Autism: Lots of clues, but still no answers
14 May 2005
New Scientist Print Edition.
THE risk of autism in twins appears to be related to the month they are born in. The chance of both babies having the disorder is 80 per cent higher for January births than December births.
This was one of the many findings presented at the conference in Boston last week. It typifies the problems with many autism studies: the numbers are too small to be definitive - this one was based on just 161 multiple-birth babies - and even if the finding does stand up, it raises many more questions than it answers. For instance, is infection during pregnancy to blame? Or another environmental factor that varies with the season? Is it the birth date that is important, or the date of conception?
While it is now clear that autism is a complicated combination of genetic and environmental factors, no one knows exactly what these environmental effects are and which genes they might be interacting with to cause autism. "We clearly have a daunting task ahead of us," Craig Newschaffer of the Johns Hopkins School of Public Health in Baltimore, Maryland, and a member of the birth-date study team, told the conference. "We have had slower-than-hoped for progress finding autism genes, and now we see an increased focus on the environment in autism."
Based on twin studies, the consensus is that, depending on how you define autism, between 60 and 90 per cent of cases are genetic in origin. And although progress has been slow, geneticists are homing in on some of the genes involved.
Eric Geschwind of the University of California, Los Angeles, announced that he has linked a region on chromosome 17 to autism, confirming earlier studies. Other recent studies have linked mutations in specific genes to some cases of autism, including two genes that code for brain proteins called neuroligins, which are involved in the transmission of nerve impulses. Further emphasising the genetic basis of the disease, Geschwind has also found that although people with autism have larger heads than the population in general, on average their heads are no larger than those of siblings and parents without the disorder.
“What if you have an infant with some genetic susceptibility to autism and then you add a toxin? It looks very disturbing”
One of the reasons progress has been slow is that many genes appear to be involved. Some people with autism have mutations in one gene or set of genes, and others have mutations in different ones. That makes it hard to carry out the statistical studies needed to pinpoint the genes involved. Geschwind thinks that the best way forward is to first identify smaller, measurable components of the disorder, and then to look for the genetic underpinnings of each one.
But it is not just about genes. "The genetics is very compelling and explains most of the risk, but there is room for the environment since genetics cannot explain it all," Geschwind says.
“The genetics explains most of the risk, but there is room for the environment since genetics cannot explain it all”
Identifying which genes and which environmental factors are interacting presents enormous challenges, says Ellen Silbergeld, an epidemiologist at the Johns Hopkins School of Public Health. "One real problem is we don't have a good idea of the disease's history," she says. She points out that our understanding of heart disease was greatly helped by the Framingham Heart Study, which followed around 5000 individuals for up to 50 years.
Autism not only lacks a similar study, but also presents more challenges, because fewer people develop autism than heart disease, and because it might be not one but multiple disorders with related sets of symptoms. "Are we talking about 10 different diseases?" Silbergeld asks.
Complicating the story still further is the apparent rise in autism. The Autism Society of America estimates that there was an increase of 172 per cent in the US in the 1990s, though the population grew only 13 per cent. It is still not clear if the rise is due to some new, presumably environmental, factor or if doctors are simply becoming better at diagnosing the disorder. But that has not held people back. "We are going through a frenzied period," says David Amaral at the University of California, Davis. "People speculate wildly about the things that could cause autism."
Some have focused on exposure to viruses or vaccines, others on pollutants such as mercury. Emitted by factories and power stations, some forms of mercury build up in animals such as fish and enter the food chain. In 2003, a small study by paediatrician Amy Holmes found much lower levels of mercury than normal in the hair of children with autism (New Scientist, 21 June 2003, page 4). Her hotly contested theory is that they have an impaired ability to excrete mercury.
In Boston, Jim Adams, a chemical engineer at Arizona State University in Tempe who has an autistic daughter and believes that mercury causes many cases of autism, presented results supporting Holmes's theory. He found that children with autism have up to three times as much mercury as normal in their baby teeth, yet lower levels in their hair.
And in December last year, Jill James of the University of Arkansas published a study showing that people with autism have raised levels of glutathione, a compound that detoxifies mercury. But as so often with autism, far more studies are needed before any conclusions can be drawn.
Others are looking at polychlorinated biphenyls, or PCBs. Banned in most countries since the 1970s, PCBs still persist in the environment and in animals and people. A team led by Tal Kenet of the University of California, San Francisco, found that one type of PCB can devastate the auditory cortex in rats, the brain area that processes sound signals according to frequency.
Some people with autism have abnormal auditory cortices and difficulty communicating verbally, despite having perfect hearing on regular hearing tests. Kenet speculates that this could be linked to exposure to PCBs or other toxic substances. Although it is far too early to confirm the link with autism, no one had ever looked at this before. "What if you have an infant with some genetic susceptibility to autism and then you add a toxin? In all it looks very disturbing," she says.
There could be a whole host of chemicals that wreak havoc on the brains of genetically susceptible individuals, team member Isaac Pessah of the University of California, Davis, told the meeting. He points out that according to the US National Toxicology Report, between 60 and 80 per cent of the chemicals we are exposed to through pesticides, cosmetics and foods have not undergone adequate risk assessment.
From issue 2499 of New Scientist magazine, 14 May 2005, page 14