Life's Greatest Secret

by Matthew Cobb

  The editors of Nature did not get the joke – to be fair, it was not very funny – and they published the letter in the 18 April 1953 issue of the journal. At the time the letter met its deserved fate and disappeared into oblivion, apart from a couple of ironic citations from sharp-eyed bacterial geneticists who picked up on the attempt at humour.46 Recently, some historians have taken this apparently brilliant insight seriously, hypnotised by the sudden appearance of the words ‘information’ and ‘cybernetics’ in a letter signed by Jim Watson. The main historian who has studied this period, the late Lily Kay, argued earnestly that the letter represented a ‘gestalt switch’ in the thinking of the scientific community.47 Despite not ‘getting it’, Kay was absolutely right: the very fact that Ephrussi was poking fun indicated that concepts and words had changed. For Ephrussi and his boozy pals, information and cybernetics were now so commonplace that they could be used in an ironic spoof. The real irony was that this squib of a joke appeared in Nature just one week before the three articles that described the structure of DNA, and seven weeks before Watson and Crick changed our view of life by allying that structure with the term ‘genetical information’, this time used in a deadly serious fashion.

  2. Spoof letter by Ephrussi and others to Nature, April 1953

  * Not all the prose is sparkling. The conclusion to chapter 2 reads: ‘I do not wish to close this chapter without indicating that ergodic theory is a considerably wider subject than we have indicated above. There are certain modern developments of ergodic theory in which the measure to be kept invariant under a set of transformations is defined directly by the set itself rather than assumed in advance. I refer especially to the work of Kryloff and Bogoliouboff, and to some of the work of Hurewicz and the Japanese school.’

  * Intriguingly, as early as the 1930s, the pioneering Cambridge physiologist Edgar Adrian was using the terms ‘information’ and ‘code’ to describe the activity of neurons – Garson, J. ‘The birth of information in the brain: Edgar Adrian and the vacuum tube’, Science in Context, vol. 27, 2015, pp. 31–52.

  –SIX–

  THE DOUBLE HELIX

  On 6 August 1945, when the atomic bomb destroyed Hiroshima, Maurice Wilkins was a 28-year-old British physicist working on the Manhattan Project. Like many of his colleagues, Wilkins had begun to have doubts about the morality of building the bomb as soon as Germany surrendered in 1945. The use of the bomb against Japan was the final straw. With his recent marriage in tatters and his love of physics poisoned by the horror of Hiroshima and Nagasaki, Wilkins returned to the UK.

  Schrödinger’s What is Life? inspired Wilkins to use physics to investigate biology, so he approached his PhD supervisor, John Randall, who suggested that he should trace how the amount of DNA doubled just before a cell divided. The two men, who worked at the new Medical Research Council biophysics unit that Randall had set up at King’s College, London, knew of Avery’s work and felt that DNA was at the very least a vital component of nucleoproteins, if not the sole genetic material. In 1947 Wilkins met Francis Crick, whose physics PhD had been interrupted by the war. Crick had also read What is Life? and had also turned to biophysics. The two men became close friends, even though they had very different personalities – Crick was a noisy, brilliant magpie, with an eye for shiny new ideas, whereas Wilkins was quiet and reserved, with an odd habit of turning away from the person he was speaking to. He was also prone to suicidal thoughts and was in psychoanalysis.1

  The friendship between Wilkins and Crick led to what is probably the most intensely studied moment in the history of twentieth century science: the discovery of the double helix structure of DNA. The events surrounding this event have been described in memoirs, biographies, exhibitions, TV programmes, countless academic articles, many inaccurate blog posts and even in a video rap contest.2* The story is of fundamental scientific importance and shows how science is an intensely human, collaborative and competitive enterprise in which luck, ambition and personality can play a central role. Above all, it was the advance that revealed the existence of the genetic code.

