Mental Activity May Affect Autism-Linked Genes
Friday, July 11, 2008
New research suggests that some cases of autism arise from defects in genes that can be turned on or off by mental activity, a finding that sheds light on the devastating condition and might eventually lead to strategies to treat it.
The findings are drawn from gene scans of about 100 Middle Eastern families in which autism is unusually common. The disorder is marked by social isolation, speech problems and strange, repetitive activities.
The study, done by a large international team and reported today in the journal Science, adds to the growing evidence that autism may result from problems in the immensely complicated process by which some networks of brain cells expand and many others die in the first few years after birth.
The fact that three of the half-dozen genes identified in the new report are regulated by "neuronal activity" -- feeling, thinking, doing -- suggests in theory that changing the experiences of autistic children could change the course of the disease.
"The genes implicated in our study are ones that interact with the environment and are involved in how the brain converts what it sees from the environment," said Christopher A. Walsh, a neurologist and chief of genetics at Children's Hospital in Boston who headed the team. "If we can activate those genes by other mechanisms, we might be able to help the kids."
Other researchers agreed that the discovery that some "autism genes" are also "experience genes" is provocative and, at some level, hopeful.
"If that is a general mechanism, then the other genes that have been identified as associated with autism should be expected to be driven by activity, as well. But we don't know that yet," said Daniel H. Geschwind, who is co-director of the Center for Autism Research and Treatment at the University of California at Los Angeles.
The study demonstrates that "environmental experiences and influences that shape postnatal brain development are not irrelevant," said Isabelle Rapin, a pediatric neurologist at the Albert Einstein College of Medicine in New York. She added, however, that "we have but the most primitive ideas about what the proteins coded by identified missing or mutated genes do."
Autism is one of the more urgent health problems of the 21st century -- and one of the more mystifying. A study by the Centers for Disease Control and Prevention last year reported that 1 in every 150 American children has an "autism spectrum disorder," which includes conditions from the severe and disabling to the mild and often-overlooked.
Some states have experienced huge increases. In California, the number of children getting services for autism tripled from 1987 to 1998 and then doubled between 1998 and 2002. In Minnesota, diagnosed autism increased from two cases per 1,000 schoolchildren in 1998 to 6.6 cases per 1,000 in 2002.
In most cases, the problems are apparent by age 3. In many, a normally developing child suddenly regresses, stops talking and becomes interested in only a few highly stereotyped activities.
Studies have suggested that genes probably contribute about 70 percent to a child's risk of developing autism. Whether environmental exposures or life events can trigger the disease is unknown. Numerous studies have found that childhood vaccinations, and a mercury-containing preservative that vaccines once contained, almost certainly do not play a role.
The new study, led by Walsh and Eric M. Morrow, a psychiatrist at Harvard Medical School, studied 104 families from the Arab Middle East, Turkey and Pakistan. In 88, the parents were cousins. The average family had two autistic children. One Kuwaiti and one Pakistani family, however, each had four.
Marriage between first cousins doubles the risk of neurological birth defects. The researchers said they think that shared ancestry would increase the risk of autism caused by recessive mutations that cause problems only when a child inherits the defective gene from both parents. By studying cases caused by such rare events, researchers can often learn about the biochemical and genetic underpinnings of the far more common cases in which there is no inbreeding.
The researchers found six genes with mutations or missing pieces. Three had been identified by Michael E. Greenberg, a Harvard neurobiologist, as members of a group of genes that are regulated -- turned up or down -- by the activity of the cells containing them and of the nerve networks those cells inhabit. Furthermore, two of the six genes are known to be involved in the growth of axons, the tendrils that nerve cells send out to contact other cells.
Many brain cell connections are prenatal, hard-wired and do not depend on experience. The new findings suggest that autism may involve problems that occur later, when nerve networks branch out and make solid connections, or loosen and are pruned back, a process essential to learning.
"Autism is a disorder of social learning for sure," Morrow said.
If there is a connection between the genes' activity and that of brain cells responding to experience, then it may be possible to stimulate the cells in a way that wakes up or gets around the defective genes.
The researchers looked for one of the gene defects in autistic children whose parents were not related and found it, but it remains unclear how applicable this study is to garden-variety autism.
Many of the Middle Eastern children had other neurological problems, such as epilepsy. Whether the activity-driven gene defects are also present when autism occurs by itself is unknown.