‘On paper it looks very simple,’ said Oren, shaking his head slowly at the memory of the months of trial and error and frustration working mostly in the realms of unseeable biology. ‘In practice, we spent a year trying things that didn’t work. Then we tried this method that worked, but we went through many false negative clones before we found one that was real.’ The trickiest part was getting the plasmid vectors to take up their new cargo, the scraps of p53 cDNA. Oren and his technical assistants screened the plasmids 10 at a time rather than individually to see if they had succeeded – a task of enormous tedium that swallowed nearly 18 months of their lives.
‘The procedure ended up being rather inefficient and the percentage of plasmid clones that indeed contained p53 cDNA was extremely low. We had to screen many hundreds of clones – actually I think about 1,800 – before we hit the first positive one.’ When that finally happened, it was one of the most exciting moments in Oren’s scientific career, he told me with a smile. He and his colleagues published their results in Proceedings of the National Academy of Sciences in 1983.
Meanwhile, back in Princeton, Arnie Levine had started collaborating with the pharmaceutical company Genentech and one of their expert cloners, Diane Pennica, to continue the search for a p53 clone once Oren had returned to Israel. Using different strategies and different cell types from Oren, they too were successful and they published their findings in the journal Virology the following year.
Others were hot on the same trail and by 1984 there were a number of p53 clones, of different cellular origin, in the scientific press and in limited circulation around the labs. Among them was one obtained at the ICRF in London by John Jenkins, one of very few people with cloning skills in Britain, where progress in manipulating genetic material had been inhibited by the controversy surrounding recombinant DNA in Europe and the US. Jenkins, who came late to science after finding himself a misfit at school, leaving early and taking up work as a general labourer and landscape gardener before returning to education, remembers fierce competition to obtain a clone. ‘Obviously people are driven by their career ambitions – but at some level there’s the pure joy of beating the opposition. I mean, it’s a competitive business! Always was; always will be.’ Like Oren, Jenkins remembers how tough a task cloning p53 was. This was the cutting edge of science: ‘Everything at the cutting edge is a challenge at the time,’ he commented.
Just how much of a challenge cloning was in the early 1980s is hard to grasp for succeeding generations of biologists. ‘My best story about that was years later,’ said Jenkins. ‘One of the people in my lab was helping me clear out my office. We came across a bound copy of my PhD thesis and she said, “Oh! Can I take this away and read it?” I said, “Sure, no problem. Bit of old history . . .” Later she brought it back and said, “I don’t believe it! You didn’t do anything, and you got a PhD for that?” I said, “Listen! You try doing that work when you have to make all your own enzymes . . . There weren’t any kits in those days, you know . . . Everything you take for granted now was just not there, so it took a long, long time – and it was tricky.”’
Oren, Levine and Jenkins had, in fact, been beaten to the finishing tape in the highly competitive race to clone p53 by a Russian scientist, Peter Chumakov, working at the Institute of Molecular Biology in Moscow. But Chumakov had published his results – in December 1982 – in Proceedings of the Academy of Sciences of the USSR, a Russian-language journal with limited readership in the West, and for some time his triumph went unnoticed beyond the Soviet Union. Back in 1979 Chumakov had just completed his doctoral thesis on SV40 and had been looking for a project that would reveal how large T antigen caused cancer when he heard of the discovery of p53. Excited by the idea that p53 might be the key to the transformation process, he decided to clone the gene to provide material for further studies.
But Chumakov faced even greater challenges than his Western competitors. His lab in Moscow had shortages of some of the most basic equipment, such as test tubes and pipettes, which he cleaned and re-used over and over again. He also had to send off to a scientist in Texas, Elizabeth Gurney, for a sample of the special antibody to p53 she had made and which he realised was vital to his cloning endeavour. He made his request without much hope of receiving the sophisticated tool. ‘I thought that even if Elizabeth decided to send the sample, a package containing a suspicious test tube would be stopped either on the USSR border, or during postal censorship that occurred regularly with mail coming from the West,’ he told me. But he was lucky. Visiting a colleague in a neighbouring lab one day, he noticed a small, foreign-looking package lying on a desk and he turned it over in casual curiosity. To his surprise and delight he found it was addressed to himself and contained a small phial of the antibody together with a letter from Gurney wishing him success.
His unexpected good fortune boosted Chumakov’s morale enormously and he was determined to complete his task – especially as he now realised, from his new contacts with the outside world, that he was not the only one trying to clone this important gene. That he was the first to succeed was finally recognised when his paper came out in English and he began to be invited to international p53 meetings. ‘There was a wonderful atmosphere of enthusiasm and hope in the lab, and we really felt lucky about being attached to those exciting discoveries in life sciences,’ he commented.
For all the research teams involved, the next step after producing a clone of p53 was to determine the sequence of the gene – the words spelled out by its base pairs, A and T, G and C, that dictate the amino acid composition of its protein. ‘Sequencing was not very difficult, even in those days,’ commented Oren. ‘It was much easier to do than cloning. But it was all manual: you had to prepare your DNA and all kinds of reagents and run gels and do everything by hand. And you had to read it and interpret it yourself. There were many errors; the method was not precise enough, but it was not a technical challenge. And by the time I got to it there was already a technician who was doing it routinely.’
