Understanding Autism

Lisa Bond spent 11 frustrating years trying to figure out why her son was the way he was. Joshua was unusually awkward, avoided eye contact, was delayed in talking, struggled mightily in school, and screamed bloody murder when his socks were changed. Doctors kept dismissing Bond’s concerns as the obsessions of an over-anxious mother.

Finally, at age six, Joshua underwent a thorough physical and psychological assessment. The diagnosis: autism. It was a difficult diagnosis for Bond and her husband to hear, but at least the family now had an answer to what Joshua’s disorder was. Still, no one could tell them why he had it. Like many parents, Bond blamed herself for her son’s troubles. Perhaps her prenatal diet hadn’t been quite perfect, she thought, or maybe she’d caused some damage when she’d slipped on the ice a month before his birth.

Still seeking answers, the family enrolled in an autism study at U of T. When Joshua was almost 12, they all agreed to undergo genetic testing, and willingly gave blood samples. The results were both surprising and reassuring: the boy had a genetic glitch on chromosome 16. Neither the parents nor Joshua’s big sister had this alteration; Joshua’s disorder seemed to be no one’s fault but the result of a random roll of the genetic dice that had likely happened at conception. And that’s when his mother finally breathed a sigh of relief. “It was just like a huge weight had been lifted off me,” she says.

Thanks to the blistering pace of genetic research, families like the Bonds will no longer have to wait years for answers, or even months. “We’ll be able to identify these kids on day one, or even prenatally,” says geneticist Stephen Scherer, director of the McLaughlin Centre at U of T and the Centre for Applied Genomics at the Hospital for Sick Children. Scherer is also a professor of medicine at U of T and one of the world’s top autism researchers. As things stand now, most autistic children in Canada aren’t diagnosed until between ages three and six – too late to get the maximum benefit from crucial early-intervention programs. Experts say early diagnosis could lead to closer monitoring of symptoms, earlier access to beneficial programs (such as for speech and social skills) and brighter outcomes. Not only that, but based on genetic research, pharmaceutical companies are working on developing new drugs for autism, now that they finally have an idea of which neurobiological processes to target. “Our genetic advances have cracked open this black box,” Scherer says.

Autism has indeed been a black box – in truth, almost an impenetrable vault, with autistic individuals locked inside and their families shut out. Increasingly called autism spectrum disorder (ASD) to reflect the diversity of symptoms and abilities, it’s more common than childhood cancers, juvenile diabetes and pediatric AIDS combined, yet has been largely a mystery (see “The ABCs of Autism“).

When Scherer, 47, started his own independent laboratory 15 years ago, he was anything but an authority on the subject. “I’d kind of heard about autism, but to be perfectly honest I didn’t know what it was,” he acknowledges. At the time Scherer, fresh from earning a PhD in genetics from U of T, was contributing to the Human Genome Project, the international study that was in the process of generating maps and sequences of all 20,000-plus genes on the 23 pairs of human chromosomes. Scherer was a world expert on chromosome 7.

Within a single week, two seemingly coincidental things happened: Scherer received a letter from a California woman whose autistic son was shown to have a swapping of genetic material from chromosome 7 to chromosome 13; and a colleague of Scherer’s at the University of Oxford in England published an autism study on a certain region on chromosome 7, which happened to be the exact region where the California boy’s translocation had occurred. Could this genetic change be a clue to the elusive cause of autism? No one in Toronto was investigating the link between genes and autism, so Scherer took the plunge. “I was essentially thrown into the pot, and it was just fascinating to me,” he says. The more he read about autism, the more he realized it was all around him. One of his best buddies from high school had an autistic brother, although the term was never used. Among the 40 children in his own six-year-old son’s playgroup, four have ASD.

Previous studies of twins suggested that autism had genetic roots, but no one knew which genes were involved. The long-held theory was that humans all have largely the same genetic profile, and that those with a given disorder such as ASD inherited common mutations in the same few genes. It was also assumed that everyone inherits two copies of every gene – one copy from each parent. But Scherer suspected that these theories didn’t tell the whole story. Since people with ASD often remain single and childless – very few form close relationships, get married, create families and pass on their genes – it seemed that at least some mutations must be arising spontaneously, and that maybe more than just a few genes were involved.

Scherer became part of the international Autism Genome Project, a consortium that would ultimately include groups in 11 countries, mostly in North America and Europe, to analyze the genes of 1,600 families worldwide who have at least two members with autism. Scherer adopted what he calls his “garbage-can approach”: he started looking closely at the data that other scientists tended to discard as insignificant. When he compared gene sequences from autistic people with the control group, instead of looking solely for predictable large differences, he began noticing tiny differences that seemed to arise spontaneously. The group found that several regions of the genome, particularly involving chromosome 11, were highly associated with autism.