  *

  The King’s College biophysics unit was initially a minor player in the small world of groups that were studying the structure of DNA. The most influential work was being carried out in Columbia University by Erwin Chargaff, who was making a detailed biochemical analysis of the relative proportions of the four DNA bases – adenine, cytosine, guanine and thymine. Between 1948 and 1951, Chargaff showed that the four bases were not present in equal amounts; his conclusion was confirmed by Avery’s arch-critic, Alfred Mirsky, who by 1949 had become convinced that the old tetranucleotide hypothesis was ‘no longer tenable’.3

  Chargaff’s insight went much further. In 1951 he summarised the results he had published over the previous three years: different tissues of the same species yielded DNA with an identical composition in terms of the proportions of the four bases, and furthermore DNA molecules showed a ‘composition characteristic of the species from which they are derived’. DNA composition was constant in all tissues of a given species, but each species had its own profile. Even more importantly, he repeated a remarkable conclusion that he had come to the year before:

  It seems that in most specimens examined until now, the ratios of adenine to thymine, of guanine to cytosine, and of total purines to total pyrimidines were not far from one.4

  The ratios of the bases actually reported by Chargaff were not always as telling as he suggested: for example, in the cow the ratios ranged from 0.75 to 0.80 for C:G and were clustered at around 1.16 for A:T. Chargaff’s analytical procedure recovered only 70 per cent of the bases, so the ever-sceptical Mirsky dismissed the seductive similarities as experimental errors.5 Even Chargaff was not certain whether the ratios had any meaning:

  As the number of examples of such regularity increases, the question will become pertinent whether it is merely accidental or whether it is an expression of certain structural principles that are shared by many desoxypentose nucleic acids, despite far reaching differences in their individual composition and the absence of a recognizable periodicity in their nucleotide sequence. It is believed that the time has not yet come to attempt an answer.6

  Chargaff was clearer about the possible significance of the nucleotide sequence, although when he gave a lecture on the subject, in the summer of 1949, he could not exclude the possibility that genes were made of nucleoproteins, not nucleic acids:

  We must realize that minute changes in the nucleic acid, e.g. the disappearance of one guanine molecule out of a hundred, could produce far-reaching changes in the geometry of the conjugated nucleoprotein; and it is not impossible that rearrangements of this type are among the causes of the occurrence of mutations.7

  The leading researcher into the structure of DNA was Bill Astbury of the University of Leeds. In 1938, Astbury and his PhD student Florence Bell had published X-ray images of DNA and had described a model in which the bases were strung perpendicularly along the phosphate-sugar backbone, like ‘a pile of pennies’. But at the Cambridge meeting of the Society for Experimental Biology in 1947, Astbury had changed his mind and had argued that the bases were in fact parallel to the phosphate-sugar backbone, as though the ‘pennies’ were laid flat, rotated at 90° compared to his previous view. At the same time, Masson Gulland’s group in Nottingham published evidence suggesting that the bases were linked by hydrogen bonds – a particularly strong and biologically widespread form of atomic bond – but Gulland could not determine whether these were bonds between nucleotides of the same DNA chain or between different DNA chains.8

  Meanwhile, at Birkbeck College in London, a Norwegian PhD student called Sven Furberg was struggling to analyse the organisation of each of the four bases found in DNA with the use of X-ray crystallography. This technique involved crystallising a sample of one of the bases and then bombarding it with X-rays for hours on end. A piece of photographic film captured the result – patches of light and dark on the film that had been produced by the
diffraction of the X-rays by the crystal structure of the sample. With a great deal of effort and no small amount of luck, these blobs could be interpreted in terms of the molecular structure of the crystal, which had diffracted the X-rays in a consistent fashion. Furberg was using some highly complex mathematics called the Patterson function to calculate a three-dimensional molecular structure on the basis of a series of two-dimensional X-ray images taken at different orientations. In a 1950 paper, Furberg concluded that Astbury’s initial insight was correct and that the bases were indeed regularly spaced perpendicular to the phosphate-sugar backbone. Furberg’s model of DNA structure was a spiral or helix, with the backbone twisting around on itself. But he could not determine the exact form of the molecule: either the backbone was in the middle, with the bases sticking outwards, or it was on the outside of the helix, with the bases all converging on the centre. Although Furberg’s idea was not published until 1952, a copy of his 1949 thesis ended up at nearby King’s College, where Wilkins was becoming increasingly interested in the molecular structure of DNA.

