Autism is often described as a spectrum – varying from severe syndromes in which the symptoms of autism are only part of the problem to the more classic autism characterized by impaired social skills, delayed language development, and repetitive or stereotyped behaviors. The genetic origin of the disorder may also occur along a spectrum.
On one hand, a severe mutation affecting a particular gene gives rise to syndromic autism, a form of the disorder that occurs along with other problems including physical and facial abnormalities, including Rett, Angelman and Fragile X syndromes.
On the other hand, a new study from researchers at Baylor College of Medicine and the Jan and Dan Duncan Neurological Research Institute (NRI) at Texas Children's Hospital proposes that several subtle mutations in different genes are associated with patients with near-normal or normal IQs and non-syndromic autism defined only by the behavioral characteristics. A report on the work appears in the current issue of Human Molecular Genetics.
High rate of mutations in autism group
The study, conducted by Dr. Christian Schaaf, assistant professor of molecular and human genetics, Dr. Aniko Sabo, assistant professor in the Human Genome Sequencing Center at BCM, and colleagues at BCM and Jan and Dan Duncan Neurological Research Institute at TCH, involved two steps. First, researchers sequenced 21 genes known to be associated with syndromic autism in 339 people who had the non-syndromic form of the disorder. They found a high rate of de novo or new mutations of the genes in the autism group as well as many cases in which individuals with autism inherited single mutated copies of two or more of the genes from their parents. They might inherit a single mutated copy of one gene from their mothers and a single mutated copy of another gene from their fathers.
In the second step, when researchers compared the genetic sequence of seven genes associated with syndromic autism in the autism patients to that of individuals without autism, they found that the rate of variation or subtle change in genes in the normal population was much lower.
Accumulated mutations can cross threshold to cause disease
The researchers call this "oligogenic heterozygosity." This means that individuals with autism may inherit two or more genes with subtle mutations. Schaaf hypothesized that the accumulation of many subtle mutations ultimately crosses a given threshold and results in autism.
"This study represents a comprehensive evaluation of the variants in the coding regions of 21 genes previously implicated in autism. Many previous studies have focused on a gene-by-gene approach, looking at large effects from a single gene, but our study suggests that groups of variants with smaller effects in multiple genes acting together might be contributing to autism," said Sabo. "Although the number of genes interrogated in this study was limited, our findings point to the importance of evaluating interactions of multiple genes in more complete datasets." Two of these larger studies are underway at BCM and the NRI.
By drawing upon advances made in syndromic autism to understand the causes of non-syndromic autism, the researchers' findings may fine-tune the focus of autism genetics from individual genes to pathways.
Many genes can contribute to autism
"People have been really humbled by the fact that there are so many autism proteins," Schaaf said. "However, a lot of these genes might still act in the same pathways or cellular subcompartments, so if we can identify four or five main autism pathways in each of which there might be 10 to 15 genes or proteins involved, then those pathways might be amenable to therapy."
In a preview in the journal Neuron published today (June 9), Schaaf and Dr. Huda Zoghbi, director of the Neurological Research Institute, explore the current state of research into the genetic origins of autism spectrum disorders, underscoring the premise that many rare variants – either gene alterations or structural genetic changes called copy number variations – have been identified in studies of people with autism across the spectrum. Citing the findings in many papers, including this one and one appearing June 8, 2011 in Science Translational Medicine, they suggest an overlap between the genetics of syndromic and non-syndromic autism.
Pathways may provide therapeutic targets
"Identifying functional relationships and interactions between various ASD (autism spectrum disorder)-associated proteins is likely to identify signaling pathways and subcellular compartments that encompass a whole subgroup of such genes. Having such rich functional pathway information might unearth common targets that are amenable to therapy," the authors wrote.
Others who took part in this research include Yasunari Sakai, Donna Muzny, Alicia Hawes, Lora Lewis, Humeira Akbar, Robin Varghese, Richard A. Gibbs and Zoghbi, all of BCM, in collaboration with Jacy Crosby and Eric Boerwinkle from The University of Texas School of Public Health at Houston. Gibbs and Zoghbi are corresponding authors of the report.
Funding for this work came from the Simons Foundation Autism Research Initiative (SFARI) and the Howard Hughes Medical Institute