to BioTechniques free email alert service to receive content updates.
Unraveling autism one de novo mutation at a time
 
Sarah A. Webb, Ph.D.
BioTechniques, Vol. 53, No. 3, September 2012, pp. 133–136
Full Text (PDF)

Affecting 1 in 88 children in the US, autism spectrum disorders represent a challenge for children and their families, as well as researchers struggling to understand the genetic basis of autism. But as genome sequencing improves and costs plummet, this technology is beginning to offer opportunities to probe the genetics of autism at a deeper level. In fact, earlier this year, four groups published studies (1-4) using exome sequencing to hunt for rare de novo mutations in children with autism. Identifying these rare mutations is important as they may occur in as many as 20% of individuals with autism. Though new treatments and diagnostic tools are still a long way off, “[these studies] show a clear path forward for gene discovery and autism, a systematic way to identify specific genes that are contributing to risks,” notes Matthew State of the Yale Child Study Center, a principal investigator on one of the studies published in Nature(1). Even as this article went to press, another paper in Nature(5) connected the number of de novo mutations in children with the father's age at conception, a finding that could have implications for the risk for autism in children.

The modern genetics of autism

Researchers have long recognized a link between genes and autism. Individuals with certain inherited single gene mutations, such as those that cause Fragile X syndrome and tuberous sclerosis, have highly increased incidences of autism as well. But the majority of children with autism do not have autistic parents, and the combination of rare genes and the heterogeneity of the disorder parallels other neurodevelopmental and neuropsychiatric disorders, says State.

One approach to understanding autism genetics has been through the use of genome-wide association studies (GWAS) that search for changes in common genes among larger numbers of individuals that might confer risk for autism. The trouble is that searching through GWAS data is like looking for a needle in a haystack. If you imagine genetic variation as a pyramid, GWAS mostly looks at the wide bottom, sifting through the largest numbers of genes, says Ben Neale of Massachusetts General Hospital and the Broad Institute and first author of the second study in Nature(2). The end result is that GWAS can often miss de novo mutations.









De novo mutations are not present in the genomes of parents, but rather arise within egg or sperm cells, leading to genetic changes in their children. The vast majority of these rare mutations have no pathological effect. Limiting searches to this much rarer set of mutations provides a kind of filter that reduces hit numbers in a screen, a way of looking at just the tip-top of that genetic variation pyramid. In 2007, a team led by Cold Spring Harbor Laboratory scientists Jonathan Sebat and Mike Wigler showed for the first time that de novo copy number variations could have implications in autism spectrum disorders (6). Ten percent of autism subjects had de novo mutations compared with only 1% of controls. “That's where the interest in de novo mutations really started,” says Dan Geschwind of UCLA, and a co-author on State's Nature paper.

If researchers locate a de novo mutation in a functional gene of a child with autism that doesn't occur in the parents or in an unaffected sibling, there exists the possibility that this gene could confer a greater risk of autism. But if that same mutation is identified in two unrelated individuals, both with autism, that confidence level increases nearly to 99.9%, says Michael Ronemus of Cold Spring Harbor Laboratory and a co-author on the paper published in Neuron(4).

Initially, researchers used microarrays in their de novo mutation studies to examine large-scale mutations on the order tens of thousands of base pairs. But even though identifying these copy number variations represented progress, such large segments often include multiple genes, says State. To really understand the neurobiology, researchers would like to examine the single gene level to investigate the connection between a specific mutation and the resulting pathology of a complex neuropsychiatric disorder. Today microarrays have a partner in the hunt for de novo mutations as advances in high-throughput sequencing and the decreasing cost associated with this technology now allow researchers to take a higher-resolution look at these genetic variants, providing the ability to identify smaller scale mutations that change protein function. In addition, it turns out that large-effect mutations are also incredibly rare, which means that researchers can use targeted sequencing to sample particular genomic regions in tens of thousands of cases and controls to identify differences.

  1    2    3