WHERE DO TOES COME FROM?

By J. Madeleine Nash/Chicago

About 360 million years ago, as any schoolchild who knows his prehistoric zoology can tell you, some adventurous fish managed to hoist themselves onto their stubby fins and crawl clumsily out of the swamps to forage for food. Once these primeval creatures were on terra firma, their offspring began to adapt to their new environment, natural selection (over tens of millions of years) favoring those that developed features well suited to life on land: paws, hooves, knees, joints, fingers and thumbs. Thus, as generations of schoolchildren have learned, did these marine creatures give rise to frogs, birds, dinosaurs and all the rest.

There's one problem with this familiar version of how our distant ancestors emerged from the sea: it's probably wrong. For one thing, newly assembled fossils -- in particular, a 360 million-year-old salamander-like aquatic animal called Acanthostega -- strongly suggest that toes and feet were developed before life climbed onto land, not after. Moreover, in shape and function, Acanthostega's fully jointed toes bear no resemblance to the spiky, fanlike fins of a fish. Scientists believe they understand how a fish's gills evolved into an amphibian's lungs. But how did fins turn into feet like these?

The answer may be in the genes. That's the tantalizing conclusion of a team of researchers from the University of Geneva in Switzerland. They have discovered that genes associated with the formation of fins in fish are the same ones that orchestrate the development of paws in mice. "Think of a mouse as a fish with limbs, or a fish as a mouse with fins," says University of Geneva developmental biologist Denis Duboule. "What a mouse does is take a fin and put something extra on top of it."

That something extra, Duboule and his colleagues suggest in the journal Nature, is provided by a special set of genes that act as master architects in a surprisingly broad range of animals, from rodents to roundworms. These gossamer strands of DNA -- known as homoeotic homeobox genes, or Hox genes for short -- lay out the embryo from head to tail, controlling everything from the development of limbs and the wiring of the spinal cord to the patterning of the gut and urogenital tracts. "What's amazing," says University of Pennsylvania paleontologist Neil Shubin, "is that evolution of complex structures appears to be controlled by this same small set of genes."

How do Hox genes pack such power? The DNA in all genes carries instructions for assembling proteins out of chemical building blocks called amino acids. What sets the proteins made by Hox genes apart is the biochemical motif known as a homeobox, a stylized string of 60 amino acids that enables Hox proteins to stick to DNA like strips of molecular Velcro and, in the process, activate still other genes. Hundreds of genes belong to the extended homeobox family, but those that are also homoeotic -- associated with changes in body parts -- are the most important. Though they are few in number (38 out of an estimated 50,000 to 100,000 genes in modern vertebrates), the Hox genes control much of what happens during embryonic development.

Only during the past decade have scientists begun to tease apart the mysteries of Hox genes. Clustered in groups of eight to 11, on as many as four chromosomes in a developing embryo's cells, these genes switch on and off in sequence. Since embryos mature from the top down, explains biologist Cliff Tabin of the Harvard Medical School, a Hox gene that turns off a bit early, or stays on just a touch longer, can make a dramatic difference in the formation of the embryo. Swans, for example, have more neck vertebrae than chickens and thus longer necks. That is because the Hox genes responsible for making neck bones stay on longer in the unhatched cygnet than in the unhatched chick.

Timing may also explain the progression of fins to feet. In tetrapods (four-legged animals), feet do not grow straight out of the leg, proceeding from the ankle out, but develop in a fanlike progression that runs from the smallest digit to the largest. In Geneva, Duboule and his colleagues tracked the activity of four Hox genes in the budding feet of embryonic mice and found precisely this pattern. By contrast, studies showed that in the zebrafish, the Hox genes switch off earlier, perhaps to ensure that a flexible fin ray (useful for swimming) will form in the place of feet. Duboule speculates that if these genes could be tricked into staying on just a bit longer, the fins of the zebrafish might sprout appendages suggestive of primitive feet.

What would a fish with feet look like? It could easily resemble the Acanthostega. Mineralized bones of this strange creature, unearthed in Greenland in 1987, tend to confirm the notion that fish did not crawl onto shores on their fins, says paleontologist Michael Coates of University College, London. Instead they probably developed limbs and feet that they used in the water for millions of years before they were capable of colonizing the land.

The transition to land was likely a gradual affair involving multiple stages of evolutionary change. The skeletons of fish, with their slender bones arrayed all in a row, are clearly ill suited for walking and running. Moreover, the muscles of fish are designed to deliver power in all the wrong places. "Think about tucking into a tetrapod [a cow, for instance] for Sunday lunch," says Coates. "The best cuts are the thighs and shoulders, the muscle motors that drive these animals along. In a fish these motors are pathetic, tiny things. It's the back and tail muscles that propel it through the water."

Duboule believes that over the eons of prehistory, Hox genes played a key role in the origin of species, facilitating the process of evolutionary change. Scientists now know, for example, that the genes that trigger the formation of hands and feet also control many other developmental processes in the posterior part of an animal -- among them, the addition of an anal opening to the digestive tract and, in four-legged creatures, the fusion of the lower vertebrae to make a pelvis. Isn't it curious, says Duboule, that fish lack a true pelvis as well as hands and feet? This suggests to him that both structures -- the appendages for walking and the bony apparatus that anchors them to the spine -- are linked at some deep genetic level that is yet to be plumbed.

Duboule concedes that "this is not even a real hypothesis," just a hunch, and that testing it will not be easy. One problem, contends Harvard's Tabin, is that Duboule and his colleagues studied "the wrong fish." Zebrafish are prolific and easy to raise under laboratory conditions, but they are advanced in evolutionary terms. A study of more primitive sea life, such as sharks or sturgeon, might yield greater amounts of evolutionary information; even better subjects would be lungfish and coelacanths, mysterious, nearly extinct creatures that lurk in the ocean depths and are the living fish closest to the fishlike ancestors of four-legged animals.

Further studies are needed to convince scientists that Duboule and his colleagues have correctly solved the fins-to-feet riddle. Other factors could be involved as well, including homeobox genes that are not Hox genes (that is, they do not affect the overall structure of an animal). Last year Sean Carroll, a developmental biologist at the Howard Hughes Medical Institute in Madison, Wisconsin, showed that a homeobox gene involved in insect-limb formation also controls the genetic signals that paint spots on butterfly wings. In essence, says Carroll, butterflies use an old gene to perform a new trick. "Evolution did not have to invent new genes," he observes. "One basic toolbox gives nature enormous potential for diversity."

The drawback for scientists is that nature's shrewd economy conceals enormous complexity. Researchers are finding evidence that the Hox genes and the non-Hox homeobox genes are not independent agents but members of vast genetic networks that connect hundreds, perhaps thousands, of other genes. Change one component, and myriad others will change as well-and not necessarily for the better. Thus dreams of tinkering with nature's toolbox to bring to life what scientists call a "hopeful monster" -- such as a fish with feet -- are likely to remain elusive. Scientists, as Duboule observes, are still far from reproducing in a laboratory the biochemical artistry that nature has taken millions of years to accomplish.