Most of us are at least superficially symmetrical--two arms, two hands, two eyes, all evenly arrayed to the left and right of a line running down the center of our body. But we are in fact essentially asymmetrical: the heart and spleen are on the left, the liver and gallbladder on the right. Some of the most essential organs are themselves asymmetrical: the right lung has three lobes, the left only two; the stomach is shaped like a bagpipe.
In part, that internal asymmetry is necessary for our survival. The heart, for example, is two pumps in one. The right half receives incoming purplish blood from the body and delivers it to the lungs, whereas the left half receives the red, oxygen-rich blood from the lungs and powers it through the aorta so that, with a little help from the blood vessel walls, it can reach the body’s farthest extremities. In their structure, the two halves are suited to their different tasks--the left half is far more muscular than the right.
Other asymmetries exist for more mundane reasons, such as to use the space inside us efficiently. The left lung has fewer lobes so that it can accommodate the heart, which nestles inside it. The liver is a complementary apostrophe to the stomach’s comma.
All in all, our asymmetry helps us survive. But there doesn’t seem to be any adaptive advantage to the particular left-right orientation of our body parts: 1 in 10,000 people, in fact, is born mirror-reversed, and the health consequences are decidedly minor. Left-heartedness appears to be an arbitrary convention, like driving on the right side of the road. Our hearts are on the left because our ancestors’ hearts were on the left, says molecular biologist Cliff Tabin of Harvard Medical School.
Thus the why of our particular asymmetry may ultimately be unanswerable. But the how of it is a hot question. As it happens, we humans are not the only animal stuck in a left-hearted mode. Among vertebrates, nearly every species, from chicken to cow, has its heart on the left. What is the mechanism that pushes the heart to the left in all these species? That’s a question for embryologists, who study the embryo as it develops from a single fertilized egg, dividing again and again into the millions of cells that make up a mature organism. It’s a question for geneticists, who study the dna blueprints that instruct those cells in how to build a body. In particular, it’s a question for Cliff Tabin: working with chicken embryos, he and his colleagues have begun to find what moves the chicken heart. And their discovery has implications for many creatures besides that one bird. The grand similarity in heart position among wildly different organisms leads Tabin and his colleagues to expect that there is a common underlying mechanism for organizing the body asymmetrically. Common dna blueprints, they believe, are giving common instructions in different organisms, even if some details differ.
To understand what makes the heart migrate left and the liver migrate right, it helps to know a bit about how embryos develop. It’s not as if a developing animal is a skyscraper under construction. In a skyscraper, a foreman surveys the blueprints and instructs workers where to go and which materials to use. Bit by bit, from the bottom up, the building takes shape. In a living body, the workers are the construction materials. Both are living cells. Each cell has a copy of the master plan tucked into its nucleus in the form of dna, like a blueprint in its back pocket.
Just as the construction foreman can’t send in the roofers before the foundation is poured, cells have to appear at the right time during the development of an embryo, laying the groundwork for their successors to build on. Though they all have a copy of the same master plan, the cells don’t all form or behave identically. They follow separate, though sometimes overlapping, sections of that plan, differentiating into the various tissues of the body. Some tissues even contain cells that die off when their work is done, like scaffolding dismantled when the building is complete.
Depending on its function, each cell reads a different part of the genetic code from the dna in the nucleus and translates it into a conglomeration of proteins, fats, and other chemicals. Some of these components build the cell or perform useful functions in the body, such as transporting oxygen. Others act like signals, carrying messages to other cells. Such signals play a big role in establishing the organism’s structure.
These chemical signals are also playing a big role in helping researchers like Tabin learn how the heart knows to move to the left. The reason is simple: chemicals can be synthesized in the laboratory and applied to embryos experimentally. This technique is especially powerful with chicken embryos, since to see the effects, all you have to do is poke a hole in the shell.
