Gene genius

One of the most important discoveries of the 20th century was that of the structure and function of DNA and proteins. Max Perutz traces the history of the work that led scientists to decipher the human genome and find the architecture of the proteins for which it codes. And now we have the knowledge, he says, the important thing is that we use it to relieve suffering and that we do no harm.

* 1938, Warren Weaver, the mathematician who headed the Natural Sciences Division of the Rockefeller Foundation in New York, submitted a report to his president in which he first coined the term molecular biology.

All organisms are chemical factories in which thousands of chemical reactions cooperate in an orderly manner to produce the miracle of life. Weaver was aware that the workhorses of the living cell are proteins - complex molecules made up of thousands of atoms - and that a specific protein is needed for each of those reactions. It was clear that their highly specialised functions were determined by the arrangement of the atoms within them, but these were unknown. Protein molecules were black boxes. Weaver channelled the foundation's money to scientists attacking that fundamental problem, and I became one of the beneficiaries of his enlightened policy.

I had decided as a 21-year-old chemistry student in Vienna to work on these then new vitamins or proteins for my PhD, and Cambridge seemed the best university for such research. I found a place as a graduate student in the physics department, the famous Cavendish Laboratory headed by Lord Rutherford, discoverer of the atomic nucleus. I landed in the sub-department of crystallography, directed by John Desmond Bernal, a charismatic Irishman whom we called Sage because he knew everything.

Together with the young Dorothy Crowfoot (later Hodgkin) Bernal had discovered that, in principle, it would be possible to unravel the atomic arrangement in proteins by X-ray crystallography, a physical method that had already been used to determine the arrangements of the atoms in minerals and simple chemical compounds. Inspired by him, I decided to have a go at haemoglobin, the breathing molecule of our red blood cells that transports oxygen from the lungs to the tissues and helps the return transport of carbon dioxide from the tissues back to the lungs. Never mind that the haemoglobin molecule contained 10,000 atoms, compared with the fewer than 50 in the molecules whose structures had been solved so far. My fellow graduate students thought me mad for taking on such a hopeless problem, and in the years to come there were to be many moments when it seemed as though they had been right. Studies of the structure and function of proteins became what was called molecular biology here in England.

At about the time I arrived in Cambridge, a young German physicist in Berlin, Max Delbruck, won a Rockefeller Travelling Fellowship to work with the famous geneticist T. H. Morgan at the California Institute of Technology in Pasadena. He had used physics to determine the dose of X-rays needed to induce a genetic mutation in the fruit fly Drosophila, and had concluded that the gene is a molecule containing no more than a few hundred atoms - but this had not told him what kind of molecule it is.

He was disappointed at first, because Morgan's extensive analyses of the genetics of the fruit fly failed to lend themselves to theoretical interpretation, but then he discovered in the basement of the same building another biologist, E. L. Ellis, who worked on viruses that infect bacteria. They were called phages. Delbruck and Ellis discovered that a single phage entering a single coli bacterium perverts its reproductive machinery, so that it bursts 20 minutes later with the release of 60 progeny phages. Analysis of the mutations arising in this simplest of self-reproducing organisms opened the immensely fruitful field of phage genetics that came to be what Americans understood by molecular biology.

Delbruck was a brilliant, tall and handsome German romantic, attracted to biology by the great Danish physicist Niels Bohr, who had given a lecture about light and life in which he predicted that the study of biology would lead to the discovery of new laws of physics. This piece of fantasy was later taken up by Erwin Schrodinger in his bestselling book What is Life?, which induced several equally romantic young physicists to switch to biology. Bohr's dream remained Delbruck's Holy Grail for the rest of his life.

After his pioneering research with Ellis, Delbruck joined forces with Salvador Luria, a young Italian bacteriologist who had found refuge from Mussolini's anti-Jewish laws in Bloomington, Indiana. The paper that was to earn them the Nobel prize 26 years later was conceived in 1942. Luria tried to discover whether bacterial resistance to phage infection was caused by an adaptive change, as many believed, or whether it arose from mutations. An adaptive change is one acquired in response to a changed environment; for example, we become adapted to life at high altitudes by a rise in the number of our red blood cells. On the other hand, changes due to mutations are passed on to future generations; a very few people carry mutations that adapt them hereditarily to high altitudes.

