By the early 1970s, the mood of biology was expectant—even heady, some might have said. The first gene ever identified in an organism had been written up in 1969—it belonged to the colon-dwelling bacteria E. coli—and it seemed only a matter of time before the genes and proteins of significant numbers of species would be decoded and biochemical explanations might exist for an organism’s every twist and turn. Biologists appeared to be on the brink of understanding how Life worked.

But during these intoxicating times, which would extend well into the future, sometimes a more realistic note was sounded. As the French geneticist and Nobel laureate François Jacob cogently offered at a scientific meeting some years later, biology still operated at a relatively primitive, one-dimensional level. Although he and his colleagues were privy to the language of genes and proteins, there was so much more to learn about how these linear sequences “give rise to two-dimensional layers of cells, which fold in three dimensions according to a precise schedule, that is, according to a fourth dimension, time.” Stressed Jacob, “We are three-dimensional objects, but

we do not even have the concepts required to handle the behavior of such objects.”

Jacob, who was based at the Pasteur Institute in Paris, made these sage remarks at a conference on teratomas and their stem cells that took place in the fall of 1982 at Cold Spring Harbor Laboratory. Nearly thirty years earlier, in 1953, at that bastion of biology on the north shore of Long Island, New York, James Watson had publicly announced his and Francis Crick’s grand finding: the structure of DNA, the chemical that genes are made of. That was the year, as well, that Roy Stevens had encountered his first mouse tumor. In his talk, Jacob glanced back just twenty years, to when a meeting on teratomas “would have consisted of a dialogue between Roy Stevens and Barry Pierce.” Stevens “literally invented” the system of using mouse teratomas and their embryonal carcinoma (EC) stem cells for exploring the parallel universe of the early embryo and its stem cells, Jacob recounted, while Barry Pierce “picked up” on this system and became the first to apply it toward “certain problems in embryonic development and cancer.” Jacob saw these unsolved realms as the “two most important problems in biology,” with a teratoma and its stem cells lying right “at the crossroads.”

Jacob had led one of the first European groups that, in the late ’60s, had rushed to investigate teratomas obtained from Roy Stevens’s Mouse Strain 129. He and others had hoped that by studying an embryo and its changing cells through the substitute world of a teratoma and its stem cells, they might mount a grand inquisition into embryological growth, which had long posed such a mystery in mammals.

As odd a stand-in as testicular tumors from mice were, they beat collecting mammalian embryos. A mouse’s early embryo (days 1-3) could be flushed from the oviduct, but it was difficult to collect enough of these specks for biochemical analysis. Then, once an embryo implanted in the uterus (day 4.5), it was hard to pick out, not lending itself to examination until some days later. Yet so much of interest was happening during these earliest days, when swiftly mul-

tiplying cells would have a profound influence on the final package. A mouse embryo’s third day, in particular, was critical. Before the day was out, the blastocyst had formed, along with the cluster of stem cells inside it. The next day, as soon as the embryo implanted, many of these pluripotent cells started differentiating as punctually as Big Ben chimes, leaving their magic behind.

To think—these same master cells sat inside the bizarre little tumors of mice. What a heyday for developmental biologists!

In the late ’60s, a box containing several of Roy Stevens’s teratoma-bearing mice crossed the Atlantic, bound for a small laboratory connected to the venerable Department of Anatomy and Embryology at University College London. As soon as he heard about Stevens’s unusual mice, Martin Evans had requested breeding stock as well as several transplantable teratomas from Stevens. Like other developmental biologists, he was full of anticipation at the prospect of having this alternative route into early mammalian development at his fingertips. Young mouse embryos were only “the size of a reasonably fat full stop,” as he describes them today, and hard to collect from inside a mother mouse. When doing his doctoral study of frog genes at University College London, Evans had discovered that early frog embryos weren’t much easier to work with. Their eggs might look commodious, but inside was mostly yolk. By comparison, Stevens’s mice and their teratomas could be shipped straight to you, making for a much more convenient study.

