Many of the mutations that are known to be associated with autism disable protein complexes that support the development and function of synapses — the communication junctions between neurons. Morgan Sheng and his colleagues at the Massachusetts Institute of Technology are investigating the role of these protein complexes to learn how their loss may lead to the social and cognitive disabilities in autism.

Two neurons talk to each other at specialized regions of the cells called synapses. To make a synapse, neurons generate extensions known as dendritic spines that touch the axon of another neuron. Scaffolding proteins, such as members of the Shank family, bring together the cytoskeleton (the cell’s internal scaffolding) and intracellular signaling factors within the dendritic spine, coordinating these pathways’ activities. Mutations in one Shank protein, Shank 3, have been found in familial cases of autism. Sheng and colleagues previously found that neurons lacking Shank1 (a close relative of Shank3) produce smaller dendritic spines and have weaker functional synapses in the mouse brain. In behavioral tests, mice that lack Shank 1 have mixed cognitive abilities reminiscent of autism: better learning of spatial information (but reduced ability to retain that memory), as well as poor memory formation in fear conditioning.

Sheng and colleagues now plan to study mice that lack Shank 1 or Shank 3 to better understand the role of this protein family in neurological and cognitive function. The researchers have found an imbalance of inhibitory and excitatory signals in Shank 1–deficient brains. Normally, this excitatory–inhibitory balance is precisely tuned to transmit information between different regions of the brain; disruptions in the balance could account for the abnormal processing of information seen in people with autism. Sheng and colleagues plan to look for accompanying changes in the social interactions of the Shank 1- and Shank 3–deficient mice, which might indicate an autism-like impairment in their behavior. In addition, Sheng and colleagues have been studying how different members of the neuroligin family of synaptic cell surface proteins — which can bind to Shank scaffold proteins — instruct neurons to make excitatory or inhibitory synapses with each other. Ultimately, it is hoped that these studies will illuminate the synaptic communications that go awry in autism, as well as potentially generate useful and representative animal models of autism for future studies.