Harvard University - Department of Molecular & Cellular Biology


by Patricia Thomas

April 6th, 2004

Josh Sanes

Jeff Lichtman

(click on image to see large version)

The synapse is where the magic happens.

Information zaps across the synapse from one neuron to the next, racing through brain circuitry that enables us to savor the aroma of freshly brewed coffee, count the cash needed to pay for it, and agonize about whether the beans were fairly traded. In the rest of the body, nerve cells use synapses to tell muscles to lift the cup and take a sip.

Joshua R. Sanes and Jeff W. Lichtman, senior faculty recruited in the Department of Molecular and Cellular Biology and the first recruits for Harvard's Center for Brain Science, are synapse experts. Sanes has made key discoveries about how these vital connections are formed, while Lichtman's forte is understanding how they are eliminated. For 15 years the two followed separate paths even though they were professors in the same department at Washington University School of Medicine in St. Louis.

Then, about six years ago, something clicked. The two scientists realized that imaging technologies pioneered by Lichtman could be used to visualize synapses taking shape and disappearing in live transgenic mice developed by Sanes. Together they have published a series of revelations about events at neuromuscular synapses, where the nervous system talks to muscle fibers. At Harvard they'll continue this work and expand their new endeavor: watching neurons communicate with neurons in the brain itself.

"When two senior scientists see that they have such a synergism, it's easy for a department to see the strength of that synergism as well," says Andrew P. McMahon, chairman of the Department of Molecular and Cellular Biology, where Sanes and Lichtman will hold faculty appointments. Separately, each has "provided new and surprising insights" into how synapses work, McMahon says. Together, they are a force to be reckoned with.

Sanes was elected to the National Academy of Sciences in 2002 and holds an endowed chair in neurobiology at Washington University. At Harvard, where he will be a professor of molecular and cellular biology in MCB, he will be the first director of the Center for Brain Science. Lichtman is a professor of neurobiology and director of the neurobiology graduate program at WUSM and will be a professor of molecular and cellular biology in MCB at Harvard.

McMahon has known Sanes for years and calls him "an engaging, enthusiastic, dynamic individual" whose broad grasp of neuroscience will benefit MCB as a whole. Right now there are few "card-carrying systems neuroscientists," and McMahon is confident that Sanes will be able to recruit top-notch people for MCB in a highly competitive environment. MCB faculty and students will also benefit greatly from Lichtman's imaging expertise. He is in the forefront of a technology revolution that makes it possible to witness molecular and cellular events as they happen, in living organisms rather than fixed samples.

Sanes and Lichtman arrived at their partnership by different routes. Josh Sanes says he was "just a little boy" when he became fascinated with schizophrenia, not because he knew anyone with this disease but because he read about it. As a Yale undergraduate he was drawn to both psychology and biochemistry, and says, "When I found neuroscience, I went 'Wow! This is just what I'm interested in!' I just didn't know it had a name." Sanes came to Harvard for graduate school in neurobiology, earning his Ph.D. in 1976. Synapse formation captured his attention during his post-doctoral years, and he focused on the interaction of neuron and muscle cell because there were as yet no tools for observing the brain itself.

Sanes entered neuroscience at the dawn of the molecular era, when the sum total of knowledge about the synapse was small: researchers knew that after an axon touched a target cell, the axon tip specialized to release neurotransmitters and the place it touched specialized so it could receive those signals. The details of this remodeling were unknown.

Advances in molecular biology, genetics, and imaging have since enabled Sanes and others to find many of "the nuts and bolts of synapse formation," including some molecules that stimulate synapse development and others that dictate how long these connections will last. It turns out that two-way signaling is needed to form a proper synapse. Agrin, for example, is an essential protein that must flow from the axon tip to the muscle. Another protein, called b 2 laminin, flows from muscle to axon tip and helps form the synapse, but isn't absolutely necessary.

Sanes distinguishes critical proteins from those "just along for the ride," by adding or inactivating specific genes in special strains of mice. All this manipulation of the mouse genome gave Sanes' team expertise that proved to be a magnet for Lichtman.

Jeff Lichtman spent the first 15 years of his career observing neuromuscular synapses in ordinary laboratory mice; like other researchers he would have preferred to study synapses in the brain, but these were inaccessible in live models. Lichtman became a neuroscientist because he wanted to know how the brain of a newborn baby, unable to do anything well at first, matures into the "magnificent and highly specialized device that is an adult human brain."

Years of observation have given Lichtman some surprising, even revolutionary, ideas. Ask most people to speculate about how a baby learns, and they will say that new connections are forged in the brain. On the contrary, Lichtman says, babies' brains start out "hyper-connected" - cells have so many competing synaptic inputs that they don't know which to obey. Over time, synapses that aren't used or reinforced fall away as the busiest, most adaptive ones grow stronger. "As we learn, we begin to focus our synaptic connections on a subset of cells and we get very good at certain things, but the price we pay is losing our ability to do others," Lichtman says.  

