The Yale Alumni Magazine is owned and operated by Yale Alumni Publications, Inc., a nonprofit corporation independent of Yale University.

The content of the magazine and its website is the responsibility of the editors and does not necessarily reflect the views of Yale or its officers.

Comment on this article

The Birth of Birth
180 million years ago, our ancestors were still hatching from eggs. Then something radical happened.
March/April 2007
by Steve Olson '78

Steve Olson '78 was a National Book Award finalist in 2002. His most recent book is Count Down.

For months, here on the third floor of Osborn Laboratory, just at the foot of Science Hill, graduate student Vinny Lynch has been trying to get platefuls of reluctant cells to grow and divide. His dissertation and PhD could depend on this project. So could an unprecedented effort to uncover the origins of one of nature’s most remarkable creations: the womb.

Lynch, who has long curly hair and a colorful tattoo of a carp that wraps around his calf, takes a petri dish from the incubator and puts it under a microscope. The cells look like translucent leaves on a bank of snow—“just like any other cells,” Lynch says. But these cells aren’t like other cells. They’re missing a crucial gene known as HoxA-11. During the development of a human embryo, this gene helps direct the construction of arms, legs, vertebrae, and —if Lynch and his advisor, Gunter Wagner, are right—the mammalian female reproductive tract.

Without the missing gene, the cells aren’t doing well. They grow slowly. They react poorly to the hormones Lynch gives them. That’s an annoyance, given Lynch and Wagner’s ambitions for this experiment. They want to take on perhaps the most hotly debated topic in evolutionary biology today: how does evolution give rise to new body parts?

Even Charles Darwin once wrote that he “felt much difficulty in understanding the origin of simple parts.” He saw quite clearly how existing parts could evolve into more complex structures. Evolution is adept at working with what’s already there—accentuating the positive (nocturnal mammals have evolved larger eyes) and eliminating the negative (Arctic hares no longer have dark fur in the winter). But how do entirely new biological features originate—the eye, hair, the uterus?

The question still dominates many debates over evolution. Creationists claim that complex structures are evidence of God's handiwork. They say that the intricate biochemistry of life, the precisely fashioned parts of organisms, and especially the self-awareness and moral sense of humans could not have simply evolved. And even many of those who accept the reality of evolution are sometimes incredulous at its creativity. As the biologist Richard Dawkins has written, “People have a hard time believing that so simple a mechanism could deliver such powerful results.” Maybe that’s one reason why just 14 percent of Americans unequivocally accept the standard scientific account of how evolution works.

Yet nature produces new body parts all the time, if you think about the word “new” in a different context. A fertilized egg slowly drifting down a woman’s fallopian tube has pretty much the same internal components as any other human cell. But just nine months later, that single cell will have produced an organism with arms, ears, blood vessels, digestive organs, a brain, and, in females, a new collection of eggs waiting to begin the process anew.

The instructions for making all these parts reside in the DNA of the fertilized egg. As an embryo develops, each new cell inherits the same DNA as in the original fertilized egg. But different segments of DNA are active in different kinds of cells. “The differences between body parts in embryological development are caused by differential expression of genes,” says Wagner, 52, a sinewy, precise, soft-spoken man. “Certain genes are only expressed in the liver and others in the brain. The key to understanding what makes a tooth or a liver or a thumb has to be related to the ways genes are regulated.”

So if you want to make a different body part, you have to change the DNA in a fertilized cell. The links between DNA, development, and evolution have been obvious since DNA was identified as the carrier of genetic information more than 50 years ago. But only in the past couple of decades have biologists been able to investigate the connection. First, they had to develop the tools of genetic engineering. When biologists could change specific genes in mouse and fruit fly embryos, they could analyze the effects on the growing animal.

Even more important, biologists needed some way of determining how DNA has changed over evolutionary time. Except in rare cases (such as with bones buried in cold caves), extinct organisms don’t leave their DNA behind. But biologists have figured out a work-around.

As scientists began to sequence the genomes of humans and other organisms, they realized that they could reconstruct much of the genomes of extinct organisms by comparing the DNA of species descended from common ancestors. For example, we now know most of the genetic code of the common ancestor of humans and chimpanzees—an ape-like animal that lived about six million years ago—because much of the DNA common to humans and chimpanzees descended to us, unchanged, from that ancestor. Where the DNA is different, scientists can usually figure out the genetic sequence of the common ancestor by analyzing the DNA of other primates.

The combination of these new techniques has allowed two branches of biology to come together to form a new field: evolutionary-developmental biology, or “evo-devo.” “There has been an explosion of findings that have been enabled essentially by genetic technologies,” says Stephen Stearns, the Edward P. Bass Professor of Ecology and Evolutionary Biology at Yale. “Evo-devo is in the process of settling a lot of important and long-standing questions by giving us a very detailed picture of how genes result in organisms.”