  In May 1950, Wilkins heard a talk by Rudolf Signer from Bern, who described a new method for creating high-quality DNA, and generously gave out samples. Wilkins initially tried flattening Signer’s pure DNA into a thin gel – it had the consistency of snot, he later recalled. Then something interesting happened:

  Each time that I touched the gel with a glass rod and removed the rod, a thin and almost invisible fibre of DNA was drawn out like a filament of a spider’s web. The perfection and uniformity of the fibres suggested that the molecules in them were regularly arranged.9

  Together with his PhD student Raymond Gosling, Wilkins made a frame from a bent paperclip (it was later upgraded to fine tungsten wire), stretched a DNA fibre across the metal and then put the sample in front of an X-ray source. To reduce background scatter by the X-rays, the inside of the camera was filled with hydrogen, and the seal between the X-ray tube and the camera was bound in a condom.10 After much fiddling about with the humidity of the sample, they produced good X-ray diffraction images on a sheet of photographic film. As with Astbury and Bell’s prewar picture, the patterns produced by Wilkins and Gosling showed a set of concentric curved and vertical lines, but their image was far sharper. It showed that when it had been pulled into a fibre Signer’s DNA was in a quasi-crystalline state: it was organised in a regular fashion with all or most of the molecules apparently arranged in the same orientation. Gosling later recalled his reaction; he was ‘standing in the dark room outside this lead-lined room, and looking at the developer, and up through the developer tank swam this beautiful spotted photograph … it really was the most wonderful thing. … I went back down the tunnels over to the Physics Department, where Wilkins used to spend his life, so he was still there. I can still remember vividly the excitement of showing this thing to Wilkins and drinking his sherry by the glass … by the gulpful.’11

  Both Wilkins and Gosling later suggested that at this moment they thought that genes were made of DNA, and that they had therefore crystallised a gene. But none of the publications from the King’s group stated matters so clearly, and even in August 1950, four months after the image was made, Wilkins expressed his uncertainty in a letter to a friend: ‘What we would really like to do, of course, is to find what nucleic acid is in cells for.’12 A year later, Wilkins showed the picture in a lecture but still suggested that nucleoproteins, rather than nucleic acid, were the physical basis of genes.13 The successful use of X-ray diffraction had in fact deepened the fundamental paradox of DNA. If it were the genetic material, then it should show variability, in order for the specific effects of genes to be expressed. But the fibre diffraction images were tantalisingly suggesting some fixed, repetitive and relatively simple structure.

  Soon after this, the X-ray tube broke. It was months before it could be replaced, so the King’s team tried to attack the problem using other techniques. Wilkins and Gosling noticed abrupt changes in the length of the DNA fibres with increasing humidity. This was remarkably similar to an effect seen in one of the most intensely studied proteins, keratin, which switched its form as the molecule stretched. In February 1951, two papers on DNA structure were sent to Nature by the King’s group. Wilkins, Gosling and their colleague Willy Seed described the stretching of DNA fibres as seen under polarised light, and concluded that as the molecule stretched when it dehydrated, the orientation of the bases changed, becoming more parallel to the phosphate-sugar backbone.14 In a second paper, a King’s PhD student called Bruce Fraser, together with his wife, Mary, used infrared measurements to confirm Furberg’s suggestion that the bases were normally perpendicular to the phosphate-sugar backbone. Above all, the Frasers reported that their data could ‘be interpreted in terms of a structure similar to that proposed by Furberg’ – a helix.15

  *

  In the related but distant world of protein structure, a helix was also causing a storm. Years earlier, Astbury had studied X-ray photos of unstretched keratin (known as α-keratin – α is pronounced ‘alpha’) and had suggested that it had a helical structure. There followed a long and intense competition between Lawrence Bragg’s group at the Cavendish Laboratory in Cambridge and Linus Pauling’s group at Caltech in Pasadena, as each tried to come up with a precise description of what became known simply as ‘the α-helix’. In a wave of papers that appeared between November 1950 and May 1951, Pauling presented exact models that accurately described the helical structure of keratin, and showed that the same structure could be found in a range of biological tissues.16 Bragg and his group were devastated. They had lost a race that had lasted for years.

