Scientists have long been captivated by the questions of how memories form and how they are represented in the brain. The answers to these questions may help researchers understand how to treat or prevent memory problems, drug addiction, and other human ailments. Thousands of changes in gene expression, neuron formation, nerve signaling, and other characteristics may be involved in the formation of just a single memory. Scientists refer to any learning-induced change in the brain as a "memory trace."
In the new study, Ronald L. Davis, Ph.D., and colleagues at Baylor College of Medicine in Houston developed fruit flies with special genes that caused the flies' neuronal connections to become fluorescent during nerve signaling (synaptic transmission). They then exposed the flies to brief puffs of an odor while they received a shock. This caused them to learn a new association between the odor and the shock – a type of learning called classical conditioning.
Using a high-powered microscope to watch the fluorescent signals in flies' brains with as they learned, the researchers discovered that a specific set of neurons, called projection neurons, had a greater number of active connections with other neurons after the conditioning experiment. These newly active connections appeared within 3 minutes after the experiment, suggesting that the synapses which became active after the learning took place were already formed but remained "silent" until they were needed to represent the new memory. The new synaptic activity disappeared by 7 minutes after the experiment, but the flies continued to avoid the odor they associated with the shock.
This is the first time that optical imaging has been used to visualize a memory trace, Dr. Davis says. "It's phenomenally powerful, like a movie appearing in front of you," he adds. The study suggests that the earliest representation of a new memory occurs by rapid changes – "like flipping a switch" – in the number of neuronal connections that respond to the odor, rather than by formation of new connections or by an increase in the number of neurons that represent an odor, he adds.
The fact that the flies continued to show a learned response even after the new synaptic activity waned suggests that other memory traces found at higher levels in the brain took over to encode the memory for a longer period of time, Dr. Davis suggests. If so, the rapid changes of nerve transmission that the researchers saw may be the all-important switch that initiates the formation of new memories.
This research suggests a previously unknown mechanism for how memories are formed, Dr. Davis says. While this study looked only at learning related to odors, this newly identified process may be at work in many other kinds of learning as well. It is likely that these or similar mechanisms are important for memory in humans and other animals, he adds.
"This is a remarkable study which uses molecular genetic approaches to visualize memory formation in a living organism. It demonstrates that, in this model system, short term memory involves the recruitment of new synaptic connections into pre-existing ensembles of synapses. It will be critical to determine whether similar principles control memory formation in higher organisms," says Robert Finkelstein, Ph.D., a program director at NINDS.
The researchers now plan to put fluorescent genes into a variety of other neurons of the brain in order to determine which ones respond to different kinds of stimuli. This will allow them to learn how the changes they identified affect higher-level neurons. They also hope to begin studying similar mechanisms in other animal models, such as mice.
The NINDS is a component of the National Institutes of Health within the Department of Health and Human Services and is the nation's primary supporter of biomedical research on the brain and nervous system.
*Yu D, Ponomarev A, Davis RL. "Altered representation of the spatial code for odors after olfactory classical conditioning: memory trace formation by synaptic recruitment." Neuron, May 13, 2004, Vol. 42, No. 3, pp. 437–449.
Journal
Neuron