BALTIMORE—Despite the fact that genetic causes are suspectedin 90 percent of the disorders that are on the autism spectrum, identifying theresponsible genes has proven tricky. However, an international group ofresearchers, led by a team at Johns Hopkins University, has added considerablyto the genomic information in this area by identifying several genetic links toautism, chief among them a variant of semaphorin 5A, whose protein productcontrols nerve connections in the brain.
Publishing in early October in Nature, the researchers note that semaphorin 5A was alreadybeen known to play a role in guiding neurons to the right connecting points inthe brain during fetal development. In addition, the team reported additionalevidence that many rare and common genetic variants contribute to autism.
It is tempting, of course, with a breakthrough like this,for people outside of the research community to get ahead of themselves and seepossibilities for predictive or diagnostic tests, or for treatments in the nearterm, notes Dr. Aravinda Chakravarti, professor of medicine, pediatrics, andmolecular biology and genetics at the McKusick-Nathans Institute of GeneticMedicine at Johns Hopkins. But this is just an important early step inunderstanding autism, he adds.
"As a society, we've gotten to the point where we want the'instant coffee'—one of the worst inventions that human creativity has everproduced," Chakravarti says. "But we cannot have good therapy, or diagnostics,or anything else like that, without understanding what it is that we aretargeting and why. We really don't know a lot of details right now, but itseems clear we're on a promising track, and I have no problem being patient andfinding out more about something we don't know much about, and wait on thediagnostics or therapies much farther down the line."
In fact, it isn't at all clear that semaphorin 5A or any ofthe other genes being investigated would even be good diagnostic or therapeutictargets; only that they are key players apparently, Chakravarti says.
He likened the current research findings to baseball, thoughhe jokes that he hates making a sports analogy out of it, saying, "We're not atthe World Series yet with this. Maybe we've just hit a home run in the sixthinning of the tenth game in the regular season. That's important, but it's notthe end of the journey, and many people will build off what we've done and gooff in their own directions as well."
In the current work, the researchers verified thatsemaphorin 5A plays a role in autism by looking at brain tissue samples fromthe Autism Tissue Program and the Harvard Brain Bank. In doing so, they foundthat the amount of the semaphorin 5A protein was significantly reduced in thebrains of those people with autism compared to brains of the non-autistic.
This suggests that autism may occur due to differences inhow nervous system connections are made in the brain, the researchers note.
The identification of semaphorin 5A, and seven additionallikely—but not yet proven—genes involved in nervous system development, cellstructure and other cell functions, was possible only because of an "extensivecollaborative effort that included inpatient samples from the Autism Consortiumin Montreal, the Autism Genome Project and additional samples from Finland andIran," the Johns Hopkins team reported when it announced the findings.
"The biggest challenge to finding the genes that contributeto autism is having a large and well-studied group of patients and their familymembers both for primary discovery of genes and to test and verify thediscovery candidates," Chakravarti notes. "This latest finding would not havebeen possible without these many research groups and consortia pooling togethertheir patient resources. Of course, they would not have been possible withoutthe genomic scanning technologies either."
Using such genome-wide scanning technologies, the teamexamined the genomes of family members, among whom more than one individual wasdiagnosed with an autism spectrum disorder. They studied 1,031 nuclear familiesin total, and 1,553 affected children, looking at 500,000 single nucleotidepolymorphisms for differences that stood out in the DNA. The information wascollected from two sources: the Autism Genetic Resource Exchange and theU.S.-based National Institute of Mental Health.
They performed two types of genetic analyses to identifygenes with both rare and common genetic variations that might contribute toautism. By studying siblings with autism, the authors discovered four regionsof the human genome—on chromosomes 6, 15, 17 and 20—where rare variants in yetunidentified genes appear to contribute to autism susceptibility. Afterexamining the genetic patterns in unrelated people with autism, the researchersdiscovered a common variation near only one gene, and that was semaphorin 5A.
The fact that only one common variant came out of this largesample study indicates that "there probably are many more that contribute toautism, but none have large effects," says Dr. Dan Arking, assistant professorof medicine at Hopkins' McKusick-Nathans Institute of Genetic Medicine."Alternatively, there may be numerous rare gene-variant containing genes."
Identifying these will require even larger collaborativestudies, some of which are already in progress, he adds.
Next steps in the research will be to continue to poolcurrent and future data with that of other groups with similar genome-widework, not just with semaphorin 5A but also with the other four genes the JohnsHopkins-led team discovered, and others that may be uncovered later by them orsomeone else.
"Thesediscoveries are an important step forward, but just one of many that are neededto fully dissect the complex genetics of this disorder," says Dr. Mark Daly, asenior associate member at the Broad Institute of Harvard and MIT and anassociate professor at the Center for Human Genetic Research at MassachusettsGeneral Hospital. "The genomic regions we've identified help shed additionallight on the biology of autism and point to areas that should be prioritizedfor further study."