dams and pups were given a single paper towel for nesting material; a metal grate prevented access to bedding.

Control animals were reared in standard housing conditions. * 1: Rats were placed in operant boxes and permitted to lever press on a fixed ratio 1 (FR1) schedule for morphine infusions.

Presses on the active lever resulted in one infusion accompanied by a 5 s light cue and then a 20 s timeout period during which the house light was off; lever presses were recorded but drug

was unavailable. Rats were tested for behavioral signs of drug seeking on day 1 and day 30 of abstinence utilizing a within-subjects design. * 2: Rats had 6 hr cocaine self-administration

access on an FR1 schedule for 10 days. * 3: RNA sequencing was conducted to delineate the effect LBN had on the transcriptional profile of the BLA in adult rats. RESULTS: All groups showed

the incubation effect, pressing more after 30 days of abstinence than 1 day. There was no difference in lever pressing between conditions in males or females. Preliminary results suggests

that LBN male rats self-administer lower levels of cocaine than controls. LBN-induced sex-specific changes in transcription. RRHO analysis revealed distinct genes upregulated and

downregulated in males and females due to LBN. CONCLUSIONS: LBN reduces morphine and cocaine drug taking in male rats but does not affect cue-driven incubation of morphine craving. LBN also

induces sex-specific patterns of gene transcription within the BLA. DISCLOSURE: Nothing to disclose. 15.4 THE FORCES THAT SHAPE US: INTERACTIONS OF EARLY EXPERIENCE, SEX, AND PUBERTY ON THE

DEVELOPMENT OF CIRCUITRY AND BEHAVIOR ANNE MURPHY GEORGIA STATE UNIVERSITY - NEUROSCIENCE INSTITUTE, ATLANTA, GEORGIA, UNITED STATES BACKGROUND: Infants born prematurely are more likely to

be admitted to the Neonatal Intensive Care Unit (NICU) where they experience upwards of 10-18 painful procedures each day, often with no anesthesia or analgesia. Both pre-clinical and

clinical studies have shown that early exposure to pain disrupts normal CNS development resulting in lifelong changes in response to stress and pain. Here we present novel data on the

effects of neonatal injury on the response to an immune challenge in adulthood. METHODS: Male and female rats were exposed to a short-term inflammatory insult induced by intraplantar

administration of 1% carrageenan (CGN) on the day of birth (P0). In adulthood (P60-P90), rats were implanted with Thermicron iButtons to monitor core body temperature; 14 days later, rats

were injected with lipopolysaccharide (LPS) to elicit an immune response. Rats were sacrificed after 24 hours or at their peak fever point and brain tissue collected for analysis of VGat,

VGlut2, Fos and prostaglandin receptor 3 within the median preoptic area (MnPO), a key site for pyrexia. Peripheral and central cytokine/chemokine levels were also determined. RESULTS: LPS

induced a febrile response in all rats beginning ~3-5 hours post injection that lasted ~5-7 hours. Early life pain (ELP) exposed males (n = 11) and females (n = 7) showed a significantly

exaggerated febrile response to LPS in comparison to handled rats (males, n = 9; females n = 11). Center of gravity analysis revealed a significant rightward shift in the ELP females (t11 = 

3.163; p = 0.009). Peak fever was also significantly higher in ELP females (F1,16 = 4.619, p < 0.0473). Analysis of EP3R, VGlut2, and VGat at peak fever revealed no differences between

groups for EP3R (F(1,18) = 0.0102; p = 0.92) and VGlut2 (F(1,18) = 0.51; p = 0.48). Two-way ANOVA of VGat signal intensity showed a sex difference with females exhibiting higher expression

compared to males (F(1, 18) = 5.777; p = 0.0272) with no effect of treatment (F1,18 = 0.036, p = 0.852). CONCLUSIONS: Our results indicate that unresolved ELP alters neural circuitry

underlying immune-based fever response that persists into adulthood and is potentiated in females. Together, these studies may elucidate a mechanism through which children experiencing

unresolved pain during the perinatal period show an increased severity of sickness behavior and attenuated immune signaling. DISCLOSURE: Nothing to disclose. PANEL 16. HUMAN NERVOUS SYSTEM

BASED TRANSLATIONAL MODELS TOWARDS DISCOVERY OF NOVEL PAIN THERAPEUTICS 16.1 PAIN TARGET IDENTIFICATION IN HUMAN DORSAL ROOT GANGLION (DRG) AND SPINAL CORD STUDIES THEODORE PRICE UNIVERSITY

OF TEXAS AT DALLAS, RICHARDSON, TEXAS, UNITED STATES BACKGROUND: Pain therapeutic development has been plagued by clinical failures. Many rationales for this have been given, but one that is

often overlooked are species differences in transcriptomes and proteomes for neurons and other cells in pain circuits. The goal of this talk will be to impart detailed information about how

human nociceptors and spinal cord putative projection neurons differ between rodents and humans and how this information can be used to improve success in analgesic development. METHODS: We

have recovered dorsal root ganglion (DRG) and spinal cords from organ donors and DRGs from people having thoracic vertebrectomy and C1/C2 fusion surgeries. Tissues have been subjected to

bulk, single cell and spatial transcriptomic assays to understand cellular transcriptomes and how these transcriptomes change in chronic pain states in people. Male and female samples have

been used for all studies, and sex differences will be presented. RESULTS: We have identified 12 subtypes of human sensory neurons and investigated how these subtypes differ between humans

and mice. Similar comparisons have been made with spinal cord neurons between mice and humans, with a focus on differences in gene expression in neurons in the dorsal horn that are likely

projection neurons. We find important species differences that have implications for therapeutic development. In the DRG, the most important difference is that human nociceptors all have

peptidergic qualities while mouse nociceptors fall into both peptidergic and non-peptidergic subsets. Major differences in GPCR and neuropeptide expression are found between mice and human,

and many such receptors and peptides are exclusively found in humans. In projection neurons in the spinal cord, human neurons express Nav1.7 mRNA and protein, while mouse projection neurons

do not. These findings may have important implications for Nav1.7 clinical development. CONCLUSIONS: There are profound differences between transcriptomes and proteomes for human versus

mouse neurons of the pain pathway. We propose that this is a primary difference that causes clinical failures. Better understanding of the molecular architecture of human nociceptors and

projection neurons can improve our ability to create effective analgesics. DISCLOSURES: Acadia Pharma: Other Financial or Material Support (Self), 4E Therapeutics, Doloromics, NuvoNuro:

Board Member (Self), Grunenthal, Merck, Abbvie, Hoba Therapeutics: Grant (Self) 16.2 COMMUNICATION BETWEEN HUMAN SKIN CELLS AND NOCICEPTORS AS A DISCOVERY PLATFORM FOR PAIN TARGETS CHERYL

STUCKY MEDICAL COLLEGE OF WISCONSIN, MILWAUKEE, WISCONSIN, UNITED STATES BACKGROUND: Keratinocytes, which constitute >95% of the epidermis, are innately sensitive to mechanical force, are

capable of releasing a wide array of neuroactive factors and form close “synapse-like” connections with intraepidermal nerve fibers. Keratinocyte activity is critical for normal sensory

neuron and behavioral responses to mechanical and thermal stimuli. Skin becomes sensitized to external mechanical and thermal stimuli following many types of tissue injury, nerve injury and

disease. The contribution of keratinocytes to this injury-associated sensitization is unknown. METHODS: We used a variety of preclinical models of pain in both sexes of animals together with

keratinocytes from skin of human female and male donors in order to determine what ion channels, receptors and signaling molecules mediate the responses of keratinocytes from normal and

tissue injury conditions to tactile and thermal stimuli. We utilize behavioral assays, calcium imaging and patch clamp electrophysiology to define the roles of keratinocytes in vitro and in

vivo. Appropriate samples sizes based on power analyses were used for all studies. RESULTS: PIEZO1 is a bona fide mechanically-gated, non-selective cation channel that is highly expressed in

skin. Keratinocytes isolated from mouse and human skin respond to the PIEZO1 agonist Yoda1, PIEZO1 expression is critical for keratinocyte mechanical sensitivity. Furthermore, we

demonstrate that loss of epidermal PIEZO1 decreases the firing rate of sensory nerve fibers in response to mechanical stimulation of the skin and blunts behavioral responses to both

innocuous and noxious mechanical stimuli in vivo. Furthermore, keratinocytes isolated from mice with chronic painful diabetic neuropathy, sickle cell disease, traumatic nerve injury and

chemotherapy-induced neuropathy exhibit sensitization of PIEZO1 in the form of increased keratinocytes responding and increased magnitudes of responses. Moreover, treatment of human

keratinocyte with compounds that induce chemotherapy neuropathy in patients acutely sensitize PIEZO1 function. CONCLUSIONS: These data demonstrate that epidermal PIEZO1 is critical for

normal touch sensation and that keratinocyte PIEZO1 is sensitized in multiple models of neuropathic pain in preclinical models and in human skin. DISCLOSURE: Nothing to disclose. 16.4 HUMAN

CANCER BIOPSIES ELUCIDATE CANCER PAIN TARGETS NICOLE SCHEFF UNIVERSITY OF PITTSBURGH SCHOOL OF MEDICINE, PITTSBURGH, PENNSYLVANIA, UNITED STATES BACKGROUND: Head and Neck squamous cell

carcinoma (HNSCC) causes severe pain, beyond what is reported in other cancer types. We hypothesized that cancer immunosurveillance is impaired by the immunosuppressive actions of exogenous

opioids and immune cell-mediated endogenous opioid analgesia could be leveraged as a novel therapeutic tactic for cancer pain treatment. METHODS: HNSCC patient-derived single cell RNA

sequencing data were used to evaluate opioid receptor and opioid precursor gene expression in tumor infiltrating leukocytes (TIL) and peripheral blood. A syngeneic mouse oral cancer

orthotopic transplant model was used to evaluate analgesic morphine on the tumor immune response as well as immune cell-mediated endogenous opioid expression. Mouse oral cancer cell line 1

or 2 (MOC1, MOC2) tumor-bearing male and female mice were assessed for all studies. RESULTS: Single immune cell RNAseq analyses from of the 26 HNSCC patients (70% male, median age=60.5) and

6 healthy donors (50% male, median age=55.5) found that opioid receptor OPRM1 is expressed by T lymphocytes; there was a significant increase in OPRM1 expression in CD8+ T cells in tumor

compared to blood (padj=0.03). HNSCC patients with low reported pain had significantly higher immune-derived PNOC expression, the precursor for opioid, nociceptin (padj = 0.002). These

clinical data led to investigation of both morphine-impaired CD8+ T cell infiltration and immune-mediated endogenous analgesia in preclinical mouse models. Immune infiltration has been

quantified in a carcinogen-induced oral cancer mouse models; tumor-infiltrating neutrophils express and release beta-endorphin suppressing nociceptive behavior. However, intratumor

neutrophils can promote tumorigenesis. Exploration of opioid gene expression in alternative tumor-infiltrating immune cell subtypes was assessed using fluorescence-activated cell sorting and

qPCR. We found that tumor infiltrating B cells express more PNOC compared to B cells isolated from sham mice. CONCLUSIONS: Together these data suggest that exogenous opioids given for pain

management may directly impair tumor immunosurveillance and weaken immune checkpoint immunotherapy (ICI) therapeutic efficacy. Alternatively, ICI may be employed to stimulate the immune

response for endogenous opioid-mediated cancer pain treatment. DISCLOSURE: Nothing to disclose. MINI PANEL 17. THE CONUNDRUM OF PERINATAL ANXIETY: USING PHYSIOLOGICAL MEASUREMENTS TO REFINE

ASSESSMENT AND TREATMENT 17.1 ANXIETY THROUGH THE PERSPECTIVE OF THE MICROBIOTA-GUT-BRAIN AXIS MARY KIMMEL UNIVERSITY OF NORTH CAROLINA, CHAPEL HILL, NORTH CAROLINA, UNITED STATES

BACKGROUND: One in five meet criteria for an anxiety disorder during pregnancy or the postpartum period; Generalized Anxiety Disorder (GAD) the most common. Anxiety can serve a role in

protecting parent and child, but anxiety is also associated with negative outcomes for parent and child. The microbiome and host cytokines hold promise to better understand anxiety and its

impacts on child development. Heart Rate Variability (HRV) as measured by respiratory sinus arrhythmia (RSA) is thought to reflect vagal activity; the vagus a key communicator of the

microbiota-gut-brain axis. The objective of this work is better understanding of the interactions of maternal microbes and cytokines in relation to maternal anxiety and infant vagal

activity. METHODS: Self-assessment of anxiety symptoms included the three questions of the Edinburgh Postnatal Depression Scale (EPDS) associated with anxiety (EPDS 3A) and the Generalized

Anxiety Disorder-7 (GAD-7); and were collected alongside maternal fecal samples at two visits in pregnancy and one visit postpartum. Infant RSA assessed by Actiheart was measured over five

minutes at rest (baseline). Maternal microbial composition was characterized by 16S rRNA sequencing and alpha-diversity was calculated by Faith’s PD, observed OTUs, and Shannon. The

microbial compositions and cytokine levels in the third trimester and postpartum of matched perinatal individuals with higher and lower GAD-7 scores were compared in network analyses. Key

microbes from these networks, prior work, and the literature (including commonly used probiotics) were chosen, and their relative abundance averaged across the visits for each parent.

Pearson’s correlations compared infant RSA and maternal alpha diversity. Regression modeling assessed these microbes in relation to maternal anxiety and infant RSA. RESULTS: In a subset of

the larger cohort of 94 perinatal individuals, 40 postpartum parent-infant dyads had infant HRV data. Network analysis in the third trimester found IL-23 as a central hub that associated

with microbes including Akkermansia muciniphila, Bifidobacterium, Faecalibacterium; which were associated with anxiety. Network analysis in the postpartum visit no longer had IL-23 as a

central hub, although IL-23 did associate with Blautia and Bacteroides; microbes including Blautia, Akkermansia, and Bacteroides were important nodes in the network and associated with

maternal anxiety. Bacteroides also associated with IL-10. When Bacteroidies uniformis, Enterococcus, and Blautia, were included in a linear regression model, all three were significantly

associated with the EPDS 3A (p = 0.05, 0.03. 0.03). Lower maternal OTUs associated with lower infant baseline RSA (p = 0.03). Lack of maternal Bifidobacterium adolescentis was significantly

associated with infant baseline RSA (p = 0.006). CONCLUSIONS: Network analysis shows how the immune system and the microbiota differ in the third trimester versus the postpartum period for

individuals with higher and lower anxiety. Utilizing a hypothesis-driven analysis of a common screening tool, three microbes with functions such as tryptophan metabolism and in inducing

inflammation were associated with maternal anxiety. Lower maternal alpha diversity associated with lower infant vagal tone; a microbe related to GABA production associated with infant vagal

tone. DISCLOSURE: Abbvie: Stock / Equity (Spouse) 17.2 INNATE IMMUNITY AND PERINATAL ANXIETY LAUREN OSBORNE WEILL CORNELL MEDICINE, NEW YORK, NEW YORK, UNITED STATES BACKGROUND: Immune

dysregulation has been linked to both psychiatric illness and pregnancy morbidity, including perinatal depression, but little is known about the immune phenotype of perinatal anxiety. Here,

we sought to identify the unique immune profile of antenatal anxiety. METHODS: Pregnant women (n = 107) were followed prospectively at 2nd and 3rd trimesters (T2, T3) and 6 weeks postpartum

(W6). Each visit included a blood draw and psychological evaluation, with clinical anxiety assessed using the Spielberg State-Trait Anxiety Scale. Multiplex assays and flow cytometry

methodology were used to examine the association of anxiety symptoms with secreted immune markers and PBMC-derived immune cells. RESULTS: K cluster means revealed three clusters of anxiety

symptomatology; due to low numbers in the highest severity group, these were collapsed into two groups: Healthy and Anxiety. Women within the Anxiety group demonstrated a lower level of

innate immune cytokines during pregnancy (β = 0.371, SE = 0.173, t = 2.14, p = 0.035), but experienced a rise in levels postpartum, whereas the Healthy group demonstrated a decline; this

difference was at a trend level .(β = 0.381, SE = 0.226, t = 1.69, p = .096) Immune cell populations differed between our two groups, where women within the Anxiety group showed a decrease

in the ratio of B cells to T cells from pregnancy to postpartum while Healthy women increased (β = 4.289, SE = 1.955, t = 2.19, p = 0.031). Women in the Anxiety group also demonstrated an

increased ratio of cytotoxic to helper T cells throughout, a modest increase in the TH1:TH2 ratio across pregnancy, and a lower ratio of TH17:TREG cells in the postpartum when compared to

the Healthy women (β = 0.537, SE = 0.243, t = 2.21, p = 0.029). CONCLUSIONS: These data suggest that the innate immune response throughout the antenatal period differs for women with trait

anxiety symptoms compared to healthy women, suggestive of a unique immune phenotype of perinatal anxiety. DISCLOSURE: Nothing to disclose. 17.3 AUTONOMIC SYSTEM MEASURES OF ANXIETY IN

PREGNANCY JULIA RIDDLE UNIVERSITY OF NORTH CAROLINA, CHAPEL HILL, NORTH CAROLINA, UNITED STATES BACKGROUND: We sought to explore the understand stress responsiveness in pregnancies with and

without anxiety during psychological stressor task using HRV, cortisol, and alpha amylase. METHODS: We followed pregnant women with and without clinical anxiety (n = 95) at four timepoints

in pregnancy and postpartum, with baseline psychiatric diagnosis determined by SCID-5 and symptoms tracked by psychological scales at each visit. At the third trimester visit, a subset of

participants (n = 54) completed a laboratory psychological stressor. We measured baseline, in-stressor, and recovery HRV. Linear mixed effects models were used to analyze HRV measures at

three time points by scales and anxiety diagnoses. RESULTS: The sample was predominantly white and high income. Mean age was 33 and mean pre-pregnancy BMI was 27.2. When controlling for age

and BMI, the two groups (anxious and non-anxious) were significantly different in their HRV RMSSD responsiveness between baseline and recovery (p = 0.0253). In secondary analysis, lower HRV

was associated with worse sleep (Pittsburgh Sleep Quality Index, p = 0.0092) and higher subjective stress (Perceived Stress Scale, p = 0.03896). There was a non-significant trend with high

subjective worry (Penn State Worry Questionnaire, p = 0.0547) and high trait anxiety (Spielberg Trait Anxiety Inventory Scale, p = 0.05367) CONCLUSIONS: Our findings demonstrate that, in

this small sample in late pregnancy, anxiety and subjective perceptions of stress and poor sleep were related to differences in autonomic functioning. Longitudinal studies with a larger

population may help to determine clinical applications. DISCLOSURE: Nothing to disclose. STUDY GROUP 19. TRIPS AND TRYPTAMINES: WHAT MAKES PSYCHEDELICS WORK IN PSYCHIATRIC DISORDERS, IF THEY

DO? BERNARD LERER*, HARRIET DE WIT, BRYAN ROTH, GABRIELLA GOBBI, MICHAEL MITHOEFER, CELIA MORGAN HADASSAH-HEBREW UNIVERSITY MEDICAL CENTER, JERUSALEM, ISRAEL STUDY GROUP SUMMARY: After a

hiatus of three decades, there is growing interest in the potential role of psychedelic drugs in treating psychiatric disorders. Although the current focus is on a small number of compounds,

particularly psilocybin and LSD, there has been a burgeoning effort to identify novel drugs, both natural and chemically synthesized compounds, to optimize therapeutic benefits. Growing

interest in psychedelic research is motivated by the limitations of current psychopharmacological treatments in psychiatry, and by frequent anecdotal reports of efficacy of psychedelic

drugs. On this background, several highly influential controlled studies have yielded promising results in major depression and posttraumatic stress disorder (PTSD), and many other studies

are in progress. These advances have fueled enthusiasm, but at the same time ignited debate including methodological concerns and uncertainty about mechanisms of action. Key issues for

discussion include: * What is the appropriate ‘placebo’ comparison, given the intense psychological effects of psychedelic drugs? * Are the trip-inducing effects essential to their

therapeutic action, and if so, what are the appropriate comparison conditions? * Is it possible to develop new drugs, or administer doses of existing drugs, that induce minimal or no

psychedelic effects but are still effective in treatment? * What are the optimal doses and dosing regimens? * Is psychotherapy crucial to the therapeutic effects of psychedelics These issues

will be the focus of this Study Group. The discussion will be led by participants with expertise in areas critical to developing psychedelics for psychiatry. Bryan Roth (University of North

Carolina) has contributed pivotally to our understanding of the pharmacology of classical and novel psychedelics and will introduce discussion on the feasibility of “non-hallucinogenic”

psychedelics. Gabriella Gobbi (McGill University, Quebec) has conducted exciting new research on the mechanism of action of low doses of LSD in anxiety and social behavior, and the potential

antidepressant effects of psychedelics in animal behavior models. She will introduce important brain-behavior mechanisms to the discussion Michael Mithoefer (MAPS Public Benefit

Corporation) will discuss the challenges in designing clinical trials with MDMA and other psychedelic drugs, and ways to mitigate concern about effective blinding. He will also address the

unknowns about the role of subjective experiences in therapeutic efficacy. Celia Morgan (University Exeter, UK) will consider the practical and ethical challenges of conducting clinical

trials of psychoactive substances and testing the role of therapy in the context of a clinical trial of ketamine assisted psychological therapy in Alcohol Use Disorder. The Study Group is

characterized by diversity in several key areas including clinical, preclinical and translational focus, geographical origin and gender, and includes a mix of ACNP Fellows and non-members.

Drs. Lerer (Israel) and de Wit (USA) will chair the Study Group and moderate the discussion ensuring active audience participation and interaction with the Study Group members. DISCLOSURES:

Back of the Yards Algae Sciences, Parow Entheobiosciences, Contracted Research (Self), Negev Capital, Consultant (Self) PANEL 20. STRIATAL MECHANISMS OF REPETITIVE BEHAVIOR AND HABIT 20.1

THE ROLE OF CORTICO-STRIATAL INTERACTIONS IN REPETITIVE SEQUENCED BEHAVIORS SUSANNE AHMARI UNIVERSITY OF PITTSBURGH, PITTSBURGH, PENNSYLVANIA, UNITED STATES BACKGROUND: Smoothly linking

individual actions into sequences is critical for execution of complex behaviors, but we still have a limited understanding of how behavioral sequences are encoded. Accumulating evidence

suggests striatal activity patterns are linked to performance of sequenced behaviors, but the role of cortical inputs in initiation and control is less clear. Our recent work showed that

SAPAP3-KOs, which display repetitive sequenced grooming behavior, have ~6 fold increase in drive to central striatum (CS) from anterolateral motor area (ALM). This suggested repeated

selection of motor programs could be caused by excessive striatal drive from ALM. METHODS: Male and female C57Bl6 mice were injected with AAV-GCaMP6m, AAV-jRGECO1a, AAV-ChrimsonR, or control

viruses and implanted with optogenetics/photometry fibers or GRIN lenses in CS or ALM to visualize Ca2+ activity during spontaneous behavior. Raw Ca2+ videos were spatially/temporally down

sampled, motion-corrected, and processed (CNMFe). Raw fluorescence was Z-scored to average fluorescence and standard deviation of the trace. Behavior was quantified using automated postural

tracking (SLEAP) and manual behavior annotation, and compared between conditions using support vector machine classifiers (SVMC). RESULTS: Surprisingly, we found a selective increase in CS

activity at initiation of grooming (and not other behaviors) using fiber photometry. Optogenetic stimulation of CS (to mimic this activity increase) evoked immediate-onset, short,

grooming-related fragmented movements distinguishable from normal grooming behavior using a SVMC. Fiber photometry also showed selective activation of ALM during grooming. Unexpectedly, peak

ALM activity correlated with grooming bout length. Stimulation of ALM-CS terminals also caused evoked grooming bouts resembling naturalistic grooming. Finally, dual-color, dual-region

photometry revealed that CS activation precedes ALM activation at initiation of grooming bouts. CONCLUSIONS: These results suggest that CS selects or initiates naturalistic grooming

behavior, while ALM plays a role in sustaining effective bouts. ALM-striatal interactions may thus be a key contributor to regulation of sequenced and repetitive behaviors. Ongoing work is

testing the causal role of ALM in sustaining vs. terminating sequenced behavior. DISCLOSURE: Nothing to disclose. 20.2 OPPOSING AMYGDALA-STRIATAL PATHWAYS ENABLE CHRONIC STRESS TO HASTEN

HABIT FORMATION JACQUELINE GIOVANNIELLO UNIVERSITY OF CALIFORNIA OF LOS ANGELES, SANTA MONICA, CALIFORNIA, UNITED STATES BACKGROUND: To make decisions, we consider our possible actions and

their consequences. This goal-directed strategy allows us to adapt but is cognitively taxing. To counter this, we also use habits. Habits are executed without thought of their consequences

and so, are more resource-efficient but inflexible. Balance between goal-directed actions and habits permits adaptive and efficient behavior. Overreliance on habit is an endophenotype of

many psychiatric conditions such as obsessive-compulsive disorder. Stress, a major contributing factor to these conditions, can prematurely promote habit. Despite this, the brain mechanisms

that allow stress to promote habit formation remain unclear. Amygdala input to the dorsomedial striatum (DMS) may mediate the influence of stress over habits. The DMS is indispensable for

goal-directed learning. Inhibition of this region reduces goal-directed learning and promotes habit. The basolateral (BLA) and central (CeA) regions of the amygdala are known stress hubs in

the brain. Excitatory BLA-DMS neurons are known to exist. Recent evidence revealed direct CeA input to the dorsal striatum and we identified these inhibitory projections target the DMS.

