Featured Profile

Gabriel Corfas, Ph.D., a neuroscientist at Children’s Hospital Boston and Harvard University, modestly characterizes an idea he had in the late 1990s as “a bit of good luck.” It was around that time that the Argentinean-born Corfas began to think that there might be a link between schizophrenia and a newly discovered growth factor in the developing vertebrate brain called neuregulin, which he had earlier helped to purify for the first time.

NARSAD’s Scientific Council saw considerable promise in this line of research and in 1998 named Corfas a NARSAD Young Investigator. The path he and colleagues have since followed in exploring the role of neuregulin exemplifies the purpose of NARSAD’s early-career funding program: to help make possible a young scientist’s pursuit of a good idea in its infancy.

Corfas, who is now an associate professor in the departments of neurology and otolaryngology at Harvard Medical School and researcher in the Neurobiology Program at Childrens’s Hospital Boston, received NARSAD’s mid-career Independent Investigator Award in 2005. This past April, he and his colleagues, including Kristine Roy, Ph.D., a 2006 NARSAD Young Investigator, published a paper in the Proceedings of the National Academy of Sciences reporting success in linking a defect in the brain’s white matter -- a defect they induced by blocking cellular signals triggered by neuregulin -- with behaviors in laboratory mice suggestive of schizophrenia. It was a finding with the potential to lead to methods of detecting schizophrenia much earlier than is now possible, as well as to new and more effective treatments.


“How Does the Brain Work, and What Goes Wrong in Disease?”

If, as the old adage has it, “chance favors the prepared mind,” cutting-edge science rather unsentimentally demands it. The career of Gabriel Corfas is a case in point, a story of perseverance not unlike that of most top scientists, who spend years satisfying their curiosity about the way things work and in doing so, preparing the ground for future discoveries.

Corfas studied biology as an undergraduate at the University of Buenos Aires. “I was always fascinated by behavior,” he says. “My main interest was how we learn, how we can think and how that can be disrupted in disease.” He determined that he wanted to use the tools of molecular biology and animal models to advance the state of knowledge. He could not do that kind of work in Argentina, and so decided to pursue his studies at the Weizmann Institute of Science, in Rehovot, Israel. He earned his Ph.D. in the department of neurobiology in 1990 for studies of the cellular and molecular mechanisms of learning and memory, using the fruit fly as a model.

Though he found this experience rewarding, Corfas was only getting started. “I felt that I needed to understand more about the function of [nerve] cells and synapses in the brain.” In 1989, as he was finishing his doctoral work, he made his pilgrimage to the United States, where he has since become a citizen. He took a position as postdoctoral fellow with Dr. Gerald D. Fischbach, at the Washington University School of Medicine in St. Louis. The following year, the lab moved to the department of neurobiology at Harvard. Under Fischbach, Corfas learned about the electrophysiological properties of synapses, his interest being drawn especially to the question of how these crucial junctions between nerve cells in the brain form and develop early in life.

This phase of his preparation perhaps more than any other determined the direction of Corfas’s career, whose overarching theme is to understand the molecular signals that regulate the interactions between neurons and the class of brain cells called glia. Glial cells, “helpers” to neurons, play critical roles in several aspects of nervous-system development, including the migration of neurons in the young brain, brain-cell differentiation and the formation and function of synapses.

It was during his postdoctoral years in the Fischbach lab that Corfas became involved in an effort to purify the growth factor that later came to be called neuregulin, or NRG. “We had been working with neuregulin since before its formal discovery, and once I started my own lab, in 1990, I hypothesized that this molecule was an important contributor to the way in which neurons and glia communicate. I decided to focus on that.”

Corfas set about making animal models that he hoped would illuminate the mechanism behind the hypothesized relationship between neurons and glia. He believed that this work was likely to shed new light on the glial cells, whose range of functions throughout the life of the organism is vast and absolutely critical to normal brain function. Thinking about this in a slightly different way, Corfas had the thought -- he calls it a lucky guess -- that neuregulin, because of its importance in activating these supporting cells, was important in schizophrenia. By 2004, he had published a “Perspective” paper for the prestigious journal Nature Neuroscience in which he discussed several ways neuregulin might contribute to schizophrenia.


A Heavy Burden of Proof

Corfas’s interest in signaling between and within brain cells placed him in good position to work with a fact that had been in circulation for some time: that defects in a gene called NRG1, which directs cells to manufacture the protein neuregulin, are much more common in people with schizophrenia than in those who do not have the disease. The critical question concerned mechanism: How does the observed flaw in the gene trigger events that cause schizophrenia? As things stand, the relation between the flawed gene and the illness could turn out to be simply a statistical coincidence.

For neuroscientists like Corfas, this is a matter that calls for rigorous proof, which can be established only by identifying the molecular mechanism by which a small irregularity in the gene somehow sets in motion a train of missteps that culminate in abnormal brain function and behaviors that clinically indicate schizophrenia. The process of proving such a series of relationships is almost unimaginably intricate, calling upon an experimenter to devise ingenious methods to raise lines of laboratory mice with specific defects in the NRG1 gene, and in which the path of subsequent causation can be documented and reproduced.


Designer Mice

In past decades, brain researchers were forced by technological limitations to work with slices of brain from postmortem tissue. They could not watch the pathological processes they were exploring develop in real time in living creatures. In the 1990s, using techniques made possible by the revolution in genome science, scientists became able to engineer laboratory mice with specific genetic defects and to observe their consequences in vivo. One approach is to “design” a mouse to grow up with one or more specific genes “turned off” or missing altogether. This is tricky work, for the genes are presumably there for a reason -- each has functions to perform, and some cannot be deleted without causing a cascade of unwanted physical and behavioral defects.

To make a mouse in which the function of neuregulin is in some way manipulated, it becomes essential to know how neuregulin goes about performing its function in brain cells. Neuregulin is primarily produced by neurons, and when released finds its way to a very specific set of protein molecules on the surface of various neuronal and glial cells in the brain.

Throughout the human brain, it has been estimated that there are 10 times as many glial cells as neurons, and Corfas has performed experiments involving several subclasses of them. In 2005, he and colleagues discovered neuregulin’s role in regulating the timing by which star-shaped glial cells called astrocytes are generated during embryonic development. Found in both white and grey matter of the brain, astrocytes, among other things, are responsible for clearing unabsorbed or excess neurotransmitter molecules from the gap between nerve-cell synapses. At an important stage early in nervous-system development, neuregulin suppresses astrocyte formation, Corfas discovered, which has the effect of enabling neurons to properly develop.

Even as they conducted their study of astrocytes, Corfas and his team were taking this work in another direction. The research with astrocytes had revealed an intricate series of intracellular events set in motion by the “docking” of the neuregulin molecule with a receptor molecule called erbB4. The erbB4 receptor and defects in the gene that gives rise to it -- like neuregulin and the gene that orders its production -- have been associated with the incidence of both schizophrenia and bipolar disorder. Putting these facts together with his interest in the relationship between white matter and schizophrenia, Corfas and colleagues began a new series of experiments.

They would try to engineer mice in which signals set in motion by neuregulin would be disrupted. The signals Corfas wanted to target this time were transmitted when neuregulin “docked” with the erbB4 receptor on a specific subset of glial cells called oligodendrocytes, or OLs. OLs wrap around the long axonal threads that project outward from nerve cell-centers. These fibrous projections hard-wire the brain. OLs produce myelin, the white, fatty coat that insulates axons in the same way that plastic or rubber coating insulates a copper wire.

The myelin sheath around axons facilitates the transmission of nerve signals and, incidentally, gives the white matter its characteristic hue. But when OLs fail to coat the axons, or when the axons lose their myelin sheath -- as in multiple sclerosis (MS), for example -- nerve signals may be short-circuited altogether or conducted through the fibers at abnormally slow speeds, causing a timing problem in the brain and throughout the neuromuscular system that results in devastating symptoms including the loss of coordination and slurred speech.

Corfas set out to interfere with the neuregulin-erbB4 signals that command OL cells to manufacture myelin and coat axons, hoping to learn about the pathology of schizophrenia. Would white-matter defects in mice be accompanied by behaviors associated with psychiatric illness? Knocking out the gene that generates the erbB receptors (pronounced “errr-bee”) was out of the question. “This is a complex family of receptors, and all of them play important roles in many different cells,” Corfas explains. “If you eliminate the receptors entirely, the animals die of heart malformations.” Instead, he developed mice with a mutant form of the erbB4 receptor, blocking its normal function. “Our strategy was not to eliminate them, but to prevent their activation, but only in one particular type of cell” -- the OL cells that manufacture myelin in the brain.


Success--and a Series of Intriguing Surprises

The experiment was successful, in ways that were hoped for, but also in ways not previously imagined by Corfas and his colleagues. When NRG1-erbB signaling was blocked in OL cells, the mutant mice developed an above-normal number of OL cells, but the cells had notably fewer branches and formed a thinner than normal myelin sheath around nerve fibers. This, in turn, slowed the propagation of signals in the affected nerves by 18 percent.

Top of page