tufts universitytufts magazine issue homepage
contact us back issues related links
features Mr. Electric The Secret Lives of Elephants Robert Robert Robert The Toolbox of Self-Deception Test Kitchen Confidential columns jumbolaya planet tufts newswire the big day departments
Portrait of Michael Levin

Mr. Electric

In search of the switches that control how body parts form, heal, and regenerate

You are slicing onions. You miss. Your skin splits and bleeds, you run it under water, you bandage it, you wait. After the pain subsides, you probably give little thought to the miracle going on under the Band-Aid. Your cut finger knits neatly back together, replacing skin and blood vessels in a matter of days. Eventually, the wound disappears completely, and you forget it was ever there.

Regeneration is one of biology’s great mysteries. Why is it that humans can repair their skin after a cut, but can’t regrow a severed finger? Why is it that starfish and salamanders can regrow their limbs but a rabbit can’t? How can some worms regrow their entire bodies from microscopic pieces of tissue, wriggling around afterward with the same knowledge they had before they were sliced?

“People have this idea that regeneration is for primitive organisms,” says Michael Levin, A92, director of the Tufts Center for Regenerative and Developmental Biology. “The fact is, it is sprinkled very capriciously throughout the tree of life. You can have a worm that regenerates everything, then you can have a very close relative that regenerates nothing. A deer can regrow meters of bone, velvet, and vasculature every single year, producing a new set of antlers in the same pattern.” Even humans have more ability to regenerate than is commonly recognized. Children under the age of eleven have regrown fingertips, for example, and adults can regenerate their livers from a small piece.

The riddle of regeneration has tantalized Levin since before he was an undergraduate at Tufts two decades ago. Now, as a brand-new member of Tufts’ biology department, he is in pursuit of the cellular switches that could allow us to regrow any part of our bodies. “You did it as an embryo,” he insists. “Forty years ago, my cells knew how to grow every structure in my body. That information has not gone anywhere. It’s still there.”

The quest for the magic switch has led developmental biologists to hunt for clues in gene expression and chemicals within cells. Virtually alone, Levin’s lab has been searching for the key in an altogether different area: electricity. The idea is not completely new. Scientists have known for more than a century that tiny electric charges, on the scale of millivolts, are emitted from skin and bone when those tissues regenerate after a wound such as that cut you got slicing onions. Along the way, however, most scientists lost interest in the charges, focusing instead on the biochemical pathways uncovered through traditional molecular genetics. Levin has pioneered research on the molecular level to show that not only are these electrical flows significant—they might also be the signals that control how and why regeneration occurs.

To understand the mechanism is to put some of that control in human hands. “By learning about the way these bioelectrical signals are naturally used to determine biological shape,” Levin says, “we can capitalize on this to alter structure rationally.” Already, his lab has used tiny electrical jolts to alter regeneration in worms, frogs, and chickens. Applied to humans, he says, the same techniques could one day repair structures damaged by birth defects, regrow amputated limbs, or even reverse the growth of cancerous tumors.

Levin talks with the easy confidence of a scientist who is used to doing the impossible. His casual dress and laid-back demeanor—he is wearing a hoodie when I meet him, and has a Simpsons blanket draped on the back of his chair—make him seem younger than his forty years. His piercing green eyes, however, match the intensity with which he pursues his research. “Probably without exception, everything of significance that we have done pretty much everyone had told me wasn’t going to work,” he says with a smile. “Everybody’s first reaction is ‘This is never going to work.’ Sometimes they are right, but if you push hard enough and pay attention, I think you can make it work.”

Prodigy may not be too strong a word to describe Levin’s early career as a scientist. He lived in Russia until he was nine years old, and even at an early age, he was fascinated to learn how complicated structures were put together. “When I was four or five, we had this gigantic black-and-white TV, and my father would take the back off so I could see what was in there. I distinctly remember thinking that all that stuff didn’t end up there by accident. Somebody knew how to put it together so that it worked. Whoever did that clearly had something special going on.” As a student at Swampscott High School on Boston’s North Shore, Levin got deep into computers. He learned difficult programming languages and made money as a designer of software for data visualization and robotic control.

Continuing in computers as an undergraduate at Tufts, he turned his mind to modeling complex systems. Yet the more he learned about computer science, the more he saw how much the field had to learn from the complexity of nature. “It’s clear that the biological world is leaps and bounds beyond anything we can duplicate in engineering,” Levin says.

