Aburst of hiring in May and June brought Vertex by the first week in July to more than forty people. In the labs, benches were divided, hoods shared, egos disjoined as new scientists, a disproportionate number of them Ph.D.’s, arrived weekly. Rivalries flared as some groups grew faster than others. There still were no permanent desks for the scientists, although some, like Navia, asserted squatter’s rights and were alternately deferred to and reviled for it. The squeamish Massachusetts economy doomed several proposals for more permanent space, forcing Boger to lease a former airplane-parts factory on Putnam Avenue, a block away. Twice as large as the original building, it wouldn’t be ready until mid-August. Then, the scientists would be divided between the two sites, kindling other jealousies. Now, Boger knew he was tightening the lid on a pressure cooker. Exalting turmoil, he thought it a good thing.

As reagents in Boger’s social experiment, the new scientists catalyzed several critical reactions. Synergy, the combined action of disparate forces and a favorite concept in drug design and of Boger’s, increased sharply. So did entropy, the tendency in all things to pull apart and degrade. If once there had been a singular Vertex identity modeled upon Boger’s—male, aggressive, entrepreneurial, brazen—that character was no longer so simple or viable. Many of the new hires were women and many came straight from graduate schools and postdocs. They had different sensibilities, different missions. The rough and tumble of business thrilled them less than Boger’s promise, taken at his word, of doing academic-grade science in attention-getting areas. Vertex’s core remained its ex-Merck contingent, but with Boger declaring all but a moratorium on hiring any more Merck scientists, its preeminence began to wane. Vertex was now filling with people who’d never heard of Max Tishler, much less deemed themselves his heirs. Many of them had been impressed by Boger at their interviews but had few direct dealings with him since. As with the scaleup of any chemical process, Vertex’s new size bred uncertainty, unfamiliarity, irreproducibility, diminished yield.

Boger’s office door remained open, with the scientists wandering in to talk while he churned through Niagaras of electronic paperwork. He never stopped working during these discussions, but benignly incorporated them into the flow of what he was doing, like a river absorbing a new tributary. His eyes remained locked on the big Macintosh IIci on his desk and his fingers tapped intermittently at his keyboard, giving the conversations a one-way, confessional air. Not that he wasn’t engaged—his mind could still move like a laser when he wanted to make a point, and he was relaxed and personable. But as they slumped in one of two chairs, so close to his desk that their knees knocked, the scientists often sensed that they were receiving only that percentage of his attention that his hypothalamus, the brain’s autonomic regulatory center, had allotted.

When there had still been little to discuss and the number of scientists was small, Boger had used these occasions to talk up the company’s business. Now he was more guarded. “Deals can be killed simply because word leaks that one is in progress,” he told the scientists via electronic mail (E-mail). If anyone asked, he counseled, “Just smile.” He, of course, smiled constantly.

A year earlier Vertex’s entire scientific staff could fit in Boger’s Toyota Camry; now it overflowed the lunchroom. Within the building, Boger communicated primarily by E-mail, preferring it to large staff meetings. Still, on July 7, he called a companywide meeting at noon. Atypically, he gave no reason, inviting rumors and, among the new people, curiosity. A few of them had never seen him lead a large group.

His mother’s son, Boger enjoys drama. He calculates flourishes carefully for maximum effect. Now, wordlessly, he switched on an overhead projector, revealing in the semidarkness a document covered except for two signatures at the bottom, his and Nagayama’s. He smiled broadly. The scientists didn’t know what they were looking at. He then revealed the rest of the page, a letter of intent between Vertex and Chugai. Again, many of the scientists didn’t understand what he was showing them.

Boger quickly put up a second transparency. Entitled “The Standard,” it was a description of the research deal between Cytel and Sandoz, the Swiss pharmaceutical giant and developer of cyclosporin. When it was announced the previous year, the deal, which brought Cytel $30 million but guaranteed Sandoz all rights to a drug and 30 percent of Cytel’s stock, was considered a record for small companies working at the juncture of immunology and chemistry.

“Pitiful,” Boger murmured.

The scientists muttered among themselves, then quieted down as Boger now put up a third transparency: “The New Standard.” It referred to the tentative agreement, signed by fax the previous night by Boger and Nagayama. The deal gave Vertex $30.25 million—“Not an accidental figure,” Boger said—and a 50 percent share of the world market. It would cost the company only 5 percent of its stock, a token amount.

“We get more money and give up half, less than half, of what Cytel did,” Boger said. “This is the world’s best deal by a factor of two.”

The room erupted. More than the money, which was crucial, the announcement rang with deliverance. Boger had promised everyone he hired that Vertex would be better than other companies: They were the original buyers of his story. And yet the constant dire need for money had discouraged many of them. They worried privately whether they had made the right choice and whether Boger was to be believed. Now, it dawned on them—some for the first time—that Boger’s optimism, his aura of glowing success, was justified. Suddenly, they were much closer to being a drug company and getting rich than their own work indicated. Corporately, if not scientifically, the company had legitimized itself much sooner than even Boger had expected.

Boger lavishly credited Aldrich as the deal’s architect and savior. “Benno helped enormously,” he said, “but ultimately even Benno thought it was a lost cause. None of the board members thought it could be done.” Aldrich was more modest. Taking no credit himself, he hoisted it all back onto Boger. A few of the scientists still distrusted Aldrich and, despite their euphoria, clucked disapprovingly under their breath.

