The autism-associated loss of δ-catenin function disrupts social behavior and prefrontal network activity

Abstract

Social behavior is imperative for survival in humans and many other species. There are various neurological disorders that indicate social impairment as a primary symptom. Previous works have suggested that synaptic structure and subsequent function can regulate social behavior, although the link between these is not yet fully understood. δ-catenin is expressed in excitatory synapses and functions as an anchor for the glutamatergic AMPA receptor (AMPAR) GluA2 subunit in the postsynaptic density. The glycine 34 to serine (G34S) mutation in the δ-catenin gene has been found in autism spectrum disorder (ASD) patients and results in a loss of δ-catenin functions at excitatory synapses, which is presumed to underlie ASD pathogenesis in humans. Our previous study using neuroblastoma cells has identified that the G34S mutation increases glycogen synthase kinase 3β (GSK3β)-dependent δ-catenin degradation to reduce δ-catenin levels, which likely contributes to the loss of δ-catenin functions. However, how the G34S mutation causes a loss of δ-catenin functions to induce synaptic dysfunction related to ASD-associated behaviors remains unclear, as does the effect of an overall loss of δ-catenin. My thesis work with colleagues reveals that synaptic δ-catenin and GluA2 levels in the cortex are significantly decreased in mice harboring the δ-catenin G34S mutation and the δ-catenin knockout (KO). In addition, the δ-catenin G34S and KO increase glutamatergic activity in cortical excitatory neurons while they decrease in inhibitory interneurons, indicating changes in cellular excitation and inhibition. This is important for brain network activity that contributes to multiple behaviors including social interaction. We in fact show that δ-catenin G34S and KO iii mutant mice exhibit social dysfunction, a common feature of ASD. Additionally, δ-catenin G34S mutant mice display altered network activity within the medial prefrontal cortex (mPFC), which likely underlies social deficit. In fact, abnormalities in prefrontal network activity has been reported in several studies of ASD. Most importantly, pharmacological inhibition of GSK3β activity reverses the G34S-induced loss of δ-catenin function effects in cells and mice. Finally, using δ-catenin KO mice, we confirm that δ-catenin is required for GSK3β inhibition-induced restoration of normal social behavior and prefrontal network activity in δ-catenin G34S mutant animals. Taken together, my work reveals that the loss of δ-catenin functions arising from the ASD-associated G34S mutation and δ-catenin KO mutation induce social dysfunction via alterations in activity at the synaptic, cellular, and network levels and that GSK3β inhibition can reverse δ-catenin G34S-induced synaptic and behavioral deficits.