From Saturday's Globe and Mail

November 2, 2007

The chicken egg has been prepped for surgery - a pea-size hole cut in the shell and covered with sticky tape. And now Hans Larsson, a McGill University researcher, removes it from the incubator, places it under a microscope and prepares to operate.

He gently peels off the tape and teases back the membranes that line the shell with tweezers. Through the eyepiece, he can see the tiny dot of a heart, steadily beating. He can also see the bud where he implants a milky bead doused in a protein. He hopes it will coax the embryo to grow a big tail. A dinosaur-like tail.

A paleontologist, Prof. Larsson spends a significant portion of his time doing traditional dinosaur hunting, digging fossils as far afield as the Arctic and Africa with jackhammers and pickaxes. But he has long been frustrated with the limitations of studying old bones and what they reveal about the mysteries of evolution.

It was by examining ancient skeletons that paleontologists learned that modern birds, including chickens, descended from dinosaurs and that their relatives include such fierce predators as Tyrannosaurus rex. What fossils don't reveal, though, is how exactly such dramatic anatomical changes first arose. How did teeth the size of bananas turn into beaks? Or mighty tails become wimpy, feathered stumps?

click thumbnail to enlarge

By manipulating genes and proteins in his lab, Hans Larsson
is trying to grow T. rex-like tails in chicken embryos.

For answers, Prof. Larsson has turned to the burgeoning field of evo-devo - or evolutionary developmental biology - a radical new approach to understanding the past.

It is based on the astonishing discovery that modern animals, including humans, share many of the same body-building genes and that some of these genes have been around for millions of years.

How these "master genes" are used during the development of an embryo, when they are switched on and in what combinations is what makes chickens look so different from their ancient relatives.

And if you play with this complex genetic choreography in the embryos of modern chickens, you should, theoretically, be able to resurrect dinosaur-like features.

All of which means that scientists such as Prof. Larson may actually be able to mimic what occurred millions of years ago inside dinosaur eggs.

This could offer answers to some of the big questions in evolution. For example, how did the first wing get built? How did dinosaurs turn into chickens? In the future, evo-devo may also reveal how the human body is formed, and what can go wrong.

But if scientists can make chickens look like dinosaurs, what other creatures might they build? What variations on the human form are possible? Along with ethical issues, evo-devo also raises humbling questions about what it means to be human and what we might look like in the future.

If we are all repositories for ancient genes, could we devolve in the same way we've evolved? Pressed by radical environmental change, say, could humans grow fur like the apes or sprout tails like the common primate ancestor we share with baboons? Could we go back to walking on all fours, or breathing underwater?

Evo-devo is often characterized as a revolution in evolution, or as a new science.

In fact, Aristotle noted that chicken embryos went through a stage in which they looked like worms. And by the 1800s, comparing the physiology of embryos was a well-established field.

Charles Darwin drew heavily on this work for evidence that modern animals, including humans, descended from a common ancestor. He noted that the embryos of many species are far more similar than the adults. He also argued that many animals show traces of their evolutionary ancestry in their early stages of development. Some kinds of snakes, for instance, grow tiny legs as embryos, hinting at forbears that walked instead of slithered.

After the publication of The Origin of the Species in 1859, embryos were all the rage among anatomists. Of special interest were creatures such as the lung fish, which can breathe air and was thought to be a link between animals that lived in water and those that walked on land.

But interest dwindled in the next century, as modern genetics came to the fore. Biologists shifted their focus to basic units of heredity - which turned out to be genes made up of DNA. What makes lobsters look like lobsters, their argument went, are lobster genes. What makes leopards look like leopards are leopard genes. Embryos are merely delivery vehicles for the genes that make each species, including humans, what it is.

"Almost everyone switched to genetics, and embryos got left behind," says Brian Hall, a researcher at Dalhousie University in Halifax who is a pioneer in evo-devo.

Then came shocking evidence of just how much genetic material species have in common. Humans share almost 99 per cent of their genes with chimpanzees and 85 per cent with mice. Even the barrel-shaped sea squirt shares 80 per cent of its genes with us.