One remarkable finding from evo-devo is how similar human beings are, on a genetic level, to other organisms. “Most animals share very similar sets of genes,” says Sean B. Carroll, an evo-devo pioneer at the University of Wisconsin-Madison. “It was thought maybe 35 years ago that novel structures had to involve new genes. It’s sort of intuitive that if you’re going to make new things, you need new genes. But that seems to be pretty much wrong. Novelty and diversity come from using the same things in new and different ways.”

Wagner’s work on the female reproductive tract began as a hunch. Several years ago, Wagner was looking at the role of HoxA-11 in the development of limbs when he came across an unusual finding. Researchers at the University of Cincinnati College of Medicine had found that female mice without a functioning HoxA-11 gene are infertile. They produce healthy eggs, which develop normally if they are transplanted into other mice. But the eggs can’t develop inside their own mothers because they can’t implant in the uterine wall.

Some scientists would ignore a tangential finding like this. Tracking down the cause of the effect was sure to be complicated; a graduate student could spend years on the project and have little to show for it. Wagner saw nothing but potential. “When something unexpected happens,” he says, “you have to pursue it.”

Pursuing the unexpected is a pretty good summary of Wagner’s scientific style. Born and raised in Vienna, he became fascinated with evolution as an undergraduate and worked as a graduate student on the mathematics of evolutionary theory. But after earning his PhD in 1979, he did two postdoctoral fellowships designed to develop his laboratory skills in neurobiology. After joining the University of Vienna faculty, he quickly became known for synthesizing mathematical analysis and hands-on lab work. Yale recruited him in 1991; the following year, he received a MacArthur “genius" grant.

Wagner, now the Alison Richard Professor of Ecology and Evolutionary Biology at Yale, has built an extraordinarily wide-ranging research group. Students and postdoctoral fellows in his lab have studied, among many other topics, the evolution of fins, the mathematics of gene interactions, limb regeneration in amphibians, and the origin of birds' wings.

Lynch was a brand-new graduate student whe n Wagner started studying the infertile mice, and he eagerly volunteered to work on the project. The evolution of the mammalian reproductive tract is an inherently interesting problem. Almost everyone has wondered at some point about the odd juxtaposition of the reproductive and excretory systems. (The classic formulation of the problem among biologists is: why is the sewer system routed through the entertainment center?) But human plumbing, at least in females, is a model of decorum compared with that of our evolutionary kin. In reptiles and birds, the female reproductive tract consists of a relatively simple tube. Eggs drop into the top of the tube from the ovary. As they make their way toward the outside world, the eggs are surrounded by the yolk, the white, and a hard shell. The tube then empties into an all-purpose excretory tract called the cloaca. In Latin, cloaca means sewer, which gives you a pretty good idea of what’s going on down there in our nonmammalian relatives.

(The penis has a different story. It has evolved a number of times in separate animal lineages. Perhaps more alarmingly, it also has disappeared a number of times among species that evolved other ways of delivering sperm to eggs.)

The earliest mammals laid eggs, like the reptiles from which they evolved. But about 180 million years ago, a mammalian species that probably looked something like a small mole or opossum began doing things differently. It started retaining its eggs in its body for part of their development. “It’s much more adaptive for the female to hold the eggs within her for longer and longer periods,” Lynch argues. “Maybe there were high tides, or bad storms in the spring, when you lay your eggs. That way, the female can control the environment.”

This was an interesting time in the evolution of mammals. Two kinds of egg-laying mammals, descendants of those that lived 180 million years ago, survive today: the platypus and the echidnas. The echidna, which looks like a hedgehog and lives in New Guinea and Australia, still has a reproductive system much like that of reptiles. But when baby echidnas hatch, their mothers provide them with milk rather than scraps of gathered food.

Another group of mammals, represented today by the marsupials, went partway toward internal development and then veered in another direction. An opossum embryo, for example, attaches to the reproductive tract of its mother. But it receives food and oxygen there for just a few days before it emerges from the birth canal, crawls into its mother’s pouch, and latches onto a teat to continue its development.

All other mammals living today are descended from ancestors that evolved ways to nurture embryos internally for weeks or months. But to carry an embryo to term, our mammalian ancestors needed a much more complicated reproductive system than in reptiles and birds. They needed the uterus, where the embryo develops. They needed the vagina, which is similar in structure to the cloaca but is apparently, Wagner says, “a complete developmental novelty.” And they needed the endometrium—specialized cells within the uterus where the embryo can attach.

The mice that lacked HoxA-11 also lacked a functioning endometrium. There could be many explanations for that. But it was possible, Wagner thought, that HoxA-11 and related genes were key in the evolution of this critical new part.

