Master Plans

First, they found that the same genes that make a wing in a fruit fly also make a wing in a butterfly. “So we saw that these genes have deeply conserved jobs in building limbs,” says Carroll. But the big surprise came next. To address Nijhout’s original question about spots, Carroll and company then looked to see which genes are turned on in a butterfly chrysalis a week or so before it emerges to unfurl its colorful wings. And they discovered, much to their surprise, that one of the master genes involved in making limbs, a gene called Distal-less, is also switched on in the center of every developing eyespot, one of the wing’s most dramatic markings.

“It just smacked us in the face,” says Carroll. “Here was this ancient gene, doing something entirely new.” In addition to its regular duty assembling the wing, it also graced the wing with spots. “So the message,” says Carroll, “is that you can teach old genes new tricks.”

And as an HHMI investigator at the University of Wisconsin, Madison, Carroll has shown again and again that much of the physical and anatomical diversity we see in the creatures around us is generated using the same genes—and their associated regulatory networks—in novel ways. What’s more, he’s found that when evolution tinkers with animal form, it often does so by tweaking the switches that control gene expression.

“Sean got a lot of notoriety for showing that the same genes involved in the patterning of limbs were co-opted to give the patterns on butterfly wings,” says Cliff Tabin of Harvard Medical School. The study made the cover of Science in 1994 and was picked up by the New York Times. “But regardless of popularization,” says Tabin, “Sean is without a doubt one of the leading people in pattern formation during embryonic development and in evolution—and he has been a true pioneer in bringing those fields together.”

“He’s really been at the forefront of figuring out the genetic and molecular changes that have led to morphological diversity,” adds Jim Skeath of Washington University School of Medicine, a former student. And he’s written eloquently about the topic in textbooks and books aimed at the general public. Carroll’s work—and his words—“have changed the way we think about how bodies evolve,” says Neil Shubin of the University of Chicago.

Since boyhood, Carroll has enjoyed collecting critters of all stripes, and has even been known to whip out a live snake during a seminar to make a point about patterning. But Carroll actually cut his academic teeth in immunology. After speeding through his undergraduate studies at Washington University in two years, he landed at Tufts, where he generated antibodies that allowed him to probe the structure and function of eukaryotic RNA polymerase II. “It was great training,” says Carroll, who took three-and-a-half years to complete his thesis. “The immunological tools were great weaponry to have as the years went on.”

When he joined Matt Scott’s lab at the University of Colorado, Boulder as a postdoc in 1983—at the age of 22—Carroll aimed that weaponry at developing fly embryos. Scott had isolated genes belonging to the Antennapedia complex, a set of genes that control the normal development of the fly thorax. Mutations in Antennapedia cause flies to form legs where their antennae should be. Although Scott had cloned the genes themselves, Carroll says, “nothing was known about their expression or their function.”

Which is where Carroll’s immunology skills came in. “Sean started preparing antibodies against the still-mysterious protein products of these genes,” says Scott, who’s now at Stanford University. “And he used the antibodies to figure out when and where the genes were active and where the proteins were present.” And he did it right. “He ended up with antibodies to die for,” says Allen Laughon, another Scott postdoc who wound up on the faculty at the University of Wisconsin. What’s more, he says, “Sean and Matt understood that with these beautiful reagents, they could look at gene expression not just in the wild-type embryo, but in various mutants.”

The approach allowed them to begin to piece together gene regulatory hierarchies and cascades, says Laughon, “and it’s a strategy that people have used ever since to figure out which genes regulate which.” So, for example, Carroll and Scott followed the expression of the Ftz gene in a series of mutants defective in body segmentation, and they were able to identify which genes act upstream of Ftz—to regulate its expression—and which genes were its downstream targets.

“I remember fondly the first day we saw the striking pattern of stripes made by Ftz,” says Scott. “We popped open some champagne and the corks made dents in our rather flimsy ceiling.” It became a lab tradition—labeling the resulting marks with the dates of notable experimental successes. “We ended up with a highly decorated ceiling,” says Scott. In fact, Carroll alone published 10 papers based on his work in Scott’s lab, including three in Cell and one in Nature. “It was a very fertile time,” says Carroll.

