To Make a Mouse

By Jon Franklin


Copyright 1999 by The News and Observer




Chapter 1: The Quest

            To get a traditional recipe for mice you have to go back to the 1600s. Jan Baptista van Helmont, the alchemist cum scientist, had a good one. You took a wad of dirty underwear and threw it in a corner. A variant of the recipe added grain. After a few days, voila! Mice appeared. Spontaneous generation, it was called.

But Andrew Xiao was not impressed. He looked up sharply from his laboratory bench, where he was attending a mixture of water, enzymes, and dissolved mouse pup toes and tail tips. He’d never heard of such a thing.

“To make a mouse,” he said, “it’s harder than that. To make a mouse takes a year. A year and a half. It’s really very difficult to make a mouse.”

He directed his attention back to the rack of tiny vials in front of him. Each contained a bit of flesh from a baby mouse. If Xiao was as lucky as he was clever, at least one of those mice would carry a brain cancer gene he had fabricated from a monkey virus.

So far, such good fortune had eluded him. He’d hoped to get his mouse in a year, but the year had come and gone and still he struggled on, his project metamorphosing first into quest and then into odyssey. Xiao’s armament of knowledge and understanding seemed pathetically insufficient. Each day he faced anew the legions of chugging polymerases, slithering ribonucleic acids, and all the other stubborn little tongue-twisting robots that populated the new world that had emerged from the opium dreams of 20th century biology.

And at the heart of his enigma was the DNA molecule, master of all that lived, barricaded behind a thicket of riddles, brooding over its primordial secrets.

So if Xiao’s existence was in many respects far removed from the abracadabra of van Helmont’s day, his was still a sorcerer’s life, full of spells and special knowledge, where recipes for mice were exchanged with the utmost seriousness and the fortunes of an apprentice like Xiao might rise or fall depending, say, on how rapidly some extract of life moved through a charged gel.

In the meantime, survival came down to rigor and focus, interminable days of study and work, endless weeks that blended together in one long agony of effort.

Not long before, Xiao had broken up with his girlfriend. She studied biology at a university several hours away and he was a Ph.D. candidate at theUniversityofNorth CarolinainChapel Hill. It was just too far. Neither had the energy to make it work.

Now it was just Xiao — and, of course, his project.

Bending over his vials, he smiled, faintly, taking joy where he could find it. Underwear, indeed.




Xiao’s mouse, if he could make it, would be different from any creature that ever lived. In the purest sense it would not be a mouse at all, or not totally a mouse because each of its cells would carry a gene alien to its murine heritage.

The gene Xiao was attempting to transplant — the “transgene” — came from a virus found in green monkeys. In Xiao’s mouse, it should induce a susceptibility to a rare but exquisitely malignant brain tumor called glioblastoma.

The recipe had all been carefully thought through by Xiao and his mentor, Terry Van Dyke. Glioblastoma, at least in humans, usually struck in midlife. If the same was true in the mouse, then Xiao’s transgenic creation would have time to procreate before it died.

Xiao would then have a line of mice that could be multiplied and shared the world over. Scientists in other institutions and other countries could study them, and perhaps make their own genetic modifications.

It didn’t matter that glioblastoma was relatively rare in humans; the important thing was that it was explosively malignant. Extremely malignant cancers tended to be the easiest to understand, and the lessons could then be transferred to the more common, slow-growing forms. Research on rare lymphomas and leukemias in the 1960s and ’70s, for example, underlay the ’90s-era treatments for more common tumors of the breast, ovary, testicle and lung.

So while it might be unlikely that Xiao’s mouse would lead directly to a cure for human cancer, the young scientist could legitimately fantasize that it would have an impact. It might prompt some other scientist, somewhere, someday, to make an observation that would trigger an insight that would lead to a procedure that would provoke a discovery that would save the lives of Xiao’s own fellow human beings.

Meanwhile, and more prosaically, in the small, technical world of mousemaking, such a mouse would be quite a trick. If Xiao could pull it off it would be yet another credit to the Van Dyke laboratory.

And then there was the final matter, far from inconsequential, of Xiao’s own Ph.D. and, beyond that, a good postdoc appointment … and then a grant … and, perhaps, some day in the dim future, a tenure-track position at a good university.




Andrew Xiao was born 27 years ago in farawayBeijing. But even as a young child he had been acutely conscious of that other, famously capitalistic nation across thePacific Ocean. It was in the family, going back generations.

“My grandfather went to missionary school,” he explained … then hesitated, thinking back, his brows furrowing. “Or maybe it was my great-grandfather, I’m not sure. One of them. But because of that we all learned English, every generation. A lot of my family, and it’s a big family, came to the West.”

To preserve its Western heritage even through the cultural revolution, the family had had to create strong traditions of its own. The parents bestowed on each chilld a secret Western name. The firstborn male received a name beginning with “A,” which is how Andrew became Andrew.

The family listened to Voice of America and the BBC, and so as a child Xiao heard, for example, the Beatles. He liked them well enough, though not so well as to buy their tapes. His ear inclined toward classical music, particularly Mozart, Beethoven and Tchaikovsky. He liked the way the masters so carefully structured the music and then layered in levels of variation to produce something that seemed deeply spontaneous. Life itself was like that.

English, meanwhile, came relatively easy to Xiao. It was the language of science and, especially, molecular biology. By the time he was in college he was reading biology in English, without bothering to translate it. Ultimately his technical English would outpace his technical Chinese to the point that he would have difficulty talking about molecular biology with a Chinese-speaking scientist.

All these things were in his favor when, in 1995, he was accepted for a master’s program at theUniversityofMaryland. But nothing could prepare him for rock music, or for Americans’ constant preoccupation with material things.

At least he knew two familiar languages. English was an old friend. And the genetic code, of course, spoke in the same whispers in either hemisphere.

InAmerica, as inChina, he had concentrated on his studies and kept to his timetable. He got his master’s, and came toChapel Hillfor his Ph.D. He finished his classwork, freeing up his entire life to focus on the laboratory and the next big hurdle, which was getting his mouse.

That was the thing: The mouse. He needed the mouse. It all came back, every time, to that mouse.




Xiao was aware, as he worked, of the intellectual tradition he inherited. There was a saying in science that each generation saw farther because it stood on the shoulders of its predecessors. The most fundamental law of genetics, that like begat like, had been known by ancient peoples, long before the hooded monk Gregor Mendel began tinkering with his sweet peas, long before he discovered that genetic information came in discrete packages.

Later those packages would be called “genes.” But the root mechanism of genetics was not discovered until the 1950s, when James Watson and Francis Crick discovered that genes were packed inside the long, double helix molecules that Xiao would one day snip apart and patch together. The language of life had an alphabet of just four letters — chemical bases that would become famous simply as A, C, G and T.

The scissors Xiao would use to do his tailoring were “restriction enzymes,” a class of proteins designed by nature to cut other proteins in different but very specific places; the men who discovered them won Nobels in 1978. The discovery of other enzymes, designed by nature to rejoin the severed ends of DNA strands, represented the life’s work of still others.

And so it was that science proceeded, generation upon generation. History dripped from every tool Xiao used, every strategy he employed, almost every thought he had. As a practical matter this imbued his work with an almost ritualistic significance, heavy with legend and scientific derring-do.

He knew his history well, but none of it weighed him down. He hadn’t been born yet when Watson and Crick figured out DNA, and was but a small child when restriction enzymes were discovered. And while Xiao might be precocious, and though he was an extraordinarily focused young man, he was still a young man, and he coped with the past the way new generations always do.

From his vantage point of 27 years of life and his perch on the razor edge of molecular biology, any process older than three years was a “traditional method.” If it went back more than a decade, it was “olden times.”




It was a cast of mind that fit the spirit of the moment. Molecular genetics was in the throes of revolution — like microbiology in Pasteur’s day, or physics in Einstein’s. Every experiment was a leap into the unknown, and every month the journals were full of yet more revelations. Anything seemed possible. At a time like this, history was not just a subject to be studied. It was something you could make.

Every day, men and women like Xiao did things that had never been done before. Sequencing new genes. Figuring out the structures of the molecules that read them. Engineering bits of DNA into virus shells, and then vectoring the result into living creatures in hopes of changing, fundamentally, what those creatures were.

It was difficult to overstate. Molecular biology had become a blur of science, engineering, art and industry; genetic information was accumulating so fast that computers couldn’t keep up. The National Science Foundation had launched a special program specifically to catalog and analyze the information molecular geneticists were collecting.

The whole enterprise twisted the mind, sowed dreams of cures for disease and industries for the new century. Human enzymes and hormones, such as insulin, could be produced by other animals and made into cheap, readily available medicines. The genetics of cancer could be discovered, and with the same tools the disease could be prevented and cured. Newspapers were punctuated with stories about advances in the understanding of cancer, heart disease … even aging.

A new world was being born, and it was happening very rapidly. Armies of molecular geneticists and biotechnologists converged on the hot spots of biology: places likeStanfordUniversity, the Massachusetts Institute of Technology, the Research Triangle. Construction crews erected steel and concrete molecular biology laboratories on university campuses; laboratory supply houses and drug companies set up shop nearby.

No one knew for sure how many people in the Triangle earned their living manipulating genetic code. But by 1999 there were a lot of them. Van Dyke, the head of the UNC laboratory where Xiao worked, guessed the number to be between 15,000 and 20,000.

In the subculture of biology, Van Dyke said, biotech workers were often known as “lab rats.” She laughed. “That’s probably because we spend our lives scurrying around laboratories.” But she and the young scientists she trained were more likely to refer to themselves as “mousers” — a word coined by Oliver Smithies, the grand master mousemaker, who worked nearby on the UNC campus.

Whatever their name, they did the hands-on work of the brave new world. If they lived a somewhat isolated existence, and were only dimly perceived by outsiders, the thing that set them apart was the very nature of their work: to re-engineer life.

But a young man in hot pursuit of his Ph.D. had no neurons to spare for such panoramic issues. Xiao had spent his life disciplining himself to focus on the task at hand, and never mind the revolution — Xiao’s own drama was unfolding, day by day, in the cramped and cluttered confines of the Van Dyke laboratory.

