Changes in Brain Shown with Learning

New Tool Speeds Study of Mammalian Protein Function

Neuro Center Ramps Up Drug Discovery Core

International Health:
Speaker Addresses Plight of Afghan Refugees

Herpes Oncovirus Reveals Trick for Dodging Immune System

Fishing for Good Health

Dopamine Receptor Puts Brake on Self-Destruction

How the Cell Decommissions Proteins

Fourth-years Now See Future

Biomedical Neuroscience Center Launches Website

HMI Announces Online Newsletter

Michael Crichton Speaks on Medicine and the Media

HST Forum Celebrates Student Research

HMS Again Ranked Top Research Med School

In Memoriam:
Juan Taveras

Honors and Advances

News Briefs

Living, and Dying, by Example

Front Page


Changes in Brain Shown with Learning

Findings on Fear Suggest Path Toward New Anti-anxiety Medications

HMS researchers have made a discovery that could help resolve one of the liveliest controversies in contemporary neuroscience--how the brain changes in learning and memory.

Researchers have argued that for memory and learning to occur, connected neurons must become more responsive to one another--so responsive, in fact, that they will continue to communicate at a high level even when they are no longer being stimulated by an external source, such as other neurons. But so far no one has been able to show conclusively that this enduring heightened responsiveness, or long-term potentiation, actually occurs in learning and memory, leading some to doubt the existence of a causal link between learning and this physiological change in the brain.

Focusing on the amygdala, Vadim Bolshakov (right) and Evgeny Tsvetkov were able to demonstrate the link between long-term potentiation and memory. Photos by Steve Gilbert

"There have been a series of progressive attempts, but there has been no demonstration of a causal interaction between the two," said Eric Kandel, a scientist at Columbia University College of Physicians and Surgeons.

McLean researchers Vadim Bolshakov, Evgeny Tsvetkov, and Bill Carlezon, working with Kandel and colleagues, report in the April 11 Neuron that they have found clear evidence of cause and effect between learning and long-term potentiation--in this case, learned fear--in the brains of rats. "I think that this is the first really important causal link," said Kandel, who won the Nobel Prize in physiology or medicine in 2000. Bolshakov and Carlezon are HMS assistant professors of psychiatry. Tsvetkov, who is first author on the study, is an HMS research fellow in psychiatry.

Pinpointing the Problem

What made the discovery possible, Kandel said, was Bolshakov's method for stimulating neurons in a brain area involved with emotional learning, the amygdala--and the tenacity to correlate his electrophysiological findings with actual behavior in live animals. "He has developed ways of stimulating single fibers in the amygdala so you can look at elementary synaptic events there," Kandel said. "It is a very powerful tool. His second accomplishment was to use these rigorous methods to relate synaptic changes to intact animal behavior."

In addition to helping settle an important scientific debate, the discovery could lead to a better understanding of a class of psychiatric disorders that affect millions of Americans--namely, anxiety disorders. Panic, phobias, posttraumatic stress disorders, obsessive-compulsive disorders, and generalized anxiety are among the most common forms of mental illness. At least some of them are thought to involve the fear system of the brain.

"It has been shown that an individual can have very poor conscious memory of a certain traumatic event but, at the same time, very strong unconscious emotional memories can be formed through a fear conditioning mechanism," said Bolshakov. "These fears, which are very resistant to extinction, can become a source of intense anxiety."

Understanding how cellular mechanisms, including long-term potentiation, work to generate such unconscious fears could someday lead to novel treatments for anxiety disorders. "The study of learned fear in the mouse and the rat is likely to provide a new generation of anti-anxiety agents that will be very useful," said Kandel.

The focus on fear conditioning proved key in linking long-term potentiation and learning. Previous attempts had focused on spatial learning in the hippocampus, a complex curlicue of tissue buried in the brain that is thought to be involved in many kinds of learning. Bolshakov and his colleagues directed their attention instead to the amygdala--a well-studied but simpler pyramid-shaped structure that is involved in producing a variety of learned emotional responses, most notably, fear. It had been largely ignored by long-term potentiation researchers in part because of the technical difficulties involved in stimulating individual neurons in the structure.

Sounds of Science

To see if learned fear involved long-term potentiation in the amygdala's neurons, Bolshakov and his colleagues trained rats to fear a certain sound. After removing the amygdala from the rats, and keeping it functioning in a salt bath, they stimulated individual neurons at very high frequency. Normally during this kind of stimulation, presynaptic neurons release glutamate in a continuous manner, which causes their postsynaptic partners to produce an enduring increase in electrical current. This prolonged heightened response is precisely what occurs during long-term potentiation. But in the postsynaptic neurons of the fear-conditioned animals, output of current barely increased above their baseline levels, suggesting that their presynaptic partners had already been stimulating them as a result of fear conditioning.

"The presynaptic inputs have already been potentiated," said Bolshakov. "You basically cannot put anything on top of them because they are potentiated already." Upon closer examination, the researchers found that high levels of glutamate were released at the presynaptic terminals, confirming that long-term potentiation had occurred.

Bolshakov and his colleagues plan to take their investigation a step further, from what occurs at the synapse to what occurs inside the neuron. For learning to produce long-term memory, long-term potentiation must result in new proteins being synthesized inside the cell. "This is another line of experiments--to go deeper, to study transduction mechanisms, the intracellular machinery that is involved," said Bolshakov.

--Misia Landau