This is supposed to be a delicacy.”

Matt Harding pinched a thymus filet, his hands encased like a surgeon’s in blood-smeared latex gloves. “Sweetbread. I’ve never had it, but I think I’ve seen too much of it to give it consideration.” He cut the filet into thin strips with a pair of surgical scissors.

Further down the lab bench John Thomson snipped aortas, looking like clam necks, from the pale meat. He plunked the slices of thymus into a double-walled steel cylinder roiling with liquid nitrogen. The pieces bobbed up dappled and lustrous.

“Looks like crabmeat, doesn’t it,” he said. “We’ll track down some lemon for you.”

Not a cattle truck but a thirtyish man in running shoes now came to Vertex each Thursday, fast-walking through the lunchroom toward the biochemistry lab with two sallow-looking bags of meat, an entrepreneur from a suburban abattoir doing a steady side business in research supply. Two months after the Barcelona conference and the opening of the labs, Thomson, Harding, and a lanky, rail-thin assistant named Matt Fitzgibbon spent the rest of the day and part of Friday cutting meat, the first stage of a protein “prep.” The setting, the job, the ethos evoked a chop-house kitchen.

“If you dipped your ’and in it,” Thomson said, nodding to the bubbling liquid nitrogen, gas spewing down the side of the cylinder, “it would turn into glass. You could smash it to dust.” He shook his hand once toward the hard epoxy countertop, then jerked it back. “The ‘frozen frog’ effect,” he said ghoulishly, recalling an undergraduate prank.

It was Thomson, the Australian, who would attempt to isolate from the stringy thymus meat the scarce, gossamer protein molecules of FKBP, the putative target for the drugs Vertex planned to design. But Harding, as codiscoverer of the protein and of cyclophilin, pitched in with the cutting, a familiar ritual to both that in Harding’s case had led to a highly promising career in immunology, and in Thomson’s, to a prized mastery of a dying art coupled with a mordant resentment of how other scientists viewed him.

On the day Harding arrived at Yale nearly a decade earlier as a postdoctoral in pharmacology, the first thing he noticed was other postdocs cutting thymus. Starzl’s recent rescue of cyclosporine had revived interest in the molecule, and Yale was among those places looking for its protein target within the body—the next step in understanding how the drug worked. The effort had been partly successful: Students in the lab of a pharmacologist named Robert Handshumacher had been able to find several possible candidates. By linking cyclosporine to a chemical tether, they’d used the drug as a magnet, fishing proteins one by one out of a crude thymus “soup.” The problem now was to find which receptor was biologically relevant and determine its chemical makeup.

It was daunting work. Proteins, as any cook knows, are promiscuous and unstable. To be active, they must be looped precisely, yet they’re held together with the molecular equivalent of spit—the combined power of their weakest members, hydrogen atoms. Heat up an egg white, and the hydrogen bonds in ovalbumin, its most abundant protein, fly apart like the rivets in a submarine that has gone too deep; the individual proteins, having lost their internal bonds, flop open and glom together in a gel. Beat the same unheated egg white with a whisk, and the surfaces of the protein molecules explode into a frothy meringue. Extracting protein without destroying it requires the precision and delicacy of an obstetrician combined with a steely equanimity and bottomless capacity for loss.

Harding, a marathoner at the time, developed a steady routine. He came to the lab early, ran six to eight miles out to the New Haven waterfront and back in the late afternoon, then worked until well after dinner. Quietly anxious, agreeable to a fault, he rarely took a day off in six months. Handshumacher’s people had brought the protein to within a couple of steps of purification but hadn’t been able to isolate it from a group of contaminants. Methodically, Harding experimented with different solvents and conditions until he was able to produce trace amounts of the pure receptor.

Doing more experiments, he and his labmates were able to calculate its molecular weight and, ultimately, characterize its sequence of amino acids, the variable links of any protein chain. Proteins are identified chiefly by their size and amino acid sequences and by other characteristics such as their behavior in water. Curiously, the protein they had discovered, cyclophilin, so named because of its affinity for cyclosporine, was highly soluble: it existed in that part of the cell between the membranelike sack and the nucleus. This would be consequential later on, as it meant that the protein was not one of those well-understood surface molecules that immunologists considered central to organ rejection, but something new, some part of the cell’s inner workings.

Harding’s paper, his first as a postdoc, appeared in Science, which ranks with Nature as one of the two most widely read journals in the world for early reports of biological breakthroughs. Though its significance was debated—“We had a structure without a function,” he explains; “a lot of people said, ‘So what?’ ”—the discovery launched him professionally, leading to a faculty appointment at the medical school and the opportunity, harder and harder to come by for a. young scientist, to do independent research in a competitive field. Yet to lure the kind of money that would enable him to support a serious lab effort and to thrive academically, Harding needed a success independent of Handshumacher. He needed his own molecule.

