The Demon Under the Microscope, page 19
Use it he did. He read every German journal in medicine and chemistry, studied business publications, tracked the patent listings, and spent eighteen hours a day compiling a master file of French pharmaceutical needs, German exports, and French imports. Every year, instead of vacationing, he attended German scientific congresses and the trade exhibitions that followed, where he could see the latest products and equipment.
France was a backwater by comparison. French medical chemistry was less an industry than a glorified hobby. The model was Louis Pasteur, a hero of the people, a selfless chemist working for the most part on his own, saving individual lives, taking a personal interest.
Fourneau was determined to bring the best parts of the German approach to France. After starting at Poulenc, he quickly discovered a synthetic alternative to cocaine, widely used (and increasingly abused) in medical procedures as an anesthetic. Fourneau’s new molecule, Stovaine (a play off his own name; in French, fourneau means “furnace”), brought the young Frenchman to the attention of the great Duisberg himself, who offered him a job running Bayer’s research operation in France. Despite the munificent salary Duisberg dangled in front of him, Fourneau turned it down.
After the excitement of Ehrlich’s introduction of Salvarsan, Fourneau accepted an offer to start the first laboratory for chemotherapy at the Pasteur Institute, married into an aristocratic Parisian family, and began implementing his ideas for competing with the Germans. Just after World War I, he even did a bit of spying, acceding to a request from the French war minister that he travel to Germany, review German industry, and report privately on the status of chemical research. He visited Bayer, was struck by the size of its humming factories and impressed by the investment Duisberg was making in research. Fourneau realized that Bayer was funneling a fortune into new-product development to make up for its war-related losses.
There was no way for the Pasteur Institute to compete—at least not directly. The two operations differed in more than sheer size. Bayer was private industry, the researchers all male, the chemists generally German, the work often secret, and the goal clearly profit. The Pasteur Institute was more a diverse enterprise based on the selfless model of the great founder, whose remains were then (and still are today) enshrined in a marble crypt in the basement of one of the institute’s buildings. Pasteur worked for the good of mankind, not for profit. His institute attracted bright young idealists from around the world, who worked in a community of scholars rather than a factory for discovery. Many of them were women. It was a collaborative, cooperative, relatively open atmosphere; there were romances at the Pasteur Institute, and gossip, personal enmity, cranky individualism. Marriages were proposed. At least one Pastorien (as they called themselves) challenged another to a duel.
Fourneau’s chemotherapeutic laboratory comprised a typically talented, polyglot group. It included a rarity in the world of chemistry in those days, a handsome married couple, Jacques and Thérèse Tréfouël (Jacques was Fourneau’s number-two man, the manager who ran the laboratory day to day; Thérèse a skilled bench scientist). There was Federico Nitti, a son of a former prime minister of Italy who went into exile when the Fascists took over. There was Nitti’s sister Filomena, who was in love with another of Fourneau’s group, Daniel Bovet, a brilliant Swiss chemist who would later win the Nobel Prize. It was, one member recalled, a wonderful place to work, a “warm-hearted laboratory.” Everyone who worked for Fourneau admired him and respected him, although he was not a typical Pastorien. Longtime laboratory directors and higher managers at the institute were often affectionately called “Père” (father) by the younger investigators. Fourneau never was. He was always “the Master” when his lab workers talked among themselves and “Monsieur Fourneau” to his face.
“The Master would often come to the counters and scrutinize all things with a look devoid of indulgence,” one of them wrote. “He would sometimes take a flask from the chemist’s hands, take out of his pocket a glass agitator he always carried, then would accelerate the reaction by heating, cooling, agitating, his right arm in the air, scrutinizing the mixture. Often the reaction would occur and, without a word, Fourneau would put the container down and leave.” Even though he never raised his voice, he was able to exert total control through his commanding presence. Sometimes the Pasteur Institute would send him problem employees, whom he had no trouble managing.
The relatively limited funding available at the Pasteur Institute did not stand in the way of discovery. Fourneau’s lab was the first to make an oral arsenical, for example, for use against syphilis. But an occasional success did not mean that his team was equipped to compete with the Germans in finding new medicines. So they worked on Fourneau’s other strategy, that of finding French versions of new German drugs and getting them quickly to the marketplace. Their first success came in 1925, when they broke the formula for Germanin. Bayer moved swiftly to repair the damage by initiating joint sales efforts with the French, but the company lost at least part of a lucrative market—and Hörlein lost whatever trust he had for Fourneau.
So when Hörlein received a letter from the Pasteur Institute requesting a sample of Prontosil “for experimental purposes,” he replied positively but asked in return for a meeting in which he and Fourneau could discuss sales and marketing of the new medicine in France. The meeting was held, the participants were polite, but nothing was resolved. The French went to work deciphering what they could of the German patent application for Prontosil. Very soon they found a way to replicate the molecule, not precisely, but closely enough to work. Constantin Levaditi, a longtime Pastorien, confirmed that the French chemical worked in animals almost as well as the Germans reported for theirs. It all happened very fast. Levaditi published his results within three months of Domagk’s paper, and the Pasteur Institute began distributing the drug to French physicians, who reported “spectacular results” using it against erysipelas and other streptococcal diseases. The recipe went to a French chemical firm that quickly began manufacturing and selling it under the trade name Rubiazol, for its red color. When he heard the news, Heinrich Hörlein started fuming. He had spent eight years and millions of marks discovering, developing, and bringing Prontosil to market. Within three months of its announcement, Fourneau had robbed him of at least some of his profits.
