p53 (17 page)

‘In the spring of 1941, German officers came to our chalet and arrested my mother. This is one of my earliest memories,’ writes Capecchi. ‘My mother had taught me to speak both
Italian and German and I was quite aware of what was happening. I sensed that I would not see my mother again for many years, if ever.’ Aged three and a half he moved in with the family next
door and joined in with the simple life of the farm. ‘In the late fall, the grapes were harvested by hand and put into enormous wooden vats. The children, including me, stripped, jumped into
the vats and mashed the grapes with our feet. We became squealing masses of purple energy. I still remember the pungent odour and taste of the fresh grapes.’

Capecchi remembers also the day when American warplanes flew low over the fields, ‘senselessly’ machine-gunning the peasants. But he does not remember how or why the money for his
support ran out, only that at the age of four and a half he had to leave the farm. ‘I set off on my own,’ he writes. ‘I headed south, sometimes living in the streets, sometimes
joining gangs of other homeless children, sometimes living in orphanages, and most of the time being hungry. My recollections of those four years are vivid . . . Some of them are brutal beyond
description, others more palatable.’

Capecchi’s mother survived the German prison camp and set out to look for her son, finding him in October 1946 in a grim hospital in Reggio Emilia where he was being treated for
malnutrition and typhoid. The two travelled together to the US, to join Lucy’s younger brother Edward Ramberg in Pennsylvania, where he was living with his wife Sarah in a commune. Sarah
taught the young Mario to read and write, and the boy now went to school for the first time. His Uncle Edward was a physicist renowned for his part in developing the first electron microscopes, and
Capecchi himself took physics and maths when he went on to college. He found his studies intellectually satisfying, but rooted too much in the past. He was looking for the challenge of the new and
‘a science in which the individual investigator had a more intimate, hands-on involvement with the experiments’.

He found both during a three-month work-study programme at MIT. It was the late 1950s. ‘There I encountered molecular biology as the field was being born,’ he writes. ‘This was
a new breed of science and scientist. Everything was new. There were no limitations. Enthusiasm permeated this field. Devotees from physics, chemistry, genetics and biology joined its ranks. The
common premises were that the most complex biological phenomena could, with persistence, be understood in molecular terms and that biological phenomena observed in simple organisms, such as viruses
and bacteria, were mirrored in more complex ones.’

Capecchi was hooked. He applied to Harvard, ‘the perceived Mecca of molecular biology’, for his graduate studies and was taken on by James Watson – discoverer with Francis
Crick of the structure of DNA in 1953 – who had told him when he asked for advice that he would be ‘fucking crazy to go anywhere else’. Capecchi observed, ‘The simplicity of
the message was very persuasive.’ He was in his element at Harvard and flourished under the ‘merciless’ but fair and extremely supportive mentorship of Watson. ‘Doing
science in Jim’s laboratory was exhilarating,’ he writes. ‘As an individual, he personified molecular biology and, as his students, we were its eager practitioners. His bravado
encouraged self-confidence in those around him . . . He taught us not to bother with small questions, for such pursuits were likely to produce small answers . . . Once you made it through
Jim’s laboratory, the rest of the world seemed a piece of cake. It was excellent training.’

After receiving his PhD, Capecchi spent another six years at Harvard before moving for the big skies and rugged open spaces of Utah, to join a new young Department of Molecular Biology being
established at the university by scientists he admired and whose vision he shared. He has remained at Utah ever since, and it was here that the knock-out mouse was created in 1989. It was just such
a mouse that Scott Lowe and Tyler Jacks used to investigate p53’s role in apoptosis.

WHAT HAPPENS IF WE ‘KNOCK OUT’ p53?

‘The University of Wisconsin is an agricultural school and our animal model was the pig. So in addition to learning how lipoproteins
⁹
interact with their receptors and how failures in that lead to high cholesterol, I also got experience with animal models – and particularly big ones.’ Scott Lowe, a
fit-looking, ruggedly built man in his late forties, smiled as he described, in a deep, melodious Midwestern accent, his entry into molecular biology. Lowe didn’t much like science at high
school and imagined he would become a lawyer. But he had taken the opportunity, while studying biochemistry and genetics among his courses as an undergraduate, to spend time in a lab and discovered
he loved it: asking questions and dreaming up ways to test ideas were fun. Research, he realised, was what he wanted to do, and after graduating from Wisconsin he managed to get into MIT, the
hothouse of bright minds and exciting science, where Mario Capecchi had discovered his calling some 30 years earlier.

