Missing Links in the Search for Causes to Autism

by Global Children's Fund

In the field of child developmental disorders, autism is the current hottest topic. Partly because of the recent debate about vaccines causing autism (they don't), and partly because its become the new ADHD; with diagnosis rates rising faster than the U.S. National debt. It can also be expensive, requiring special schooling or behavioral therapy, and taking a toll on both children and their families. This combination of public exposure and rising rates of diagnosis has left researchers scrambling to find autism's causes and, hopefully, cures.
Scientists and doctors have always suspected that biological or genetic factors might play a role, and past research has managed to turn up a few clues about genetic factors. Now a new study adds more evidence to the genetic-autism argument, by pinpointing errors in the way a child's DNA is coded that may lead to the disorder.
When our genomes are copied, it isn't a perfect process. In fact, nobody's genome is ever replicated perfectly. Sometimes we end up with deletions in certain areas, other times we end up with duplicate copies of a portion of the genome, and in other cases we inherit areas where the coding may be scrambled. These random mutations can alter our genetic expression, for better or worse, if they happen to hit along a section of the genome that is important for development.
Most of the time, such random mutations are benign. In fact, a great deal of the human genome is obsolete, encoding for things that are no longer active in human development. Sort of like your garage, the human genome is full of junk that's been collected over the years but is no longer used or needed. For example, some human beings are born with remnants of a tail, because a random copying error happens to "turn on" an area of the genome that encodes for such backside appendages-which have long since been abandoned from our lineage.
These gene alterations, known as copy number variants, have also been linked to conditions such as schizophrenia or other diseases. Previous research has also revealed that those with autism may have slightly more copy number variants than normal people. The current study uncovered no more such variants in autistic people than in non-autistic people, but discovered them in key areas among some in the autistic group. The researchers analyzed the genetic makeup of 996 people with autism and 1,287 without. More than 5,000 copy number variants were found in those with autism, usually cases in which portions of DNA were missing. Many people with autism had areas where large chunks of DNA was missing.
You and I likely have just as many errors or deletions in our genome, but since they don't hit important genes that code for the development of things important to language or social interaction, we don't develop autism.
By themselves, each specific variant was fairly rare, with even the most common one occurring in no more than 1% of the people in the study. Each person analyzed seemed to have a distinct set of genetic variations, suggesting that every person may have their own genetically unique version of the disorder. Though varied in nature, the affected genes all tended to influence similar biological processes: particularly those involved with brain development and functioning.
In some cases, the deleted genes had a strong link to autism: just a single copy error was enough to push a person over the autism threshold. Others had to be inherited together along with other copy number variants in order for autism to develop. One gene with an especially strong link to autism was DDX53-PTCHD1, which is located on the X chromosome. Some women naturally carry a deletion of this gene on one of their X chromosomes. This doesn't doom their children to autism, because such women will almost always carry a healthy version of the gene on the other X chromosome, so the child inherits a healthy copy. But if a random coding error occurs, deleting this area of the healthy gene, the child is left with no healthy copies to draw from and will develop autism.
The group was also able to identify several previously unknown areas of the genome that seem to be involved in the formation of connections between brain cells. All together, 25 places were identified in the present study that may be linked to autism. The hopes are that this information might be used for earlier diagnosis, or down the road, possibly even to treat the condition through some type of gene therapy. Yet such things are still a ways off. Scientists need to better understand how all these factors work together to cause the disorder before any definitive diagnostic tests or gene therapies come about.
Lest you believe that the source of autism has now been explained, think again. Even with these newly discovered genetic markers, scientists are still only able to pinpoint genetic causes in about 10% of autism cases overall. "What causes the other 90% of the cases is still on the table," says geneticist Steve McCarroll of Harvard Medical School, in an interview with Science News.
Many problems still exist in the study of autism, not the least of which being that we still can't pinpoint precisely what autism is. What we group under the generic label of autism is in all likelihood not a single condition but a broad array of different disorders with multiple causes that tend to result in similar symptoms. This is why the diagnosis is officially labeled as "autism spectrum disorder." Each of these variants along the spectrum may be its own unique condition with its own unique cause.
It's not the miracle parents of autistic kids are hoping for, but every little advancement helps. If nothing else, this study shows the complexity of the problem, and how naive it is to try and blame autism on a quick and easy scapegoat such as vaccines.