In fact, in the month since Chugai had first responded positively to Schmidt’s ultimatum, Aldrich had been engaged in one continuous negotiation and at least once had kept it from collapsing entirely. “I have to restrain myself,” Boger had said upon reading Chugai’s initial response in June, “but there are no issues in this.” Worse perhaps than issues, however, were the fine points, nuances, understandings, and subtleties of language—“hidden grenades,” Boger called them—that Aldrich now addressed in a series of daily, multipage memos back and forth to Japan. From the time he came in each morning until he left after dark, he worked on sharpening Vertex’s demands, then slipped them into its fax machine before going to the gym. When they rolled out of a similar machine in Tokyo minutes later, Chugai’s lawyers and licensing people attacked them as Aldrich worked out and tried fitfully to sleep at home; they shipped their own end-of-the-day fax just as they were leaving work that night and Aldrich was arriving at Vertex the following day. Segue upon segue, over thirteen time zones, it rolled on this way, a textbook negotiation, hurtling twice as fast as it would with a company next door, and proceeded thus until two days earlier, when a glitch arose.

A midlevel functionary in the Chugai licensing office suddenly challenged certain aspects of the deal that had already been agreed upon. Aldrich fired back angrily that if Chugai changed its position, the deal was off. Aldrich held his breath. He knew the potential price for his adamance. It was not inconceivable that the deal would unravel and that it would be perceived as his fault. “Rich got a bit grayer,” Boger recalled. Fifteen hours later a three-paragraph fax arrived with Chugai’s answer. “I apologize for raising and addressing issues in a rather slovenly way,” the author of the previous fax wrote, impaling himself in a way that no American manager ever would. Several hours later, with the debris finally cleared and the deal back on track, Boger and Nagayama signed the letter of intent, paving the way for the scene in the lunchroom.

Aldrich, recalling his experience at Biogen and IG, cautioned the scientists not to count the deal as done until the final papers were signed and Chugai’s check cleared at the Bank of Boston. Much could still go wrong.

But for Boger the full impact of the deal was already clear: It was his deal; the mantle of glory, Schmidt notwithstanding, was all his. It was Boger’s vision, his scientists, his system of science, that Chugai had been willing to pay for at such a premium and on such apparently unfavorable terms. And it was he who would benefit. He now had a powerful ally, someone who could take Vertex’s molecules to market and yet would leave him in charge. He had money, enough to afford Vertex time to meet his most ambitious goals. And he had validation.

Especially this last. Fourteen months earlier, at the first meeting of the board of directors, he’d said he’d be happy to do a $500,000 deal in the company’s first two years. Now it had done a deal sixty times that size—120, if one accepted Boger’s analysis comparing it with Cytel’s. If there had ever been any doubt about Boger’s promise or his competence to run a company, it was now shattered. Schmidt, who had always admired Boger’s business acumen, began singing his praises loudly to the rest of the board. “Joshua is the best fella in this field I’ve seen,” he drawled, “and he’s as good as anybody starting a new company in any field I know of. He and I work together in a negotiation like two musicians that have been playing the same duet all their lives.” Occasionally going so far accidentally as to call Boger Jock, after his friend and mentor Jock Whitney, he smothered any further suggestion about bringing in seasoned management.

Boger now had a secure mandate to run Vertex’s business and science as long as he pleased. He was in total, unmitigated control—god mode.

“I’m going to have to resist the temptation to rent a sedan chair for the board meeting next Tuesday,” he said.

•  •  •

Tom Starzl strode through the mobbed liver transplant unit on the fifth floor of the Falk Clinic. The suite was small, semimodern—four examining rooms, a sparsely furnished office that doubled as a conference room, a cramped waiting room with a TV dulcetly delivering Joan Rivers from high atop a Formica shelf. In the waiting room, hollow-eyed patients, some with walkers, huddled like refugees on a steamer. Along the walls stood relatives, friends, transplant candidates, case workers from donor banks. Inside, surgeons—many still dressed in blood-spattered scrubs and running shoes after operating all night—raced between rooms followed by nurses, case managers, residents, and interpreters and demanded updates, prognoses, indications, charts, reports. At up to fifteen transplants a week—more, until recently, than the rest of the country combined—the clinic joined the magnificent aspirations of Lourdes with the grit and commotion of an inner-city emergency room.

Starzl is king here; FK-506 is king. After the conference in Barcelona eight months earlier, he was besieged with requests for the drug. He tried to keep up, but the demands on his energies soon became homicidal. There was too much to do. He was running the transplant service, pushing ahead with multiorgan grafts, directing new experiments with autoimmune diseases and cellular and intestinal transplants, and administering several burgeoning clinical trials that compared FK-506 with cyclosporine for patients receiving new livers, kidneys, and, soon, hearts. His Monday night FK-506 meeting ballooned. Starting at 7:00 P.M., when Starzl entered crisply in his customary windbreaker and slacks and look of eternal stamina, it routinely lasted past midnight.