  There were still some kinks to be ironed out, including explaining how the α-helix seemed to coil round itself, but by early 1952 these had been fixed by both Pauling and by an infuriatingly brilliant and garrulous newcomer at the Cavendish, Francis Crick. Crick was vaguely studying for his PhD on the molecular structure of haemoglobin, including the mathematical theory behind the X-ray diffraction data produced by helical molecules.17

  Pauling’s description of the structure of the α-helix was a tour de force, but it was far more impressive to chemists than it was to biologists. The structure provided no explanation of keratin function. Max Delbrück was particularly scornful, as he later recalled: ‘the α-helix, even if correct, had not provided any biological insights’.18

  *

  Randall was acutely aware that the King’s group lacked the skills required to interpret X-ray crystallography data. In late 1950, he recruited a British woman researcher in Paris who was using X-ray diffraction to study the molecular structure of coals. The initial plan was that she should study the structure of proteins in solution, but then Wilkins fatally suggested to Randall it might be a good idea for her to work on the X-ray diffraction analysis of DNA. Randall agreed and sent a letter to his new researcher, 30-year-old Rosalind (pronounced ‘Ros-lind’19) Franklin, explaining the change of plan – ‘nucleic acid is an extremely important constituent of cells and it seems to us that it would very valuable if this could be followed up in detail’.20

  Randall’s letter to Franklin, which Wilkins did not see for several decades, was the source of several tragic misunderstandings between Wilkins and Franklin. In the letter, Randall stated that Franklin would be the only researcher studying DNA with X-rays – ‘as far as the experimental X-ray effort is concerned there will be at the moment only yourself and Gosling’. Franklin understandably concluded that she would have sole control of her research. But Wilkins was still interested in using X-ray diffraction on DNA and had no idea that Randall apparently wanted him to hand the project over to Franklin. When Franklin arrived in the laboratory, Wilkins was on holiday, and when he returned he discovered that his PhD student Gosling was now working with the new arrival without any explanation. This uncomfortable situation could have been solved by talking, but Wilkins and Franklin were the victims of an immediate and appalling clash of personalities. Wilkins was quiet, diffident and hated a
rguments; Franklin was forceful and thrived on the rough and tumble of intellectual debate. Her friend Norma Sutherland recalled: ‘Her manner was brusque and at times confrontational – she aroused quite a lot of hostility among the people she talked to, and she seemed quite insensitive to this’.21 Wilkins was intimidated by Franklin’s character and bewildered by her refusal to work with him. Franklin, in contrast, was irritated by Wilkins’s rather limp behaviour and by his ignorance of the fundamentals of X-ray diffraction. Their working relationship was doomed before it began.

  It is not known why Randall wrote his letter to Franklin in the way that he did. Wilkins later wondered whether it was an attempt to sideline him from the project, either because Randall wanted to keep DNA to himself or because he was frustrated by Wilkins’s slow progress. Perhaps Randall hoped to get the best out of both researchers by setting them up to compete with each other. Whatever the case, the inherent personality differences between Wilkins and Franklin were horribly amplified by their totally different impressions of their respective roles, and their inability to simply talk about things.22

  *

  At the end of May 1951, Randall was due to speak at a dull-sounding ‘Symposium on submicroscopical morphology in protoplasm’, to be held at the Naples Marine Biological Station. He was unable to attend, so he sent Wilkins in his place. In his talk, Wilkins explained their work but presented it in terms of nucleoproteins, not nucleic acids: ‘when living matter is to be found in a crystalline state, the possibility is increased of molecular interpretation of biological structure and processes. In particular, the study of crystalline nucleoproteins in living cells may help one to approach more closely the problem of gene structure.’23 At that point he showed the DNA X-ray diffraction image he had taken with Gosling. In the audience, a gangly 23-year-old American suddenly paid attention. His name was Jim Watson. He later described the scene: