Share this article on:

Autism Gene Discoveries May Support Behavior Therapy


doi: 10.1097/01.NT.0000337651.15261.83
Back to Top | Article Outline


Discovery of several genes mutated in autism in consanguineous marriages in the Middle East suggest that a variety of brain mechanisms converge on neuronal connections to produce the disorder.

Investigators at Boston's Children's Hospital and international collaborators identified a half-dozen mutated genes in autistic people in Middle Eastern families. The implicated genes have diverse roles, yet all seem to mediate activation of neuronal genes, they reported in the July 11 Science.

The researchers say the findings support some current approaches to autism that may activate and maintain connections among nerve cells. More specific therapy for the disorder may one day be possible, they suggest, because the different genes affected in these families are not deleted but still present, even though they are not working properly.

“It tells us the problem involves gene regulation,” said Seung-Yun Yoo, PhD, one of the investigators who was a postdoctoral fellow funded by the National Alliance for Autism Research (currently Cure Autism Now) in the laboratory of Christopher A. Walsh, MD, PhD, at Children's Hospital. “It's not that they lack these genes,” Dr. Yoo said, “but expression may be dysregulated during neural activity.”

“Therefore we can probably alter this with therapy,” said Dr. Yoo, who is now a management consultant in life sciences. Early intervention is helpful in autism, she said, adding that the intensive behavioral therapy used for autistic children might activate genes that lead to proper synaptic development.



Evidence so far suggests autism is a heterogeneous disorder with a strong genetic component, Dr. Yoo said. To boost the likelihood of finding mutations, the Boston area Homozygosity Mapping Collaborative for Autism turned to families with consanguineous marriages, which increase the likelihood of birth defects a hundredfold. Using a search strategy called homozygosity mapping, the team sought gene changes that would appear in the same place on both chromosomes in individuals with autism. Because people who inherit a disease mutation are likely to also inherit the nearby gene variants on the chromosome, these loci surrounding the disease locus will tend to be the same (homozygous) in affected individuals. Searching for homozygous segments in diseased individuals from a marriage of cousins therefore helps to locate the disease gene.

Back to Top | Article Outline


The team recruited 104 families with autistic members from Middle Eastern countries, Turkey, and Pakistan; 88 of them included marriages of cousins. Collaborating researchers trained in each country to use standardized, validated research diagnostic tests to identify individuals with autism. Members of the Boston team also traveled to Dubai, Kuwait, Saudi Arabia, and Turkey to directly confirm the diagnosis. They found five chromosome deletions affecting at least six identifiable genes in people with autism in these families.

Affected genes included protocadherin 10 (PCDH10, necessary for axon growth), and c3orf58, which the study investigators now propose to call DIA1, or deleted in autism 1. The activity level of these genes decreases in response to neuronal activity. A heterozygous nonsense mutation in NHE9 (sodium/hydrogen exchanger 9) was found in two American brothers from unrelated parents; both patients suffer from autism and seizures. Rare coding changes of NHE9 were more common in North American autistic patients with epilepsy than controls or autistic patients without epilepsy, Dr. Yoo said.

Michael Greenberg, PhD, a coauthor of the paper and director of the neurobiology program at Children's Hospital-Boston and his colleagues had already identified these three genes — c3orf58, NHE9, and PCDH10 — while looking for genes that activate or deactivate in neurons in response to synaptic traffic.

Together with prior studies linking gene changes with autism, the new study suggests that a variety of brain mechanisms can be altered to produce the disorder. These now include defective cell adhesion, endosome traffic, glutamate transmission, and protein turnover.

Back to Top | Article Outline


“Using these related families is helpful; if a mutation occurs, “it will be on both copies of chromosomes,” said Huda Y. Zoghbi, MD, professor of pediatrics, molecular and human genetics, and neuroscience at Baylor College of Medicine in Houston, TX, and a Howard Hughes Medical Institute investigator. “These findings will help researchers focus on a handful of genes that are clearly causing a clinical picture that involves autism and related features, such as developmental delay and epilepsy. It is a different twist, to capitalize on the consanguinity.

“Now that you have these candidate genes, you can look at the larger population of sporadic autism and see if any have mutations in these genes,” Dr. Zoghbi continued. “It gives you an inroad to begin to explore the contribution of these genes to sporadic autism.

Figure. A

Figure. A

“The implication is that now you can begin to understand something about the biology of the disease,” she said. “It gives you pathways that could be the culprit.”

Nancy Minshew, MD, professor of psychiatry and neurology at the University of Pittsburgh, who was not involved in the research, noted that the findings dovetail with a decade or more of work that has progressively defined autism as a problem involving information processing. With improved technology for studying the brain and diagnosing autism in verbal individuals, physicians are becoming better able to narrow the focus on autism, to study how neural connections form and refine as infants develop.

Researchers now widely accept autism as a disorder of connectivity, said Dr. Minshew, who is principal investigator and director of the University of Pittsburgh's NIH Autism Center of Excellence. Hints that autism might involve excess brain connections first appeared in a publication in 1984 that documented an increase in brain weight with autism. Others found an increase in head circumference, total brain volume, and total white and gray matter in autism, with onset in the first year of life, Dr. Minshew said, but noted that other research has failed to confirm any problem with pruning in autistic children.

Back to Top | Article Outline


The new findings add to a number of genes already implicated in autism. It is not surprising that many different genes could lead to a common pathophysiology, Dr. Minshew said, given the complexity of systems circuitry in the brain. Most geneticists are holding out for a common principle: “Is there going to be a new gene defect for every family or will there be some common genetic principle that is aberrant?”

She calls the report “an interesting genetic strategy, and genetic strategies are beginning to pay off.” But Dr. Minshew hesitates to accept the suggestion by the study authors that environmental manipulation or intensive training might circumvent the gene changes.

“Children with autism are born into and grow in the same environment as their peers and neighbors. Their parents are not doing anything different,” she said. “The (investigators) have found these genes, but they are a long way from knowing how that connects to behavior, or how that relates to intervention.

“Without the work that came before it to define this disorder as one of connectivity and the work developing diagnostic methods and criteria, this study would not have been possible. These researchers stepped into the middle of a paved road and had the benefit of much technology,” Dr. Minshew said. “Right now it's a lot of conjecture, but twenty years from now, this disorder and how to approach its treatment will look entirely different.”

Back to Top | Article Outline


Morrow EM, Yoo SY, Walsh CA, et al. Identifying autism loci and genes by tracing recent shared ancestry. Science 2008;321:218–223.
    Minshew NJ. The new neurobiology of autism: cortex, connectivity, and neuronal organization. Arch Neurol 2007;64(7):945–950.
      Lander ES, Botstein D. Homozygosity mapping: a way to map human recessive traits with the DNA of inbred children. Science 1987; 236(4808): 1567–1570.
        ©2008 American Academy of Neurology