By 1970, a handful of other scientists, including Gordon Sato and Boris Ephrussi, had shown that the stem cells—embryonal carcinoma cells—of mouse tumors could be kept alive in a dish, dividing and dividing. A standard culture medium was all that was called for. As much as Roy Stevens and Barry Pierce had pored over terato-

mas, they’d never grown their stem cells long-term; those they inspected were mostly dead ones that were present in tissue slices they had shaved from the teratomas of dead mice and mounted on slides. And so, knowing that a teratoma’s stem cells could survive in vitro was a very big deal, for only then would scientists obtain a realistic picture of how stem cells in mammal embryos grew and differentiated. To make EC cells flourish in culture, the researchers added feeder cells, just as Howard Green in Boston would do a few years later to ensure that his skin keratinocytes kept multiplying without differentiating. “Differentiation for a stem cell is the kiss of death; once it differentiates, it’s no longer a stem cell,” cites Evans. Knighted in January ’04, Evans—a jaunty Englishman whose bushy eyebrows rival Francis Crick’s—is now at Cardiff University where he directs the School of Biosciences.

Within earshot of the traffic along London’s busy Gower Street, Evans plunged after key questions about EC cells, questions that Jacob later summed up at Cold Spring Harbor: “Do EC cells really differentiate in vitro? How many types [of cells] do they produce? Can they be induced to differentiate?” Science-related enigmas had attracted Evans since his toddler days. He wasn’t much older than four when the sight of workmen at his family’s home in Hertfordshire mixing water with cement had made him so curious and determined to find out why cement hardened that his parents had ordered him to keep his distance, for fear he would get in the way. He grew up loving chemistry and expected to become a chemist, until courses in high school and then at Cambridge drew him toward biology.

For the latest inquiry, Evans’s jumping off place was Barry Pierce and Lewis Kleinsmith’s crucial 1967 landmark experiment that demonstrated beyond a shadow of a doubt that the different cell types present in a living mouse’s teratoma indeed arose from one cell and one cell only. The first order of business was to grow up large populations of EC cells as a few other labs were also trying to do. Evans laced his EC cells with feeder cells and got them growing. Rigorously cloning, or reproducing, a single type of cell with the aim of

getting a crowd of its progeny wasn’t a snap, however. To begin with, cells are small and can glomp on to each other, and it was hard knowing you were starting with one particular kind of cell—in this case, a stem cell—which is what cloning a cell and getting pure populations is all about.

Arriving on the scene in 1973 was Gail Martin, Evans’s first postdoctoral fellow, who had developed a knack for culturing cells while obtaining her Ph.D. in molecular biology from the University of California, Berkeley. Martin devised a method whereby she looked into a microscope and carefully picked up a single cell with a micropipette—making sure she had a stem cell, and not a stem cell stuck to a muscle cell, for instance—before dropping it into a well and growing its progeny in the company of feeder cells. Before long, the Evans lab had thriving batches of tumor stem cells that Evans and Martin kept undifferentiated by splitting the populations before they became too dense, just as Howard Green in Boston was doing with his skin cells. It was all part and parcel of learning how to manage cells outside of the body.

Next on the agenda: Would their flourishing EC cells differentiate into as many cell types in vitro as they did when inside a teratoma? Recollects Evans, “I plated out some cells, and I went away and left instructions with my tech to put more medium on the cells from time to time.” More plasma, more vitamins. When he came back, a surprise was waiting for him. “The cells were differentiating magnificently—beating cardiac muscle cells, cartilage, nerve, skin cells, and more.” As soon as the cells had started piling up, they began to differentiate, which was entirely consistent with what happens to stem cells inside the embryo when they start piling up.