With the information provided by sequencing and a potentially endless supply of clones – some of the whole gene, others of important segments of the gene – researchers were now ready to start investigating how p53 functions within cells.
CHAPTER SIX
A Case of Mistaken Identity
In which we: a) discover that we humans also have would-be oncogenes in all our cells; b) learn that what turns oncogenes nasty is mutation; and c) hear that an oncogene is what almost everyone – erroneously – concluded p53 to be.
***
Scientists, like anyone confronting a new problem, will start with what they already know. Neither in theoretical speculation nor at the bench do they often sail beyond sight of the shore.
Horace Freeland Judson
As they set about their investigations into how p53 functions, thoughts of oncogenes were uppermost in many researchers’ minds, since the first such gene in human DNA had been discovered just months earlier, in 1982. Once again Bob Weinberg, one of the pioneers of oncogene research, was there at the cutting edge, and here we need to retrace our steps a little to see how the oncogene story developed.
Weinberg was born in Pittsburgh, Pennsylvania, and raised in the United States by a mother, father and grandparents who had arrived as refugees from Europe in the late 1930s. His father, a dentist in Germany, had seen the writing on the wall as the Nazi threat grew and had been smuggling money out of the country to a brother in the Netherlands. When he finally fled to the US in 1938, he had enough money saved to retrain and re-certify as a dentist in his new home. Though the younger Weinberg had no direct experience of Fascism, the suffering of his family in Europe, where many of them did not escape or survive the Holocaust, percolated into his consciousness and he grew up with a keen sense of the precariousness of life. ‘One thing my father said to me was, “It’s okay to be successful and productive, but don’t be too visible lest the guys in the brown shirts come in one night and take you away.”’
Although this man with the bushy moustache, self-deprecating humour and Woody Allen-ish face insists he’s happiest in mud-caked overalls, pottering in his garden or splitting logs, he is undoubtedly the most visible of a bunch of scientists in three different labs who all made the momentous discovery of the first human oncogene at the same time.5 Undeterred by the scandal caused by the fraudulent Canadian research described in Chapter 2, he and his team continued to investigate the phenomenon they had discovered and reported in 1979 – that rodent cells contain would-be oncogenes that can be activated by chemical carcinogens as well as by infection with viruses. By 1981 his lab, with that of Geoffrey Cooper at Harvard Medical School in hot competition, had shown that oncogenes are to be found in human tumour cells too.
Both labs had used the same ingenious and relatively new technique called ‘transfection’ in their research. This involves taking pure DNA from tumour cells, chopping it up and putting it in solution with normal cells that are the researcher’s target. By tinkering with the solution, the scientist can induce the cells to take up some of the pure DNA floating around and to integrate it into their own genomes. He or she then cultures the cells in Petri dishes and looks for evidence that they have turned cancerous. The cells in our bodies talk to each other constantly: normal cells obey instructions that ensure orderly growth and usually stop them dividing once they have formed a single layer in the Petri dish and filled the available surface space. Cancerous cells, however, are unruly: they talk anarchy, pile on top of one another and form haphazard clumps, known as foci, in the dishes. Thus Weinberg’s and Cooper’s teams looked for foci – and found them.
Homing in on the specific gene or genes responsible was an infinitely painstaking process of elimination. It meant repeating the experiment over and over again with different chunks of naked DNA from the tumour cells. But the three teams racing against each other got there within a year, publishing their results in the summer of 1982. They discovered that the oncogene in the human tumour DNA was one named ‘Ras’. Mirroring the story of the oncogene Src discovered by Varmus and Bishop, Ras turned out to be homologous (that is, very similar, and suggestive of a common evolutionary ancestor) to an oncogene found in a virus – in this case the rat sarcoma virus that had been discovered in the 1960s.
The following year, 1983, Mike Waterfield at the ICRF in London found that a virus that causes sarcomas in monkeys had, some time way back in evolutionary history, hijacked a gene from us that is involved in growth and repair, and is especially important in healing wounds. This was further proof that we humans have in us genes that are part and parcel of our normal DNA – and have a regular job to do in our cells – that can, under certain circumstances, cause us harm.
But what are the circumstances? At the time the first human oncogene was found, no one yet knew that carcinogens – whether they be chemicals, infections or radiation – typically work by scrambling our DNA. Bob Weinberg, however, had a hunch that mutation would be the key to activating would-be oncogenes, and once again he faced down the doubters who pointed out that some chemical carcinogens don’t actually cause mutations. ‘I was not troubled,’ he commented wryly in an essay on the discovery of human oncogenes. ‘I thought that a good simple idea should not be undermined by complicated facts.’
Weinberg’s hunch proved correct, and he and his co-discoverers soon found that the Ras gene – which has a central role in orchestrating the growth, division and differentiation of our cells in the normal course of events – can be turned nasty by an alteration in a single nucleotide, representing just one letter in its protein recipe. In a sensitive position in the gene, such a mutation can result in Ras being permanently ‘switched on’, heedless of any signals it may be receiving, thereby driving the cell to grow and divide unchecked. It was an important revelation, another vital insight into the mechanics of cancer at the most basic level. Nature hailed the discovery of the human Ras gene and its activating mechanism by naming 1982 ‘The Year of the Oncogene’.