If Starzl had once insisted on directing every aspect of the drug’s testing, he now had no choice. Fujisawa, working with the FDA and European regulatory agencies, had been scheduled to begin offering FK-506 at sixteen other sites in April but had failed to develop acceptable protocols. It was now July, nearly a year after the patient revolt in Pittsburgh, and still no one else had the drug. With the world’s transplanters clamoring for FK-506 and unable to get it and exhorting Starzl for more data to support his claims, his work with FK-506 grew more controversial than ever. Once again, his credibility was at stake. “Starzl’s a giant; I’ll be surprised if he doesn’t win a Nobel Prize,” said Dr. Ronald Busuttil, chief of liver transplantation at UCLA. “But the joke has already gone around: ‘FK-506? It’s a unique drug. It only works in Pittsburgh.’ ”

Characteristically, Starzl responded by immolating himself in work. He pushed himself as hard as he always had, but he was now sixty-four and the pace took its toll. He looked strained and gaunt, with liver spots freckling his worn face. He had always been a notoriously irresponsible eater: flying out of the operating room in his scrubs and blood-spattered disposable booties to gorge on donuts or heaps of french fries smothered with melted cheese at the Original, a hot dog place and student hangout a block from the medical center. That hadn’t changed, though he was also now eating a lot of pizza; his office, a third-floor walkup, was above a Pizza Hut. A former three-pack-a-day smoker, he had quit ten years earlier after suffering chest pains but had maintained the nervous energy and compulsiveness of the reformed addict. A competitor once observed that Starzl’s energy quotient is “so far off the scale of most humans that it is almost unbelievable,” but now there was also a measure of desperation to his drive as if he was racing not only against himself and the world, but against time. Working furiously, he was bogged down increasingly on the phone, in meetings, and with paperwork. He came to the liver clinic not to see his own patients—he seldom had time to operate anymore—but as a general whose few relaxations from command included visiting the front. It was here, and only here, that the full measure of FK-506 could be gauged.

“You look outstanding,” he gushed to an energetic woman in her mid-forties. She was sitting on an examining table, amid an entourage that spread to the clogged doorway and beyond. Her exposed torso bore the signature scar of the liver-graft recipient: a cross-hatched, upside-down T running from her sternum to below her navel and from hip to hip. Two months after her transplant, the scar had begun to darken, a rusty color now, no longer pink, and she had stitch abscesses. She complained of soreness, but said she couldn’t have the abscesses drained that afternoon; she had to take her daughter to the doctor. The woman’s hair was thinning slightly, and she had some tingling in her hands. Otherwise, she said she felt fine, an assessment that seemed to be borne out by her appearance. She looked no more haggard than any other middleaged parent of teenagers. Starzl ordered her off acyclovir, a powerful antiviral drug, and bactrim, an antibiotic. He had already taken her off prednisone. Because of the generalized effects of suppressing the immune system and the toxicity of cyclosporine, most transplant patients take up to a dozen other drugs, many of them toxic in their own right, for as long as they live. With no signs of rejection, the woman would now take only FK-506. Starzl was elated. “That’s our objective,” he told her. “To liberate you from us.”

Passing through the throng, Starzl swooped in next on a burly man with a thick black beard and volunteer fire department cap who was being examined by another surgeon. Three and a half months ago the man had been rejecting his new liver. He was jaundiced, his kidneys were failing, and he shook uncontrollably, Starzl believed, from the neurotoxic effects of cyclosporine, which is known to affect the brain. He was also taking heavy doses of prednisone. Now the man was off both drugs, his liver and kidneys had revived, and the tremors were nearly gone. Starzl wanted them gone completely.

“We’re going to drop your FK dose to three milligrams twice a day,” he said. “It’s what we give kids.”

In the hall Starzl was buttonholed by a young visiting oncologist. She asked if he would be willing to recommend FK-506 for a patient dying of leukemia. “Sounds interesting,” he said. “Maybe we can get a compassionate.” Because FK-506 was still approved only for certain transplant patients, all other requests had to be sanctioned individually by the FDA, which depended heavily on Starzl’s opinion. Some he approved pro forma: For instance, the drug had already been shown to work dramatically in extreme cases of psoriasis, clearing up within weeks rampant, open sores that resisted all other treatment. Starzl was determined to test FK-506 against every disease for which there was a clear scientific rationale, especially those like juvenile-onset diabetes, which had shown a response to cyclosporine. It was well established that cyclosporine, if given soon enough, could stop type A diabetes from developing in children. The problem was, the drug was too toxic to be given over time. The relief, tragically, was only temporary. Starzl was convinced that FK-506, because it was less toxic, could cure the disease permanently and was pushing hard to begin a clinical trial in Pittsburgh. He was not indiscriminate, however. He told the oncologist about a Michigan doctor who asked for FK-506 to treat a rare tumor. Starzl had declined. “Sounds like a shark,” he said.

The more patients Starzl saw, the more he was convinced of FK-506’s singular potential, even when its side effects mimicked cyclosporine’s. In the next room, a distraught black woman, thirtyish and wearing a red dress, told him the drug had caused her to have hallucinations; FK-506, like cyclosporine, seemed to be hitting an unspecified receptor in the brain. Terrified, the woman had asked for her dosage to be cut back, but Starzl observed from her most recent biopsy the aggregation of white cells that indicated she was beginning to reject her graft.

“We’re seeing some nibbling,” he told her, his manner clipped but not unfeeling. “We’re heading for disaster unless we do something about it. With this drug, in a few weeks, you can go off everything else, and I think that’ll help. But you’re being undertreated.” Reluctantly, the woman agreed to a higher dose.

Viewing complaints about side effects from the extreme vantage point of life and death, Starzl, as an experimental transplanter seeing the most desperate patients, has always had a higher threshold for considering toxicity a problem than many of those he treats. Certainly, he is willing to overlook many side effects more readily than most other doctors—a disparity that now had critical implications for FK-506 and for Vertex. Hallucinations and kidney poisoning were one thing if you were going to die; they were another if you had an itchy scalp or were an eight-year-old diabetic. In other words, as Starzl was finding—and as Boger had predicted ten months earlier—FK-506 was a remarkably benign drug compared with the multiple traumas of transplant surgery and cyclosporine, but it still might be too toxic for most autoimmune diseases. Second-generation molecules would undoubtedly be needed.

