Life's Greatest Secret
Page 2
The outcome of this collaboration was a joint German-language publication that appeared in 1935, called ‘On the nature of gene mutation and gene structure’, more generally known as the Three-Man Paper.19 The article summarised nearly forty studies of the genetic effects of radiation and included a long theoretical section by Delbrück. The trio concluded that the gene was an indivisible physicochemical unit of molecular size, and proposed that a mutation involved the alteration of a chemical bond in that molecule. Despite their best efforts, however, the nature of the gene, and its exact size, remained unknown. As Delbrück explained in the paper, things were no further on from the alternatives posed by Morgan in 1919:
We will thus leave unresolved the question of whether the individual gene has a polymeric form that arises through the repetition of identical structures of atoms, or whether it exhibits no such periodicity.20
The Russian geneticist Nikolai Koltsov was bolder than Delbrück or Morgan. In a discussion of the nature of ‘hereditary molecules’ published in 1927, Koltsov, like Delbrück, argued that the fundamental feature of genes (and therefore of chromosomes) was their ability to replicate themselves perfectly during cell division.21 To explain this phenomenon, Koltsov proposed that each chromosome consisted of a pair of protein molecules that formed two identical strands; during cell division, each strand could be used as a template to produce another, identical, strand. Furthermore, he suggested that because these molecules were so long, the amino acid sequences along the proteins could provide massive variation that might explain the many functions of genes.22 However perceptive this idea might look in the light of what we now know – the double helix structure of DNA and the fact that genes are composed of molecular sequences – Koltsov’s argument was purely theoretical. Furthermore, it was not unique – in a lecture given in 1921, Hermann Muller picked up on a suggestion by Leonard Troland from 1917 and drew a parallel between the replication of chromosomes and the way in which crystals grow:
each different portion of the gene structure must – like a crystal – attract to itself from the protoplasm materials of a similar kind, thus moulding next to the original gene another structure with similar parts, identically arranged, which then become bound together to form another gene, a replica of the first.23
In 1937, the British geneticist J. B. S. Haldane came up with a similar idea, suggesting that replication of genetic material might involve the copying of a molecule to form a ‘negative’ copy of the original.24 Koltsov’s views were initially published in Russian and then translated into French, but like Haldane’s speculation they had no direct influence on subsequent developments.25 Koltsov died in 1940, aged 68, having been accused of fascism because of his opposition to Stalin’s favoured scientist, Trofim Lysenko, who denied the reality of genetics.26
Koltsov’s assumption that genes were made of proteins was widely shared by scientists around the world. Proteins come in all sorts of varieties that could thereby account for the myriad ways in which genes act. Chromosomes are composed partly of proteins but mainly of a molecule that was then called nuclein – what we now call deoxyribonucleic acid, or DNA. The composition of this substance showed little variability – the leading expert on nucleic acids was the biochemist Phoebus Levene, who for over two decades explained that nucleic acids were composed of long chains of repeated blocks of four kinds of base (in DNA these were adenine, cytosine, guanine and thymine – subsequently known by their initials – A, C, G and T) which were present in equal proportions.27 This idea, which was called the tetranucleotide hypothesis (‘tetra’ is from the Greek for four) dominated thinking about DNA; it suggested that these long and highly repetitive molecules probably had some structural function, unlike the minority component of chromosomes, proteins, which were good candidates for the material basis of genes simply because they were so variable. As Swedish scientist Torbjörn Caspersson put it in 1935:
If one assumes that the genes consist of known substances, there are only the proteins to be considered, because they are the only known substances which are specific for the individual.28
This protein-centred view of genes was reinforced that same year when 31-year-old Wendell Stanley reported that he had crystallised a virus, and that it was a protein.29 Stanley studied tobacco mosaic disease – a viral disease that infects the tobacco plant. Stanley took an infected plant, extracted its juice and was able to crystallise what looked like a pure protein that had the power to infect healthy plants. Although viruses were mysterious objects, in 1921 Muller had suggested that they might be genes, and that studying them could provide a route to understanding the nature of the gene.30 Viruses, it appeared, were proteins, so presumably genes were, too. During the 1930s, many researchers, including Max Delbrück, began studying viruses, which were considered to be the simplest forms of life. Whether viruses are alive continues to divide scientists; whatever the case, this approach of studying the simplest form of biological organisation was extremely powerful. Delbrück, along with his colleague Salvador Luria, focused on bacteriophages (or ‘phage’) – viruses that infected bacteria, and in the 1940s an informal network of researchers called the phage group grew up around the pair as they tried to make fundamental discoveries that would also apply to complex organisms.31
Stanley’s discovery caused great excitement in the press – for the New York Times it meant that ‘the old distinction between life and death loses some of its validity’. Although within a few years valid doubts were expressed about Stanley’s claim that he had isolated a pure protein – water and other contaminants were present and, as he admitted, it was nearly impossible to prove that a protein was pure – the overwhelming view among scientists was that genes, and viruses, were proteins.32
The most sophisticated attempt to link this assumption with speculation about the structure of genes was made in 1935 by the Oxford crystallographer Dorothy Wrinch. In a talk given at the University of Manchester, she suggested that the specificity of genes – their ability to carry out such a wide variety of functions – was determined by the sequence of protein molecules that were bound perpendicularly to a scaffold of nucleic acids, a bit like a piece of weaving. As she emphasised, however, ‘there is an almost complete dearth of experimental and observational facts upon which the testing and further development of the hypothesis now put forward must necessarily depend.’ Nevertheless, her conclusion was optimistic, as she encouraged her colleagues to explore the nature of the chromosome and of the gene:
