Miracle Cure

by William Rosen

  —

  In September 1941, Florey, having completed his proselytizing in the United States, returned home to England to continue his own investigations, leaving Heatley behind to work at the Northern Lab. He was, in consequence, not on hand when the tiny vial of brown penicillin powder that saved the life of the septicemic Anne Miller in March 1942 had, in some sense, marked not just the birth of the antibiotic age, but also the end of European, and especially British, preeminence in the field. The following month, Florey opened a package from the United States expecting to find his long-awaited kilogram of penicillin. It contained only five grams; less, in fact, than the amount that Merck had rushed to New Haven Hospital to save Mrs. Miller’s life. The Americans were keeping everything they could produce for their own research; despite Florey’s tireless campaigning during the summer and fall of 1941, British research on penicillin still relied on what Britain could produce at home.

  While Heatley and Florey were traveling across North America drumming up support for their original request—that elusive kilogram of penicillin needed back in Oxford for clinical testing—the Dunn School had remained the most important antibiotic research center in the world, though still reliably dysfunctional. Chain had learned that his colleagues were leaving for the United States only when he noticed their readied luggage. Not only had he not been invited; he hadn’t even been informed. He was furious.

  The rationale—that the purpose of the trip was to promote industrial manufacture of the drug, which was Heatley’s special province, not to discern its structure, which was Chain’s—was strong, but unpersuasive. Though Florey understood, earlier than most, that penicillin would never realize its potential either as a therapy or as a scientific breakthrough until its structure was completely understood, he had misunderstood his émigré colleague. Chain was even more avid for recognition than Florey, and feared that he would be cheated out of “his” Nobel Prize—yes; he was already dreaming of a call from Stockholm—unless he promoted his involvement in the discovery.

  He wasn’t just sensitive to slights concerning his scientific importance. He was equally sensitive (probably more understandably) about his status as a Jewish émigré. Though Chain could not yet know the fate of the mother and sister he had left behind—both would die in 1942, likely in the Theresienstadt concentration camp—he was well aware that the world had rarely been so dangerous for Jews. In his own words, “One could not trust any undertaking given by Florey. I gave in in the end . . . to avoid any action which could provoke latent anti-Semitism, which was very widespread. I had to bear in mind the fact that the Jewish community would not be very pleased if controversies of any kind with anti-Semitic overtones came into the open. . . . I have always considered, and still do, that Florey’s behaviour to me in the years 1941 until October 1948, when I left Oxford . . . was unpardonably bad.”

  During the first months of 1942, Chain and Abraham finally perfected the chemical technique required to produce a stable penicillin salt. The tiny quantities of the drug still being produced by Heatley’s jury-rigged factories were producing better and better results on patients from the Radcliffe Infirmary, and, at the instigation of Winston Churchill, Britain’s drug companies were finally stepping up to the challenge. By the fall of 1942, Kemball, Bishop was sending 150 gallons of broth to Oxford every ten days, which allowed the Dunn—by far the most productive penicillin “factory” in Britain—to supply sufficient research material to Ethel Florey during 1942 and well into 1943.

  Nonetheless, the other Dr. Florey recognized that the center of gravity in penicillin production was shifting irrevocably to the United States, the only place with sufficient penicillin for clinical trials. Florey’s original research program had consisted of three key objectives: finding more potent penicillin strains, increasing manufacturing capacity, and unlocking the compound’s chemical structure. Two of them were rapidly outgrowing England. Peoria was well on its way to identifying the most promising strains of Penicillium, and—as will be seen—American industry was about to be enlisted in manufacturing the precious stuff in quantity. The third goal, however, remained on the Dunn School’s research agenda: discovering penicillin’s elemental components, its structure, and its mechanism of action—how it worked.

  As can’t be repeated too often, it’s a great deal easier to discover a compound’s ingredients than to understand how they fit together. Fairly rudimentary tools, for example, can identify the presence of sugar, flour, butter, and eggs in a cake; they will tell next to nothing about how the chains of proteins, carbohydrates, and fats link together. And they tell nothing at all about oven temperature or baking time. So, too, with the chemistry of penicillin. The ultimate goal of all the researchers was a method of producing penicillin in a factory, rather than growing it in fermentation tanks, to ensure both consistent quality and increased quantity. It was the same way that Ehrlich created Prontosil: analysis, then synthesis.

  This is not to imply that analysis of penicillin was a trivial matter. Teams on both sides of the Atlantic struggled for more than a year to establish the type and number of elements that composed the penicillin molecule. As early as 1940, the Dunn team had demonstrated that, like so many organic molecules, penicillin always contained carbon, hydrogen, nitrogen, and oxygen. But from the beginning, sulfur had also been appearing in samples of the compound. Some samples, though not all. The inconsistencies could have been due to mistakes in the chemical analysis, or the fact that sulfur was just another impurity that was introduced during the growth or extraction processes.