The similar toxicity profiles were important for another reason. The range of side effects Starzl was seeing in patients with FK-506, though lesser in degree, was almost identical now to those with cyclosporine. In a way, their duplication was even more intriguing than the fact that the drugs worked alike. Whatever their targets, however they behaved on a molecular level, the two drugs were so similar that they were likely to be interfering with the same chemical pathway. It was extraordinary. To understand how they worked, what they did, would perhaps unlock one of the most basic secrets of how the body defends itself. It even might help explain how cells—all cells—transmit information internally, one of the premier mysteries of biomedicine.

Starzl felt himself reaching across a gulf that no one had ever bridged: a surgeon drawn across the whole expanse of medicine to plumb the most basic questions of molecular immunology, of life. He had spent his career groping in the dark, feeling his way along the walls of a tunnel, going farther than anyone else, and now he was near some transfixing light that seemed to indicate its source. More than a drug, FK-506 was a beacon, a probe, and Starzl was determined to track it, no matter where it led. “Transplantation,” he began saying suggestively, “may just be a footnote to this entire story.”

If Starzl was insatiable before, this new larger mission drove him to flights of religious ferver. He seemed mesmerized by FK-506, enchanted, consumed. After two hours, nearing the end of the clinic and still pumping, he marched past the reception area and beckoned, like an evangelist who can’t stop pleading for souls after an all-night revival, the one or two patients still in the waiting room. “Come on back, somebody,” he waved. “I’ll steal a room.”

And yet he was also exhausted and prey to an unfamiliar lassitude. He had to push himself harder and harder just to keep up. In June, he took his first vacation in ten years, traveling with his wife, Joy, to Hawaii. He then went on to Japan to give a series of talks on FK-506, which, ironically, because of Buddhist strictures against violating corpses and an absence of laws governing brain death, is barely used in that country. Kidney transplants from live donors are sanctioned, but otherwise patients like those that Starzl routinely restored to a normal life with the aid of Japan’s first great drug were, as he saw it, cruelly and needlessly relegated to die for want of legal organs.

Returning from overflow lectures in Osaka and Tokyo, Starzl drove to his office the morning of Saturday, July 11—four days after Boger’s announcement of the Chugai deal—to plow through paperwork that had piled up while he was away. Halfway up to the second floor above the Pizza Hut, he collapsed. “The slightest movement,” he later wrote, “caused a cylinder of fire beneath the breast-bone which erupted like a volcano into [my] neck.” He dragged himself, inch by inch, to the second-floor landing, where he lay sweating and panting for an hour. He then did the same to get to the third floor. He eventually pulled himself into an upright position at his desk. For the next twelve hours, speaking breathlessly into two dictaphones, he answered three weeks of mail before stumbling downstairs and driving home.

The next day doctors discovered a 99 percent blockage in Starzl’s dominant right coronary artery and told him he risked a heart attack unless he had a bypass operation at once. He refused. He and his team were now rushing to finish more than forty papers in time for the international meeting in San Francisco in mid-August of the Transplantation Society, the same organization that had sponsored the conference in Barcelona. It was to be the most rigorous hearing to date on FK-506, made all the more pressing—and critical for Starzl—by Fujisawa’s failure to expand its clinical trials. Surgery was out of the question, Starzl said. He agreed to an angioplasty—expansion of the artery with a balloon-tipped catheter—but so reluctantly that he dislocated his shoulder resisting the straps on the operating table. “I needed,” he explained, “to be in San Francisco.”

Starzl returned to work two days later, but his stamina was diminished. All summer long, his conditioned worsened. By the time he reached San Franscisco, he was in constant pain, yet he dragged himself through the conference, winning adherents with his frank appraisal of the new drug. He flew home immediately afterward and was operated on the next day.

“Between the angioplasty and the San Francisco meeting it was kind of touch and go,” he conceded, returning to work the following week. “But it was worth the flip of the coin, I thought.”

•  •  •

If Starzl’s collapse had any effect on Vertex, it could only be construed positively: Chasing FK-506, the company stood to gain from any delay in its clinical progress, however small or ill derived. Even without knowing FKBP’s structure, Vertex’s chemists had begun making small molecules that bound to it almost as tightly as FK-506. Such binding is the first, primitive measure of a molecule’s effectiveness and in no way indicates whether it will become a drug: In the evolution of safe, active pharmaceuticals, good binders have the same relationship to good drugs as troglodytes to brain surgeons. But Boger knew that science is an incremental progression, won and lost by inches, and that every gain is worth optimizing. “This class of compounds would be so easy to manufacture,” he said about Vertex’s tightest binders. “I’d be happy to have an FK-506 look-alike eighteen months behind Fujisawa. Theirs is going to cost $1000 a gram; ours is going to cost $1000 a tankload. Even if our drug has the exact same therapeutic profile, I can wipe them out.”

His prediction, of course, precluded the use of structure-based design. Still without protein crystals, the company was several months away at least from having the kind of detailed information about FKBP that would influence what molecules it made; more—perhaps much more—time would be involved if FKBP turned out not to be the relevant target and the process had to be restarted with another protein. Though it certainly was conceivable, given Starzl’s resistance to comparison testing with cyclosporine and Fujisawa’s mishandling of the regulatory agenda, that Vertex would have a drug on the market less than two years after FK-506, it was unlikely that such a molecule would be designed in a way that proved Boger’s theories or brought him closer to his scientific goals.