The chromosome is not a phenomenon belonging to a closed field. Rather it should take its place among the objects worthy of being treated with all possible subtleties and refinement of concept and technique belonging to all the sciences. A concerted attack in which the full resources of the world state of science are exploited can hardly fail.33
*
In the 1930s, most geneticists were not particularly concerned with finding out what genes are made of; they were more interested in discovering what genes actually do. There was a potential link between these two approaches. As the Drosophila geneticist Jack Schultz put it in 1935, by studying the effects of genes it should be possible ‘to find out something about the nature of the gene’.34 One of the scientists who took Schultz’s suggestion very seriously was George Beadle, who had studied the genetics of eye colour in Drosophila in Morgan’s laboratory, alongside the Franco-Russian geneticist Boris Ephrussi. When Ephrussi returned to Paris, Beadle followed him to continue their work. Their objective was to establish the biochemical basis of the mutations that changed the eye-colour of Drosophila flies. Beadle and Ephrussi’s experiments failed: the biochemistry of their system was too complicated, and they were unable to extract the relevant chemicals from the fly’s tiny eyes. They knew the genes that were involved, and they knew the effect they had on eye colour, but they did not know why.
Beadle returned to the US, determined to crack the problem of how genes could affect biochemistry, but equally certain that he had to use an organism that could
be studied biochemically. He found the answer in the red bread mould Neurospora. This hardy fungus can survive in the near absence of an external supply of vitamins because it synthesises those it needs. To gain an insight into the genetic control of biochemical reactions, Beadle decided to create Neurospora mutants that could not synthesise these vitamins.
Together with microbiologist Edward Tatum, Beadle followed Muller’s approach and irradiated Neurospora spores with X-rays in the hope of producing mutant fungi that required added vitamins to survive, thereby opening up the possibility of studying the genetics of vitamin biosynthesis. Beadle and Tatum soon found mutants that were unable to synthesise particular vitamins, and published their findings in 1941.35 Each mutation affected a different enzymatic step in the vitamin’s biosynthetic pathway – this was experimental proof of the widely held view, going back to the beginning of the century, that genes either produced enzymes or indeed simply were enzymes.36 When Beadle presented their findings at a seminar at the California Institute of Technology (Caltech) in Pasadena, the audience was stunned. He spoke for only thirty minutes and then stopped. There was a nonplussed silence – one member of the audience recalled:
We had never heard such experimental results before. It was the fulfilment of a dream, the demonstration that genes had an ascertainable role in biochemistry. We were all waiting – or perhaps hoping – for him to continue. When it became clear that he actually was finished, the applause was deafening.37
In the following year, Beadle and Tatum suggested that ‘As a working hypothesis, a single gene may be considered to be concerned with the primary control of a single specific chemical reaction.’38 A few years later, a colleague refined this to the snappier ‘one gene, one enzyme hypothesis’. There was support for this view from work on human genetic diseases such alkaptonuria – in 1908 Archibald Garrod had suggested that this disease might involve defective enzyme production. But Beadle and Tatum’s hypothesis met with opposition at the time, partly because it was known that genes have multiple effects, while their hypothesis – or rather, the ‘one gene, one enzyme’ catch-phrase by which it came to be known – seemed to suggest that each gene did only one thing: control an enzyme.39
*
Trinity College sits in the heart of Dublin, its grey three-storey neoclassical buildings positioned around lawns and playing fields. At the eastern end of the campus there is another grey building, built in 1905 in a rather different style. This is the Fitzgerald Building, or the Physical Laboratory as it is called in deeply engraved letters on the stone lintel. On the top floor there is a lecture theatre, and in the late afternoon of the first Friday of February 1943, around 400 people crowded onto the varnished wooden benches. According to Time magazine, among those lucky enough to get a seat were ‘Cabinet ministers, diplomats, scholars and socialites’, as well as the Irish Prime Minister, Éamon de Valera.40 They were there to hear the Nobel Prize-winning physicist Erwin Schrödinger give a lecture with the intriguing title ‘What is life?’ The interest was so great that scores of people were turned away, and the lecture had to be repeated the following Monday.41
Schrödinger had arrived in Dublin after fleeing the Nazis – he had been working at Graz University in Austria when the Germans took over in 1938. Although he had a reputation as an opponent of Hitler, Schrödinger published an accommodating letter about the Nazi takeover, in the hope of being left alone. This tactic failed, and he had to flee the country in a hurry, leaving his gold Nobel medal behind. De Valera, who was interested in physics, offered Schrödinger a post in Dublin’s new Institute for Advanced Studies, and the master of quantum mechanics found himself in Ireland.42
On three consecutive Fridays, 56-year-old Schrödinger walked into the Fitzgerald Building lecture theatre to give his talks, in which he explored the relation between quantum physics and recent discoveries in biology.43 His first topic was the way in which life seems to contradict the second law of thermodynamics. Since the nineteenth century it has been known that, in a closed system, energy will dissipate until it reaches a constant and even level: physicists explain this in terms of the increasing amount of disorder, or entropy, that inevitably appears in such systems. Organisms seem to contradict this fundamental law because we are highly ordered forms of matter that concentrate energy in a very restricted space. Schrödinger’s explanation was that life survives ‘by continually sucking orderliness from its environment’ – he described order as ‘negative entropy’. This apparent breach of one of the fundamental laws of the Universe does not cause any problems for physics, because on a cosmological scale our existence is so brief, our physical dimensions so minute, that the iron reality of the second law does not flutter for an instant. Whether life exists or not, entropy increases inexorably. According to our current models, this will continue until the ultimate heat death of the Universe, when all matter will be evenly spaced and nothing happens, and it carries on not happening forever.