Gail Martin reached the same destination by a different route. When she took EC cells off feeder cells, they aggregated into little round balls and similarly started to specialize. Roy Stevens had seen these clumps of cells, termed embryoid bodies, form when he injected stem cells into the abdominal cavity of mice. Staring into her microscope one day, Martin experienced one of those rare moments in a

scientist’s life when the fog completely lifts. “It suddenly clicked” that these clumps “were mimicking normal development,” she recounts. Upon aggregating, they made an outer layer of endoderm, the same way that the early embryo’s inner cell mass did after the embryo arrived in the uterus. That the tumor stem cells “had retained this knowledge” of what their normal counterpart cells did in the developing embryo “was mind-boggling,” says Martin, who presently directs the Developmental Biology Program at the University of California, San Francisco. Here was more proof that a teratoma’s stem cells and an embryo’s stem cells were mirror images of one another, even though one type was normal and the other malignant.

By 1975, Martin Evans’s London lab, as well as François Jacob’s Paris lab, had demonstrated that a single cell in a dish could differentiate into an array of cell types. Knowledge was extending. “If aggregation stimulated stem cells to differentiate, preventing their aggregation kept them from differentiating,” notes Gail Martin. Furthermore, the addition and withdrawal of feeder cells could significantly influence this outcome. Martin Evans would increasingly recognize the value of selecting the right serums and optimizing certain culture conditions, so that he could maintain control over his stem cells in culture. It was a balancing act of sorts.

Would the average pedestrian walking down Gower Street past the university’s anatomy building have cared a whit that a few flights up a creaky wooden staircase scientists were able to grow undifferentiated cells that generated a potpourri of specialized cells? They might have cared had someone proposed the following. Given the fact that you can refurbish a lawn with grass seed or rebuild a forest with transplanted saplings, try imagining doctors repairing the human body with the very cells that compose it. Regenerative medicine was a fairly new concept, yet its antecedents were everywhere to be seen. One notable example was organ transplants, the numbers of which were on the rise now that researchers had begun to understand how to appease the immune system. The first successful human organ transplant had transpired at what is now the Brigham and Women’s

Hospital in Boston as far back as 1954. A kidney had been transferred from one identical twin to another. As the 1970s lengthened, bone marrow and liver transplants were also on track and supported the idea that cellular tissues were transplantable.

The progress borne out in Martin Evans’s London lab and at other benches created new challenges. Since researchers had stem cells from mouse tumors living in a dish, they now wanted to try their hand at directing these charmed cells. One wish, as seen, was simply to keep EC cells self-renewing without differentiating. As long as teratoma stem cells held to this pattern, they stayed pluripotent. Yet they seemed intent on differentiating, and indeed were genetically predisposed to do so. Another trick that researchers hoped to master was how to steer these stem cells down a specific path toward a certain fate. By the mid- to late ’70s, they had identified various chemicals that could do this to a certain degree. Retinoic acid, for instance, could induce EC cells to differentiate into endoderm; dimethylsulfoxide changed them into muscle cells. This second trick, however, was a good deal harder than the first one.

Researchers were trying to make stem cells do things in a dish that they likely didn’t do in their natural setting. Maintaining them in an undifferentiated state, for instance, “is asking them to freeze their differentiation program,” notes Virginia Papaioannou. When put through hoops by researchers, stem cells in culture became artifacts of a contrived system, and couldn’t be considered as normal as they once had been back in their real home.

Because embryonic stem cells from mouse tumors bore such a strong resemblance to those in embryos, for the time being scientists had plenty to occupy them. Still, the normal kind—stem cells that resided in mammalian embryos—clearly couldn’t remain out of reach

forever, not at the rate that other biologists were descending into the embryo. Some of these bold forays are worth a brief visit, especially because they put the spotlight on stem cells that much more brightly.

Better culture recipes and exceptionally fine needles and other tools invented in the late ’50s and ’60s enabled embryologists to test young mouse embryos in ways that, risky as they seemed, did not necessarily harm them. For instance, a minuscule eight-cell mouse embryo could be removed from its mother’s oviduct, further grown and kept alive in a petri dish, and later returned to the uterus of a foster mother, who would give birth to the resulting baby mouse. The early mammalian embryo—a mere speck in a dish—was becoming accessible, which gave scientists ideas and the freedom to experiment. Now that egg cells, sperm cells, and embryos could be housed in dishes, experiments were inevitably headed toward in vitro fertilization, the ability to fuse egg and sperm in culture and create an embryo.

In the ’50s and ’60s as well, a longtime question about the embryo had found some answers. As an embryo grew and its embryonic cells lost potential—their fate pulled in the direction of one tissue or another—were their genes permanently lost or deactivated once they were no longer needed? Or did an individual’s genes, an entire set of which exists in most every cell, remain useable throughout life, every gene capable of encoding a complementary protein at a moment’s notice? If genes were lost, “it could explain how cells differentiated,” notes Marie Di Berardino, an eminent developmental geneticist and cloning authority at Drexel University in Philadelphia. An early theory based on observations in worm embryos that “pieces might be breaking off the main part of a chromosome,” with genes literally discarded each time a cell took another step toward a particular fate, imparts Di Berardino.

A useful test would be to extract the nucleus of a cell taken from an older embryo and put it into an oocyte, a one-cell unfertilized egg, whose own nucleus had been removed. Would the older substitute nucleus—really its genes—know to respond to chemicals within

the egg cell that tell the nucleus to start dividing and catapult it into a developing embryo? If so, it would reveal that genes in the older cell hadn’t been lost, but were still in place to set off all the bells and whistles that drive normal development.

Robert Briggs and Thomas King at the Institute for Cancer Research in Philadelphia (now the Fox Chase Cancer Center) put their shoulders to the experiment beginning in 1950. They took the nuclei of cells from slightly older frog embryos and implanted them into frog eggs. Indeed, something in the eggs’ cytoplasm reprogrammed the nuclei and enabled them to start the growth of a new creature from scratch, and the historic outcome was perfectly healthy tadpoles. When Briggs and King repeated the experiment using the nuclei of cells from even older individuals, in most of these cases the development of new tadpoles ran amuck. Still, their progress left the impression that the cloning of creatures wasn’t a ludicrous pursuit, at least as far as frogs and other less complicated animals went. Hardly anyone imagined mammals would ever be cloned. “In the beginning—in the ’50s—there were all those jokes about cloning,” recalls Di Berardino. “There were drawings of Tom King pictured as multiple people with a golf club in his hand. No, we never thought it would be done in mammals. The sole thing driving the work back then was basic research.”

“I seem to remember that it was thought unlikely that mammalian nuclear transfer would be successful for a long time, if ever,” recalls John Gurdon at the University of Cambridge. A mammal’s egg was so much smaller than a frog’s egg that replacing its nucleus would surely do it damage.

Yet it was Gurdon, when he was in the Department of Zoology at Oxford, who made the cloning of all kinds of animals that much more of an eventuality. In 1962, he reported that upon removing the nuclei of intestinal cells of tadpoles, which he took to be differentiated cells, and planting them in egg cells, many among them developed into normal tadpoles. There would be a long debate over this. Some colleagues suggested that perhaps Gurdon had inadvertently

scooped stem cells out of the intestine, an organ rich in stem cells, and not differentiated cells as he thought. Not until forty years later—in 2002—would a research team vindicate Gurdon, proving that his previous science was without holes. Differentiated cells truly did retain the use of every gene. “Briggs and King, and then Gurdon showed us that a nucleus of a cell from a later stage was not irreversibly committed to what it was doing, but that it could go back and restart development. This gave us the concept that the nucleus retained the potential to do everything,” provides Virginia Papaioannou at Columbia.

The modern era of cloning—or “nuclear transfer,” as scientists refer to it—had arrived. The basic strategy is to move a nucleus from the cell of one individual into an egg cell taken from the same, or another, individual in order to produce a genetic replica of the cell’s donor. Theoretically, the donor cell can come from any part of an embryo, fetus, or adult, although the older the donor cell, the less successfully its nucleus, when transferred, appears to be able to revert to a totipotent state.

