Drawing Out Autism

Using brain imaging and advanced genetic tools, a new MIT collaboration is trying to show how genes could mold the autistic.

Scientists know volumes about the language and social problems that plague autistic children. Yet so far they've been unable to build a clear picture of the possible underlying genetic causes and neurological deficits. Now a group of MIT neuroscientists is attempting to unravel the intricacies of the disorder.

Autism is considered the fastest-growing developmental disorder in the United States. Typically diagnosed within the first three years of life, it leaves a person with profound problems in social interaction and communication.

Studies have shown that autism has a strong genetic component. Today, researchers think the disorder may involve 30 to 40 genes, which each exert a small effect, in combination with environmental factors. Large-scale studies of autism-affected families have revealed some candidate genes -- but little is known about how these genes contribute to the social and behavioral problems that characterize autism.

"The most pressing question is the causes of autism," says Andy Shih, chief scientific officer of the National Alliance for Autism Research, based in New Jersey. "It's crucial to understand how the genetic differences translate into the phenotype."

With the aid of a $7.5 million grant from the Simons Foundation, the MIT researchers will combine imaging and genetic tools to better understand this link.

"We will look for a relationship between gene variation and variation in the brain," says John Gabrieli, an MIT neuroscientist. Gabrieli will use fMRI, a type of MRI that shows which areas of the brain are active when people think about specific problems, to compare brain activity in normal individuals and in those with different forms of the suspected autism genes. Specifically, his group will look at how people deal with social functions, by imaging brain activity in response to faces and facial expressions.

Mriganka Sur, the neuroscience professor who heads the project, will look at similar genes in a variety of mouse models. By either blocking or boosting the action of these genes during critical periods of mouse development, researchers can determine the function of the genes, as well as the key timing for therapy.