In his own lab in Madison, Carroll continued to chase the genes that generate body parts, especially “adult” features like wings, legs, and body color patterns. He looked in a menagerie of creatures in addition to flies and butterflies, and found, for example, that the Distal-less gene—and the regulatory elements associated with it—controls outgrowth of all sorts of limbs across the animal kingdom, including the legs of flies and butterflies, the feet of marine worms, and the siphons on sea squirts. “If you want to make a novel appendage, you just take this outgrowth circuitry and snap it into place,” says Tabin.

But how do animals use the same genes and genetic circuits to do different things? The key, Carroll has found, lies in the regulation. “Body building genes play many different roles at different times in development. So they generally have vast regulatory regions,” he says. And those regions might contain a dozen or more switches that include different binding sites for the sets of regulatory proteins that control when and where a gene is expressed. Evolution, then, fine tunes the function of these genes by fiddling with these regulatory switches.

“Sean was one of the people who really championed the idea that you could change morphology by changing the cis-regulatory elements of developmental genes,” says Yale’s Scott Weatherbee, Carroll’s former student. “And he’s pushed it to the forefront to the point where folks take it for granted that that’s how different morphologies arise.”

The earliest proof for this paradigm of cis-regulatory evolution came from studies the Carroll lab conducted on pigmentation in different Drosophila species. “Fruit flies differ remarkably in body pigmentation,” says Antonis Rokas of Vanderbilt University, a former postdoc. “Some species are all black or tan or yellow. Others have various kinds of spots or pigments or stripes. And Sean’s lab has been systematically dissecting the genetic basis for that variation.” The team has shown, for example, that changes in the cis-regulatory sequences of a gene called Yellow are responsible for the patterns of spots—or the lack thereof—on the wings and bodies of Drosophila melanogaster and related species.

The work was fueled, in large part, by Carroll’s enthusiasm and determination. “Sean has a real instinct for identifying the most informative experiments,” says Patricia Wittkopp of the University of Michigan, another former student. “It doesn’t matter if you’ve never done it before—or if others have failed. If it’s the definitive experiment, then Sean encourages you to push hard to try to get it to work.” Like her colleagues who cloned the butterfly genes, Wittkopp had to develop tools for studying the regulation of Yellow in different Drosophila species. “That really was the strength of the paper in the end,” she says of the resulting article, published in Nature in 2005.

In addition to being experimentally rigorous, articles from the Carroll lab are also visually captivating. “Sean knows how to grab a person’s attention, give them a compelling story,” says Laughon. “And what could be more compelling than a gorgeous and informative image of a fluorescently stained butterfly wing or fly embryo?”

Not to mention—the guy knows how to write. “He’s just phenomenally clear—and fast,” says Skeath. “I’d give him a draft and within an hour he’d give back comments. It got to the point where I would wait until I thought he’d left for the day before dropping off my draft, just so I could get away from it for a night. He’s just off the standard curve with regards to writing.”

And perhaps with regards to where the papers appear. Indeed, publishing in Science, Nature, or Cell was the goal, says Jim Williams of the Nature Technology Corporation in Lincoln, Nebraska, a former postdoc whose work led to back-to-back cover articles in Nature and Science in 1994. “In a way, anything not in those three journals was a bit of a failure. Because if it wasn’t global enough to be of general interest, you didn’t get to the big picture.”

Rokas agrees. “Sean used to say that life is too short to spend it trying to answer trivial questions. He’d really force you to think big and to try to find problems that are truly exciting,” he says. “I think that’s why so much of his work appears in these high-caliber journals: the studies are designed to be high impact from the get-go.”

Which is good for Carroll’s trainees—and for folks in the field. “He’s one of those authors that when you see one of his papers, you make a point to read it,” says Shubin. “Because you know it’s going to be something that will change the way you think.”