PROJECT SUMMARY More than half of all individuals in treatment for substance use disorder (SUD) will relapse. Inflexibility in selecting between familiar, habitual behaviors that have been rewarded in the past (drug seeking) and novel strategies that might be more advantageous (rehabilitation) may be a factor that preserves SUD. The goal of this proposal is to identify neural factors supporting goal-directed action selection, which could provide insight into therapeutic targets for disorders in which goal-oriented action selection is impaired. The dorsomedial striatum (DMS) is a brain region that receives and integrates glutamatergic input from cortical and subcortical regions required for goal-directed action selection. However, the factors in the DMS responsible for coordinating this incoming information remain incompletely understood. One candidate factor is the melanocortin-4 receptor (MC4R), a high-affinity receptor for α-melanocyte-stimulating hormone. MC4R regulates GluA2-AMPA receptor (GluA2-AMPAR) localization on dopamine D1 receptor-containing medium spiny neurons in the striatum. Inhibiting MC4R reduces the expression of repetitive, familiar behaviors and improves flexible action selection in mice. Thus, MC4R seems well-positioned in the DMS to integrate incoming glutamatergic signals and control flexible, goal-directed action. I will test the hypothesis that MC4R presence controls striatal-dependent action selection by regulating the cellular localization of GluA2-AMPARs in the DMS (Aim 1). Next, I will identify incoming projections that terminate on Mc4r+ neurons in the DMS and identify which projections are necessary for goal-directed behavior conferred by Mc4r silencing (Aim 2). In Aim 1, I will combine pharmacological inhibition of activity-dependent GluA2-AMPAR internalization with site-selective infusions of an MC4R agonist and measure the capacity of mice to engage in flexible action selection via instrumental conditioning assays. I will thus determine whether MC4R activity controls action selection via regulation of GluA2-AMPAR localization. Then, I will use synaptoneurosome preparations and quantify the levels and localization of multiple AMPA and NMDA glutamate receptor subtypes, providing a comprehensive perspective on MC4R control of glutamatergic receptor subunit expression in the DMS. In Aim 2, I will use rabies virus-mediated trans-synaptic tracing to create a brain-wide map of inputs onto Mc4r-expressing cells in the DMS. I will then use combinatorial viral vector strategies to test whether projections from specific brain regions, like the orbitofrontal cortex, influence action selection controlled by MC4R presence in the DMS. These experiments will identify the presynaptic partners necessary for striatal MC4R to influence an animal’s propensity to flexibly seek goals vs. engage in familiar routines. Impact. The proposed work has tremendous translational value, given that an over-reliance on inflexible, habit-like...