Even more exciting – instead of having the usual two copies of genes, some people had only one, while other people had three. This suggested that some of the key directions in the body’s “instruction manual” were missing or duplicated, which would affect development. Scherer and his co-discoverer, Charles Lee of Harvard Medical School, dubbed these differences copy number variants, or CNVs. Since one particular gene, neurexin 1, plays a major role in determining how nerve cells in the brain communicate, it made sense that CNVs of this gene might be a cause of autism. This didn’t constitute proof, but it was certainly a smoking gun.

Further, the researchers reported that various types of CNVs are present in all of us, and that in fact they’re the most common type of human genetic variation. Our own individual collections of CNVs, they surmised, contribute not only to our physical and mental development but to our personal uniqueness, as well as offer clues about development of disease and human evolution. “We were the first in the world to find these new genetic variations in everybody,” Scherer says. “It was a real eureka moment.” When the study was published in 2007, it generated worldwide publicity. The international journal Science named the copy number variant theory the “breakthrough of the year.”

Scherer’s research has huge implications for the doctors who see autistic patients. “Three or four years ago we were ready to throw up our hands and say we’ll never find an answer,” says Wendy Roberts, co-director of the Autism Research Unit at SickKids and a professor of pediatrics at U of T. “But these are some of the most exciting and important autism studies ever done.” One of Roberts’ jobs is diagnosing ASD in children, but it can take up to three or four years to see a specialist and begin intervention. The kids are often four or five by the time they get publicly funded treatment. “Having to wait that long for help is criminal, considering you’ve got the most chance of making a difference early on,” she says.

Roberts, a developmental pediatrician, notes that if ASD is diagnosed in an infant, research has shown that early intervention by therapists and parents can “push” the child into enjoying certain activities, or “pull” him away from his obsessive preoccupations. “We’ve got an amazing ability to affect expression of genes,” she says, “and we believe that with early intervention we’re pushing the best possible gene expression at every stage.” By the time many kids are diagnosed, that window has already started to close, she says, when the “hard-wiring” sets in and some genes are turned off for good.

Building on its findings, Scherer’s lab published a study in 2008 that described CNVs found on other genes related to autism. The work also identified a new ASD-related region on chromosome 16. With improvements in the technology they were using to analyze the genes, the researchers kept finding more and more genes involved, on several different chromosomes. In 2010 the consortium published another paper, this time in the prestigious journal Nature. The largest- ever study of its kind, it identified even more genes involved, including SHANK 1 and 2, neuroligin 3 and 4 and patched-related 1. In total, the identified genes account for at least 15 per cent of autism cases. Scherer says the research is moving so quickly that within five years they’ll have identified all the genes involved – he estimates there are more than 100 – and that these will ultimately account for at least 90 per cent of all autism.

After most of the genes have been identified, Scherer says researchers will have to decide on a new direction – perhaps investigating how the environment affects the proteins that influence how brain cells develop and interact. “There certainly could be some environmental agents that trigger genetic changes or alter how the proteins work in the cell, but we haven’t found those yet,” Scherer says. Whatever these environmental culprits may be, they’re likely not childhood vaccines, as some people believe. The 1998 British study that linked childhood vaccines to autism was retracted last year for being incorrect and dishonest, and numerous studies since have shown no increased risk at all. (Scherer had his own two children fully vaccinated.) It is known that chemical compounds such as valproic acid and thalidomide have been linked to autism. But Scherer says it’s necessary to identify all the ASD genes first, before investigating how the environment may affect them.

Here’s another tricky thing about those genes: even if the same genes are involved in many people, ASD individuals could each have their own personal combination of genetic glitches – and therefore, their own unique version of autism. The genetic variations don’t become common in the population because they’re not passed on by the oftenchildless people with ASD. “They kind of go extinct,” Scherer says. “So with this whole concept of rare genetic variations, the studies suggest that each individual who has a genetic form of autism has their own specific genetic form.” It certainly would explain why no two people with ASD are alike.

But if everyone has his or her unique syndrome, wouldn’t that make it almost impossible to come up with drugs and other therapies? Not at all, Scherer says. So far, the genes identified seem to involve the same biological networks in the brain, which suggests common areas for drug makers to target. “My dream is that there will be a pill that can help alleviate at least a few of the core deficits in autism in some individuals, and maybe all of the deficits in others,” he says.

Funding for Scherer’s research has come from various international public sources, including Genome Canada, the Ontario Genomics Institute and the U.S. National Institutes of Health, along with several private sources. Last summer Scherer’s lab received $8 million to sequence the genomes of 1,000 ASD children in Ontario, to discover new genes and to help families get into assistive programs, with a goal to ultimately testing all autistic children in Ontario. Right now, parents are lining up to be part of the study. “To see it all come together is so satisfying,” says Scherer. “But I’m not going to retire until we find a cure.”