Fourneau’s group continued to study the drug. In July, Jacques Tréfouël gave Bovet and Nitti the task of testing Prontosil/Rubiazol as well as a series of the Pasteur Institute’s own chemical variations for effectiveness against strep in mice. Between July and November 1935, working around the usual summer holiday (during which the institute virtually shut down), they perfected their animal-test system. Bovet and Nitti found, as Domagk had found before them, that much in mouse experiments depended on the particular strain of Streptococcus being used; they got variable results until they started using a particularly vicious strep culture they obtained from a victim of childbed fever at the Hôtel Dieu. Against this germ they tested eighty chemicals, including many azo compounds. Their findings fully confirmed the German results.
Then came something entirely unexpected. “It happened on November 6, 1935,” Bovet wrote. “We received a large number of mice (forty) and placed them in glass jars in groups of four. They all received an intra-peritoneal injection of highly virulent streptococcal culture. One group was kept as a control group and the other received a sufficient dosage of the coloring agent described by Bayer [Prontosil]. After that we treated seven groups orally with the products we had synthesized in the lab. But I only had seven new products and we had an extra group of four mice. Why, I asked, not just try the product common to all these products, para-amino-phenyl-sulfonamide?” The long technical string at the end of Bovet’s sentence is a chemical name for pure sulfanilamide, the side chain the Germans used to awaken their azo dyes.
Bovet’s casual decision would change the history of medicine. Everything the Germans had been testing at Bayer involved linking sulfa to something else, almost always an azo dye. Everything the French had been trying was done the same way. Testing pure sulfa alone was something the Germans had nosed around—they had tested molecules that were close to pure sulfa, yet they always appeared to just miss the exact molecule itself—but never quite accomplished. Then came four extra mice and a bit of handy chemical. Bovet and Nitti separated the groups of mice; marked them with stains of picric acid, a yellow spot on the tail for one group, patterns of yellow spots on the sides or back for others; infected them with strep; injected them with chemicals; and observed them on their bedding of dried oats. Each dead mouse received a cross in the lab journal, each one who lived got a V. The next day the control group, the unprotected mice, were dead or dying, as expected, as were almost all of those treated with the experimental chemicals. Only one of the new French azo dyes seemed to be working. Those mice treated with Prontosil were alive, also the expected result.
The surprise came in the last jar, which contained the four extra mice, the group treated with pure sulfanilamide. They were not only alive, Bover recalled, but were “doing great.” It was one of those results that no one quite believed until retests proved the point—retests that Bovet and his coworkers performed immediately after informing Fourneau. It was difficult to accept, but it appeared to be true: Simple sulfanilamide, a colorless, common, unpatentable, off-the-shelf chemical used by the pound in the dye industry, was as potent a medicine as the German wonder drug Prontosil. The implications were immediately clear to the bright young men and women in Fourneau’s laboratory. “As we wrote the last ‘V’ on the report, we had already realized that the future belonged to ‘colorless products,’” Bovet wrote. “From that moment on, the German chemists’ patents had no more value whatsoever.”
The finding meant much more than money. It offered explanations, opening the way to understanding why attaching sulfa to many types of azo dyes resulted in an active medicine, while dyes without sulfa were far less active, if they had any medicinal value at all. It pointed toward an answer to the mystery of why Prontosil worked in live animals but not in the test tube, with the French team immediately hypothesizing that sulfa had to be released from the rest of the Prontosil molecule in order to become active. That could happen in the body of an animal, where enzymes in the body could split the Prontosil molecule into two pieces, releasing pure sulfa as the medicine. The dye did nothing but stain the skin. The sulfa cured. In the test tube, there were no enzymes to split the Prontosil, and the sulfa would not be released. This discovery, which the French explored and confirmed, opened up an entirely new field of medicine, in which substances could be “bioactivated” in the body.
Most important, however, was the discovery that the world’s most effective antibacterial medicine was also among the simplest ever found. Everyone had been searching through all these complicated dyes, tinkering around the edges, while the real power was in a colorless add-on. As Bovet later put it, the Germans’ complicated red car had a simple white engine.
Fourneau was elated, but he refused to take credit for the finding. A few days after confirming their first experiment, Bovet and Nitti brought him a copy of a report they had drafted for publication. They expected that he would accept his name where they had placed it, at the top of the list of authors. He surprised them by crossing his name off entirely. “The reasons—or maybe the feelings—which made Fourneau do that were hard to understand for those close to him,” Bovet wrote. “Was it a generous gesture towards younger colleagues, in order to help their careers? Or was he being deferential to and respectful of H. Hörlein and the scientific staff of Bayer, and worried he might hurt their scientific prestige by this publication?”