Hans Driesch is often credited with being the first scientist to clone an animal, since he forced apart the cells of a sea urchin embryo and ended up with multiple individuals. This overlooks Abraham Trembley, however, who, when he cut a hydra in two and produced two hydra, was a bona fide cloner many years before Driesch himself was so much as an embryo. “No one had ever even guessed the possibility that a snip with scissors could make two animals from one,” wrote Trembley’s biographer John Baker. Briggs, King, Gurdon, and others brought a newer mode of cloning to the fore, however, one that introduced the idea of reproducing complex vertebrates from just one cell. This was a thunderous departure from the regular route of reproduction and its joining of two cells, one from male and one from female.

Still other experiments in the ’60s displayed the extraordinary versatility of an embryo and its embryonic cells. Reminiscent of Driesch’s 1890s demonstration that each cell of an early embryo has

the pizzazz to become a full individual, in the early ’60s Beatrice Mintz and Andrzej Tarkowski independently showed that you could combine two mouse embryos in a foster mouse’s uterus and end up with one, and only one, normal-sized mouse, whose salt-and-pepper fur was the only visible token of its dual-embryo beginning. Animals made from two or more embryos were dubbed “chimeras,” after the ancient Greeks’ famous amalgam of lion, goat, and serpent. Like Driesch’s flexible sea urchin embryos, Tarkowski and Mintz’s chimeras were valuable proof that an “embryo’s cells were not determined in what they could do; they responded to their environment,” describes Virginia Papaioannou. Cells in the merged embryos had done whatever it took to end up as one normal-sized mouse, a sign of how tightly regulated an embryo in the uterus is and also how versatile its inner stem cells are.

Mouse chimeras made possible an experiment in the mid-’70s that completely riveted the attention of numerous biologists, gesturing as it did toward an ingenious new avenue into an animal’s genes and the diseases locked therein. The experiment, already described in Chapter 2, was yet another example of the early embryo’s flexible, alterable cells. Scientists injected stem cells from a mouse tumor into a mouse blastocyst, depositing them next to its inner mass of stem cells, and in a good many instances the added tumor stem cells, far from jeopardizing the embryo’s growth, helped form a normal mouse pup. The tumor cells, instead of promoting tumors, fell under the spell of normal development. Here again was a case of cells, or really the genes scrunched inside them, changing in response to their environment. In other instances, however, the normal environment failed to matter; because of the added malignant cells, the embryo would acquire tumors.

While the public mightn’t have been ready to appreciate this experiment, biologists saw it as a potential tunnel into a mammal’s tens of thousands of genes, about which so little was known. Scientists might target a specific gene, modify it in a stem cell, put the stem cell into a mouse embryo, and thereby cause a change in the

growing mouse. Any gene of the thousands present in each cell could be targeted, its function explored. Embryologists might be able to study genes that steered development; medical researchers might be able to knock out and study genes associated with a disease. Some distant day, maybe this same approach could be parlayed into fixing disease. “Even back then, the thought of using this technique for cell therapy was very much on our minds. That was a driving force,” says Papaioannou, a member, along with Martin Evans, of one of the three teams whose 1974-75 work attempted this promising new approach. Although EC cells were malignant, it seemed safe to use them as vehicles for modifying genes, since when added to the embryo they appeared to contribute to the making of normal mice.

All of these marvelous visions depended on an EC cell’s gene modification, once inside an embryo, becoming part of the animal’s germ line—its eggs or sperm. That is, the modified stem cells would have to differentiate into, among other things, egg and sperm cells. Only then would future generations have this genetic change. Yet when attempting this goal, although one group claimed success, François Jacob, Martin Evans, and others had no luck. The egg and sperm cells derived from stem cells of teratomas couldn’t start an embryo. Their chromosomes, scientists began to concede, must be intrinsically flawed—apart from any gene alterations inflicted on them. This realization caused researchers bitter disappointment, and work on this front would slow to a crawl by the late ’70s.

