Scientists Identify Brain Region That May Give Rise to Schizophrenia


by Carl Sherman

November 6, 2009

Schizophrenia is as mysterious as it is devastating. No one knows just what happens within the brain to cause the disease’s characteristic hallucinations, delusions and cognitive deficits. Now researchers have identified a small region in the hippocampus where an early increase in activity—before symptoms become marked—might represent the beginning of the disease process. The results could lead to new targets for drug treatment and new ideas for strategies to prevent the disease.

Schizophrenia is often preceded by a “prodromal” phase, when individuals begin to exhibit symptoms but fall short of the criteria for a psychiatric diagnosis. They might, for instance, withdraw socially, or hear their names in the sound of the wind rather than an outright hallucination. Some studies suggest that treatment—including medication—during these early stages may delay the onset of full schizophrenia, reduce its severity or prevent it altogether.

But not all prodromal individuals go on to develop the disease—just 35 percent within 2.5 years, says Thomas McGlashan, a professor of psychiatry at Yale University who works with such patients. In the face of this uncertainty, most clinicians believe the risks of side effects from treatment outweigh potential benefits and have adopted a conservative wait-and-see approach. “For the ‘false positives,’ which might be the majority, you’d be unnecessarily giving powerful agents that can have severe side effects,” McGlashan says.

To investigate why the prodromal phase progresses to schizophrenia in only one out of three patients, researchers from Columbia University used a high-resolution variant of functional magnetic resonance imaging (fMRI) to compare the brains of 18 people with established schizophrenia with 18 healthy controls. In the disease group, they found increased blood flow, suggesting higher levels of activity, in the orbitofrontal cortex and in a small section—the CA1 subfield—of the hippocampus. In the dorsolateral prefrontal cortex, meanwhile, blood flow decreased.

Was one of these areas the place where schizophrenia began? “Our assumption is that no matter how complex or chronic a disorder is, there will always be one region of the brain most vulnerable to it,” says Scott Small, a professor of neurology at Columbia University and senior author of the research, which appears in the September 2009 issue of Archives of General Psychiatry. Once dysfunction starts, however, it spreads broadly throughout the brain. “It’s very hard to distinguish primary from secondary areas affected by the disorder when you look at someone with full-blown schizophrenia.”

Indeed, when the researchers examined 18 young people who had prodromal symptoms—some of whom presumably had the disease in a very early stage—the group as a whole differed from the healthy controls only in the CA1 subfield.

The evidence was stronger when the researchers reanalyzed the data two years later. By then, seven in the prodromal group had been diagnosed with schizophrenia. These people had shown significantly higher CA1 activity when first measured compared with the “false positives.”

Further analysis of the fMRI data added another connection between this brain region and psychiatric disorder—increased blood flow in the CA1 subfield mirrored the severity of symptoms. Delusions, for example, tended to be more severe in patients with higher CA1 activity.

New tests and treatments?

Besides providing insight into the pathological process, pinpointing where schizophrenia begins could help doctors detect the disease in its earliest stages—and initiate treatment when it is likely to do the most good.

“This is a really exciting paper,” says James M. Stone, a clinical lecturer and honorary consultant in psychiatry at the King’s College London Institute of Psychiatry. “The fact that hyperactivity in the CA1 area of the hippocampus both predicted transition to psychosis and correlated with symptoms—it needs replication, but it’s amazing stuff.”

The paper has both theoretical and practical implications, McGlashan adds. “It probably has equal interest for people interested in what this disorder does to the brain, and in those who would like to know how good a diagnostic signal it is.”

A biological test to better predict which prodromal patients will progress to psychosis is “the golden fleece for clinicians,” but while the predictive value of the new findings is statistically significant, even if replicated they would fall short of the level of accuracy that might justify aggressive early treatment. But they’re “an important step in that direction,” McGlashan says. “It’s a good new avenue to begin looking at.”

Cheryl Corcoran, director of the Center of Prevention and Evaluation at Columbia and a co-author of the paper, agrees that talk of clinical application is premature. But “it would be interesting to look at CA1 and other potential risk factors in tandem.” Electrophysiological changes, neuropsychological profiles or genetic markers teamed with high-resolution fMRI, “might have really good predictive value.”

More generally, the findings have provocative implications for our understanding of the underlying biology of schizophrenia. The CA1 area is particularly responsive to the neurotransmitter glutamate, which is the primary neurotransmitter in most brain regions. As Stone points out, the glutamatergic system has become a subject of growing interest for schizophrenia researchers. A recent animal study at Columbia, for instance, found that mice genetically modified to reduce glutamatergic neurotransmission were resistant to the effects of drugs that otherwise induced psychotic-like symptoms.

Traditional thinking about schizophrenia has emphasized another neurotransmitter, dopamine; antipsychotics, new and old, are believed to work largely by blocking dopamine receptors. “These findings suggest we might want to think beyond antipsychotics and try other medications in young people at risk,” Corcoran says. “They suggest we consider that there are stages of illness, and that each stage might be responsive to different pharmacological treatment.”

It may be worth exploring the efficacy for very early schizophrenia of existing drugs (such as the anticonvulsants lamotrigine and gabapentin) and amino acids (such as D-serine and glycine) that affect glutamatergic neurotransmission, Corcoran says. (A clinical trial of D-serine for prodromal individuals has in fact been underway since early 2009.)

 The fMRI findings could guide drug development along more precise lines than these, Small suggests. “Identifying this specific region raises the question, why is CA1 ‘hot’ in schizophrenia?” he says. . “If we can find molecular reasons why glutamate is elevated in that area, find what—an enzyme perhaps—is abnormal in the glutamate pathway, it might provide a new drug target.” -30-