Thus, BLA and CeA are well positioned to differentially regulate the ability of stress to influence behavioral control. METHODS: We developed a task to model chronic stress-induced

instrumental habits in male and female mice and coupled it with in vivo pathway-specific neural activity monitoring and manipulation techniques. RESULTS: We found that stress promotes habits

in males and females. DMS projecting-BLA and -CeA neurons show opposing activity during instrumental learning. CeA-DMS activity opposes goal-directed learning and is necessary for stress to

promote habits. Conversely, BLA-DMS activity opposes this and is sufficient to rescue goal-directed control in stressed mice. CONCLUSIONS: These data reveal a function for the BLA-DMS and

newly identified CeA-DMS pathways. While BLA-DMS activity promotes the learning that supports flexible, goal-directed actions and prevents stress from promoting habits, CeA-DMS activity

enables stress to promote habits. These findings have important implications for psychiatric conditions influenced by stress and characterized by maladaptive habits, such as

obsessive-compulsive disorder and substance use disorder. DISCLOSURE: Nothing to disclose. 20.3 CELL TYPE SPECIFIC CONTRIBUTIONS OF STRIATAL PROJECTION NEURONS TO HABIT LEARNING MELISSA

MALVAEZ UNIVERSITY OF CALIFORNIA, LOS ANGELES, LOS ANGELES, CALIFORNIA, UNITED STATES BACKGROUND: Optimal behavior relies on a balance between two distinct strategies; one goal-directed in

which the consequences of actions are considered, and one habitual, where tasks are conducted without forethought of their consequences. The balance between these systems allows adaptive and

efficient behavior, however, dysfunction in this balance can lead to symptoms characteristic of several psychiatric diseases. Goal-directed and habit learning are known to rely on the

anatomically distinct dorsomedial (DMS) and dorsolateral (DLS) striatum, respectively. However, the subregion-specific contribution of the two major subtypes of striatal output neurons, the

direct (dSPNs) and indirect (iSPNs) striatal projections neurons, is unknown. METHODS: We used subregion-specific chemogenetic approaches to modulate dSPNs and iSPNs in mice, as well as

cellular resolution calcium imaging to monitor cell-specific activity, during instrumental habit learning. RESULTS: In the pDMS, both dSPNs and iSPNs mediate goal-directed learning.

Chemogenetic inactivation of dSPNs (WT, n = 7; D1-cre, n = 9; Two-way RM ANOVA with Bonferroni MCT: p = 0.009) or iSPNs (WT, n = 9; A2Acre, n = 7; Two-way RM ANOVA with Bonferroni MCT: p = 

0.031) during instrumental training prevents goal-directed learning, while activation of dSPNs (WT, n = 7; D1-cre, n = 7; Two-way RM ANOVA with Bonferroni MCT: p = 0.030) or ISPNs (WT, n = 

11; A2A-cre, n = 9; Two-way RM ANOVA with Bonferroni MCT: p = 0.050) promotes goal-directed behavior even after overtraining. Conversely, in the DLS, we found that dSPNs (WT, n = 10; D1-cre,

n = 11; Two-way RM ANOVA with Bonferroni MCT: p = 0.046), but not iSPNs (WT, n = 12; A2A-cre, n = 11; Two-way RM ANOVA with Bonferroni MCT: p > 0.999), mediate the acquisition of habits.

In the pDMS, both dSPNs (n = 466-538) and iSPNs (n = 411-499) are active during instrumental performance. dSPNs maintain representation of action initiation throughout training and the

transition to habit, while iSPNs become recruited to the initiation of instrumental performance as habits form. CONCLUSIONS: dSPNs and iSPNs differentially contribute to behavioral strategy

in a subregion-specific manner. In the pDMS the two major striatal outputs coordinate the transition to habits. These findings challenge the canonical views of striatal function and identify

a neural substrate of adaptive strategy learning. DISCLOSURE: Nothing to disclose. 20.4 PLASTICITY OF STRIATAL FAST-SPIKING INTERNEURON NETWORKS IN HABIT FORMATION NICOLE CALAKOS DUKE

UNIVERSITY MEDICAL CENTER, DURHAM, NORTH CAROLINA, UNITED STATES BACKGROUND: Striatal fast-spiking interneurons (FSI) occupy a powerful position within the striatal microcircuitry to

influence striatal output. The behavioral transition from goal-directed to habitual requires dorsomedial (DMS) and dorsolateral (DLS) striatal regions. DLS FSI activity is required for

habitual performance of a learned task and FSI excitability positively correlates with habitual, as opposed to goal-directed behavior. Here we examine the cellular and circuit mechanisms

contributing to FSI excitability in habitual mice. METHODS: Adult parvalbumin (Pv) Cre mice were trained in a lever press task. Habitual behavior was quantified by sensitivity of the

behavior to outcome devaluation. For electrophysiology, acute brain slices were prepared after devaluation test. For behavioral tests of FSI activity, adult Pv Cre mice received intracranial

AAV for Cre-dependent expression of Halo-Tag. Drugs Acutely Restricted by Tethering (DART) methodology was used to concentrate a modified GABAR antagonist, gabazine at FSIs. The sufficiency

of GABAR inhibition on FSIs to modify habitual behavior was tested by infusing drug on the day of devaluation testing. RESULTS: Intrinsic membrane excitability differed between FSIs in

goal-directed and habitual mice, with habitual mice FSIs being more excitable (p < 0.05). FSI excitability differences were similar between dorsomedial and dorsolateral striatal regions.

Notably, chamber control mice showed FSI excitability similar to habitual mice. Preliminary findings suggest that selectively reducing GABAR transmission in FSIs may be a sufficient

mechanism to correspondingly influence habitual behavior (p = 0.0009). CONCLUSIONS: Our results provocatively update working models for habit learning and expression. First, rather than a

model of acquiring plasticity with continued training from the goal-directed to habitual periods, data from naïve control mice support a model in which plasticity that dampens FSI

excitability occurs in the early stages of goal-directed learning, followed by a return to “baseline” levels of FSI excitability in habitual mice. Second, even though DLS and DMS circuits

are generally associated with habitual and goal-directed behavior, respectively, our results indicate that at least at the level of FSI plasticity, it occurs similar across these two

regions. DISCLOSURE: Nothing to disclose. PANEL 21. BAD WRAP: OLIGODENDROCYTES AS A COMMON CELLULAR SUBSTRATE FOR GENETICALLY-IDENTIFIABLE NEURODEVELOPMENTAL DISORDERS IN HUMANS AND MOUSE

MODELS 21.1 DEVELOPMENTAL DISRUPTION OF WHITE MATTER MICROSTRUCTURE IN 22Q11.2 COPY NUMBER VARIANTS: A CROSS-SPECIES INVESTIGATION CARRIE BEARDEN UNIVERSITY OF CALIFORNIA, LOS ANGELES,

CALIFORNIA, UNITED STATES BACKGROUND: Chromosome 22q11.2 deletions(22qDel) are among the greatest known risk factors for schizophrenia (SCZ). In the largest multisite diffusion-weighted

magnetic resonance imaging (dMRI) study of 22qDel to date, we recently found widespread white matter (WM) microstructure alterations, characterized by higher fractional anisotropy (FA)

relative to healthy controls (HC) in callosal tracts, but lower FA in long association tracts, suggesting altered frequency of layer 2/3 projection neurons. We also found intriguing gene

dosage effects, involving opposing patterns of WM disruption in the reciprocal 22q11.2 duplication, which is putatively protective against SCZ. Here, we investigated the neurodevelopmental

trajectory of these disruptions in a cross-species investigation. METHODS: Our human study included 594 participants: 334 with 22qDel (mean age 16.88 ± 6.43, 153 females) and 260 healthy

controls (HC, mean age 16.55 ± 8.01, 123 females). Across species we used tract-based spatial statistics to interrogate WM. Given findings of nonlinear dMRI trajectories in healthy

individuals, we fit Poisson nonlinear models for each group separately. In parallel, we conducted dMRI studies in the LgDel mouse model (Juvenile WT = 22, LgDel=13; Adult WT = 21; LgDel=18;

both sexes), using ex-vivo overnight imaging at 7 Tesla in PFA perfused samples. Animals were injected with Gsk3β in early postnatal days, shown previously to rescue cognition in a 22q11.2

mouse model. RESULTS: Human 22qDel carriers showed robustly increased FA in callosal regions of interest (Cohen’s d = 0.37-0.58; all p < 2.69e-5). Further, the mean age of peak FA was

significantly older in 22qDel vs HC (t = -3.03; p < .007). Consistent with human findings, adult LgDel mice had significantly higher FA in callosal fibers relative to WT(t > 2.1,

cluster-corrected at p = 0.01). This effect was also present in juvenile mice, although its magnitude was slightly smaller (2- way ANOVA, genotype, p < 0.02; F(1,70) = 20.79). Lastly, FA

alterations were not rescued by Gsk3β antagonism (t(36) = 0.88, p = 0.39). CONCLUSIONS: Findings of developmental WM disruptions, particularly impacting callosal fiber tracts, are consistent

across species. Delayed WM maturation in 22qDel may be secondary to altered axonal diameters, a precursor of a delayed myelination process in 22qDel. DISCLOSURE: Nothing to disclose. 21.2

ENHANCING MYELINATION AS A THERAPEUTIC INTERVENTION FOR AUTISM SPECTRUM DISORDER BRADY MAHER LIEBER INSTITUTE FOR BRAIN DEVELOPMENT, JHMI, BALTIMORE, MARYLAND, UNITED STATES BACKGROUND:

Autism spectrum disorder (ASD) is a genetically heterogeneous disorder with convergent symptomatology that suggests the potential for identifying common dysregulated pathways that could be

amenable to therapeutic intervention. Mutations in Transcription Factor 4 (TCF4) cause a syndromic ASD known as Pitt-Hopkins Syndrome (PTHS). We have previously shown that transcriptional

and biological profiling of PTHS mouse models indicated a disruption in oligodendrocytes (OLs). Moreover, we demonstrated a convergence of differential expression related to OL biology was

also observed in two additional ASD mouse models (Pten and Mecp2). Remarkably, the eigengene of these convergent genes was effective at separating postmortem ASD brain samples from controls.

Taken together, these results indicate that defects in myelination are a common pathophysiology in ASD and opens the door to developing ASD therapies that targets myelination. METHODS:

Clemastine (CF) is an antihistamine that is effective at promoting myelination through its block of muscarinic receptors. We tested the effects of CF (10uM) on primary OL cultures derived

from Tcf4+/tr and WT littermates (male and female), as well as in vivo using daily i.p. injections (10mg/kg) between postnatal day 28 and 42. RESULTS: In vitro CF treatment of Tcf4+/tr OL

cultures resulted in a significant increase in the OL density compared to vehicle treated cultures (F = 5.34, p = 0.03, n = 6-10 samples/genotype/treatment). In vivo administration of CF

also significantly increased the OL population (F = 6.78, p = 0.014, n = 8-13 animals/genotype/treatment) compared to vehicle treated mice. Moreover, in Tcf4+/tr CF treatment increased the

proportion of newly formed uncompacted myelin as measured by electron microscopy (t = 4.19, p < 0.0005, vehicle n = 79, CF n = 89) and rescued electrophysiological deficits in compound

action potentials (t = -2.49, p = 0.03, vehicle n = 11, CF n = 14). Remarkably, CF treatment also rescued behavior in Tcf4+/tr mice (F = 25.03, p < 0.0001, n = 8

animals/genotype/treatment). CONCLUSIONS: Treatment with CF, an FDA approved compound, is effective at rescuing the OL population, electrophysiology and behavior in a PTHS mouse model. These

findings provide novel preclinical data that suggests CF or other remyelinating agents may be beneficial to PTHS patients and potentially other related ASDs. DISCLOSURE: Nothing to

disclose. 21.3 ROLES OF OLIGODENDROGLIAL SODIUM CHANNELS IN MYELINATION AND AUDITORY PROCESSING DEFICITS RELATED TO AUTISM JUN HEE KIM THE UNIVERSITY OF TEXAS HEALTH SCIENCE CENTER, SAN

ANTONIO, TEXAS, UNITED STATES BACKGROUND: Auditory processing abnormalities are common features of neurodevelopmental disorders such as autism spectrum disorder (ASD). However, the causes

and mechanisms have not been sufficiently explored. The gene, SCN2A, encoding the alpha subunit of the voltage-gated Na+ channel 1.2, is highly linked to ASD. This presentation will

highlight novel data on how loss of oligodendroglial SCN2A impacts myelination and neural connectivity in the auditory system, leading to auditory processing disorders. METHODS: We used a

novel Scn2a conditional knockout mouse (Scn2a cKO) to specifically delete SCN2A in oligodendrocytes (OLs), by crossing tamoxifen -inducible Pdgfra-CreER with Scn2a floxed mice. All mice were

in a C57BL/6 genetic background. For auditory brainstem responses (ABRs) test, control and Scn2a cKO mice (20 mice per group, both sexes) were anesthetized with 2% isoflurane during

recording (1 l/min O2 flow rate). ABR recordings was done in a sound attenuation chamber. RESULTS: To determine the role of SCN2A in OL development and axonal myelination, we evaluated the

population of OLs and myelin properties in the auditory brainstem using transmission electron microscopy (TEM). We found that SCN2A deletion impairs OL differentiation. In TEM analysis,

Scn2a cKO mice showed a higher g-ratio in the brainstem, compared with control (control:0.81 ± 0.065, n = 1154 axons vs cKO: 0.85 ± 0.002, n = 432 axons; 5 mice/genotype, t = 7.233, p < 

0.0001), indicating that axons have a thinner myelin in cKO mice. Alterations in myelination impact neuronal activity and axonal conduction. The nerve terminals in Scn2a cKO mice displayed a

number of action potential failures in response to axon fiber stimulation (at 200 Hz, control:25 ± 6.1%, n = 11 cells vs Scn2a cKO: 40 ± 6.8%, n = 16 cells in cKO). Finally, in vivo ABRs

showed Scn2a cKO mice had a reduced threshold (control:35 ± 3.7 dB, n = 23 mice vs cKO: 23 ± 4.1 dB, n = 13 mice, t = , 1.943, p = 0.044) and an elevated amplitude of ABRs (control: 1.9 ± 

0.22uV, n = 23 mice vs cKO: 3.0 ± 0.26 uV, n = 13 mice, t = 3.483, p < 0.01), suggesting that Scn2a cKO mice have auditory hypersensitivity. CONCLUSIONS: SCN2A is important for

oligodendrocyte differentiation, myelination, and neuronal activities in the auditory nervous system. Loss of SCN2A in OLs during early development can cause auditory processing disorders

related to ASD. DISCLOSURE: Nothing to disclose. PANEL 22. SEXX, BRAIN, AND CARDIOMETABOLIC HEALTH: WHAT DO THEY HAVE IN COMMON AND WHY SHOULD PSYCHIATRY CARE? 22.1 SEX-SPECIFIC REGULATION

OF AUTONOMIC FUNCTION – IMPACT OF GONADAL HORMONES AND IN UTERO GLUCOCORTICOID EXPOSURE TABEN HALE UNIVERSITY OF ARIZONA, PHOENIX, ARIZONA, UNITED STATES BACKGROUND: Prenatal insults leading

to increased fetal glucocorticoid exposure can sex-selectively impact future disease risk. We have shown that when pregnant rat dams are treated with the glucocorticoid, dexamethasone

(DEX), for the last 4 days of gestation, female-specific changes in stress-responsive cardiovascular function, autonomic function, depression-, and anxiety-like behaviors are detected in

their adult offspring. The present study evaluated the degree to which the sex-specific changes in autonomic function are due to the activational effects of gonadal steroid hormones.

METHODS: Pregnant rat dams were administered DEX (0.4mg/kg per day, s.c.) or vehicle (VEH) on gestation days 18-21. Rats underwent a gonadectomy (GDX) or sham surgery post-puberty; 1-week

later radiotelemetric transmitters were implanted for direct recording of heart rate variability (HRV) from arterial pressure waveforms to assess autonomic function. HRV was measured 2-3

weeks post GDX, and response to acute stress was assessed using a restraint tube for 20 minutes. Testing was performed on diestrus in sham females. Frequency domain analysis of HRV was

performed during baseline and stress periods, with high frequency (HF) power reflecting parasympathetic and low frequency (LF) power reflecting both sympathetic and parasympathetic outflow.

RESULTS: There was a main effect of GDX resulting in reduced HF power (F (1, 21) = 4.946, p = 0.037) and LF power (F (1, 21) = 4.581, p = 0.044) in females. In females there was a

significant interaction for LF/HF ratio (F (1, 21) = 5.314, p = 0.031), whereby LF/HF was increased during stress period in VEH, but not DEX-exposed rats. LF/HF was increased following GDX

in DEX, but not VEH exposed females (F (1, 21) = 4.130, p = 0.055). In males, there was no significant impact of GDX, prenatal exposure, or stress on HF and LF HRV. LF/HF was significantly

increased during stress in males (F (1, 23) = 4.784, p = 0.039). CONCLUSIONS: Findings revealed sex-specific regulation of autonomic function, whereby activational effects of gonadal

steroids impact autonomic control of cardiac function in females, but not males. In utero DEX exposure induced dysregulation of sympathovagal balance in female rats that was exacerbated by

gonadal hormone loss. Findings may lead to future sex-selective therapeutic considerations particularly in post-menopausal women. DISCLOSURE: Nothing to disclose. 22.2 TRAJECTORIES OF RISK

FOR AND RESILIENCE AGAINST NEURODEGENERATIVE DISEASES OF AGING: LESSONS LEARNED FROM MIDLIFE NEURO-TRANSITIONS ROBERTA BRINTON UNIVERSITY OF ARIZONA, COLLEGE OF MEDICINE, CENTER FOR

INNOVATION IN BRAIN SCIENCE, TUCSON, ARIZONA, UNITED STATES BACKGROUND: Midlife aging transitions can be critical periods for neural circuit reorganization, metabolic reprogramming and

function which can provide insights into windows of opportunities to promote resilient vs vulnerable brain aging. The menopause in women and the andropause in men can be a critical period

for determining risk for age-associated neurodegenerative disease and initiating the prodromal phase of disease. METHODS: RNA-Seq was conducted on rat hippocampal RNA at 100 bp paired-end on

NovaSeq. Metabolomics from rat brain and plasma was performed utilizing Global Metabolomics and Complex Lipids mass spectrometry. Mmedical claims study included women with or without claim

records of hormone therapy medications. Relative risk ratios and 95% confidence intervals for combined Alzheimer’s, Parkinson’s, Multiple Sclerosis and ALS were determined. RESULTS:

Transcriptomic analysis revealed chronological and endocrinological aging specific shifts in bioenergetic systems of biology that were paralleled by bioenergetic dysregulation in midlife

aging female brain. Metabolomic and lipidomic analyses revealed dynamic adaptation of the aging female brain from glucose centric to utilization of auxiliary fuel sources that included amino

acids, fatty acids, lipids, and ketone bodies. At the population level of analysis, in 379,352 women with or without claim records of hormone therapy, use of HT was associated with

significantly reduced risk for combined neurodegenerative diseases of aging (RR 0.42, 95% CI0.40–0.43,P < 0.001).Greatest reduction in risk of AD, and dementia emerged in patients aged 65

years or older. CONCLUSIONS: Midlife aging transitions can be critical periods of neural circuit reorganization, metabolic reprogramming, brain function and therapeutic interventions to

prevent neurological and psychiatric disorders of aging. DISCLOSURE: Nothing to disclose. 22.3 EVALUATION OF SEX DIFFERENCES IN THE GENETIC RELATIONSHIP BETWEEN CARDIOVASCULAR DISEASE,

DEPRESSION, AND CHRONIC INFLAMMATION LEA DAVIS VANDERBILT UNIVERSITY MEDICAL CENTER, NASHVILLE, TENNESSEE, UNITED STATES BACKGROUND: Cardiovascular disease (CVD) and depression are highly

comorbid. Prior research finds that polygenic scores (PGS) which increase the odds of depression, also increases the odds of cardiovascular disease, but do so to a greater extent in females

than in males, after accounting for canonical CVD risk factors. In contrast, coronary artery disease (CAD) PGS does not associate with clinical depression (OR, 0.93 [95% CI, 0.87-1.00]; P = 

0.07). These findings raise the hypothesis that canonical CVD risk factors (e.g., lipids, hypertension, etc.) may have differential importance in males and females and that

sex-differentiated biology may illuminate new relationships between depression and CVD. METHODS: Motivated by prior results, we employed a data-driven approach (laboratory measurement-wide

association study, i.e., LabWAS) in a retrospective study design. We first tested the relationship between depression PGS and 315 routinely collected laboratory measurements available in the

VUMC electronic health records of 70,704 individuals of primarily European ancestry and 12,384 individuals of primarily African ancestry. Using linear regression models, we corrected for

the cubic splines of age at lab measurement, top 10 principal components from genetic data, and sex (in the sex-combined models). We then further adjusted for the presence of clinical

depression or anxiety disorders. Finally, we stratified the data by sex and tested the models separately in males and females. RESULTS: After multiple testing correction, the LabWAS of

depression PGS implicated significant associations with multiple elevated immune markers. Upon stratification by sex, all markers except for WBC count showed sex differential association

with depression PGS (see Statistics). Lastly, mediation analyses (N = 20,749) demonstrated that markers of inflammation mediate ~30% (p = 0.017) of the risk for CAD that is conferred by

genetic risk for depression. CONCLUSIONS: We show that multiple markers of inflammation are associated with the genetic risk for depression, even in the absence of clinical depression.