Through that hole you can see the chick’s primordial heart, which looks like a tiny length of pipe, beginning to form 30 hours after the egg is fertilized. It’s the first organ to become left-right asymmetrical: 5 hours later it begins to bend right, to form the heart’s characteristic shape--as long as nothing interferes with it. Until recently no one knew what chemicals might be used to interfere, since there were precious few clues to what chemical signals the embryo’s cells were using to talk to each other. Biologists mostly just observed developing organisms. Until about ten years ago, says Tabin, all we did was get incredibly sophisticated in our descriptions. Then Tabin and his colleagues found the first hint of the mechanism that arranges a chick’s organs asymmetrically. In the last couple of years, they have begun to crack the hitherto secret code of the body’s master plan.
As with any code, knowing just a few words doesn’t tell you much. At first you don’t even know what they mean. Sometimes the same code words occur in wildly different contexts. This is the case with the signal for left-heartedness: its discovery emerged from the apparently unrelated area of limb development.
Chicken wings share a basic structure with human arms: they start with a humerus, branch into an ulna and a radius, then diverge into metacarpals and digits, albeit only three, corresponding to our index, middle, and ring fingers. In 1968 embryologist John Saunders, then at Marquette University in Milwaukee, set out to investigate the way digits formed from a bulge of embryonic tissue called the limb bud. Digits-- fingers, toes--do not simply spring, fully formed, out of the bud. Instead the chick limb resembles that of its amphibian ancestors, webbed with excess tissue, most of which then dies off. Saunders wondered how early the between-finger tissue’s fate was sealed. He decided to extract a bit of webbing tissue from the rear of the limb bud where it was growing and move it to the front, toward the chick’s head, to see whether it would take along its doomed to die fate or discover new life there. He found neither.
Instead Saunders found that the plug of tissue he transplanted caused the wing to grow twice as many digits. What’s more, the new digits appeared as mirror images of the original ones. That is, the new digits appeared as ring, middle, and index digits next to the regular index, middle, and ring digits of the original embryonic wing. Somehow the transplanted tissue carried with it not only the signal to make digits but also an inherent polarity. The powerful signal, whatever it was, came not from the pre-digit tissue but from the transplanted region, which the researchers called the zone of polarizing activity, or zpa for short.
So many of the things in the limb that we’re finding molecules for now, Saunders was involved in discovering in the first place, says Tabin. Still, Saunders’s contribution was necessarily limited. There’s only so much cutting and pasting of embryos you can do, says Tabin. Inevitably, someone would have to look for the chemical identity of the signal that was establishing the polarity of the limb bud.
As soon as Saunders discovered the ZPA, British embryologist Lewis Wolpert proposed that ZPA cells secreted the polarizing signal into their immediate surroundings. The signal then went to work organizing local pre-digit tissue into digits. In Wolpert’s model, each digit’s destiny depended on how much of the chemical reached it. Nearby digits received the signal loud and clear, becoming the chick equivalent of ring fingers; digits farther away picked up just enough of the signal to become index fingers. The concentration of the signal, whatever it was, determined what the fingers would become.
But what was the signal? Scientists tried random chemicals off the shelf to see if they mimicked the zpa activity. No good. Researchers in one lab ground up 2,000 chick zpas trying to isolate it. That led nowhere. Candidates were proposed, but every candidate had arguments against it. The path leading to the molecular signal grew cold. The discovery, when it finally arrived, came from a surprising direction, from a different phylum, in fact--the insects.
A prolific breeder with a short generation time, the fruit fly Drosophila melanogaster has been the organism of choice for studying genetics for 100 years, and Drosophila studies have shed light on many of the characteristics we share with flies. Unfortunately, the body plan of a fly and that of a human do not have much in common. Sure, we both have heads and appendages, but the similarity would seem to end there. Fly hearts are even symmetrical. One might not be surprised to find four-legged vertebrates (such as frogs and dogs) using identical chemical signals, since such similarities could conceivably be traced to a common ancestor. But fruit flies? To find the last common ancestor between vertebrates and arthropods (such as flies), you have to go back about 550 million years, to a time before animals had even crawled onto the land. The fossil record shows that the last common ancestor, whether it was a worm, an eel, or a jellyfish, did not even have limbs.
