Mutations in Shank3 are strongly associated with autism spectrum disorders and neural circuit changes in several brain areas, but the cellular mechanisms that underlie these defects are not understood. Homeostatic forms of plasticity allow central circuits to maintain stable function during experience-dependent development, leading us to ask whether loss of Shank3 might impair homeostatic plasticity and circuit-level compensation to perturbations. We found that Shank3 loss in vitro abolished synaptic scaling and intrinsic homeostatic plasticity, deficits that could be rescued by treatment with lithium. Further, Shank3 knockout severely compromised the in vivo ability of visual cortical circuits to recover from perturbations to sensory drive. Finally, lithium treatment ameliorated a repetitive self-grooming phenotype in Shank3 knockout mice. These findings demonstrate that Shank3 loss severely impairs the ability of central circuits to harness homeostatic mechanisms to compensate for perturbations in drive, which, in turn, may render them more vulnerable to such perturbations.

Mice are opportunistic omnivores that readily learn to hunt and eat insects such as crickets. The details of how mice learn these behaviors and how these behaviors may differ in strains with altered neuroplasticity are unclear. We quantified the behavior of juvenile wild type and Shank3 knockout mice as they learned to hunt crickets during the critical period for ocular dominance plasticity. This stage involves heightened cortical plasticity including homeostatic synaptic scaling, which requires Shank3, a glutamatergic synaptic protein that, when mutated, produces Phelan-McDermid syndrome and is often comorbid with autism spectrum disorder (ASD). Both strains showed interest in examining live and dead crickets and learned to hunt. Shank 3 knockout mice took longer to become proficient, and, after 5 days, did not achieve the efficiency of wild type mice in either time-to-capture or distance-to-capture. Shank3 knockout mice also exhibited different characteristics when pursuing crickets that defied explanation as a simple motor deficit. Although both genotypes moved at the same average speed when approaching a cricket, Shank3 KO mice paused more often during approaches, did not begin final accelerations toward crickets as early, and did not close the distance gap to the cricket as quickly as wild type mice. These differences in Shank3 KO mice are reminiscent of some behavioral characteristics of individuals with ASD as they perform complex tasks, such as slower action initiation and completion. This paradigm will be useful for exploring the neural circuit mechanisms that underlie these learning and performance differences in monogenic ASD rodent models.

Mice are opportunistic omnivores that readily learn to hunt and eat insects such as crickets. The details of how mice learn these behaviors and how these behaviors may differ in strains with altered neuroplasticity are unclear. We quantified the behavior of juvenile wild type and Shank3 knockout mice as they learned to hunt crickets during the critical period for ocular dominance plasticity. This stage involves heightened cortical plasticity including homeostatic synaptic scaling, which requires Shank3, a glutamatergic synaptic protein that, when mutated, produces Phelan-McDermid syndrome and is often comorbid with autism spectrum disorder (ASD). Both strains showed interest in examining live and dead crickets and learned to hunt. Shank 3 knockout mice took longer to become proficient, and, after 5 days, did not achieve the efficiency of wild type mice in either time-to-capture or distance-to-capture. Shank3 knockout mice also exhibited different characteristics when pursuing crickets that could not be explained by a simple motor deficit. Although both genotypes moved at the same average speed when approaching a cricket, Shank3 KO mice paused more often, did not begin final accelerations toward crickets as early, and did not close the distance gap to the cricket as quickly as wild type mice. These differences in Shank3 KO mice are reminiscent of some behavioral characteristics of individuals with ASD as they perform complex tasks, such as slower action initiation and completion. This paradigm will be useful for exploring the neural circuit mechanisms that underlie these learning and performance differences in monogenic ASD rodent models. 3 Significance StatementDuring early development, activity-dependent plasticity mechanisms shape brain circuits.Shank3 is a synaptic protein that is mutated in Phelan-McDermid syndrome and is usually comorbid with autism spectrum disorder. Prior research shows that mice deficient in Shank3 exhibit abnormalities in a plasticity mechanism called homeostatic synaptic scaling. Here, we explore whether Shank3 knockout mice can learn to hunt crickets. We find that they can, although they pause more during their hunting and do not accelerate towards the cricket as rapidly. Future studies may be able to trace the neural circuits responsible for these differences, shedding more light on the causes of Phelan-McDermid syndrome and autism.