Furthermore, increased inflammation accounts for a substantial portion of the risk for CVD that is conferred by depression genetics. Finally, sex differences point towards inflammatory

processes as important mediators of the comorbidity between depression and CVD. DISCLOSURE: Nothing to disclose. 22.4 OPTIMIZATION OF NON-INVASIVE VAGAL NERVE STIMULATION FOR THE MODULATION

OF MOOD AND CARDIOVASCULAR FUNCTION IN MAJOR DEPRESSION AND HYPERTENSION RONALD GARCIA MASSACHUSETTS GENERAL HOSPITAL, HARVARD MEDICAL SCHOOL, CHARLESTOWN, MASSACHUSETTS, UNITED STATES

BACKGROUND: Our recent studies have suggested that a novel, non-invasive neuromodulation technique called Respiratory-gated Auricular Vagal Afferent Nerve Stimulation (RAVANS) may

effectively modulate the stress response circuitry and have beneficial effects on cardiovagal activity and mood regulation. However, optimal stimulation parameters for this technique have

not been established. Here we present the results of studies aimed to identify frequency-dependent effects of RAVANS on the regulation of depressed mood and anxiety symptoms in female

patients with major depression (MDD) and sex- and frequency-dependent effects on the regulation of blood pressure values in patients with hypertension. METHODS: Study 1 included women with

recurrent MDD in an active episode (n = 13, 30.5 ± 6.0 years). Subjects underwent five stimulation sessions, during which they received exhalatory-gated stimulation at frequencies of 2, 8,

30, and 100 Hz or sham stimulation, in a randomized order. Electrodes were placed over vagal-innervated auricular regions (cymba concha) in the left ear. Subjects completed a Beck’s

Depression Inventory (BDI) and State-Trait Anxiety Inventory (STAI) at the beginning and end of each stimulation session. Study 2 included subjects with hypertension (n = 20, 54.5 ± 6.2

years) in which the effects of five different RAVANS stimulation frequencies (2, 10, 25, 100 Hz or sham) on blood pressure levels were evaluated. RESULTS: In MDD subjects, RAVANS

administration at a 100 Hz frequency resulted in a significant reduction in depressive (BDI score) (β = -3.45, p = 0.026) and anxiety symptoms (STAI score) (β = -6.13, p = 0.034). In

hypertensive subjects, RAVANS at 2 and 100 Hz frequencies resulted in a significant reduction of diastolic blood pressure (β = -5.10, p = 0.004; β = -8.28, p = 0.003) and mean arterial

pressure values (β = -5.89, p = 0.003; β = -7.45, p = 0.02), but only in black female participants. CONCLUSIONS: Our results demonstrate the optimized, frequency-dependent effects of RAVANS

administration on the modulation of mood and anxiety symptoms in MDD, and blood pressure regulation in hypertensive subjects. Furthermore, these findings indicate a potential differential

response to this stimulation based on sex and race with important implications for the use of this technique in the treatment of comorbid MDD and cardiovascular disease. DISCLOSURE: Nothing

to disclose. STUDY GROUP 23. QUO VADIS: THE IMPLICATIONS OF SMALL EFFECT SIZES IN POPULATION NEUROIMAGING STUDIES MARTIN PAULUS, WESLEY THOMPSON, CHUN-CHIEH FAN, BRENDEN TERVO-CLEMMENS, HAE

KYUNG IM, MONICA ROSENBERG, CYNTHIA ROGERS, ROMAN KOTOV, ALEXANDER SHACKMAN, DAMIEN FAIR LAUREATE INSTITUTE FOR BRAIN RESEARCH, TULSA, OKLAHOMA, UNITED STATES STUDY GROUP SUMMARY: Large

scale, population-representative imaging studies are beginning to examine the utility of structural and functional neuroimaging to better understand the neural mechanisms underlying

psychiatric disorders and to provide predictive biomarkers that could be pragmatically useful. These studies have yielded the basis for a much-needed correction to the problems of small

samples, capable of providing more accurate estimates of brain-behavior association effect sizes. However, the tools and techniques that are applied to the large studies to evaluate the

utility of these brain-behavior associations are just beginning to be developed. One crucial question is what measures should be used to assess the “practical” utility of imaging data when

effects are too small to be of individual patient level clinical importance (e.g., for neuropsychiatric outcomes). Moreover, these tools and techniques need to provide answers to the

questions whether brain-behavior effects are truly small in a mechanistic (causal) sense, or whether they are they only apparently small, e.g., due to random or systematic measurement error,

or whether they are due to group heterogeneities that leads to small average effect sizes? In this Study Group, we will focus on four main challenges of how to formulate a taxonomy of

analytic designs for neuroimaging-behavioral studies going forward: 1) how to think about power and sample size (e.g., when do you need very large datasets, and when can you get by with

smaller studies?); 2) generalizability to populations of interest (external validity); 3) estimation of effect sizes potentially biased by confounding, measurement error, and selection; 4)

application of results of neuroimaging studies to diverse/understudied groups. Our panel of experts will focus on these issues, with the aim of moving beyond the current realization that

effect sizes are often (though not always) quite small in brain-behavior relationships. Emphasis will be placed on a path to move forward and for providing guidance to stakeholders (e.g.,

NIH, neuroscience researchers) on how to fund and conduct studies going forward that maximize efficiencies and opportunities for scientific and clinical utility. DISCLOSURES: UpToDate:

Royalties (Self), Spring Health: Board Member (Self), Engrail Therapeutics: Employee (Spouse) PANEL 24. EXTENDING PSYCHIATRIC GENETICS TO MULTIPLE ANCESTRIES 24.1 THE LATIN AMERICAN GENOMICS

CONSORTIUM: INCREASING THE REPRESENTATION AND ENGAGEMENT OF LATINX/HISPANIC INDIVIDUALS IN PSYCHIATRIC GENOMICS STUDIES PAOLA GIUSTI-RODRIGUEZ UNIVERSITY OF FLORIDA, COLLEGE OF MEDICINE,

GAINESVILLE, FLORIDA, UNITED STATES BACKGROUND: Psychiatric genome-wide association studies (GWAS) are making clear progress in identifying variants linked to psychiatric disorders and

related traits. However, nearly 80% of all published GWAS still contain only European ancestry individuals. Admixed individuals in particular are routinely excluded from genetic studies due

to concerns over population structure, contributing to the concerning health disparities documented across ancestries. METHODS: Incorporating non-European and mixed-ancestry populations in

genetic analyses can help to identify genetic effects that generalize across populations, pinpoint causal variants, and ascertain population-specific genetic variants. RESULTS: The Latin

American Genomics Consortium (LAGC) was founded in 2019 to accelerate psychiatric genetics research in Latinx/Hispanic populations. The LAGC includes over 100 active members, representing 8

Latin American countries/territories, as well as the U.S. This initiative aims to not only address the enormous underrepresentation of Latinx/Hispanic individuals in psychiatric genomics

studies, but to facilitate access to training and resources, and engage in collaborations globally. CONCLUSIONS: To address the complexity of GWAS analysis in Latinx populations and conduct

well-powered genetic studies, we have assembled an international team of experts equipped with sophisticated tools that will allow us to adequately account for admixture. We have secured

access to multiple relevant datasets from the US and Latin America and are conducting meta-analyses focused on psychiatric traits captured across datasets. We continue to work on procuring

access to additional cohorts and bringing in members from other Caribbean and Latin American countries. DISCLOSURE: Nothing to disclose. 24.2 THE POWER OF INCLUDING ALL ANCESTRIES: NEW

METHODS AND RESULTS KAI YUAN MASSACHUSETTS GENERAL HOSPITAL, CAMBRIDGE, MASSACHUSETTS, UNITED STATES BACKGROUND: The scale of non-European genomic data has started to grow in recent years.

Methods to leverage genomic data across multiple ancestries, however, have been lagging. Most studies used naïve methods for multi-ancestry studies, failing to realize the full potential of

the data. Here we present two new methods, with applications to psychiatric genetics, that leverage data across ancestries for variant discovery (fine-mapping) and polygenic risk prediction

(PRS). METHODS: Statistical fine-mapping refines a GWAS locus to a set of likely causal variants (credible set). We developed a novel method for cross-ancestry fine-mapping integrating data

from multiple ancestries through explicitly modeling population-specific linkage disequilibrium (LD) and accounting for multiple causal variants in a genomic region. PRS are less effective

when ported across populations. We developed the first principled Bayesian PRS method, PRS-CSx, that jointly models GWAS summary statistics from multiple populations to improve

cross-ancestry PRS. PRS-CSx couples genetic effects across populations via a shared continuous shrinkage prior, enabling more accurate effect size estimation. RESULTS: We applied our new

fine-mapping method to traits from the Taiwan Biobank and UK Biobank (UKBB). Compared with fine-mapping in UKBB alone, we significantly reduced the size of credible sets and increased the

PIP of the most probable variant. We then applied our method to schizophrenia genetics data of East Asian (EAS) and European (EUR) ancestries. Compared with the published fine-mapping

results we reduced the size of credible sets in 70% of the loci. Similarly, we showed PRS-CSx substantially improved the prediction accuracy even if only a small non-European GWAS was

included. The median R2 increased by 76% for EAS individuals when the Biobank Japan samples were added to the UKBB EUR samples to train the PRS. By integrating schizophrenia genetics data

from EAS and EUR, we more accurately predicted schizophrenia risk in EAS individuals, showing 52% and 97% improvement in the liability R2 relative to PRS constructed using EAS or EUR summary

statistics only. CONCLUSIONS: We developed two new methods that model genomics data across multiple ancestries to realize the full potential of the emerging non-European resources.

DISCLOSURE: Ono Pharma: Consultant (Self), Xian Janssen Pharmaceutical Ltd.: Honoraria (Self), Biogen: Contracted Research (Self) 24.3 ENHANCED INTERPRETATION OF SCHIZOPHRENIA GWAS OF

DIVERSE POPULATION WITH BRAIN REGULATORY ARCHITECTURE WITH AFRICAN AMERICAN AND EAST ASIAN DESCENT YU CHEN THE BROAD INSTITUTE, CAMBRIDGE, MASSACHUSETTS, UNITED STATES BACKGROUND: Previous

genetic studies of schizophrenia (SCZ) focused on disproportionate majority of European Population. GWAS of non-European populations highlighted the importance of generalizability of genetic

studies. Yet, the scarcity of expression quantitative trait loci (eQTL) data of non-European population brains restricted our understanding of how population genetic diversity is involved

in the risk of schizophrenia. METHODS: We analyzed genotype and RNA-seq of African Americans (AA, n = 158) and Europeans (EUR, n = 408) from the PsychENCODE consortium and East Asians (EAS,

n = 217) from the Chinese Human Brain Bank. We compared population differences of eQTLs from three analyses: differences in allele frequency, population-specific variants, and heterogeneity

in eQTL effects. We next used the brain eQTLs to explain SCZ GWAS from diverse ancestries and observed that eQTLs better explained SCZ GWAS in the matched population. We also prioritized

risk genes via transcriptome-wide association analysis and colocalization in the matched populations. RESULTS: In total, 1,996,544 significant independent eQTL signals involving 11,622 genes

were identified. 173,211 eQTLs (8% of all eQTLs) involving 1,386 genes are novel captured in the non-EUR populations. The trans-ancestry differences represented by the 110,855 (64% of all)

novel eQTLs can be explained by allele frequency differences. Although the effect size of eQTLs showed a strong correlation across populations, we identified 913 (2% of all) eQTLs involving

ten genes that have opposite eQTL effects across populations. Fourteen novel SCZ candidate genes were identified based on EAS population eQTL with EAS SCZ GWAS. CONCLUSIONS: This

characterization of brain regulatory architecture across diverse populations highlighted the importance of studying brain eQTL from non-EUR populations and provides a comprehensive resource

for novel insights into schizophrenia. DISCLOSURE: Nothing to disclose. 24.4 LEVERAGING LOCAL ANCESTRY TO ENHANCE TRANSCRIPTOME-WIDE ASSOCIATION STUDIES IN ADMIXED AND MULTIETHNIC

POPULATIONS ERIC GAMAZON VANDERBILT UNIVERSITY MEDICAL CENTER, NASHVILLE, TENNESSEE, UNITED STATES BACKGROUND: We have previously shown that use of local ancestry in genome-wide association

studies (GWAS) in multiethnic populations may improve characterization of the heritability of complex traits and lead to more accurate mapping of genetic associations. METHODS: We develop a

framework for transcriptome-wide association studies (TWAS) in admixed and multiethnic populations. We investigate the methodological implications of a local ancestry aware TWAS (LA-TWAS).

TWAS is a two-stage procedure, which exploits information on the genetic regulation of gene expression to conduct gene-level association testing. In the first stage, TWAS trains predictive

models of gene expression using local genetic variation; in the second stage, the association between the genetically determined expression generated from each gene model and the trait under

study is evaluated. RESULTS: We show that LA-TWAS leads to significantly higher transcriptome prediction performance than conventional TWAS approaches. LA-TWAS jointly estimates the

heritability of gene expression and the ancestry specificity attributable to local genetic variation. In addition, in systematic application to BioVU, Vanderbilt’s DNA biobank linked to

electronic health records (N > 100k), LA-TWAS leads to significantly increased power for discovery of gene-level associations across a range of genetic architectures and degree of

admixture. We present an application to neuropsychiatric traits in African Americans. We present both cross-population and population-specific gene-level findings. CONCLUSIONS: Thus,

integration of local ancestry enhances modeling of gene expression and identification of expression-mediated complex traits. Our study contributes to ongoing efforts to enhance portability

and generalizability of genetic methodologies to traditionally underrepresented populations. DISCLOSURE: Nothing to disclose. PANEL 25. BEYOND RESTING STATE: CONNECTIVITY-BASED ANALYSES OF

EVOKED BRAIN RESPONSES AS AN OPTIMAL APPROACH TO UNDERSTANDING BRAIN-BEHAVIOR CLINICAL ASSOCIATIONS 25.1 CHARACTERIZING BRAIN-PHENOTYPE RELATIONSHIPS ACROSS BRAIN STATES IN HEALTH AND

DISEASE ABIGAIL GREENE YALE SCHOOL OF MEDICINE, NEW HAVEN, CONNECTICUT, UNITED STATES BACKGROUND: Predictive modeling may reveal the macroscale neural bases of complex phenotypes and

identify individualized targets for clinical intervention. This talk will demonstrate the impact of brain state and measure bias on the performance, interpretation, and utility of

brain-phenotype models. METHODS: First, we used task- and rest-based functional connectivity (FC) to predict intelligence measures from 2 datasets (sample 1): the Human Connectome Project

(HCP; n = 515, 241 male) and the Philadelphia Neurodevelopmental Cohort (n = 571, 251 male). Next, we used psychophysiological interaction (PPI), predictive modeling, and novel inter-subject

PPI analyses in the HCP (n = 703; sample 2) to explore how tasks increase predictive model accuracy. Finally, we used task- and rest-based FC from a new, demographically and clinically

heterogeneous dataset (n = 129; sample 3) to predict performance on 16 neuropsychological measures and explored in whom models failed. RESULTS: In sample 1, task-based models consistently

outperformed rest-based models (P = 0.018 via Mann-Whitney U test), with the best task explaining 12.8% of the variance in fluid intelligence (P < 0.001) and the best rest explaining 3.9%

(P = 0.05). In sample 2, FC significantly predicted intelligence (all P < 0.01) independent of task-evoked activation. Activation is useful for prediction only if the task is related to

the predicted phenotype (r = 0.96, P < 0.05). Further, some tasks are more useful for prediction than others (e.g., 3.4-12.8% in HCP sample 1), with the best task varying by sex: emotion

models outperformed working memory models in females, while the opposite was true in males (all P < 0.02). In sample 3, we explored such group-specific models, finding that model failure

is reliable in a subset of participants, phenotype specific (r = 0.49, P < 0.0001), generalizable across datasets (all P < 0.0001), and related to sociodemographic variables such as

education and race. CONCLUSIONS: Tasks amplify phenotype-relevant patterns of brain activity via distributed, task-general denoising. Care must be taken, however, in interpretation, as

models often reflect not unitary cognitive constructs, but rather stereotypical profiles. Together, this work furthers optimization of brain-based predictive modeling to reveal the neural

bases of cognitive processes relevant to health and disease. DISCLOSURE: Nothing to disclose. 25.3 TRIANGULATING MULTIMODAL REPRESENTATIONS OF AFFECTIVE EXPERIENCES DURING NATURALISTIC MOVIE

VIEWING LUKE CHANG DARTMOUTH COLLEGE, HANOVER, NEW HAMPSHIRE, UNITED STATES BACKGROUND: Emotions reflect coordinated, multi-system responses to events and situations relevant to survival

and well-being. These responses emerge from appraisals of personal meaning that reference one’s goals, memories, internal body states, and beliefs about the world. Consequently, vmPFC

activity involved in processing these subjective appraisals appears to be highly idiosyncratic across individuals. We developed a novel computational framework to characterize the

spatiotemporal dynamics of the vmPFC in processing our ongoing experiences. METHODS: Participants watched the 45-minute pilot episode Friday Night Lights while undergoing fMRI in Study 1 (n 

= 13) and Study 2 (n = 35) and while we recorded their their facial expressions (Study 3, n = 30) and 16 subjective emotional feelings (Study 4, 183). In study 5 (n = 14), we recorded local

field potentials from stereotactic electrodes implanted in patients undergoing intervention for epilepsy. In study 6, we performed a meta-analysis across n = 38 previously published datasets

to identify the consistency of inter-subject correlations across the brain across a variety of naturalistic stimuli. RESULTS: Overall, we observed very little evidence of synchronization in

brain activity in the vmPFC compared to sensory cortex. This finding replicated across Study 1, 2, 5, and 6. We used a hidden markov model (HMM) to identify discrete latent state changes

across different measurement modalities and performed an inter-experiment factor analysis across measurement modalities to triangulate changes in affective states that manifested across

Studies 1-5. We found that the vmPFC slowly transitions through a series of discretized states that broadly map onto affective experiences. Although these transitions typically occur at

idiosyncratic times across people, participants exhibited a marked increase in state alignment during high affectively valenced events in the show. CONCLUSIONS: Our work suggests that the

vmPFC ascribes affective meaning to our ongoing experiences and that affective states can be objectively inferred based on patterns of brain activity. Furthermore, we observed strong

evidence that participants are rarely occupying the same affective state in response to observing the same stimuli, which poses new theoretical and methodological challenges for studying

affective experiences. DISCLOSURE: Nothing to disclose. 25.4 CLINICAL RELEVANCE OF TASK-EVOKED PATTERNS OF FUNCTIONAL CONNECTIVITY SARAH YIP YALE SCHOOL OF MEDICINE, NEW HAVEN, CONNECTICUT,

UNITED STATES BACKGROUND: Emerging data from basic human neuroscience indicates that functional connectivity data acquired during task performance (vs. resting state) is optimal for accurate

brain-behavior modeling of trait associations. This approach has only recently been introduced to clinical research. METHODS: In Study 1, we tested the utility of different types of

task-based fMRI data for predicting outcomes among poly-substance-using individuals using a connectome-based machine learning approach. In Study 2, we used a similar approach to test whether

the accuracy of predictive models of future alcohol use also differed as a function of task type—hereafter referred to as brain state—in a longitudinal sample of youth (N~1,200). In Study

3, we present ongoing work using a dense sampling approach (i.e., weekly scanning over 2 months) to study changes in brain-based signatures of abstinence (such as those identified in Study

1), as assessed across multiple brain states among individuals in methadone treatment. RESULTS: Study 1: Findings demonstrated that the accuracy of machine learning models of abstinence

during treatment (% of drug-negative Utox samples over three months) differed as a function of brain state. Reward-related brain states were specific for predicting future cocaine use,

whereas inhibitory-related brain states were specific for predicting future opioid use. Based on this striking distinction—that different brain states are specific for predicting different

substance-use behaviors in the same individuals—we next tested the effect of brain state in predicting risky alcohol-use in youth. Study 2: Results indicated significant sex divergence in

the accuracies of brain-behavior models of future alcohol-use severity, such that female-only models consistently outperformed male-only models. Specifically, female-only models successfully

predicted future and current severity across both reward-related and inhibitory control states. In contrast, male-only models were successful in predicting current severity using

connectivity data acquired during inhibitory control—but not reward—brain states, indicating brain-state specificity of predictive models of alcohol-risk in males only. CONCLUSIONS: Brain

state manipulation via performance of in-scanner tasks is optimal for prediction of complex, real world, clinical outcomes. DISCLOSURE: Nothing to disclose. MINI PANEL 26. MINORITY STRESS

AND ITS MECHANISMS: FROM SYSTEMIC AND PROXIMAL SOCIAL CONTEXT TO MENTAL HEALTH DISPARITIES IN SEXUAL AND GENDER MINORITY YOUTH 26.1 ADOLESCENTS’ SEXUAL ORIENTATION AND BEHAVIORAL AND NEURAL

REACTIVITY TO PEER ACCEPTANCE AND REJECTION: THE MODERATING ROLE OF FAMILY SUPPORT KIRSTY CLARK VANDERBILT UNIVERSITY, NASHVILLE, TENNESSEE, UNITED STATES BACKGROUND: Adolescents who face

peer rejection are more likely, compared to their accepted peers, to anticipate and react severely to peer rejection (Downey at el., 1998) and to display blunted neural reactivity to peer

acceptance (Rappaport et al., 2019), which has been linked with higher depressive symptoms (Kujawa et al., 2019). Sexual minority adolescents frequently endure stigma-related peer rejection,

yet scant research examines sexual orientation differences in behavioral and neural reactions to peer feedback. Further, while family support can buffer negative mental health impacts of

peer rejection, no studies have examined the protective (i.e., moderating) role of family support in studies assessing sexual orientation and behavioral and neural reactivity to peer

feedback. The present study examined associations between sexual orientation, behavioral and neural reactions to peer feedback, and the potentially protective role of family support.

METHODS: In a community sample of adolescents approximately 15 years old (47.2% female; n = 36 sexual minority, n = 310 heterosexual) participants completed the Island Getaway task while EEG

data were recorded (Kujawa et al., 2014). The Island Getaway task is a computerized social interaction task where, across several rounds of game play, participants vote to reject or accept

peer co-players (who are, in fact, pre-set computerized co-players) and receive rejection and acceptance feedback from these peer co-players. Voting data were used to calculate behavioral

measures of ingratiation (i.e., total number of votes to ‘keep’ co-players who had voted the participant off in the previous round) and alliance-building (i.e., the total number of votes to

‘keep’ co-players who had voted to keep the participant in the previous round). The reward positivity, an event-related potential (ERP) thought to index reward processing, measured

individual differences in neural reactivity to peer acceptance versus rejection (i.e., residual RewP). Family support was assessed through the Multidimensional Scale of Perceived Social

Support (MSPSS; Zimet et al., 1988). RESULTS: Sexual minority adolescents reported significantly lower family support (M = 20.44) than heterosexual adolescents (M = 22.98, p = 0.013). Linear

regression models showed that sexual minority adolescents engaged in a significantly higher average number of ingratiation efforts than heterosexual adolescents (b = 1.31, p = 0.046). Main

effects for alliance-building and residual RewP were not significant at p < 0.05. An interaction between sexual orientation and family support predicted residual RewP (p-value = 0.039).

