Griffith noted that similar effects had previously been observed in anthrax; the author of the anthrax study, O. Bail, had suggested that the effect involved ‘the inheritance of the capsule-forming substance’. Griffith thought instead that the polysaccharide capsule carried by the dead S cells was being used as a kind of template by the R bacteria to make more capsules. Rather then being inherited, he argued, the ‘specific protein structure’ of the virulent pneumococcus was the cause.
Whatever the explanation – and there was no real evidence as to what was going on – Griffith’s wealth of data and the rigour of his experiments were overwhelming. Almost straight away, Neufeld in Germany replicated the result.* Avery’s group also began to study the effect, getting transformation to occur in a Petri dish rather than in a mouse. By the early 1930s they had made a breakthrough, extracting a substance from S pneumococci that could transform R bacteria. The Avery group called this substance ‘the transforming principle’, and the rest of Avery’s working life was focused on identifying its nature. Shortly after he began studying transformation, Avery received the first of many international awards, the Paul Ehrlich Gold Medal, but severe illness prevented him from attending the ceremony in Germany. For years, Avery had been suffering from Graves’ disease, or hyperthyroidism. This made his eyes bulge out, left him feeling tired and depressed, and gave him a tremor that made it difficult to carry out the delicate and precise microbiological procedures that were his stock in trade. In 1934, Avery went into hospital and had his thyroid removed. It took him months to recover, and it was more than a year before he regained the weight he had lost.
During the summer of 1934 a Canadian physician called Colin MacLeod joined the Rockefeller Institute Hospital, attached to the pneumonia service. When Avery returned from his sick leave, the pair began investigating the chemical nature of the transforming principle. A little more than a year later, Avery explained to his new colleague Rollin Hotchkiss where he thought their study might be going. Hotchkiss recalled:
Avery outlined to me that the transforming agent could hardly be carbohydrate, did not match very well with protein and wistfully suggested that it might be a nucleic acid.
5
There were no clear results to back up Avery’s hunch, as MacLeod’s work had not been conclusive. This caused a problem – the young Canadian needed to strengthen his curriculum vitae with some published articles, so he worked instead on the effectiveness of the new sulphonamide antibiotics. The Avery group did no further research on transformation until 1940.
Despite the fact that the method for separating the transforming principle from bacterial cells had been published, no scientists took up the challenge. This was not because people did not know about or appreciate the significance of pneumococcal transformation. In 1941, the leading evolutionary geneticist Theodosius Dobzhansky published the second edition of his influential book
Genetics and the Origin of Species.
In a chapter entitled ‘Gene mutation’, Dobzhansky described the work of Griffith and Avery and claimed that their findings were ‘not unduly surprising from the standpoint of genetics’, as the change from the R to the S form could be understood in terms of a mutation. More challenging was Griffith and Avery’s demonstration that transformation could take place through contact with a killed sample – Dobzhansky reassured his readers that this ‘extravagant’ finding was ‘conclusively proved’.
6
Dobzhansky emphasised that the transformed strains did not merely acquire ‘a temporary polysaccharide envelope of a kind different from that which their ancestors have had, but are able to synthesize the new polysaccharide indefinitely.’ Dobzhansky’s conclusion was that contact with the transforming principle had somehow induced a mutation in the R bacteria, and that this could lead to the use of targeted mutation to study gene function:
If this transformation is described as a genetic mutation – and it is difficult to avoid so describing it – we are dealing with authentic cases of induction of specific mutations by specific treatments – a feat which geneticists have vainly tried to accomplish in higher organisms … geneticists may profit by devising experiments along the lines suggested by the results of the pneumococcus studies.
7
Dobzhansky was not claiming that the transforming principle was a gene, but the attention he paid to it showed that Avery’s research was widely known and was seen as important.