Stuart Schreiber, though more secure than Harding professionally, also was interested in new molecules for similar reasons. For fifty years the great achievements of organic chemistry were in synthesizing biologically active molecules, and Schreiber, still in his twenties and already a full professor at Yale, was conceded to be one of the coming masters of the art. Like Boger, who had just had his own first great success at Merck, Schreiber was drawn to cyclosporine not for what it did, but for how it did it—its structure. Yet he also was blocked. Major work on the synthesis of cyclosporine had already been done in the lab of his department chairman, Sam Danishevsky. Danishevsky was Schreiber’s chief sponsor. Much as he was ambitious, Schreiber was careful not to be thought uncollegial.

A leader in studying cyclosporine, and now cyclophilin, Yale pressed ahead. All universities covet such leading positions for the money and prestige they invite, and Yale, long Harvard’s inferior in the sciences, relished especially any project on which it had an edge. It began collaborating with Merck, taking money in exchange for reagents and data and, not incidentally, bringing into contact Boger and several of those, including Harding, whom he would soon hire. The university also sought large-scale federal funding of the type that grants an institution instant benediction as a center of a particular field. As junior members of the team, Harding and Schreiber often found themselves sitting together in the back at group meetings, talking about science and falling into friendship.

“In October 1988 there was an NIH [National Institutes of Health] site visit,” Harding recalls. “I already had my faculty position—actually I was half an assistant professor, and half a postdoc, which served everybody’s best interests but mine—and the Warty paper had come out.” (A Starzl conscript named Vijay Warty was the first to speculate publicly that FK-506, like cyclosporine, worked through a discrete binding protein in the body.)

“Stuart had become interested in FK, which was natural, and we were at this meeting where Bob [Handshumacher] was talking about the derivative of cyclosporine we had used to discover cyclophilin. I don’t know if Stuart was aware of how easily you could pull protein out of a crude mixture, but when he saw this little reagent we had, he started drawing, and his eyes lit up. He kept saying, ‘Fujisawa. Fujisawa. Fujisawa has this compound.’ Then he said, ‘We’re making this, and I’m pretty sure we could put something on to make a reagent you could couple to this column.’

“He knew I needed to do something new, and he said, ‘If we could make something, would you be interested?’

“I said, ‘Yeah. In one experiment we could find Fujiphilin.’ ”

Schreiber by then had been “called” to Harvard. Harding’s first experiments, meanwhile, were a “dismal failure,” which, he says, “was good. I always like experiments that don’t work the first time, because if they do, they never work again.” In February 1989, the Nature papers about the role of cyclophilin in protein folding and the burst of attention that followed suddenly lifted Harding’s standing as codiscoverer—his molecule now had the cachet of an important biological function. Then, in March, Harding ran a thymus extract through a one-inch column and in two days came up with what he was looking for—a bright band on a photographic gel that indicated a protein with a strong molecular affinity to FK-506. He sent the gel to Yale’s protein chemistry lab, which conducted a quick search of the data bases. Harding’s protein was brand new.

For someone in Harding’s position, the discovery was a bonanza, provided he and Schreiber published their results first. All that was likely to follow—important publications, funding, patent royalties, more lab space and students, international recognition, wealth, travel, star power, and, perhaps, after all that, tenure—hinged utterly on not being beaten to press. Yet almost as soon as he was done with his experiments, Harding began hearing rumors that Merck had also discovered a high-affinity binding protein for FK-506. In science, Harding knew, winner takes all; there are no silver medals. The only saving grace is that, because of the stakes, there are an inordinate number of ties.

“The race was on,” Harding recalls. “I was trying to get some of the material, purify it, and prove that it was really a specific binding protein. And of course the obvious thing to ask was whether [like cyclophilin] it catalyzed protein folding. If you didn’t ask that question, you didn’t deserve to be a scientist.”

With Schreiber calling every day from Cambridge, Harding raced to finish his experiments, then drove to Harvard with a draft of the paper. In Rahway, meanwhile, Boger’s former collaborator on FK-506, a Canadian-born M.D./Ph.D. named Nolan Sigal, was similarly pushing hard to finish up. Sigal’s group submitted its paper to Nature on June 16, three weeks ahead of Yale and Harvard. Though the two articles were finally published together that October, as Boger was leaving for Barcelona, the Merck paper appeared first in the magazine, formally signaling its priority. The distinction was noted carefully only among scientists. To the world at large, copublication signified a tie, and Harding, listed as first author, would thereafter be known as a discoverer of FKBP. Because universities customarily own all discoveries in their labs, Yale and Harvard filed jointly for the patent.