Schrödinger encountered far greater difficulties when he came to discuss his second topic: the nature of heredity. Like Koltsov and Delbrück before him, Schrödinger was struck by the fact that the chromosomes are accurately duplicated during ordinary cell division (‘mitosis’ – this is the way in which an organism grows) and during the creation of the sex cells (‘meiosis’). For your body to have reached its current size there have been trillions of mitotic cell divisions and through all that copying and duplicating the code has apparently been reliably duplicated – in general, development proceeds without any sign of a mutation or a genetic aberration. Furthermore, genes are reliably passed from one generation to another: Schrödinger explained to his audience that a well-known characteristic such as the Hapsburg, or Habsburg, lip – the protruding lower jaw shown by members of the House of Hapsburg – can be tracked over hundreds of years, without apparently changing.
For biologists, this apparently unchanging character of genes was simply a fact. However, as Schrödinger explained to his Dublin audience, it posed a problem for physicists. Schrödinger calculated that each gene might be composed of only a thousand atoms, in which case genes should be continuously shimmering and altering because the fundamental laws of physics and chemistry are statistical; although overall atoms tend to behave consistently, an individual atom can behave in a way that contradicts these laws.44 For most objects that we encounter, this does not matter: things such as tables or rocks or cows are made of so many gazillions of atoms that they do not behave in unpredictable ways. A table remains a table; it does not start spontaneously turning into a rock or a cow. But if genes are made of only a few hundred atoms, they should display exactly that kind of uncertain behaviour and they should not remain constant over the generations, argued Schrödinger. And yet experiments showed that mutations occurred quite rarely, and that when they did happen they were accurately inherited. Schrödinger outlined the problem in the following terms:
incredibly small groups of atoms much too small to display exact statistical laws … play a dominating role in the very orderly and lawful events within a living organism. They have control of the observable large-scale features which the organism acquires in the course of its development, they determine important characteristics of its functioning; and in all this very sharp and very strict biological laws are displayed.45
The challenge was to explain how genes act lawfully, and cause organisms to behave lawfully, while being composed of a very small number of atoms, a significant proportion of which may be behaving unlawfully. To resolve this apparent contradiction between the principles of physics and the reality of biology, Schrödinger turned to the most sophisticated theory of the nature of the gene that existed at the time, the Three-Man Paper by Timoféef-Ressovsky, Zimmer and Delbrück.
As Schrödinger explored the nature of heredity for his audience, he was forced to come up with an explanation of what exactly a gene contained. With nothing more than logic to support his hypothesis, Schrödinger argued that chromosomes ‘contain in some kind of code-s
cript the entire pattern of the individual’s future development and of its functioning in the mature state.’ This was the first time that anyone had clearly suggested that genes might contain, or even could simply be, a code.
Taking his idea to its logical conclusion, Schrödinger argued that it should be possible to read the ‘code-script’ of an egg and know ‘whether the egg would develop, under suitable conditions, into a black cock or into a speckled hen, into a fly or a maize plant, a rhododendron, a beetle, a mouse or a woman.’46 Although this was partly an echo of the earliest ideas about how organisms develop and the old suggestion that the future organism was preformed in the egg, Schrödinger’s idea was very different. He was addressing the question of how the future organism was represented in the egg and the means by which that representation became biological reality, and suggesting these were one and the same:
The chromosome structures are at the same time instrumental in bringing about the development they foreshadow. They are law-code and executive power – or, to use another simile, they are architect’s plan and builder’s craft – in one.47
To explain how his hypothetical code-script might work – it had to be extremely complicated because it involved ‘all the future development of the organism’ – Schrödinger resorted to some simple mathematics to show how the variety of different molecules found in an organism could be encoded. If each biological molecule were determined by a single 25-letter word composed of five different letters, there would be 372,529,029,846,191,405 different possible combinations – far greater than the number of known types of molecule found in any organism. Having shown the potential power of even a simple code, Schrödinger concluded that ‘it is no longer inconceivable that the miniature code should precisely correspond with a highly complicated and specified plan of development and should somehow contain the means to put it into operation.’48