The Puzzle of Mental Illness : New Research Can’t Link Genes to Two Major Disorders

New research has raised serious doubts about previously proposed genetic linkages to two serious forms of mental illness.

Researchers from the National Institute of Mental Health and UC San Diego reported in Thursday’s Nature that they were unable to confirm previous data linking manic-depressive illness--which causes as many as 2 million Americans to experience radical mood swings between elation and depression--to a faulty gene.

Nor have researchers been able to reproduce British studies demonstrating a link between a gene on chromosome 5 and schizophrenia, which affects one in every 100 people by causing hallucinations and delusions and inappropriate emotional responses.

The failure to confirm these earlier studies, neuroscientists say, indicates the great difficulty of determining a precise cause for such complex and variable diseases as the many forms of mental illness.

But even though an editorial in the same issue of Nature says, “This leaves us with no persuasive evidence linking any psychiatric disease to a single (gene),” some geneticists are undaunted.

“Fundamentally, we still believe strongly that mental illness has a very, very strong genetic component,” said Steven M. Paul of the NIMH, one of the authors of the new study. “We just haven’t found the genes yet.” Paul and others say their work actually suggests that many types of mental illness may be caused by more than one gene.

In an effort to locate such genes, the NIMH is beginning a major study, at 10 U.S. medical centers, of families with a history of manic-depressive illness, schizophrenia and Alzheimer’s disease.

Although he called the failure to confirm the earlier findings “depressing,” geneticist Kenneth Kidd of Yale University noted that “they are injecting a healthy dose of reality into the field.”

Kidd said that because of the initial, apparent ease with which researchers had linked manic-depressive illness and schizophrenia to genes, “virtually every psychiatric department in the country is trying to set up a DNA laboratory to do these kinds of studies.”

Now, he said, the pendulum is swinging back to the middle ground as researchers are learning that “it is going to take a lot of effort” to find the genes.

The original study linking manic-depressive illness to a faulty gene, reported in February, 1987, was conducted among the 12,000 members of the Old Order Amish community in Lancaster County, Pa. All are descended from 15 to 25 couples who emigrated from Europe in the early 18th Century.

Manic-depressive illness is no more common in the Amish than in the general population. But the large family size, clearly established paternity, and social and religious prohibitions against use of drugs and alcohol in the community make it an ideal group to study hereditary factors in genetic disorders.

In the original study, psychiatrist Janice A. Egeland of the University of Miami School of Medicine and her colleagues studied DNA (deoxyribonucleic acid, the genetic blueprint of humans) from an Amish family with 81 members, 19 of whom had manic-depressive illness. They found that all 19 affected individuals had an identical stretch of DNA on the short end of chromosome 11, one of the 46 chromosomes (23 pairs) in which all genetic information is concentrated.

They did not, however, identify a specific gene that causes the illness.

In work that led to the new study, Paul and his colleagues were studying a gene that had previously been linked indirectly to depression and, possibly, to mania. The researchers wanted to find out whether a defective form of the gene might be present in the Amish individuals with manic-depressive illness, but the first few individuals whose cells they studied did not have the gene they were looking for.

Because no other research group had been able to find families that share this same genetic defect, this finding persuaded them to restudy the whole group.

But when they did so, they found that the family “had changed in several ways,” said neurogeneticist John R. Kelsoe of UCSD, who wrote the Nature paper along with Paul and others. Most important, he said, two individuals who had been healthy at the time of the earlier study had now developed manic-depressive illness--and these individuals did not have the stretch of DNA shared by the 19 other affected members.

Incorporating these changed diagnoses into the original study “does not eliminate the evidence for a genetic linkage, but it reduces it below the level of statistical significance,” said Kidd, who also participated in the study.

The researchers also studied a set of 31 extended family members, eight of whom have manic-depressive disease or related illnesses, who hadn’t been included in the original study and found no evidence at all for the linkage to chromosome 11 in this group, making the original conclusion weaker still.

The group is now studying two hypotheses, Kelsoe said. One is that the original linkage was “a false positive, not an error, but a statistical error,” he said. The second, and perhaps more likely, hypothesis is that more than one gene can cause the disease. In fact, researchers have always argued that mental illnesses, with their great variability, are probably caused by more than one gene.

Support for the latter view is found in the fact that the 30 extended family members share a common ancestor, a Mennonite woman who married into an Amish family 100 years ago and who suffered from some form of mental illness. It is possible, Kelsoe said, that her illness was caused by a second gene and that the presence of that gene in the family has complicated the interpretation of the results.

Reports at recent meetings suggest that the problem with the putative schizophrenia gene is more severe. Psychiatrist Hugh Gurling of the University of London and his colleagues reported last November that they had studied seven British and Icelandic families in which 39 of the 104 family members had the disease.

They reported that the suspected schizophrenia gene seemed to lie between two well-known genes on chromosome 5. Unfortunately, researchers in five other laboratories have since reported that they have been unable to find the link to these two genes in other families.

Gurling has also not been able to provide other researchers with DNA from the families he studied, so no one has been able to attempt to replicate his results. “The burden of proof is on Hugh now,” Paul said.

Psychiatrist Lewis Judd, director of the NIMH, said the new evidence about both putative genes is “very disappointing.” But, referring specifically to the manic-depressive study, he said, “It doesn’t mean that there isn’t a gene present. It just means that the likelihood of it being on chromosome 11 is significantly diminished.”

Judd also noted that, “because we believe the evidence of genetic influence is so strong,” workers at the NIMH are systematically mapping all the chromosomes of the Amish to try to locate the gene. In their new program, they are also aggressively seeking other families with the most common mental illnesses to expand the search for causative genes.

Judd remains confident. “It is clear to all of us that, given all the evidence (about genetic links), if studies are carefully and systematically done, the linkages will appear and genes will be identified for specific mental disorders.”