Decomposing family support by sexual orientation interaction into simple slopes demonstrated that the sexual orientation difference in residual RewP was significant at low levels of family

support (t = -2.68, p = 0.007) but not at mean (t = -1.28, p = 0.202) or high levels of family support (t = 0.29 p = 0.770). CONCLUSIONS: Results found that sexual minority adolescents show

distinct behavioral and neural reactions to peer feedback and suggest that sexual minority adolescents with low family support exhibit particularly blunted neural reactivity to peer

acceptance, which has been implicated in the development of depressive symptoms. These results may have implications for developing preventive interventions, especially for sexual minority

adolescents with low family support. DISCLOSURE: Nothing to disclose. 26.3 MEDIAL PREFRONTAL CORTEX ACTIVITY TO REWARD OUTCOME MODERATES THE ASSOCIATION BETWEEN VICTIMIZATION DUE TO SEXUAL

ORIENTATION ON DEPRESSION IN YOUTH KRISTEN ECKSTRAND UNIVERSITY OF PITTSBURGH, PITTSBURGH, PENNSYLVANIA, UNITED STATES BACKGROUND: Sexual minority youth and young adults (SMY) are 3x more

likely to experience depression than heterosexual peers. Minority stress theory posits that this association is explained by sexual orientation victimization, which acts as a stressor to

impact depression. Emerging evidence suggests that stress increases depression severity by altering activity in neural reward systems. For those vulnerable to the effects of stress,

victimization may worsen depression by altering activity in neural reward systems. However, imaging studies among SMY have primarily focused on neural differences by sexual orientation

identity alone and no studies have examined the influence of sexual orientation victimization on neural reward system function. The purpose of this study was to examine whether neural reward

activity moderates the relationship between sexual orientation victimization and depression, and whether these patterns differed by sexual orientation. METHODS: 81 participants ages 16-22

years old (41% SMY, 59% self-reported female sex, 52% racially marginalized) were enrolled into the study. Participants completed self-report measures of depression using the Center for

Epidemiological Studies-Depression (CESD) scale; anhedonia using the Snaith Hamilton Pleasure Scale (SHAPS); and identity victimization experiences based on sexual orientation, race, and

gender. Participants underwent functional magnetic resonance imaging during a standardized monetary reward task to measure neural response to reward. Neural reward activity was determined by

significant neural response (pFWE <0.05) to reward>neutral outcome within the Neurosynth reward mask in SPM12. BOLD activity was extracted for subsequent analyses. Age was included as

a covariate in all models. RESULTS: Bilateral ventral striatum (VS), medial prefrontal cortex (mPFC), dorsal anterior cingulate cortex, and right orbitofrontal cortex were significantly

activated during reward outcome. SMY experienced higher severity depression (β = 14.96, p = <0.001, 95CI = 9.68, 20.25) and anhedonia (β = 5.98, p = 0.017, 95CI = 0.79, 7.75), and more

victimization events based on sexual orientation (t[1,75]=-5.69, p = <0.001). Higher depression severity was also associated with more victimization based on sexual orientation (β = 0.95,

p = 0.047, 95CI = 0.14, 1.87). More sexual orientation victimization predicted higher mPFC response to reward (β = 0.18, p = 0.04, 95CI = 0.01, 0.36). In an interacting moderation model (R2

 = 0.493, p < 0.001) mPFC reward activation moderated the relationship between sexual orientation victimization and depression (β = 4.79, p = 0.04, 95CI = 0.22, 9.37), with higher mPFC

activation and higher sexual orientation victimization being associated with higher depression. Sexual orientation did not moderate this interaction. No other regions of significant neural

reward activity moderated orientation victimization-depression relationships. CONCLUSIONS: Depression in SMY was related to sexual orientation victimization, but only in the context of

higher mPFC activation, a pattern observed in depressed youth. These novel results provide evidence for neural reward sensitivity as a vulnerability factor for depression in SMY, suggesting

mechanisms for disparities, and is a first step towards a clinical neuroscience of minority stress in SMY. DISCLOSURE: Nothing to disclose. MINI PANEL 27. SEROTONIN REUPTAKE INHIBITORS

(SRIS) AND PREGNANCY: SERIAL MONTHLY CHANGES IN PLASMA SERTRALINE CONCENTRATIONS, THE USE OF PLATELET SEROTONIN TO ASSESS SRI BIOEFFECT, AND THE ASSOCIATION OF MATERNAL AND CORD BLOOD

PLATELET CONCENTRATIONS WITH FETAL-NEONATAL NEUROBEHAVIOR 27.1 CHANGES IN SERTRALINE PLASMA CONCENTRATIONS ACROSS PREGNANCY AND POSTPARTUM AND RELATIONSHIP TO CYTOCHROME P450 2C19, 2C9 AND

2D6 ENZYME ACTIVITIES KATHERINE WISNER NORTHWESTERN UNIVERSITY SCHOOL OF MEDICINE, CHICAGO, ILLINOIS, UNITED STATES BACKGROUND: Serotonin reuptake inhibitors are prescribed to 5-8% of

pregnant persons. Sertraline (SERT) is the most frequently prescribed drug, although few data about its pharmacokinetics in pregnancy is available. The primary metabolic pathway is the

conversion of SERT to its weakly active metabolite, desmethylsertraline (DMSERT). SERT metabolism is catalyzed by CYP3A4, CYP2C9, CYP2C19, CYP2B6 and CYP2D6. CYP3A4 comprises more than 30%

of the Phase I metabolic capacity and is the primary driver of SERT metabolism. The goal of this study was to characterize plasma SERT concentration to dose (C/D) ratios across pregnancy and

postpartum as well as evaluate the effect of pharmacogenetic variability on SERT elimination. METHODS: This prospective observational cohort study was conducted in pregnant people receiving

maintenance SERT therapy and who elected to continue during pregnancy. The study was conducted at 3 NICHD-funded Obstetrical-Fetal Pharmacology Research Center sites. Pregnant people with a

singleton pregnancy ≤18 weeks gestation and with at least one previous episode of MDD were eligible. Blood samples were obtained 24-hours post-dose every 4 weeks across pregnancy and twice

postpartum for measurement of plasma concentrations of SERT and DMSERT. We analyzed the changes in trough plasma SERT C/D ratios as well as plasma SERT to DMSERT (S/DS) ratios (a measure of

metabolic capacity). Plasma SERT and DMSERT concentrations were measured by high performance liquid chromatography-tandem mass spectrometry. We explored the effect of genetic variability in

enzymes responsible for SERT metabolism on the SERT C/D ratios. Participants were genotyped for variants in CYP2C9, CYP2C19, and CYP2D6, using commercial allelic discrimination assays with

Taqman probes. RESULTS: Pregnant people (N = 47) were enrolled. Overall mean SERT C/D ratios decreased early in pregnancy and remained low until after delivery. In the last 4 weeks of

pregnancy, the mean SERT C/D ratio, 0.25 (95% CI, 0.19-0.3) was less than the postpartum ratio, 0.32 (95% CI, 0.27-0.37). The model-estimated mean S/DS ratios were highest after birth,

demonstrating that SERT elimination in pregnancy was comparatively increased. C/D ratios in those with functional CYP2C19 activity did not significantly change, while ratios in those with

poor or intermediate CYP2C19 activity decreased by 51%. This decrease likely reflects increased SERT clearance by CYP3A4. Although CYP2C19 activity decreases in pregnancy, those with

poor/intermediate CYP2C19 phenotypes do not decrease activity to offset the increased activity of other CYPs. CONCLUSIONS: An increase in SERT metabolism and declining concentrations in

pregnancy compared to postpartum occurred for all genetic phenotypes. Although some experts recommend CYP2C19 testing prior to SERT therapy in nonpregnant adults to avoid adverse effects

from elevated concentrations in people with CYP2C19 poor and intermediate metabolizers, these same individuals, when pregnant, are at risk for undertreatment as SERT concentrations decline.

No meaningful SERT concentration differences were associated with CYP2C9 and CYP2D6 polymorphisms. DISCLOSURE: Nothing to disclose. 27.2 ASSESSMENT OF SEROTONIN REUPTAKE INHIBITOR BIOEFFECT

BY MEASUREMENT OF PLATELET SEROTONIN GEORGE ANDERSON YALE UNIVERSITY SCHOOL OF MEDICINE, NEW HAVEN, CONNECTICUT, UNITED STATES BACKGROUND: Serotonin reuptake inhibitors (SRIs) are widely

used to treat depressed mood and anxiety. Limited neuroimaging studies indicate that typical clinical doses of the SRIs result in approximately 80% occupancy of the central serotonin (5-HT)

transporter (SERT). Researchers have also attempted to assess the SRI inhibitory bioeffect by examining decreases in platelet 5-HT level. This more accessible approach depends upon the fact

that all platelet 5-HT is taken up via SERT over the platelet’s 10-day lifespan. The platelet measure provides a time-averaged index of SRI bioeffect and has advantages compared to plasma

drug level when assessing SRI exposure. We will overview the theoretical basis for the measure and its practical applications and present a systematic analysis of the reported apparent

bioeffects of SRIs. Of special relevance to this panel is the use of platelet 5-HT in assessing maternal and infant SRI exposure when SRIs are used to treat maternal perinatal depression.

METHODS: A systematic literature search used search terms including platelet 5-HT, whole blood 5-HT, antidepressant, selective reuptake inhibitor, selective serotonin reuptake inhibitor,

serotonin-norepinephrine reuptake inhibitor, as well as individual SRIs. Bibliographies of identified papers were also searched for relevant papers. Studies were included if mean pre-drug

and on-drug 5-HT values were given, if standard clinical doses were used, and if treatment lasted for at least two weeks. The mean value reported for the on-drug 5-HT value as a percent of

baseline was extracted and the mean, standard deviations (SDs) and median values of the study means were calculated for the combined SSRI group and for individual SRIs with 3 or more

studies. RESULTS: Mean (SD) and median on-drug 5-HT values as a % of baseline for the different groups are given in Table 1., along with number of studies included in each group. The on-drug

5-HT values in the clomipramine group were significantly lower than in the combined SSRI group: Kruskal-Wallis H statistic 6.13, p = 0.013 Table 1. SRI Bioeffect: On-Drug Platelet 5-HT

Values as a Percent of Baseline GROUP n MEAN (SD) MEDIAN Combined SSRI 38 17.4 (11.3) 14.2 Clomipramine 15 12.3 (7.7) 8.0 Fluoxetine 10 14.6 (7.8) 14.6 Paroxetine 8 11.4 (10.7) 9.7

Fluvoxamine 4 13.8 (2.1) 14.1 Sertraline 4 19.6 (13.5) 14.5. CONCLUSIONS: This systematic review of the use of platelet serotonin to assess the bioeffect of clinical doses of SRIs indicated

that: 1) studies have found mean and median on-drug 5-HT values to be 8-20% of baseline values; 2) clomipramine showed significantly greater reductions in platelet 5-HT compared to other

SSRIs. The overall results are consistent with limited neuroimaging data and within vitro evidence. The clomipramine results are consistent with prior indications that typical clinical doses

of clomipramine lead to relatively greater SERT inhibition. In general, the results support the use of platelet 5-HT to assess SRI bioeffect. DISCLOSURE: Nothing to disclose. 27.3 FETAL

NEUROBEHAVIOR IN THE CONTEXT OF BIOEFFECT FROM GESTATIONAL SELECTIVE SEROTONIN REUPTAKE INHIBITORS AMY SALISBURY VIRGINIA COMMONWEALTH UNIVERSITY, RICHMOND, VIRGINIA, UNITED STATES

BACKGROUND: More than a third of women with depressive disorders in pregnancy are treated with selective serotonin reuptake inhibitor (SSRI) antidepressants during the gestational period.

The benefits of treating depression in pregnancy with SSRIs has been shown to outweigh potential risks to the fetus. However, gestational SSRI exposure alters serotonin signaling and has

been associated with adverse respiratory, gastrointestinal, and neurobehavioral signs in the newborn. Although these signs diminish over the first postnatal month, recent evidence suggests

potential long-term changes in serotonin-related functions. However, there is no consensus on the mechanisms of these early or later alterations. Therefore, this study examined fetal

neurobehavior at the time of SSRI exposure compared to no exposure, with and without concurrent maternal depression and associations between fetal measures, SSRI bioeffect, and later infant

outcomes. METHODS: Pregnant women (N = 243) were enrolled into a prospective, longitudinal observational study examining the impact of prenatal SSRI use on fetal and infant neurodevelopment.

Women were enrolled in second and third trimesters and completed standardized assessments to determine depression diagnosis and gestational SSRI use, dosage, and timing. Standardized fetal

assessments were conducted on one or two occasions between 26- and 36-weeks gestation using ultrasound-based video recordings. Women who were using SSRIs concurrent with the fetal

assessments (N = 50) and who were not using SSRIs at any time in pregnancy (N = 133; Total N = 183) were included in these analyses. At the time of delivery, samples of maternal peripheral

blood (N = 103) and fetal cord blood (N = 97) were collected in a subset of participants to measure maternal and fetal whole blood platelet serotonin levels and SSRI blood concentrations.

Fetal bioeffect was previously confirmed with lowered platelet serotonin levels in women using SSRIs. Mixed models were used to examine group differences for fetal measures; correlation

analyses were used to examine relationships between continuous variables. RESULTS: Compared to fetuses of women with (N = 61) and without depression (N = 72) and no SSRI use in pregnancy,

fetuses with SSRI exposure (N = 50) had lower fetal movement quality, higher jerky to smooth movement ratio, shorter quiescent periods, and fewer fetal breathing movements. There was a

significant negative relationship between fetal movement quality and maternal platelet serotonin levels at birth. Gestational age at birth was positively correlated with cord blood platelet

serotonin levels, but not maternal levels. CONCLUSIONS: These results suggest that the adverse signs observed in some infants after delivery may be present prior to birth and removal of the

SSRI exposure; suggesting that withdrawal or adaptation may not be the primary mechanism. Fetal neurobehavioral patterns found here suggest immaturity of motor and respiratory function which

may have implications for later development. We will discuss these findings in relation to previously reported findings in the newborn period. DISCLOSURE: Nothing to disclose. PANEL 28. HOW

THE EXPOSOME INFLUENCES SUICIDAL BEHAVIOR 28.1 SUICIDAL IDEATION IN RESPONSE TO NEGATIVE LIFE EVENTS: THE ROLE OF EMOTION REGULATION MARIA OQUENDO PERELMAN SCHOOL OF MEDICINE UNIVERSITY OF

PENNSYLVANIA, PHILADELPHIA, PENNSYLVANIA, UNITED STATES BACKGROUND: A subset of suicidal individuals responds to negative events with increased suicidal ideation (SI), placing them at

greater risk for suicidal behavior during those periods. Whether these SI increases relate to lack of coping strategies such as emotion regulation (ER) is unknown. We hypothesized that those

responding to negative events with lower SI increases would engage in ER more during exposure to painful memories. METHODS: Individuals with Major Depression (n = 27) participated in an

fMRI task, recalling painful autobiographical memories, and engaging in ER when instructed. Ecological Momentary Assessment (EMA) at 3, 6, 12, 18 and 24 months after baseline provided 7 days

of responses about stressors, affect and SI. A neural decoder trained to detect ER effects in fMRI data from 77 participants identified a network comprising inferior frontal gyrus,

orbitofrontal cortex (OFC), Anterior Cingulate Cortex (ACC), frontal operculum, left caudate, and insula. A linear mixed effect regression tested whether the ER neural decoder output

moderated the effect of stressors on epoch-to-epoch SI change, adjusted for time, lagged values for SI and affect, and for 2-way interactions between lagged affect and decoder, and lagged

affect and stressors. The neural decoder output was residualized for bold activity. RESULTS: ER decoder output moderated negative events’ impact on SI; those who appeared to use ER more

robustly when recalling negative memories responded to life events with lower changes in SI. This was in contrast to findings in those with lower ER during memory recall (interaction term:

value = -3.87; S.E. = 1.34; df=3126, t = -2.89; p = 0.004). The ER decoder’s moderation of SI change in response to stress remained significant after adjusting for depressive symptoms before

the life event through main effect and interactions. CONCLUSIONS: These pilot findings suggest that those who respond to life events with lower changes in SI may be able to harness ER as a

coping strategy by engaging OFC and ACC when confronted with painful stimuli. Interventions for suicidal individuals such as dialectical behavioral therapy, which teach emotion regulation

are available. Studying whether individuals who respond to life events with greater SI can leverage these interventions to display less propensity towards increased SI in response to stress

would be instructive. DISCLOSURE: C-SSRS: Royalties (Self), Bristol Myers Squibb: Employee (Spouse) Otsuka, MindMed, Alkermes, Sage Therapeutics: Advisory Board (Self), St George’s

University: Board Member (Self) 28.2 CONTRIBUTION OF INDIVIDUAL AND STRUCTURAL PSYCHOSOCIAL EXPOSOME TO SUICIDE ATTEMPTS IN DIVERSE YOUTH RAN BARZILAY PERELMAN SCHOOL OF MEDICINE UNIVERSITY

OF PENNSYLVANIA, PHILADELPHIA, PENNSYLVANIA, UNITED STATES BACKGROUND: Rates of youth suicidal behavior are rising, most steeply among Black youth. Distal and proximal psychosocial stressors

are key suicidality risk factors. The network of environment and social determinants of health is referred to as the “exposome”. We use exposome data to better understand youth suicide risk

and its related racial disparities. METHODS: We combine multi-layer data of individual exposome with geocoded neighborhood exposome in 3 settings: (i) Electronic health record (EHR) of

Children’s Hospital of Philadelphia (CHOP), where we complied data from N = 19,879 youth (Mage=15, 55% Black) presenting to emergency department. (ii) Research and clinical integrated

dataset of 1,514 youth from the Philadelphia Neurodevelopmental Cohort (PNC) study that have longitudinal data on suicide attempts from CHOP EHR. (iii) Research data from the Adolescent

Brain Cognitive Development (ABCD) Study that follows ~12k diverse adolescents across 21 sites in the US. RESULTS: In CHOP data, 10% of patients reported past suicide attempt, with higher

rates in Black (10.9%) Vs. White (8.5%) youth (P < .001). Prevalence of stressors associated with suicide attempt were different between races, with Black youth experiencing 63% more

physical bullying, 29% more sexual abuse, 2-fold more romantic partner violence and 3-fold more involvement in fights. Conversely, White youth experienced 27% more cyberbullying (all P’s 

< .001). In PNC, we found an interaction effect of exposure to assault with geocoded neighborhood measures in association with suicidal ideation (Wald = 8.19, p = .004). Longitudinal

analyses of PNC youth with data on suicide attempts years later revealed that geocoded neighborhood measures (e.g., percent of vacant lots) are among the top ranked predictors in models

predicting future suicide attempt. In ABCD, peer stressors like discrimination, bullying and cyber-victimization were associated with suicide attempts (controlling for demographics, odds

ratios, 95%CI = 1.3, 1.2-1.4; 1.6, 1.5-1.7; 3.7, 2.8-4.8, respectively, P’s < .001). CONCLUSIONS: Results converge across three different data sources to suggest a critical role for

exposome in youth suicide attempt risk. Data highlight the potential of integrating individual with structural exposome to study youth suicide attempt, its related racial disparities, and

optimize its prediction. DISCLOSURE: Taliaz Health: Advisory Board (Self), Taliaz Health: Stock / Equity (Self), Zynerba Pharmaceuticals: Advisory Board, Consultant (Self) 28.3 CHILDHOOD

ADVERSITY AND RECENT LIFE STRESS BRAIN RELATIONSHIPS IN DEPRESSION AND SUICIDE J. JOHN MANN COLUMBIA UNIVERSITY, NEW YORK, NEW YORK, UNITED STATES BACKGROUND: Childhood adversity and recent

life stressors increase major depression and suicidal behavior risk in adulthood. Little is known of genetic and epigenetic mediators. The serotonergic system is abnormal in suicidal

behavior. We therefore examined relationships between serotonin (5-HT) system components and childhood adversity (Study 1) and recent life stress (Study 2) in major depression and suicidal

behavior. METHODS: Study 1: N = 192 MDDs, and N = 88 healthy volunteers (HV), majority female, were assessed for childhood adversity and recent life changes and genotyped for rs6295 in the

5HT1AR gene. DNA methylation was assayed at three upstream promotor sites (-1017, -1007, -681). 5-HT1A receptor binding (BPF) was quantified by PET with [11C]WAY100635 in a subgroup (N = 

119). Study 2: Recent life stress was measured by Ecological Momentary Assessment (EMA) over one week in 25 depressed MDD participants (12 suicide attempters, majority female) before

undergoing serotonin transporter (5-HTT) and 5-HT1A receptor PET imaging. RESULTS: Study 1: Recent stress correlated positively with methylation at the -681 CpG site promotor site, adjusted

for diagnosis, sex and age, and had positive and region-specific correlations with 5-HT1A BP in MDD, but not HVs. In MDDs, but not in HVs, methylation at only the -1007 CpG site had positive

and region-specific correlations with 5-HT1A BP Childhood adversity was not associated with DNA methylation or 5-HT1A BP. Study 2: In suicide attempters, but not in non-attempters,

region-specific associations between both 5-HTT (p = 0.006) and 5-HT1A (p = 0.01) BPND and EMA stress emerged. 5-HT1A BPND correlated positively with EMA stress in attempters in 8/10 regions

(p’s<0.05) and negatively with 5HTT binding in one region. CONCLUSIONS: Study 1 Recent stress increased 5-HT1A receptor BP via DNA methylation of inhibitory transcription factor binding

sites, thereby impacting MDD psychopathology and the risk of suicidal behavior. Study 2: Spatially localized lower 5-HTT binding and widespread higher 5-HT1A binding indicate impaired

serotonin signaling in suicide attempters and not in non-attempters. Epigenetic relationships suggest a pre-existing genetic or epigenetic effects of childhood adversity, but this was not

detected. DISCLOSURE: Research Foundation for Mental Hygiene: Royalties (Self) STUDY GROUP 29. BRIDGING THE GAP BETWEEN PRE-CLINICAL AND CLINICAL STUDIES: THE PROMISE AND CHALLENGES OF

FORWARD AND REVERSE TRANSLATIONAL APPROACHES JENNIFER BLACKFORD*, MARISA SILVERI, DANNY WINDER, TALLIE Z. BARAM, MICHAEL YASSA, MOHAMMED MILAD, NED KALIN, DANIEL PINE, KERRY RESSLER, WILLIAM

CARLEZON, ISABELLE ROSSO, JENNIFER BLACKFORD MUNROE-MEYER INSTITUTE, UNIVERSITY OF NEBRASKA MEDICAL CENTER, OMAHA, NEBRASKA, UNITED STATES STUDY GROUP SUMMARY: There have been extensive

advances in our knowledge of the neurobiology of mental health disorders over the past decade. The translation of findings from bench to bedside, however, remains slow, costly, and is often

fraught with substantial limitations. There has been much discussion around how to improve components of the translational process—for example, by using different animal models, developing

novel and innovative experimental tasks, or selecting appropriate research tools. Nevertheless, a major challenge in translational research has been the wide gap between the scientists

conducting preclinical animal studies and those conducting clinical human studies. Most scientists focus exclusively on only animal or human models, leading to silo mentality. Among factors

isolating camps include the use of different research approaches, methods, and terminology, as well as presenting data at conferences and publishing findings in journals most associated with

one line of research. Funding mechanisms and study section assignments also contribute to separation of animal from human studies, and vice versa. Despite these longstanding challenges, new

approaches are beginning to emerge. This study group will highlight work by scientists dedicated to bridging gaps between animal and human studies through several perspectives. The first

perspective is from scientists who have cross-trained in both human and experimental model research. A second perspective arises from scientists collaborating tightly across research

modalities. Combining these perspectives can help mitigate silos by developing transdisciplinary language, methods, and interpretations. Especially compelling are scientific partnerships

that provide a foundation for rapid and iterative forward and reverse translational approaches. Forward translation has been a hallmark of the bench to beside approach. However, reverse

translation from human to animal models is equally important, as identifying the power and limitations of experimental systems and matching human questions to appropriate models is crucial.

An iterative approach, that supports real-time forward and reverse translation, shows great potential for moving the field forward. A combination of reverse and forward translation can

enhance identification of causality and mechanisms in animal models that can then be tested in human disease, and can encourage the application of findings from human studies to develop

novel approaches and hypotheses to be tested in preclinical studies. An iterative, transdisciplinary process is critical for advancing construct-validated investigations of disorder-relevant

neural circuitry, with the ultimate goal to guide the identification of novel treatment targets, development of novel therapeutics, and enhancement of current therapies. This study group

will include scientists who have cross-trained in animal models and human studies (Ressler, Milad, Silveri) and pairs of scientists who are actively collaborating to achieve iterative

forward and reverse translations across their research programs (Winder-Blackford, Kalin-Pine, Baram-Yassa, Carlezon-Rosso). DISCLOSURE: Nothing to disclose. STUDY GROUP 30. MECHANISMS OF

DYSREGULATED SLEEP AND CIRCADIAN RHYTHMS ACROSS THE PSYCHIATRIC SPECTRUM: CLINICAL STUDIES TO DECIPHER THE TRANSDIAGNOSTIC ROLE OF SLEEP IN MENTAL AND COGNITIVE HEALTH ELLEN LEE*, JENNIFER

GOLDSCHMIED, KATHLEEN MERIKANGAS, BRANT HASLER, BRYCE MANDER, BENGI BARAN, MICHAEL IRWIN, REBECCA HENDRICKSON, AJ SCHWICHTENBERG, DAVID KUPFER UNIVERSITY OF CALIFORNIA, SAN DIEGO, LA JOLLA,

CALIFORNIA, UNITED STATES STUDY GROUP SUMMARY: Sleep and circadian rhythm disturbances are core features of most neuropsychiatric disorders, but can also be viewed as modifiable targets with

important implications for the progression of psychiatric disorders as has been shown by the improvement in depressive symptoms following cognitive behavioral interventions for sleep.

