His cynicism had hardened by 1945, the year in which he was elected Fellow of the Royal Society. At that time, he saw ‘no evidence to justify the existence of the tetranucleotide’ and noted the emerging reports that the four bases were liberated in unequal amounts when DNA was broken down. These revolutionary results had to be confirmed, but Gulland already looked forward to the post-tetranucleotide era when DNA, freed from the shackles that had held it back for thirty years, could be reexamined to see if it served specific and fundamental roles in biology. This was why he was so excited by the assertion that pneumococcal genes were made of DNA. Like Avery and McCarty, Gulland saw implications that went far beyond bacteria and would be ‘of great importance in the fields of genetics, virology and cancer research’.
In his lecture at Cold Spring Harbor, Gulland presented results recently obtained by three of his PhD students and then tried to make sense of what they had found. First, J.C. Threlfall had developed an especially gentle method for extracting pure DNA from calf thymus. He found his DNA molecules to be bigger than previously reported, suggesting that they were closer to the ‘native’ form in living nuclei. The calculated molecular weight was around 3 million, and the molecule’s length had stretched to over 700 times its width.
Next, H.F.W. Taylor had exposed DNA to acids and alkalis. Others, including Miescher, had long since done this but Taylor used a highly accurate method (electrometric titration) across a wider range than ever before, from pH = 2 (acidic enough to dissolve iron) to pH = 14 (drain cleaner). This revealed clues that DNA had withheld until then. Two separate populations of chemical groups were apparently buried deep inside the DNA molecule, and could only be unmasked by strong acid or strong alkali, respectively.
The third PhD student, Michael Creeth, used two physical methods to probe the DNA molecule across this wide range of pH. He found that the normally viscous DNA solution suddenly became runny at the same critical pH points where the hidden chemical groups were revealed. Creeth also recorded changes in refractive index (the ability to bend light) when DNA solutions were forced to flow through a narrow glass tube. For DNA solutions at neutral pH, the refractive index measured in the direction of flow differed markedly from that measured across the stream. Outside the pH cut-offs, that difference was greatly reduced.
Gulland’s interpretation of what happened at very high and low pH was elegant and masterly. He suggested that a standard solution of DNA was highly viscous because it contained a tangle of very long molecules. When the DNA solution was pumped through a narrow tube, the string-like molecules would be forced to line up in the direction of flow, like a handful of raw spaghetti dropped into a straight-sided glass. This mass orientation would explain the peculiar optical properties of the flowing solution. If we visualise the sheaf of spaghetti from above (along the ‘line of flow’) and then from the side (across the ‘flow’), we can appreciate how light could be bent differently in those directions.
Gulland suggested that the dramatic changes in viscosity and refraction at extremes of pH were caused by a single phenomenon: the rupture of specific links which normally hold the DNA molecule in shape. Breaking these bonds would distort the DNA molecule and simultaneously expose the chemical groups that the bonds usually concealed. The features of the titration curve enabled him to identify the crucial links as ‘hydrogen bonds’. This force, discovered during the 1920s, pulls certain atoms in a molecular structure towards hydrogen atoms. Hydrogen bonding is weak – just a handclasp, compared with the handcuff of the ‘covalent’ bond which permanently locks atoms together – but it is strong enough to bend flexible molecules so as to bring together the atoms of attraction. Gulland also worked out that the all-important hydrogen bonds must tie together bases in the DNA structure, although details – such as whether particular bases were connected – remained beyond reach. His analysis was a huge leap forward, even though its full significance would not be realised for several years. This was the first suggestion that hydrogen bonds hold DNA in shape, a crucial understanding that helped to launch the final assault on the structure of the double helix.
Hydrogen bonding came into one of the questions that followed Gulland’s talk – and if we could be transported back to that moment, we might have believed that we were about to witness a turning point in science. Arthur Pollister, Alfred Mirsky’s co-investigator, first sketched out the background to his question: the recent evidence that chromosomes were made of tightly coiled threads of nucleoproteins. This prompted Pollister to speculate that ‘the hierarchy of coils actually extends down to a molecular spiral’. As DNA was the only linear molecule in chromosomes, it was the obvious candidate. Therefore, would Dr Gulland care to comment on the possibility that ‘DNA can be thrown into a regular helix by evenly spaced linkages, perhaps hydrogen bonds, between different points along the molecule?’
Evidently, Gulland had not wondered whether DNA might be helical rather than linear. He thanked Pollister for his ‘interesting and stimulating suggestion’ and gave a waffly answer that did not mention a helix at all.
The rest of the summer of 1947 was a whirl for Masson Gulland, because his dream job – Professor of Biochemistry in his beloved Edinburgh – came up. Armed with his FRS and heavyweight curriculum vitae, he threw all his energy into getting the Chair – and was devastated when they appointed someone else. The normally calm and urbane Gulland was so disgusted that he gave up academia for good; he resigned from his professorship in Nottingham, and moved to London as the first FRS to direct research at the Institute of Brewing. Gulland’s team was shattered by his abrupt departure. The group quickly fell apart, leaving the three PhD students to write up their theses while looking for new jobs. Within six months, it was as though Nottingham had never nurtured a research interest in the nucleic acids.
By then, Masson Gulland was out of circulation, because of a vicious twist of fate. During the war, while researching seaweed as a source of weather-resistant materials, he had founded the Scottish Seaweed Research Association. In late October, he went up to Scotland for an Association committee meeting and was desperately unlucky in choosing the train home. On 27 October 1947, the 11.15 from Edinburgh to London King’s Cross left the tracks just south of Berwick on Tweed; the twenty-eight fatalities included Professor Masson Gulland FRS.
The timing of Gulland’s death allowed the proceedings of the 1947 Cold Spring Harbor Symposium to be dedicated to the memory of ‘this most courteous and charming man’. Erwin Chargaff wrote that Gulland’s death was ‘a sad and irreparable loss to us all’ – but science moves on, and Gulland’s insights into hydrogen bonding in the DNA molecule were soon forgotten.
And an astounding flash of inspiration that hit one of his PhD students went completely unnoticed.
Magic numbers
André Boivin also left Cold Spring Harbor with the fingertips of Fate imprinted on his shoulder. For now, though, he returned to Strasbourg where he and the husband-wife team of Roger and Colette Vendrely were exploring the question of how much DNA was in spermatozoa and ova. Assuming that DNA was the stuff of genes, the quantity of DNA should be halved in these germ-cells, in parallel with the loss of half the number of chromosomes found in every other tissue. Vendrely had devised a laborious method for measuring DNA in single nuclei isolated from bovine tissues and spermatozoa. Allowing for inevitable wobbles in a fiddly technique, the results confirmed their prediction: the DNA content of a single spermatozoon was estimated at 43 per cent of that in nuclei from ordinary tissues.
Mirsky and Ris had started similar experiments and soon found the same result, but the Strasbourg team beat them into print with what became known as the ‘Boivin-Vendrely Rule’ (1948), which states that the nuclear content of DNA is constant in all tissues of a given species, and is halved during germ-cell formation. Even though Boivin and Mirsky were in complete harmony mathematically, their interpretations reinforced their respective beliefs. Boivin saw this as strengthening the case that genes were made of DNA; Mirsky
conceded that DNA was ‘part of the genetic material’ but insisted that ‘this does not mean that the gene consists of nothing but DNA’. Plus ça change, plus c’est la même chose.
Thereafter, Boivin’s luck changed. An American lab failed to replicate his transformation experiments with E. coli, and wrote to ask why. Out of character, Boivin did not answer their letters. His legendary energy had run out, for reasons that were unexplained until cancer was diagnosed in mid-1948. The malignancy was relatively slow to progress, but untreatable. André Boivin died in early July 1949, just over two years after his presentation at Cold Spring Harbor. He was commemorated in the medal which bears his name, awarded annually by the Institut Pasteur in Paris where he had been Deputy Director before moving to Strasbourg.
A fitting memorial can also be found in the Proceedings of the Twelfth Cold Spring Harbor Symposium, in Boivin’s brief ‘glimpse’ into the workings of the cell. Genes, obviously made of DNA, must replicate themselves automatically during cell division. He sees DNA as the ‘primary directing centre’ in the nucleus, which controls the ‘secondary directing centre’ consisting of RNA. In turn, RNA regulates the ‘tertiary directing centres’ – the ‘enzymatic equipment of the cytoplasm’ which run all the processes of life.
Boivin’s ‘glimpse’ was just one of hundreds of ideas thrown out during the meeting, and his ponderous language may have obscured the significance of what he was trying to say. This might explain why the proceedings did not record any interest in his vision of how genes reproduce themselves and create life. But Boivin’s hierarchical scheme – DNA controls RNA, which controls proteins – is precisely the chain of command that we recognise today.
Golden ratios
Meanwhile, Erwin Chargaff had been struggling to pursue his vendetta against the tetranucleotide. His challenge was to pull the four bases out of the soup of DNA breakdown products, and to measure the levels of each one accurately and reproducibly.
Timely inspiration came when Ernst Vischer, a visiting biochemist from Miescher’s former department in Basel, went to hear a lecture about a new method to separate and assay individual amino acids. The lecturer was Archer Martin from Leeds, who with Richard Synge† had discovered that each amino acid migrates at its own speed through a sheet of damp filter paper and forms a specific spot that could be cut out and analysed. Vischer saw the potential for adapting this ‘paper chromatography’ to measure amounts of bases rather than amino acids. Starting with a strip of filter paper which Martin kindly tore off one of his demonstration experiments, Chargaff and Vischer sat down to make it work.
Within weeks of the Cold Spring Harbor meeting, they had used paper chromatography to measure each of the bases in DNA from calf thymus and spleen as well as yeast and tuberculosis bacilli. Consistent patterns began to emerge, which became stronger as they improved the method and diversified into DNA from a wider range of sources; these included human spermatozoa, in an exhaustive experiment that began with volunteers keen enough to produce (collectively) three-quarters of a pint of semen.
The ‘mathematical regularities’ in the proportions of the bases in DNA were later glorified with the title ‘Chargaff’s Rules’, but their discoverer was initially cautious about reading too much into his results, which he regarded as ‘noteworthy, though possibly no more than accidental’. To anyone with an open mind, Chargaff’s First Rule sounded the death knell of the tetranucleotide. None of the DNA sources showed the expected 25:25:25:25 per cent ratios for A:G:C:T that were dictated by Levene’s hypothesis. In calf thymus, for example, the A:G:C:T ratios were 30:20:20:30 per cent. Chargaff’s Second Rule states that the base composition of DNA is ‘characteristic of the species’ – i.e. constant for all the tissues of a given species – but varies between species. For example, tuberculosis bacilli showed a markedly different split from calf thymus, with A:G:C:T of 35:15:15:35 per cent.
Figure 18.3 Chargaff’s Rules.
Chargaff’s findings elevated DNA to a new plane of credibility and potential biological importance. ‘The results served to disprove the tetranucleotide hypothesis,’ he wrote, and some people (but not all) began to believe him. His conclusion that DNA was species-unique also raised the spectre which Mirsky had done his best to exorcise: that DNA might, after all, have enough intrinsic variability and specificity to be the stuff of genes.
These results also threw out a wonderfully cryptic clue that would take several years to solve but would eventually form a magical union with Gulland’s vision of hydrogen bonding inside the DNA molecule. One summer evening in 1948, Chargaff called Vischer over to his desk, where he was pondering a table of the base ratios from different DNA sources. Chargaff had noticed that the proportions of G and C were similar in all sources of DNA – and so were the proportions of A and T, even though the amounts of G and C could differ widely from those of A and T (look again at the percentages above, and then at Figure 18.3).
The remarkable constancy of the base ratios – C = G and A = T – seemed too robust to be a coincidence, but did it mean anything? Like almost everyone else, Chargaff was programmed to think of DNA as a single, long string-like molecule with the bases stacked one above the other as in Astbury’s pile of pennies. On that basis, DNA could consist, for example, of variable numbers of couplets of A+T and C+G lying along the molecule.
There was certainly no reason to consider the possibility that the paired bases – A with T, C with G – might belong to separate strands of DNA.
* This became Chernovtsy in the Ukraine – the town where Nikolai Vavilov was arrested in 1940.
† Later, the only Nobel laureate to be convinced that he had seen the Loch Ness Monster.
19
TWISTS AND TURNS
Arthur Pollister’s seemingly prophetic question at Cold Spring Harbor, about whether the DNA strand could be twisted into a helix, was not the bolt from the blue that it might appear. Evidence that some molecules formed a spiral had been accumulating for over thirty years, and it was only a matter of time before someone wondered if the same might apply to DNA. This is therefore a good moment to introduce the shape which a dictionary might define as a three-dimensional curve on a cylindrical surface.
You can draw a helix by using your forefinger to follow the hands around an imaginary clock face just in front of your nose, and continuing to trace the circle while you smoothly stretch out your arm. Your fingertip will describe a right-handed helix, and will also show you its radius and pitch, the two fundamental measurements of all helices. The radius is simply half the width of your imaginary clock face, and the pitch – the length of each complete turn along the helix – is the horizontal distance that your finger travels each time it passes twelve o’clock.
Natural and artificial helices are everywhere, in and around us. Obvious man-made helices include corkscrews, spiral staircases and the decorative twists inside the stems of expensive eighteenth-century wine-glasses. Nature makes imaginative use of helices, at all levels from the macroscopic to the submicro-scopic. The archetype is the garden snail (Helix in Latin), not forgetting ammonites and winkles. The cucumber plants in Gregor Mendel’s garden could hoist themselves off the ground by extending tendrils which coil into a tight helix around a support and then shorten. The delicate white flowers of the lady’s tresses orchid wind up the stem in a neat right-handed helix. Some beautiful helices can be seen under the microscope, including delicate diatoms and – looking like tiny corkscrews that have drilled into the tissues – the bacteria that cause syphilis and gastric ulcers. Smaller still are the helical flagella that propel bacteria and excited Bill Astbury. And a thousand times tinier are the helices that bring shape and function to countless molecules, including proteins and the nucleic acids.
The first hints that some molecules are helical came during the mid-1930s from simple synthetic polymers. Their X-ray diffraction photographs showed a regular periodicity that could not be explained by a recurring feature along a straight molecule; instead, the structure was ‘
most likely spiral’. It was assumed that forces such as hydrogen bonding, acting between certain atoms within the molecule, lifted the structure into the third dimension and held it in a stable spiral.
Some surprising helices emerged when precise knowledge of the lengths and angles of the links between particular atoms (e.g. carbon to nitrogen, and carbon to hydrogen) was applied to familiar molecules that had been assumed to lie flat, as if drawn on a page. The American biochemist Maurice Huggins realised that alpha-keratin could not have the two-dimensional zigzag structure that had made Astbury famous, because unopposed atomic forces acting along one side would bend the molecule into an arc and wreck any tendency to form long fibres. The structure that best fitted the constraints of bonding was a helix. The difference can be easily appreciated with the help of two open-spiral corkscrews and a sledgehammer. Bash one of the corkscrews flat and you have a zigzag, like Astbury’s alpha-keratin. The unbashed corkscrew is still a helix, like the configuration that Huggins favoured.
Astbury heard the bad news about his structure of alpha-keratin in May 1937, when he visited Huggins’s lab in Rochester, New York, during his American lecture tour. There was worse to come, because Huggins had concluded that ‘none of the models discussed by Astbury’ was viable, for the same reason. When confronted, Astbury vigorously defended his flat zigzags – and left Huggins wondering if he had ever heard of hydrogen bonding.
By now, time was running out for the molecular structures that had carried Astbury to the top of his field. All of his ingenious deductions were about to be made redundant, not by Maurice Huggins but by a brilliant and bored man with a head cold, who decided to while away a couple of hours with a pencil and paper.
Unravelling the Double Helix Page 27