Since the 1980s, scientists have also found that the same genes (or very similar ones) govern embryonic development in animals that are nothing alike. The construction of the human heart, it turns out, is directed by a gene that gives insects their primitive hearts. Our eyes are built with the help of the same gene that directs the formation of eyes in flies and frogs.

In many animals, the same family of genes - called the Hox family - determines which end is up, or which part of the embryo will be the head. Hox genes also tell cells in the embryo which kind of appendage they should grow - arms, legs, antennae or wings.

And the Hox genes date back half a billion years, evolutionary biologists say, to the ancestors of all the major animal groups now found on Earth.

Nature, it seems, doesn't like to upgrade construction equipment that is working well. Why invent a new way to build a leg or an eye when the old way has been working just fine for hundreds of millions of years?

In other words, while no one has sequenced dinosaur DNA, it is highly likely that the same genes that directed the embryonic development of T. rex do the job in the modern chicken. And if all animals - ancient and modern - share the same body-building genes, it should be possible to get a chicken embryo to grow a monster tail.


This may also explain the occasional blast from the past we see in modern humans: the rare cases of babies born with hairy faces, extra nipples, short tails or webbed fingers and toes - ancestral traits called atavisms.

According to reports in the medical literature, human tails are sometimes made up of fatty tissue. But in other cases they have extra vertebrae and muscle and can move. Most are removed after birth.

Geneticists who have studied this kind of abnormal development say the genes involved are the same ones that direct the process of laying down the vertebral column in all animals with spines. But this process stops earlier in us than in vertebrates with tails - or than they once did in our ancestors - except in rare cases where the timing is somehow thrown off.

A Mexican family offers another example of how ancient traits can resurface, Dalhousie's Prof. Hall says. He wrote a paper about them for the British journal Nature in 1995 that was accompanied by a picture showing a six-year-old boy whose entire face, other than his eyes and lips, is covered with black hair so dense it looks like fur.

The boy is part of a large family and over five generations 19 of his relatives were born with excessive body hair, especially on their upper torsos and faces.

Prof. Hall says it is likely that the mutation geneticists found on one of their chromosomes triggers an ancestral distribution and density of body hair.

Atavisms have been well documented in other species too. Whales, which descended from terrestrial mammals, are sometimes found with tiny bones that look like hind limbs. Horses are sometimes born with three toes, like their ancestors.

These throwbacks used to be an embarrassment to evolutionary biologists, Prof. Hall says. How could evolution move backward?

Now, however, they are seen as potent evidence of how much genetic potential we retain. Snakes may have lost their legs and humans may have lost their tails, but that doesn't mean the ability to make these structures has disappeared.

And in the distant future, says Prof. Hall, who has published dozens of papers about evo-devo and what is considered the seminal text in the field, humans may once again be furry or grow tails if environmental conditions favour those developments and the traits get passed down for many generations.

This would not happen at the pace of Hollywood movies though. In the futuristic Kevin Costner thriller Waterworld, mutant humans develop fins and gills in time to cope with the melting ice caps.

In the real world, Prof. Hall says, "it would take a long time."


In Prof. Larsson's laboratory, meanwhile, experiments with atavisms take about five days of incubation.

He soaks small beads in a yellowish liquid containing a protein produced by one of the genes involved in building animal embryos. Then he implants them in about two dozen eggs.

He suspects that dinosaurs - with their huge tails - used far more of this protein than modern chickens and that this affected other genes. He hopes that by tinkering with chicken embryos he may mimic, however briefly, the process that gave T. rex its mighty appendage.

Like a small but growing number of paleontologists, Prof. Larsson was attracted to evo-devo because it opens up just this door to recreating the past. So although the 36-year-old - who studied at McGill and the University of Chicago - learned the techniques of traditional paleontology and how to describe ancient life, he has also mastered the tools of molecular biology and embryology.