With virtually no biology behind him, he approached the developmental biologist Susan Ernst to do an independent study in her lab. “The first day he came in to talk with me about the possibility of doing a project together, he had a notebook filled with reference after reference,” Ernst recalls. “It was very easy for me to see that Michael was special. He had been thinking about living systems in a different way than most people were at the time.” Coming out of physics and engineering, he was more interested in knowing what information cell groups exchanged during an embryo’s self-assembly than in studying how specific genes influenced development, the vogue in developmental biology at the time.

He took magnetic coils—borrowed from the physics department—and wrapped them around sea urchin embryos to observe how electromagnetic waves affected their growth. The results showed those kicks of magnetism could change the animals’ rate of cell division and early development. Before Levin, no student in Ernst’s lab had been the lead author on a research paper; Levin was the lead author on two of them.

After graduating from Tufts, Levin carried on his exploration of cell communication as a Ph.D. student in the genetics department of Harvard Medical School, in the lab of the developmental biologist Cliff Tabin. At the time, Tabin’s lab was doing research on genes that control the development of the nervous system and limbs of chick embryos. One of the molecules involved in the process enabled Levin and collaborators to identify how a chick embryo knows its right from left in early development—for example, how it knows to put the heart on one side and the liver on the other. That might seem like a simple distinction, but it’s a thorny problem for a developing embryo. While “top” and “bottom” can be determined by gravity, there is no external force that can says which side is which.

Previously, developmental biologists thought that left and right must be determined later in development, when organs are being formed. But a postdoc in Tabin’s lab discovered that a gene known as Sonic hedgehog (so called because it gives fruit fly embryos a spiky appearance) was expressed in a tiny spot only on the left side very early on. Levin volunteered to look into the phenomenon. Indeed, he found a number of consistently left-sided and right-sided genes and showed that the cascade in which they progressively turned each other on and off ultimately determined the position of the heart, stomach, and other internal organs.

In his postdoctoral work with Mark Mercola, in the cell biology department of Harvard Medical School, Levin traced the process back to an even earlier stage of development, discovering that the genes themselves know which side to express themselves on because of flows of electric charge across the embryo. Cells were known to have “gap junctions,” little submarine hatches that line up to allow small molecules to pass from one cell to another. What Levin found is that these gap junctions line up in a continuous path across the entire early embryo. They form a conveyor belt that passes signals all the way across. Interrupt this long-range communication path—as Levin did—and organs grow on either side of the body at random.

Levin formed a hypothesis: this path along which the left and right sides communicated was the equivalent of an open electrical circuit, pointing the way to the right side of the animal as surely as a neon arrow. Remarkably, this process occurred when the embryo had just sixteen cells—long before the cascade of genes that ultimately directs the organs of the developing animal, and much sooner than biologists had previously thought possible.

Like Susan Ernst at Tufts, Levin’s advisers at Harvard were impressed by his willingness to follow a few vague clues down the road of a particularly risky hypothesis. “He is one of the most creative people I have ever had in my lab,” says Tabin. “He enjoys working on the frontiers of science.” But, Tabin adds, the frontiers are inherently risky. Levin “is more likely than most people at his stage to do things that are fundamentally important and have a real breakthrough. He is also someone who could slave away at a problem that turns out to be less significant than he had hoped.” Lucky for Levin, his experiments have mostly been the former—though he admits there have been blind alleys that haven’t led to publication.

After his success in explaining the electrical flows behind left-right patterning, Levin went in for a closer look at the battery itself. He began with a simple question: If the movement of charges could determine the shape and position of organs during early development, what would happen if you artificially changed the pattern of these electrical signals after injury? This round of experiments used the African frog Xenopus, a muck-dwelling amphibian that, like most frogs, can regenerate its tail and legs while it’s a tadpole but not as an adult frog.

On a tour of his lab, Levin ushers me into a swampy-smelling room he calls the “Frog Farm.” Inside, a couple of hundred frogs laze in tanks filled with greenish water. “They probably just got fed,” Levin says, pointing out a shelf full of granular frog chow. “They are just chilling out.” A room next door is piled with Tupperware containers holding the frog’s progeny—transparent tadpoles that are the real focus of the experiments.