At MIT Lowe met Tyler Jacks, a young researcher who had picked up Capecchi and colleagues’ new technology with enthusiasm and was busy creating transgenic mice of all kinds to investigate
cancer-related genes. Jacks had made some knock-out mice in which various tumour suppressors had been deleted and he was asking the simple and obvious question: do the animals get cancer? He had a
mouse model with p53 knocked out, but he had been beaten to it in his experiments by another scientist who had been investigating the same question, so his p53 knock-out mice were sitting around
with not much to do. Jacks was happy to let Lowe suggest alternative experiments with them.

Lowe was already fascinated by apoptosis. He had done some work with cell cultures, watching it happen, to his great surprise, in response to oncogene activity, and he was not sure what role, if
any, p53 was playing. Perhaps the knock-out mice could answer this question. He had also seen Oren’s paper in
Nature
about his temperature-sensitive mutants and the experiments he
had done with leukaemia cells which killed themselves when the thermostat was set too low and his mutant p53 morphed into wild type. ‘Moshe Oren’s experiment was very exciting, but it
over-expressed the p53 gene, and one always has to worry that that might be artefactual – you don’t know if it does what it appears to do in real life,’ Lowe told me when I
visited him at Memorial Sloan Kettering Hospital, where he has a lab on the eleventh floor looking out over the dramatic roofscape of New York City. Over-expressing the gene is like using a
sledgehammer, I suggested. ‘Yeah, that’s right. Cells are sick, they die, right? So it wasn’t clear. Particularly since the view was still very strongly that p53 was a checkpoint
gene.’

To test whether p53 induces apoptosis in real life Lowe decided to concentrate on the thymus gland, a particularly sensitive organ in which programmed cell death was already known to play a
role. The thymus gland produces the thymocytes, important cells of the immune system that give rise to the T cells. Autoimmune diseases occur when T cells start to attack the body’s own cells
rather than fending off invading organisms. Scientists had discovered that to prevent this happening, thymocytes that might cause problems are weeded out naturally by apoptosis as part of the
thymus gland’s regular cycle of production and quality control. Barbara Osborne, an immunologist on sabbatical at MIT, had suggested this might be an ideal system for Lowe and Jacks to
study.

There are many stimuli known to drive thymocytes to commit suicide, explained Lowe. ‘So we decided to line up all of the treatments that were known to trigger this response and then
compare the normal mice with the p53-deleted ones, to see if any were defective when p53 was gone. We treated them a lot of ways and mostly the cells died normally, whether p53 was there or not.
But the one treatment that was different was radiation, which is known to damage DNA. You could take a machine that would produce gamma rays – the same kind of rays you’d use to treat
patients in radiotherapy – and if you irradiated the p53 knock-out cells, they didn’t die nearly as effectively as the others. So that was sort of genetic proof that p53 was critical
for an apoptotic programme – not as a general principle for all cells, but a very specialised subset: ones which had been irradiated.

‘And so it fit the model in a way. That DNA damage could activate p53 was known, but in all the other cell types tested so far it led to a checkpoint arrest. Here the cells
died
.’ This was decisive evidence that p53 was involved in apoptosis in real life, and it was an enormously important finding. Lowe and Jacks – whose p53 knock-out mice had
also helped Kastan confirm his hypothesis that p53 leaps into action when DNA is damaged – were riding high. ‘We knew instantaneously that this was a huge result, and that it was going
to sail into a very prestigious journal,’ said Lowe.

Then came a crushing blow. On the far side of the Atlantic, Andrew Wyllie had teamed up with Alan Clarke, another transgenic mouse man, at Edinburgh University; they had been working on exactly
the same experiment with thymocytes and come up with exactly the same results. In the race to publish first the two teams came neck and neck; both had submitted their papers to
Nature
and,
in a move that took Lowe by surprise, both papers appeared together in the same edition of the journal in April 1993. ‘There was a nice summary by David Lane that highlighted how important
this result was . . .’ concluded Lowe gamely, unable to hide a note of keen disappointment even a couple of decades later.