Harding was elated. He was on his way. But if he was happy in his career, he was miserable in his job. He was working at the medical school under surgeons who depended on him for test results involving their transplant patients. To them, as for Starzl, a junior faculty member in immunology was prized for his technical ability first and his ideas second. Far from aggressive, Harding found himself running assays for doctors rather than doing his own experiments. Frustrated, and encouraged by Schreiber about the possibility of further collaborations, he wrote Boger in July asking for an interview at Vertex.

•  •  •

A small tableau graces the bookshelf above John Thomson’s desk at the end of his lab bench: a jar of Kraft Vegamite, the bitterly salty soy paste Australians favor over peanut butter, and three bottles from a chemical supply company labeled, in plain black type, Caffeine, Nicotine, and Ethyl Alcohol. The bottles, though meant to be ironic, represent Thomson’s muses. He gulps sludgelike coffee that no one else at Vertex will touch, smokes unfiltered Camels, and drinks hard. In the lab, in winter, Thomson wears Ray-Bans, battered running shoes, tight blue jeans, and even tighter T-shirts; a favorite one bears a picture of Calvin, the boy in the comic strip Calvin and Hobbes, and says: “I hate everybody. As far as I’m concerned, everyone on the planet can just drop dead.” The introduction to his doctoral thesis contains the prologue to Goethe’s Faust, in which Mephistopheles mocks the existence of a good soul.

Like Faust, who sought redemption in applying science for a larger good, Thomson is a creature of the laboratory. It is his sanctuary, his crucible, his cave. Yet his attachment to it separates him from his peers. With science drawing more and more from the norms of business, individual success now most often correlates with how much time a researcher spends on the phone, on airplanes, and in meetings, not in the lab. As Harding had discovered at Yale, doing experiments one becomes the “hands” of others whose work is moving ahead of one’s own.

Boger, who had hired almost exclusively people who were on the verge of moving up sharply in their careers, was determined to reverse this trend. To get full value from his scientists and to get them talking to one another and help block the formation of hierarchies, which he considered a dull waste of talent antithetical to good science, he insisted that even his most senior researchers make a commitment to working at the bench. Boger personally designed the labs without private offices so that even those making $80,000 a year had to lug their briefcases from one communal desk to another to write letters or make phone calls. Boger’s “social experiment,” as the scientists called it, was another unpopular idea, like making use of the SAB, that most of them tolerated begrudgingly. They assumed it would fail and hoped that it would be sooner rather than later—except, notably, Thomson. “I’m proud to be called a blue-collar scientist,” he said.

Fortunately for Thomson, Harding and Merck appeared to have discovered the same FKBP, which saved him from the abyss of a thoroughly blind search. They agreed on a number of key points—how tightly it bound to FK-506, its chemical sequence, that it accelerated protein folding—which Vertex could translate into analytical tools. Eventually, Thomson would use those tools to determine if what he had was what he was after. But he didn’t have them now, and Harding and Merck had also had their differences. Harding’s protein had a molecular weight of 14,000—it weighed 14,000 times as much as a hydrogen atom—while Merck’s weighed 11,000. In the first stage of Thomson’s search, FKBP’s known characteristics were the only signposts. That some were confusing and others unavailable hardly cheered him.

Thomson plunged ahead with what he knew: the approximate size of the protein. Like Harding, he had chosen the thymus because it was the organ in which the preferred receptor of a drug whose chief effect was immunosuppression was likely to be most abundant. Shrunken and vestigial in adults, the thymus is draped illustriously over the heart in infancy and is one of the seats of the immune system, churning out those cells that orchestrate the body’s defenses, the T cells (T being for thymus). Of the tens of thousands of distinct proteins in each T cell, only a handful were likely to be in the 11,000 to 14,000 molecular weight range. By creating a clarified protein soup and running it through a series of submicroscopic filters, Thomson initially hoped to capture FKBP solely on the basis of its size. In other words, he would pan for it.

In a dozen years of wet-lab experiments going back to the University of Melbourne, Thomson had come to know animal tissue, which is coarse and matted, and protein, which is lacey and ephemeral, and could separate the two without harming the protein as well as perhaps anyone in the world. It was grueling, unglamorous work. Tierchemie, runs an old German expression, ist nur Schmierchemie (Animal chemistry is just the chemistry of slimes and messes). Observing once the aftermath of a small prep in Schreiber’s lab at Harvard, Thomson had taken note of the flecks of wet tissue covering the walls.

