Study clarifies link between 15q duplication and autism
Children who carry an extra copy of the 15q11-13 region of the genome usually have autism and sleep troubles, as well as distinctive brain-wave patterns and facial features, according to a report published 14 March in Autism Research1.
The study is the largest to date characterizing the effects of a rare type of duplication in this region. This large chromosomal region, dubbed '15q,' is the most common site for autism-related DNA deletions and duplications.
It's also famous for its complexity: Losing 15q on the maternal chromosome leads to Angelman syndrome, with developmental delay, seizures and a happy demeanor. Deletion of the paternal copy causes Prader-Willi syndrome, characterized by intellectual disability and obesity.
Carrying duplications in the region is a strong risk factor for autism. These duplications come in two types. Most common is 'isodicentric,' which stems from an extra chromosome and leads to two extra copies of 15q. The new study focuses on the rarer type, called 'interstitial' duplication, which occurs within a chromosome and results in just one extra copy of the region.
Children carrying interstitial duplications tend to have milder symptoms than those with isodicentric duplications, making it all the more difficult to find them.
Larry Reiter and his colleagues at the University of Tennessee Health Science Center evaluated 14 children aged 3 to 16 with interstitial duplications, making the study the largest published sample of this genetic abnormality. Previous reports were case studies of just one or two children.
"It’s important because this is a disorder that hasn’t been that well characterized," says Janine LaSalle, professor of medical microbiology and immunology at the University of California, Davis, who was not involved in the study. Still, she says, the findings leave unanswered many questions about the complex syndrome.
"It helps clear some things up, but it's not the easiest genetic disorder to study," LaSalle says. "We want it to be simple, but it's not."
Reiter's team recruited the children from across the country and performed all of the clinical tests in Tennessee. These included a genetic screen, two standard diagnostic tests — the Autism Diagnostic Observation Schedule (ADOS) and the Autism Diagnostic Interview-Revised (ADI-R) — and an electroencephalogram (EEG), which uses electrodes on the scalp to measure brain waves.
For ten of the children, the duplication came from their mothers. Nine of these children have autism and one was not tested.
The fact that at least nine of the ten children with maternal duplications have autism "is huge," says Reiter, associate professor of neurology at the University of Tennessee.
One reason that's striking is because most deletions and duplications linked to autism have partial penetrance, meaning they sometimes lead to the disorder and sometimes don't.
Given the new results, "I can’t think of a more penetrant cause of autism," says Ed Cook, professor of psychiatry at the University of Illinois at Chicago, who was not involved in the study. Even if 15q's penetrance doesn't turn out to be 100 percent, Cook adds, "it's quite a bit higher than anything else that I’ve seen in the literature."
This gene is paternally imprinted, meaning that the paternal copy is silenced. If UBE3A were the primary cause of autism symptoms in the 15q duplication syndrome, you would expect it to affect only the children whose duplication came from their mother.
It's not quite that simple, however.
Of the four children in the study whose duplication came from their father, two have autism and two do not. Because of the small numbers, "we can't quite get a handle on the paternals yet," Reiter says.
It may be that epigenetic influences, which affect the expression of genes without changing the underlying DNA code, are somehow un-silencing the paternal copy of UBE3A. "My guess is there's an epigenetic explanation," LaSalle says.
LaSalle's lab has focused on chromatin, which includes the proteins that help package DNA in the nucleus, in this region. "There's bizarre chromatic dynamics going on where the maternal and paternal chromosomes seem to communicate with each other," she says. "Having too many partners in the dance seems to alter the whole locus."
It may also be that one or more of the two dozen genes in the 15q region are driving the autism symptoms.
For example, the region holds several genes that code for parts of the gamma-aminobutyric acid (GABA) receptor, a protein that helps dampen brain signals. Some studies have identified people with autism who carry rare genetic variants in these GABA genes.
The EEG results also implicate GABA. Of the 14 children in the study, 10 show a distinctive pattern in their EEGs: spikes of so-called 'beta' brain waves, which move at 18 to 22 cycles per second. The same pattern is seen in the brains of children who are taking benzodiazepines, drugs that activate GABA receptors in the brain.
The unusual EEG patterns may be caused by the over-expression of the GABA-related genes in the 15q region, notes Gregory Barnes, director of the pediatric epilepsy monitoring unit at Vanderbilt University in Nashville, Tennessee. But it may also be a result of excess UBE3A or other genes in the region, he says.
Last year, Barnes and his colleagues reported that children with Angelman syndrome show a variety of abnormal EEG patterns, though they did not observe excess beta waves2.
Reiter says he plans to continue to study children with interstitial duplications. He aims to scan their brains using magnetoencephalography, or MEG, and investigate whether there are links between the size of their duplications and their expressive language skills.
"For me, it's important to understand phenotype-genotype correlations," Reiter says. "What is it that UBE3A does? What kind of autism is it?"
News and Opinion articles on SFARI.org are editorially independent of the Simons Foundation.
1. Urraca N. et al. Autism Res. Epub ahead of print (2013) PubMed
2. Vendrame M. et al. Epilepsy Behav. 23, 261-265 (2012) PubMed