The Boston Globe
Health/Science
May 3, 1999

You take the thigh bone and connect it to the wing bone...

By Karin Jegalian, Globe Correspondent


Like they do in supermarkets, chicken eggs come to Cliff Tabin's lab at Harvard Medical School in plastic foam containers. But these eggs are fertilized, and with a little genetic manipulation, biologist Malcolm Logan, Tabin's colleague, can make the embryonic chicks inside grow drumsticks where their wings should be.

The deformed chicks - whose forelimbs bend like legs, have claws like toes, and lack the feathers of wings - are never permitted to hatch. But their existence reveals some long-sought clues about the mystery that drives all developmental biologists: how does a single cell, the fertilized egg, generate the intricate form of an animal?

Manipulating embryos so that they grow limbs in the wrong places may seem like work fit for a Frankenstein, but Tabin points out that these experiments are the only way to learn what purposes particular genes serve.

Ultimately, some biologists say, a thorough understanding of how arms and legs grow could have an eminently practical application - the regeneration of human limbs.

The recent results out of Tabin's lab and those of researchers in California, Montreal, and Japan also illuminate what goes wrong in certain birth defects - malformations of the arms are among the most common.

Limbs begin to ''bud'' out of a vertebrate embryo when the embryo still resembles a tube, with few obvious features. The limb bud is a small swelling on the side of the embryo and grows outward from its tip, just as trees grow fresh leaves from the tips of their branches. To form a human arm, for example, the bone of the upper arm and the tissues that surround it are laid down first. As the tip of the limb grows outward, the forearm is formed next, and finally the hand.

Biologists have long known that the fin of a fish, the wing of a bird, the leg of a dinosaur, and the arm of a human being all are braced by similar bones. In all of these creatures, a single bone extends from the torso, joins a segment with more than one bone - the forearm in a human - which in turn connects to a segment with many bones - in humans, the hand. Nature repeats this ancient theme over and over, like a Baroque composer recapitulating a melody with endless, subtle variations.

But the similarities in the limbs among species go deeper than the bone. Over the last two decades, biologists have learned a great deal about the molecules that direct limb growth. These molecules have tended to be versatile characters, playing the same basic role in a chick's wing, for example, as they do in the chick's leg or, for that matter, in a human arm or human leg.

In general, developmental biologists have so far struggled to answer broad questions - what distinguishes an animal's head from its tail, its back from its belly, what controls where a limb sprouts from the body, and what makes the limb asymmetric. In comparison, what makes forelimbs - the arms, in humans - and hindlimbs - the legs - distinct is a relatively subtle question.

Reports published in the journals Genes and Development in February, Science in March, Development in April, and two papers in the current Nature have begun to unveil the cast of molecules that distinguish forelimbs and hindlimbs. By tinkering with individual genes, scientists have been able to turn chicken legs into wings as well as wings into drumsticks. In two of the recent papers, similar experiments on mouse embryos have made their hindlimbs more like forelimbs.

Hans-George Simon studies limb regeneration in newts, animals whose legs and tails can regrow if lopped off, at Northwestern University Medical School in Chicago. He can imagine a future where human limbs are regenerated as well, but his immediate goals, as for most developmental biologists, are not exactly pragmatic.

''In developmental biology there are areas of pure research,'' he says. ''There are so many interesting miracles that help us understand ourselves.''

In the specific miracle of making distinct forelimbs and hindlimbs - say, our arms with their dexterous, sensitive hands, our strong legs - three genes appear to be key.

''These genes must be at the very top of the cascade,'' says Juan Carlos Izpisua Belmonte of The Salk Institute in La Jolla, California.

The idea is that these genes are part of a hierarchy of genetic control; they presumably turn on another set of genes, which in their turn activate still others. Work out of the Izpisua Belmonte lab, analyzing two of the key genes, appeared in the April 29, 1999 issue of the journal Nature.

The three newly discovered genes, with the awkwardly scientific names Pitx1, Tbx4, and Tbx5, are turned on in the flank of the embryo before the limb buds even form. Two of the genes, Pitx1 and Tbx4, are turned on where hindlimbs develop but not in forelimbs, while one, Tbx5, is turned on in developing forelimbs but not hindlimbs.

Discovered in mice, these genes are shared by all tested vertebrates, ranging from humans to fish, and may well define limb identity in all these species, whether the ''limb'' involved is an arm, leg, wing, or fin. The studies show that the genes found exclusively in hindlimbs do promote hindlimb formation, and the gene found exclusively in forelimbs does promote forelimb formation.

''Together, the full set of papers represents a very significant advance,'' says Tabin whose lab research was published in the March issue of Science. The question of what makes forelimbs and hindlimbs different was among the fundamental mysteries in developmental biology, and until recently, no one had a clue.

In fact, defects in the forelimb-specific gene, Tbx5, cause Holt-Oram syndrome, a condition that can leave people with ''flipper-shaped arms,'' similar to the limbs in people whose mothers took the drug thalidomide during pregnancy. The syndrome also has relatively mild forms - for example, giving people just an extra joint in the thumb.

Curiously, the same gene is also turned on in an embryo's heart, and people with Holt-Oram tend to have structural defects in their hearts - typically, a hole in the wall that divides the two halves of the heart - as well as in their arms or hands.

Heart and limb problems of various kinds are the most common forms of congenital malformations, and the two problems are often linked, notes Izpisua Belmonte.

''It is a mystery why,'' he says.

Still, the roles the Tbx5 gene plays in these different parts of the body can probably be teased apart. In March, geneticists led by Harvard Medical School's Christine Seidman reported that they had sorted out 10 different forms of the human Tbx5 gene in people with Holt-Oram syndrome. They showed that some forms of the gene can lead just to heart problems, while others lead just to limb malformations.

Izpisua Belmonte thinks the research on limbs may hold clues about the development of other parts of the body. After all, an embryo tackles the problem of growing similar but subtly different structures in more than one place - for example, to produce the more than two dozen vertebrae that make up the backbone and the five fingers of each hand.

Now, Izpisua Belmonte would like to see whether the same genes that specify limb identity are important elsewhere. And he'd like to know what activates the genes that are found only in budding hindlimbs or specifically in budding forelimbs.

Despite the recent progress, biologists are still far from being able to say what leads to the enormous variety of limb structures among different species. Researchers know many similar forces that make the limbs of turtles and kangaroos, for example, but not what sets these species apart.

Simon concludes, ''There's more to the whole story, but we have the first inroads now. I think we're getting into a very exciting time.'' Now, if they can only figure out what makes a wing different from a hand.

This story ran on page C01 of the Boston Globe on 05/03/99.