Researchers to track path of mental illness from gene on up

By Cynthia Lee and Mark Wheeler

When you get down to it, treatment for most mental illnesses remains a matter of trial-and-error. Despite a revolution in designing new drugs, we still don't know exactly how they work, and there's no way to predict exactly which drug will be most helpful for people with schizophrenia, bipolar disorder or other syndromes.

In fact, just coming up with a diagnosis can be exceedingly tricky, said neuropsychologist Robert Bilder, chief of medical psychology-neuropsychology at the Semel Institute for Neuroscience and Human Behavior.

"It's a little ironic that we diagnose people by talking to them," Bilder said. "With some of these disorders, like schizophrenia, communication may be disturbed as part of the syndrome. Yet the diagnosis is based mostly on a conversation."

Bilder and 51 other UCLA researchers who represent an unusual mix of clinicians and basic scientists are hoping to revolutionize the way we define mental illness in the future by identifying its biological foundations — down to the genetic level.

"Everyone agrees that if behavior is disturbed, then we have to find a signal that's responsible for that somewhere in the brain. The question is, what brain systems should we be looking at?" Bilder asked.

Researchers from the Semel Institute, the Geffen School of Medicine, the College, the Samueli School of Engineering and Applied Science, and the School of Public Health are coming together as "a dream team" to search for that answer.

Zeroing in on such mechanisms as memory and self-control, they are going to follow the progression that leads from gene to protein to cell to different circuits of the brain where these mechanisms reside and where behavior results.

"If we understand the biology of a disorder at the genetic and molecular levels, then we can start focusing more rationally on the drugs or behavioral treatments that will help," Bilder said.

So instead of diagnosing someone with schizophrenia, for example, doctors may one day be able to say this person has a certain degree of dysfunction in the brain system that controls memory and another degree of dysfunction in the brain system responsible for language generation.

Rather than lumping together people under the label of schizophrenia, which can have 100 different causes, doctors would instead pinpoint and quantify the strengths and weaknesses found in different brain systems.

"If we can do that, we're going to provide a much better, more accurate and biologically meaningful representation of what people are experiencing," Bilder said. "It will both advance our understanding of the causes of these problems and enable better treatment."

To pull off this high-wire challenge, Bilder and his colleagues have created the new Consortium for Neuropsychiatric Phenomics. Developing an interdisciplinary "research team of the future" was particularly important to the National Institutes of Health Roadmap Initiative, which recently awarded the consortium a $22-million grant over five years.

Phenomics is the systematic study of phenotypes. Just as a genotype describes your collection of genes, a phenotype describes everything that those genes produce — from the molecule to the mind. While the Human Genome Project gave us a map of the human genome, decoding the human phenome — discovering everything that gene variations produce — "is going to occupy science for the next century," said Bilder, director of the consortium.

"It's both exciting and daunting," said Nelson Freimer, co-director and a geneticist who, with a colleague, laid out the basic concept for the Human Phenome Project four years ago.

"It has taken years of preparation to make this happen," Freimer said. "We're able to do this because the study of neuropsychiatric diseases, together with the basic science that underlies them, is one of UCLA's great strengths."