Life's Greatest Secret

by Matthew Cobb

  This was one of Boivin’s last lectures. The cancer he had been suffering from returned and he died in July 1949. In the meantime, doubts began to be raised about his reports on transformation in E. coli – researchers in the US were unable to replicate his findings, and his original strains were lost.32 Despite – or perhaps because of – Boivin’s bold statements and prophetic vision, an air of disbelief accumulated around his discoveries. This took years to dissipate – his findings were eventually confirmed in the 1970s, and his vision of the nature of heredity and the future of biology were also shown to be true.33

  *

  By the end of the 1940s, support for the hypothesis that DNA had a fundamental role in heredity had grown much stronger. In the summer of 1950, the cell biologist Daniel Mazia gave a lecture at Woods Hole Marine Biology Laboratory that summed up many people’s thinking. Mazia could not be absolutely certain that genes were made of DNA, but it certainly looked that way: ‘The “physical basis of heredity” is something in the chromosome which may or may not be DNA, but which follows DNA for all practical purposes’, he said.34 Following Boivin, Mazia outlined four criteria that had to be met by what he called the vehicle of heredity, whatever its chemical composition might be. There had to be the same amount in every diploid cell of a given species, that amount should double just before ordinary cell division, and it should be halved during the creation of haploid sex cells. It should be stable, it should be capable of specificity and it should be able to be transferred from one cell to another and act like a gene. Proteins failed at the first hurdle – there was no evidence that the levels of protein in cell nuclei were the same in all tissues of an organism. All that proteins had going for them was that they were known to be complex. Both Chargaff and Gulland had suggested that nucleic acids could vary by the sequence of the bases, perhaps providing a source of complexity. While still not excluding the possibility that proteins were involved, Mazia concluded: ‘DNA is the most likely candidate so far for the role of the material basis of heredity.’35

  A couple of months earlier, in April 1950, Mazia had chaired a session at a special conference on the biochemistry of nucleic acids, held at the US Atomic Energy Commission’s Oak Ridge Laboratory in Tennessee. Among the speakers was Arthur Pollister, who with Mirsky had criticised Avery’s conclusions two years earlier. Pollister was changing his tune; he enthusiastically reported the data from Boivin’s laboratory that showed that the amount of DNA was constant in all diploid cells of a given species, before discussing the idea of a ‘DNA-gene’ and raising the possibility that the chemical structure of the gene was within reach. Nevertheless, Pollister was not completely convinced: the complexity of gene function led him to suggest that ‘important genic components other than DNA remain to be discovered.’36

  Another speaker at the Oak Ridge meeting was Erwin Chargaff, who was acquiring some of the most telling evidence in favour of Avery’s conception of the role of DNA. Chargaff had been an early supporter of Avery’s hypothesis, and in 1950 he presented data on the precise base composition of nucleic acids, using paper chromatography to identify the bases by weight. He found that the proportion of the different bases was constant in all tissues of any species, but differed wildly between species. As Chargaff pointed out, these data disproved the tetranucleotide hypothesis, to the extent that anyone still thought it was true.37 DNA was clearly not boring.

  The 1951 Cold Spring Harbor symposium was on the topic ‘Genes and mutations’, and one of the speakers was Harriett Ephrussi-Taylor, who took the opportunity to survey the seven years that had passed since Avery, MacLeod and McCarty had published their landmark paper. She was downbeat:

  Considering the interest which was aroused by the publication of the results of the chemical and biochemical study of the capsular transforming agent of pneumococcus, it is surprising that so few scientists are at present working in this field.38

  As she admitted, transformation was difficult to study – for example, transformation in Boivin’s E. coli system ‘occurred only with some irregularity’ – and many researchers were not familiar with the pneumococcal system in which transformation had first been described. She glumly concluded that the study of transformation remained isolated: ‘as yet,’ she said, ‘no bridge can be seen leading over into classical genetics’.

  Ephrussi-Taylor’s lament was related to what now appears to be an odd feature of genetics research in the second half of the 1940s – many biologists, including geneticists, simply did not ‘get’ Avery’s result. Not only did they not immediately accept that genes were made of DNA, they did not even attempt to test the hypothesis in the systems they studied. For example, Max Delbrück first heard of Avery’s breakthrough in May 1943, when his Vanderbilt colleague Roy Avery showed him the letter from New York that announced the discovery. Delbrück later recalled his ‘total shock and surprise’ at the contents of the letter, ‘which I read there standing in his office in the spring sunshine’.39 But despite his ‘shock’, Delbrück did nothing. He did not start studying the role of DNA in viruses, nor did any of his colleagues, even though they were all intimately aware of the results that were coming out of Avery’s laboratory. Delbrück later explained that the suggestion that genes were made of DNA left them nonplussed. As he put it ‘you really did not know what to do with it’.40 With the easy wisdom of hindsight, this lack of interest in what led to the most remarkable biological discovery of the twentieth century looks remarkably short-sighted. And at one level, it was. The phage group did not react in the way that Lederberg, Boivin and others did. Their diffident attitude was one component of the failure of Avery’s discovery to immediately transform biology.41 Avery’s findings now look so obvious, and yet many scientists at the time responded to them with hostility or bemusement.

  One of the scientists who did not immediately embrace Avery’s findings was the young Gunther Stent, who worked with Delbrück. In 1972, Stent sought to explain the lack of widespread recognition of the importance of the Avery group’s discovery by suggesting that the result was ‘premature’.42 This term does not explain anything; in fact it obscures the historical reality of how Avery’s work was received, and it does not explain why some scientists accepted the finding but others rejected it. There were two valid criticisms of Avery’s suggestion that DNA was the hereditary material in the transforming principle, each of which gradually became weaker. First, there was the issue of potential protein contaminants, which led the Avery group to employ increasingly precise techniques, the results of which all indicated that protein contamination was not the cause of transformation. Second, there was the conundrum of how exactly specificity might be represented in what were supposed to be boring molecules – if DNA was essentially composed of four bases, a way needed to be discovered that enabled it to bring about the almost infinitely different effects produced by genes. The leading chemists of DNA such as Gulland were happy to imagine that specificity could reside in DNA through the sequence of bases, or their proportions, but this had yet to be demonstrated. Nevertheless, as time wore on, there were fewer reasons not to accept Avery’s findings.

  Some scientists had strong personal reasons to reject the DNA hypothesis. Mirsky’s career was based on the study of nucleoproteins and he was clearly not going to give up his world view without a fight. Through his articles, his lectures and his interventions at conferences, Mirsky sowed doubt among the undecided. Similarly, Wendell Stanley turned a blind eye to the work of the Avery group, even though he too had been familiar with it before publication. In 1936, Stanley had crystallised the tobacco mosaic virus and announced that it was a protein; this was finally shown to be wrong in 1956 – the hereditary material in this virus is in fact RNA, and small amounts of RNA in his protein extract accounted for his results. In 1946, Stanley won the Nobel Prize in Physiology or Medicine for his mistaken claim; he later said that he ‘was not impressed’ by Avery’s discovery – otherwise he would have immediately tested tobacco mosaic virus RNA for specificity. I
n 1970, he concluded, somewhat shamefacedly:

  I have searched my memory and have failed to find any really extenuating circumstances for my failure to recognize the full significance of the discovery of transforming DNA.43

  The diffident response of the main members of the phage group – Delbrück, Luria and Hershey – had a rather different source, and all three of them later explained their behaviour in the same way: they were interested in genetics, not chemistry, and so they simply did not realise the potential implications. Typically robust, Delbrück said:

  And even when people began to believe it might be DNA, that wasn’t really so fundamentally a new story, because it just meant that genetic specificity was carried by some goddamn other macromolecule, instead of proteins.44

  Luria recalled: ‘I don’t think we attached great importance to whether the gene was protein or nucleic acid. The important thing for us was that the gene had the characteristics that it had to have.’45 In 1994, Hershey explained that their focus was simply elsewhere – ‘as long as you’re thinking about inheritance, who gives a damn what the substance is – it’s irrelevant.’46 Ironically, Hershey is now best known for his attempt to resolve the issue of whether proteins or DNA are the basis of heredity, an experiment that students are now taught settled the question once and for all, even though it did not.

  *

  Alfred (‘Al’) Hershey was a tall, skinny taciturn man with a toothbrush moustache and bad teeth. Although he was renowned for working long into the night, he was not solely focused on science – he often took afternoon naps and in the summer he would disappear for weeks on end, sailing his yacht on Lake Michigan. Like everyone else in the phage group, Hershey had followed the discussions around the chemical nature of Avery’s transforming principle. In May 1949, Hotchkiss sent Hershey an update on his progress in excluding any possible protein contamination from the DNA extracts of the transforming principle; after looking at the data, Hershey wrote to the younger man: ‘The experiments are very beautiful. … My own feeling is that you have cleared up most of the doubts.’47 But like Luria and Delbrück, Hershey’s initial interest in Avery’s experiments was unfocused – the members of the phage group could not see how chemistry could help them understand genetics.

  Nevertheless, as phage researchers tried to understand how viruses reproduced, the question of chemistry became increasingly pressing. By 1949, electron microscope images had shown that a viral infection begins with the virus sitting on the outside of a cell; in ways that were unclear, the virus then took over the cell’s metabolic system and ‘lost its identity’ – no viruses could be detected inside the cell for a period, while the viral structures that were still on the outside of the bacterium lost their infective power. It was possible to burst viruses by subjecting them to a sudden change in the concentration of their surrounding medium; all that remained were ghost viruses – empty protein shells that would happily adhere to the outside of a bacterium but were not infectious. Researchers had begun to use radioactive tracers to explore this phenomenon – by growing phage and bacteria on radioactively labelled medium, radioactive phosphorus was taken up by nucleic acids, and radioactive sulphur could be used to mark proteins. It was therefore possible to track the fates of the two components of the phage virus, namely DNA and protein, by using radioactivity.

  By late 1950, several phage researchers had begun to sketch out a hypothesis about the roles of DNA and protein in virus replication, explicitly acknowledging that Avery was right. John Northrop of Berkeley concluded one of his articles with an outline of this thinking:

  The nucleic acid may be the essential, autocatalytic part of the molecule, as in the case of the transforming principle of the pneumococcus (Avery, MacLeod, and McCarty, 1944), and the protein portion may be necessary only to allow entrance to the host cell.48

  Roger Herriott of Johns Hopkins University wrote to Hershey:

  I’ve been thinking – and perhaps you have, too – that the virus may act like a little hypodermic needle full of transforming principles; that the virus as such never enters the cell; that only the tail contacts the host and perhaps enzymatically cuts a small hole through the outer membrane and then the nucleic acid of the virus flows into the cell.49

  Thomas Anderson later recalled:

  I remember in the summer of 1950 or 1951 hanging over the slide projector table with Hershey, and possibly Herriott, in Blackford Hall at the Cold Spring Harbor Laboratory, discussing the wildly comical possibility that only the viral DNA finds its way into the host cell, acting there like a transforming principle in altering the synthetic processes of the cell.50

  It was in this context that Al Hershey, together with his new technician, Martha Chase, decided to settle the matter. Hershey had recently moved to Cold Spring Harbor and had equipped his laboratory with the latest radioisotope technology.51 Chase, who was only 23 years old when she joined Hershey, had a round face and short hair; generally she was as reserved as her boss, but she was nonetheless prepared to complain loudly about her low pay.52 Their experiments, which were published in the Journal of General Physiology in the middle of 1952, have since taken on an iconic quality.53 They are reproduced in textbooks and are presented as a turning point, because they are now seen as showing that genes are made of DNA. The reality is rather different.

  The Hershey and Chase paper describes several experiments in which they tried to identify the functions of protein and nucleic acid in bacteriophage reproduction. First, they confirmed and extended previous findings about the function and composition of ghost phage, which they showed were made of protein, were not infectious, could still attach to bacteria, and protected their DNA contents from enzyme attack. They next demonstrated that when the phage settled onto a bacterium, it injected DNA into the cell. All this supported Herriott’s hypodermic needle hypothesis, but there was no proof of what the DNA actually did, nor could they be certain that no protein entered the bacterial cell.

  The final experiments are those most often taught to students today, but they are usually described inaccurately. They all involved the use of a Waring Blender, or Blendor as the Waring company trademark had it. This device was employed to agitate the viruses and their bacterial hosts, and the experiments that used it are now often known as the Blender experiments. This apparatus is often called a kitchen blender, which conjures up some kind of retro 1950s domestic device, all chrome and glass. Sadly this was not the case. Although the Waring company did make kitchen blenders, the apparatus used by Hershey and Chase was a highly specialised, unstylish bronze-coloured piece of laboratory equipment about 25 cm high that could run at speeds of up to 10,000 r.p.m. – much faster than anything you would have in your kitchen. It was not simply a centrifuge, it also produced what Hershey and Chase described as ‘violent agitation’, which they used to shake the protein-rich viral ghosts from the outside of the host cell. Using radioactive sulphur, they showed that they could remove up to 82 per cent of the phage protein from their preparations by separating out the ghost phage; a similar experiment with radioactive phosphorus showed that up to 85 per cent of the virus DNA was transferred into the bacterial cell.

  Students are now generally taught that these experiments provided the evidence that DNA is the genetic material, but in fact they did no such thing, nor did Hershey and Chase claim that they did. The problem faced by Hershey and Chase was similar to that encountered by Avery and his colleagues, but in spades. Hotchkiss had reduced the protein component in his version of Avery’s experiment to effectively zero (at most 0.02 per cent), and still people did not accept his findings; in Hershey and Chase’s extracts around 20 per cent of the protein was still floating around. It was quite possible that some of this protein played a role in the reproduction of the virus. Furthermore, as Hershey and Chase put it, none of the experiments proved anything more than that DNA had ‘some function’ in viral reproduction. The paper concluded with Hershey’s typical terseness, beginning with the question of ‘adsorption’, or
how the virus sticks to the outside of the bacteria:

  The sulfur-containing protein of resting phage particles is confined to a protective coat that is responsible for the adsorption to bacteria, and functions as an instrument for the injection of the phage DNA into the cell. This protein probably has no function in the growth of intracellular phage. The DNA has some function. Further chemical inferences should not be drawn from the experiments presented.54

  Hershey remained troubled by his findings and later admitted, ‘I wasn’t too impressed by the results myself’.55 When he first presented his experiments in a small laboratory seminar at Cold Spring Harbor, he expressed his surprise that protein apparently had no function inside the infected cell. And when he made his first public presentation of the results, at the June 1953 Cold Spring Harbor meeting, speaking after the double helix structure of DNA had been described, Hershey was still sure that DNA could not be the sole carrier of hereditary specificity. He addressed this issue head-on by summarising the evidence from Avery, Boivin, Taylor and himself as follows:

  1.The amount of DNA in chromosomes is consistent in a species, not in a given kind of tissue in different species.

  2.DNA can transform bacteria.