Exploring the Mysteries of Memory
By Eric Sauter
We might as well start with the California sea slug, Aplysia californica—a large, hermaphroditic mollusk that Darwin described as "dirty-yellowish colour, veined with purple…"—for it was the sea slug that got Mark Mayford, an associate professor of cell biology and a member of the Institute for Childhood and Neglected Diseases at The Scripps Research Institute, interested in the study of memory in the first place.
"I was always interested in the brain and you can't really begin to understand how the brain works without understanding how memory works," he said. "But as a practical matter, it was seeing a lecture by Eric Kandel—he won the Nobel Prize in 2000 for his brain studies using sea slugs. That's because sea slugs have very simple circuitry that's easy to study—how are you going to figure out memory unless you know what synapses to look at? The human brain has trillions of synapses, which basically rules that out."
So, Mayford, a Midwesterner who did both his undergraduate and graduate work at the University of Wisconsin, moved to New York in 1997 to do his postdoctoral work in Kandel's laboratory at Columbia University. Aplysia californica captured his imagination long enough for one major paper on how the mollusk's synaptic connections became altered through the neurotransmitter serotonin. Married and with three young daughters, Mayford also met his wife in Kandel's laboratory—she did her postdoctoral work at Scripps Research.
From Columbia, Mayford first went to the University of California in San Diego, later joined Scripps Research in 2000 and moved away from sea slugs.
From Mollusks to Molecules
"My scientific direction has changed three times," he said. "I'm a molecular guy and it's hard to do molecular research with Aplysia. For one, the cells have to be grown in seawater. In the end, there were just too many molecular approaches that were impossible to take."
As a result, Mayford switched to mice during his time in Kandel's Columbia laboratory. The technology of producing genetically altered mice was just getting under way and Mayford found that an attractive lure. Also, there was the prevailing theory of memory at the time that held most everyone in its sway—long-term potentiation or LTP. This is the process of increasing the strength of the synapse or what Mayford calls "an activity dependent change in synaptic strength."
Neuronal activity follows a sensory stream, Mayford explained. For example, a beam of light falls on the retina, neurons in the eye react to this stimulus, and they connect with neurons in the brain. That communication starts a chemical cascade that leads to our perception of that beam of light. Because neurons have different capacities to fire at different rates, it depends on just how many synaptic inputs they're receiving; synapses are the points where neurons (and other cells) communicate to one another, the process through which the nervous system controls the rest of the body. If the neuron receives enough signals, the neuron fires and signals other cells.
This is where the process of memory begins.
"As neurons fire and release neuron transmitter," Mayford said, "it is believed that LTP strengthens the connections between the neurons by increasing the response of the second cell to the neurotransmitter. It's a phenomenon that is observable primarily in experimental environments and a great many people like it as the mechanism for learning and memory. It's what's we call Hebbian, that is, its works according to the theory that a learning mechanism should be dependent on combined cell activity—that memory would occur if the synapses between two cells grew stronger if they repeatedly fired at the same time."
The Hebbian theory is often short-handed as cells that fire together, wire together.
Hebbian or not, at least for Mayford, LTP had its limitations as well.
"There is data to support the role of LTP in memory, as well as data that complicates the notion that it is deeply involved with long-term memory," he said. "First, the study of LTP is part of an artificial process. It's a phenomenon people could study only if they looked at it in vitro—it's been observed in the brain only recently. For me, it's just too simplistic to say there is a one-to-one correlation between LTP and long-term memory."
Mayford's studies provide a framework that surrounds a number of molecules that are essential for long-term memory. His biggest problem—the problem of all memory research—is putting what he knows together to create a clear understanding of what's actually happening inside the brain as memory forms. For a molecular guy, this is as close as you can get to what poets might call deep blue mystery stuff.
Mayford has gathered up what he learned from the sea slug—his most recent paper was partly influenced by his work with sea slugs—and from LTP—which informs the various molecules that he manipulates within the 20-plus genetically altered mouse models he and his laboratory colleagues have created over the years—and has taken his own dive into that deep blue mystery.
Mayford has shown that there are certain molecules that are important to the formation of memory—certain kinases, neuroreceptors, and transcription factors—and he knows that they are important at the macro level. When these molecules get disrupted in the brain, memory gets destroyed.
His most recent work involves identifying the exact cells that are involved in the creation of specific memories and looking at whether a selected memory can be disrupted through single active neurons.
"We can pinpoint which regions of the brain regions are involved in certain neuronal activity," he said, "but I want to know which specific neurons are involved, how they change relative to the ones that are not active, and, eventually, how to manipulate them."
Brain Patterns 'R Us
When asked how he sees that sort of memory manipulation playing out in humans—the basis of nearly every Philip K. Dick short story, not to mention the bleeding heart of obsessive romanticism affecting every artist from Proust to Adam Sandler—Mayford simply won't go there.
"The brain patterns we study are memory," he said. "I don't really believe in free will. The brain is made up of a group of chemicals and our behavior is derived from those. Of course, we have complex behavior patterns. But is getting or wanting to get a coffee free will? When you get to the biology of it, it's hard to imagine how that could be. Are you controlling neuronal activity when you think? Not really. My words, my answer to your questions are a response stimulated by the question. Wanting a cup of coffee means I'm getting some sort of signal that I don't have enough caffeine in my system."
What Mayford is ultimately interested in knowing is the exact nature of the molecular changes that underlie the formation of memory because that is how the normal brain works. And from there, finding out the nature of all the things that go wrong in some illnesses—and, potentially, drug targets for therapeutic interventions. Schizophrenia, for instance.
"What is a hallucination?" he said. "It's probably just a bad pattern of neurons. At a cursory level, we want to learn how to manipulate these memories to see which neurons we need to study. And then we want to look at those neurons and how they change over time. Does memory involve changes in neuronal structure or is it something else? Why do some memories last a lifetime?"
Mayford is on the case.
Send comments to: mikaono[at]scripps.edu