No one could know then just how important a find Ras would turn out to be. We now know that this gene is mutated in around a quarter of all human tumours, including roughly a half of all colon cancers and 90 per cent of pancreatic cancers. But as scientists continued to investigate oncogenes – of both viral and cellular origin – they became aware of a vital caveat: a single oncogene acting alone cannot create a tumour; in order to mess up a cell’s machinery enough to cause cancer, oncogenes need to co-operate with one another. Researchers had no idea initially why this should be, only that it was so. When, for example, Ras is put into cells together with Myc – another powerful oncogene found in the DNA of a virus as well as in animals, including us – the effect is clear cut and often dramatic.
p53 LOOKS LIKE AN ONCOGENE
In 1984, when researchers began working with the new p53 clones to find out how the gene functions, they quickly concluded it too was an oncogene. This was one of the first ideas they tested and they did so by taking the classic experiment – pairing the powerful Myc with another oncogene – but replacing the Myc with p53 to see if it had the same effect. ‘We put p53 with Ras; some others put it with other oncogenes; and we all got positive results,’ said Moshe Oren. ‘From the beginning the results were similar to Myc, though less dramatic and less efficient. So the feeling was: p53 is an oncogene. Not as good as Myc, but okay, Myc is the king; this is just a regular knight!’
An essential condition for cancer to get going is that cells become ‘immortalised’ – that is, able to override the limits set by nature to the number of times they can divide. An individual oncogene acting alone can accomplish this vital step, which leaves the cell on the brink of cancer, vulnerable to a second hit by another oncogene that will take it all the way. Jenkins’ group at the Marie Curie Institute, tucked away in the southern English countryside in Oxted, Surrey, found that their p53 clone was able to immortalise cells.
Another indication that what they were dealing with was an oncogene came from the fact that it produced a dramatic abundance of protein – what the scientists call ‘over-expression’ – in the cells it transformed into cancer. ‘At that time it was quite standard to conclude that proteins that are over-expressed in cancer cells must be produced by oncogenes,’ says Oren. ‘That’s how viruses transform; that was in the textbook; that’s what everybody thought. Varmus and Bishop got a Nobel Prize around that time for their work on Src, and the general dogma was: oncogenes are things that are over-expressed in cancer cells and that’s what makes them cancer cells. So here we have a perfect candidate: here’s a protein that’s over-expressed in cancer cells, right? p53 looks just like any plain oncogene.’
However, while everybody else seemed to be getting nice clear-cut results with their clones, Arnie Levine back in Princeton was drawing a blank, time after time, and growing mighty frustrated. ‘We saw Moshe’s group publish with Weinberg on the fact that it transformed cells; we saw Jenkins publish that it transformed cells, and Crawford also; and we couldn’t reproduce those results with our clone. We were devastated!’ he told me. ‘I had two people in the lab, a graduate student named Phil Hinds and a postdoc named Cathy Finlay, who were working on this, and to watch other people publishing and not be able to reproduce their results . . .’
Like everyone else, Levine’s team were conducting experiments designed to confirm their hunch – based on all the bits of evidence thus far – that p53 was an oncogene, and they naturally interpreted their singular lack of results as failure, pure and simple: they had a dud clone. Something gnawed at the back of their minds, however. As the various cloners published their gene sequences, Levine’s team noticed that their own clone had a difference in one codon – representing just one building block in its protein – but they had dismissed it as insignificant. Sequencing at that time was still a ‘hands-on’ process, where people not machines produced and interpreted results, and you had to allow a margin for human error.
‘At that time a few sequencing errors w
ere not something to be proud of, but it wasn’t a major crime,’ Oren concurred. ‘You see a difference here, a difference there, it doesn’t really matter – because by and large we can see what the protein looks like, and biologically we see that this one and that one do the same thing, so we’re fine. So we didn’t spend much time thinking about the significance of a difference in one amino acid. I tended to think their amino acid was wrong and mine was right, but we’re dealing with the same thing, so who really cares?’
Levine and his team cared very much, as they agonised over their failure to get results. They asked Oren, as a one-time member of their lab, if they could have a copy of his clone to try their luck. Oren happily obliged, and sure enough the Princeton team reproduced his results. Now they all knew that single codon difference was significant. It began to sow niggling doubts in their minds. At one of their periodic meetings, Oren was in Levine’s office in Princeton when Phil Hinds produced a list he had compiled of the sequences of all the clones the various labs had published in the literature. The only clone that had no biological effects, recalled Oren, was the one that Arnie Levine’s lab had produced. ‘This was very revealing because at the end of this session it was clear that all the people who were getting biological effects were using clones of mutant p53 – it was kind of obvious from the sequences. Arnie’s was the only one that was not mutant. He had a clone of the normal (or ‘wild-type’) gene.
p53 Page 6