The laboratory was probably originally conceived by some architect as a single large room. But in the realpolitik of academic science, where space was money and status combined, open spaces didn’t stay that way. Van Dyke’s laboratory had been divided into a warren of walkways lined with black laboratory benches and, above and below, cupboards.

The space was relatively orderly, but to the uninformed eye it was a clutter of solutions, glassware and instruments. Refrigerator doors were covered with the kinds of things that occupied a genetic engineer’s mind: children’s drawings, jokes from American Scientist, genetic schematics of plants and animals. Scientists in white lab coats sat at their benches, focused on what they were doing, or moved, preoccupied, through the passageways. There was little noise: the distant whir of a centrifuge, the click of computer keys, soft music from a radio. And in the air, the faint but ineradicable scent of mouse urine.

Xiao’s allotted space included a desk in a far corner, next to a window, and perhaps five feet of adjacent lab bench. This was where he would make his mouse and, having made it, would characterize its genetics.

It was all mapped out. It had to be. Times were tough at universities. Only one out of every four qualified postdocs was getting a faculty research job.

So he would have to be spectacular. He would have to use his mouse, or something he had learned from it, to develop a really hot line of research. That, with luck, should lead to five or six years of postdoctoral research which, in turn, should attract enough federal money to serve as a dowry when he went looking for a tenure-track position at a university.

First, though, the mouse. He had to have the mouse.

It was the way things were generally done, in his field. Certainly in Van Dyke’s lab.

Yes, he said, there were cases where a young researcher like himself, through no fault of his own, had failed to produce the requisite mouse. The masters of the laboratory usually figured out some way to keep the student around a few more years, to try again and again …

Xiao sat on his stool, elbow on the hard black surface of his laboratory bench, considering for a fleeting instant the fate of the Ph.D. student who failed to get his mouse. Then he shook his head emphatically.

“No,” he said. “That would not be good. I will get the mouse.”




Once, not all that long ago, it had been enough for a Ph.D. candidate just to get a transgene into a mouse and have it “take,” producing a functional transgenic mouse. But as the technology advanced, the requirements ratcheted up.

Xiao’s mouse was expected to be medically useful, which meant that the transgene had to be a powerful one — powerful enough, in this case, to cause cancer.

That changed the nature of the exercise. As soon as you undertook to transplant a truly powerful gene, unexpected things started to happen. The transgenic fetus aborted before reaching term. Or it was born with grotesque malformations. Or it was quickly consumed by spreading cancer.

Such problems often had to do with differentiation, the process by which fetal cells parted company and went down separate metabolic pathways, some of them becoming, say, liver cells, and others becoming heart or brain cells. Differentiation was one of biology’s central enigmas, and so the problem had the full attention of scientists.

The way life worked, each cell had a full complement of genes. Yet, once differentiation occurred, most of those genes turned off. A liver cell made only liver proteins and a brain cell made only brain proteins. And that was sometimes a problem with transgenic mice: The transplanted gene would express itself in every cell of the mouse’s body. Life, at least higher forms of life, could not tolerate that.

But how did each cell know what DNA code to read, and what to ignore? The answer, when it came, followed a familiar logic — familiar, at least, to computer programmers.

Messages sent over the Internet had “headers,” which were basically address labels instructing servers where to send the code. Genes turned out to have similar headers to designate which cell type they had evolved for. The header, which appeared directly in front of the gene, said, “If you are a brain cell …” or “If you are a kidney cell …”

DNA was read by proteins called polymerases, which chugged up and down the DNA strand reading some genes and ignoring others. At least part of their discrimination was based on the header that preceded the gene. Different cells had slightly different kinds of polymerases, and if a kidney polymerase, for example, read a header that said “If you are a skin cell …,” then it skipped the following gene.

Biologists didn’t call the “if you are” sequences “headers,” however. They called them “promoters,” because they promoted the manufacture of a protein. So it was that liver polymerases read genes with liver promoters in front of them, brain polymerases read genes with brain promoters in front of them, and so on throughout the thousands of different tissue types in the living animal.

By the time Xiao arrived inChapel Hill, mousers were just perfecting their ability to manufacture specific headers and attach them to the genes they transplanted, thus ensuring that the proteins those genes coded for would be produced only in the target organ.

And that, now, was Xiao’s task. The cancer gene he was using was a powerful one, and in its unrestricted form it would immediately express all over the mouse’s body, and the mouse would die in the womb. That was why, to date, no mouse model of glioblastoma had been made.

He would transcend this difficulty by splicing a promoter in front of the transgene, so that it would only be read in the glial cells, in the supporting structures of the brain.

Or that, at least, was the theory that had brought Xiao to the threshold of his Ph.D.

But now, for some reason he could not seem to fathom … the mouse eluded him.




In his childhood, Xiao and his classmates had worshipped scientists the way American adolescents worshipped rock stars. To be a scientist, if you were a Chinese boy, was to play the greatest game there was, to confront Nature herself, on her own terms, and wrest from her the secrets that would make human beings healthier and more prosperous. No calling could be higher.

In Xiao’s case, it was a dream that could actually come true. His father was an electrical engineer and two of his uncles were research scientists. He could do it too. He knew he could do it, if he just worked hard enough, studied long enough, made it the priority in his life.

So he would definitely be a scientist, but he didn’t know what kind until a high school teacher introduced him to the intricate, unseeable world of molecular biology. The young man was fascinated to discover that the liquids, lumps and goos that made up living tissues really weren’t what they appeared to be. What the eye saw was meaningless. The eye functioned on the wrong scale.

If you wanted to understand life, you had to focus down farther than the eye could see or the retina could record, beyond the reach of the light microscope, beyond even the resolution of the electron microscope. Life proceeded on a scale so small that only one instrument could record it: the human mind.

On that scale, everything was made of molecules, which were combinations of atoms. The molecules bounced around like magnetic balls, attracting and repelling one another, and producing reactions that translated to fire and thunder, plastics and computer chips.

Most common molecules were simple. Water was composed of only three atoms, two of hydrogen and one of oxygen. But the molecules of life were huge. Many of them contained hundreds of thousands of atoms linked together in structures that folded in on themselves and became hideously complicated.

Such molecules were industrious little machines — proteins, mostly, complete with moving parts: arms, legs, grapplers, pincers and cutters. Tractor proteins pulled loads of material from one part of the cell to the other. Others served as winches and cranes. Some welded other molecules together, fabricating new proteins, while others cut apart broken proteins to recycle their parts.

Still other molecules were, in effect, self-contained messages. Growth hormones were proteins with a particular shape; when their presence was detected by special receptor proteins on the surface of the cell, the cell began to grow and divide. Hormones of various other shapes instructed cells to stop growing, to change metabolic functions, to take on glucose fuel … or even to die.

If a young man could focus his mind down far enough he could glimpse one of the most fundamental truths of existence … that the very act of “living” was done by these workaholic little machines.

To learn how these machines worked was to be indoctrinated into what was for all practical purposes a secret order of scholars devoted to the contemplation of a single magical molecule: deoxyribonucleic acid. One filament of DNA strung together thousands of messages spelling out the recipe for men and for microbes, and everything in between. In recent generations these scholars, by sheer force of mind over complexity, had created a new science — even a new kind of wizardry.

For life, being a dance of machines, was not immutable. It could be changed, redesigned, even created.

To get a grasp of what “living” meant on the scale of molecules, the apprentice wizard had to master and apply knowledge from a whole collection of other disciplines, from chemistry and physics to computer science. But it was clear early on that Xiao had what it took: intelligence, energy, vision, patience, endurance — and, of course, desire.

The desire was an integral part of it all. For while it was true that the new biology was very complicated and mechanistic, and that it required great discipline and effort, there was also something indefinably magic about it … and Xiao was definitely bewitched.




The decision to go into molecular biology dramatically raised the educational bar for young Xiao. As the son of intellectuals in his nation’s capital city, he had been assured a good education and, eventually, entrance into a college of some sort. But if he was going to play such a high-stakes game as biochemistry, the college would have to be a very good one, indeed.

What’s more, he would have to graduate at the top of his class because the next step was to get admitted to a master’s program in theUnited States. That’s where biology was happening.

InBeijingthe schools were rated, as were the students, and only one out of 15 students could get into the best schools. Once in, every course was mandatory and every year was make or break, because only a fraction of each class was allowed to go forward into the next grade. After three years in high school, students took a national exam; their scores determined who would go to which college.

In this way Xiao’s youth was willingly sacrificed on the altar of science. He studied very hard and, in succeeding, earned the privilege of studying harder, later. He hacked at the code of life, memorizing the names of all the little robots, the builders-up and the tearers-down of metabolism. He got the hang of thinking in nanometers and megabases and daltons.

He read and digested everything he could find about the nucleus of the cell and the library it contained — the grammar and punctuation of DNA, and exactly how the polymerases read it.

In bacteria, the genetic material mixed freely with the protoplasm of the cell, but in more efficient creatures the DNA was walled off behind a fatty, lipid membrane — the nucleus. Behind that membrane the double helices were tended by several classes of librarian enzymes that catalogued, read, translated and repaired the code.

When new robots had to be manufactured, the librarians read off the pertinent bit of DNA and created a ribbonlike ribonuclease, which contained the design for the robot. The ribonuclease then wormed its way through the nuclear membrane and out into the cell proper, where the robot factories were.

If this was a source of fascination to Xiao, it was also his toolbox. Scientists who re-engineered cells couldn’t do so directly — the clockwork mechanisms were too tiny, too intricate and too fast-moving. The engineers had to isolate the various robots and subvert them to their own uses.

Viruses were favorite tools because they were natural hijackers. A virus was nothing but a stretch of genetic code surrounded by a protein capsule designed to punch through the cellular membrane of its prey. Then the viral DNA commandeered the protein factories to make new virus capsules. It also insinuated itself into the cellular library, where it snipped out bits of the code that might have allowed the cell to defend itself.

For decades experts thought viruses were the smallest living things, but by the time Xiao was in high school they’d discovered a parasite even smaller. It was called a “plasmid,” and it was nothing more than a loop of DNA — pure genetic code. The only thing a plasmid could do was make another plasmid, and it could only do that inside the cytoplasm of a bacterium.