The work appeared under the names of the Tréfouëls, Bovet, and Nitti, in the Comptes Rendus des Séances de la Société de Biologie (Reports of the Society of Biology) at the end of 1935. They kept it brief, not because they thought their paper was unimportant but because they knew that the secretary of the Société de Biologie, in order to pack in as many communications as quickly as possible, measured each submission with a centimeter ruler and discarded whatever was too long.
The French team then set to work digging further into their discovery. Now that they knew where the active portion of the molecule was, they could move away from dyes entirely and attach sulfa to many different side chains at different points, then measure the effects. They were looking for ways to make sulfa stronger or safer and also for clues into how it worked its magic. They made many different sulfa-containing compounds that varied terrifically in terms of chemical personality—solubility, surface tension, melting point—but many of which stopped strep. If the sulfa portion was lost or altered, however, the effect went away. They made a number of variations on the sulfa molecule as well, moving pieces around, seeing what happened. All of their results were quickly published. Secrecy was not an issue.
It was wonderful that this powerful, inexpensive medicine was now available, but for a year after the Pasteur Institute announcement, no one marketed it seriously in its pure form as a medicine. Because it was not patentable, it was difficult for major chemical or drug firms to see a way to make much of a profit from it. It was not until months after the Pasteur group’s first publication on sulfa that the president of Rhône-Poulenc, an industrial supporter of Fourneau’s laboratory, visited the Pasteur Institute to hear about it. After talking with the researchers he decided to launch Septazine, a variation on pure sulfa that he felt was different enough to allow patenting—and hence profits. Septazine reached the marketplace in May 1936. In June a British group confirmed the French findings. There was still no word from the Germans. “We at the Pasteur Institute were extremely anxious to find out the reaction of the Bayer researchers to our discovery,” Bovet said. But no reaction was forthcoming—at least not publicly.
CHAPTER FOURTEEN
AS SOON AS it appeared, the first paper by the Pasteur Institute team was translated into German and circulated within Bayer. According to one of the German chemists, it “struck like a bombshell . . . . The excitement caused by this paper can only be appreciated by those who witnessed it.” The revelation that the sulfa side chain was the active ingredient in Prontosil rather than the dye set off a string of responses that started with denial: Certainly there had to be something wrong with the French results. After Klarer and Mietzsch provided Domagk with pure sulfanilamide and his tests showed that it worked—not only worked but, on a gram-for-gram basis, was twice as effective as Prontosil—the Bayer team moved to confusion: Mietzsch and Klarer in particular were stunned. How could the medicine work if it was not a dye? This undid everything Ehrlich had taught. Then came recrimination: How, after three years of research, had they missed a critical discovery that the French had made within a few weeks? Then finger-pointing: At least one observer within Bayer noted a “confrontation between Mietzsch and Klarer on the one hand and Domagk on the other,” in which, presumably, each side tried to assign responsibility for the embarrassing French finding.
Finally the Germans found their way to acceptance. Something had gone wrong, their group had executed a major scientific pratfall, but it was done, and they had to deal with the results. Who was to blame? No one and everyone. The chemists had provided Domagk with hundreds of sulfa-containing chemicals, almost all of which worked. Within three months of discovering Prontosil, they had sent him Kl-820 and Kl-821, molecules in which sulfa was not attached to an azo dye. Domagk, for his part, believed that he had run his tests flawlessly. Almost every time they tested an azo dye with a sulfa side chain, it killed strep; almost every time it did not, the effect was absent or greatly reduced.
Perhaps there were enough confounding results—tests in which an occasional azo dye without sulfa proved somewhat effective, usually on bacteria other than strep; tests in which, for no apparent reason, a sulfa-containing azo dye that should have worked did not—to throw them off, keep them focused on azo dyes instead of sulfa.
Because many documents at Bayer (including all records of upper-level administrative deliberations) are not publicly available, it is impossible to know exactly what happened. In hindsight it appears that the German laboratory data pointed clearly toward the French results and that the Germans, at least to some degree, knew it. The Bayer chemists had followed their intuition and, just like the French, within months of discovering Prontosil had developed molecules like Kl-820 and Kl-821, compounds that contained sulfa but no azo dye. They gave the molecules to Domagk, Domagk tested them, and a confirmed, positive set of results from at least one of these non-dye molecules was ignored, dismissed, or buried within the company. The team seemed well on their way to finding what the French found. Then they stopped.
Or did they? As early as November 1932, within weeks of first making the molecule that became Prontosil, Klarer in his internal reports was already referring to the sulfa side chain as “the active group” and adding that Mietzsch was helping him to produce sulfa in large quantities. Weeks before submitting the patent for Prontosil, Klarer noted that he had prepared at least two compounds in which the sulfanilamide group was attached to non-azo molecules, adding that he planned further work on them. Domagk’s laboratory notebook includes no mention of any tests on those substances.