For years, Levin and his fellow researchers have been inserting RNA into cells of the developing frog embryo. The RNA produces proteins called ion channels, or similar structures known as proton pumps, which flick ions across the cell membrane, creating an excess of positive or negative ions in the cell. This dramatically alters the cells’ electrical properties. Cutting off a tadpole’s tail, Levin and his fellow researchers found that they could stop it from growing back if they inserted RNA that produced a mutant protein, one that prevented a particular enzyme— V-ATPase—from pumping hydrogen ions.

More impressively, Levin’s lab also showed that by inserting a proton pump into the cells surrounding a wound, they could spark regeneration later in development, past the stage where tadpoles can normally sprout new appendages. This was the first demonstration that muscle, spinal cord, and blood vessels could be completely regenerated—arranged in the right pattern in a normal appendage—by the molecular tweaking of bioelectric properties.

Not stopping there, the researchers experimented with other ion transporters, and hit pay dirt with a particular potassium ion channel that allowed the tadpole to grow extra legs or eyes. Outside Michael Levin’s sunny office near the Medford/Somerville campus is a large photo of a six-legged frog, just one of the creatures the lab has created by inserting these channels in multiple cells of the developing organism. Determining exactly which transporters cause the development of which organs will require further research. “Right now we don’t have a good handle on how we tell it to grow an eye rather than a leg or a tail,” Levin says. “But ion flows and electrical properties seem to induce different shape-changing events.”

Frogs aren’t the only animals that Levin has successfully manipulated. In another room in his lab, he opens a refrigerator to reveal more stacks of Tupperware, which seem to contain grains of dirty rice, but are actually full of flatworms called planaria, wriggling in water. “Poland Spring, in fact,” he deadpans. “I tried to get money out of Poland Spring because it turns out the flatworms refuse to live in any other kind of spring water.” These worms can grow back their heads and tails after being cut into pieces, but how they “know” which structure—head or tail—to grow at each end has been a mystery. As with the frog embryos, Levin was able to show that regeneration depends on physiological signals moving long distances to tell the wound what to do. Once the tail is cut off, signals flow through gap junctions from the wound to the head to ensure that it is still there. Block that flow, and the worm will grow another head. To grow a tail, you block another ion transporter. Just by making these electrical changes, Levin’s group has been able to grow worms with tails on both ends, heads on both ends, or even four-headed worms from multiple cuts.

Such machinations are more than just idle curiosity—by disturbing the natural process, the scientists can begin to understand how it works, and ultimately how it can be controlled. “Somebody said it’s like figuring out a magic trick. The magician does a sleight of hand, and you say, ‘My God, he has a pigeon in his hand,’ but you learn pretty much nothing. The minute the guy makes a mistake and you catch a glimpse of the pigeon up his sleeve, now you are starting to see how the trick is done.” In this case, the pigeon up the sleeve is the six-legged frog or the two-tailed worm. “If you know how to build a structure, then you know how to fix it when something goes wrong from birth defects,” Levin says. “If you know how to make a six-legged frog, that means you may know how to make a leg grow in adulthood.”

Although Levin’s group is the first to integrate bioelectrical techniques with molecular biology, they are by no means the first to explore the effects of electrical charges on animals. As far back as 1786, the Italian scientist Luigi Galvani discovered that animals reacted to a charge when he noticed that the muscles of a dissected frog twitched whenever he touched a nerve with his electrostatically charged scalpel. (Among those inspired by his work was a young novelist named Mary Shelley, who used it as the basis for reanimating a human corpse in her book Frankenstein.) Other Victorian scientists were able to expand on his work to show that bioelectrical charges had other effects on living systems, such as speeding the healing of bone fractures. After the turn of the twentieth century, the biologist Thomas Hunt Morgan confirmed that animals had inherent electric charges when he measured electrical polarity in earthworms. In the 1920s, a Danish scientist, Sven Ingvar, showed that when chick embryos were placed in an electric field, they grew toward the charge.

By 1950, though, research into bioelectricity fell by the wayside when the field of biology as a whole took two divergent paths. The cutting edge focused on biochemistry, leading to modern molecular biology and genetics; the other, less “sexy” offshoot focused on physiology, continuing to explore bioelectricity only in the case of the electrical signals transmitted by neurons in the nervous system.