So in 1980, when Christiane Nüsslein-Volhard and Eric Wieschaus at the European Molecular Biology Laboratory in Heidelberg, Germany, identified key genes involved in establishing the body plan of fruit flies, vertebrate developmental geneticists didn’t pay much attention. Even five years ago, few biologists realized that the same signals that lead to wing development in flies might have something to do with chicken wings or human arms. Presumably a link to the developing heart would be even less likely. Yet one of the genes that Nüsslein-Volhard and Wieschaus found helped Tabin uncover the signals that put chickens’ hearts on the left.
At the time Nüsslein-Volhard and Wieschaus discovered the fly genes, the functions of the developmentally important genes were not obvious. Most of the genes turned up because flies that were missing them died before hatching or were born misshapen, often exhibiting bizarre deformities. The researchers called one such gene hedgehog, after the bristly, porcupine-like appearance of mutant fly embryos that lack it. While normal embryos are divided into orderly segments, these mutants look like a single chaotic pincushion. When cells translated its code, the hedgehog gene produced a signal molecule that maintained polarity in the fly’s body segments. In other words, hedgehog was a signal that, like the zpa factor, told the front of the segments that they were the front.
In 1984, Swiss researchers showed that widely divergent species such as the fruit fly, the frog, and the mouse shared small segments of developmentally important genes. This was utterly surprising to biologists, who scrambled to apply techniques then available to identify more and more counterparts--called homologues--of these genes in other organisms. With the advent of the quick-copying method called polymerase chain reaction (pcr) in 1985, the process of finding these genes and cloning them--making multiple copies of them--got much faster, and by the early 1990s a gold rush was on, as scientists prospected for key homologues in higher animals.
I attended a meeting at Ringberg Castle in Bavaria in early 1993, recalls Tabin, and I mentioned to Phil Ingham that I was interested in identifying and cloning vertebrate hedgehog genes using the Drosophila gene as a starting point. Ingham, a developmental biologist at the Imperial Cancer Research Fund, working in Oxford at the time, was one of the first people to understand aspects of hedgehog signaling in Drosophila, recalls Tabin. The subject made him uncomfortable. Well, you know, Cliff, we’re trying to clone vertebrate hedgehogs ourselves, said Ingham. Just then Andy McMahon, a British scientist then at the Roche Institute of Molecular Biology in New Jersey, walked up and said, Actually, Cliff, we’re trying to clone vertebrate hedgehogs, too.
Suddenly, recalls Tabin, it seemed like everyone at Ringberg was trying to clone vertebrate hedgehogs. But the trio quickly realized that they were the only ones, at least at that meeting. Later, in Ingham’s palatial room at the castle, they agreed to join forces in tracking down the vertebrate counterpart to the fruit fly gene. Each was working with a different organism--Tabin with the chick, McMahon with the mouse, and Ingham with the zebra fish. By collaborating, they increased their chances of finding a hedgehog homologue without encroaching on one another’s scientific turf.
Tabin and his partners pulled off a double coup: not only did they find vertebrate hedgehogs, but they showed that the genes they’d discovered were identical to the long-sought zpa factor. First, McMahon found a mouse gene that was clearly a homologue to the Drosophila gene--the first vertebrate hedgehog. By searching for something that looked similar in large libraries of chick dna, Tabin and his co-workers found two related chick genes and then a third.
As these new hedgehogs began to pop their bristly heads up all over, in mice, chickens, and zebra fish, they needed names. The scientists in Tabin’s lab first called the initial two genes they had identified A and B; then they began calling them by the names of actual hedgehogs found in nature. But developmental biologist Bob Riddle, a postdoctoral fellow in Tabin’s laboratory at the time, had a different idea for one of them. It would have been boring to name it ‘common European.’ I wanted to find something that fit our lab. And in Cliff’s lab, we like music. Loud music. We’re a loud lab. When I saw an ad in a magazine for a Sonic the Hedgehog video game, I thought, ‘Sonic--this is us.’
Sonic, it turned out, was the gene they had all been waiting for. First of all, the protein it produced was the zpa factor. Sonic hedgehog protein turned up near the chick’s version of the ring finger, but not near its version of the index finger. The protein’s distribution over time matched the known distribution of the zpa activity. For the coup de grâce, Tabin’s team showed that Sonic protein could mimic what the zpa does in a developing limb, doubling the digits with the characteristic reverse polarity.