But plasmids were wonderfully malleable. DNA could easily be grafted into them and taken out again. And they were durable, so durable that they could exist outside their host cell. They were perfect for storing and duplicating genes.

Life was an amazing thing, so complex it seemed like magic, yet every process was specific, definable, learnable. It just took determination, and that was something Xiao had. He studied, he worked, he took one step at a time. And then, finally, in the fullness of time and a place halfway around the world, he was an understudy mousemaker working furiously to earn his Ph.D.




But no matter how much you knew, it was still hard to make a mouse. The recipe was complex, exacting and tedious, and had to be followed without error. The engineer could never see the object he was engineering. Everything was blind. Yet a single mistake … or one piece of bad luck … and it was all for nothing. So making a mouse was to make many mice, most of them rejects.

And so it went, endless, meticulous toil in the laboratory, the days coming and going in a flash, blending into weeks, the weeks becoming months, summer and fall, winter and spring. Again and again Xiao followed the recipe, varying it slightly, teasing the system, thinking that a different host mouse might help, or that slightly different procedures would produce a significantly different result. And still, yet still, no mouse.

The basic recipe used a cancer gene that came from another laboratory. To multiply it into usable quantities, he first spliced it into the DNA of a plasmid. Then he teased the plasmid into bacteria and grew the bacteria out in a flask so that the plasmids, and their transplanted gene, would multiply as well. Then he purified the plasmids and harvested the gene by adding restriction enzyme “scissors,” which neatly and accurately cut the cancer genes out of the plasmid loops.

That was the first step.

The entire process took a long time. Genetic material was purified, amplified, passed through electrified gels and inserted into mouse eggs, which were in turn inserted into host mice. Days disappeared into the black hole of scientific concentration as Xiao finished one batch and started another, worried over host mice, ran genetic tests on the resulting pups. So far, all for nothing.

Oh, he had come close, all right. But never close enough.

Sometimes it seemed like bad luck followed him. He’d produced a number of transgenic mice over the past few months, but they’d all died in infancy.

The biggest heartbreak had come just a few months earlier. Xiao had isolated some viral gene fragments, and a steady-handed co-worker had injected them into the fertilized eggs collected from a line of black mice. The eggs were then implanted into the womb of a white mouse.

The good news was that the pregnancy took. The bad news was that most of the pups were born with deadly malformations of their brains. Only one tiny male survived.

Xiao had cut off the tip of its tail and one of the digits of its front paw — enough to run a genetic test. The result was positive: The mouse carried the cancer gene.

What’s more, after a few weeks’ growth it became clear that it also had a brain tumor. Its movements were curtailed, and for a time it lost its hair. The gene was expressing! Extraordinary metabolic things were happening in the tiny black biosystem.

Xiao was ecstatic. He haunted the mouse facility, hovering over his creation, waiting for it to reach maturity so that he could breed it.




From a scientist’s point of view the nice thing about mice was the same thing that was so awful about them from most other points of view: They bred like, well … like mice. A female mouse could conceive at just four weeks, and 19 days later give birth to a new generation. The black mouse was a male, though, and it took males longer to mature: two interminable weeks longer.

But the time finally passed and, as soon as he could, Xiao introduced a female into the cage and waited for signs that she was pregnant.

A day passed, two. Nothing happened.

Puzzled, Xiao removed the female and replaced her with another. But more days passed, and still nothing happened.

He tried a third female. Nothing.

He checked the male over, physically. It was visibly intact, which was to say that it had testicles and there were no outward signs of malformation.

He tried another female, out of a forlorn hope, but by that time he wasn’t surprised when nothing happened. He already knew, in the pit of his stomach, that the precious black mouse was sterile. It was a mouse, and he needed a line of mice. It was another failure.

What was wrong? Why did most of his mice die? Why couldn’t the rest reproduce? Was he making some mistake?

There was no way to know, nothing to do except let himself be disappointed for a moment … and then move on. To make a mouse was to make a lot of mice. He knew that.

But this many?

The next time he would just have to work harder, be smarter, make no mistakes. He would vary the recipe in some critical way. In fact, as he thought about it, he already had a glimmering of what he would do.

And then he was busy again, making another mouse.


Chapter 2: A recipe for life


To make a mouse took many things: a trained mind, a state of the art laboratory, a steady hand. But more than anything it took faith — faith in things forever unseen, too small to be captured in the lens of a microscope, too insubstantial to be manipulated directly … things that had been deduced, induced and theorized into existence, then held in the mind by dint of model and metaphor.

The cancer gene Andrew Xiao wanted to transplant into his mouse was all those things. You couldn’t just “put” it into a cell. Billions would fit into a drop of water, and the water would still look like … water.

So the genetic engineer, to do surgery on a gene, had to use nature’s own tools. He had to map out a plan and follow it, adding one substance to another, incubating them, trusting and believing all the time that the invisible little machines in the solution were doing what he’d predicted they would do.

Mousers swapped genes back and forth, from laboratory to laboratory. For convenience of handling, they spliced them into plasmids — those tiny hoops of DNA that lived in bacteria. Sometimes Xiao received genes suspended in a drop of liquid in one of the test tubes his profession favored: little vials less than an inch long, equipped with snap-lock caps to protect the precious snippets of code.

But just as often, he’d get his genes in a regular envelope. Inside would be a piece of paper, with a note scribbled on it, and somewhere on the paper would be a little circle. That would be where the other scientist had placed a drop of water with plasmids in it. The plasmids were hardy buggers; they’d be there, all dried out. Xiao would just have to add liquid and they’d be ready to go.

Life was like that. Durable.

Programming a mouse-to-be, like programming a computer, involved many steps, most of which were relatively simple for someone who had the training … and the faith. To make a new mouse, Xiao began with a vial of plasmids — circles of code, in computer talk. Plasmids were all small. A bacterium might have a million base pairs of code, a virus 100,000, but plasmids in their natural state ranged from 5,000 down to 2,000. Those favored by Xiao and his colleagues were toward the small end. The cancer gene was about 500.

Xiao’s first job was to insert the address code — the so-called promoter, that would allow the cancer gene to be read only in the brain. The address had to go immediately in front of the cancer gene, so the hoops had to be cut in exactly the right place. To do that, Xiao added a drop containing a “restriction enzyme,” a scissor protein that could grab onto the plasmid hoop and read around it until it found a particular bit of code, and then cut the hoop there. In this case the scissor enzyme was designed to cut at the beginning of the cancer gene.

The broken loops were mixed with a drop containing the short bursts of DNA address code — the promoter. Yet another drop contained a protein whose specialty was rejoining broken DNA strands. These proteins floated through the vial, grabbing raw ends of DNA and fusing them. They did not discriminate. Sometimes, in the process, they simply repaired the plasmids. Other times, they stuck the end of a plasmid strand onto the end of a promoter strand; when those particular plasmid loops were put back together, the promoters were there, right in front of the cancer genes — precisely where Xiao wanted them.




Xiao’s cancer gene was a fragment of DNA found in a monkey virus long before Xiao was born. The virus was Simian Virus 40, or “SV 40” for short, and how it came to the laboratory inChapel Hillwas a classic tale of molecular genetics.

It began in the 1950s, as scientists were frantically working to make a polio vaccine. The polio virus had been identified, but no one could figure out how to grow it — and without quantities of virus, no vaccine can be devised. The breakthrough came when two scientists atJohnsHopkinsUniversityfound they could grow the virus by incubating it in the kidney cells of the green monkey. The viruses, and fragments of the cells they grew in, eventually went into the manufacture of early polio vaccines.

It was discovered later, however, that those monkey cells had harbored a second virus — which would eventually be named SV 40. Those viruses had ended up in the vaccines given to humans. Worse, subsequent research showed SV 40 could cause cancer.        Whether it actually posed a risk to the recipients of the vaccine would remain controversial for the rest of the century, but as cancer replaced polio as the most dreaded disease, curiosity about cancer genes soared. They were called “oncogenes,” and the one in the SV 40 virus became an early target of research.

Eventually the virus’ DNA was sequenced and the cancer gene identified. How it caused cancer took years to tease out, but some of the final work was done by aPrincetonteam that included Terry Van Dyke, who now ran the laboratory where Xiao was trying to build his mouse.

Back then, Van Dyke, who was herself learning to be a mousemaker, was a junior member of a project focused on a rare human malignancy called retinoblastoma. Retinoblastoma, usually shortened to “RB,” was a strange and deadly tumor that erupted from the retinas of babies and very young children.

The victims turned out to be missing a particular gene, which was quickly dubbed the “RB gene.” Scientists later discovered that it was a tumor-suppressing “suicide gene.” The gene was a safety device nature had built into the genome, and what it said, in genetic code, was, “If this cell begins dividing rapidly, make a protein that will kill it.”

The discovery of the RB gene was another confirmation of what many scientists had thought for decades. The body periodically produced cancer cells, but those cells were normally killed off by natural anti-cancer mechanisms. The RB gene was an anti-cancer mechanism.

At this point, the story of the RB gene collided with the story of the monkey virus cancer gene. The cancer gene, it turned out, coded for a protein that went looking for the RB gene and, when it found it, latched onto it and muffled it. In knocking out one of the body’s key cancer-controlling mechanisms, it set the stage for cancer. The muffled suicide gene could no longer tell a runaway cell to die.

With each bit of knowledge, the stakes mounted. A mouse that lacked the RB gene, by predictably getting a kind of cancer that humans also got, would allow scientists to conduct all manner of meaningful experiments. It was easy to fantasize that such experiments might lead fairly directly to prevention or cure.

But as often happens, what seemed straightforward in theory turned out to be much more complicated in the laboratory. It turned out that disabling the RB gene was devastating to the mouse. The fetuses aborted, were stillborn, or died shortly after birth from cancers arising all over their bodies.

By Xiao’s time, Van Dyke’s generation of scientists were in charge of the laboratories, and the science of molecular genetics had grown steadily more sophisticated. The genetic manipulations were more delicate, and more transgenic animals survived. Procedures that had once been impossible were now routine, and a new crop of scientists — Xiao’s generation — were learning to direct genetic expression in increasingly specific ways.

