Studying these cells could lead to new treatments for diseases ranging from gastrointestinal disease to diabetes.
For a long time, researchers thought that if neural connections in the brain weren't built during a certain critical developmental period, they weren't built at all. By adulthood, our brains were thought of as hard-wired computers with no ability to change.
But Center for Learning and Memory researcher Elly Nedivi, who is seeking to zero in on the exact genes used in learning and memory, points out that we are made of biological materials that do not last a lifetime. The brain makes new proteins all the time, and she has found molecules in the mature brain that clearly have to do with structural rearrangements.
These changes are not as dramatic as those that take place during early development, but they are there. This means that even as adults, the more we use our minds, the more robust they can be. In the long run, scientists may be able to "turn on" specific genes to allow parts of the brain to fill in for those that have been damaged by accident, congenital problems, surgery or senility.
"If you could identify genes that are expressed at much lower levels in, say, older brains, you can say that maybe part of the deterioration you see in older, diseased brains has to do with the fact that specific genes are not there. Then you could try to see if there's any way that you could increase their expression or compensate for them," said Dr. Nedivi, assistant professor in brain and cognitive sciences.
When we practice an action such as riding a bike or playing the piano or studying for the SATs, the connections between brain cells called synapses become stronger. Connections that do not get used eventually dwindle and disappear.
But Professor Nedivi wanted to know what is happening physically in these connections, and what makes a synapse that knows it's being used respond more strongly.
"What does it mean when we say that a connection is strong? Does it mean that there are more contact points between the neurons? Does it mean that with each signal there's more transmitter going out? Does it mean that the receiving side, the post-synaptic side, is more sensitive to this signal now? We want to understand the mechanism that creates strong synapses," she said.
"When we started, we had actually no idea of what we would be looking for. But we said we'll find out which molecules get turned on through brain activity, and through the kind of functions that these molecules carry out, we will be able to learn something about the mechanism."
PINPOINTING 'ACTIVE' GENES
Professor Nedivi and her research team thought long-term changes must involve "turning on" a single gene or handful of genes. The driving force for the changes must be the electrical activity that occurs when neurons are sending a signal or "talking" to other neurons. This activity can come from either sensory stimulation or from thinking.
The team looked for molecules that only give a signal when the neurons are active. The test they conducted, which turned out to be an order of magnitude more sensitive then conventional screens of this type, gave Professor Nedivi a big surprise: she found a whopping 360 genes that are turned on by brain activity, some of which might be involved in learning and memory.
"This made us change the way we thought about the problem. We realized that when synapses or connections were getting strengthened, you actually turned on this whole ensemble or gene set," she said.
"It would easier if there were just one gene involved, but I think that finding 360 makes more sense. Something as complex as learning and memory may need more than one type of threshold to be crossed in order to achieve a result. You may need different triggers for different genes or different levels of triggers.
"We found out that even though we were getting really a huge amount of genes turned on, it was not like every gene in your brain gets turned on at the same time when these neurons are active. It could be that the first stimulus turns on a subset of this group of genes that are kind of trigger-happy. Then, if you get a second stimulus within a certain amount of time, you get a second wave that could lead to longer term changes."
The type of molecules they saw the most were related to structural change, such as adding membrane and other synaptic components, or using proteases to eat away at the matrix to make room for growth.
"You have growth factors that induce the neurons to grow and signaling molecules that are related to that," Professor Nedivi said. "These things hinted to us that maybe what was happening, as a result of activity, was that there was actually structural change -- that when the synapse is getting strong, it means that it's actually building more contact. You can imagine how a synapse would be stronger if there were more contact points between the pre- and post-synaptic cells."
Out of the 360 genes, Professor Nedivi zeroed in on a handful, which she called candidate plasticity genes or CPGs. She experimented with overexpressing one of these in frogs and saw that it led to obvious structural changes in neurons. The research team is now setting up experiments in a mammalian system.
THE BIOLOGICAL COMPUTER
The fact that Professor Nedivi had uncovered structural changes in an adult brain was "a little bit surprising and somewhat controversial in the sense that for a long time, people have been thinking that the adult brain is hard-wired -- that any of these changes that we think about as learning and memory are happening at a diffused network level rather than as outright growth and new connections, new processes," she said.
Unlike a computer, the brain is made of proteins. "It cannot be that a connection that you are born with is going to be that same connection exactly for your whole life," Professor Nedivi said. "It means that even just to keep the connection as it is, you have to maintain it by making new proteins."
Like a well-used highway that gets repaired and widened more frequently than a quiet country lane, the maintenance appears to be regulated by the level of activity. An active neuron gets more general maintenance to the point where connections are actually increased.
While researchers agree that the adult brain has some plasticity -- there are examples of stroke victims recovering function in spite of brain damage -- seminal work has indicated that most of this gets established during the critical period, which for humans is the first few years of life.
Professor Nedivi says that while this period is indeed critical for setting up some of the brain's basic wiring, "there is probably just a little more degree of freedom than was previously thought," she said. "The good news is that you can induce that high level of activity reminiscent of the critical period.
"The more active you are, the more you're exposed to sensory stimulation and the more you use your mind, the more you can actually influence the strengthening of connections in your brain, even as an adult," she said.
This work is funded by the National Eye Institute, the National Science Foundation and the Ellison Foundation.
A version of this article appeared in MIT Tech Talk on May 31, 2000.