  Or, possibly, sulfur might have been central to its effectiveness. Chain and Abraham, along with other organic chemists from Oxford’s Dyson Perrins Laboratory, were determined to solve the sulfur puzzle. By then, Chain, using a technique known as partition chromatography on silica gel, was able to produce penicillin salts that were between 70 and 90 percent pure, which he could then oxidize—that is, turn into acids. Although what was left behind when penicillin was broken down to simpler compounds—formally “degradation products”—weren’t of any therapeutic value, they did have one extremely important aspect: The degradation products, primarily penicillamine, penillic acid, and the penilloaldehydes, were all crystals.

  Crystals could be analyzed.

  As far back as the seventeenth century, the polymaths Robert Hooke and Johannes Kepler were independently speculating that crystals like gemstones, common salt, and even snowflakes were structurally similar, each with visibly flat faces meeting at regular angles, all in a repeating series. As scientists realized that all matter is composed of atoms, they also concluded that macroscopic crystals reflected a microscopic structural similarity: The atoms that composed them had to be organized in some regular, and therefore decipherable, structure. Structure would explain, for example, why pure carbon can take the form of both the soft graphite in a wooden pencil and a diamond.

  Even as late as the 1920s, crystallography was exiled to the disreputable Department of Mineralogy, which was mostly about taxonomy: classifying all of the earth’s naturally occurring crystals, a task on which the Oxford mineralogist Thomas Vipond Barker spent his career. In 1942, one of his students began working with Chain and Abraham on the analysis of penicillin. Her name was Dorothy Crowfoot Hodgkin, still, as of this writing, Britain’s only female science Nobel laureate.

  When Dorothy Mary Crowfoot arrived at Oxford’s Somerville College in 1928—she didn’t marry Thomas Hodgkin until 1937—the science of X-ray crystallography was barely a decade old, though the principle of diffraction had been known for centuries, since the Scottish mathematician James Gregory noticed that rays of sunlight split into a spectrum of colors when passing through a bird’s wing; light-wave spectroscopes had been used to analyze chemical elements ever since. X-rays, on the other hand, which had been discovered by the German physicist Wilhelm Roentgen in 1895, travel in waves that are thousands of times shorter than light waves. This meant that the p
risms and other tools used to split light waves were too coarse a filter, about as useful for diffracting X-rays as a fishing net was for catching gnats. In 1912, though, another German physicist, Max von Laue, proposed that the tightly packed and repetitive atoms found in crystals—copper sulfate, for example—could do the job. Three years later, the father and son British physicists William Henry Bragg and Lawrence Bragg described the math that could decode the pattern of myriad dots recorded on a photographic plate after scattering X-rays through a crystal: If the X-ray wavelength and the intensity of the image were known, the molecular structure of the crystal could then be calculated. Table salt, for example, as the Braggs showed, has a formula of one atom of sodium plus one of chloride: NaCl. But the X-ray diffraction pattern shows that it is organized in alternating cubes, an atom of sodium surrounded by six chlorine atoms, next to a single chlorine surrounded by six sodium, repeating one after the other.

  Credit: Getty Images/Howard Clements

  Dorothy Crowfoot Hodgkin, 1910–1994

  Dorothy Crowfoot was barely fifteen, and still living in Khartoum—her father, John Winter Crowfoot, was employed in the Egyptian Education Service of Britain’s Colonial Office—when Dr. A. F. Joseph, a chemist working for Wellcome Laboratories, gave her a book by Sir William Bragg on X-ray diffraction, and so kindled her fascination with crystals. At Oxford, the fascination turned into a passion. Under the tutelage of Robert Robinson, a future chemistry Nobelist and the head of the Dyson Perrins lab, and especially J. D. Bernal, she mastered the technique of using X-ray crystallography to map the internal structure of organic molecules several orders of magnitude more complex than Bragg’s table salt.* The list included cholesterol, testosterone, progesterone, pepsin, and a dozen more, most particularly insulin, whose structure Hodgkin began investigating in 1934, and finally resolved thirty-five years later, in 1969.

  By 1940, Hodgkin had begun a relationship even more fruitful and long-lasting than the one with Bernal, accepting her first grant, in the amount of £1,000, from Warren Weaver of the Rockefeller Foundation (though, Weaver, whose eye for talent had already found Howard Florey and Ernst Chain, described her request for a grant “for research on proteins and viruses, with the object of carrying out more general and fundamental investigations . . .” as an “uncertain proposal at the present moment”). It was the beginning of a nearly thirty-year relationship with the Rockefeller Foundation, one of the longest in the foundation’s history.

  Also in 1940, her extraordinary talents—and her team at Oxford’s Laboratory of Chemical Crystallography—were recruited to the penicillin project. Even before the publication of the August 1940 article in the Lancet, she had been given a heads-up about the miraculous discoveries going on at the Dunn School, when she encountered Ernst Chain on a walk along Oxford’s South Parks Road, just after the first mice had been successfully treated with penicillin. Chain promised, “Some day we will have crystals for you.” It took more than a year for the promise to be redeemed, but in November 1941 Hodgkin wrote to her husband, “I’ve just come back from visiting Chain and now it’s 10:30 P.M. I’m feeling disgustingly cheerful as a result of my visit. . . . Apparently [penicillin] hasn’t yet been crystallized after all. . . . Chain seemed quite keen to let me have some stuff”—the degradation products—“and I’d simply love to try.”

