Restoring glutamate receptosome dynamics at synapses rescues autism-like deficits in Shank3-deficient mice

Shank3 monogenic mutations lead to autism spectrum disorders (ASD). Shank3 is part of the glutamate receptosome that physically links ionotropic NMDA receptors to metabotropic mGlu5 receptors through interactions with scaffolding proteins PSD95–GKAP–Shank3–Homer. A main physiological function of the glutamate receptosome is to control NMDA synaptic function that is required for plasticity induction. Intact glutamate receptosome supports glutamate receptors activation and plasticity induction, while glutamate receptosome disruption blocks receptors activity, preventing the induction of subsequent plasticity. Despite possible impact on metaplasticity and cognitive behaviors, scaffold interaction dynamics and their consequences are poorly defined. Here, we used mGlu5–Homer interaction as a biosensor of glutamate receptosome integrity to report changes in synapse availability for plasticity induction. Combining BRET imaging and electrophysiology, we show that a transient neuronal depolarization inducing NMDA-dependent plasticity disrupts glutamate receptosome in a long-lasting manner at synapses and activates signaling pathways required for the expression of the initiated neuronal plasticity, such as ERK and mTOR pathways. Glutamate receptosome disruption also decreases the NMDA/AMPA ratio, freezing the sensitivity of the synapse to subsequent changes of neuronal activity. These data show the importance of a fine-tuning of protein–protein interactions within glutamate receptosome, driven by changes of neuronal activity, to control plasticity. In a mouse model of ASD, a truncated mutant form of Shank3 prevents the integrity of the glutamate receptosome. These mice display altered plasticity, anxiety-like, and stereotyped behaviors. Interestingly, repairing the integrity of glutamate receptosome and its sensitivity to the neuronal activity rescued synaptic transmission, plasticity, and some behavioral traits of Shank3∆C mice. Altogether, our findings characterize mechanisms by which Shank3 mutations cause ASD and highlight scaffold dynamics as new therapeutic target.