Sleep/circadian factors are intricately tied to the neural circuitry of emotion regulation, reward processing, cognitive control, and sensory processing, suggesting trans-diagnostic

mechanistic roles in the development and maintenance of neuropsychiatric disorders. This study group will unite nine of the field’s most relevant researchers–balanced by gender and career

stage–to discuss sleep/circadian rhythms impairment across depression, bipolar disorder, schizophrenia, post-traumatic stress disorder, neurodegeneration, autism, and substance use. The

non-ACNP member participants have expertise in aspects of sleep and/or circadian rhythms in depression (JG), substance use (BH), neurodegeneration (BM), psychotic disorders (BB), autism

(AS), and post-traumatic stress disorder (RH). We propose that the study of sleep and circadian rhythms is essential to the investigation of the pathophysiology, maintenance, and treatment

of psychiatric disorders. We will address the following questions: * 1. A broad range of underlying biological mechanisms connect sleep/circadian rhythms to cognitive and psychiatric

functioning – e.g., inflammation, amyloid and tau deposition, slow-wave activity-mediated neuroplasticity changes, alterations in neural circuitry underlying reward function and cognitive

control, oxidative stress, and sympathetic dysregulation. What is the current evidence of how these sleep/circadian-related mechanisms overlap with the pathophysiology of psychiatric

conditions? * 2. Accurate assessment of sleep, circadian rhythms, and associated mechanisms are a longstanding challenge for researchers. Gold standard approaches like In-lab polysomnography

for sleep and in-lab melatonin or core body temperature assessments for circadian rhythms are expensive, and burdensome. Novel wearable and environmental sensors can offer complementary

information about sleep and circadian parameters, including night-to-night variability of sleep, over time and under naturalistic conditions. What are the strengths and limitations of

different sleep/circadian assessments? How acceptable and feasible are such approaches for psychiatric populations? * 3. Sleep/circadian rhythm disorders are often attributed to primary

psychiatric disorder, and thus, underdiagnosed, and untreated. Psychiatric treatments can also adversely affect sleep architecture and quality. How can clinical studies systematically

examine the impact of sleep/circadian treatments on psychiatric symptoms and of psychiatric treatments on sleep/circadian problems? Which treatment approaches improve both sleep/circadian

function and psychiatric symptoms? Which treatment approaches that target sleep/circadian function show promise in preventing psychiatric disorders such as depression? DISCLOSURE: Nothing to

disclose. PANEL 31. NEUROBIOLOGICAL PATHWAYS FROM EARLY ADVERSITY TO PSYCHOPATHOLOGY 31.1 DEVELOPMENTAL TIMING AND KEY DIMENSIONS OF EARLY ADVERSITY: MULTIVARIATE ASSOCIATIONS WITH WHITE

MATTER MICROSTRUCTURE DYLAN GEE YALE UNIVERSITY, NEW HAVEN, CONNECTICUT, UNITED STATES BACKGROUND: Delineating links between early adversity, neurobiology, and psychopathology is critical to

identifying trajectories of risk and resilience following adversity exposure. Microstructural remodeling of white matter pathways represents one potential mechanism linking childhood

adversity exposure with psychopathology across development. Given vast heterogeneity in the nature of early adversity and outcomes, conceptual models of early adversity have increasingly

focused on key dimensions of adversity exposure. The ways in which specific aspects of adversity—such as the developmental timing of an exposure—may uniquely shape developing white matter

have yet to be elucidated. METHODS: N = 107 adults (ages 18-30) underwent a diffusion-weighted imaging scan and completed a modified version of the UCLA PTSD Reaction Index designed to

examine key dimensions of lifetime stress exposure. We examined multivariate associations between specific features of adversity exposure (developmental timing, threat, deprivation,

predictability, controllability) and 43 white matter tracts using regularized canonical correlation analysis. We then examined associations between the canonical modes that were identified

and symptoms of psychopathology. RESULTS: Results identified 11 canonical modes that differed significantly from chance (Bonferroni corrected; p < .01). Several meaningful modes emerged.

One mode predominantly indexed adversity characterized by deprivation that was experienced across childhood and adolescence and alterations in the optic radiation tracts. A second mode

predominantly indexed adversity that was perceived as predictable and controllable and experienced in adolescence and alterations in the cingulum cingulate. Further, the canonical mode

characterized by deprivation was associated with internalizing symptoms in adulthood (p = .01). CONCLUSIONS: Our findings suggest that sensory processing tracts may be most affected by

experiences characterized by deprivation, whereas frontolimbic and frontostriatal tracts may be more sensitive to experiences characterized by controllability and predictability. These

results highlight the utility of multivariate approaches for parsing heterogeneity related to early adversity and indicate that such techniques can yield clinically meaningful information.

DISCLOSURE: Nothing to disclose. 31.2 NEUROSTRUCTURAL ALTERATIONS FOLLOWING TRAUMATIC EXPERIENCES DURING CHILD AND ADULTHOOD MAURIZIO SICORELLO CENTRAL INSTITUTE OF MENTAL HEALTH, MANNHEIM,

GERMANY BACKGROUND: Animal studies have shown that adverse experiences, and their neuroendocrine mediators, can lead to volumetric changes in hippocampus and amygdala. Similar volumetric

alterations have also been observed in humans suffering from posttraumatic stress disorder and, for the hippocampus, even healthy trauma-exposed individuals. Recent evidence suggests that

alterations can be dependent on the neurodevelopmental timing of events in childhood and adolescence. Still, these studies have thus far not compared effects to traumatic experiences

occurring in adulthood and the specificity to psychopathology vs trauma-exposure is often not tested. METHODS: A total of 155 women were allocated into one of six age-matched groups

according to timing of traumatization (childhood vs adulthood) and psychopathology (PTSD vs trauma-exposed healthy vs trauma-naïve healthy). Volumes of amygdala and hippocampus were compared

between these groups. Six additional exploratory regions of interest (ROI) were included based on a recent meta-analysis. Statistical tests were controlled for the number of ROIs, with

additional correction for pair-wise post-hoc comparisons. RESULTS: There was a significant disordinal interaction between timing and group for amygdala volume (p < .001, η2 = .08): For

childhood trauma, participants with PTSD had lower volumes than healthy trauma-naive controls in both hemispheres (left: p = 017, g = 0.81; right: p = .31, g = 0.75). For adulthood trauma,

the PTSD group had larger bilateral volumes in comparison to both trauma-exposed controls (left: p = .002, g = 1.01; right: p = .013, g = 0.83) and trauma-naive controls (left: p = .002, g =

 1.01; right: p = .002, g = 0.73). The interaction between time and group was not statistically significant for the other regions (all p > .08). CONCLUSIONS: Timing of traumatization was

associated with amygdala volumes throughout the lifespan, with opposite effects dependent on age at trauma occurrence. These findings have to be viewed in light of fundamental confounds that

distinguish trauma during child- and adulthood. Regardless of these confounds, the opposing effects for amygdala volume have important implications for the interpretation of

trauma-contingent structural findings in this region. DISCLOSURE: Nothing to disclose. 31.3 LOOKING INTO TROUBLED WATERS: CHILDHOOD EMOTIONAL MALTREATMENT AND NEURAL RESPONSES TO PROLONGED

GAZING INTO ONE’S OWN AND OTHERS EYES BERNET ELZINGA LEIDEN UNIVERSITY, LEIDEN, NETHERLANDS BACKGROUND: Eye contact is crucial for the formation and maintenance of relationships, not only

with others, but also with oneself. Childhood emotional maltreatment can disrupt the development of positive views about self and others. One potent way the psychological consequences of

emotional maltreatment may become apparent in daily life is when gazing into one’s own or other people’s eyes. METHODS: To investigate how emotional maltreatment may affect people’s

responses when looking into one’s own or others’ eyes adult participants (mean age=49.9 years; n = 79, n = 35 men and n = 44 women) viewed videos of direct versus averted gaze of themselves

and an unfamiliar adult while neural activity was recorded using fMRI, as well as self-reported mood and time spent looking into one’s own and others’ eyes using an eye tracker. Mood and

gaze responses were analyzed in R (R Core Team (42), version 3.6.1) with lme4 for mixed model analysis. Significance for analyses on mood and gaze responses was set at p < 0.05

(two-tailed) and Cohen’s d effect sizes were calculated for significant effects. To examine associations between CEM and participants’ neural responses to self and unfamiliar adult, two

separate regression analyses were conducted in SPM including CEM as covariate of interest and neural responses to either self or unfamiliar other adult as outcome. All whole-brain results

were corrected for multiple comparisons with Family-Wise Error (FWE) cluster correction at p < 0.05 (with a p < 0.001 cluster-forming threshold). RESULTS: Participants who experienced

more maltreatment reported lower mood after videos of direct, but not of averted gaze (of themselves and unfamiliar adults). Individuals who reported higher levels of maltreatment exhibited

increased activity in ventromedial prefrontal cortex when looking at videos of themselves. Maltreatment did not correlate with gaze or neural responses during eye contact with an unfamiliar

adult. CONCLUSIONS: These results suggest that childhood emotional maltreatment may have a long-lasting impact on social-affective processes during eye contact and on self-referential

processes that may particularly surface when looking into one’s own eyes. These findings may be an interesting point of departure for clinical interventions to strengthen self-views.

DISCLOSURE: Nothing to disclose. PANEL 32. FROM BIG DATA TO PERSONALIZED PREDICTION IN PSYCHIATRY 32.1 PERSONALIZED VULNERABILITY INDEXES FOR SEVERE MENTAL ILLNESSES: TRANSLATING BIG DATA

FINDING TO THE LEVEL OF INDIVIDUAL SUBJECT PETER KOCHUNOV MARYLAND PSYCHIATRIC RESEARCH CENTER, UNIVERSITY OF MARYLAND SCHOOL OF MEDICINE, ELLICOTT CITY, MARYLAND, UNITED STATES BACKGROUND:

Big Data research in major depressive (MDD), bipolar (BD), and schizophrenia spectrum (SSD) disorders led to stable and reproducible neuroimaging deficit patterns. We propose the Regional

Vulnerability Index (RVI) to quantify an individual’s brain-wide similarity to the expected SMI patterns, analogous to polygenic risk scores (PRS) that measure an individual’s similarity to

genome-wide risk patterns. We hypothesized that RVI is a brain intermediary between genome and symptoms and is sensitive to both genetic and environmental risks for mental disorders. We

tested this hypothesis in clinical samples as well as using it to evaluate the impact of genetic and environmental factors on development of illness in children. METHODS: Sensitivity and

specificity analysis in UKBB sample (N = 37285) evaluated RVI in mental illnesses and compared it to PRS. RVI-SSD was evaluated as a predictor of treatment resistance in sample of N = 122 of

SSD subjects. The RVI-MDD was tested as predictor of higher symptoms of depression in N = 8,089 UKBB subjects with clinical and sub-clinical MDD. Adolescent Brain and Cognitive Development

(ABCD) (N = 8940, age=10 years) was used to evaluate if ancestral history (AH) for SSD, perinatal risk factors and negative early life experiences can lead to higher individual similarity

with the adult SSD brain patterns in children. RESULTS: RVI for mental illnesses were sensitive and specific, more so than PRS. Treatment-resistant patients showed significantly higher

RVI-SSD than that in treatment-responsive patients (p = 0.01). Group-average increase in depressive symptoms was mirrored by increases in RVI-MDD (r = 0.85). In children, RVI-SSD had

captured more variance from family history of SSD, prenatal and negative life events than any whole-brain or individual regional brain measurements. Higher RVI-SSD scores in children were

already associated with worse performance on the overall cognition (p < 10-20), working memory (p < 10-10) and processing speed (p < 10-5). CONCLUSIONS: Big Data analyses provided

stable and reproducible illness deficit patterns that can be used to derive individual indexes of similarity. RVI can be used translate these patterns into sensitive and specific biomarkers

that can be used in clinical and epidemiological samples, as well as to evaluate and quantify risk factors for mental illness in ABCD cohort. DISCLOSURE: Nothing to disclose. 32.2 PREDICTING

COGNITIVE IMPAIRMENT AND AMYLOID POSITIVITY USING A DATA-DRIVEN VULNERABILITY INDEX LAUREN SALMINEN KECK SCHOOL OF MEDICINE UNIVERSITY OF SOUTHERN CALIFORNIA, MARINA DEL REY, CALIFORNIA,

UNITED STATES BACKGROUND: Composite neuroimaging measures have shown greater utility in identifying cognitively normal individuals at risk for Alzheimer’s disease (AD) when compared to

region-of-interest (ROI) methods that focus on a single structure (e.g., hippocampal volume). Different “metaROI’s” have been identified in the literature, but the structures included in

these measures vary across studies and are typically based on a priori assumptions of the temporal sequence of AD pathophysiology. Here we evaluated the utility of a data-driven measure of

individual vulnerability to differentiate stable cognitively normal controls (N = 184) from unstable cognitively normal controls who later developed MCI or dementia (N = 128) in the ADNI and

OASIS-3 datasets. METHODS: We created two regional effect size templates (REST) using Cohen’s d effect sizes for group differences in 68 FreeSurfer-derived regional cortical thickness and

surface area measures, and 7 subcortical volumes between clinically determined controls (n = 836) and dementia cases (n = 452), and separately between amyloid- controls (n = 168) and

amyloid+ dementia cases (n = 403) in ADNI and OASIS-3. Two regional vulnerability index (RVI-Dx, RVI-DxAB) scores were then calculated as the total correlation between the 75 regional Z

scores and REST in a test set of 312 stable and unstable controls from both datasets. RESULTS: Random effects linear regression showed that unstable converters had significantly higher

RVI-Dx (p = 0.018) and higher RVI-DxAB (p = 1.6e-10) than non-converters, adjusting for age, sex, ICV, and site as a random effect, indicating greater vulnerability to dementia. Both RVI

measures predicted clinical conversion with 79% AUC using logistic regression, but RVI-DxAB better predicted amyloid positivity (78.1% AUC) compared to RVI-Dx (65.9% AUC). CONCLUSIONS:

Collectively these results support the use of data-driven methods for quantifying individual vulnerability to dementia, as well as amyloid positivity. DISCLOSURE: Nothing to disclose. 32.3

COMBINING BIG AND CLINICAL NEUROIMAGING DATA: ANALYTIC STRATEGIES FOR STUDYING CATEGORICAL AND DIMENSIONAL NEUROIMAGING MARKERS OF MENTAL ILLNESS VINCE CALHOUN GEORGIA STATE UNIVERSITY,

ATLANTA, GEORGIA, UNITED STATES BACKGROUND: The neuroimaging community has seen a number of large open data collection efforts which have vastly increased the amount of data available to the

community. However, the use of such data for studying mental illness is not trivial and comes with major challenges including 1) balancing computational resources and analytic flexibility,

2) harmonization/standardization across studies, 3) integrating categorical versus dimensional approaches to (multimodal) data. In this session I will present specific examples of strategies

to address each of these points, with a focus on leveraging big data without ‘losing the forest for the trees.’ METHODS: We apply three computationally efficient data-driven approaches

(spatially constrained independent component analysis (ICA), constrained joint ICA, and a pre-trained deep neural network) to capture information about resting fMRI, structural MRI, and

polygenic risk score (PRS) data in the large UK Biobank and ABCD datasets. We evaluate results relative to several transdiagnostic multisite clinical datasets including schizophrenia, major

depression, autism spectrum, schizoaffective disorder, and bipolar disorder. RESULTS: PRS associated neuroimaging patterns derived from typical data were associated with robustly decreased

function and volume in frontotemporal, hippocampus complex, thalamus and insula. These patterns predict schizophrenia from controls with high accuracy, and predict cognition and symptoms for

schizophrenia across four independent cohorts. Functional network connectivity stability was highly predictive of the individual and predictive of cognitive and psychiatric scores. The

similarity of the functional patterns to clinical/control references showed a strong negative skew towards the clinical reference which was also significantly associated with prodromal

scores. Finally, the deep fusion results were highly predictive diagnosis, outperforming unimodal prediction. CONCLUSIONS: Strategies for working with large heterogeneous multimodal

neuroimaging data are numerous and fraught with trade-offs. Here we show several ways to efficiently work with large typically developing or aging datasets in combination with clinical data.

Results highlighted potentially important relationships between neuroimaging genomics data and the mental illness spectrum. DISCLOSURE: Nothing to disclose. 32.4 DOES MULTIVARIATE MODELLING

IN PEDIATRIC BIG DATASETS YIELD REPRODUCIBLE AND REPLICABLE RELATIONSHIPS BETWEEN CORTICAL THICKNESS AND PSYCHOPATHOLOGY? HAJER NAKUA UNIVERSITY OF TORONTO, MISSISSAUGA, CANADA BACKGROUND:

Identifying latent variables (LVs; linked dimensions between two data matrices) can delineate brain-psychopathology relationships across the general population. These relationships can help

describe potential risk or resilience factors of mental health trajectories across adolescent development. Canonical correlation analysis (CCA) and partial least squares (PLS) are two

approaches that identify LVs based on maximizing the correlation or covariance between them, respectively. Here, we compared the identified LVs as detected by CCA and PLS, and the

reproducibility of each LV, to assess whether one approach produces more robust relationships compared to the other. METHODS: T1-weighted imaging and psychopathology data were accessed from

the Adolescent Brain Cognitive Development dataset (n = 8643; ages=9-11). The brain matrix consisted of cortical thickness estimates from the Desikan-Killiany Atlas. The psychopathology

matrix consisted of 11 subscale scores from the parent-reported Child Behavioral Checklist. CCA and PLS models were separately applied to both matrices. Statistical significance of each LV

was assessed using permutation testing. Reproducibility of each LV was assessed using split-half resampling. RESULTS: The first LV from CCA and PLS models revealed distinct linked dimensions

between cortical thickness and psychopathology domains (singular values:CCA = 0.12,PLS = 0.35, p < 0.001) with CCA identifying spatially distributed positive and negative relationships.

Split-half resampling did not reveal reproducible relationships. Post-hoc analysis linking cognitive task performance measures to cortical thickness revealed that summary scores from the NIH

toolbox had a reproducible first LV for CCA and PLS models (singular values:CCA = 0.2,PLS = 0.47,p < 0.001). CONCLUSIONS: CCA and PLS did not yield reproducible relationships between

cortical thickness and psychopathology. The results of the post-hoc analysis may suggest that population-level variation of cortical morphology can be better delineated by variation in

cognitive performance instead of psychopathology. However, delineating reproducible brain-psychopathology relationships may be optimized by exploring clinically relevant subtypes within

large-scale general populations instead of examining relationships across the sample. DISCLOSURE: Nothing to disclose. PANEL 33. NEUROCIRCUITRY OF SOCIAL-EMOTIONAL BEHAVIOR 33.1 BEHAVIORAL

AND DOPAMINERGIC SIGNATURES OF RESILIENCE ANNEGRET FALKNER PRINCETON NEUROSCIENCE INSTITUTE, PRINCETON, NEW JERSEY, UNITED STATES BACKGROUND: Chronic stress can have lasting adverse

consequences in some individuals, yet others are resilient to the same stressor. While previous work has found differences in the intrinsic properties of mesolimbic dopamine (DA) neurons in

susceptible and resilient individuals after stress was over, the causal links between DA activity during stress, dynamic stress-evoked behavior, and individual differences in susceptibility

and resilience are not known. METHODS: We subject mice to a 10-day chronic defeat stress and then perform post-hoc tests of affective state. We record high speed video during stress and

perform supervised and unsupervised behavior analysis to detect behavioral signatures of resilience (~14m video frames, n = 32 males, n = 8 females). In addition, using fiber photometry, we

record the activity of the VTA-NAc and the SN-TS projecting populations across defeat and during post-hoc testing (n = 19 males, n = 8 females). Lastly, we develop a novel system to perform

fast behavior-triggered optogenetic perturbation. RESULTS: We find that resilient (R) and susceptible (S) individuals employ different behavior strategies during defeat. In males, resilience

was associated with fight-back. Using unsupervised analysis individuals could be classified into R or S groups. Female behavior during defeat revealed a single cluster that was

significantly correlated with R. In addition, we find that NAc-DA activity (but not TS-DA activity) predicts R in proximity to the aggressor during defeat in both sexes. This correlation

persists during the post-hoc test of social avoidance. We find that both closed-loop triggered stimulation timed to an R-associated behavior and open-loop stimulation during defeat similarly

bias individuals towards resilience. Both of these regimes also similarly reorganized behavior towards R-associated patterns of behavior during defeat. CONCLUSIONS: Overall we find that

resilient and susceptible individuals are differentiated by their behavior strategies during defeat and by their dopaminergic correlates. Activation of NAc-DA during defeat timed to an

R-associated behavior or untimed bias individuals towards resilience and reorganize behaviors towards patterns of resilient behavior. This provides evidence for a causal link between DA

during defeat and resilience. DISCLOSURE: Nothing to disclose. 33.2 OPTOGENETIC STIMULATION OF ESR1-EXPRESSING NEURONS IN PVT REVEALS A PUTATIVE NEURAL SUBSTRATE FOR OUTGROUP, BUT NOT

INGROUP, AGGRESSION BRANDY BRIONES UNIVERSITY OF WASHINGTON, SEATTLE, WASHINGTON, UNITED STATES BACKGROUND: The Paraventricular Thalamus (PVT) is a critical node for integrating sensory

features and emotional state information within novel and aversive contexts. However, few studies have investigated integration of social information in the PVT in these contexts. Navigating

novel social environments is complex, multisensory, and cognitively demanding, where mammals often rely on social heuristics to guide behavior. This can lead to increased perceived threat

and aggression towards an unfamiliar outgroup member, a phenomenon known as ingroup bias. While sex hormone receptors have been shown to modulate appetitive aggressive behavior, the neural

circuits, and mechanisms critical for ingroup versus outgroup social behavior remain to be determined. METHODS: Using a multiplexed hybridization chain reaction assay, we characterized sex

hormone receptor-related genes to identify distinct cell clusters in the PVT of male (n = 3) and female (n = 4) mice. We virally targeted sex hormone receptor-expressing neurons in the PVT

of Esr1-Cre + /- mice to investigate the role they play in defensive and social behaviors. After testing a variety of behavior assays with optogenetics (n = 8), we uncovered a male-specific

aggression phenotype and focused the remainder of our study on males (n = 9-10). RESULTS: We analyzed ~10,000 cells across the anterior-posterior axis of the PVT. Estrogen and androgen

receptor-related genes Esr1, Esrra, and Rora, were highly co-expressed (>50%) in central and posterior PVT neurons, corroborated by expression patterns of viral-labeled neurons in

Esr1-Cre + /- mice. Optogenetic activation of Esr1+ PVT neurons elicited aggressive attacks time-locked to light stimulation towards outgroup (indicated by differences in coat color), but

not ingroup, mice (p < 0.05), providing us with a putative neural circuit to dissect outgroup aggression further. CONCLUSIONS: From the data we have collected thus far, our results

suggest that activation of central and posterior PVT neurons expressing estrogen and androgen receptor-related genes gate intergroup aggression. Future directions of our study will aim to

investigate 1) in vivo Esr1+ PVT neural activity during freely moving social behavior in the presence or absence of circulating hormones and 2) whether estrogen and/or androgen receptors in

the PVT are necessary for outgroup aggression. DISCLOSURE: Nothing to disclose. 33.3 IDENTIFICATION OF TOUCH NEURONS UNDERLYING DOPAMINERGIC PLEASURABLE TOUCH AND SEXUAL RECEPTIVITY ISHMAIL

ABDUS-SABOOR COLUMBIA UNIVERSITY, NEW YORK, NEW YORK, UNITED STATES BACKGROUND: The pleasure of a partner’s caress or a child’s embrace begins with mechanical signals transduced by neurons

in our skin. Despite the centrality of socially rewarding touch in our daily lives, the neurons in the skin that detect social touch and shape the valence of perception generated in the

brain, remain unknown. This gap in knowledge is critical, especially when considering the nature of neurodevelopmental disorders like autism spectrum disorder, where gentle touch and

socially rewarding behaviors are aversive. METHODS: Transdermal optogenetic stimulation of sensory neurons was performed on 8-16 week-old male or female mice that were habituated to mesh

platform under a plastic chamber for 1 hour on each of two days prior to behavioral testing. Experimenter was present with lights, camera, and laser running during habituation to mimic

entire sensory experience of test day. RESULTS: A population of sensory neurons labeled by the G-protein coupled receptor Mrgprb4 detect stroking touch in mice, however, these neurons have

never been implicated in any natural social behaviors. Here, we study the social relevance of Mrgprb4-lineage neurons by genetically engineering mice to allow activation or ablation of this

population and reveal that these neurons are required for sexual receptivity and sufficient to induce dopamine release in the brain. Even in social isolation, optogenetic stimulation of

Mrgprb4-lineage neurons through the back skin is sufficient to induce a conditioned place preference and a striking dorsiflexion resembling the lordotic copulatory posture in females. In the

absence of Mrgprb4-lineage neurons, female mice no longer find male mounts rewarding sexual receptivity is supplanted by aggression and a coincident decline in dopaminergic release in the

mesolimbic reward pathway. CONCLUSIONS: In summary, the work described here has revealed the sufficiency of peripheral inputs to regulate social reward independent of context and other

sensory cues. We believe this study draws new attention to the importance of elucidating skin-brain circuits, analogous to the importance of gut-brain circuits. Moreover, this work points

towards the therapeutic potential of peripheral manipulations for enhancing intact or impaired social reward systems, including sexual receptivity, or simulating social reward during periods

of isolation. DISCLOSURE: Nothing to disclose. 33.4 GENETIC VARIATION IN OXYTOCIN RECEPTOR (OXTR) ALTERS EXPRESSION OF OXTR AND AUTISM SPECTRUM DISORDER (ASD) RISK GENES IN REWARD CIRCUITRY

