Hodgkin Solves the Structure of Penicillin

Date 1944-1949

Dorothy Crowfoot Hodgkin used a computer to work out the X-ray data of penicillin, becoming the first to use a computer in direct application to a biochemical problem.

Locale Oxford, England

Key Figures

• Dorothy Crowfoot Hodgkin (1910-1994), English crystallographer, who won the 1964 Nobel Prize in Chemistry
• Barbara Wharton Low (b. 1920), English chemist, who assisted in the penicillin project

Summary of Event

The method Dorothy Crowfoot Hodgkin used to solve the problem of the molecular structure of the antibiotic penicillin consisted of three scientific disciplines that had, to that time, little common ground: synthetic organic chemistry, X-ray crystallography, and the new field of computer technology. Penicillin was discovered in 1928 by Alexander Fleming at St. Mary’s Hospital in London. Fleming’s original paper on penicillin, published in 1929, described potential uses of the antibiotic as well as toxicity tests performed on animals and on some sensitive human tissue. Ernst Boris Chain and Baron Florey at the University of Oxford began working with penicillin in 1935, with Chain attempting to extract and purify sufficient penicillin for Florey’s clinical testing. They published the results of a series of successful treatments in 1941. Time brought worldwide attention to the new wonder drug in 1941, but noted that because penicillin was difficult to extract, it had limited usefulness until it could be prepared with less expense or synthesized.

Because of the potential antibiotic activity of penicillin and its low toxicity, it was invaluable on the battlefields of World War II. An enormous effort that involved scientists in academia, industry, and government was undertaken by the English and American governments in 1942 to increase the supply of penicillin. The research effort moved in two directions: improved means of producing natural penicillins and the synthesis of an artificial penicillin.

The development of a synthetic route to penicillin required a knowledge not only of the chemical formula of penicillin but also of its molecular structure. The usual method of structure determination by organic chemists at that time was to subject the compound to harsh treatment and then study the degradation products. The analysis of these fragments would allow reconstruction of the original molecule, a jigsaw puzzle approach.

Among the proposed structures for the penicillin molecule, two were considered leading contenders by the end of 1942. The favorite of Sir Robert Robinson, a well-known organic chemist of the University of Oxford, was a combination of two rings known as oxazolone-thiazolidine. Nevertheless, the structure had a serious problem: The formula predicted that the molecule should contain a basic group that was never found by titration. A second possible structure was suggested, a different assembly of rings known as a beta-lactam-thiazolidine structure. This was an unfamiliar structure that had never been found in natural products. No known reactions could explain how the observed degradation fragments could fit to give the beta-lactam structure. It seemed that the traditional organic method of structure determination was incapable of resolving the problem. To settle the dispute, a sample of penicillin was sent to Hodgkin at the end of 1944 for a single-crystal X-ray study of penicillin.

The discipline of X-ray crystallography , in which Hodgkin spent her scientific career, had its beginning in Munich in 1912, when Max von Laue showed that X rays behaved as electromagnetic waves with short wavelengths and that X rays scattered by a row of atoms in a crystal could produce interference patterns. The intensity of the various spots in the pattern and their spacing depended on the arrangement of the atoms in the crystal. Sir Lawrence Bragg in England formulated a simple law by which he was able to explain successfully all the spots from Laue’s patterns. This development provided a new tool for examining the internal arrangements of atoms in crystals and provided solutions, almost immediately, for many inorganic structural questions.

The general method involved in crystallographic study is repeated comparison of experiments and calculation of reflections from various planes of the crystal. If the structure of the crystal and the wavelength of the X rays are known, it is easy to predict the diffraction pattern. It is much more difficult to deduce the crystal structure from the observed pattern. For very simple structures, a trial-and-error method is appropriate, but as more complex structures are studied, the calculations involved become extremely tedious and time-consuming.

The object of a crystal-structure determination is to ascertain the positions of all atoms in the basic unit of the crystal. The process involves collection of data, solution of the phase relations among the scattered X rays (determination of a trial structure), and refinement of this structure. In the 1940’s and 1950’s, the time involved in the difficult task of collecting three-dimensional data (the first step) was short in comparison with the time required to solve the phase problems (the second step). This depended on the complexity of the problem and on the luck, perseverance, and intuition of the investigator. The third step required such a large number of calculations that it was ignored, except for the simplest of structures.

Hodgkin applied her particular qualities of precision, astute mathematical analysis, and special imagination to the penicillin problem as soon as crystals were prepared for photographing. Three crystals were prepared for study: sodium benzylpenicillin and the isomorphous potassium and rubidium salts of benzylpenicillin.