Now, at Vertex, he decanted the snap-frozen pieces of thymus into an industrial-strength blender, which pulverized them without spraying. Then, adding water, he reconstituted the mixture, producing a salmon-colored bisque. He poured the bisque into plastic jars, carrying them into another room where he placed them in a high-powered centrifuge, which spun out much of the fat and some of the larger, heavier proteins. As with every step, Thomson had to weigh the necessity of treating certain constituents harshly enough to remove them from the broth while gently leaving the FKBP intact and unperturbed. To this end he used as few solvents as possible and kept the mixture, which could change even at room temperature, refrigerated for all but short stretches. This required him to work long hours in a forty-degree Fahrenheit cold room, which he did in his T-shirt, slipping out only for an occasional swig of coffee or a cigarette.

Once Thomson started something, he didn’t rest. When the first stage of the prep gave way to the more difficult step of clarifying the extract, he immediately set in with a new and more demanding series of experiments. Proteins tend to be colorless. If Thomson was going to obtain a mixture from which he could single out individual proteins, he had to eliminate everything else, especially the fats and waxes that would gum up his microfilters and gave the bisque its pinkish cast. Generally, these can be pulled out with organic solvents like ether and chloroform, but so can protein. Over a period of days, Thomson, working mostly in the cold room, washed and rewashed the extract until it was almost clear.

To manage the widening effort and also because it suited him, Thomson began staying at Vertex two and three days at a stretch, then three and four, then four and five. As the work expanded, so did his compulsiveness. During the day he raced between his bench and the cold room, running several preps simultaneously, he and Fitzgibbon, his assistant, becoming a two-man factory. At night he kept up his vigil while washing glassware, a purifying ritual that often lasted several hours and was still going on when others began arriving in the morning for work. His hands became chafed and swollen from being in detergent and in solvents. His feet swelled from his being on them for twenty-four hours or more between those occasions when he would collapse in a settee in the lunchroom and, with his Ray-Bans still on, sleep for a few minutes sitting up. His face assumed a fluorescent-induced pallor.

A complex mixture of choice and necessity propelled Thomson ahead. More than anyone else at Vertex including, possibly, Boger, he was a true believer in structure-based design—he imagined Vertex in utopian terms. He passionately admired Boger and the all-for-one, lab-centered, egalitarian social ethic he espoused within the company, and he believed Vertex would be limited in its success only by a shortage of pure, high-quality protein: protein that he would now provide. At the same time, Boger and Aldrich had begun negotiating a possible collaboration with Glaxo, Inc., the giant British pharmaceutical company, which planned to send a delegation in mid-February. Boger had called this a “camel’s nose” visit, with Glaxo sniffing Vertex’s goods, and was pressing hard to have “as much on the tent wall as we can.” Thomson had vowed publicly to have pure FKBP to show Glaxo. He had five weeks. “I’d stuck my neck out,” he says.

As with almost all of Thomson’s actions during this time, there was a Faustian motivation to his resolve. Unlike Harding, he had not come to Vertex borne on a succession of triumphs. The previous two years in America had been a downward spiral, punctuated by impulsive acts he would come to regret. Though he had done superb work resulting in several important publications, his postdoc at MIT ended badly with a falling out with his lab chief. Thomson, who was married and had two young children, began seeing another woman. His marriage broke apart, with his family returning to Australia. He then was hired by Boger at Merck, but Merck’s lawyers were unable to arrange a work visa for him, and he ended up taking an academic post in Wyoming for a year. His girlfriend dropped him soon after.

Thomson came to Vertex “in a very angry sort of a situation” and “determined to make a new start.” But after a smooth beginning, his troubles resumed. He met another woman, but she left him for Jeff Saunders, a Vertex chemist who had become Thomson’s closest friend, in a reversal that many others at Vertex apparently knew about before Thomson did. Blaming himself for the breakup of his marriage and his stalled career and now feeling mocked and betrayed, Thomson began drinking heavily. On a night in early November, he got drunk, climbed on his motorcycle, and roared off from an east Cambridge bar on his way back to Vertex. He went most of a block on one wheel before skidding on a puddle and careening out of control.

Thomson totaled the bike and tore up his hands, but his rage was undiminished. Three weeks later he went out drinking again with another chemist, John Duffy. This time Thomson drove his car. “I got shitfaced,” he recalls, “real rip-roaring drunk. I don’t remember leaving. I don’t remember the accident at all. I just smashed into another car head on. I went through the windscreen, smashed my face up, and totaled the car. I was looking through my eyelid.”

Thomson was taken to Massachusetts General Hospital, then to the neighboring Massachusetts Eye and Ear Infirmary, where doctors sewed up his eye and removed dozens of glass slivers from his face. “I remember waking up,” he says, “and saying, ‘I need to call someone at work and tell them I’m not coming in.’ ” Two days later he was back at the bench, resuming his pursuit of FKBP and seeking redemption, as Faust had, in the one thing he hadn’t destroyed—his work. His right eye was swollen and bandaged behind his Ray-Bans, and his vision was blurred. For months afterward, throughout his ever-longer sieges in the lab, he continued to pick pieces of glass from the stigmata on his face. He avoided all but perfunctory contact with most others in the lab, keeping to himself.