IN PRAIRIE VOLES LARRY YOUNG EMORY UNIVERSITY SCHOOL OF MEDICINE, ATLANTA, GEORGIA, UNITED STATES BACKGROUND: Variation in oxytocin receptor (Oxtr) expression increases diversity in social

behavior within and across species. SNPs in human OXTR are associated with autism spectrum disorder (ASD) phenotypes and functional connectivity of reward circuitry in ASD. In monogamous

prairie voles, variation in Oxtr expression in nucleus accumbens (NAc) is associated with variation in pair bonding, alloparentingr and resilience to neonatal social neglect. A set of

non-coding Oxtr SNPs largely explain individual variation in Oxtr expression (r2 > 0.7) in NAc. Here we explore links between Oxtr SNPs, Oxtr expression, and transcriptional profiles in

NAc. METHODS: To investigate links between Oxtr SNPs, altered NAc Oxtr expression, and transcriptomic profile, we used ATAC-seq, bulk and single nucleus (sn)RNA-seq (N = 6-8/group, both

sexes). The combination of these techniques allowed us to compare differences in chromatin accessibility and transcriptomic profile between genotypes (low- and high-NAc Oxtr expression) and

across brain regions that show variable (NAc), stable (insular cortex, INS) or no (superior colliculus, SC) Oxtr expression. RESULTS: ATAC-seq revealed that SNP in the Oxtr intron are more

likely to be near areas of open chromatin than 5’ flanking SNPs and are better candidates for directly affecting Oxtr expression. RNA-seq showed differential expression between low- and

high-NAc Oxtr expressing genotypes, in both bulk RNA and snRNA-seq data, with Oxtr as a top hit (q < 1x10-7). The differentially expressed gene sets are highly enriched for ASD risk genes

(SFARI 2.0 database) in both bulk and snRNA-seq data sets (p < 0.0001). Most genes that are differentially expressed between genotypes localize to glia, which show little if any Oxtr

expression. CONCLUSIONS: SNP-dependent gene expression may be the consequence of direct effects of the SNPs on trans gene expression, of individual variation in OXTR signaling in the NAc, or

a combination of both. We are using CRISPR/Cas9 to differentiate between these possibilities. Our results suggest that variation in OXTR signaling in the NAc alters circuit

wide-transcriptional profiles, enriched for ASD risk genes, in diverse cell types. This study will be discussed in the context of our rsfMRI studies in ASD subject showing that intranasal

oxytocin normalizes reward circuit connectivity in ASD. DISCLOSURE: Levo Therapeutics, Inc: Consultant (Self) PANEL 34. SUBSTANCE USE DISORDER TREATMENT WITH PSYCHEDELIC DRUGS 34.1

BREAKTHROUGH THE BARRIERS: PSYCHEDELICS TO CONTROL THE CYCLE OF DYSFUNCTION INDUCED BY OPIOIDS AND COCAINE NOELLE ANASTASIO UNIVERSITY OF TEXAS MEDICAL BRANCH, GALVESTON, TEXAS, UNITED

STATES BACKGROUND: A record >107,000 fatalities due to drug overdoses was reported in the U.S. for the twelve months preceding April 2022. Opioids account for most deaths, but deaths

involving cocaine also steadily increased. Despite availability of medications for opioid use disorder, this polysubstance epidemic has crystalized the need to identify novel therapeutics to

improve the odds of successful recovery. Recently, there has been a resurgence of interest in serotonergic psychedelics for the treatment of psychiatric disorders. While a clinical trial of

the psychedelic psilocybin for cocaine use disorder is ongoing (NCT02037126), the nature and directionality of serotonergic dynamics and adaptations in the 5-HT2A receptor (5-HT2AR) system

engaged by opioids and cocaine are elusive. We tested the hypothesis R-(-)2,5-dimethoxy-4-iodoamphetamine [(-) DOI] will disrupt opioid and cocaine-evoked effects in preclinical rodent

models. METHODS: Two cohorts of Sprague-Dawley rats (Cohort 1, n = 13; Cohort 2, n = 16) were utilized. Cohort 1 was subjected to the drug discrimination paradigm to model the interoceptive

effects of the opioid oxycodone to establish ligand potency and effectiveness in vivo. Cohort 2 was trained to self-administer cocaine (0.25 mg/kg/inf) on an FR1-FR5 schedule of

reinforcement to model human cocaine-taking and drug reinforcement. Selectivity was ascertained in the presence of the 5-HT2AR antagonist M100907. RESULTS: Substitution tests indicated that

all (-)-DOI doses (0-0.2 mg/kg) resulted in percent oxycodone-lever responding significantly different from the training dose of oxycodone (p < 0.05). (-)-DOI (0.1, 0.2 mg/kg) reduced the

percent oxycodone-lever responding vs vehicle (p < 0.05), which was blocked by M100907 (p < 0.05). (-)-DOI reduced cocaine infusions (p < 0.05) at all doses tested (0-0.3 mg/kg)

without altering the number of inactive lever presses (n.s.). M100907 prevented the (-)-DOI-induced suppression of intake (p < 0.05). CONCLUSIONS: The psychedelic (-)-DOI via actions at

the 5-HT2AR does not substitute for oxycodone, suppresses the stimulus effects of oxycodone as well as the reinforcing effects of cocaine. Taken together, these studies provide support for

expanded interrogation of the efficacy of 5-HT2AR agonists in counteracting the behavioral effects of opioids and cocaine. DISCLOSURE: Nothing to disclose. 34.2 PSILOCYBIN IN THE TREATMENT

OF COCAINE USE DISORDER PETER HENDRICKS UNIVERSITY OF ALABAMA, BIRMINGHAM, ALABAMA, UNITED STATES BACKGROUND: Cocaine use disorder (CUD) affects millions of people in North America as is

associated with significant medical and psychiatric comorbidity. However, there are no approved pharmacotherapies for CUD. The 5HT2A receptor agonist psilocybin shows evidence of

transdiagnostic efficacy for a range of mental health conditions and may represent a promising treatment for CUD. METHODS: 40 participants with CUD received 4 sessions of “preparatory”

psychotherapy consisting of client-centered counseling and cognitive-behavioral therapy for CUD before being randomized to receive either psilocybin (25 mg/70 kg) or diphenhydramine (100

mg). Following drug administration, participants received 4 sessions of “integration” psychotherapy focused on translating insights from their drug experience to adaptive changes in

cognition and behavior. Assessments occurred 90 and 180 days after end-of-treatment. RESULTS: Whereas the two conditions reported no differences in percentage of cocaine use days prior to

drug administration, following drug administration those in the psilocybin condition reported a significantly greater percentage of abstinent days than those in the diphenhydramine

condition. CONCLUSIONS: Psilocybin shows preliminary signals of efficacy in the treatment of CUD. Future study with larger samples is needed to further interrogate psilocybin in the

treatment of CUD, with attention to putative mechanisms of action. DISCLOSURE: Bright Minds Biosciences, Eleusis Benefit Corporation, Reset Pharmaceuticals: Advisory Board (Self) 34.3

PSILOCYBIN IN THE TREATMENT OF TOBACCO USE DISORDER MATTHEW JOHNSON JOHNS HOPKINS UNIVERSITY SCHOOL OF MEDICINE, BALTIMORE, MARYLAND, UNITED STATES BACKGROUND: Early research from the 1950s

to 1970s investigated classic psychedelics in the treatment of substance use disorders. Over the last 20 years research has once again investigated psychedelics as potential medications.

METHODS: The presenter is conducting randomized, open-label, comparative efficacy study randomizing 100 treatment-seeking cigarette smokers to a single psilocybin session or a course of

nicotine patch treatment (n = 50 per group), both combined with cognitive behavior therapy. Both sexes were included. Point-prevalence abstinence including at 6- and 12-month follow-up

visits was biologically verified with urinary cotinine and breath carbon monoxide samples. These interim data were analyzed via chi-square (2-tail, alpha = .05). RESULTS: Current results (n 

= 61) show better results for psilocybin (n = 31), with a 12-month biologically confirmed 7-day point-prevalence abstinence rate of 52%, compared to 27% for nicotine patch. CONCLUSIONS:

Current results are promising and if confirmed would suggest a potential role for psilocybin treatment in tobacco smoking cessation. DISCLOSURE: AJNA Labs LLC, AWAKN Life Sciences Inc,

Beckley Psychedelic Ltd, Entheogen Biomedical Corp, Mind Medicine Inc, Silo Pharma: Advisory Board (Self), Field Trip Psychedelics Inc., Otsuka Pharmaceutical Development and

Commercialization Inc: Consultant (Self) PANEL 35. CLINICAL RESEARCH ON NOVEL PHARMACOLOGIC AND IMMUNOLOGIC APPROACHES TO TREATING CANNABIS USE DISORDER AND OPIOID USE DISORDER 35.1 EFFECT

OF AEF0117, A SIGNALING-SPECIFIC INHIBITOR OF THE CB1 RECEPTOR, ON THE REINFORCING AND SUBJECTIVE EFFECTS OF SMOKED CANNABIS MARGARET HANEY COLUMBIA UNIVERSITY, NEW YORK, NEW YORK, UNITED

STATES BACKGROUND: Cannabis use disorder (CUD) is an escalating problem, and there is no FDA-approved medication to facilitate its treatment. AEF0117, a signaling-specific inhibitor of the

cannabinoid type 1 (CB1) receptor (Aelis Farma), may show promise as a potential pharmacotherapy. AEF0117 selectively inhibits THC activation of one intracellular signaling pathway (MAPK)

without affecting CB1 agonist binding or THC’s other signaling cascades. In healthy volunteers, AEF0117 was safe and well-tolerated. Here we describe a Phase 2a placebo-controlled, human

laboratory study testing the effects of AEF0117 on cannabis’s positive subjective effects and its self-administration. Evidence of precipitated withdrawal was also evaluated by measuring

sleep, food intake and mood. METHODS: Nontreatment-seeking cannabis users completed two 5-day inpatient periods separated by a ≥ 14-day outpatient washout. Two doses of AEF0117 (0.06 and 1

mg/day) were tested in two distinct cohorts (n = 13/cohort). Participants received AEF0117 or placebo once/day in counterbalanced order for 5 days. Capsules were administered at 09h00 and

participants smoked a controlled amount of cannabis 3.5 hours later (12h30). Ratings of subjective effects were assessed over 2 hours. Participants had four opportunities (from 14h30 to

20h30) to self-administer cannabis. RESULTS: Daily cannabis smokers (1F,25M), averaging 3.3 + 1.7 grams/day, completed the study. In analysis of the crossover design, AEF0117 (1 mg/day)

significantly reduced both the positive subjective effects of cannabis and its self-administration (p < 0.05), while the 0.06 mg dose did not. However, a significant sequence by treatment

interaction (p < 0.01) showed that AEF0117 had effects after 14 days of washout. A parallel-group analysis looking at the first treatment period only confirmed a significant (p < .01)

reduction in the subjective and reinforcing effects of cannabis by AEF0117. AEF0117 was well tolerated, with no evidence of precipitated withdrawal. CONCLUSIONS: AEF0117 safely and robustly

attenuated the positive subjective and reinforcing effects of cannabis in participants with CUD. Favorable pharmacokinetic and safety profiles combined with the human laboratory data

support an upcoming Phase 2b, placebo-controlled multi-site clinical trial testing AEF0117 in patients seeking treatment for CUD. DISCLOSURE: Pleo Pharma: Advisory Board (Self), Pleo Pharma:

Stock / Equity (Self) 35.2 MULTISITE RANDOMIZED, DOUBLE-BLIND, PLACEBO-CONTROLLED, PARALLEL GROUP STUDY OF THE EFFICACY, SAFETY AND TOLERABILITY OF THE FATTY ACID AMIDE HYDROLASE (FAAH)

INHIBITOR JZP150 (PF-04457845) IN ADULTS WITH CANNABIS USE DISORDER (CUD) DEEPAK D’SOUZA YALE UNIVERSITY SCHOOL OF MEDICINE, WEST HAVEN, CONNECTICUT, UNITED STATES BACKGROUND: With the

liberalization of cannabis laws, the rates of and demand for the treatment of cannabis use disorder (CUD) are anticipated to increase. However, there are no FDA-approved or clinically

accepted treatments for CUD. One approach to treating CUD is by potentiating endocannabinoid (eCB) signaling. Fatty acid amide hydrolase (FAAH) inhibitors which increase levels of

anandamide, a principal eCB, reduce cannabis withdrawal in THC-dependent animals. Compared to cannabis and THC, FAAH-inhibitors do not have psychoactive or rewarding effects, are not

associated with tolerance and do not increase the abuse liability of other addictive drugs. PF-04457845 aka JZP-150 is an orally active, long-acting, potent and selective FAAH inhibitor. In

a completed proof of concept (POC), single-site, double-blind, randomized, placebo-controlled study (n = 60), JZP150 reduced CWS, reduced cannabis consumption, and attenuated stage 3 sleep

deficits in CUD participants. METHODS: Building on the positive results of the completed POC study, the efficacy, safety and tolerability of JZP150 in reducing cannabis use was evaluated in

a 4-site randomized, double-blind, placebo-controlled, parallel-group, outpatient clinical trial comparing JZP150 (4mg) and placebo in individuals with DSM-5 CUD. Participants received

motivational interviewing for 2 weeks prior to randomization and had to attempt to quit cannabis within the first week of treatment. Participants received 4 mg of JZP150 or placebo PO QD for

8 weeks during which time they were evaluated weekly for efficacy, safety and tolerability with measures of cannabinoid exposure (self-report and toxicology), problems related to cannabis

use, sleep, mood were assessed. In addition, daily video assessments of adherence to study medication, cannabis use and safety were conducted by cellphone. RESULTS: The last subject was

randomized on 5/5/2022 meeting the target of 234 subjects (85 female) with 171 completers thus far. The study has been safe and well-tolerated with a low dropout rate (10%). All data

collection will be completed by July 2022 with data analysis planned for August 2022. This is the largest clinical trial to date testing the efficacy, safety and tolerability of a FAAH

Inhibitor for CUD. CONCLUSIONS: JZP150 is safe and well-tolerated. The efficacy results of the study will be discussed. DISCLOSURE: Biohaven: Consultant (Self), Jazz Pharmaceuticals:

Advisory Board (Self), Psilocybin Migraine: Patent (Self) 35.3 SAFETY, IMMUNOGENICITY, AND PRELIMINARY EFFICACY OF OXY-SKLH, A VACCINE THAT TARGETS OXYCODONE, AND EXPLORATORY BIOMARKERS OF

OPIOID USE DISORDER AND VACCINE EFFICACY IN HUMANS RACHEL LUBA COLUMBIA UNIVERSITY AND NYSPI, NEW YORK, NEW YORK, UNITED STATES BACKGROUND: Illicit opioid use contributed to the majority of

drug-related overdose deaths in 2021. This presentation will highlight new data from a Phase 1a/1b study of OXY-sKLH, an opioid vaccine that targets oxycodone (OXY). We also will describe an

independent clinical study to identify biomarkers that can be used to select participants who may be “good responders” to the vaccine. METHODS: Unvaccinated individuals with opioid use

disorder (OUD; n = 23; 21M, 2F) and non-drug-using volunteers (n = 22; 19M, 3F) were enrolled in our biomarker study. Blood samples were collected to analyze opioid-specific B cell

lymphocyte population subsets by antigen-based magnetic enrichment paired with multiparameter flow cytometry, cell sorting paired with sequencing of the B cell receptor, and evaluation of

cytokine panels and opioid-specific serum IgG antibodies. The Phase 1a/1b study of OXY-sKLH has enrolled 11 individuals with OUD (10M, 1F). OXY was administered pre- and post-vaccination to

examine the efficacy of OXY-sKLH to reduce subjective ratings and physiological responses. The pharmacokinetic (PK) profile of oxycodone also was assessed. RESULTS: The frequency of

opioid-specific B cell population subsets was higher in subjects with OUD compared to controls (p < 0.01). The frequency of B cells specific for heroin/morphine (M) were more prevalent

than the OXY-specific B cell population. No difference in OXY- and M-specific serum IgG antibodies have been found between groups. Decreases in selected cytokines were observed in those with

OUD compared to controls (p < 0.05) and significant correlations were observed between several cytokines and heroin use in those with OUD (p < 0.05). Similar immunoprofiling will be

performed in participants in the Phase 1a/1b study. The OXY-sKLH vaccine appears to be safe with only minimal adverse effects. Oxy-specific IgG titers and concentrations are elevated in a

subset of participants for > 40 weeks; the PK and preliminary efficacy data will be presented at the conference. CONCLUSIONS: Selected immune-related biomarkers, such as opioid-specific B

cells and cytokines, may be predictive of OUD. These biomarkers also may be predictive of vaccine efficacy, which would be consistent with our data collected in rodents. Thus far, the

vaccine has proven to be safe, and some participants are exhibiting long-lasting antibody levels. DISCLOSURE: BioXcel Therapeutics, Janssen: Contracted Research (Self), Go Medical,

Intra-cellular Therapies, Lyndra: Grant (Self), Alkermes: Other Financial or Material Support (Self), Clinilabs, Mallinckrodt, Nektar, Otsuka: Consultant (Self), Opiant: Advisory Board

(Self) 35.4 MECHANISM OF ACTION OF ITI-333, A NOVEL MODULATOR OF MU OPIOID, SEROTONIN, AND DOPAMINE AND FOR THE TREATMENT OF OPIOID USE DISORDER SANDRA COMER COLUMBIA UNIVERSITY AND NYSPI,

NEW YORK, NEW YORK, UNITED STATES BACKGROUND: New medications are needed to treat opioid use disorder (OUD) by easing somatic symptoms of drug withdrawal and mitigating dysphoria and

psychiatric comorbidities that drive opioid use/abuse. We describe properties of ITI-333, a novel molecule combining high affinity binding (Ki < 50 nM) to receptors that individually

participate in substance use disorders and psychiatric comorbidities: mu opioid (MOP), serotonin 5-HT2A, and dopamine D1 receptors. METHODS: ITI-333 was tested in cell-based assays on

recombinant receptors and in male mice and rats for analgesia, and somatic and behavioral signs of opioid withdrawal, heroin reinstatement, tolerance/physical dependence/self-administration,

and safety. Data were analyzed by ANOVA. RESULTS: Cell-based assays demonstrate ITI-333 is a biased MOP receptor partial agonist, acting as an antagonist to block effects of high doses of

morphine in pain and motor activity models; acting alone as an analgesic in acute and inflammatory pain models. ITI-333 induces cAMP accumulation (agonism) at human recombinant MOP receptors

while blocking effects of a full MOP agonist, DAMGO (antagonism). Further, ITI-333 is a biased agonist at MOP receptors, acting as an antagonist on beta-arrestin pathways mediating opioid

side effects. ITI-333, (0.01-1mg/kg, SC, n = 10/group), produces naloxone-sensitive analgesia in tail flick (TF) assay, while attenuating morphine analgesia in TF (0.1-1mg/kg, SC, n = 

10/group). ITI-333 blocks morphine hyperactivity (32mg/kg, SC, n = 6/group) without affecting spontaneous activity. ITI-333 (0.3, 1, and 3mg/kg, SC) dose-dependently suppresses signs of

opioid withdrawal precipitated by naloxone injection in oxycodone-dependent mice (i.e., oxycodone dosed for 8 days; increasing daily doses of 9-33 mg/kg, n = 10/group). Chronic ITI-333 (0.3

or 3mg/kg, SC) does not induce tolerance or physical dependence in rats; acute doses of ITI-333 (0.3 or 3mg/kg, SC, n = 8/group) do not induce GI or pulmonary side effects. ITI-333

(0.003-0.01mg/kg, IV, 8/group) is not self-administered by heroin-maintained rats. ITI-333 was safe and well-tolerated in a first-in-human safety study and continues in clinical development

as a treatment for OUD. CONCLUSIONS: Data support a role for ITI-333 in mitigating opioid withdrawal symptoms with no significant indications of abuse liability. DISCLOSURE: Intra-Cellular

Therapies Inc: Employee (Self), Intra-Cellular Therapies Inc: Stock / Equity (Self) PANEL 36. MEMORY MECHANISMS RELEVANT TO THE DEVELOPMENT AND MAINTENANCE OF POSTTRAUMATIC STRESS DISORDER

36.1 SEX-DEPENDENT ROLE FOR PREFRONTAL CORTICAL INHIBITORY INTERNEURONS IN THE REGULATION OF STRESS-INDUCED IMPAIRMENTS IN FEAR EXTINCTION IN RATS STEPHEN MAREN TEXAS A AND M UNIVERSITY,

COLLEGE STATION, TEXAS, UNITED STATES BACKGROUND: Hyperarousal is a core symptom in patients with post-traumatic stress disorder (PTSD). Elevated stress underlies deficits in

extinction-based behavioral therapies in PTSD patients. Considerable work indicates that the infralimbic (IL) division of the medial prefrontal cortex (mPFC) is a neural substrate for

stress-induced extinction impairments. Here we examine the possibility that stress-induced activation of IL inhibitory interneurons contributes to stress-induced extinction impairments in

male and female rats. METHODS: Stress-induced activation of mPFC interneurons was assessed by exposing male (n = 12) and female (n = 12) Long-Evans rats to footshock and quantifying

parvalbumin (PV) and Fos expression in the IL and prelimbic (PL) cortex. The contribution of PV interneurons to the stress-induced extinction impairment was assessed by chemogenetically

inhibiting these cells in male (n = 21) and female (n = 20) rats. We expressed an inhibitory DREADD in PV-Cre transgenic rats prior to fear conditioning and immediate extinction; extinction

retrieval was tested 48 hr later. To validate this strategy, a subset of male (n = 14) and female (n = 13) rats received the DREADD ligand CNO or VEH prior to footshock and Fos expression

was quantified in DREADD-expressing mPFC cells. The percentage of freezing was used as the behavioral measure of fear. RESULTS: Consistent with our hypothesis, footshock significantly

increased Fos expression in IL PV + cells in both sexes. Chemogenetic inhibition of mPFC PV + interneurons had no effect on conditioned freezing during either fear conditioning or immediate

extinction. Unexpectedly, this treatment led to a potentiation of freezing during the drug-free extinction retrieval test in male, but not female, rats. Chemogenetic inhibition of mPFC PV + 

interneurons significantly reduced footshock-induced Fos expression in those neurons, confirming the functional efficacy of the manipulation. CONCLUSIONS: These results reveal that mPFC

inhibitory interneurons play a sex-specific role in the regulation of extinction under stress. Tonic increases in IL excitability resulting from PV inhibition appear to compromise IL

function and impair extinction learning. DISCLOSURE: Nothing to disclose. 36.2 LATENT ASSOCIATIVE STRUCTURES FACILITATES TRANSFER OF LEARNED FEAR IN HUMANS JOSEPH DUNSMOOR UNIVERSITY OF

TEXAS AT AUSTIN, AUSTIN, TEXAS, UNITED STATES BACKGROUND: Animals often rely on pre-existing associations to facilitate transfer of new learning. For instance, once a stimulus becomes

predictive of an aversive outcome, pre-associated stimuli acquire the capacity to elicit a defensive response, known as sensory preconditioning. In rodents, this form of fear generalization

requires the hippocampus and orbitofrontal cortex. Mechanisms of sensory preconditioning in humans are unclear. Insights could offer clues on how fear broadly generalizes to a host of

stimuli following aversive events (e.g., trauma). Here, we used a novel sensory preconditioning task during fMRI to investigate latent associative networks involved in fear generalization.