It was a coup for McGill to lure him back from the United States, where top paleontologists are far better funded than in Canada. And Prof. Larsson, born in Alberta and raised in Ontario, is happy to be back. He loves that his job combines the thrill of the hunt with the challenges of experimental science.

"A few of us are realizing that paleontology can be so much more," he says.

But figuring out how to get a chicken to grow a dinosaur-like tail isn't easy. It is a bit like a novice baker trying to make bread with no instructions and knowing only a fraction of the basic ingredients, say water and salt. What would you do first?

If you had a recipe, it would tell you to mix the yeast with warm water and a bit of sugar, then add it to the flour and salt. Next, knead the dough and let it rise in a warm place.

Without instructions, you might guess that you need yeast, but throw it in at the last minute. Or use a cup of sugar instead of a few teaspoons. What if you used a bucket of flour? Or something that looks like flour but isn't edible, such as plaster of Paris? Or several cups of salt? You could end up with anything from biscotti to bricks.

The basic ingredients for assembling a dinosaur tail are the proteins produced by the common body-building genes, but like the novice baker Prof. Larsson has only a few of them on his grocery list. He is trying to figure out the rest and get the timing and combinations for the recipe right.

Luckily, it is relatively easy to play with proteins. "It is amazing, in a crude scale, the kinds of things we are doing, manipulating timing of expression of proteins, turning them up or turning them down," he says. "You turn this dial to get these ancestral patterns."


At this stage, Prof. Larsson is waiting for clear signs of what exactly his chicks will become.

The tray of embryos in his laboratory fridge - operated on earlier - look pink and otherworldly. And their tails appear to be significantly longer than in normal embryos. But they are clearly still too stumpy to qualify as dinosaurian. Chicken tails typically have four to five free vertebrae, while dinosaur tails involve 40 to 60.

Still, if Prof. Larsson succeeds with the tails, he will work on getting the embryos to grow dinosaur-like hands, feet and skulls. And while his work is confined to embryos right now - they are all killed before they hatch - if he can create chickens that look like dinosaurs, he says, he will want to let them grow unhindered, at least once.

He knows, however, that some tough ethical questions will have to be answered first. "Should this organism be allowed to live? Should we even be doing this to an embryo with the intent of letting it live? Those questions have not been brought in the public eye yet," he says.

To put his work in context, biologists have been creating grotesque and troubling mutants for decades as they attempted to understand how embryos develop. The researchers figuring out the details of the Hox genes, for example, ended up with fruit flies that had feet where their mouths should have been.

And while ethical questions hang in the air, Prof. Larsson's contemporaries continue to explore the frontier between the past and the present, trying to produce evolutionary throwbacks in their labs.

Last year, researchers at the University of Wisconsin and the University of Manchester coaxed chicken embryos to briefly develop alligator-like baby teeth. Eventually, the field may move into manipulating developmental genes in human embryos.

"It is probably not going to be done for many, many decades," Prof. Larsson says, "but we certainly have the potential to make a human with 60 back vertebrae [almost twice as many as we now have] or six limbs."

Other evo-devo scientists are probing how turtles first grew shells, or why storks have such big beaks and sparrows such little ones. Their work involves comparing the genes that are at work during various stages of embryonic development in different, but related species.

For example, Prof. Hall is trying to figure out how some kinds of fish traded a set of fins for powerful suckers that allow them to hang onto rocks in strong currents and even climb waterfalls.

So when should we start the conversation about this type of work and where it may lead? Some might say it is already too late, that the proverbial genie is out of the bottle. Prof. Larsson argues that it is better to wait until we know what is possible.

The goal of evo-devo, he says, is not to create bizarre hybrids of ancient and modern creatures. It is to produce the complete book of life, one that would explain where modern animals come from and how our different bodies took shape. It would chart, in exquisite detail, how a single-cell organism developed into what Darwin described in the final passage of his most famous book as "endless forms most beautiful and most wonderful."

"That's the holy grail," Prof. Larsson says.

Anne McIlroy is The Globe and Mail's science reporter.

Fair Use Notice and Disclaimer
Send questions or comments about this web site to Ann Berlin,