“I was real screwed up at the time,” he says. “Plus I didn’t have a vehicle. I couldn’t afford one and I didn’t want to drive: I was afraid I’d hurt someone.

“I just wanted to make things work and I knew what needed to be done. I didn’t have wheels. I didn’t have a lot of friends I could trust. And I had a lot of work to do. So I stayed here and I did it.”

•  •  •

To Boger, the impending Glaxo visit meant two things, money and benediction, each more vital than the other. Vertex needed a corporate partner to carry the immunophilins project forward and to help salve its burn rate. It also had to show the drug industry and investors that someone (it didn’t especially matter who: Nissin, the largest Japanese noodle manufacture, was another early prospect) took the company seriously enough to put up a stake in it. That Vertex’s suitor was Glaxo, the most dramatically successful company in Europe throughout the previous decade and the strongest challenger to Merck’s hegemony in drug research, sweetened Boger’s calculations dramatically.

The drug industry, powered by the huge profits of the 1980s, had lately embraced the concept of “strategic alliances.” Fervently in vogue, they were thought to solve a common problem: in a fractionated marketplace spanning countless diseases, an explosion in knowledge, and thousands of laboratories, no big company was big enough and no small one clever enough to go it alone. Kissingeresque marriages of convenience were a necessity. For Boger to be entertaining overtures from Merck’s chief rival, a company with more than $2 billion pouring in annually from the world’s best-selling medicine, automatically enhanced his position. Glaxo had never done a large research deal before—a “virgin,” one former manager called it—but Boger was unworried. He was sure it would come around.

But Glaxo also was uneasy. Boger’s ambiguous relationship with Merck troubled some company officials, who fretted a Merck “backlash.” Had Boger’s separation been clean? they asked: as one of them put it, “Could Merck ever come back at Josh?” It was a reasonable concern. Boger had gone to great pains to “lobotomize” himself against what he knew about Merck’s research, refusing to discuss anything that wasn’t published. Yet by choosing as his inaugural project a target molecule he had previously pursued at Railway, he invited inevitable questions. Merck’s senior management was known to be furious about the decision. (“Does Vertex have a single project that wasn’t first formulated here at Merck?” Ed Scolnick, Merck’s head of research, would fume more than three years later; he refused to elaborate.) And so Boger remained, at least for now, a defector, highly prized yet suspected on all sides. As much as Glaxo and other big drug companies might like to know what Merck was doing in immunophilins, they preferred not to know it from an impoverished prospective partner whose bills they were being asked to pay and whose fortunes they were considering hitching to their own.

In fact, Boger knew little. He knew Merck’s approach in such situations, which was to take a molecule like FK-506, synthesize it, and have its chemists begin making molecules that were similar, testing them for activity. But that required a method of testing, something no one in the world had prior to the discovery of FKBP several months after Boger had already resigned. His own ideas for improving the molecule were considered radical and went untested while he was still at Rahway. He was confident they would not have been pursued since.

Boger and Aldrich visited Glaxo during a week-long swing through Europe in December, and Boger had come away impressed enough to consider the company a possible competitor—another reason for seeking a collaboration. Yet he knew Glaxo was probably uninterested in Vertex’s theories about drug design, being even more firmly committed than Merck to discovering its drug leads through screening. What seemed to interest Glaxo most about Vertex, besides Boger himself, was what Vertex might know about the biology of FK-506 and, in particular, Vertex’s relationship with Schreiber—a relationship that, unknown to Glaxo or anyone else outside the company, was now starting to unravel.

After the first SAB meeting in October, Aldrich had confronted Boger about Schreiber’s ambiguous relationship with the company. Prior to the meeting, Schreiber had insisted he had no interest in making drug molecules at Harvard. But in exploring how FK-506 blocked FKBP, he and his group had now begun making compounds that mimicked parts of the drug. Aldrich, as head of business development, failed to see the distinction between Schreiber’s molecules, which Schreiber claimed were only for experiments, and intentionally designed drugs. To him, Schreiber represented only severe, multiple contaminations. If one of Schreiber’s graduate students made a molecule similar to one discussed at Vertex, the company could be dragged into a ruinous patent fight with Harvard. If a Vertex chemist made a molecule similar to an idea Schreiber claimed he had while driving to work and mentioned accidentally at an SAB meeting, Vertex could be forced into paying the university steep royalties. Either way, Aldrich told Boger, Harvard now had a lien on the company.