Xiao’s mouse recipe, for example, called for the expression of the cancer virus gene only in the specialized cells that made up the support structures of the brain. These cells were called glial cells, and so Xiao wanted to attach a promoter that said, “If you are a glial cell.” Then, in theory at least, the mouse would be normal except for its susceptibility to glioblastoma.

He assumed that the reality would probably be more complicated than that, but … one never knew. There was always the hope.

In any event it was becoming increasingly clear that the RB gene, and others like it, played a key role in cancer development. They needed to be studied. If they could be adequately understood, that understanding would surely dovetail with research on other fronts. Someday, maybe someday soon, it could all be put together. If it could be put together for glioblastoma, then it could also be put together for breast cancer, colon cancer, prostate cancer, lung cancer and the other big killers.

So the trail was hot, and Xiao was on it.




The next step was to introduce the plasmids into bacteria cells so that they could be multiplied.

The word Xiao used was not “multiply,” but “amplify.” His generation of genetic engineers, who spent their lives manipulating biological code, had begun to pick up jargon from electrical engineers who worked with computer code. The two codes were not exactly the same, but they had similarities … and the similarities seemed to grow more glaring by the day. At some level, code was code was code.

The task was to get the plasmid code into a bacteria so it could copy itself. Normally, plasmids could not penetrate the cell membranes of bacteria. Cell membranes were strong, but they were made of lipids, a class of fats … and a fat, by another word, was grease. Detergents cut grease.

Xiao went to the nearby incubation room and fetched a batch of E. coli, the ubiquitous gut bacteria that served as the microscopic equivalent of the laboratory mouse. This particular strain of E. coli was somewhat special because it was not resistant to antibiotics. Further into the recipe, this would make a critical difference.

Xiao held the test tube containing the bacteria up to the light and carefully, ever so carefully, added a bit of detergent.

It was a delicate maneuver. Too much detergent, and the cells would rupture and die. Too little, and nothing would happen. Hunched over his laboratory bench, eyes focused on the culture, Xiao shot for the exact medium: Enough detergent to make holes in the membranes, holes big enough to allow the plasmids to float into the cell.

Then he introduced the plasmids to the culture and waited. The law of averages was on his side: Some of the plasmids were bound to find their way into a bacterium. And some of those plasmids would have both the address code and the gene in them.

Before the next step, Xiao gave the cells a few minutes to repair themselves. Mousemaking was all carefully choreographed and timed. Life lived at its own rate, waited for no scientist, was ready when it was ready and not before.

Xiao concentrated totally on his work. On the desk a few feet away from his lab bench, his notebook lay open to the recipe.

Occasionally he stepped over to the desk and paused thoughtfully, considering the next steps.

Xiao was constantly reading books and scientific papers, consulting with other scientists, conferring with the head of the laboratory. It was important to know not only what his own immediate colleagues and competitors were doing but also the state of allied disciplines, especially those involving genetic control of growth, development and aging. He needed to understand the daily problems they faced because some of those problems might be solutions to him.

Take, for example, the plasmid. Plasmids were nature’s gift to the genetic engineer, but to the rest of the biomedical world they were mostly pests. They first came to scientific attention as carriers of genes that conferred antibiotic resistance. It was via plasmids, scientists believed, that resistance to drugs like penicillin was transferred from bacterium to bacterium. Indirectly they were a threat, if not to human existence, then to human health and civilization. On the other hand, Xiao could hardly function without them.

Over the years, the plasmids themselves had been engineered. The ones Xiao was using at the moment came not only with the cancer gene but also a second gene that coded for resistance to ampicillin.

Xiao looked at the soup of bacteria he was working with. He had started with a uniform strain of bacteria, but now there would be three. Most of the bacteria cells would have been unaffected by the earlier procedures, and would be unchanged. Others, a small minority, would have plasmids in them. Of those, some would have plasmids with only the cancer genes in them. Finally, some tiny fraction would have plasmids with both the cancer gene and the promoter, or address code, that would allow the gene to be read only in the glial cells of the brain. Those were the ones he wanted.




The first step would be to get rid of the cells that had no plasmids. That would be easy … and was why Xiao had originally chosen a line of bacteria susceptible to ampicillin. All the plasmids carried the ampicillin-resistant gene. So the cells without plasmids in them would be sitting ducks for the antibiotic.

With his left hand Xiao picked up a petri dish filled with a bluish nutrient gel that contained ampicillin. With a deft motion of his wrist he smeared his cell samples across the surface of the gel. In minutes the bacteria without plasmids in them would die. But those with plasmids in them would have the ampicillin-resistant gene, and they would prosper.

It was an ingenious trick — “cool,” in the language of Xiao’s generation.

The petri dish spent the night in the warm incubation room, where the plasmid-containing cells could grow and multiply. The surviving cells, being few, were scattered across the surface of the gel, and could divide without impediment. Where at first there was one cell, in 45 minutes there were two, and in an hour and a half there were four, then 16, 32, 64, 128 … the next morning, the surface of the gel was speckled with fuzzy white dots the size of pinheads.

It was a turning point in the experiment. Each dot was a thriving colony of bacteria, and each bacterium was chock full of plasmids, some with just the cancer gene in them and some with both the cancer gene and the promoter. Xiao would sort them out later, but to do that he needed more to work with than fuzzy dots.

He held the petri dish for a moment, gazing at the bacteria. He never questioned what the dots were, or if the ampicillin had done its work, or if the spliced-together cancer gene and promoter were really somewhere in the plasmids that were in the bacteria that were in the colonies in the dish. Faith. They were there; he believed it.




When he was younger, Xiao was often faintly surprised when one of his experiments actually worked. But now his faith in the unseeable was stronger and his expectations of himself much higher. He expected success. He even expected luck because luck … well, there was a saying in science: Fortune favored the prepared mind.

The prepared mind was the mind that studied, the mind that rehearsed, the mind that lay awake at night imagining how tiny unknown machines would react to this or that theoretical circumstance. Xiao was nothing if not prepared.

Now, delicately, he scooped out one of the fuzzy dots with a small spatula and dropped it into a test tube full of nutrients. Then, cleaning the spatula, he transferred another dot to another test tube, and then another.

He did everything slowly, carefully, his eyes on the minutiae of the moment, thinking through every step, keenly aware that if he made one mistake he might not know until he had wasted six weeks.

In this way, lab time became different from other time.

In the laboratory, the accumulation of seconds into minutes and minutes into hours often passed unnoticed. Outside the window near his desk, a bulldozer snorted and growled, excavating for yet another science building. In the hall outside the laboratory other scientists, equally preoccupied, carried small trays between the labs and incubation rooms and the rooms that housed the gene-amplifiers and analytical equipment. Over Xiao’s right shoulder a white mouse explored the limits of its plastic universe with a tiny paw and an inquisitive pink nose.

When Xiao finished distributing the colonies of E. coli, he sealed the test tubes and took them to the incubation room. There a purring machine would agitate them gently while heating coils warmed them, maintaining a precise 37 degrees Celsius.

The rest of the day passed quickly. He had things to read, supplies to order, his Ph.D. exams to study for. And then there were always mice to tend. Outside, the bulldozer labored on.

The felt life of a scientist was a ballet of the mind, full of effort, things to do, things to do, things to do. Over the years Xiao had developed the biotechnologist’s signature ability to focus on the work in front of him until it was done, and then shift his focus to a completely different task.

The rhythms of his life followed no clock but that provided by the experiments themselves. He was there when he needed to be, and if a reaction was scheduled to ripen in the wee hours, or if that’s when he could get access to a machine or instrument he wanted … well, then he was there. He had to be there, but it was a duty mixed with equal parts desire. Because work and life were as one. He wanted to be there, to find out what would happen.

Meanwhile an undergraduate, working at a nearby bench, had a question. Xiao listened thoughtfully, then helped solve the problem. He moved from bench to desk, and back again, studying his notebook, making entries, looking up the specifications for a reagent. Eventually the bulldozer outside fell silent, its operator gone home. The sun set. It was hours later when Xiao closed his notebook, cleaned off the top of his lab bench, and went back to the apartment he shared with two roommates.




It was a busy life, crammed full of responsibility and interest, leavened with curiosity, driven by ambiguous mixtures of ambition, obligation, competitiveness and a deep yearning to contribute. But even in all the busy hours, the human mind always found time to worry … and to dream.

The worry was obvious: What if he didn’t get his mouse?

Genetics was nature’s game of chance, and while hard work and depth of knowledge could shave the odds in favor of the prepared mind, odds were still, irreducibly … odds.

If the genetic dice rolled the wrong way, Xiao was pretty sure Van Dyke wouldn’t toss him out of the program. He’d have an opportunity to try, and try again, and ultimately he’d get his mouse and his Ph.D. But by then he might be behind the game, behind his generation, even just … that worst of all scientific fates … behind.

So he worried. And, worrying, he went over it again and again.

His earlier mice had failed, and they had failed often enough that more than simple luck might be operating. The promoter limited the cancer gene to expressing only in the glial cells of the brain, which should ensure that the transgenic mice were healthy, at least until their glial cells went berserk. Yet the knowledge of how genes were addressed was still young and full of ignorant guesses. Perhaps, even with the glial-cell promoter, the gene expressed somewhere else, in some unknown fashion. And perhaps that expression led to subtle conditions that, as scientists so delicately put it, were “incompatible with life.”

He might not be failing, in short, because he was unlucky. He might be failing because he was trying to do something impossible. In that case, he could spend his life trying and get no mouse.

So he and Van Dyke were mulling over other strategies. Some months ago they had come up with an idea. It was a very complicated idea, and it pushed the envelope. But it might work. It was a fallback.




Like most genetic engineering strategies, the fallback plan took advantage of a natural happenstance of metabolism. Bacterial cells had a critical gene that began and ended with a short but very distinctive burst of code. These bursts were, in effect, punctuation. In this case, they were biological parentheses.

A certain bacterial virus had capitalized on this by evolving a gene for an attack protein called Cre. Cre was a very smart robot. It chugged up and down the bacterial DNA until it found those parentheses. Then it clipped them both out … and everything in between.