Their probings of teratomas had proven so productive in so many ways, biologists had nearly forgotten that something might be wrong with the stem cells of these malignant growths. Now that they couldn’t get past this fact, there was all the more incentive to do what no one had done: isolate the real thing, normal stem cells from nor-

mal mouse embryos. “When I moved to UCSF in 1975,” Gail Martin recounts, “lots of people were trying to [fetch] stem cells directly from embryos. People were concerned that something was wrong with EC cells and thought it would be really great to make cultures of normal embryo cells.” Scientists’ newfound ability to grow a teratoma’s stem cells was a big part of this readiness, according to Martin. If EC cells could survive in a dish, stem cells from embryos probably would as well.

The general location of stem cells in the early mouse embryo was no secret. But attempts in the late ’70s by a number of investigators, including Martin Evans, who had relocated to Cambridge University’s Department of Genetics in ’78, and Gail Martin, who had been given her own lab at UCSF, kept ending in failure. Hampering their efforts was the scarcity of stem cells in an embryo’s inner cell mass, as well as no clear idea of the ideal time to harvest these cells. Evans came to think that the right time was the far side of the blastocyst stage, “before or about the time of implantation,” he mentions today, which added the hardship of hunting down and retrieving the dot of an embryo from the uterus. To “find this tiny patch of embryo” made for “a difficult dissection,” he relates. A scientist also had to have prepared the proper culture to put the cells into once they were out of the embryo, or face losing them. Evans sought a culture mix that would encourage the stem cells to multiply, not differentiate; yet he was retrieving them at a stage when implantation signaled them to start differentiating.

In early 1980, Evans had a chat with a colleague at Cambridge that would prove to be a turning point. As he told Matthew Kaufman, a lecturer in anatomy and an up-and-coming authority on mouse development, he had been refining his methods, but try as he might, he still hadn’t been able to sustain the growth of stem cells from mouse blastocysts in culture. The catch might be, he felt, that a blastocyst’s inner cell mass simply provided too few cells to work with. If more existed, it might raise the odds of converting at least a few cells to long-running cell lines. “Martin was desperate to get cell

lines going, and I offered him a suggestion—a trick,” recollects Kaufman. There actually was a way to increase the number of stem cells in a mouse embryo’s inner cell mass, he told Evans, a strategy that had been written up in a paper or two. It involved removing a female’s ovaries immediately after she got pregnant. Without this source of estrogen and progesterone, the early embryo failed to implant and remained afloat in the mouse’s uterus, its inner cell mass continuing to increase in size. It could wind up with double, triple, even quadruple the number of stem cells it normally has, according to Kaufman, who is now at Edinburgh University.

Kaufman’s recommendation was right on the mark. By interrupting implantation in this fashion, Kaufman and his technician created blastocysts that in a microscope looked like Graf Zeppelins, as Kaufman describes them, or little sausages. These little oblongs were put into vials with tissue culture medium and carted the brief distance from Kaufman’s lab in the Department of Anatomy across Cambridge’s medical campus to Evans’s lab in the Department of Genetics. There, Evans placed them in his own specially prepared culture medium, complete with feeder cells. A few days later, using a hair-thin bore pipette, he picked apart these tiny mouse embryos, dispersing the cells of their inner cavities in culture. After a few more days, there they were—little colonies of pluripotent cells growing away! He knew at once they were stem cells, he says, because “I was so used to looking at EC cells” in teratomas. By repeating the experiment, he eventually cultivated fifteen stem cell lines.

Evans’s foresight, tenacity, and fastidious refinement of culture ingredients had paid off. For the very first time, science had embryo-derived stem cells from a mammal lying in a dish and “marking time,” Evans observed some years later. When they were hidden away in the embryo, these cells were caught up in “an inevitable time progression,” he described. “But taking them out into tissue culture, into the petri dish, they’re marking time. And we can keep these cells, still with their ability to develop in any way, as millions and millions of cells in our petri dish.”