Luria was perplexed by the extreme variability of the numbers of resistant bacteria present in different cultures of the same organisms, until the correct explanation dawned on him one night while watching a game machine. If the change from susceptibility to resistance was a random event due to mutations, then a mutation occurring early in the life of a culture would give rise to a large clone of resistant bacteria, while several mutations arising later would each produce only small clones. Luria wrote to Delbruck, telling him of his idea; Delbruck put it into mathematical form and proved rigorously that the distribution of resistant bacteria in Luria's different cultures was consistent only with their being due to random mutations, and that those mutations occurred with a constant frequency of one in 250 million bacteria divisions. These results opened up the field of bacteria genetics. Just like Delbruck and Ellis's earlier results on phage, they involved nothing that could not have been found out years earlier; the only new ingredient was clear thought.

Meanwhile, another bacteriologist, Oswald Avery, and his associates, Colin MacLeod and Maclyn McCarty at the Rockefeller Institute of Medical Research in New York, had discovered that genes are made not of proteins, as most scientists believed, but of deoxyribonucleic acid, DNA for short. It was the unsung climax of a 30-year saga, of a piece of research begun in order to discover why some patients died from pneumonia while others recovered, with no thought about the nature of the gene. DNA was discovered by Swiss biochemist Friedrich Miescher in 1872, but until Avery, MacLeod and McCarty's discovery in 1944, no one knew what it does. Without that crucial and totally unexpected discovery, there would be no DNA double helix and no deciphering of the human genome.

In 1947, an 18-year-old genetics graduate from the University of Chicago called Jim Watson joined Luria to work for his PhD. One day in 1951, he opened the door to Kendrew's and my office at the Cavendish Laboratory and asked: "Can I come and work here?" By then my seemingly hopeless undertaking to solve the structure of haemoglobin by X-ray crystallography had attracted several enterprising young men; among them were John Kendrew, a chemist like myself, and Francis Crick and Hugh Huxley, both physicists. We shared the belief that the nature of life could be understood only by getting to know the atomic structure of proteins, and that physics and chemistry would open the way, if only we could find it.

Proteins are made of long chains composed of 20 different kinds of links, called amino acids. In 1958 my Cambridge colleague, Fred Sanger, received his first Nobel prize for inventing chemical methods that allowed him to determine the sequence of the different amino acids along the two chains of the antidiabetic hormone insulin, and for showing that this sequence was specified by a gene. It was a sensational discovery, because it had been uncertain whether the amino acids are arranged in a definite sequence, and it was the first indication of what genes actually do. Determining the sequence of the amino acids in proteins is now known to be their main function.

In his bestselling book, The Double Helix, Watson portrays himself as a brash western cowboy entering our genteel circle, but this is a caricature. Watson's arrival had an electrifying effect on us, because he made us look at our problems from the genetic point of view. He asked not just, "what is the atomic structure of living matter?" but foremost, "what is the structure of the gene that determines it?" Watson found an echo in Crick, who had begun to think along similar lines.

Crick was 34, Watson 22, a whizz-kid from Chicago who had entered university aged 15 and got his PhD in genetics at 20. They shared the sublime arrogance of men who rarely met their intellectual equals. Crick was tall, fair, dandyish and talked volubly, each phrase in his King's English strongly accented and punctuated by eruptions of laughter that reverberated through the laboratory. By contrast, Watson went around like a tramp, making a show of not cleaning his one pair of shoes for an entire term (an eccentricity in those days), and dropped his sporadic nasal utterances in a low monotone that faded before the end of each sentence and was followed by a snort.

To say that they did not suffer fools gladly would be an understatement - Crick's comments would cut like daggers at non sequiturs and Watson demonstratively unfolded his newspaper at seminars that bored him. Watson had put Crick's mind to the structure of DNA, yet their relationship was something of teacher and pupil because there was little that Watson could teach Crick, but much that Crick would teach Watson. Crick had a profound understanding of that hardest of the sciences, physics, without which the structure of DNA would never have been solved. This crucial fact is obscured in Watson's book, but Watson had an intuitive knowledge of the features that DNA ought to have if it were to make genetic sense.

Crick and Watson often achieved most when they seemed to be working least. They did an immense amount of hard work, studying while hidden away, but when you saw them they were more likely to be engaged in argument and apparently idle. This was their way of attacking a problem that could be solved only by a tremendous leap of the imagination, supported by profound knowledge. Imagination comes first in both artistic and scientific creations. But in science Nature always looks over your shoulder. To paraphrase Winston Churchill, "in science you don't need to be polite, you only have to be right".