The combination of Schreiber’s personality and ambition posed a more immediate threat. He brimmed over. Like Boger, who was more circumspect but just as loquacious when it served his needs, Schreiber loved to talk about himself and what he was thinking, and he now had the world’s ear. “I’m not concerned that Stuart will find a compound that will compete with ours,” Boger said. “But I’m tremendously concerned he may tell everyone in the world what we’re doing. Two of the companies we saw in Europe said they had telephone calls from Stuart within the past couple of weeks. His phone bills must be enormous.”

Boger feared that Schreiber was “persistently naive” about the need for secrecy; Aldrich thought him calculating and opportunistic. A “good Stu/evil Stu” polemic ranged sporadically between them for months while they tried to negotiate an agreement with Harvard that would give Vertex exclusive rights to any discoveries relating to immunophilins in Schreiber’s lab. Schreiber, for his part, favored such an arrangement, since it also included a collaboration he coveted. The big prize with any new protein was solving its three-dimensional structure. But Schreiber had no X-ray crystallography lab of his own and no ready access to one. For the chance to work with Manuel Navia and Vertex’s biophysics group, Schreiber was happy to grant Vertex sole rights to whatever he produced.

The obstacle was Harvard. Ethically, the commercialization of academic science had given the university, as the standard bearer of American education, an immense institutional heartburn. No entanglement was harder to swallow than the specific type of alliance that Vertex and Schreiber were now proposing, an exclusive license between a professor and a company in which he or she owned stock. Such arrangements slashed at every presumption of liberal education—free and impartial inquiry, pursuit of learning for its own sake, the sanctity of knowledge—and replaced them with a tangled system wherein a faculty member stood to profit directly from the work of students. Harvard, though it had negotiated two or three such deals over the last decade, recoiled ritualistically each time one came up and was balking now at Vertex’s proposal.

In December, Boger, Aldrich, and Boger’s brother Ken, Vertex’s lawyer, met at Harvard with officials of the Harvard patent office to discuss a compromise. In preparation for the meeting, Schreiber had sent them the terms of a $10 million research agreement he and two professors at the medical school had recently signed with Hoffman-LaRoche, the large Swiss-owned drug company. In his cover letter, Schreiber had said he saw no conflict between his work with Roche and his commitment to Vertex. Now, however, Joyce Brinton, Harvard’s head of licensing, was reviewing a proposal that Schreiber had written earlier, in which he listed, among those research interests that Roche might want to support, his group’s work in chemistry and immunology. In other words, FK-506.

Boger was stunned. Schreiber, who had a major equity stake in Vertex and was privy to its science, had licensed his work in immunophilins to a direct competitor; Vertex, Boger thought, might as well have invited Roche to its lab meetings. What’s more, Schreiber hadn’t told him. Schreiber considered the overlap the result of “miscommunication.” Moving quickly to correct it, he agreed to restrict the Roche agreement to work in another area, AIDS, in which he had suddenly become interested. Still, as word spread among Vertex’s scientists, their anger and confusion soared. Pressure mounted on Boger to act.

Boger was caught in a contradiction of his own making. He had wanted Schreiber involved in the company, a decision that had already paid off with the recruitment of Harding and others and was now of defining interest to Glaxo. Schreiber’s marquee value, which remained Vertex’s alone despite the Roche deal, was indispensible for the sort of joint venture Vertex now needed to survive. Yet Vertex had nothing if not its ideas for designing a novel immunosuppressant, and Schreiber, who knew most of them, was acting like a free agent. Demonizing Schreiber had its advantages. Having a famous academic collaborator who appeared to care only about himself reinforced Vertex’s culture of industrial chauvinism, and it helped to close ranks behind Boger, who, despite his success in recruiting, still had to sell himself internally by proving he could play in the blood sport of big-time science. But Boger was walking a fine line: making good enemies was one thing when they loomed at a distance. It was another when they were in your huddle.

No other recent situation so rankled Boger, who, ever since, as a child, he had watched his father squander his business dealings, had developed a conservative’s heated loathing for the expropriation of intellectual property. Boger didn’t think Schreiber would contribute significantly to Vertex’s science, yet Schreiber was moving so fast it was impossible to rein him in. That Harvard, which had done nothing toward creating the company, was also now in a position to share in Vertex’s future appalled Boger as much as if it were some new, especially heinous form of street crime.

Resisting Aldrich and the scientists, Boger determined that Schreiber would stay but in name only. From now on, he would be brought out for such goodwill displays as the Glaxo visit but otherwise was to be treated like any other competitor. Boger ordered all lab books updated and signed daily to document the conception of ideas—a defensive measure anticipating future lawsuits. As for direct contact with Schreiber, Boger counseled his usual rule for negotiating with rivals: “Tell them only what they need to hear so that they’ll tell you what you need to know worse.”

“I think,” Boger told the scientists, “everybody should be very friendly to Stuart and listen to what he has to say.”

•  •  •

Boger didn’t inform Schreiber of the change. He worried that Schreiber might bolt, negotiating with Glaxo on his own. He had other reasons for keeping Schreiber in the dark. Now that Schreiber and Vertex had both begun making compounds, the stakes for each of them had risen exponentially. Despite his denials, few things troubled Boger more than the possibility of Schreiber’s group developing a promising drug lead before Vertex—a prospect Schreiber believed he already might have achieved.

In December, Schreiber’s group had synthesized a new molecule that he called FK-506 binding domain, or 506BD, which was central to Boger’s concerns. FK-506 is what chemists call a macrocycle—as its name implies, a large ring. It has 126 atoms and a pronounced, hornlike gaff. Drawn conventionally (Schreiber, practicing a different aesthetic, draws it upside down), it looks like this:

As shown in the figure at the left, FK-506 has two regions. At left is the binding core, which connects the molecule to its target. The right side, the effector domain, remains exposed. It extends from the protein surface (the northwest to southeast axis in the drawing) to interact with other molecules. The balls in the picture depict atoms.

As a chemist, Schreiber cared little that FK-506 was a powerful drug. What intrigued him, indeed what drove him to a kind of solipsistic rapture, was something else: its extraordinary resemblance to another microbial compound, rapamycin, which not only was immunosuppressive but also inhibited cells from reproducing.

Isolated in 1975 from a soil sample from Easter Island, rapamycin was considered a possible anticancer agent, but again Schreiber was uninterested. What consumed him was the almost mystical fortuity: two microbial compounds, sharply similar structures, yet apparently different modes of action. As with any molecule, the big prize academically was figuring out how it worked—that and, of course, being first. Here, quite possibly, was a three for one opportunity. Schreiber couldn’t help but be enthralled.

Thus 506BD. It contained only those structural pieces common to FK-506 and rapamycin, and Schreiber and his colleagues theorized that 506BD was the part of each drug that bound to its target. However, when they tested it, they discovered that although it blocked protein folding as well as FK-506, it was “dead”—biologically inactive. In other words, something else appeared necessary to make FK-506 a drug.

Schreiber was ecstatic. He began to theorize that what accounted for FK-506’s activity was not its ability to block the protein-folding action of FKBP (as Vertex and everyone else still assumed), but some other function. This he attributed to an effector region—that part of the molecule not associated with binding. Using 506BD as a platform, he and his students began building outward to replicate the missing piece.

Boger maintained that he, not Schreiber, had first proposed making 506BD at a meeting at Vertex in September, although getting credit for the discovery didn’t concern him nearly so much as what would happen now that Schreiber had begun announcing 506BD to the world. Though Boger discounted the molecule as the basis for drug design and dismissed Schreiber’s conclusions, he believed other drug companies would be drawn to the theory. And that, Boger feared, could be shattering for Vertex.

Boger had little but contempt for what he called the “monkeys with typewriters” approach to medicinal chemistry practiced by virtually all major drug companies. Legions of chemists systematically changing every atom and subgroup on a molecule in order to improve it was not Boger’s idea of intelligent science. Like screening, the approach was largely statistical—how many chemists making how many derivatives for how long—and Boger didn’t like gambling when the odds were against him.

But conventional medicinal chemistry could also, Boger knew, be extremely successful, and it favored those companies with the most and best chemists. Merck had routinely been able to bludgeon competitors simply by throwing more chemists at a project.

This was what troubled Boger now. Without its own legions—Vertex had only seven chemists making compounds; Merck, which guarded the information as if it were a state secret, more than 400—and without any structural data, Vertex was left with its own risky strategy for drug discovery, a strategy that Boger had devised. The approach was to explore not miniscule variations in the structure of a molecule, but to carve it up wholesale in an effort to discover those minimal portions that accounted for its activity. Such “small-molecule” derivatives—attempts to reduce a drug to its minimal size and weight while still retaining its effectiveness—were controversial among drug developers, but Boger believed in them categorically and counted on others not to do so. Small molecules were Boger’s religion, the intellectual basis for his faith in structure-based design, the critical fault line in his break with Merck and the industry. Indeed, if Vertex had a “secret formula,” it was the half-dozen small-molecular blueprints that he and the chemists—some in discussions with Schreiber—had designed.