An ocean and continent away, Gail Martin at UCSF was on track as well. Five months after Evans and Kaufman published their important news, a paper by Martin in the Proceedings of the National Academy of Sciences reviewed how she had reached the same milestone, albeit by a different route. Instead of beefing up a mouse embryo’s inner cell mass, Martin had retrieved stem cells from the inner cell mass and jump-started them by growing them in a “conditioned” medium that teratoma stem cells had previously grown in, a technique she learned in graduate school. Proteins released by the teratoma cells made the embryo cells thrive. In her December ’81 report, she referred to her cells as “embryonic stem cells,” a name that would give way to the nickname “ES cells,” paralleling a teratoma’s EC stem cells, and one that would stick.

Neither of the special techniques that helped Martin Evans and Gail Martin with their pioneering work of isolating pluripotent stem cells from mouse embryos—implantation delay or conditioned medium—are necessary today. A scientist can now simply take an embryo out of a pregnant mouse, put its stem cells on feeder cells, and grow them. “Who knows why this didn’t work to begin with,” says Martin. Beginning experiments are always hard, she attests. And then one day, all the essentials fall into place.

The chief motivation for collecting normal stem cells from normal mouse embryos had been the hope that these cells could serve as vehicles for manipulating genes in embryos. Martin Evans persevered toward this goal, with long hours logged by postdoc Elizabeth Robertson and grad student Allan Bradley. (Robertson is now a principal research fellow at the Wellcome Trust Centre for Human Genetics in Oxford; Bradley, the director of the Wellcome Trust Sanger Institute in Cambridge, England.)

By the time of the Cold Spring Harbor conference in the fall of 1982, the one at which François Jacob made the point that biology was still at a primitive one-dimensional level, Evans and his lab crew were actually in the midst of finding their way down the rabbit hole into a second dimension. By ’83, they had injected embryonic stem cells into a mouse blastocyst, whereupon the cells’ progeny got into the mouse’s germ line, and therefore into the mouse’s descendants. By ’87, they had extended this experiment. Altering an embryonic stem cell’s gene in culture by means of a retrovirus, they then placed the changed cell into a mouse embryo. “So we had engineered the first mouse to carry a disrupted gene,” cites Allan Bradley.

In the early to mid-’80s, two other scientists—Mario Capecchi at the University of Utah in Salt Lake City and Oliver Smithies at the University of North Carolina in Chapel Hill—had separately applied their wits to the feat of gene targeting, or replacing one form of a gene with a modified form. They had worked out an alternative way of installing a gene change, ingeniously hooking on to the fact that DNA is a twine of two corresponding threads, each gene having two copies. When introduced into a chromosome, a modified gene automatically seeks out its alter ego and lands in its proper place on the chromosome.

This long string of advances meant that biologists now had the supreme luxury of exploring the genome one gene at a time. The concept of a “knockout” mouse had been born. Virtually any known gene among thousands could be singled out, altered in an embryonic stem cell, and that stem cell could then be slipped into an early embryo. “We could now do real experimental genetics in the mouse,” elaborates Evans. If you altered a gene, “What happened? What changed?” Today, more than 5,000 different knockout mouse strains exist that are helping investigators analyze genes connected to heart disease, inner-ear defects, anemia, and cystic fibrosis; genes that influence the risk for cancer; genes that make the immune system work; genes that regulate early development.

All this came about because back in 1953 a scientist on the

Maine coast bumped into a chinchilla mouse with an inherited burden. For scientists, there was something doubly engaging about that mouse. First, its teratomas got people interested in the wonderously versatile stem cells found in these tumors. Second, the early embryos of Strain 129 mice just happened to give rise to normal stem cells much better than embryos of other mouse strains, which led to these stem cells being used for the making of knockout mice. In fact, up until recently most knockouts were Mouse Strain 129 kin. If it weren’t for these mice and the genomic depths they’ve opened up, observes Elizabeth Robertson at Cambridge, “We’d still be floundering around in the dark.”

This page intentionally left blank.