Watson and Crick solved the structure of DNA in March 1953. In August the same year, I found the key to the solution of protein structures by X-ray crystallography, and from that moment I knew that Kendrew and I were not wasting our lives on a hopeless problem. Kendrew had begun to work on myoglobin, a simpler relative of haemoglobin that stores oxygen in muscle. By 1957 he obtained the first, rough, three-dimensional picture of that molecule which made him famous overnight. In 1959 he was able to construct a model that revealed the positions of most of its 2,500 atoms, and in the same year my collaborators and I were able to build the first 3-D model of haemoglobin. Kendrew's and my solutions of the first protein structures earned us the Nobel prize for chemistry, while Crick, Watson and Maurice Wilkins received the prize for physiology and medicine, for their double-helical model of DNA. Rosalind Franklin, whose X-ray diffraction pictures provided the proof that the model was right, would have been a strong candidate had she not died of cancer in 1958.

Protein crystallography, the crazy enterprise on which I embarked for my PhD 63 years ago, has now come of age. New protein structures are being solved at a rate of nearly 2,000 a year, and plans are afoot to raise that number fivefold in order to determine the structures of all the proteins specified by the probably more than 60,000 genes in human chromosomes. All self-respecting pharmaceutical firms now employ protein crystallographers, because the atomic structures of proteins involved in diseases allow new drugs to be developed by rational design. Several effective new drugs against Aids have been developed in this way. The next century is likely to bring the design of new proteins for chemistry, agriculture and medicine.

The genetic information of all living organisms is written in a four-letter code on long chains of nucleic acids. The human genome contains about 3,000 million letters which code for about 60,000 genes - equivalent to a library of thousands of volumes - but their writing is on the atomic scale, a million times smaller than the smallest things visible to the eye, and this allows all the enormous amount of information that specifies a human being to be packed into a single sperm and a single egg.

Fred Sanger won his second Nobel prize in 1980 for inventing an ingenious chemical method for deciphering genes. This has been automated and robots can now decipher them at a rate of about 100 letters an hour. Hundreds of these robots in 15 different centres are fast deciphering the entire human genome and are expected to complete their task by 2002.

The functions of many of the proteins specified by many of the genes deciphered so far are still unknown. When they do become known, what benefits and dangers will that knowledge present? Medical scientists are sanguine about the benefits: they expect them to lead to deeper understanding and more rational treatments of many diseases, from mental disease to cancers. The genes for the most common inherited diseases have already been identified and deciphered, but our hope that this would soon lead to effective treatments has so far been disappointed.

There are three possible approaches to treatment of genetic diseases: either by drugs, or by somatic or germ line therapy. In somatic therapy you try to treat a patient with a malfunctioning gene, say a child with cystic fibrosis, with the healthy gene. In germ line therapy the healthy gene is introduced into the fertilised egg or into early embryonic cells. Somatic gene therapy raises no ethical concerns, but has proved unexpectedly difficult. Most current attempts are aimed at cancer cells rather than at genetic diseases. Germ line therapy is ethically unacceptable, because animal experiments have shown it to be too hazardous.

This is one of the crucial arguments against the creation of genetically engineered humans. Another is that most human traits are determined by a multiplicity of genes whose effects are later tuned by environmental factors, which would render the outcome of any genetic interference uncertain. For these and other reasons, any attempt at creating genetically modified humans is now - and should remain for the foreseeable future - illegal.

Identification of the genes for certain inherited diseases has brought some important benefits. Thalassaemia is a crippling, often fatal, blood disease that is transmitted recessively. If both parents are healthy carriers, one in four children is likely to inherit it from them. Isolation of its genes has allowed it to be diagnosed at an early stage of pregnancy and has given parents the choice of termination and waiting for the conception of a healthy baby. This has brought immense relief to many couples and has led to the near-eradication of thalassaemia from regions such as Sardinia and Cyprus, where it used to be endemic.

On the other hand, there is concern that antenatal diagnosis could one day be abused, to rid couples of babies of the wrong sex or simply of babies with unwanted physical traits.l Max Perutz won the Nobel prize for chemistry in 1962 with John Kendrew. He is based at the Medical Research Council's Laboratory of Molecular Biology, Cambridge.