The early electron density projections, based on the phase angles obtained from the isomorphous salts, indicated the presence of a heavy scattering center, identified as sulfur, in the crystal structure. Armed with these few clues and the tentative structures, Hodgkin and Barbara Wharton Low proceeded with a trial-and-error analysis. They constructed scale models using heavy wire, and by studying the shadows made by illuminating the model with parallel light beams, they were able to record new atomic coordinates. Further calculations followed. Hodgkin and Low frequently compared their results with those of C. W. Bunn and A. Turner-Jones, who were using an optical diffraction method in studying the sodium salt of benzylpenicillin. Eventually, the two groups agreed on the beta-lactam structure as the correct one for the penicillin molecule.

Significance

X-ray crystallographic analysis of even the simplest molecules requires many mathematical computations. When the work on penicillin began, the computing equipment available was grossly inadequate. By the mid-1940’s, Hodgkin was able to borrow an old IBM card-punch machine, without which the refinement of the penicillin structure would have been almost impossible.

Hodgkin continued to make use of computers of varying degrees of complexity to solve even larger molecules. It was becoming clear that understanding the chemistry of life processes required a detailed stereochemical knowledge of the compounds involved. By 1956, Hodgkin had shown by the penicillin work and by solving the structure of vitamin B₁₂ that it was possible to use X-ray crystallography alone to determine even very complex structures. As the size and complexity of the molecules to be analyzed increased, the need for more advanced computers increased also. Fortunately, computer technology kept pace with the demands of other fields of science.

During the 1960’s, many organic chemists worried about the gradual encroachment of X-ray crystallography into what had been an important preserve of organic chemistry: the elucidation of structure. It was becoming evident by that time that natural product structures, especially those that contained new features, could be determined more quickly by X-ray analysis than by traditional degradative methods. The chemists feared the loss of new discoveries frequently brought about by the process of classical structural analysis, particularly in the area of chemical synthesis In fact, organic chemistry has survived its crisis rather well. Freed from the responsibility of structural determination, organic chemists have turned, with great success, to other endeavors.

As interest among organic chemists in degradation studies waned, their energy was channeled into other areas, particularly synthetic and mechanistic studies. Various types of bonds were studied during this period, especially those formed in organic molecules and their influence on geometry: angles, bonding and nonbonding distances, and planarity of certain groups. The hydrogen bond also received much attention after crystal X-ray analysis showed that some pairs of electronegative atoms (oxygen, nitrogen, fluorine), at least one of which was bonded to a hydrogen, approached closer than expected. The importance of the hydrogen bond was recognized at this time, particularly its influence in determining secondary structures of biological molecules.

The use of the computer in modeling chemical compounds has gone far beyond the computational stage. The development of graphics capability allows models of molecules to be viewed on screen and manipulated in many ways. New drugs are frequently computer-designed to mimic natural molecules and to act as positive or negative inhibitors of many body processes. The introduction of the computer truly has revolutionized chemistry.

Bibliography

Burke, John G. Origins of the Science of Crystals. Berkeley: University of California Press, 1966. A short history of the science of crystallography. Burke clearly explains the terms used in crystallography, as well as the basic theories.

Crowfoot, D., C. W. Bunn, B. W. Rogers-Low, and A. Turner-Jones. “The X-Ray Crystallographic Investigation of the Structure of Penicillin.” In The Chemistry of Penicillin, edited by Hans T. Clarke, John R. Johnson, and Sir Robert Robinson. Princeton, N.J.: Princeton University Press, 1949. A rather technical article by Hodgkin and colleagues that describes the tedious process by which the structure of penicillin was determined. For readers with an average background in mathematics and general science.

Dodson, Guy, Jenny P. Glusker, and David Sayre, eds. Structural Studies on Molecules of Biological Interest. New York: Oxford University Press, 1981. Contains technical articles and articles about the early work of Hodgkin and her importance to organic structural analysis. Many of the latter are the reminiscences of friends and colleagues.

Ewald, P. P., ed. Fifty Years of X-Ray Diffraction. Utrecht, the Netherlands: Oosthoek’s Uitgeversmij, 1962. A detailed description of the development of X-ray crystallography written for the fiftieth anniversary of the discovery of X-ray diffraction by crystals.

Ferry, Georgina. Dorothy Hodgkin: A Life. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press, 2000. An excellent, detailed biography of Dorothy Crowfoot Hodgkin.

Lax, Eric. The Mold in Dr. Florey’s Coat: The Story of the Penicillin Miracle. New York: Henry Holt, 2004. Relates the story of the discovery of penicillin and its development into a useful drug. Sheds light on the personalities of the scientists involved in its discovery.

Sheehan, John C. The Enchanted Ring: The Untold Story of Penicillin. Cambridge, Mass.: MIT Press, 1982. An interesting and readable book on the history of the attempt to synthesize penicillin by the person who eventually accomplished the feat. Discusses the many conflicts that arose along the way, from the crediting of the original discovery by Fleming to the disagreements over the structures proposed by organic chemists before Hodgkin’s solution.

Straus, Eugene W., and Alex Straus. Medical Marvels: The One Hundred Greatest Advances in Medicine. Amherst, N.Y.: Prometheus Books, 2006. A history of medicine and medical discoveries with chapters on penicillin and other antibiotics.