[1] 2 Next »
March/April 2009
A Hole in the Genome
A small chunk of DNA linked to schizophrenia, mental retardation, and autism may change the way we think about disease.
By Emily Singer



Credit: Martin O’Neill
Go about 145,000,000 bases (or "letters") down the long arm of chromosome 1 and you'll come to 1q21.1, the genetic address of a small but important piece of DNA that is particularly prone to mistakes. When chromosome 1 is duplicated during normal cell division (say, in creating sperm or eggs), short, repetitive bits of DNA within this stretch are all too likely to mistakenly pair up, raising the chances that the new cells will have extra or missing copies of specific pieces of DNA.
Those small mistakes can have a big impact on people who carry them. Several studies in the last year have found that missing or extra pieces of DNA in the 1q21.1 region put the bearer at risk for a surprisingly broad range of psychiatric and neurological disorders, including autism, schizophrenia, and mental retardation. The discovery that one piece of DNA can lead to such diverse outcomes is opening new avenues in the study of disease. Rather than focusing solely on finding a common genetic flaw in everyone with a particular disease, researchers have begun to examine the various consequences that the same genetic flaw may have in different people. These studies suggest that even patients with different diagnoses may share common biological problems. "It's been eye-opening," says Mark Daly, a geneticist at the Broad Institute in Cambridge, MA, "because it's made us realize that in searching for the molecular basis of disease, it may be profitable to search for connections between seemingly unrelated phenotypes." Last year, Daly and his colleagues identified a section of DNA on chromosome 16 that also raises the risk of several different brain disorders, suggesting that this pattern may be common in the genetics of disease.
Physicians have long known that structural abnormalities in our genomes--deletions, duplications, and rearrangements of large stretches of DNA--trigger developmental problems and disease. Down syndrome, for example, results from an extra copy of chromosome 21. But over the last few years, new kinds of microarrays--small slides dotted with specific sequences of DNA--have begun allowing scientists to efficiently search the genome for architectural flaws too small to be visible with a microscope. These errors, called copy number variations, are distinct from the single-letter changes that until recently have been the focus of most research into genetic variation. Ranging in size from one thousand to more than one million base pairs, they can encompass part of a gene or one or more entire genes.
The far end of region 1q21.1, which at about one million bases long constitutes a tiny percentage of the roughly 3.2 billion pairs of letters that make up human DNA, harbors just one of the genome's many "hot spots"--so called for their tendency toward structural instability. But in this region, structural abnormalities--especially missing sequences--seem particularly troublesome. Intrigued by this mysterious morsel of DNA, Heather ­Mefford, a pediatric geneticist at the University of Washington in Seattle, compiled data on variations in 1q21.1 from clinical genetics labs around the world. She found that 25 patients in a sample of more than 5,000 people with autism, mental retardation, or other congenital abnormalities were missing the same chunk within the region. While that is a small percentage, no one in a similar-sized group of healthy people carried that particular mistake, meaning that the deletion is the likely cause--or at least partial cause--of the patients' problems. Studies by other researchers have linked similar changes in the region to schizophrenia, as well as to abnormal head size and accompanying developmental delays.
Different studies linking 1q21.1 to mental retardation, autism, and schizophrenia all identified deletions or duplications in approximately the same region. That's because this particular stretch is flanked by repetitive sequences prone to rearrangement. It contains at least eight known genes, the functions of which are mostly unknown. "This region of the genome must clearly have one or more genes that are important for normal cognitive development," says Mefford, whose research was published in the New England Journal of Medicine in October.



Scientists hope that identifying the underlying mechanisms affected by the missing or duplicated piece of DNA will provide new targets for drug development. But at this point, it's not clear whether it's one gene or several that raise the risk of disease, or how deletions and duplications of the same piece of DNA can trigger outcomes as different as schizophrenia and mental retardation.
The findings do hint that autism, schizophrenia, and mental retardation have common biological underpinnings, a conclusion that has some precedent. Children with mental retardation often have psychiatric and behavioral problems as well, although these may be undiagnosed or underappreciated in the face of their cognitive deficits. And some families may have a history of mental illness, but not of a specific illness.
Mental retardation, autism, and (to some extent) schizophrenia are developmental diseases, diagnosed in childhood or adolescence. So identifying a common biological flaw may shed light on the crucial components of neural development and suggest ways to help when that development goes awry. Perhaps a disruption in the 1q21.1 region of the chromosome inherited from one parent can send some fundamental developmental process off course. The ultimate impact might depend on environmental factors, variations in other parts of the genome, or the version of the gene inherited from the other parent. Someone whose genome has mistakes in other regions that are important for brain development and cognitive function might end up with mental retardation. Someone whose genome is largely intact, but who has a mutation in a gene linked to autism, may end up with high-functioning autism.
A better understanding of the molecu­lar consequences of errors in 1q21.1 and other recently identified hot spots may help redefine autism and schizophrenia and even change the way they are diagnosed. Both disorders cause a wide range of symptoms, and they are currently identified through behavioral and cognitive tests. Physicians may now be able to augment that diagnosis with the results of genetic testing. Only a small percentage of people with autism or schizophrenia will carry a particular genetic variation. But researchers hope that as more copy number variations are linked to these disorders, such genetic characterizations will become useful tools for predicting the best treatment for a given patient.
"At one time in the history of medicine, when you had a cough and an infection of the lungs, they called it pneumonia," says James Lupski, a physician and scientist at Baylor College of Medicine. Now we know that pneumonia is actually a group of different diseases, both bacterial and viral, that must each be treated differently. Eventually, someone developed a way to distinguish bacterial pneumonia from other forms, ­Lupski says, and that set the stage for the development of different treatments.
A diagnostic test that can detect copy number variations already exists: array CGH, the same test scientists use in research studies. It is currently used in clinical genetics labs to diagnose unexplained cases of mental retardation, developmental delay, and, increasingly, autism as well. It's not yet clear how to use the results to guide treatment--especially in disorders such as autism, for which no drugs are available to treat the root cause. But when it comes to other disorders, scientists are optimistic. "We have lots of effective psychiatric drugs, but it often takes weeks to find the right one," says Lupski. "Could this simple characterization predict the one that works best? That alone would be of tremendous benefit to patients."