METHODS: Participants (N = 27) learned that a category (preconditioned stimulus, PS; animals or tool, counterbalanced) predicted a picture of a square, while another PS category (tools or

animals, respectively) predicted a circle. Next, one shape (conditioned stimulus, CS + ) was paired with an aversive electrical shock while the other shape (CS-) was unpaired. Subjects then

viewed novel exemplars (PS + ) from the category originally paired with the CS + and exemplars (PS-) from the category originally paired with the CS-. Subjects returned 24 hours later for a

surprise memory test of PS + /PS- exemplars encoded before and after threat conditioning the previous day. RESULTS: Participants acquired conditioned learning to the CS + , expressed through

skin conductance responses and differential (CS + > CS-) activity in the thalamus, insula, and dorsal anterior cingulate cortex (all fMRI activity p < .001 corrected for multiple

comparisons at p < .05). During the fear generalization test, there was differential hippocampal and orbitofrontal activity to the PS + > PS- (p < .001, two-tailed t-test). Finally,

participants had a 24-hour memory bias for PS + > PS- items encoded prior to fear conditioning (p = .02, two-tailed t-test). CONCLUSIONS: Results show how latent associative structures

support transfer of learned fear. This extends evidence from rodents on the role of the hippocampus and OFC, and offers insights into key neurocircuity implicated in the higher order

transfer of emotional learning. Sensory preconditioning may be a valid but underappreciated model for the widespread and idiosyncratic nature of fear generalization following aversive

events, such as trauma. DISCLOSURE: Nothing to disclose. 36.3 THREAT-RELATED AROUSAL IMPACTS THE HIPPOCAMPAL REACTIVITY IN PTSD RISK AND RELATED MEMORY PHENOMENOLOGY VISHNU MURTY TEMPLE

UNIVERSITY, PHILADELPHIA, PENNSYLVANIA, UNITED STATES BACKGROUND: A growing body of research implicates threat-related arousal and hippocampus (HPC) deficits as major determinants of

developing post-traumatic stress disorder (PTSD). Research in this domain has often been siloed, treating hyper-arousal and HPC dysfunction independently. Recent work supports a model by

which threat-related arousal impairs HPC function, biasing memory away from relational processing. I will present two studies characterizing how threat-related arousal interacts with the HPC

to predict PTSD symptoms, and a follow-up study in a normative cohort to unpack how threat-related arousal influences the phenomenology of episodic memory for threatening events. METHODS:

In Study 1, fMRI markers of HPC threat reactivity, fear-potentiated startle (FPS), and PTSD symptoms 2 weeks following a traumatic event in an emergency room cohort of 117 individuals as

part of the AURORA study. Interactions between FPS and HPC threat reactivity were assessed to see how they predict PTSD symptoms. In study 2, threat-related arousal during an in-person

haunted house visit and 1-week free recall were collected in a cohort of 42 normative individuals. Interactions between threat-related arousal and free recall content and comprehensibility

were characterized. RESULTS: Study 1 showed that decreased HPC threat reactivity was associated with PTSD symptoms two weeks following trauma exposure (n = 117; p < 0.05,

Bonferroni-corrected). Further, the relationships between decreased threat reactivity in the HPC and PTSD symptoms was limited to individuals showing high FPS, a marker of hyper-arousal (p 

< 0.05, Bonferroni-corrected). This study did not evaluate threat’s influence on memory phenomenology. In study 2, we demonstrated that threat-related arousal during an in-person haunted

house biased memory recall towards sensory features of the environment rather than action or contextual features (n = 47; p < 0.01). Further, we found that the memories of individuals

that showed greater threat-related arousal were rated as less comprehensible and more confusing by a naive set of listeners (n = 200; p < 0.01). CONCLUSIONS: Together, our findings

characterize how threat influences HPC function, such that arousal biases information processing away from the HPC resulting in greater PTSD symptomology and the construction of more

fragmented narratives. DISCLOSURE: Nothing to disclose. 36.4 SUCCESSFUL MEMORY ENCODING IN TRAUMA-EXPOSED CHILDREN DEPENDS ON CORTICAL SENSORY REGIONS, RATHER THAN THE MEDIAL TEMPORAL LOBE

JENNIFER STEVENS EMORY UNIVERSITY SCHOOL OF MEDICINE, ATLANTA, GEORGIA, UNITED STATES BACKGROUND: Childhood trauma is a primary risk factor for intrusive memories associated with PTSD. The

medial temporal lobe (MTL) subserves memory formation and is known to be highly sensitive to damaging effects of stress, but little previous research has addressed MTL function in trauma or

stress-exposed children. METHODS: Mother-child pairs were recruited through the primary care waiting rooms of a large publicly funded urban hospital. Children ages 8-14 participated in an

fMRI task probing MTL-dependent memory encoding, viewing scene stimuli with negative, positive, or neutral affective valence. After a 30-minute delay, children indicated their memory for

those scenes in a cued recall task. MTL regions of interest included the amygdala (AMY) and hippocampus (HC). RESULTS: Children showed greater recall for negative and positive scenes,

relative to neutral (ps < .001), irrespective of trauma exposure (p > .05). MTL engagement during scene encoding was associated with the enhancing effect of emotion on memory, but the

association depended on trauma load (trauma * recall enhancement for negative scenes, AMY: F = 6.96, p = 0.02, HC: p = .001; positive scenes, AMY: p = 0.05, HC: p = 0.005). In children

reporting fewer traumas, recall was positively associated with MTL engagement, similar to healthy adults in prior work. In children reporting more traumas, recall was negatively associated

with MTL engagement. Exploratory analyses showed that higher trauma instead predicted engagement of the left middle temporal sulcus for negative>neutral scenes, and primary visual cortex

for positive>neutral scenes, pFWE < .05. CONCLUSIONS: MTL engagement in encoding predicted emotion-related memory enhancement only among children who experienced fewer traumas. Those

with higher trauma appeared to have preserved recall, potentially explained by greater engagement of alternative routes to memory encoding in visual and multimodal cortex. We posit that this

may represent a neurodevelopmental adaptation to a high-trauma environment, such that emotional information is routed toward more stable cortical storage that may influence the earliest

phases of sensory processing to help children quickly respond to salient environmental cues. Such adaptations may also decrease memory flexibility and context sensitivity, increasing later

risk for intrusive symptoms. DISCLOSURE: Nothing to disclose. MINI PANEL 37. EMERGING INTERSECTIONAL TECHNOLOGIES FOR PROBING MOLECULAR MECHANISMS OF NEUROTRANSMITTER CO-TRANSMISSION IN

BRAIN AND BEHAVIOR 37.1 DISSECTING THE ROLES OF DOPAMINE/GLUTAMATE CO-TRANSMISSION IN SEX DIFFERENCES UNDERLYING SELECTIVE DOPAMINE NEURON RESILIENCE ZACHARY FREYBERG UNIVERSITY OF

PITTSBURGH SCHOOL OF MEDICINE, PITTSBURGH, PENNSYLVANIA, UNITED STATES BACKGROUND: Parkinson’s disease (PD) is characterized by dopamine (DA) neuron loss but some DA neurons are more

protected than others. Clues for DA neuron resilience in PD come from a subpopulation of medial VTA DA neurons that co-transmit glutamate. These cells are spared in PD and express the

vesicular glutamate transporter 2 (VGLUT2). Additional insights come from sex differences in DA neuron loss as men are more affected, and at earlier ages of onset vs women. Yet, mechanisms

for these sex differences and implications for resilience remain unknown. We discovered resilient female DA neurons express more DA neuron VGLUT2 vs males and these sex differences are

conserved from flies to rodents to humans (Buck et al 2021). We found that DA neuron VGLUT2 enables DAergic vesicles to tune content during firing (Aguilar et al. 2017). Thus, VGLUT2 permits

adequate DA neurotransmission to be maintained in spite of DA neuron loss. Considering the key role of mitochondrial dysfunction in PD, we posit VGLUT2 improves mitochondrial resilience to

oxidative stress in DA neurons. METHODS: Intersectional imaging/INTRSECT2.0: VGLUT2-cre;TH-flpo mice were injected with Boolean-gated intersectional viruses to selectively label TH + /VGLUT2

 + vs TH + only neurons. Cleared brains were imaged via ribbon scanning confocal microscopy (RSCM). Intersectional DA Neuron VGLUT Assay: In flies, we used an intersectional DA neuron VGLUT

luciferase reporter to measure sex-specific changes in DA neuron VGLUT expression. Reactive Oxygen Species (ROS) Imaging: We tested DA neuron mitochondrial ROS via the MitoTimer biosensor in

both sexes, including after DA neuron VGLUT knockdown with RNAi. All studies were IACUC-approved and included 6+ males and females/group. RESULTS: We combined high-resolution RSCM imaging

of INTRSECT-labeled mouse brains with neural network analysis of projections. This identified a pattern of TH + /VGLUT2 + synapses distinct from TH-only neurons. To dissect sex-specific

mechanisms for resilience of TH + /VGLUT2 + neurons, a genetic screen in flies showed that mitochondrial metabolism genes are the primary hits for VGLUT2-mediated DA neuron survival in

response to insults by DA neurotoxin paraquat (PQ). In line with VGLUT-mediated sex differences, males show 40% more DA neuron mitochondrial ROS and lower VGLUT expression vs females (p = 

.005). In response to PQ, dVGLUT KD exacerbates mitochondrial ROS and DA neuron loss with males more than females (p < .0001). These data suggest sex differences in VGLUT2 expression

underlie DA neuron resilience to mitochondrial stress. CONCLUSIONS: DA/glutamate neurons constitute a subpopulation of DA neurons unique from purely DAergic cells. Among these DA/glutamate

co-transmitting neurons, levels of VGLUT2 expression mediate sex differences in DA neuron survival by regulating mitochondrial oxidative stress in response to insults. Thus, higher levels of

DA neuron VGLUT2 in females diminish overall ROS to boost resilience compared to males. Collectively, our data suggest a key mechanism for VGLUT-mediated resilience is its ability to reduce

mitochondrial ROS in DA neurons. DISCLOSURE: Nothing to disclose. 37.2 INTERSECTIONAL GENETICS TOOLS REVEAL CELL-TYPE SPECIFIC TRANSCRIPTIONAL NETWORKS OF CO-TRANSMITTING DOPAMINERGIC AND

GLUTAMATERGIC NEURONS RYAN LOGAN BOSTON UNIVERSITY SCHOOL OF MEDICINE, BOSTON, MASSACHUSETTS, UNITED STATES BACKGROUND: Accumulating evidence suggests that distinct subpopulations of neurons

are capable of co-releasing multiple neurotransmitters to regulate an array of fundamental physiological processes and behaviors. Co-transmitting neuronal subpopulations may be involved in

disease-related processes including substance use disorders and neurodegeneration. The advent of new genetic tools enables the investigation of co-transmitting neurons in health and disease

with unprecedented resolution. Through collaborative efforts, we have aided in the development and application of novel intersectional genetics approaches for studying multiple

subpopulations of neurons in the same brain of the mouse, called intronic recombinase sites enabling combinatorial targeting (INTRSECT 2.0). Our proposal will present new findings on the

transcriptional networks in specific subpopulations of cells in the mouse brain that co-transmit dopamine and glutamate and their relevance for drug reward and neuronal resilience. METHODS:

Male and female mice expressing both Th-Flp (Tyrosine Hydroxylase) and Vglut2-Cre (Vesicular Glutamate Transporter 2) were injected (n = 8 per genotype per sex) with virus cocktail that

included AAV8-nEF-CreON-FlpOFF-EYFP (Th + /Vglut2+ cells) and AAV-nEF-CreOFF-FlpON-mCherry (Th+ cells). Following 5-6 weeks of viral expression, mice were sacrificed, and reporter labeled

cells were sorted for EYFP, mCherry, and cell-type specific surface markers to parse neurons from microglia (i.e., Cd45/Cd11b). RNA was extracted from sorted cells, then processed for

RNA-sequencing, followed by a computational pipeline based on differential expression, pathway enrichment, and gene network module building. RESULTS: Transcriptomics analysis identified ~600

transcripts enriched in Th + /Vglut2+ cells in mouse midbrain (adjusted p < 0.01), with significant associations to presynaptic release and vesicle fusion processing includes

SNARE-dependent vesicle fusion (p < 0.05). Many of the sex-specific transcriptional networks were related to metabolism and mitochondrial activity. Gene targets specific to Th + /Vglut2+

cells were prioritized based on cross-species data integration of similar cell types in flies and humans, identifying 200 candidates (AW Fisher meta-analysis of overlap, p < 0.05). These

candidates were then screened using flies by the TH-GAL4 system knockouts for locomotor phenotypes related to psychostimulants, opioids, and toxin-induced neurodegeneration. Highly ranked

gene candidates from cross-species data integration revealed potential mechanisms in Th + /Vglut2+ neural cells related to drug reward and neurodegeneration, particularly opioid use disorder

and Parkinson’s Disease. CONCLUSIONS: Our findings use innovative new transcriptomic pipelines developed to employ in INTRSECT-labeled cell populations to specifically define

dopamine/glutamate co-transmitting neurons at the level of transcriptional networks. We identified major pathways related to mitochondrial function, reactive oxygen species, bioenergetics,

and presynaptic vesicle trafficking and release in TH + /VGLUT2 + neurons of the midbrain, along with the identification of potential upstream regulators and modulators of Vglut2 expression.

These candidates may be involved in various disease-related processes in the brain, including substance use disorders and neurodegeneration. DISCLOSURE: Nothing to disclose. 37.3 EFFECTS OF

AGING ON DOPAMINE-GLUTAMATE NEURONS PROJECTING TO THE LATERAL ENTORHINAL CORTEX SUSANA MINGOTE ADVANCED SCIENCE RESEARCH CENTER, NEW YORK, NEW YORK, UNITED STATES BACKGROUND: Dopamine (DA)

neurons in the ventral tegmental area (VTA) play a critical role in motivation, learning and memory. Is it well established that DA neurons in the VTA co-express the vesicular glutamate

transporter 2 (VGLUT2) and use glutamate (GLU) as a co-neurotransmitter. The GLU co-transmission adds greater dynamic signaling range and heterogeneity to the actions of DA neurons, although

its function is still not well defined. Fate-mapping experiments in mice have shown that almost all of DA neurons expressed VGLUT2 during early development. Using intersectional strategies

to selective label DA-GLU neurons, we found that the density of DA-GLU neurons is reduced to 30% in adulthood (Mingote et al., 2019). It is not known if the density of VTA DA-GLU neurons

continues to decrease with aging and its effects, mental function. Considering that DA-GLU neurons send dense projections to the lateral entorhinal cortex (LEC; Mingote et al., 2015), a key

region for spatial memory and particularly sensitive to age-related neurodegeneration, we investigate if the density of DA-GLU terminals in this region was altered in old-aged mice. METHODS:

We used INTRSECT 2.0 viruses (Fenno et al, 2021) to selectively labeled DA-GLU and DA-only neurons with different fluorescent proteins in 3-, 12- and 24-months old mice. The viruses were

injected into the VTA of TH-Flp/+::VGLUT2 cre/+ neurons, and animals were perfused one month after the stereotaxic surgeries for neuronal quantification. The brains were processed for either

immunohistochemistry or multiplex in situ hybridization (RNAscope). All studies were IACUC-approved. RESULTS: We found that the number of the DA-GLU neurons in the VTA decreased between 3-

and 24-months old mice, while the density of DA-only neurons significantly increased (p < 0.05). In addition, the number of VGLUT2 transcripts in the remaining DA-GLU neurons also

significantly decreased in 2-year-old mice (p < 0.05). We then focused of the effects of aging on the projections to the LEC. We found that the density of DA-GLU axons in the LEC

significantly decrease by over 60% in 24-months old mice. CONCLUSIONS: Our findings suggest that the DA neurons go through a process of neurotransmitter switching during aging, in which

DA-GLU neurons become DA-only neurons. In addition, the extensive decrease in the density of DA-GLU axons in the LEC, revealed regional-specific alterations that may contribute to changes in

cognitive skills in aging. DISCLOSURE: Nothing to disclose. STUDY GROUP 38. EARLY CAREER SCIENTIST WORKSHOP: TOOLS FOR BUILDING YOUR SUCCESSFUL CAREER ANGELA OZBURN*, SADE SPENCER, NII

ADDY, MARIA OQUENDO, DAMIEN FAIR, M. MERCEDES PEREZ-RODRIGUEZ OREGON HEALTH & SCIENCE UNIVERSITY, VA PORTLAND HEALTH CARE SYSTEM, PORTLAND, OREGON, UNITED STATES STUDY GROUP SUMMARY: Few

of the research skills learned in our training translate to an independent academic or industry career, where under-represented minority (URM) scientists often face discrimination and

additional systemic barriers to success. Despite a 9-fold increase in the number of URMs earning PhDs in the past 2 decades, there has been little progress in recruitment and retention of

URM faculty. Moreover, URMs generally receive their first R01 at a later age than do majority awardees, where applications with white PIs have higher probability of being discussed in study

section and a 2-fold advantage in funding success (compared with URM PIs). Further challenges exist due to a limited focus on cultivating critical thinking abilities (and other desirable

skills e.g., problem solving, project management, persuasive communication, etc.), however it is difficult to self-identify and articulate these proficiencies beyond the training

environment. This URM led study group will help fill gaps in the transfer of some of the skills most relevant to embarking upon and maintaining a successful career in science. A labyrinth

metaphor can be used to describe the journey of an early career researcher. On the one hand, this is a misleading metaphor because there is no single path to a successful scientific career.

On the other hand, like navigating a labyrinth, a successful career requires strategic planning, intuition self-assessment, leadership skills, and time. This study group brings together an

esteemed and diverse panel to share strategies they have developed during their journeys. Each panelist is an accomplished researcher in addition to being an effective mentor. Dr. M.

Mercedes Perez-Rodriquez, Associate Professor of Psychiatry (Icahn School of Medicine at Mt. Sinai), will introduce the concept of building a business plan to provide a road map for your own

scientific career. Dr. Nii Addy, Associate Professor of Psychiatry (Yale School of Medicine) and host of the engaging Addy Hour podcast, will discuss what it means and how to chart a path

towards succeeding as your authentic self. Dr. Maria Oquendo, Professor and Chair of Psychiatry (University of Pennsylvania Perelman School of Medicine), will share her journey and provide

insight on developing leadership skills for managing teams and projects at all professional levels. Dr. Damien Fair, Professor of Pediatrics and Co-Director of the Masonic Institute for the

Developing Brain (University of Minnesota), will introduce self-assessment tools and metrics that can be used to identify areas for further development. Upon the introduction of these tools,

Drs. Angela Ozburn (Associate Professor, Oregon Health, and Science University) and Sade Spencer (Assistant Professor, University of Minnesota) will facilitate an interactive discussion

between the panelists and the audience members to expand upon these concepts. In alignment with the commitment of ACNP to education and training, this study group provides a unique career

development opportunity to focus on skills commonly overlooked in traditional sessions. Deliverables from this study group will include tangible new tools, techniques, and strategies as well

as advice for cultivating existing skill sets that can be implemented by a diverse group of early career researchers. DISCLOSURE: Nothing to disclose. PANEL 39. COMPUTATIONAL PSYCHIATRY

INSIGHTS INTO COGNITION AND MOTIVATION IN SCHIZOPHRENIA, BIPOLAR DISORDER AND MAJOR DEPRESSION: PROGRESS FROM THE NEW CNTRACS CONSORTIUM 39.1 BEHAVIORAL MARKERS OF COGNITIVE IMPAIRMENT

ACROSS PARADIGMS IN PSYCHOPATHOLOGY MOLLY ERICKSON THE UNIVERSITY OF CHICAGO, CHICAGO, ILLINOIS, UNITED STATES BACKGROUND: People with serious forms of psychopathology such as schizophrenia

and bipolar disorder consistently exhibit impaired performance on a range of cognitive tasks. Using computational modeling, it is possible to estimate functioning in specific aspects of

cognition that are shared across these tasks, thus permitting a more nuanced conceptualization of cognitive impairment in these disorders. We predicted that people with mood and psychotic

disorders would exhibit core impairments in precision, sustained attention, and memory capacity that are shared across tasks and account for variance in IQ and functional outcome. METHODS:

People with schizophrenia or schizoaffective disorder (SZ; N = 64), bipolar disorder with psychotic features (BP; N = 42), major depressive disorder (MDD; N = 40), and a group of healthy

control participants (HC; N = 76) completed three tasks assessing episodic memory, working memory, and visual perceptual precision. Using computational modeling, precision and attention

lapse rate was estimated separately for each task. Memory capacity was also estimated from the working and episodic memory tasks. Principal components analysis was then used to calculate

three factor scores (precision, attention lapsing, and memory capacity). RESULTS: SZ participants exhibited trend-level reductions in precision, as well as significantly higher rates of

attention lapsing and reduced memory capacity. By contrast, MDD participants exhibited normal precision, sustained attention, and memory capacity, with BP participants showing intermediate

impairment. IQ was significantly correlated with the precision factor (r’s = 0.42 to 0.59; p’s < 0.001) and the memory capacity factor (r’s = 0.56 to 0.67; p’s < 0.001) for HC, SZ, and

BP. The attention lapsing factor was only correlated with IQ for HC and SZ (r’s = -0.32 to -0.43; p’s < 0.01). None of the factors were significantly correlated with functional outcome.

