WEDNESDAY, April 4 (HealthDay News) -- Genetic mutations that arise spontaneously, as opposed to being passed through generations, could play an important role in the development of autism, new research suggests.

Three research teams sequenced the genes of children with sporadic autism, meaning it did not run in their families, and compared the sequences with those of their parents and siblings. Their results were published in three separate articles on April 4 in the journal Nature.

"We found that 15 to 20 percent of sporadic patients could be explained by 'de novo' [or spontaneous] mutations," said Evan Eichler, professor of genome sciences at the University of Washington in Seattle, who led one of the studies.

Overall, the researchers identified hundreds of spontaneous mutations in gene sequences that they predicted would upset the function of the genes in the children with autism. Most of them were only found in single patients.

The good news, according to Eichler, is that the multitude of affected genes seems to belong to just a handful of pathways -- involved in, for example, development or cognition. That suggests that mutations in a diverse set of genes could have a similar biological effect.

"This kind of investigation is of tremendous value for understanding the genetic architecture of risk for autism," said Andy Shih, vice president of scientific affairs at Autism Speaks, a national advocacy group.

"We can probably explain genetic risk factors that might lead to autism in less than 30 percent of the population" from previous research, Shih added.

Autism spectrum disorders, which include both mild and serious forms of autism, affect one in 88 children in the United States, according to just-updated statistics from the U.S. Centers for Disease Control and Prevention.

The study led by Eichler involved 677 individuals representing 209 different families, each with one child who had sporadic autism. The researchers found a total of 126 spontaneous mutations that they predicted would have a severe effect on the genes in which they occurred.

Eichler's team relied on samples from the Simons Simplex Collection, a nationwide project that gathers blood and DNA from children with sporadic autism and their unaffected family members.

A second study, led by researchers at Yale University in New Haven, Conn., looked at 238 families from the Simons project, some of which overlapped with Eichler's study. They identified 125 de novo mutations that would change the readout of genes among the children with autism and 87 among their unaffected siblings.

The third study in the trio found that just under half of children, both with and without autism, had readout-altering spontaneous mutations, but that the mutation rate was comparable between children with autism and their unaffected siblings. This study was led by researchers at Harvard Medical School, Massachusetts General Hospital in Boston and the Broad Institute in Cambridge, Mass., and involved 175 sets of children with autism and their parents.

Altogether, this body of research suggests that the frequency of de novo mutations is not significantly higher in children with autism, but that the types of mutations, which occur by chance, are more detrimental in children with autism than those in their unaffected siblings, Eichler said.

Eichler's team also found that de novo mutations were four times as likely to lie on DNA strands inherited from the father, and that the number of mutations increases with paternal age.

The possible paternal influence suggests that many of the de novo mutations originate in the father's sex cells, which give rise to sperm. Presumably, mutations would be more likely in paternal sex cells rather than maternal ones because paternal cells continue to divide throughout a man's lifetime, giving them more chances to pick up mutations.

Although the link with paternal age agrees with some epidemiological studies that have found higher rates of autism among children of older parents, it probably only plays a part in a modest 10 percent or so of patients, Eichler said.

From the three studies, mutations in two genes, called CHD8 and KATNAL2, emerged as likely autism risk factors because they were found in more than one patient. "Almost never did we see lightning strike the same place twice," Eichler said.

Unlike the myriad mutations identified in genes involved in neuronal development, these mutations could have more universal effects on regulating gene expression, cell growth and differentiation.

These two genes were also among the 49 genes that fell into the same biological pathway, Eichler's team found. "This one is a monster pathway," Eichler said, because it involves the largest number of identified de novo mutations.

The gene mutations identified in this research "underscores that autism is a complex interplay between genes and the environment," Shih said. For example, CHD8 can control the expression of other genes in response to environmental stimuli.

This research is also a reminder that autism is a group of related disorders involving many genes in different pathways, he added. "Each of the genes [in these studies] seem to confer only a small risk, and are only readily found in a small percentage of individuals with autism," he noted.

However, Shih said, these studies show that with more genetic analyses involving more patients, "there could be some unifying principles revealed that could allow identification of individuals at risk of autism and guide therapeutics."

More information

To learn more about autism, visit the Autism Society of America.