Dorothy Hodgkin, one of the main founders of protein crystallography, possessed a unique mixture of skills that allowed her to extend the use of X-rays to reveal the structures of compounds that were far more complex than anything attempted before.

Victory in Europe Day in Oxford, 8 May 1945. The war in Europe was over, and thousands of people lined the streets to celebrate. One woman making her way through the cheering crowds had even more reason to be triumphant. Dorothy Hodgkin held in her hands a model of wires and corks so frail she struggled to protect it from the celebrations, yet the information within this model would help to protect many of these people, and countless more, in years to come.

Hodgkin had just solved the structure of penicillin, and not even a crowd of thousands could have prevented her from getting to the nearby Dunn School of Pathology to show her discovery to an equally excited colleague, Ernst Chain.

Five years earlier, on a morning in May 1940, Hodgkin had bumped into an unusually animated Chain outside the Dunn School of Pathology. Chain told Hodgkin that he and his colleague Howard Florey had just discovered something amazing. Florey and Chain were experimenting with penicillin, the bacteria-killing substance that Alexander Fleming had somewhat fortuitously discovered in moulds in 1928. Since Fleming’s discovery, several researchers tried and failed to purify active forms of penicillin, and had resigned themselves to the fact that penicillin might be interesting to the bacteriologist but would be of no great practical importance in medicine, until the experiment that Chain was so excited about.

Florey and Chain had injected eight mice with lethal doses of bacteria, but four of them also received extracts of penicillin. The untreated mice died within one day, but to Florey and Chain’s excitement the penicillin-treated mice survived for several days or weeks. For the first time since Fleming’s discovery it seemed possible that that penicillin extracts might be used as a remedy, but Chain knew that knowing the molecular structure of penicillin would be crucial. “Some day we will have crystals for you,” Chain promised Hodgkin.

A Rare Talent

Hodgkin had just turned thirty, but was already known as arguably the most outstanding X-ray crystallographer of her time. The pioneering work of Max von Laue and the father-and-son team of William Henry Bragg and William Lawrence Bragg during the early 1910s had opened up the possibility that the pattern of spots formed on photographic plates as a result of passing beams of X-rays through crystals could be used to find out the three-dimensional structures of compounds. Scientists entering this new discipline, called X-ray crystallography, were busy working out how to crystallize ever more complex compounds, and how to take good X-ray photographs.

Hodgkin began her career in Cambridge in 1932, working in the lab of John Desmond Bernal. Bernal was a true pioneer in these early days of X-ray crystallography, consistently pushing the boundaries of what was possible using this method. Bernal was the first to discover that keeping protein crystals wet produced great X-ray photographs. In 1934, Bernal had succeeded in producing an X-ray diffraction photograph of the digestive enzyme pepsin, which caused a sensation at that time. Bernal and Hodgkin confirmed the finding, and their 1934 paper on pepsin in Nature marked the beginning of protein crystallography.

The charismatic Bernal, who was the scientific inspiration for Hodgkin, was impressed by his young student’s skill and determination from the beginning. Hodgkin showed a natural aptitude for even the most tedious and convoluted aspects of X-ray crystallography, like the thousands of complex mathematical calculations by hand that were needed to create a Patterson map, a form of contour map that could be used to define the distances between atoms in a crystal.

Hodgkin started her own lab at Somerville College, Oxford, in 1935. Within a year she produced her first X-ray photograph of insulin, after which she is said to have wandered the streets of Oxford in a daze, until advised to go home by a kindly policeman in the early hours of the morning. Hodgkin then moved onto more complex compounds called sterols for her PhD research, and soon after she solved the structure of cholesterol. So, when Chain told her about the need to identify the structure of penicillin, Hodgkin, not unjustifiably, did not think this would be beyond her capabilities.

Perhaps this confidence was just what was needed. It was wartime, and countless wounded soldiers suffering from bacterial infections would benefit from a drug like penicillin. Trials with penicillin were beginning to show amazing results. In February 1941, the first human patients were treated, and doctors witnessed resurrection-like cures of death-bound patients with relatively impure extracts. With insufficient supplies of the drug, however, the treatment had to be discontinued prematurely in some cases. So, the development of a penicillin treatment became a top priority for the authorities both in the United States and United Kingdom, and an unprecedented network of academic and industry scientists collaborated to reach this goal.

Chemists were still struggling to isolate penicillin in a pure and active form. It was not until July 1943 that they knew the composition of the active ingredient. After systematically breaking down penicillin into smaller pieces, chemists knew it consisted of 27 atoms: 11 hydrogen, 9 carbon, 4 oxygen, 2 nitrogen atoms and 1 sulphur atom. The trouble was that this combination of atoms could form two very different structures, and chemists couldn’t decide which structure was more likely. Some chemists were convinced the structure contained two five-membered rings connected by a single bond, known as a thiazolidine-oxazolone. Others were equally sure it was a four-membered ring fused to a five-membered ring, known as a beta lactam.

“No absolutely unequivocal conclusion could be derived from it,” Ernst Chain explained in his Nobel Lecture. “The final solution of the problem of the structure of penicillin came from crystallographic X-ray studies.”

