Researchers say they are close to identifying two genes that go awry in schizophrenia.
New York Times, July 5, 2002
Free registration required
Despite years of false leads, setbacks and unsustained claims, researchers hope they are now starting to close in on some of the genes that go awry in schizophrenia, a devastating mental disease that affects two million Americans. Discovery of such genes would illuminate the fundamental mechanisms of the disease and might lead to new treatments.
From analyzing the genetics of 270 Irish families, each with several schizophrenic members, researchers at Virginia Commonwealth University in Richmond and two Irish institutes say they have found clues that point to a specific gene as a possible cause of schizophrenia. The gene, recently discovered in another context, is known as dysbindin and lies on the sixth of the 23 pairs of human chromosomes.
Separately, scientists at Decode Genetics, an Icelandic company that searches for disease-causing genes, has found a gene called neuregulin-1 on chromosome 8. Mutations in the gene are highly correlated with schizophrenia in about 15 percent of Icelandic patients.
Both reports are to be published in The American Journal of Human Genetics, which put the Richmond group's paper online yesterday.
Geneticists regard schizophrenia as a complex disease, meaning one to which many different mutated genes are thought to contribute.
Although scientists are now skilled at identifying the genes involved in single-gene diseases, complex diseases are far harder to analyze because they are not inherited in a recognizable pattern. It is the complex diseases, like cancer, Parkinson's, diabetes and stroke, that are the most common.
Very few of the responsible genes in complex diseases have been identified, and the two proposed schizophrenia genes, if confirmed, would be among the first.
Both the Virginia and the Decode teams worked by analyzing small runs of DNA known as markers along the genome, or total DNA, of their patients. They then used statistical methods to see which markers the schizophrenic patients seemed to have inherited more often than did unaffected relatives.
The markers, which are like milestones along each chromosome, have a fixed position and so can be used to define a region of the chromosome where a schizophrenia-causing mutation may lie.
These candidate regions, called loci by geneticists, are often large and may contain many genes, any one of which could hold the suspect mutation. Finding a locus is hard enough, and a whole new problem to narrow down the responsible gene within it.
In the past several years many researchers have claimed to have detected a schizophrenia locus. But the evidence for these loci is often weak, in part because of the expense of gathering as many patients as good statistics requires, and several claims have not been confirmed.
But with the human genome sequence available, as well as catalogs of common genetic variations, research groups have started to hit on the same handful of loci, suggesting that they could be seeing a true signal.
"I think there is a limited number of chromosomal regions for which there is converging evidence for schizophrenia. Each of these genes are in those regions," said Dr. Douglas Levinson, a schizophrenia expert at the University of Pennsylvania, referring to the new suspect genes.
Both new genes are plausible candidates for being involved in schizophrenia because each is active in brain cells. Dysbindin, the gene identified by the Richmond group, is involved in the operation of the synapses, the points where one neuron wires itself into another.
The Irish patients were identified by Dr. Kenneth S. Kendler of Virginia Commonwealth and two Irish physicians, Dr. F. Anthony O'Neill and Dr. Dermot Walsh. The project started in 1983, Dr. Kendler said, and involved visiting every psychiatric hospital in Ireland and Northern Ireland except for downtown Belfast, then a war zone. His former colleague, Dr. Richard E. Straub, now of the National Institute of Mental Health, performed the genetic analysis.
Dr. Straub found genetic variations in the dysbindin gene that were more common in the schizophrenic patients. Curiously, they are all in introns, the spacer regions of the DNA that lie between the working parts of the dysbindin gene. The Richmond team is not sure that any of the intron changes is the causative mutation of schizophrenia and is analyzing the working parts more closely.
"Soon we will know if this is another false positive or a real signal," Dr. Kendler said.
Less is known about the candidate schizophrenic gene found in Iceland, which has so far been described only in a patent application. Dr. Kari Stefansson, chief executive of Decode Genetics in Reykjavik, said the Richmond group's dysbindin gene seemed to work in conjunction with the cell's receptor for neuregulin-1, the product of the gene implicated in Icelandic schizophrenics.
"If you are willing to stretch yourself, standing on your toes, you could see these genes as part of the same story," Dr. Stefansson said, suggesting that both genes might be involved in the regulation of synapses and that mutations in either could cause miswirings in the brain's neuronal circuits.
Though Icelanders share some Celtic ancestry with the Irish, Decode had not seen any linkage to schizophrenia at the dysbindin site among its patients, Dr. Stefansson said. But the negative finding meant little, in his view, because his method allowed only the most significant genes in schizophrenia to be mapped, not minor ones.
Dr. Steven Warren, editor of the journal that is publishing the two articles, said the next important step was to see if other scientists could confirm the results.
"We're getting more compelling genes that are making some sense," Dr. Warren said, "so I think it's worthwhile thinking we might finally be getting close to some genes that predispose people to this important disease."
Schizophrenia has a clear genetic component because it runs in families. But genetics cannot be the entire explanation because if one identical twin has the disease the other has only a 30 percent to 50 percent chance of developing it. This suggests that something in a person's environment interacts with their genetic makeup to trigger the disease.