In the 1970s and 1980s, some scientists, including the physiologist Lionel Jaffe, worked to put developmental biology and bioelectricity back together. Jaffe and his student Richard Borgens, for example, showed that limb regeneration in frogs and newts could be spurred by applying electric currents. But they hit a wall when they couldn’t explain how the electricity actually caused limbs to regenerate. “We had a relatively limited toolkit, and this is where Michael Levin comes in,” says Ken Robinson, a professor emeritus at Purdue University who was active in the field in the 1980s. “He put together the tools of modern genetics and molecular biology with the physiology.” The two collaborated on many experiments after Levin finished his postdoctoral work and began his own lab at the Harvard-affiliated Forsyth Institute.

Malcolm Maden, a regeneration biologist at the University of Florida who specializes in using chemical cues to explore regeneration in salamanders, says Levin “has reignited the field again—with explanations, which is always better.” He theorizes that bioelectrical cues precede the kind of chemical stimuli he himself has studied.

Of course, adding limbs to a frog is a far cry from regrowing limbs in humans. But Levin’s experiments provide a case that, theoretically at least, it could be done. If a tadpole can be made to grow a tail past the point of development where it’s ordinarily possible, then people could be made to regrow their fingertips beyond childhood.

Levin and other Tufts researchers—such as David Kaplan, the chair of biomedical engineering—have begun tackling bioelectricity in humans. In research published last November, Levin’s lab, in collaboration with Kaplan and his doctoral student Sarah Sundelacruz, showed that adult human stem cells—those basic seeds of all body tissues—undergo changes in electric charge before they differentiate. By inserting ion transporters into stem cells, the researchers were able to control when the cells changed into bone or fat.

Levin believes such control could be extended to mature human cells. With the right genetic key, cells could be returned to a plastic state in which they would develop skin, blood, or bone just as surely as a tadpole regrows a tail or a deer sprouts new antlers.

It’s an enticing proposition, this idea that humans could regrow legs, eyes, or fingers damaged in an accident. Levin talks about the possible future applications of these bioelectrical techniques as excitedly as a four-year-old child discovering the inside of a television set. “Once we know what the signals are, you can provide the right signals at the right time to cause the tissue to grow,” he says. “Our job here is to figure out what those signals are.” Such knowledge would of course transform medicine. “What we currently do as a society,” Levin says, “is invent increasingly complicated and expensive ways to patch up a sinking ship as the patient ages.” How much more sensible to replace your worn-out systems with brand-new healthy ones.

And just as Levin’s work could give medicine the power to make organs grow, it might also provide the power to stop organs from growing when tissues turn cancerous. “You can view cancer as a sort of disease of geometry,” says Levin. “A tumor is a collection of cells that has failed to obey the patterning instructions from the rest of the body and is growing out of control.” His lab has already created cancer-like cells in Xenopus frogs by disrupting the flow of potassium ions in embryonic stem cells. “Obviously, you want to go in the opposite direction,” Levin says, “but if you can figure out the switches, you can learn how to control them.”

But as promising as all these discoveries are, Levin admits that bioelectrical impulses are only part of the regeneration puzzle, which most scientists agree will require electrical and chemical cues to work properly. So far, Levin’s work suggests that the electrical signals often occur first, that they are important “control knobs” that initiate the process. The bigger question, so far unanswered, is how they interact with the later molecular cues that tell the developing body part exactly how to form—with, say, skin on the outside, bone on the inside, and blood and guts in the middle.

Cliff Tabin of Harvard Medical School, for all his apparent pride in his former Ph.D. student (he calls Levin’s work “very significant” and contends that developmental biology “is very different for Mike being in it”), reserves judgment about the importance of bioelectric signals. “I wouldn’t discount that they play some role,” Tabin says. “The question for Mike is, Are these things where the key decisions are being made? Or important, but not the lynchpins in the process? The only way to find out is to make some experiments.”

No matter what those experiments find, the research itself satisfies Levin’s hunger to understand how complex systems work. “Even if we were to find out this has nothing to do with humans whatsoever, that makes not a whit of difference to me,” he says, not entirely convincingly. “This is something we are going to want to know about, how living tissues process information. And if humans don’t do it, that’s fine. It is fascinating for other reasons.” Still, if these techniques do eventually enable humans to grow back an arm or cure a cancer, it wouldn’t be the first time Levin did something that everyone told him couldn’t be done.

MICHAEL BLANDING is an award-winning magazine writer whose work has appeared in The Nation, The New Republic, Boston Magazine, and the Boston Globe Magazine. “Mr. Electric” is his second cover story for Tufts Magazine.

  © 2009 Tufts University Tufts Publications, 80 George St., Medford, MA 02155