Indeed, as a demonstration project for designing drugs, FK-506, with its devilish receptor and uncertain biology, was beginning to seem a much poorer choice than the company’s other prospect, HIV. As its name implies, structure-based design requires a reasonable certainty about how molecules are shaped and how they fit together. Researchers need to know the exact protein target, correctly configured, and the precise correlation between the way it “talks” with other molecules and the biological activity they hope to affect. With FKBP, in the summer of 1990, science had none of this information; with HIV protease, it had it all. There was no proof that if you inhibited the protein folding action of FKBP, you automatically had an FK-506-like drug. However, it was now all but certain that if you placed a well-designed molecule into the active site of HIV protease, you could cripple the virus’s reproductive ability and thus slow the spread of AIDS. You might not destroy the virus, but you would wound it gravely.

The problem with designing protease inhibitors was not how to make them but how to make them into drugs. Practically all of the research to date—“thousands of man-years,” observed Boger—suggested that the best way to shut down enzymes that cut and slice proteins is with peptides, small chainlike molecules made from amino acids. Peptides can be made to infiltrate and disarm aspartyl proteases, like HIV’s, with remarkable precision. But peptides are useless as drugs. They’re fragile, easily broken down in the gut. They can work ideally in a test tube, as almost every major drug company has shown, and yet if you ate grams of them, continuously, almost none would make it to their destination within cells, leaving science with a tantalizing challenge: to make molecules that look and act like peptides but are structurally different. Called peptide mimetics, such molecules are the grail of that considerable portion of AIDS research focused on inhibiting HIV protease.

In terms of pure science, there were many at Vertex who believed HIV protease was custom-made as a test project for the new company, an opportunity for them to succeed where screeners and medicinal chemists were doomed to fail. It had everything: known biology, a solved protein structure, established chemistry, well-developed model systems employing similar enzymes. More, Vertex seemed uniquely positioned to exploit it. Navia’s solution of the structure at Merck; Boger’s leading work in inhibiting aspartyl proteases; the company’s high-tech approaches, theories about small molecules, and integrated labs—all suggested a powerful edge, if not an outright franchise.

Yet Boger himself was less than convinced. From a business standpoint, AIDS—with its muscular competition between industry giants, each with scores of able researchers and a multiyear head start—was still no place for a small start-up company, he believed. Nor was it at all clear that nonpeptidal protease inhibitors worked as drugs—no company had ever made one—or that one of those behemoths, particularly Merck, wasn’t infinitely better suited than Vertex to produce the first such molecule. Boger proceeded cautiously. He authorized Roger Tung, a former Merck chemist who was eager to begin working on his own project, to make the Janssen compounds that had prompted Navia’s “hallucination” in March. He asked Dave Livingston to develop an enzyme assay so that they could be tested. But he stopped short of a full-scale call for crystallographic quantities of the protein. Notoriously hard to produce, in part because, by definition, it begins cleaving itself almost as soon as it’s made and because unlike FKBP it’s unavailable through tissue extraction, HIV protease can only be obtained through large-scale fermentation or by Herculean chemical synthesis. Vertex, Boger believed, hadn’t the personnel or facilities to spare for either.

Navia, understandably dismayed, pressured Boger for more enzyme, but he was chronically stretched by the demands of his career and the variety of his interests and was no longer the project’s main champion.

Mark Murcko was. Five feet seven inches, sandy-haired, stocky, with a thick mustache, Murcko, thirty-one, was the last scientist Boger had hired from Merck and the one he’d gone to the most trouble to get. He was also the most pivotal. A molecular modeler and computational chemist, he stood between crystallography, which churned out huge amounts of data with up-to-the-minute technology and computers, and chemistry, which made molecules with almost no technology more advanced than an espresso machine. If, as Boger said, Vertex was set up to generate and use more information in drug design than anyone else, Murcko stood astride that river of data at its widest, wildest point. The colossalness suited him. Fast-talking, glib, bristling with intelligence and enthusiasm, he had all the attributes (low-to-the-ground construction, broad-beamed confidence, gravelly appearance, hectoring banter, high schmooze quotient) of a catcher in baseball.

If Vertex was to design drugs, Murcko and people like him would design them. On his first day at the company, in May, he sat down at one of a bank of Silicon Graphics workstations that Boger had ordered in anticipation of his arrival and worked, his hands flying over the keyboard, until 3 A.M.—putting himself instantly in league with Thomson and Yamashita, with whom he became fast friends. He had kept up a similar pace ever since, programming, simulating molecular activity, setting up calculations that could “pin” Vertex’s computer network for days on end and would easily pin it forever if other researchers hadn’t also been clamoring to use it. Murcko’s phrase for this was “speculative science.” It’s impossible to know, of course, how molecules, which vibrate at billions of times per second, actually connect with one another. But with big, fast computers and three-dimensional computer graphics, it’s at least possible to venture educated guesses.

For instance, it’s understood that molecules, adhering to the laws of nature, favor those conformations that require the least energy. Thus there are “good” interactions—those that conserve energy—and “bad” ones—those that tax it. Similarly, individual atoms, which Murcko thinks of as “like Nerf balls, a little squishy,” smush or repel one another on the basis of an amalgam of almost unknowably small, yet measurable, forces—heat, from the formation and destruction of individual atomic bonds; gravity, exerted by subatomic particles; electric charge. In imagining how molecules bind, Murcko couldn’t predict—at least not without a crystal structure showing it—how a drug molecule and a protein might fit together, but he could speculate mightily on how, based on certain proclivities, they might want to.