  She quickly found that the degradation products were extremely difficult to work with: tiny crystals that were, in Hodgkin’s words, “immersed in a gummy fluid [and] extremely hygroscopic [i.e., attracting water molecules from the environment], so that they are practically impossible to leave out in the air for more than a few minutes or so.” They did, however, reveal one of penicillin’s secrets: Sulfur was an essential element of the compound, which seemed to be a molecule containing nine atoms of carbon, eleven of hydrogen, four oxygen, two nitrogen . . . and one atom of sulfur.*

  But how were the atomic bricks arranged into a structure that was able to stop infectious pathogens cold? X-ray crystallography was a powerful tool for investigating such questions, but not an easy one. Creating the picture formed by the X-rays scattered by penicillamine crystals onto a photographic plate was only the first step, and as likely to be confusing as to be revealing.

  The reason is that the intensity of the dots captured on a photographic plate measures the amplitude of the wave of the X-ray as it bounced off the faces of the compound’s crystalline lattice, from which a complete picture of the crystal could, in principle, be derived. More electron density, higher amplitude. But a dot that made a dense, high-intensity impression on the plate could just as easily be the result of two waves that were both arriving at the same frequency—the same number of waves per second—and amplitude. When those kind of signals arrived “in phase” with one another, they reinforced the density. In the same way, a low-density image could be produced by two waves that were out of phase and canceled one another out.

  Phases: In the top graphic, two waves are in phase—that is, completely overlapping. In the middle, the waves are out of phase, and, in the bottom, completely out of phase.

  Solving the potential confusion demanded a subtle and tricky mathematical technique, one that had originally been developed in the early nineteenth century to describe how waves of heat travel in three dimensions over time. As an example, if a match is held under the center of a coin, the coin’s center will heat up; after the match is extinguished, the heat will dissipate in an ordered way, with the center of the coin getting colder, and the edges warming, until an equilibrium is reached. This phenomenon, however, is an extremely complex business, which is why, in the early nineteenth century, the French mathematician Jean-Baptiste Joseph Fourier helped to develop a technique that turns complicated wave functions like these into a collection of the simple sine or cosine waves that are taught to high school trigonometry students.

  Unfortunately, while X-ray amplitudes are relatively easy to chart with a simple molecule—remember the Braggs’ table salt—they rapidly become more complicated as molecules get larger. The electron density photographs start to resemble three-dimensional weather maps, complete with eddies, currents, and whirlpools. Diagramming a molecule the size of penicillin—essential for a full understanding of its structure—and calculating which waves were in or out of phase, would take Hodgkin more than two years.

  In the meantime, on both sides of the Atlantic, progress was accelerating in the race to produce penicillin in useful quantities: more animal trials, increasingly sophisticated production methods, and, of course, more precise maps of the mysterious compound. Most of the key work was invisible to the public at large; though, on the heels of the Dunn team’s August 1941 Lancet article, the September 15, 1941, issue of Time magazine reported:

  A marvelous mold that saves lives when sulfa drugs fail was described in the British Lancet last month by Professor Howard Walter Florey and colleagues of Oxford. The healing principle, called penicillin, is extracted from the velvety-green Penicillium notatum, a relative of the cheese mold. Although it does not kill germs, the mold stops the growth of streptococci and staphylococci with a power “as great or greater than that of the most powerful antiseptics known.”

  The article didn’t really generate much in the way of follow-up. Neither, implausibly, did the cure of Anne Miller in March 1942. From the standpoint of scientists like Heatley, Coghill, Florey, and even Chain, that was as it should be. With barely enough penicillin being produced for the minimum number of human trials, the last thing anyone needed was to raise the hopes of every patient in the English-speaking world. But, as the tide of war against the Axis seemed to slowly shift in favor of the Allied armies, the public, in the United States and Britain particularly, was primed to embrace a lifesaving miracle, the world’s first antibiotic drug.

  And, of course, the hero who discovered it.

  At the end of August 1942, the Times of London published a brief editorial promoting a greater degree o
f public investment in penicillin development. Either out of ignorance or delicacy, the editorial declined to lionize (or attack) any scientist or public official by name.

  It’s impossible to know what reaction the editorial might have generated on its own. However, on August 31, the Times published the following letter:

  Sir,—In the leading article on penicillin in your issue yesterday you refrained from putting the laurel wreath for this discovery around anyone’s brow. I would, with your permission, supplement your article by pointing out that, on the principle of palma qui meruit ferat,* it should be decreed to Professor Alexander Fleming of this research laboratory. For he is the discoverer of penicillin and was the author also of the original suggestion that this substance might prove to have important applications in medicine.

  As the careful reader might have guessed from the phrase, “this research laboratory,” the letter was signed:

  Almroth E. Wright

  Inoculation Department, St. Mary’s Hospital