As Wyllie tells it, the discovery of p53’s role in apoptosis following radiation of the thymus was for him also one of the most thrilling moments in his scientific career. ‘And we so
nearly missed that one!’ he mused when I spoke to him at the meeting in Sheffield. Wyllie’s inclination was to bombard the p53 knock-out mouse with steroids as a killing stimulus for
the thymocytes, because this was closest to a natural scenario. But, significantly, steroids don’t cause DNA damage. ‘So we used the steroid and got nothing. There was no difference in
effect between having p53 and not having p53. The cells died on schedule the same as the controls.’ So Wyllie and Clarke decided, belatedly, to try radiation on their knock-out mice thymus
cells as well, since DNA damage was the hot topic at the time. ‘The effect was entirely different,’ he commented. ‘In the presence of radiation and p53 you got beautiful,
reproducible cell death, which was apoptosis. And if you did the same experiment with radiation and without p53 the cells did not die . . . That was a golden moment, absolutely! That paper was
written over a weekend,’ said Wyllie with a grin.

ENGINEERED MOUSE RAISES TRICKY QUESTIONS

Fast-forward some dozen years to the first decade of the 21st century and the lab of Gerard Evan, whom we first met at the start of this book, marvelling at the extreme rarity
of cancer in the multitudinous cell population. Evan is something of a maverick. From time to time he is prone to dropping bombshells that shake the foundations of prevailing opinion, and he did so
in 2005–6. Recognising the vital importance of understanding what each component of the cell does in the greater scheme of things, he is an enthusiastic mouse man. ‘The cells in a
tumour are all interacting with one another. It’s not as if they’re all doing their own thing and just decking it out,’ he explains. ‘They’re all talking to one
another, reacting and interacting with the normal cells in the body, and educating and instructing them. And that’s just by virtue of the fact that as a multi-cellular organism the way we
hold together is that our cells talk to each other all the time.’

Evan adapted the gene-targeting technology to make an even more sophisticated version of the p53 knock-out mouse. He replaced the animal’s natural p53 with a doctored version of the gene
that he could toggle at will between two states – from inactive to functional and back again – by giving and withdrawing a specific drug (a hormone) that controls the
‘silencer’ on the gene.

p53 as the ‘guardian of the genome’ that leaps into action in response to DNA damage was well established. But Evan’s experiments had led him to question whether its function
as a tumour suppressor was as straightforward as that paradigm suggested. He wanted to test it – and in so doing he proposed a heresy. For his experiment he needed a cancer known to be
induced by DNA damage, and he chose leukaemia because he knew from experience that if p53 is not functional, mice will develop this type of cancer very quickly after radiation. He knew too that,
conversely, mice will be protected from leukaemia even after radiation so long as p53 is present and functioning properly. That much was clear.

But Evan and his team wanted the answer to a very simple question: ‘Do you need p53 around at the moment the DNA is sustaining the damage that causes the cancer in order to protect against
cancer?’ Or to put it another way, at what stage during the gradual development of a tumour does p53’s activity become critical to protecting us from runaway disease? Teasing out the
answer, however, was not quite so simple. It involved looking at two different scenarios with their engineered mice – one in which p53 was active and functioning during radiation treatment,
and the other in which the gene was silenced during treatment.

The experiments for scenario number one brought no surprises at first; the scientists saw what they expected: the animals got very sick, there was ‘mega-death’ in their lymphoid
organs, their bone marrow, their gut – all sites of fast-dividing cells. ‘This was always thought to be the price you pay for p53 getting rid of the damaged cells, right?’
comments Evan. ‘So you get all of this: animals get sick from the mass dying of cells, but they recover.’ However, once p53 had done its job in response to the DNA damage caused by the
radiation, the researchers ‘switched the gene off’ again, and here they did get a surprise. ‘Blow me down, the incidence of cancer was like you never had p53 there at all! So none
of that pain had any gain in terms of tumour suppression – it was, like,
irrelevant
.’