*
In October 1940, MacLeod and Avery returned to the problem of identifying the nature of the transforming principle. To help with their analyses, they needed a powerful ultracentrifuge that could separate bacterial contents from the rearing medium – as the sample was spun round at high speeds, the heavier molecules sank to the bottom more quickly, concentrating compounds with a similar weight into a narrow band. The Rockefeller Institute had built some of these devices, using a design developed by the Swedish scientist Theodor ‘The’ Svedberg.
8
Avery’s everyday needs were not so demanding – initially his group simply needed to obtain large quantities of bacteria. The solution was to adapt a kitchen cream separator made by the Sharples company. The Sharples, as it was called in the lab, consisted of a tube that was the size of a thick cucumber – about 5 cm in diameter and 25 cm long. There was one problem: the tube was not tightly sealed, and tiny gaps in the apparatus meant that every time it was used, the room became full of an invisible aerosol of potentially lethal bacteria. Sharples was therefore placed in a specially constructed containment device that could be sterilised before opening.
9
Even so, using the equipment safely was no easy matter. After centrifugation, the cake of bacteria that had accumulated at the bottom of the tube had to be removed – this was impossible to do cleanly and ‘one would see small flecks of white material fly in one direction or another’, recalled a lab member.
10
All of the cake was handled with towels soaked in germicide and then heated at 65°C before it was studied further, in an attempt to reduce the risk to lab members.
11
This messy and dangerous procedure so distressed the fastidious Avery that he would leave the lab when the Sharples was in action.
The group soon found that adding calcium chloride to the liquid transforming principle produced a white precipitate that contained most, if not all, of the transforming activity: adding white precipitate from smooth bacteria to a rough colony would transform it into a smooth colony. This white substance was very powerful – even at 1/1,000 dilution it could still transform a rough colony. At the beginning of 1941, MacLeod noted that the white precipitate contained both the polysaccharides typical of the smooth capsule and nucleic acids – DNA and its close relative, ribonucleic acid or RNA. When MacLeod added an enzyme that was known to destroy RNA, this had no effect on the transforming activity of the extract, strongly suggesting that RNA played no role in producing the power of the white material. In April 1941, in his six-monthly report to the Rockefeller Institute, Avery described the progress he and MacLeod had made and hinted at the potential implications:
This study is being continued with the hope that knowledge of this important cellular mechanism may lead to a better understanding of the principles involved in certain induced variations of living cells, not only of the pneumococcus, but also those of other biological systems.
12
*
In the summer, MacLeod left the Institute and another young physician, Maclyn McCarty, joined the Avery group. By the end of November 1941, McCarty had shown that if he used an enzyme to remove the polysaccharide, the extract nevertheless retained its transforming activity, showing that – as expected – the polysaccharide was not involved. That apparently left just two possibilities: proteins or DNA.
In December 1941 the Japanese attacked Pearl Harbor and the US entered the war. The Avery group shifted its work towards more practical aspects of pneumonia as the disease began to appear among US troops. Nevertheless, McCarty continued with his research, and in January 1942 he found that if alcohol was added to the transforming principle, a stringy white material appeared that contained 99.9 per cent of the transforming activity. It soon became apparent that this stringy stuff also contained most of the DNA that was present in the sample.
Two floors above Avery’s office was the laboratory of Alfred E. Mirsky, one of the world’s leading experts on nucleic acids. Mirsky gave the Avery lab some mammalian DNA extracted from the thymus gland, the traditional source of DNA, and they compared it with the white stringy stuff produced by alcohol precipitation of the transforming principle. The two substances seemed to be very similar. McCarty took an extract of transforming principle that had been treated with enzymes to remove both proteins and polysaccharides, and placed it in an ultracentrifuge. After spinning the sample for a few hours at 30,000 r.p.m., a gelatinous ‘pellet’ appeared at the bottom of the tube, containing the heaviest components of the extract. This contained all the transforming activity of the original solution and was apparently composed entirely of DNA.
