Abstract
Psychosis is a hallmark of schizophrenia. It involves a collection of symptoms that are typically associated with disrupted dopamine signaling and emerges during adolescence or early adulthood. Most schizophrenia-associated genes, however, involve glutamatergic or other ubiquitous targets that do not explain the latent expression of psychosis or dopaminergic abnormalities. Here, we describe an etiologically relevant model for the adolescent onset of dopamine-related dysfunction in schizophrenia. We focused on GRIN2A, the gene encoding the GluN2A subunit of the NMDA receptor, as both the common loss-of-function variants and the rare missense variants in this gene are risk factors for schizophrenia. We find that GluN2A levels distinctly decline in dopamine neuron-containing regions throughout adolescence while remaining stable in other regions. This suggested that adolescent dopamine neurons may be particularly vulnerable to further reductions in GluN2A caused by a damaging variant of GRIN2A . Consistent with this idea, we find that selective knockout of Grin2a in adolescent rat dopamine neurons results in a psychosis-relevant behavioral phenotype. This manipulation also reduced dopamine release in response to unexpected outcomes in young adults, reflective of prediction error signaling abnormalities observed in the clinical population. These data provide mechanistic insight into how GRIN2A mutations may contribute to the delayed onset of dopamine-related symptoms and provide a model for identifying course altering treatments for schizophrenia.