Such a tool might be perfectly suited to Xiao’s dilemma. If the transgenes were doing something to kill his mice in the womb, or make them sterile if they chanced to survive, then it might be possible to scramble the code in the transgene in such a way that it could be unscrambled later.

The mouse recipe Xiao had been using had a header that said, “If you are a glial cell read the following message,” followed by, “Make a protein that will muffle the RB gene.”

In the scheme he was developing, he would build a new piece of code, composed of the two parenthetical bursts and a chunk of nonsense in between. Something that made no sense at all, like (dp#&dsw). Then he would put it between the header and the cancer gene.

So the modified gene would read, “If you are a glial cell,” (dp#&dsw) “make a protein that will muffle the RB gene.”

The cell could not decipher such a sequence, so the resulting mouse should gestate and be born without problems … and it should be able to reproduce. Then, when Xiao wanted a brain tumor to express, all he would have to do would be inject a gene for the Cre protein into the mouse’s brain.

The Cre protein would find the parentheses, and clip out the nonsense phrase. The monkey virus gene would then express in the mouse’s glial cells, making the protein that would muffle the RB suicide gene … and cancer, in a short while, should erupt.

The idea was a good example of just how sophisticated molecular biologists had become at using nature for their own purposes. Such a mouse would be a tour de force, cooler than cool — and Xiao, like most scientists, had a weakness for such feats.

On the other hand, it was oh, so very complicated. It pushed the technology to the limit. It would take time. Meanwhile, Xiao’s Ph.D. would be held hostage.

He had to hope that the old way would work. Just one more try. Maybe this time.




Under the conditions maintained in the incubation room, life was explosive. Yesterday there had been a few E. coli cells in each test tube. Now, when Xiao held the tube up to the light, a visible cloud of cells hung like a thick fog in the warm liquid.

In each one of the invisible cells that made up the fog, there would be a plasmid. In some test tubes it would be the cancer plasmid, the code that said “make a protein to muffle the RB gene.” In others it would say, “if you are a glial cell, make a protein to muffle the RB gene.” The latter was what he wanted, and he had enough quantity now to analyze, and determine which was which.

The analytical method he used was called “electrophoresis,” and back in the days before Xiao, which was to say the olden days, it had revolutionized biochemistry. Now, of course, it was a “traditional method.” Xiao didn’t think about it, he just used it — unless he was asked by a younger student, or an outsider.

Then he would explain that electrophoresis was based on two simple facts: First, DNA molecules were negatively charged. Second, some were heavier than others.

It was second nature to him. You made up a plate of gel, an unflavored version of what might be found on a dinner table. You attached a negative electrode to one end and a positive electrode to the other. You made a row of little wells in the negative end, and you put your samples in those. You added markers to make the results visible. You turned on the juice.

And then, before your eyes, the substances in the wells would begin moving slowly through the gel, toward the positive end. Since the different molecules all had characteristic weights, they moved at different speeds. They arranged themselves in lines, like little robot marching soldiers, each line representing a different type of molecule. It was something you could see with the naked eye. The weights and charges and movements had all been worked out by someone else and printed in a book. It was so straightforward that a moderately gifted middle school student could, and often did, use electrophoresis to sort out and purify proteins and genes.

Now Xiao took samples from each flask of bacteria and processed them to remove all the proteins, fats and various other materials until all he had left was DNA — plasmid DNA and E. coli DNA. A drop of DNA from every test tube went into its own well, and Xiao turned on the electricity.

As the molecules started moving through the gel, the huge strands of bacterial DNA hardly budged. But the lightweight plasmid DNA moved right out in front … and, within minutes, it was clear that the plasmids in some samples were moving faster than the plasmids in other samples.

That told Xiao which test tubes of bacteria he could discard: The fast samples. They were faster because they were lighter in weight, and they were lighter in weight because they didn’t have the promoter in them. The others did, and they were the ones he wanted.

To make a mouse, you had to believe.




Xiao seeded the remaining bacteria into flasks of nutrients and put them back in the incubation room to multiply. What remained was essentially a refining job, separating out the plasmids from the bacteria and the genetic sequence he wanted from the plasmids.

Digestive enzymes chopped up the cells and a few minutes in the centrifuge concentrated the DNA in the bottom of a test tube. Restriction enzymes read their way around the plasmids, finding and snipping out all genes that read, “If you are a glial cell, make a protein to muffle the RB gene.”

Finally Xiao was back to the electrophoresis apparatus, only this time with much more material. The gene he wanted, being the shortest in the mix, immediately moved out in front. When the separation was great enough Xiao turned off the electricity and, ever so carefully, cut out the line he wanted, gel and all.

The gel and marker were easily separated, leaving a few drops of clear liquid — liquid that had to have the precious gene in it. Xiao gave the drop to Hua Wu, a biotechnician with a steady hand and experience with the delicate surgical process of harvesting, manipulating and processing live mouse eggs.

The process was not much different from that practiced in human fertility clinics. Hua extracted the eggs she would use from the ovaries of a mouse, then took the genes Xiao had prepared and, using a microscope and a tiny syringe, injected a bit of the liquid into each mouse egg. Then the eggs went into the ovaduct of a receptive female mouse.

The rest was a bookmaker’s game. The genes would or would not enter the nuclei of the eggs; and, if they did, they would or would not incorporate themselves into the mouse DNA.

The odds were good because egg cells, being stem cells, were exquisitely receptive. But they were still odds, or luck, or fate, or statistics. The dice rolled, and Xiao’s future hung in the balance.

Meanwhile, his part was to care for the mother mouse, and to hope for the best. Nature, if “nature” was the proper term, would do the rest.


Chapter 3:  Of Mice and Men


Under normal circumstances, Andrew Xiao made rounds on his mice once a day. But at times like these, when a female was incubating what he hoped were transgenic fetuses, he went more often.

The UNC animal facility was located across the street from the building where he and his colleagues spent their days and sometimes their nights making mice. To gain admittance Xiao had to carry his laptop up several flights of steps, cross to another building, unlock the door and walk down a hallway. Another door got him into an anteroom, from which five more doors opened into mouse rooms. Their cages were visible through thick, wire-reinforced glass.

In the anteroom Xiao covered his clothes with sterile coveralls. Surgical booties went on his feet, a paper cap on his head, a mask over his nose and mouth. He wiggled his fingers to work them into the rubber gloves. Then he could proceed through the last door.

The mice lived in shoe-box size plastic containers stacked on two shiny, rubber-wheeled racks. Each mouse had a separate genetic history and a dossier. In terms of time, money, creative energy and medical hope, they were among the university’s most valuable possessions.

Xiao’s entry sent the mice into a frenzy. They squeaked in tiny, high-pitched voices, scurried around, stood on their hind legs and stuck their twitching noses through the grid that topped their cages. They had no fear of humans in white uniforms. Somewhere, far back in their ancient pedigrees, there was surely a field mouse or a house mouse that had lived in incipient terror of cats and owls. But that was a very long time ago, even in human years, and long forgotten by the mice.

After they fell under the evolutionary shadow of humans, they lost the brown coats that once helped them blend with the earth. Some turned white, some black; color didn’t matter in a universe without predators. Now their natural habitat was more properly a plastic world, with its laboratory-issue water bottle, manufactured bedding and the largess of a keeper, who brought them food pellets chock full of all the nutritional elements necessary to murine health and well-being. If these mice couldn’t have lasted an hour in nature, the fact was of no relevance. Few places on Earth were as far from nature as this room.

The cages were constantly being moved by their keepers and by other scientists, so Xiao had to search for his mouse. The dollies rolled silently, and with little effort. On the front of each plastic cage was a piece of tape with a code on it.

Xiao found the cage he wanted, removed it and took it to a stainless steel counter across the room. The top slid back.

The female mouse came to him eagerly. He picked her up in his gloved hand. Nine days after implantation, he could feel the swell of her uterus.

So far, so good. In there somewhere, warm and almost alive, there grew a tiny, semitransparent thing that nature never made: a transgenic fetus, a mouse with a short stretch of monkey virus gene in its cells. That was the future. At least, Xiao hoped it was.

In biology, nothing was ever certain. There were always unexpected and unforeseen things. The engineering of life was a strange new frontier, the designs untried, the fates fickle.

Xiao put the mouse back in the box and typed notes on the computer. There always had to be notes, log entries, documentation, every detail, every event, every act.

Several times each day the young scientist returned to the facility, trudged up the stairs, donned sterile clothing and checked his mouse. Watching and wishing made time slow down, but however slowly the days might pass, they passed, and with each sunrise the belly of the female mouse was larger than it had been the day before.

Finally, when 19 days had come and gone, Xiao peered down into the plastic box at a tiny, unruly pile of pink life, wriggling and squirming, eight tiny mouths gaping for a teat, each one possessed of its own stubborn will to live, to grow, to make yet more mice — a housekeeper’s nightmare. But to Xiao they were beautiful things, one of the most sophisticated products of the highest civilization the human race had ever achieved. And maybe, just maybe, one of them was his Ph.D.

To find out, he would need a tissue sample from each mouse. Newborns were too small to be picked up — and they didn’t have any tissue they could spare.

He waited, not patiently, as the days passed. After 10 days, they were big enough.

One at a time he lifted each pup delicately from its nest and snipped off a tip of its tail and one of its front toes.

The tails were simple to clip, but the toes took a good eye and a steady hand. A mouse could easily live without one toe, and the loss of a tail tip was inconsequential. But Xiao only wanted one toe from each, and it had to be a different toe for each individual. That’s how he would identify them later if one of them turned out to be the mouse he was after.

Each mouse’s tail and toe went into a tiny, prelabeled blue plastic vial. The vials had snap-on tops, which Xiao closed with a gloved finger. Each made an audible pop as it sealed.




A scientist’s job wasn’t to dream, or to gawk at the passing scenery. A scientist’s task was to get on with it, do the analysis, get the numbers, redefine the problem, move on to the next step. But even a molecular geneticist sometimes had to stop and wonder over the complexity of it all.

Life must have been a very simple thing, in the beginning — something that coalesced in a warm spring, or on some damp clay flat, or maybe in a tidal pool. No one knew what message that first syllable of genetic code had uttered, but it must have been exuberant beyond measure: It was the biological equivalent of the big bang.