In the summer of 1942, the suggestion that the transforming principle was made of DNA became stronger when McCarty and Avery showed that enzymes that destroyed transforming activity also affected Mirsky’s DNA samples, and enzymes that had no effect on DNA did not affect the activity of the extract. This should have led to great excitement, but Mirsky was unimpressed. As he explained to the Avery group, the transforming principle could not be made of DNA because nucleic acids were all alike. As their late Rockefeller Institute colleague Phoebus Levene had argued more than three decades earlier in his tetranucleotide hypothesis, the components of nucleic acids – the two kinds of base, the purines (adenine and guanine) and the pyrimidines (cytosine and thymine; thymine is replaced by uracil in RNA) – were present at similar levels. Although DNA was known to be a component of cell nuclei, its apparently boring nature meant that it was not thought to have biological ‘specificity’ – the term used at the time to describe the unique effects of a particular molecule. Proteins, in contrast, were extremely varied, and could be active even at very low levels. It was quite possible that despite all the treatments to remove proteins from their extracts, minute amounts of very powerful protein molecules remained, Mirsky explained.
Although most scientists agreed that DNA did not have the necessary variability to have specificity, some were not so sure. In July 1941, at the annual Symposium on Quantitative Biology held at Cold Spring Harbor Laboratory out on Long Island, Jack Schultz pointed out that the supposed uniformity of nucleic acids was based on a single data point – all the DNA that had been studied had been taken from the thymus gland of cows. The suggestion that DNA structure was uniform could be accepted ‘only as a first order approximation’, he argued: ‘much new data is necessary before we can exclude the possibility of specificities in the nucleic acids themselves’.
13
However, Schultz was firmly convinced that genes were made of what he called nucleoproteins – proteins that were known to be tightly associated with nucleic acids in the chromosomes.
In 1943, Mirsky underlined the growing sense of mystery surrounding nucleic acids when he wrote an article that was supposed to sum up current knowledge about nucleoproteins. Strikingly, he had little to say about proteins and instead concentrated on nucleic acids, and above all on DNA. Mirsky described how the nucleic acid component of a solution could be identified by its reaction to ultraviolet radiation; this was due to the responses of the pyrimidine and purine bases, which apparently lay in rings, perpendicular to the central axis of the molecule. Using this procedure, it was possible to show that, in animals and plants, chromosomes were largely made of DNA, and that DNA was also present in bacteria and in viruses. Finally, it seemed that nucleic acids were involved in both metabolic processes and the replication of chromosomes. Although there was no direct evidence for any link between proteins and genetic functions, Mirsky nevertheless concluded that the proteins found with DNA were at the heart of heredity:
The great accumulation of desoxyribose nucleoproteins in the chromosome strongly suggests that these substances either are the genes themselves or are intimately related to the genes.
In retrospect, virtually all the evidence that Mirsky summarised indicated that DNA was basis of heredity, and yet – like everyone else outside the Avery lab – he argued that genes were made of proteins that were bound up with DNA. He could not see what now appears obvious because there seemed to be no way in which DNA could contain the kind of variability that was necessary to produce the wide range of genetic effects. For Mirsky, Levene’s suggestion that DNA was composed of a monotonous repetition of the four bases was ‘a definite restriction in possible variation among the desoxyribose nucleic acids’.
14
Despite these arguments, Avery and McCarty were increasingly convinced that DNA was the transforming principle and therefore the main component of genes. To prove their point, production of the stuff had to be stepped up – it took 200 litres of bacteria to produce just 40 milligrams of stringy white precipitate. By this stage, McCarty had been called up to active duty by the Naval Reserve research unit. Feeling he should do something related to the war effort, McCarty asked to be put on a more practical project relating to disease treatment, but was told not to worry and to return to Avery’s lab – the main difference that his call-up made was that he now went to the lab in uniform.
In April 1943, Avery’s report to the Rockefeller Institute Board explicitly framed the problem of transformation in terms of genes for the first time. The transforming principle ‘has been likened to a gene’, Avery wrote, and the polysaccharide was like ‘a gene product’. He explained: