Human Behavior

Research Ethics
Experimental Therapeutics
Research Briefs
Around & About

One patient with panic disorder carries a list of emergency rooms in his car so when he feels his heart pounding and he can't catch his breath, he can get to the nearest hospital, even though he knows he's really having a panic attack, not a heart attack.

Another patient with social anxiety disorder pretended to leave his house for school, but then climbed back into his bedroom window and spent the rest of the day in the closet to avoid his fear of feeling foolish in front of his classmates.

These individuals, like others with panic or social anxiety disorder, make life-changing adjustments to avoid situations that might produce fear or anxiety.

Genetic and epidemiological studies show that panic and anxiety disorders have a strong heritable component, but as with most other disorders with multiple gene etiology, researchers have had little success uncovering the responsible genes. Part of the problem is that the disorders' pathophysiology, what actually changes in the brain, is unknown.

In spite of these challenges, a multidisciplinary group of Columbia researchers believes it can identify the heritable genetic changes that underlie at least some anxiety disorders using an alternative approach: the study of a perfectly normal behavior – learned fear. A $6.7 million grant from the National Institute of Mental Health supports five related projects led by Drs. Eric Kandel, Rene Hen, Abby Fyer, Myrna Weissman, and Conrad Gilliam.

"As an example of learned fear, imagine that a gazelle comes to a pond for a drink where it meets a tiger and barely escapes with its life. The pond then becomes associated with the gazelle's innate fear of tigers, and the gazelle learns to be wary around ponds," says the group's leader, Dr. Conrad Gilliam, professor of genetics and development and psychiatry and director of the Columbia Genome Center. "Learned fear is a highly conserved adaptive behavior documented in species ranging from snails to humans. Moreover, evidence suggests that this fear-related behavior will share genetic determinants with at least some types of human anxiety disorders. It's our window into the study of panic disorder and anxiety."

That window has already been partially pried open by decades of research into the brain circuits that control learned fear. In a classic laboratory experiment, mice hear a sound and at the same time are shocked with a small electrical charge to their feet. They soon learn to associate the sound with the shock; when they hear the sound, their heart rate soars and they freeze, even when no shock is administered.

The cellular circuitry inside the mouse's brain that mediates this behavior is generally understood. It starts at the ear, where two pathways carry information about the sound to the back of the amygdala, a pair of almond shaped structures in the cortex. The synapses between cells in the pathways strengthen as the mouse learns to connect the sound with the shock. In essence, the mouse has created a "fear" memory. When it hears the sound, the amygdala quickly interprets messages from the strengthened pathways and sends messages out to many other parts of that brain that increase heart rate and freeze the animal in its tracks.

Human circuitry also works the same way so researchers hope that learning about genes and proteins in the cellular circuits of mice will enable them to understand differences in fear learning among people. The team's neurobiologists are now looking for fear-enhancing genes in mice – Dr. Kandel, University Professor, is looking for genes involved in learned fear, and Dr. Hen, associate professor of pharmacology in the Center for Neurobiology, is looking for genes involved in the innate fear that animals are born with.

Dr. Kandel has already found one learned fear gene in the amygdala and the pathways leading into the amygdala. In a paper published last December in Cell, Dr. Gleb Shumyatsky in Dr. Kandel's lab reported on gastrin-releasing peptide (GRP), a protein that limits the strength of the fear memory. By inactivating GRP's receptor, the lab could strengthen the synapses in the amygdala fear circuit and increase the mouse's fear.

Do more subtle changes in the GRP protein, its receptor, or other genes make people susceptible to panic or anxiety disorders? Using a test designed to measure fear learning in humans, that question may be answered by the team psychiatrist, Dr. Fyer, and genetic epidemiologist, Dr. Weissman, both of the Department of Psychiatry and the New York State Psychiatric Institute. Dr. Fyer will test people without panic or anxiety disorders. Dr. Weissman plans to evaluate individuals from families with multiple cases of panic disorder or social anxiety for fear learning and anxiety traits.

Data from these evaluations will be analyzed in Dr. Gilliam's lab to detect heritable gene changes related to variation in normal fear learning and to human anxiety traits and disorders. In a complementary approach, Dr. Abraham Palmer, associate research scientist, will use mouse strains bred for good and poor fear learning to locate mouse genes that influence fear learning. Finally, bioinformatic approaches will be used to detect fear-related patterns of gene expression.

By the end of the four years, the team hopes to identify critical fear-related genes in mice and in humans and the beginnings of a mechanistic understanding of normal fear learning and processing. The more ambitious goal, Dr. Weissman says, "is to put everything together to translate from fear learning in animals to human panic and anxiety disorders."