Everything that lived and breathed on the Earth today was, in the inner fastness of its genome, an acidic echo of that first gene, amplifications of it, riffs on the amplifications, melodies grown from the riffs … chemical symphonies composed of long, twisted, invisible strings of code, complicated and convoluted, tough as wire, persistent as the turn of the planet, ubiquitous and various, living and making more of itself everywhere you looked, from glaciers to hot springs, Beijing to New York City to Chapel Hill … all of a family, from plasmids to mouse to genetic engineer.

The amplitude was astonishing. You could make a plasmid with only a few thousand base pairs, but it took 6 million to make a bacterium, and a hundred million to make a fruit fly. The blueprint for mammals, like Xiao and his mice, ran in the neighborhood of three gigabases — enough to fill a hard drive.

By comparison, the sequence of the cancer gene he was trying to transplant into his mouse was only a few hundred base pairs long. Once it was incorporated into the mouse genome, it was a needle in a haystack. Was it there, or wasn’t it?

A few decades ago, such a thing would have been unfindable. In 1999, a doctoral student could do it in less than a week. But that didn’t mean he could fathom the answer.

According to the tests, one of the baby mice was transgenic.


Xiao was puzzled — puzzled, and a little skeptical. The readings were unequivocal but, given his record of failure, he still couldn’t resist a wave of doubt. One mouse. Out of all the eggs injected with the cancer gene and implanted in the host mouse, out of eight pups brought to term, only one was born positive.

What happened to the other ones? Why was this so hard?

As he mulled it over, several thoughts came to mind. It could be that this particular gene was, for some obscure reason, difficult to transplant. Or maybe the gene expressed in some subtle, unknown way. Perhaps the mother’s metabolism somehow sensed the alien character of the implanted fetuses and rejected them. So the only mice that survived were the normal ones.

But, in that case, why did the one positive mouse survive?

Could it be a mosaic?

When a mouse egg was injected with a transgene, the hope was that the transplant would be immediately filed in the cell’s genome, or genetic library. That way, all the subsequent generations of that cell, as the egg became a fetus and the fetus became a mouse pup, would have a copy of the gene.

But if the gene didn’t immediately incorporate into the genome, but hung around, it might incorporate later into one of the daughter cells but not all of them. The result would be a mosaic, with the transgene cropping up in some tissues but not others. The result was a bizarre creature, partly one thing and partly another.

But it was not unusual. When scientists first learned to analyze human chromosomes they were shocked to find that a substantial number of people were sexual mosaics: They had an arm or leg, say, that was of the opposite sex. Functionally, it didn’t make any difference; as a practical matter their sex was determined by the genetic makeup of their gonads. But they were mosaics, nonetheless.

It was odd. Nature was sometimes so fastidious, sometimes so sloppy. Perhaps there was some deep philosophical meaning there, but to a young mousemaker, plugging away at his mouse while his Ph.D. receded into the distance, there was only one interpretation that mattered: Nature could once again give him his mouse … and then take it away. A mosaic mouse might test positive for the transgene, but she was unlikely to pass it on.

In that case, he would have to start all over, yet another time. That, or he would have to reconsider the whole recipe. He might have to switch recipes, and make the more complex mouse he and his boss, Terry Van Dyke, had discussed — the mouse with the garble in the transplanted code.

But it was not a pleasant line of thought. Producing a garbled gene was more complicated, and then later, when he wanted the cancer to grow, he’d have to be able to take the garble out. The more complicated an experiment was, the more things could go wrong.

Best not to leap ahead; better to assume the positive mouse really was positive, and proceed from there. Breed her, and find out. Maybe her pups would be positive.

It was fortunate that she was a female; females were sexually mature at four weeks. When the time came, Xiao introduced a mate. Nine days after that, he could feel the pups growing inside her uterus. Ten days later, she bore nine pups. Ten days after that, he took the tissue samples.

He did it at the end of a long shift at the bench — after midnight. The campus was dark as he crossed the street and entered the animal facility. A half-hour later, there were nine sets of tails and toes floating in individual vials.

Then, exhausted after another marathon day, Xiao slipped out of the room, took off his whites, snapped the gloves into the trash can, took the precious blue vials across the way, put them in the freezer and went home.


                Hope springs eternal: That was part of life, too, and so the following day began in anticipation. First thing, Xiao thawed the toes and tails and added a drop of digestive enzyme to each vial. Then the rack of vials went into a warm water bath, to encourage the little scissor proteins to do their demolition work. While they did, Xiao occupied himself elsewhere; then, several hours later, he took the rack of vials out of the water bath.

                The toes and tail tips had swelled up like tiny popcorn puffs. The white was mainly collagen; the mouse protein would have been chopped up into its component amino acids, which would be dissolved — and invisible — in the water.

                Xiao held the vials up to the light. Tiny hairs could be seen floating in the aqueous mix. Hair didn’t digest well.

                More to the point, neither did DNA. So, with the protein chopped up, the mouse DNA should now be floating freely.

                But for this day, Xiao had gone as far as he could go. He had checked the sign-up sheets for an instrument that was critical to his analysis, only to find that the schedules were full. He would have to wait until tomorrow.

                The instrument in question was called a “PCR,” for “polymerase chain reaction.” It was a squat little rectangle the size of a desktop printer, and about as imposing. But the PCR, perhaps more than any other item in the Van Dyke laboratory, had come to symbolize molecular genetics — its growth, its possibilities, and the remarkable cleverness of the men and women who dared to engineer life.

                The PCR was, essentially, a copy machine for genetic code.

                When presented with a DNA sample and the specifics of the code a scientist was looking for, the PCR would hunt through the DNA, find the target gene, and copy it over and over.

                At the heart of the process was an enzyme from an otherworldly bacterium, Thermus aquaticus, which lived only in hot springs. For the bacterium to reproduce, one of its polymerases, or reader proteins, had to read and copy DNA at temperatures too high for normal polymerases to function.

                From this fact a clever scientist, Kary Mullis, concocted the recipe that changed molecular biology. First you heated the DNA until it unwound and came apart. Then, before it could cool down, you arranged for the hot spring polymerase to copy the gene segment you wanted to amplify. Then the process was repeated, heating and cooling, heating and cooling, each time doubling the amount of the gene you were looking for. Finally the mixture was so rich you could separate your gene out with gel electrophoresis.

                The PCR machine quickly became a necessity for molecular geneticists working in widely disparate subdisciplines, from the study of cancer to the manufacture of artificial viruses to cure genetic disease. Mullis received the Nobel Prize in 1993.

                That was a glorious step forward for science, but most science was far from glorious. Most science was like most of life, muddling through, and too much muddling could wear a man down. Now, in Chapel Hill, six years after Mullis got his Nobel, Xiao would have happily settled for a mouse.

                Perhaps, he told himself, this time he had it. But he wouldn’t know until he could get access to the PCR, and the first opening in the schedule was tomorrow. He would have to wait.

                Meanwhile, the nine vials went back in the freezer.


                The next day Xiao was back at his bench, bending over his toe and tail samples. He thawed them with the warmth of his hands and breath.

                It was a normal day in the bustling Van Dyke laboratory. Ten feet down the bench from where Xiao sat, another lab worker used a pipette to transfer liquid from a bottle to a similar tray of vials. From the far end of the laboratory, a half dozen rows of benches away, came the sound of fingers hammering on a keyboard. One of Xiao’s fellow students was in the final throes of putting together his dissertation. Xiao’s own dissertation, of course, was still only a fantasy.

                It had all seemed straightforward, in the beginning. He would make his mouse. Then he would do the defining experiments on it, showing that the transplanted cancer gene had bestowed upon its progeny a genetic predisposition for brain cancer. Then he would write and defend his dissertation. Then he would go on to a postdoctoral fellowship somewhere. Then he would …

                But now he was exactly where he had been a year earlier, working on his mouse.

                Meanwhile, the critical thing was to make no mistakes.

                When his samples thawed he popped open the tops of the vials and stared into them, thinking. Then he opened a bottle of buffer and, taking up a pipette in his right hand, carefully added a drop of buffer to each one, adjusting the acidity. He closed each vial with a snap, picked up the tray, slipped off his stool and headed toward the centrifuge at the other end of the lab.

                It was all routine, repetitive, each step simple enough, no magic to it at all. Xiao inserted the vials and turned on the centrifuge. The machine spun up with a vibrant hum, then took several minutes to coast to its long slow stop. Xiao removed the tubes and held them up to the light. A tiny white pellet could be seen at the bottom of each one: mouse DNA.

                Somewhere, in one of those vials, there might be a tiny piece of cancer-causing DNA from the genome of a virus that once, long ago, lived in the kidney cell of a green monkey. At least, one could hope.

                Then Xiao was back at his bench. Carefully, he poured the excess liquid out of each tube — and, with it, many of the chopped-up pieces of mouse cell protein and lipids. He added an alcohol wash, and carried the vials back to the centrifuge. Again the centrifuge spun up, whirred for a few minutes, then coasted to a stop.

                Careful step followed careful step. Dissolve and precipitate, drain off the supernatant, agitate and centrifuge, until what was left was nothing but code.

                As the morning wore on, Xiao’s bench became cluttered with the paraphernalia of mousemaking … bottles of solution, pipette tip containers, small instruments of various sorts. He didn’t like to talk while he was doing such fussy work because conversation sometimes made him forget where he was in the recipe. Then he had to throw it out and start over.

                If there were nine vials of toes and tails to test, Xiao needed a total of 22 vials. In addition to the nine important ones, there had to be nine controls. In those, the PCR would search for a common mouse gene that Xiao knew had to be there; if they weren’t positive, he would know that somewhere along the way, he had thrown out the baby with the bath water. The last four vials were controls for the PCR itself so that Xiao would know for sure it was working properly.

                When the PCR was finished, Xiao would drop samples from all 22 vials into a row of wells on a sheet of gel. Since such small bits of code would be transparent, he would add a tracer that would fluoresce under ultraviolet light. A positive result would show up later on the electrophoresis gel as a curved line above one of the wells. It looked like nothing so much as an eyebrow.

