Boston scientists find autism genes linked to learning

Researchers from Boston have discovered six new genes implicated in autism. The genes normally make new brain connections needed for learning, but their absence or silence apparently places them among many mutations that lead to the devastating disorder, which is marked by trouble with communication and social interaction.

Writing in tomorrow's issue of Science, Dr. Christopher Walsh (left) of Children's Hospital Boston and his co-authors say in some of the genetic mutations they found, the genes were present but the on/off switches they controlled were broken.

"People think of genetic diseases as immutable and untreatable," Walsh said in an interview. "Studies like ours and others give more hope we might not need to replace genes one by one, but find other ways of activating the genes that might be silent."

Walsh said the Science paper combines work done in his lab on genetic mutations that cause autism with research by his Children's colleague Michael Greenberg on brain cell activity changing gene activation. The two scientists both presented their work at a meeting in January.

"I showed my list of autism genes and he showed his list of plasticity of genes," Walsh said. "We found a lot of genes on both lists and that occurred much more frequently than expected by chance."

That led to their hypothesis that in autism, brain cells are unable to turn genes on to make new brain connections.

"People have been studying the molecular correlate of learning in the brain. They suspected for a long time that defects of that process are probably very important to the brain and disorders like autism," Walsh said. "It's still surprising to see that connection made explicit and see a child who has the mutation."

The researchers studied large Middle Eastern families in which cousins had married and the incidence of autism was high. Genetic analysis showed missing DNA in parts of the genome linked to autism. They discovered six genes along these stretches that are part of the molecular network involved in learning and memory.

Symptoms of autism arise between age 1 and 2, when the child's developing brain is refining connections between its nerve cells triggered by experiences, Walsh said. Genes that are not functioning normally do not allow this process to take place.

The research may also explain why some autistic children improve after repetitive interventions that perhaps jump-start previously turned-off genes.

"Sometimes the genes aren't completely inactive. We know that intensive training or enriching of the environment in animal models has ways of turning genes on that would normally be silent," Walsh said.

Developing drugs to target these genes also holds potential, he said.

Dr. Steven E. Hyman, provost of Harvard University, a neuroscientist, and former director of the National Institute of Mental Health,
praised the study.

"The groundbreaking work is particularly significant because it supports a model of autism as a condition that disrupts the brain's ability to form new connections in response to experience," he said in a statement. "This has been an area of intense research and these new findings begin to bring the pieces together."