In 2001, Wagner and Lynch set out to determine whether changes in HoxA-11 helped bring about the evolution of the endometrium. For this experiment, they didn’t need petri dishes. They used computers.

HoxA-11 is one of a large group of genes that biologists are studying intensively. The Hox genes are the classic example in evo-devo of very old genes that have learned new tricks. Originally discovered in fruit flies, these genes were later shown to coordinate the development of body parts in all animals, from sea anemones to humans. Essentially, a Hox gene acts like the director of a movie. It tells hundreds of other genes to turn on or turn off. In turn, these active and inactive genes then determine where a cell should go during development and what that cell should be. Furthermore, each Hox gene is like a director who works on several movies at once. A given Hox gene might direct the construction of arms and legs, fingers and toes, and the digestive tract. And in true Hollywood style, a Hox gene can run amok: when cancer cells begin to proliferate, Hox genes are often at the center of the chaos.

For their research, Wagner and Lynch first obtained the DNA sequence of HoxA-11 from several organisms, including humans, mice, opossums, chickens, and fish. (The sequences for some organisms were already available in databases; for others, Wagner and Lynch sequenced the gene themselves.) They reasoned that organisms without a uterus, like chickens and fish, should have a HoxA-11 gene similar to the ancestral version. In contrast, mammals like humans and mice should have a version of the gene that includes the directions for making a uterus. Marsupials like the opossum or kangaroo should be intermediate cases, since they have some of the components needed for internal development but not all.

When a gene like HoxA-11 evolves, the process leaves a telltale trace. Changes in a DNA sequence—caused by chance copying errors when a cell divides—may have no effect on the protein encoded by a gene. These silent changes happen more or less at random and at a constant rate, which establishes a sort of ticking clock that can be used to time evolutionary processes. Other DNA changes do alter proteins. Many of these changes are harmful, and the organisms that have them die or fail to reproduce. But some protein-altering changes enable an organism to have more offspring. Over successive generations, these changes can become more common. By comparing the rate of beneficial changes with the background rate of neutral changes, geneticists can estimate when the changes to a gene proliferated because they gave the organism an advantage. In short, they can find out when an evolutionary change to a gene took place.

Wagner’s hunch paid off dramatically. When he and Lynch compared the DNA sequences of animals with and without a uterus, they found a burst of evolutionary change in HoxA-11 about 180 million years ago, just when the fossil record suggests that the uterus was beginning to take shape. The finding, says Jeffrey Innis, a Hox expert at the University of Michigan, “demonstrates the power of relating new genomic sequences to evolutionary and anatomical information.”

It was the first time that change in a single gene had been tied to the evolution of a novel body part—confirmation that a small change in a very old gene had been essential in creating something new.

Lynch’s current experiment is an attempt to build on this work and find out how HoxA-11 does what it does. The cells in his incubator are mouse uterine cells from which the HoxA-11 gene has been removed. As a result, their output of proteins has changed. They can’t make the crucial proteins of the endometrium that enable an embryo to attach.

Lynch has dozens of petri dishes in the incubator, and millions of cells in each dish. (A sign on the lab door reads, “Capacity: 10¹².”) If he can get these cells to grow well, he and Wagner are planning to reinsert the HoxA-11 gene from various organisms into the cells. Their hypothesis is that HoxA-11 from a mouse or a human should return the cells to their fecund state: the mix of proteins they produce will be the same as that of normal mouse uterine cells.

But HoxA-11 from a chicken should fail to change the proteins. And the genes from an opossum and a platypus should fall somewhere in between.

The problem is that removing the HoxA-11 gene appears to have altered the cells in some more fundamental way. They don’t seem to realize any longer that they are uterine cells; they don’t respond to the hormones that would trigger normal uterine cells to mature. To Lynch, it suggests that the missing HoxA-11 must have an unknown function earlier in the cell’s development. “They’re having an identity crisis,” he says.

It’s a setback that might take some time to solve, Lynch acknowledges. But he doesn’t seem to mind. For one thing, he has learned from Wagner that unexpected outcomes can provide rich scientific insights. Already, he points out, “we’ve learned something cool—that HoxA-11 is needed for cell identity.”

And Lynch knows that he and Wagner are working on one of the great mysteries of biology. They’re showing that biological creativity does not require a divine act of creation. It just requires time. Evolution is itself an experimentalist. It tries a new arrangement of molecules to see what will happen. If the experiment fails, it is discarded. But every once in a while an experiment succeeds, and on that successful result, new experiments can be conducted. The mechanism is simple, but it creates wonders.  

©1992–2012, Yale Alumni Publications, Inc. All rights reserved.

Yale Alumni Magazine, P.O. Box 1905, New Haven, CT 06509-1905, USA. yam@yale.edu