Hodgkin received the first useful penicillin crystals in autumn 1943 from Robert Robinson, who had brought some back from a recent visit to the laboratories of the US pharmaceutical company Squibb. Researchers at Squibb had succeeded in creating wonderful crystals of benzylpenicillin (later known as penicillin G) as a sodium salt, isolated from mould growing on a melon in Illinois. But Hodgkin needed more crystals to solve the structure and decided she also needed crystals from two different penicillin G salts, potassium and rubidium, so that she could compare the three different diffraction patterns.

With the help of the director of the Royal Institution, Sir Henry Dale, a US military aircraft brought an additional 10 milligrams of the sodium version of penicillin to the UK in February 1944. The Medical Research Council persuaded the UK company ICI in the UK to supply Hodgkin with the rubidium version of penicillin, and Squibb supplied some potassium penicillin. Due to the highly sensitive environment that all parties were operating under, Hodgkin was informed “that all information on the chemistry and production is now on the secret list and may not be transmitted to anybody either in this country or abroad.”

Hodgkin and her team joined forces with ICI’s Charles Bunn. In their two labs they passed X-ray waves through the different penicillin salt crystals, and recorded each of their diffraction patterns on photographic film.

Uncovering the Structure

Hodgkin set to work uncovering the information about penicillin’s structure that lay within the patterns. For simple compounds, like penicillin, the arrangement of atoms can be determined from the position and brightness of the regular array of spots in the diffraction patterns.

In Hodgkin’s day, converting a diffraction pattern into the structure of a molecule relied as much on inferences and perspiration as it did on hard and fast rules. As X-rays pass through a crystal, they are deflected off the electrons that orbit atoms, so X-ray crystallographers can only use the diffraction patterns to create a map of the density of electrons within the crystal, from which they must infer the positions of the atoms. Calculating these electron density maps is a complex mathematical problem which today is easily solved by computers. When Hodgkin was solving these structures, however, computers were only in their infancy, and handling the data and translating the diffraction patterns into electron densities required a great deal of hard work.

According to her peers, Dorothy Hodgkin succeeded where others failed through a combination of creative qualities and profound chemical knowledge. She followed an intuitive sense for how to position atoms in space and had a feeling for the structures she was dealing with. In the case of penicillin, researchers struggled to interpret correctly the electron density maps that held the secrets to its structure. Hodgkin solved this even though the two-dimensional electron density maps from the different crystals conflicted with each other, simply by drawing projections of the sodium and rubidium structures to the same scale, placing them on top of each other, and rotating them until she found a position in which many of the peaks coincided – which, she concluded, must represent the molecule.

The molecule in question had the beta lactam structure. To confirm this result, she used the computing services of the Medical Research Council, which in those days was a Hollerith punch card machine. The machine calculated that the electron density of penicillin in three dimensions had a beta lactam core. It was a bittersweet moment for one of the Medical Research Council’s researchers: John Cornforth had boldly stated that if the beta lactam structure was correct he would give up chemistry and grow mushrooms instead. Thankfully for him he did not fulfil the wager, as he ended up winning the Nobel Prize in Chemistry in 1975 for his work on the stereochemistry of enzyme-catalysed reactions.

Hodgkin’s structure of penicillin was, as Ernst Chain said in his Nobel Lecture, a considerable achievement. “For the first time the structure of a whole molecule has been calculated from X-ray data,” said Chain, “and it is the more remarkable that this should have been possible in the case of a substance having the complexity of the penicillin molecule.” Knowledge of the structure finally opened new avenues for creating and developing semi-synthetic derivatives of penicillin – such as the cephalosporines – that sparked the creation of antibiotic treatments.

Dorothy Hodgkin was keen to share her success with her mentor Bernal. As she later wrote: “I remember sitting on the steps of the Royal Society … talking to Bernal and I was telling him that we had solved the structure of penicillin. He said ‘you will get the Nobel prize for this.’ I said ‘I would far rather be elected a Fellow of the Royal Society’ and he said ‘that’s more difficult’.”

In fact, the opposite became true. Two years later, Hodgkin was elected Fellow of the Royal Society. It took another seventeen years, and solving the structure of vitamin B 12, before she was awarded the Nobel Prize for Chemistry in 1964. For her colleague and friend Max Perutz this recognition was long overdue. Perutz, who had received the Prize for Chemistry two years before Hodgkin for his work in discovering the structure of haemoglobin, said: “I felt embarrassed when I was awarded the Nobel Prize before Dorothy, whose great discoveries had been made with such fantastic skill and chemical insight and had preceded my own.”

Bibliography

Ferry, G. Dorothy Hodgkin. A life. London 1998.

Glusker, J. P. Dorothy Crowfoot Hodgkin (1910-1994). Protein Science 3, 2465–2469 (1994).

Chain, Boris Ernst: The Chemical Structure of the Penicillins. Nobel Lecture, March 20, 1946.

Hodgkin, Dorothy Crowfoot: The X-ray analysis of complicated molecules. Nobel Lecture, December 11, 1964.