CONCLUSIONS: SZ exhibited reduced memory capacity, higher attention lapse rates, and reduced precision across all three tasks, with BP showing intermediate impairment and MDD performing

comparably to HC participants. Computational modeling revealed shared cognitive processes across tasks that exhibited significant correlations with IQ, but not functional outcome in the

patient groups. DISCLOSURE: Nothing to disclose. 39.2 CAN APPLYING COMPUTATIONAL MODELS TO BEHAVIORAL PERFORMANCE PROVIDE RELIABLE AND VALID INFORMATION ABOUT SPECIFIC DIMENSIONS OF

COGNITIVE FUNCTION? STEVEN LUCK UNIVERSITY OF CALIFORNIA, DAVIS, CALIFORNIA, UNITED STATES BACKGROUND: Fitting computational models to behavioral data from cognitive paradigms can allow us

to quantify specific aspects of cognitive function, but are the resulting variables sufficiently reliable for use in large-scale studies of psychiatric populations? To address this question,

we fit a computational model to data from a working memory task and an episodic memory task, allowing us to separately estimate different factors that contribute to task performance

(attention, storage capacity, and precision). METHODS: Data from the CNTRaCS project were available for 149-239 men and women (both neurotypical individuals and people with mental

illnesses). Participants completed a working memory task (memory loads 1 and 5) and an episodic memory task, both of which required memory for locations in a circular space. A computational

model was used to estimate lapses of attention, memory storage capacity, and memory precision. Noncomputational aggregate performance measures were also derived for each task. RESULTS: The

split-half reliability was found to be very good (r = .87–.94, p < .0001) for all computational parameters except the estimate of memory precision in the episodic memory task (r = .69, p 

< .0001). Reliability for most of the computational parameters was nearly as high as reliability for the noncomputational aggregate measures (r = .92–.95, p < .0001). In addition, the

pattern of correlations across computational parameters indicated that they measured different aspects of task performance. For example, working memory precision was highly correlated for

memory loads of 1 and 5 items (r = .73, p < .0001), but the correlation was much lower between lapses of attention and precision (r = .30–.35, p < .0001). External validity was also

good: storage capacity was strongly correlated with the Wechsler Test of Adult Reading for both the episodic memory task (r = .49, p < .0001) and the working memory task (r = .53, p < 

.0001). CONCLUSIONS: These findings show that using computational models to isolate specific aspects of psychological function from behavioral data can yield reliable and valid measures of

psychological function. The present models make it possible to distinguish between lapses of attention, insufficient memory storage capacity, and imprecise memory contents, factors that may

vary across diagnoses and respond to different treatments. DISCLOSURE: Nothing to disclose. 39.3 PREDICTING ATTENTION LAPSES ACROSS PSYCHOPATHOLOGY USING EEG MEGAN BOUDEWYN UNIVERSITY OF

CALIFORNIA, SANTA CRUZ, CALIFORNIA, UNITED STATES BACKGROUND: Lapses of attention are thought to contribute to the impairments in performance on a variety of cognitive tasks that have been

observed in psychiatric populations. Our goal was to extract and examine neural markers of the attention lapse construct in people with schizophrenia, bipolar disorder and major depression,

as well as healthy controls. Following from previous work from our group and others that has connected periods of inattention during cognitive tasks to changes in alpha-band activity (8-13

Hz), a measure derived from EEG recordings, we used pre-stimulus alpha activity as a measure of attention lapses. Our hypothesis is that pre-stimulus alpha power predicts behavioral markers

of attention lapsing, thereby accounting for group differences in performance on multiple cognitive tasks. METHODS: EEG was recorded while participants with schizophrenia (N = 35), bipolar

disorder (N = 25), major depression (N = 35) and demographically matched healthy control participants (N = 35) completed a set of cognitive tasks, including visual perception, episodic and

working memory tasks. Logistic linear mixed effects regression models were used to predict response accuracy as a function of diagnostic group and pre-stimulus alpha power. RESULTS: Across

tasks, all patient groups showed higher error rates compared to healthy controls (ps<0.05). In addition, increased pre-stimulus alpha activity significantly increased the likelihood of

making an error in all groups (ps<0.05). CONCLUSIONS: The pattern of results supported our hypothesis, demonstrating that a biological measure of attention lapsing significantly predicts

response accuracy across cognitive tasks in all groups. Participants with a psychiatric disorder also showed higher rates of attention lapsing as compared to healthy controls. Overall, the

results indicate that attention lapsing contributes to differences in performance across cognitive tasks in a range of major mental health disorders. DISCLOSURE: Nothing to disclose. 39.4

COMPUTATIONAL MODELS GO ONLINE: OPPORTUNITIES AND PITFALLS OF TASK-BASED EPIDEMIOLOGY ANGUS MACDONALD UNIVERSITY OF MINNESOTA, MINNEAPOLIS, MINNESOTA, UNITED STATES BACKGROUND: Our focus on

understanding computational models of performance in serious mental illnesses is complemented by an exploration of performance profiles across the general population. Community samples such

as HCP, UK Biobank, ABCD and All of Us will play a central role in understanding psychiatric risk factors by linking adaptive and maladaptive biological mechanisms. To date, such samples

show a heavy reliance on self-report questionnaires and computational models of behavioral performance remained under-represented in this revolution. It remains largely unknown how to

validly extend online data collection to uncontrolled settings. METHODS: Parallel to in-lab task development, we sampled n = 834 participants with valid responses to obtain preliminary data

on measures of working memory, episodic memory, and sensory precision in a manner that allowed us to measure the precision of the underlying representation (in all cases as locations along a

circle). Participants completed two one-hour sessions, including self-report and validity measures and a subset of behavioral tasks counter-balanced across subjects. In addition to

self-reported validity items, tasks included trials to test data quality and attention lapses. RESULTS: Online tasks replicated key in person patterns of performance, in particular neural

precision and attentional engagement. Precision measures across tasks were highly correlated as were measures of attentional lapsing, suggesting common underlying behavioral mechanisms. We

used conservative split-half cross-validation to establishing relationships with individual differences. Working memory capacity and precision related to the cognitive-perceptual factor of

the Schizotypal Personality Questionnaire in both exploratory and validation samples. Attentional lapsing also predicted this factor. CONCLUSIONS: While neural precision was drawn upon in

diverse functions, differentiating this from attentional impairment remains a crucial goal. Schizotypal traits related to these impairments, and served to validate precautions used in

collecting these samples. While the utility of questionnaire-based online studies is increasingly common, these findings show a path for large-scale, task-based epidemiology to link

computational parameters to personality traits and psychopathology risk factors. DISCLOSURE: Nothing to disclose. PANEL 40. DISENTANGLING THE NEURAL MECHANISMS UNDERLYING COGNITIVE CONTROL –

RELEVANCE TO, AND BIOMARKERS OF, DYSFUNCTION 40.1 A ROLE FOR THE CLAUSTRUM IN COGNITIVE CONTROL BRIAN MATHUR UNIVERSITY OF MARYLAND SCHOOL OF MEDICINE, BALTIMORE, MARYLAND, UNITED STATES

BACKGROUND: Cognitive dysfunction is a feature of several neuropsychiatric disorders that is characterized by disruptions in cortical networks. Our inability to treat cognitive dysfunction

reflects our poor understanding of cortical network biology. A leading model for how widespread cortical network nodes co-activate, into networks is recruitment of a subcortical nucleus that

synchronizes cortical network nodes. The subcortical nucleus the claustrum is herein hypothesized to serve this role. METHODS: We test this hypothesis using resting-state functional

magnetic resonance imaging (n = 51; voxel-wise correction for multiple comparisons FWE p < 0.05) and functional circuit mapping (n = 1,050 neurons; subgraph extraction using permutation

test) in mouse of both sexes. In humans we implemented the multi-source interference attention task (MSIT) and experimental pain in (n = 34 healthy subjects of both sexes) and applied linear

mixed effect model analyses. RESULTS: At rest, multiple frontal cortical areas are significantly (voxel-wise correction for multiple comparisons FWE p < 0.05) functionally connected with

both the claustrum and posterior cortical regions. Using these cortical regions as guides, we find using channelrhodopsin-assisted circuit mapping that these frontal areas connect, via the

claustrum, to these posterior cortical regions. In particular, the anterior cingulate and prelimbic prefrontal cortices statistically cluster in their connections with posterior cortices

(cohen’s d > 1.66). Probing whether MSIT or experimental pain, which both known to induce synchronous frontoparietal cortical network activity, activate the claustrum we found that

stimulus onset at both the MSIT and experimental pain (q = 0.0047) elicit significant claustrum activation in human subjects. CONCLUSIONS: These results support a model in which the

claustrum is recruited to support frontoparietal cortical networks via common glutamatergic input to frontoparietal cortical network nodes. DISCLOSURE: Nothing to disclose. 40.2 PREFRONTAL

CORTEX PARVALBUMIN NEURONS UNDERLYING ATTENTION - RELEVANCE FOR NEUROPSYCHIATRIC DISORDERS TYLER DEXTER UNIVERSITY OF WESTERN ONTARIO, LONDON, CANADA BACKGROUND: Abnormal oscillatory

activity in the prefrontal cortex is a hallmark of neuropsychiatric disorders such as schizophrenia. These abnormalities may arise from dysfunctional parvalbumin interneuron (PVI) function,

a class of inhibitory neurons that regulate high-frequency oscillations, and underly the cognitive impairments. Here, we investigated how PVIs in the medial prefrontal cortex (mPFC) of mice

contribute to attention under non-pathological conditions and disease states. To model aspects of schizophrenia, we assessed the potential of targeting prefrontal PVIs to improve cognition

in a 22q11.2 microdeletion syndrome (22q) mouse model. METHODS: To assess attention, we used the touchscreen rodent continuous performance task (rCPT). For Exp 1, male and female mice were

bred to express channelrhodopsin (ChR2) on PVIs and implanted with bilateral optic probes in the mPFC. For Exp 2, adult male and female 22q mice and littermate controls were assessed on the

rCPT. Additionally, 22q mice were genetically modified for optogenetic control of PVIs. These mice received bilateral injections of a cre-dependent ChR2 virus into the mPFC and were assessed

on the rCPT prior to optogenetic stimulation of prefrontal PVI. In vivo optogenetics was used to stimulate prefrontal PVIs at a high (30hz) or low (5hz) frequency during the response phase

of the task. RESULTS: Exp 1 shows that 30hz stimulation of prefrontal PVIs significantly improved animal’s ability to discriminate the correct image during the rCPT, while 5hz stimulation

significantly impaired these processes. In Exp 2, we observed a significant reduction in rCPT performance in both male and female 22q mice, specifically in the ability to detect and

discriminate target stimuli. Finally, preliminary data shows that 30hz stimulation of prefrontal PVIs in 22q mice and controls significantly improves rCPT performance. While 22q mice

performed significantly worse than controls at baseline, 30hz stimulation rescued this impairment. CONCLUSIONS: We demonstrate that prefrontal PVIs regulate attention in a frequency

dependent manner, and that high frequency activation enhances cognitive performance in non-pathological mice. Further, direct targeting of prefrontal PVIs may serve as a therapeutic avenue

of interest for improving cognition in neuropsychiatric disorders. DISCLOSURE: Nothing to disclose. 40.3 MICE WITH AN AUTISM-ASSOCIATED R451C MUTATION IN NEUROLIGIN-3 EXHIBIT INTACT

ATTENTION ORIENTING BUT ATYPICAL RESPONSES TO METHYLPHENIDATE AND ATOMOXETINE IN THE MOUSE-POSNER TASK EMMA BURROWS FLOREY INSTITUTE OF NEUROSCIENCE AND MENTAL HEALTH, PARKVILLE, AUSTRALIA

BACKGROUND: Atypical attention orienting has been associated with some autistic symptoms, but the neural mechanisms remain unclear. The human Posner task, a classic attention orienting

paradigm, was recently adapted for use with mice, supporting the investigation of the neurobiological underpinnings of atypical attention orienting in preclinical mouse models. The current

study tested mice expressing the autism-associated R451C gene mutation in neuroligin-3 (NL3) on the mouse-Posner (mPosner) task following administration of clinically-relevant attention

modulators. METHODS: Male NL3 and wild-type (WT) mice (n = 8-10 per group) were trained to sustain a nose-poke to a central square on a touchscreen until a validly or invalidly cued target

was displayed. The cue was a peripheral non-predictive flash in the exogenous task and a central spatially predictive image in the endogenous task. The effects of methylphenidate (MPH) and

atomoxetine (ATO; latin-square design; n = 8-10 per group) on performance were assessed. Mann-Whitney U tests were conducted to compare sessions to criterion in training and genotype and

drug effects were analysed by generalised linear latent and mixed models (GLLAMM) with robust standard error estimation and with individual mice treated as random effects. RESULTS: In both

tasks, mice were quicker and more accurate in the validly versus invalidly cued trials, consistent with results in the human Posner task (validity). Administration of MPH increased the

difficulty of disengaging exogenously oriented attention only in NL3 mice, (reduced accuracy and increased omission errors in invalid trials). MPH enhanced endogenous orienting in both NL3

and WT mice (i.e., increased correct responses and reduced omission errors). In contrast ATO impaired exogenous orienting in both WT and NL3 mice (i.e., decreased response times and

increased incorrect responses) and decreased accuracy of endogenous orienting through more omission errors in WT mice only. CONCLUSIONS: This study presents the first investigation of

attention orienting in a preclinical model of autism and expands our understanding how the NL3R451C mutation may lead to selective alterations in attentional processes. DISCLOSURE: Nothing

to disclose. 40.4 TRANSLATIONAL NEUROPHYSIOLOGICAL MEASURES TO EVALUATE COGNITIVE CONTROL FOLLOWING PRENATAL ALCOHOL EXPOSURE IN RODENTS JONATHAN BRIGMAN UNIVERSITY OF NEW MEXICO HEALTH

SCIENCES CENTER, ALBUQUERQUE, NEW MEXICO, UNITED STATES BACKGROUND: Fetal Alcohol Spectrum Disorders (FASD) are still the most common type of neurodevelopmental syndrome and cognitive

control has been consistently reported to be impaired in affected individuals. The 5 Choice Continuous Performance Task (5C-CPT) has recently been shown to measure attention and cognitive

control in rodents and show similar cross-species neural signatures via EEG. METHODS: Male and female offspring (12-16 per sex/trt) from a drinking-in-the-dark model (10% EtOH with 0.066%

saccharine for 4hrs/day throughout gestation; BAC: ~90mg/dL; controls=0.066% sweetened water, “SAC”). At 8-10 weeks, mice trained to touch white target stimuli. Next, mice underwent

stereotactic surgery and fitted with dura-resting skull screws targeting medial prefrontal, parietal, and motor cortices. After recovery non-target trials (5 stimuli presented; withholding

of response was rewarded) were added at a 2:1 ratio for 5 days followed by recording at a 5:1 ratio for 12 days. ANOVA was used to examine main effect of treatment, sex, and interaction

effects for dependent variables. RESULTS: On 2:1 there was a significant difference for sex and trt, [Session: p = 0.96, Sex: p = 0.04; Trt: p = 0.002; Sex x Trt:, p = 0.06]. The false alarm

rate was significantly different for sex and trt with an interaction [Session: p = 0.74, Sex: p < 0.0001; Trt: p < 0.0001; Sex x Trt: p = 0.04]. For 5:1, PAE mice made more false

alarms [Session: p < 0.0001, Sex: p = 0.59; Trt: p = 0.001; Sex x Trt: p = 0.25]. EEG revealed high frequency enhancement in correct rejection trials compared to target correct trials

[ROI: p = 0.20; Sex: p = 0.54; Trt: p = 0.65; Sex x Trt: p = 0.04] and a strong low frequency enhancement in target-post-error response trials compared to target-post-target [ROI: p = 0.005;

Sex: p = 0.31; Trt: p = 0.22 Sex x Trt: p = 0.11]. CONCLUSIONS: PAE spared attention on target-only trials while impairing performance when non-targets were added. EEG found an increase in

frontal theta in all mice for post-error correction as and increases in posterior beta during choice conflict. Analysis of posterior beta revealed a sex x trt effect with increased power

during correct rejection trials in female PAE mice. These results show that 5C-CPT can detect post-error signal changes consistent with human EEG studies and that alterations in cortical

activity via EEG may be used as a potential biomarker of PAE. DISCLOSURE: Nothing to disclose. STUDY GROUP 41. ELIMINATING MENTAL HEALTH AND SUBSTANCE USE STIGMA: THE TIME IS NOW MARK

RAPAPORT*, DARRELL KIRCH, SETH KAHAN, WILLIAM SMITH, ARTHUR EVANS, ELLEN KAHN, STEPHENIE LARSEN, JOHN MACPHEE, CHRISTINE MOUTIER, BRANDON STAGLIN, GAHAN PANDINA HUNTSMAN MENTAL HEALTH

INSTITUTE, SALT LAKE CITY, UTAH, UNITED STATES STUDY GROUP SUMMARY: Mental health and substance use disorder (SUD) stigma, ignorance, shame, and fear have caused our patients to be

marginalized, at times criminalized, and disenfranchised in terms of both psychiatric and physical care. This stigma and prejudice has also led to decreased funding for clinical care,

training and research. The study group will discuss employing the social impact model to create a sustained grand challenge to eliminate mental health and substance use disorder stigma. The

urgency and opportunity to create a grand challenge will be discussed by Dr. Kirch while the theory and implementation model will be presented by Mr. Kahan, an international expert in

leading grand challenges. Dr. Smith will summarize his work on the impact of racism and micro-cultures on not only creating stigma but limiting successful strategies to eliminate stigma. Ms.

Kahn will discuss the work that the Human Rights Campaign is doing at the interface between mental health and SUD challenges and the LBTQ + community while Ms. Larsen will demonstrate how

Encircle has created safe spaces throughout the Mountain West where LBGTQ + youth can be accepted and receive mental health and SUD care. Mr. MacPhee will describe the exciting work that the

JED foundation is doing on high school and college campuses to eliminate stigma and provides evidence-based assessment and resources to our educational system. Dr. Moutier will describe the

long-term work that ASFP has done to breakdown the ignorance, shame, guilt and societal judgement that frequently accompanies both people who attempt to kill themselves and the family

members of people who die because of suicide. Dr. Evans will describe the thoughtful strategies that the American Psychological Association have implemented in their efforts to decrease

stigma. Dr. Pandina will discuss the lifelong work that he and his team has done to battle stigma- from his support of One Mind to the creation of the Science over Stigma movement. While Mr.

Staglin will discuss the systematic yet innovative work that he and One Mind have done employing podcasts, video and social media to work on the elimination of stigma. By the end of the

study group the participant will have learned about the National Academy’s work on mental health stigma, some of the tremendous work already being done by many individual organizations, why

a growing number of people believe that now is the time to create a sustained national movement to eliminate mental health and SUD stigma and prejudice, and how the participant can become

involved in this effort. DISCLOSURE: Nothing to disclose. PANEL 42. DEVELOPMENTAL CRITICAL PERIODS FOR SOCIAL PROCESSING: WINDOWS OF VULNERABILITY AND OPPORTUNITY 42.1 WORTH THE RISK?

EFFECTS OF DEVELOPMENT AND EARLY CARE QUALITY ON LATERAL HABENULA INVOLVEMENT IN INFANT SOCIAL BEHAVIOR FLEXIBILITY MAYA OPENDAK KENNEDY KRIEGER INSTITUTE, BALTIMORE, MARYLAND, UNITED STATES

BACKGROUND: Flexible social behavior is critically important during early life, when environmental demands are in constant flux. On the other hand, heightened circuit plasticity during this

period also renders the infant uniquely vulnerable to environmental influences that can provide a template for lifelong social behavior. However, the circuit mechanisms linking early

experience to lasting social behavior patterns remain unclear. Here, we focus on the ontogeny of the lateral habenula (LHb), a key negative regulator of dopaminergic signaling, in social

behavior in typical and perturbed development. METHODS: In all studies, we used the Scarcity-Adversity model of Low Bedding (SAM-LB) from PN8-12, in which the dam is given limited nesting

materials, with equal #s male and female pups used. Expt 1. Habenulae were dissected at PN18/ PN28 and processed via Western blot. Expt 2. PN18/ PN28 rat pups were injected with 14C

2-deoxyglucose (2-DG, 20 μCi/100 g, s.c.) and then received mild tail shocks (0.5mA) or no shocks, alone or with a social partner present every 5 mins for 40 mins, followed by brain removal

for autoradiography. Expt 3. PN3 pups were bilaterally transduced in LHb with pAAV-hSyn-hM4D(Gi)-mCherry or pAAV-hSyn-mCherry. PN18/28 pups underwent peer sociability tests with/without

ambient predator odor. RESULTS: We observed an increase in CaMKIIβ expression in PN28 pups compared to PN18 pups, regardless of rearing condition (p < 0.001). Metabolic imaging data

showed that post-weaning, social presence and shock (threat) cues were associated with increased 2-DG metabolism in the lateral (p = 0.007) and medial habenula (p = 0.009); these stimuli did

not increase Hb engagement at PN18 or in SAM-LB-reared pups. Hb network connectivity varied with rearing, social presence, and threat (p = 0.003). Chemogenetic inhibition of the LHb at PN28

increased social approach when threat odor was present (p = 0.016). SAM-LB was associated with a robust decrease in social approach behavior in both sociability tests at PN28, which could

be rescued by LHb inhibition via DREADDs (p = 0.045). CONCLUSIONS: Development and early care influence LHb involvement in processing social and threat cues. Whereas weaning typically

triggers habenula-dependent inhibition of social approach under threat, early adversity engages LHb-inhibited approach even in the absence of threat. DISCLOSURE: Nothing to disclose. 42.2

MECHANISMS LINKING JUVENILE SOCIAL ISOLATION AND ADULT SOCIAL BEHAVIOR DEFICITS HIROFUMI MORISHITA ICAHN SCHOOL OF MEDICINE AT MOUNT SINAI, NEW YORK, NEW YORK, UNITED STATES BACKGROUND:

Loneliness is becoming increasingly recognized as a threat to mental health. Juvenile social isolation (JSI) is known to disrupt social behavior in adulthood, but little is known about the

neural mechanisms linking these two events. Our previous study suggests that there is a juvenile sensitive period when isolation will induce medial prefrontal cortex (mPFC) abnormalities

including parvalbumin positive interneurons (PVIs) in adult mice. Of note, these circuit abnormalities were not present at the end of the isolation, raising the question of when and how

JSI-induced social deficits emerge over the course of development. METHODS: Developmental progression of JSI-induced social dysfunction was investigated by the 3-chamber sociability test (n 

= 12-21mice). Affiliative behavior and aggression were also tested among cage mates. To modulate activity of mPFC-PVIs during a re-grouped housing period, male mice were injected with either

excitatory DREADD or mCherry at postnatal day (p) 14, isolated for 2 weeks (p21-35) and given CNO in their drinking water for 2 weeks (p35-49). Mice were then subjected to 3 chamber

sociability test, open field, elevated plus maze, or light dark box in adulthood (n = 9-10mice). Only male mice were used as female JSI mice did not show sociability deficits. RESULTS: We

found that JSI-induced social dysfunction in the 3-chamber test is delayed (not fully emerging until p50) and the sociability deficit, where subjects interact with novel mice, is preceded by

negative social interactions between JSI cage mates during the first week after the end of isolation. A chemogenetic modulation of dmPFC-PVI activity for a 2-week adolescent period (p35-49)

at the start of re-housing leads to a long-term rescue of sociability (p = 0.045) without altering anxiety or locomotor behaviors. CONCLUSIONS: These results suggest that the prevailing

“social deprivation model”, where adult social deficits are attributed to disruption of developmental processes occurring during the isolation period, should be supplemented by the

“developmental mismatch model”, where social deficits are attributed to disruption of developmental processes occurring after the isolation period. Our study also highlights a novel

adolescent sensitive window to prevent the emergence of social behavioral deficits in adulthood. DISCLOSURE: Nothing to disclose. 42.3 SEX DIFFERENCES IN THE IMPACT OF JUVENILE STRESS ON

SOCIAL BEHAVIOR AND VENTRAL HIPPOCAMPAL-NUCLEUS ACCUMBENS ACTIVITY FRANCIS LEE WEILL CORNELL MEDICAL COLLEGE, NEW YORK, NEW YORK, UNITED STATES BACKGROUND: Impairment in social behavior is a

key feature of many psychiatric disorders. However, less is known about the effects of juvenile stress on the development of circuits mediating social behaviors, especially in females who

are more likely to be diagnosed with mood disorders. These studies use animal models for early life adversity to identify sensitive periods in social development. METHODS: To quantify

development of ventral hippocampus (VH) inputs to nucleus accumbens (NAc), anterograde tracer PHA-L was injected into VH of P23, P30, and P60 male (m) and female (f) mice (n = 6/group).

Terminals in NAc were quantified as a measure of VH neuron projection density. Separate male and female mice underwent 10 days of chronic unpredictable stress (CUS) (P22-P31). Three-chamber

test and free social interaction (FSI) tests began at P70 (n = 9-11/group). In-vivo calcium imaging was used to record activity in NAc-projecting VH neurons during FSI in adulthood (n = 

7–11/group). RESULTS: There was increased fiber density in the NAc at P30 compared to P23 or P60 in males (p < .0001, p < .0001) and females (p = .023, p < .0001). Additionally,

females had increased fiber density compared to males at P23 (p < .0001). At P70, in the 3-chamber test, all control (CTL) mice and CUS males spent more time interacting with a social

partner than an object while CUS females did not (m: ctl: p < .001, cus: p < .01; f: ctl: p < .0001, cus: p = .854). In FSI, regardless of sex, CUS mice spent less time interacting

with a social partner than CTLs (p = .04). For fiber photometry, neural activity was higher 2 seconds post interaction initiation than 0.5 seconds pre for stressed females (p < .001) but

not CTLs (p = .162). Additionally, CUS females had increased activity 2 seconds post social interaction relative to CTLs (p = .044). In males, CTL (p = .003) and CUS (p = .001) mice showed

increased activity from pre to post social interaction initiation, but there was no difference between groups in the 2 seconds post interaction initiation (p = .894). CONCLUSIONS: We

identified a juvenile sensitive period for sociability that coincides with changes in VH-NAc connectivity. Females that underwent juvenile CUS show elevated activity in the VH-NAc circuit

upon initiation of social interaction. Together, these findings inform our understanding of sex-differences in sensitive windows for social development. DISCLOSURE: Nothing to disclose. MINI

PANEL 43. INTERACTIONS BETWEEN INSULIN AND DOPAMINE IN THE BRAIN: IMPLICATIONS FOR METABOLISM ACROSS HEALTH AND DISEASE 43.1 EFFECT OF SHORT-TERM OVEREATING OF HIGH CALORIC SNACKS ON

INSULIN RESPONSIVENESS OF THE BRAIN IN NORMAL-WEIGHT MEN STEPHANIE KULLMANN INSTITUTE FOR DIABETES RESEARCH AND METABOLIC DISEASES OF THE HELMHOLTZ CENTER MUNICH AT THE UNIVERSITY OF

TÜBINGEN, TÜBINGEN, GERMANY BACKGROUND: Insulin resistance (IR) is the hallmark feature of obesity and type 2 diabetes with detrimental effects in the periphery and the central nervous

system. Numerous studies confirm that the disruption of insulin responsiveness in the human brain influences the pathology of metabolic, psychiatric, and neurodegenerative diseases. Insulin

action in the brain regulates synaptic plasticity and dopamine signaling. Human brain imaging studies show that central insulin action affects region-specific activity and functional

connectivity in the hypothalamus and the mesocorticolimbic circuitry influencing food intake and non-homeostatic feeding for pleasure. It is currently unclear whether short-term overeating

hampers insulin sensitivity of the brain in healthy individuals and whether this relates to changes in peripheral metabolism. METHODS: Brain insulin sensitivity was assessed by functional