That discovery alone made Sonic hedgehog one of the most important signals ever found in vertebrate development. Secreted by the zpa and a few cells elsewhere in the developing embryo, it affects the fate of cells near and far. Sonic hedgehog is a special kind of signal, one that is involved in shaping all manner of structures--the muscles, the spine, and the brain, to name just a few. Simply discovering the zpa factor would have made Tabin a name in developmental biology. But there was still the niggling problem of asymmetry--a problem about which he was soon to become an expert.
When Tabin’s team examined which areas of developing chick embryos were producing Sonic hedgehog protein, they noticed something unusual: there was a lot more Sonic on the left side of the embryo than on the right. A year later, Mike Levin, a graduate student in Tabin’s lab, found that Sonic protein appears on the left side early on, about 18 hours into chick development. A few hours before that, another gene begins producing protein, but only on the right, and a few hours after, both Sonic and the second gene switch off on their respective sides and a third, left- sided gene begins to produce its protein. The activity of the third gene, Levin and Tabin showed, depends on the second, and the second on the first. The three are part of a pathway of genes, each triggering the next in line.
The researchers suspected that this pathway might be responsible for asymmetrical development, at least in the chick. The best way to prove this was to alter it. In the summer of 1995, Levin and Tabin implanted a plug of cells engineered to churn out the Sonic hedgehog protein--one of the two I’m left signals--and placed them on the right-hand side of developing chick embryos. Suddenly, heart location became random in these embryos. The primordial heart tissue could no longer tell left from right-- and the heart moved left or right with equal probability.
Since it has such a compelling effect in chicks, and since it was initially identified as a homologue of a mouse gene, one might expect Sonic hedgehog to play an equally key role in mouse development. And because mice are physiologically more similar to humans than chickens are, such a finding would suggest that Sonic hedgehog has something to do with the human heart moving to the left.
However, things are not that simple. For starters, no one has found that Sonic hedgehog genes are producing their protein asymmetrically in mice (or zebra fish, for that matter), though plenty of people have looked. Tabin points out that perhaps they have not looked hard enough--the gene may be switched on only fleetingly, and the subject is harder to study in mice because developing mouse embryos are not as accessible as chick embryos. Human embryos are less accessible still, for obvious reasons.
But in a larger sense, it does not matter. Scientists were quick to recognize that the asymmetry pathway Tabin and his colleagues identified is an important one, perhaps even the only one, at least in birds and mammals. The parallels between embryos of different species are just too striking for it to be otherwise. Evolution is lazy, says Lewis Wolpert. Once it’s got a good technique, it sticks with it. And early evidence seems to indicate that at least one of the three genes in the pathway--not Sonic hedgehog--has a homologue that produces its protein asymmetrically in the mouse.
Now that the pathway has been identified, biologists all over the world are scrambling to find the so-called upstream genes that trigger the asymmetrical production of Sonic hedgehog protein and the other two gene products in the unfolding chick embryo.
There could be a hundred upstream genes, says Tabin. Or there could be one. But even if you could trace the source of left-right asymmetry back to the very first gene that produces its protein asymmetrically and to the first cell in which that gene goes to work, you would still be left with a question: How did that cell know it was supposed to be on the left?
One possible explanation involves the way some molecules naturally bend or spiral. There are a lot of oriented molecules in the cell, says Tabin. dna, for example, forms a right-handed helix. Many proteins twist to the left, while others twist to the right. Some of these molecules may be inherited differently because they are differentially distributed from a mother cell to two daughter cells, though it’s random speculation to try to say which molecule.
But the lack of answers about the ultimate source of asymmetry-- or even Sonic’s apparent lack of asymmetrical expression in the mouse--does not bother Tabin. Five years ago, there wasn’t a prayer of asking these questions, he says. Sonic has allowed us to ask them.
And asymmetry is just the beginning. By the end of my career, declares the 42-year-old Tabin, we will understand development. Not just describe it. Not just describe it on a molecular level. But we’ll understand the logic behind it. We’ll understand in a very real sense how you go about making an organism.