Boger believed that 506BD, although dead, would broadcast Vertex’s strategy to the world. It would point toward those other small molecules that Vertex had made and that, although probably novel, would not be hard to imitate and expand on—something Boger believed no company would try without prompting from someone as effusive, as influential within the field, as Schreiber. Boger couldn’t believe Schreiber didn’t see that. And yet Vertex couldn’t afford to lose him. Other drug companies would be pounding down his door. And Schreiber, in his ebullient way, would want to accommodate them.

On the second Friday in January 1990, Schreiber visited Vertex for the first time in weeks. A meeting with the full SAB was scheduled for the next day, and Schreiber had come with one of his postdocs to report on some new work—“trying to put some cookies back in the jar,” suggested Harding, who had begun to be torn by shifting loyalties and was growing increasingly uncomfortable because of it.

Schreiber seemed no less ill at ease, for other reasons. Although he had helped recruit Boger to lead the company, he was increasingly unimpressed by Vertex’s scientific production, especially Boger’s decision to try to make drugs that were inhibitors of FKBP. “A nonstarter,” he called the effort. (What else Vertex might have made compounds to do, given that the true mechanism was unknown, remains unclear.) He was convinced that Vertex’s scientists were concentrating on the wrong parts of the molecule, and he saw no gain in being associated with their problems. It wasn’t that he distrusted them, as a growing number of them distrusted him. He disrespected them—in Schreiber’s view, a far harsher judgment.

What Schreiber and Boger thought, they colored the world with and their mutual displeasure now set them irremediably apart. Black was the other’s white. Not only, would Schreiber say, had he not disclosed the company’s secrets as Boger believed and feared, he would have been embarrassed to do so. “Other people’s studies were irrelevant to me,” he says. “It’s absolutely absurd that I would talk about Vertex because they never had any work to talk about.” (In fact, Schreiber by now had been taken unawares out of Vertex’s communications loop and didn’t know what the company was doing. His disdain reflected accurately what Boger and the scientists were telling him.)

Meeting now with Harding and David Livingston, Vertex’s chief enzymologist, in the cramped conference room, Schreiber seemed tentative, awkward. He repeatedly reached for something in his open briefcase, then pulled his hand away as if reminding himself that he couldn’t touch it. The temptation grew, with Schreiber muttering “dammit” at one point before bringing back his hand sharply. After forty-five minutes, Schreiber finally pulled out two documents—a preprint of a recent article he had written on cyclophilin and one by a competitor addressing many of the same issues. After a minute, he returned his own paper to its file, apparently feeling free to share his competitor’s work, which was in press, but not his own.

After Schreiber left, the tension of the past weeks suddenly erupted in a round of gripe sessions that stalked Boger wherever he went. The Roche deal, 506BD, Harvard’s hard line, the next day’s SAB meeting, the decision to keep Schreiber on—Boger was called sharply to account for each.

“Digging up your neighbor’s yard because you want a bigger garden is one thing,” groused Harding, standing in an angry knot with three or four others, “but Stuart’s rototilling the whole neighborhood.”

“The problem is he tells people whatever he’s thinking,” said chemist Jeff Saunders.

“The problem is he’s making compounds,” Boger said. “We can control Stuart’s life when he’s here, but he has another life we don’t control.” Boger paused. “The problem . . . the problem is that they’re exactly the same.”

“We should follow Stuart to every meeting he goes to,” Saunders said.

Hal Meyers, a chemist and one of Schreiber’s first doctoral students at Yale, nodded approvingly. “It might work. Stuart doesn’t document anything.”

“Good,” Boger snorted. “We document, he doesn’t. At least we don’t want to change Stuart’s method of operation in that regard.”

“Can’t he be informed that there’s fame and there’s fortune?” Saunders asked.

There was a collective sneer. “He thinks he can have both,” Dave Armistead, another chemist, said.

“When we can walk into a major drug company and do a half-hour seminar on our own findings,” Boger said, “Stuart’s name won’t mean anything to us anymore. But until then, his capital is too valuable to squander.

“Stuart’s value and his liability to us are on approaching curves. Unfortunately, his liability factor is not as great as his value right now. But those curves aren’t flat. They’re going to cross.”

For months Boger’s stock answer to questions about competition from Schreiber was that Vertex would soon surpass him. Schreiber would cease to be relevant, Boger said, the day Vertex was first to solve the structure of FKBP—when Vertex had vital data that no one else, including Schreiber, had. Boger had no doubt this was true, but now, in the gloaming of a late Friday afternoon in January, after weeks of mounting frustration, he turned impatiently on his heel and headed back to his office.

“The world is competitive enough,” he said, “without competing with yourself.”

Two hours later, during the customary Friday afternoon beer hour, John Thomson, stumbling and glazed, approached Boger in the lunchroom.

“We’ve got the pure protein,” he said.