                All nine controls should have eyebrows. Of the four PCR controls, two should have eyebrows and two would not. And of the nine test samples … if one or more of those had eyebrows …

                Xiao focused on his work, using the pipette to add a drop of this, a drop of that.

                Each vial got a full drop of nucleotides, the raw material for DNA: adenylic acid, thymidylic acid, gguanylic acid, cytidylic acid … A, T, G, C, the alphabet of life. The PCR would need them to make copies of the cancer gene. Assuming, of course, it was there.

                Someone had turned on a radio and tuned it to a public broadcasting station. The music of Bach drifted through the laboratory. Xiao worked very slowly, methodically. If he made a mistake, it was all for nothing. But since everything was invisible, he wouldn’t know that until the very end, when there were no eyebrows, or they were in the wrong places.

                Worse, when he made a mistake he often had no way of knowing what the mistake was, or how many there had been. He might not even know that he had, in fact, done something wrong because when you were making mice there was always that wild card — statistics, chance, luck. But it didn’t matter: You got no cigar, and no Ph.D.

                Finally, Xiao sat up straight, stretched, and briefly contemplated the 22 vials in front of him. Then he snapped them shut, picked up the tray, slid off his stool, and headed toward the PCR.

                The equipment room was jammed with exotic-looking apparatuses. The laboratory’s two PCRs occupied a central position, sitting side by side on a table. One was already humming away, the LCD readout on its face flashing numbers as the temperatures rose and fell and the cycles clicked over.

                Xiao crossed the room to the other machine, lifted its face, and inserted the vials into recesses made for them. Then he closed the lid with a click, and the sequence began. It was that simple.

                It would take the PCR three and a half hours to do its work.


                Xiao walked slowly back to his bench. Laboratory work bred patience and, in any event, there was always something to do.

                As he cleaned up his bench, a coworker, who was also from China, brought him a petri dish with E. coli grown out on it — genes destined for a future mouse. The two examined the dish and exchanged a quick burst of Chinese. The woman flashed a satisfied smile, laid the dish on Xiao’s bench and left.

                As he worked, Xiao mused about the present moment in biology.

                “I’m fascinated by what I’m doing, of course, and right now it’s all new. But it won’t be long until it’ll be old-fashioned, too. That’s the way biology is.

                “Think of the people in the former generation, or the former two generations of biologists. When they figured out what DNA looked like, for example, or discovered restriction enzymes … at every step there were some people who thought, then, ‘OK, that’s it, now we’ve figured out how biology works.’

                “Yet now, 20 years later, we definitely haven’t figured it out. We still know just a very little bit of what biology really is. So I’m excited about what I’m doing now, but I know my research is going to be old-fashioned someday. Someday. Not now, though. That’s the process. You have to try a lot of things for the next generation, for the next generation after that. Even if this thing I’m doing doesn’t work, those people can know. So they will try something else. I’m just in the middle of a process. Yes, it’s a revolution. But it’s a long one.”

                The hours passed slowly. Xiao sat at his desk with his lab notebook in front of him, almost in a trance, occasionally he stirred to make an entry with small, neat strokes of his pen. A laboratory notebook was a legal document as well as a personal one, and Xiao’s was filled with his textbook English. On many of the pages he’d taped square photographs with lines of fuzzy, elongated dots running across the bottoms. Earlier PCR results.

                Failures, of course. All of them.

                Now a fresh plate of gel waited on a nearby counter for the samples to come out of the PCR. In this case, there would be 22 dots, but the nine that counted would be on the left.

                Would there be eyebrows over any of them?

                That’s what it came down to, in the end. If you were a mouser, the most important phenomena in your life were smudges in gel — smudges, or the lack thereof. That’s what made or broke a career. More abstractly, of course, the same smudges might be the difference between life and death for human beings, cancer patients Xiao had never heard of, and who were perhaps still unborn.

                But smudge or no smudge, good work or bad luck, victorious or heartbreaking, the photographs always went into Xiao’s notebook. And so they were all there, Xiao’s past mice, all the negative results, photographs visibly thickening the notebook, a permanent testament to the perseverance of a careful young man who wanted so very much to become a master mousemaker.

                In the room down the hall, the PCR went through its mindless cycles, one after the other, heating the mouse genes, copying them, cooling them off to recombine them, then repeating the process.

                To make a mouse was to make many mice, and whatever the expertise of the mouser many were bound to be failures. But the mouser never knew, and so each experiment carried its own thrill. There was always the possibility, the hope, that this mouse would be the mouse, and this day the day.

                When he was finished with his notebook Xiao occupied himself with housekeeping chores. Thoughtfully, he made a “to do” list. The pad of paper he wrote on was supplied, gratis, by Perkin-Elmer, the company that made the PCR. Large printing on the side of the paper touted the corporation’s “applied biosystems.”

                Of all the things it took to make a mouse, one of them was of course money. That meant commerce, and profits. Companies like Perkin-Elmer were growing exponentially, feeding on theories, possibilities, news stories of breakthroughs, the desperation of the sick and dying, the dreams of people like Xiao and his boss, Van Dyke.

                Then, finally, it was time.

                In the PCR room, the cycles ceased. Xiao popped open the lid and retrieved the samples. He carried them carefully down the hallway, to where the gel was waiting. Part of a vial went into each well. He flipped the switch on the transformer, and it began to hum.

                Xiao bent over and peered at the gel plate. He couldn’t tell much in regular light, but he looked anyway. He looked at everything carefully because a scientist did that. Otherwise, one would never see what one did not expect.

                In a few minutes he turned off the current, disconnected the wires, and carried the plastic plate of gel through the hallway and into the corner room where he inserted it into a special metal box, where the gel would be illuminated by the black light and scanned by a UV camera.


                The moment of truth was not complex. There was a click. Xiao stared at the screen of an attached computer, then hit the “print” button. The machine purred and exuded a small square of black photographic paper, a line of white dots visible across its bottom edge.

                Xiao looked at it without touching it. The 22 samples were all there, each represented by an elongated white dot at the bottom of the photograph. The positive controls were all positive, each with its eyebrow, and the negative controls were all negative. So the PCR was in proper order.

                But over the nine dots that mattered, the dots that might have held Xiao’s future, there was no second line of migrating DNA, no eyebrow, no nothing.

                It was very quiet in the building. Somehow, while Xiao was working, night had fallen and his coworkers had gone home. Somewhere, a telephone rang.

                Xiao stared at the offending photograph with hard-focused black eyes, as if by sheer force of will and intelligence he could somehow change the image.

                Finally, he drew a deep breath and grimaced.

                “It didn’t work,” he said, his voice laden with disgust. “I made a mistake, somewhere.”

                In the background, the telephone rang and rang and rang. Xiao stared at the dots. He would just have to try again. And maybe again after that.

                Then the mousemaker turned and walked out of the room and down the hall. He walked with shoulders slumped, head bowed, eyes on the floor. Halfway down the darkened hallway, his hands found their way into his pockets.

Chapter 4:  To move a mountain


Western culture had a myth about eternal failure. A Corinthian king, Sisyphus, had offended the gods and so they sentenced him to spend eternity pushing a boulder up a mountain. But every time he neared the top, the boulder rolled down and he had to begin again.

Andrew Xiao certainly had his Sisyphean moments, but the apprentice mousemaker’s disposition favored a different legend, this one from his own Chinese culture. It was about a man named YuGong, who had a mountain in front of his house. The mountain was inconvenient, so YuGong set about with a shovel to move it. His neighbors laughed at him, but he was sanguine. “If I can’t finish it,” YuGong said, “my sons will finish it. If my sons can’t finish it, my grandsons can finish it. If my grandsons ….”

So Xiao was not given to despair, and he had come too far, worked too hard, to quit. If he had failed to create a line of mice, he had learned at least one thing that before had been only supposition: The gene did, in fact, cause cancer. The sterile mouse had gotten it. If only Xiao could somehow pull off the whole trick.

He and Terry Van Dyke, the head of the laboratory, discussed the problem at length. She understood that the failures were not Xiao’s fault; they went beyond bad luck, or a beginner’s awkward laboratory practices — and, in any event, Xiao was no awkward beginner. He had gone through the process of making his mouse so many times he was an expert at it.

In any event, enough failures added together became informative. Most likely, the mouse was simply unmakable.

The probable cause of failure was that the promoter was allowing the cancer gene to be read in some population of cells in addition to the glial cells, and that it did something in those cells to kill the transgenic fetuses. In that case, he would have to resort to a deeper level of wizardry. He would have to definitively silence the gene, and establish a line of mice that carried it. Then he would have to find a way to selectively unleash the gene.

Xiao had hoped to keep it simple, but it was nature who ruled. So he would move on to plan B: the mouse with the garbled cancer gene.

The gene he had been transplanting into his mice instructed the cells to make a protein that muffled, or disabled, the “suicide gene” that killed any cell that turned cancerous. It also had a promoter, which made it readable only in the brain cells. In effect, the gene had said, “If you are a glial cell, make a protein to muffle the suicide gene.”

Xiao’s new mouse would have the same promoter and gene. Only this time, he would insert garble between the promoter, or “if you are” phrase, and the cancer gene, the “muffle the suicide gene” phrase.

The garble itself would be a nonsense syllable, something no polymerase could read. Putting it in would involve an additional set of steps, but the ingenious part of the experiment would be cutting the nonsense syllable out.

They would use a virus gene that some other laboratory had discovered in a phage, a type of virus that infected bacteria. The phage gene coded for a robot protein called Cre that homed in on its host’s DNA and read down until it found a bit of code called “LoxP.”

LoxP sequences came in pairs and functioned like parentheses. When the Cre protein found one parenthesis it grabbed onto it, marking the place, and then read ahead until it located the second. Then the protein clipped out both parentheses and everything in between. It finished the job by fusing the remaining DNA back together.

What had begun as an evolutionary tool for the phage had become an editing tool for genetic engineers.

So the new recipe would call for the combination of a five-part code. First would come the promoter, the “if you are a glial cell” sequence. After that would come the first parenthesis. Then would come the nonsense code, designed to make the whole construction unreadable. The second parenthesis would be grafted on after the nonsense code and, finally, the cancer gene itself would go on the end.