Murcko is an expert at calculating the activities of atoms. He came to biochemistry, not as almost everyone else at Vertex had, through an interest in life science, but through a passion for computers and a curiosity about the physical aspects of chemical reactions—the “underlying reasons, the mechanisms, the properties of things, the fundamental forces.” The computers were first. As a twelve-year-old growing up in Fairfield, Connecticut, a Bridgeport suburb, in the early 1970s, he took a field trip to a science center that had the force, judging from the loving detail with which he retells it, of a revelation. By ninth grade, he was spending twenty hours a week after school and at night writing programs in the computer lab at Fairfield University, where he eventually went to college and majored in chemistry. He then went to graduate school at Yale. There, midway through his dissertation, he became interested in the molecular life of drugs.

“If you believe you understand something fundamental about the way molecules interact and you want to find out if you do understand it,” he says, “now, pick a difficult system: Imagine a complex system where your confidence can be tested.” Practically by default, proteins, the most complicated of all molecules, became that system. In the fall of 1985, Murcko, who knew nothing about protein structures or drug design, began applying to drug companies, which not surprisingly were mystified by his interest in them.

“I would go into an interview with a representative of a Pfizer or a Glaxo or a Lilly and they would look at my résumé and say, ‘Well, you’ve obviously done interesting things but you don’t seem to have the kind of background that would lead somebody into the pharmaceutical industry.’ And then the representative from Merck came, and it was Joshua. And he starts telling me about the modeling that he himself had done. . . . This was incredible to me. He was the only one who seemed to appreciate that you didn’t have to be a medicinal chemist to contribute to this work.”

As a pioneer in the use of molecular modeling, Merck recruited Murcko avidly, though in fact Boger had already persuaded him that there was no other place to be. Joining the company in the spring of 1987, just as Boger was beginning to organize Merck’s rational drug design effort, he was assigned to its West Point labs. With Boger now in Rahway, the two never worked together directly, but Boger continued to track Murcko’s career, and shortly after he left to found Vertex, he called him. To that point, Boger had failed to win only one scientist whom he’d set his sights on, but Murcko was not easily converted. After persuading him to go to Merck, Boger was now trying to pull him away, and Murcko, his catcher’s crouch set, dug in. It took three months and a hefty draft of founder’s stock—three months in which Vertex’s senior scientists were regularly platooned to call Murcko every other day—to extract him. For Boger, it was a tour de force, a major victory against Merck’s best hands, a personal triumph, although one that would also leave him troubled that he had perhaps ruffled Mother Merck one too many times. Seeing the battle for Murcko in Machiavellian terms, he worried about the cost of such a prize.

Says Murcko: “Joshua has the unfortunate situation of having a psychiatrist and two lawyers as brothers, and that, combined with his own scheming, complex, twisted intellect, causes him to see plots and patterns where there probably are none. I can’t swear that there’s nothing to that notion, but it strikes me as a little paranoid.” On the other hand, Merck was noticeably put out by Murcko’s decision. Unlike previous defectors to Vertex, who were given up to a month to leave—and perhaps reconsider—Murcko was told initially to be out in four days. Not wanting to waste several hours driving, he flew to Boston the day he finished at Merck, leaving his wife, Kathy, a teacher, to complete the school year back in New Jersey.

Murcko had expected to begin work in immunophilins. “No one ever said to me in an unequivocal way, ‘Yes, we have crystals and we’re, oh, fractions of an inch from getting a refined structure,’ ” he says, “but the clear impression was that Vertex was very far along on that path, and that turned out not to be true.”

And yet he was far more dismayed by the alternative. For the previous six months, he’d been working eighty to one hundred hours a week designing HIV protease inhibitors at Merck. With major-project status, HIV was discussed in rare detail at regular Monday morning meetings, and Murcko had attended those sessions. He was privy to all Merck’s strategies, leads, data, talking points—and legally sworn not to disclose them. Assuming, as Boger himself had, that Vertex would never work in AIDS, he had comforted himself that he would be free of any conflicts in his new job. Yet now, asking Navia on his first day what other areas the company was considering besides FKBP, he was stunned to learn that it had a feeler project in HIV.

“I thought I was going to have a heart attack,” he says. “I don’t know what a heart attack feels like, but this was a heart attack. Here I was, fresh out of Merck, which I hadn’t exactly left on the best of terms, and this icy, steel ball just knots up in my stomach, and I start twitching and drooling as it turns out they want me to start modeling right away on HIV. I couldn’t do it. I felt as if anything I did had the potential for being a legal problem.

“I felt really uncomfortable. But I talked to Joshua and other people here and what I ended up deciding was that it was a situation where I had to use my best judgment on a case-by-case basis. It was unfair to me to simply say, ‘Well, since you’ve worked on HIV somewhere else, for the rest of your life you can’t work on it.’ ”

Storming Vertex’s computers—and still with no structure of FKBP to work with—Murcko hurled himself into modeling HIV-protease inhibitors that were determinedly unlike any he’d designed, or seen, at Merck. He placed himself in a kind of deep denial, “lobotomizing” himself to forget what he knew, a tortured bit of self-surgery. “At one point, one of the chemists showed me an idea for a compound, and I knew before he was done talking exactly how to synthesize it and how it would perform in several assays. I put my hands over my ears and walked out. It hurt. That really hurt.” (Later, Tung concluded that Murcko’s repeated self-disqualifications cost the company up to six months.) Meanwhile, by mid-July he was also immersed in a series of modeling experiments using the Janssen compounds and another potent antipsychotic, Haldol, that was being promoted as a powerful inhibitor of the enzyme.