Neurobiologists used to believe that the life of a synapse was far shorter than the life of a human being, and that synapses constantly formed and disappeared at about the same rate in the adult brain. But Lichtman's meticulous observations show that many synapses last a lifetime, or close to it. This is good when these connections are crucial for walking, or remembering the names of loved ones. Pruning still goes on, however, which explains why a person's interests narrow with age and how Alzheimer's disease or some other neurodegenerative process can gain the upper hand.  

Synapse elimination caught Lichtman's fancy when he was an M.D.-Ph.D. student at Washington University in the late Seventies. In contrast to most biologists of his generation, who run hypothesis-driven, experimental labs, Lichtman is a classic descriptive biologist. When dealing with something as "insanely complicated" as the brain, he subscribes to Yogi Berra's famous maxim that "You can observe a lot by watching."

Lichtman has now been watching synapses in an era marked by rolling revolutions in imaging technology. He has been among the first to catch each of these waves, as light microscopes familiar to high school biology students were joined by electron, confocal, two-photon, and other increasingly exotic microscopes. Today, Lichtman uses videomicroscopy to capture dynamic pictures of synapse activity in live animals.

"Jeff Lichtman is acknowledged to be one of the world's leaders in imaging," says Harvard MCB chairman Andy McMahon. Lichtman says he found the invitation to join the new Center for Brain Science irresistible because one of his mandates "is to use the generous resources of Harvard to create a state-of-the-art imaging center, perhaps the best in the world."

Lichtman and Sanes joined forces in the late Nineties to solve a problem that plagued both their labs: in the nervous system's tightly packed tangle of wiring, it is nearly impossible to determine exactly where the long, snaky axons of a single neuron are going and what cells they are talking to. Lichtman wanted to watch axons and target cells interact; Sanes wanted to manipulate them. "We agreed that a good starting place would be to make some mice where neurons glow," Lichtman recalls.  

Sanes' expertise in mouse genetics enabled the collaborators to insert the gene for green fluorescent protein (GFP), a widely used laboratory tool, into mouse neurons. Permutations soon followed, and today they have 30 lines of mice with neurons that glow in four different colors and a startling array of patterns. In one of these transgenic strains, all the branches of a single neuron glow bright yellow against a dark background - making it easy to see what cells these axons are communicating with.

In another line of mice, Lichtman has been able to record a dramatic competition between a yellow axon and a blue one, as two neurons fight to maintain a synapse on the same target cell. This struggle lasted for days, with the lead going back and forth, until one suitor vanquished the other. What the winner did next came as a total surprise. "The remaining neuron instantaneously took over the same synaptic sites that had been occupied by the axon that left," Lichtman notes. "It looked like the invasion of one neuron into the territory of the other." Another study revealed that losers are eventually consoled: a neuron that fails on one front will eventually make a match elsewhere. Like people, Lichtman says, "all neurons win some competitions and lose others."

At Harvard, Sanes and Lichtman will develop technologies for studying larger numbers of synapses simultaneously, so they can understand how neurons link together to form circuits. The long-term goal of systems neuroscience is to connect these circuits and bridge the huge knowledge gap that separates physical events in the brain from complex human behaviors, Sanes says. They'll start with more manageable gaps, such as how a honeybee finds its way back to the hive or why a mother mouse nurtures her babies. Even these may take a decade or so to work out, he predicts, "but I think these are solvable problems that are plenty interesting."

Technical innovation is an important part of their mission. The two are already working to develop what they call "brainbow mice" - animals where sophisticated genetic manipulations will allow them to observe the activity of as many as ten neurons at once, each glowing a different color. In the more distant future, Lichtman believes that infrared and ultraviolet labeling, combined with more powerful microscopes and computers, will open even bigger windows into the life of the brain. Sanes also hopes that nanotechnology will yield tiny probes that can monitor the activity of hundreds of neurons at once.

Ultimately, what Sanes wants is a panoramic view of the brain. "Too much biology has been devoted to mapping a lot of trees and a lot of leaves without taking a good look at the forest," he notes. It's no use to try and explain behavior by making a "wiring diagram" of nerve cells and their connections, Lichtman agrees. Investigators who work with the tiny nematode, C. elegans, demonstrated this by laboriously tracing the connections of every nerve cell in this simple creature - only to realize they were no closer to understanding its behavior.

What systems neuroscience must do, Lichtman says, is uncover the fundamental principles that organize the brain. Although there are many different types of neurons and numerous specialized brain regions, "having a better understanding how one kind of circuit is put together can be generalized, and used to postulate how learning, action, and sensation all take place."

View Jeff Lichtman's Faculty Profile

View Joshua Sanes's Faculty Profile