The gene should be totally silent until the nonsense buffer was removed. Xiao would do that by injecting the code for the Cre protein into the brain of the adult mouse.

The Cre would then do what it had evolved to do. It would find the two parentheses and clip them out, along with the nonsense buffer. The result would be the juxtaposition of the promoter and the cancer gene, which should then be readable, and should produce glioblastoma at the site of the injection.

Neither Xiao nor Van Dyke had ever heard of anyone using a recipe anything like that, but it should work.

Xiao couldn’t help but be fascinated by the plan. If successful, it would be quite a performance. Such a mouse, even in theory, was enough to make a scientist appreciate how far the field of genetic engineering had come.

In practice, though, he was still starting over. There was no denying that. Yet again, his Ph.D. was receding farther and farther into the future.

But there were no other options. Resolutely, he set his mind on the mouse with the scrambled gene.




To make a mouse, you had to plan it out, exactly what you would do when, and how. Whatever else might be said, Xiao had gotten quite good at such plans. He would work it out.

But Robert Burns, some 200 moldering years earlier, had made a cogent observation about the best-laid plans of mice and men, and what was true in his lifetime was still true on the frontiers of genetics in 1999. Before Xiao could compose his new recipe, he was interrupted by events several hundred miles to the north.

The Van Dyke laboratory worked in close collaboration with a team of brain cancer specialists headed by Dr. David Louis, a Harvard scientist who worked atMassachusetts GeneralHospital. The Louis team had isolated a stretch of the human genome that often mutated in the tumor cells of glioblastoma patients. Now researchers needed to know what the genes had originally said.

The stretch of DNA at issue was several hundred thousand base pairs long, so there were a number of genes involved; it wasn’t clear how many. There could be 10 or more in a sequence of that length. And even if they knew what the genes were, they wouldn’t have a clue what they did under normal circumstances … or failed to do in a tumor cell.

Such questions could have been answered by isolating and sequencing the genes, but with so many base pairs that could take a very long time. Years.

But with a mouse …

It was a mousemaker’s kind of problem. With a mouse, that whole section of DNA could be knocked out. If the scientists inBostonwere right, the mice should get cancer.

At that point a good mousemaker could start adding the genes back, in bits and pieces, until they made a mouse that no longer got cancer. Then they would know exactly what sequence was responsible.

It was a great opportunity for the lab, but it would have to be done quickly because other laboratories were working on it too. What Van Dyke needed now was an experienced mousemaker, someone who was fast and accurate and stubborn as dirt.

Andrew Xiao.

And so the decision was made. It didn’t matter that he had failed to make the mouse; there would be plenty of other mice. And while it was easy to forget in the daily frustration of laboratory work, the object of the exercise had never really been to make a mouse. The object had been to make a scientist.




Life was like that: complicated and dramatic, full of blind alleys and unexpected turns, moments of confusion and moments of insight — totally unpredictable. Whether you were man or mouse, existence was a drama. When the drama got complicated and circumstances overwhelmed you, it sometimes seemed fateful, even magical.

It had always been that way. Biology had seemed magical, until it became possible to look deeply into it. Then it became clear that muscle and blood were really amalgams of busy little robots doing the bidding of their master molecule, DNA. Instead of magic, there was molecular chemistry.

The positive side of that was that molecular chemistry was understandable in a way that magic was not. If you paid your dues, you could learn how it worked. Then you could rewrite the code, which changed the little robots, which of course changed the mouse. One figment of molecular biology could redesign another. Humans could make mice.

It wasn’t godlike, though. There was too much complexity. You went along, making the biochemical equivalent of magic, thinking you were in control of it, and then all of a sudden it didn’t work anymore. You didn’t know why. That’s when you realized that no matter how much you understood, you didn’t know the half of it.

What life was, mostly, was persistent. Science was like that, as well. If nature didn’t want to follow your plans, you changed your plans. You persisted. That, in the end, transcending all the confusion, error, heartbreak and failure, was what life was about. You persisted and, like YuGong, set about moving the mountain.

As for Xiao’s mouse, and his dissertation, that could be finessed. Van Dyke expected each Ph.D. candidate to make a mouse. But she could make an exception. In other laboratories Ph.D. candidates didn’t actually make their own mice. A high-level technician made the mice. The apprentice mouser designed the mouse, supervised, and then wrote it up for the scientific press. That was plenty for a dissertation.

So Xiao gave the new recipe he had constructed to Hua Wu, the technician, while he himself shifted his focus to theBostonmouse.

Xiao looked in on Wu whenever he could, but she was the one who followed the recipe, chopping up the plasmids and folding in the genes, adding the genetic parentheses and the garbled code … combining, separating, incubating, testing, purifying, centrifuging, testing. Eventually the fruit of Xiao’s recipe and her work was summed up and condensed out into minuscule droplets of clear liquid. Those went into mouse eggs, and the mouse eggs went into a female mouse with a ready womb. Then, in the fullness of time, Wu presented Xiao with a litter of black mice, several of which tested positive for the new, buffered transgene.

Xiao studied them carefully, looking for signs of abnormality, and found none. They seemed perfectly healthy. The test was to see whether they bred true.

When he mated them, pregnancy ensued in normal, almost instantaneous, mouse fashion.

Nineteen days after the first mating, two females delivered a squirming, healthy pile of 19 pups. They would tell the tale. If one had the transgene, Xiao had his mouse.

As busy as he was with theBostonproject, Xiao kept close tabs on the growing litter. As newborns they had been pink, but as the days passed the wriggling ball of baby mice sprouted black hair. Finally, at 10 days, Xiao visited the mouse facility and left with 19 vials of toes and tails.

The first day he could get access to a PCR after that was a Thursday. In the morning of that day he digested, centrifuged and purified the DNA from the tails and toes, and by noon the DNA was in the PCR and the readouts were flashing.



There was time, this day, for lunch at a nearby restaurant. Xiao lingered over his food, thinking broadly about his chosen field and its academic and commercial future.

There were jobs in private industry, and in time there would be more, but Xiao wasn’t after money. He wasn’t against it, and it would be pleasant to have the things that money could buy, but his Chinese heart didn’t beat to a capitalist rhythm. He was set on a life of discovery.

That meant a professorship somewhere. First, of course, he had to do a postdoc in some good lab, and to get a postdoc he obviously first had to get a Ph.D. And to get a Ph.D. …

Things changed, but they also stayed the same. He no longer had to actually make the mouse … but he still had to have one.

It was a balmy, pleasant day, the kind of beautifully magicNorth Carolinaday that a young man fromNorthern Chinacould appreciate. It was a good day to walk back to the laboratory.

The sidewalk ran in front of the university’s new computer building, a subject that turned the mind to the dynamics of technological revolutions. A generation ago, or perhaps a little longer, computer scientists had been much like mousers and other biotechnologists were now, working in a field full of promise but not yet really in the money. Then the revolution hit the marketplace, and new buildings grew on campuses.

Now “CPU” and “gigabytes” were words that came easily to almost every American tongue. Could “gigabases” be far behind?

Xiao was cautious about his own future. Basic research lagged far behind the money, and Xiao’s generation had had the misfortune to be born into an age of tight academic funding. For years, in fact, the job market had been terrible, though he thought things seemed to be opening up somewhat now.

Still, he worried, one had to be very, very good to get a permanent research job. Even being good wouldn’t be enough. You also had to be working on something hot, something that senior people could see would spur excitement. Excitement translated into grants. If you couldn’t bring in money, you couldn’t go far.

In Xiao’s business, the thing that triggered excitement was an interesting mouse. A line of interesting mice.




When he arrived back at the building the lights were still flashing on the PCR. There was still time to make rounds on the mice. So he packed up his computer, climbed the stairs, walked across the street and suited up.

The mice whose DNA was being analyzed had grown to the size of grapes. They were compacted together in a writhing pile of baby mouse flesh, the mice seeming no worse off for the loss of a toe and tail tip. The mother mouse stood on her hind legs and looked at the scientist through the plastic box.

Xiao scattered the pile and examined the pups, one by one. Each set of observations had to be typed into the computer. Ten minutes passed, 20. Finally, it was time.

Back in the lab building, the PCR had finished its work. Xiao snapped open the top, removed the tray of vials, and carried them briskly back to his corner of the laboratory. He set them beside the tray of gel and opened them. First he added the liquid that would make the DNA fluoresce under the black light. Then, changing the tip of his pipette with every use, he carefully transferred several drops of genetic material from each vial into a well in the gel.

The samples being tested for the cancer gene went on the left side, the controls on the right. Xiao touched the electrical switch and the transformer hummed. Invisibly, the molecules heeded the beckoning of the electrons, and moved through the gel.

A few minutes later Xiao carried the gel into the room where the ultraviolet box was kept. The gel went inside the box; Xiao closed the door and peered intently at the monitor on the attached computer. The sample blobs, visible now, were hazy, grainy smears on the screen.

Xiao said nothing, quickly making technical adjustments to the instruments. Then he touched the print button. The photographic machine hummed, and the black-and-white photograph emerged slowly out of its slot. It tried to curl down over itself, but Xiao’s right thumb held it against the body of the machine, so he could see clearly.

The controls were all positive, as they always were, testifying to the accuracy and completeness of the test. The 19 bright dots that represented the mouse pups ran in a clean line across the left side of the print.

A number of them had distinct eyebrows above them.

Xiao stared at the photographic readout for a long heartbeat of a moment, counting. Seven, eight … nine eyebrows.

The magic of life was its persistence. It always seemed to find a way. It worked on a scale too fine to be seen by the human eye, but it worked incessantly. Invisible things happened, or did not happen, that resulted in the long clingy threads of DNA, the hoops of plasmids, the enzymes that cut and the enzymes that sewed. It was a dance of valences, of atomic combinations choosing and then changing partners, of nucleotides spinning off ribonucleic streamers, and the streamers making proteins, and the proteins making the mouse. And, of course, the man.

Even joy was biochemistry, or so it was said — though in the sober, controlling world of science that kind of reaction was seldom given free reign.

Andrew Xiao, gazing at his results, allowed himself the tiniest of smiles.

“I got it.”