From the outset, the work proceeded slowly. Now that he had committed himself, Murcko began haranguing Boger—“Joshua the indeflectable,” he called him—about the need for protein. He reminded him incessantly that without real structural data on either project, he couldn’t begin the actual process of designing drugs, and that Vertex, already behind on HIV, could only catch up by exploiting such information. Meantime, he ran endless simulations, devouring masses of computer time. Asked during this period how much computating power he considered optimal, he deadpanned: “Infinite.”

His chief ally was Yamashita, with whom he bore the same ambivalent relationship as Yamashita had earlier had with Thomson, when Thomson was still trying to isolate FKBP. Bound by a common belief in structure-based design, brought together by long, late-night theoretical conversations that often stretched until morning, depended upon equally by Boger, they became close confidants in Murcko’s “fraternity of grief.” And yet Murcko was stalled in his own work because of delays in crystallography. He didn’t want to pressure Yamashita—Mason pressured himself enough—but like the last runner in a relay, he was forced to watch in exasperation each time crystallography stumbled and fell behind.

On a Thursday night in late July, Yamashita mentioned to Murcko that he and Navia had grown several new crystals with Thomson’s protein and that he was planning a third diffraction attempt early the next day before the labs filled with onlookers. At home later that night, Yamashita logged onto Vertex’s computer at about 1 A.M. to monitor some calculations. There was a memo from Murcko in his directory. It concluded with a snatch of dialogue from Star Wars, which Murcko had seen a dozen times and could quote at length:

OBIWAN: Vader was seduced by the dark side of the Force.

LUKE: The Force?

OBIWAN: An energy field produced by all living things. It surrounds us, penetrates us, it binds the galaxy together. (much later)

HAN SOLO: I been all around the galaxy, kid, and I seen all kinds of strange things, but I’ve never seen anything to make me believe in some all-powerful “force” that controls everything.

Publicly, Murcko thought it best to humor Yamashita, though privately he was less sure. “If that crystal doesn’t diffract,” he said earlier that night, “a lot of people around here are going to be crushed.

“I would watch Mason, follow him around, take his car keys.”

•  •  •

Thomson could work in a forty-degree cold room all morning in a T-shirt, but Yamashita, having spent the last fifteen years in Hawaii and Los Angeles, hated cold. He bundled up in a stadium coat. Even before Murcko’s inadvertent reminder, the steel-walled meat locker where he and Navia grew many of their crystals had always made him think of Darth Vader, its unseen compressor exhaling nightmarishly like a cancer patient dying into a microphone. At 7 A.M. Friday, midsummer, Yamashita stood blowing into his hands amid its steel racks and Styrofoam coolers and boxes of supplies from the Spectrum corporation—“laboratory products for the third millenium”—preparing to mount, for the third time, presumptive crystals of FKBP.

It was painstaking work, and Yamashita was wise to ward off any sudden shivers. The crystals, still tiny though substantially larger than those he and Navia had first grown in April, floated luminously in droplets of mother liquor: Under the microscope, they looked like coffin-shaped diamonds suspended in water. A protein skin on the surface of each drop needed to be peeled back without disturbing the crystals. A slip could easily set him back weeks, a reversal he preferred not to consider.

The unintelligible but hypnotic lyrics of Ireland’s Cocteau Twins blared through his headset as Yamashita hovered over the double-barreled eyepiece of his microscope, steady as a gem cutter. He clasped in his right hand a small plunger that was attached at one end to a glass capillary tube, 0.5 millimeter in diameter. Slowly retracting the molecular skin with the end of the tube, he then poked it into a clear area of the liquid. Finding a spot near a well-formed, isolated crystal, he subtly withdrew the stopcock with his thumb and forefinger until the crystal rose halfway inside the tube. Sealing one end of the tube with hot wax, he inserted into the other a hair-thin filament wick, which he dipped in mother liquor before sealing that end as well. He then repeated the process two more times with two other crystals that had been grown in different conditions and were more lozenge shaped. Together, the three toothpick-sized tubes looked like tiny glass barbells with tear-shaped orange wads of gum on the ends, a barely visible silverfish floating amidships in each.

“There,” he said, referring to the crystals. “That ought to keep them happy.”

Stung by the pain and embarrassment of his last diffraction attempt, Yamashita had worked hard in recent weeks to control his feelings, both high and low. He affected a seasoned and, given how desperate he was to succeed, anomalous midrange view, an I’m-OK-you’re-OK emotional competence that fit him like an oversized suit. Still, pleased by the success with mounting the crystals, Yamashita was buoyant as he entered the X-ray lab. Murcko had come in early to join him, and Navia was once again out of town: at the very least, if the crystals didn’t diffract, he wouldn’t have to repeat June’s excruciating psychodrama.

“That’s very good,” Yamashita said, observing several bright spots on the monitor, “I’m impressed. Manuel grows very good crystals.” It was just the signal he was looking for, the one that had evaded him both times previously, irrefutable proof that the crystal was protein. But there was bad news as well: If the number of spots indicated that he had gotten a protein lattice to diffract with, their distance from an established reference point at the left of the screen also showed that the grid was of a poor quality. Yamashita might be able to solve the molecular structure of the protein with such a crystal, but his data would be only marginally reliable. Would it be the correct structure? He wouldn’t know. Could Vertex depend on it as a template for drug design. Probably not. “This is actually the toughest result we could have had,” he confessed to Murcko. “It means we’re going to have to work on it, but it’s going to be very hard to work on.”

If the outright failure of his previous diffraction attempts had left Yamashita distraught, the ambiguity of this one was far kinder. It was a start. He knew that. And though it was a disappointing one, he also knew it meant the structure could be solved. He had heard through the grapevine that Merck, too, had crystals but no structure, which meant that though he and Navia might be behind, they were still in the race. Despite his concern about their quality, the pinhead-sized crystals he had mounted on the beam that morning contained information he now knew to be worth keeping, and he began instantly collecting data. He was calm, assured, relatively composed, if privately worried about what lay ahead.

Boger, arriving by 8:30 and immediately visiting the X-ray lab, was ecstatic by comparison. “Great,” he said, hearing the result. He paraphrased Archimedes. “Give me a place to stand,” he said, “and I’ll move the world.”

With the discovery of information he considered most vital for designing a better drug than FK-506—Vertex’s long-sought-after “place to stand”—now being only a matter of time, Boger felt he could move the world. Unlike Yamashita, he wasn’t primarily interested in who got the information first and received credit for it, though that clearly mattered to him. What mattered most was that it could be gotten. It had taken Navia less than three months from the time Merck crystallized HIV protease to solve its structure, and he had vowed that he and Yamashita would take even less than that with FKBP. By Thanksgiving, perhaps, they would have the native structure of the enzyme. After that, they would begin feeding Murcko a succession of structures, first FK-506 bound to active site, then Vertex’s own proprietary compounds—structures that would show how changes in Murcko’s “fundamental forces” affected the biological activity of the molecules.

It was all laid out before them now. Thomson’s hellish struggle with the protein, the difficulties in getting crystals, Murcko’s impatience—in the weeks ahead, Boger watched them dissolve in the face of what was from here on, a vastly more predictable, if not always orderly, process. Getting diffraction-grade crystals was the prime rate-limiting step in crystallography. Now that Vertex had them, its scientists would solve the structure. They would make better molecules. Indeed, they already were. By mid-August, while both Boger and Aldrich were in Japan working out the final details of the Chugai deal, the chemists received word from Harding that one of their molecules was now one hundredfold less potent than cyclosporine in switching off T cells. Like the binding and enzymatic assays that showed Vertex’s compounds behaving much like FK-506, regulating cellular activity in test tubes was still a primitive measure. And Vertex’s molecule was weak—too weak to become a drug. But making a molecule that was active at all was another major step. Vertex now had compounds that appeared to be patentable and that had a desired biological effect—a drug in form, if not content. And Vertex was gathering information about its presumed target, information that would help improve it.

Boger, returning from another trip to Tokyo, was as confident as the scientists had ever seen him. “If in the next six to nine months we can improve by a factor of thirty in cells, six months after that we have a preclinical candidate,” he said. “A year after that we have a clinical candidate. That’s three years from start to finish—at least two years ahead of where I thought, in my most wildly optimistic projections, we could be.”

It had always been implicit in Boger’s story that time is money in the drug industry and that structure-based design, in addition to all its other virtues, would vastly increase the speed (thus reducing the cost) of discovering new drugs and bringing them to market. Says Murcko: “Molecular modeling is largely a game of making mistakes faster than your competitor can”—screening ideas, quickly and efficiently, at a keyboard, rather than laboriously scouring dirt samples by sifting through foul-smelling, labor-intensive fractionation broths. Satisfied that Vertex was setting a record pace in all departments, Boger departed on August 15 with a foot-high stack of publications under one arm and gripping a Macintosh laptop with the other and started out on his first vacation since he and Kinsella began plotting the company. He and Amy took the three boys to North Myrtle Beach, slowing down to make the trip in two and a half days, for their first time off together in more than twenty months.

•  •  •

There was nothing to stop him now. In less than a year, Boger had accomplished all that he had set out to and more. He had known that before Vertex could make a drug it had to make a deal, and he and Aldrich and Schmidt had now brought in the deal of the decade, or so he could argue. The company had protein by the Thomson Unit, crystals that diffracted, and patent applications pending on novel, tight-binding molecules that showed discrete biological activity. It had the germ of a second project, and companies were interested in that, too. Perhaps most significant, it had a powerful ally, Chugai, contributing money, expertise, and credibility—“benedicting,” to use a word of Boger’s, the company. “Together we now have the biggest research program in the world in this area,” he told the scientists. “We had the best, but not the biggest. Now we’re bigger than Merck, bigger than anyone.” With the exception of permanent labs, Boger had delivered on every one of his major promises, no matter how grandiose or improbable.

It was time to assert the prerogatives of his new station: “the night of the long knives,” Aldrich called it. On September 26, 1990, within hours of receiving the last signed legal documents from Chugai, Aldrich sent two letters to Harvard by bonded courier, backed up by certified mail. The first was to Martin Karplus, a founding member of the SAB, who was attempting to solve the structure of FKBP with protein from Schreiber’s lab. The second was to Schreiber. They were termination notices. Aldrich wrote officiously, without explanation, that Vertex now considered its relationship with both of them untenable and that as of December 31, 1990, they would no longer be associated with the firm. Vertex, he wrote, would buy out their unvested stock, although they could keep what they already owned—75,000 shares apiece. Though the letter didn’t say so, the stock was the price of Vertex’s freedom, all but guaranteeing that Schreiber and Karplus wouldn’t complain too loudly or counter with lawsuits.

For Boger, the worst part of an otherwise charmed year, his yawning distrust of Schreiber and the need to debase himself with Harvard, the one matter that had persistently eluded his control and provoked his anger, was over. The negotiations with the Harvard patent office were moot, discontinued, as what Schreiber did with his compounds and protein were no longer Vertex’s affair. If Schreiber was a loose cannon, he was no longer on the company’s deck. Where he pointed, as with all things Boger couldn’t control, ceased instantly to matter to him, or so he would make it seem.