The award was set up by Cancer Research UK to recognise a lifetime’s work relevant to cancer. It was a very pleasant surprise and a considerable privilege, I felt, to receive an award from an organisation which I hold in very, very high regard. Cancer Research UK, or its two precursor charities which were Cancer Research Campaign and Imperial Cancer Research Fund, between them have supported my research for a third of a century and it’s an organisation I’ve got to know very well for that reason. So it’s an award I therefore appreciate out of proportion to any others. In other words, it’s one from an organisation I have an utmost respect for. Also it’s nice to feel that that particular organisation which has done so much for cancer patients has taken that view so it’s a very nice award to have received and I’m delighted that they’ve chosen me.
Can you summarise the work you have been covering?
It’s difficult to do justice to try and summarise what was intended to be an award for a lifetime achievement in 30 minutes, or I think 38 minutes, I was allowed 40 minutes and took 38. So inevitably it’s very difficult to ensure that you can do justice to any of the things. So I’ve attempted to do a sketch of some of the highlights and those were concerned with several areas, one of which was control of cell proliferation – how cells decide when to divide and what the mechanisms of how they divide faithfully are. So part of my talk was looking at how cell proliferation is controlled, a crucial underlying aspect of cancer. But some of the proteins we found during those studies turned out to be exceptionally promising markers for cancer diagnosis and cancer screening. They are markers which are retained in cancer cells when they’re shed from the surface of the tumour but they’re not retained in normal tissues when they shed cells into the lumen. For example, the cervix which sheds cells which are sampled in the smear test, normal cervix does not contain these markers whereas premalignant or malignant cervix does so it becomes possible to stain cancer cells a different colour from the normal cells because they contain these markers related to DNA synthesis and cell proliferation in general. So that was one of the themes that I covered, there were others but it starts getting a bit complicated if we try to review them all.
How important are basic studies to understanding cancer and generating agents?
That is exemplified by the case I’ve just offered but there are others. It’s very difficult to predict which types of research are going to produce patient benefit, very difficult indeed. The classic example of this is, of course, from an old friend of mine who is sadly no longer with us, César Milstein, who invented monoclonal antibodies while trying to do something completely different. What he was attempting to do at the time he had no idea would be of benefit to cancer patients by producing a wonderful range of antibodies such as Herceptin, for example, and many others used in cancer therapy now. So it’s very difficult to predict which experimental approaches will be beneficial.
An argument I use to explain this principle to funding bodies is a painting of a watermill by Constable. I think it illustrates beautifully what you need to do good translational research for patient benefit. You need the translational machinery to turn it into something that benefits the patient but, in addition, you need a constant stream of basic science new ideas coming in, the water that turns the wheel. If you’ve got both of those in the same place you can do far more for patient benefit than if you have them separate or if you only concentrate on the application and don’t generate the constant flow of new ideas to drive the machinery.
There is a case to be made for abstract biological research?
Extrapolate it to absurdity - that becomes quite dangerous because obviously we can’t fund all possible branches of science, nor should we try. But there is definitely a case for having blue sky science going on in the same place and in close proximity to and in good dialogue with people who are trying to directly benefit the patient. That synergy can be immensely powerful.
What are the problems with inadequate models?
This supplies more to the basic side of what I was talking about where some of the things, some of the topics, I covered were how things moved from the cytoplasm into the cell nucleus and how things move from the nucleus back to the cytoplasm. In both of these cases the prevailing models have proved to be inadequate; one refers to work we did way back in the early 1980s where we found that the widely published view of how proteins accumulate in the cell nucleus was completely wrong and it was great fun overturning it. But wrong models can hold back progress for a very long time. For mRNA export from the nucleus to the cytoplasm, for getting information out from the nucleus to transcribe the information from DNA, that hasn’t been wrong, it’s just been inadequate. We think of it as a bulk flow of RNA being made from DNA and going out to the cytoplasm to be turned into protein but in fact there is a whole series of selective and crucial fast-track priority routes to get information out of the cell nucleus and this has been completely underestimated. I think it’s an interesting field to see if it can be exploited in any way. An example of this is that, and this is work we’ve done collaboratively with the group of Ashok Venkitaraman in Cambridge, and the example is any proteins involved in the most accurate form of DNA repair in response to DNA damage, those proteins are made from messenger RNAs that have a priority pathway out of the nucleus, a dedicated factor which a signalling mechanism says, “We’re damaged. Send out the message to make sure we’ve got plenty of this protein to repair the damage.” That’s a level of regulation which we had no idea existed until a few years ago. So the view that everything goes out by the same route is wrong, there are differential, preferential pathways out of the nucleus.
Do these models need to be improved as you go along of can something be done at the beginning?
I think they get improved as they go along. In my talk yesterday I confessed, on reviewing what I’ve done over my research career, I confessed to realising, rather to my horror, that I appear to have been the Victor Meldrew of the cell nucleus in saying, “I do not believe it,” and finding that some of the models were wrong. But they are being corrected as we go. But it’s crucial that when a model is inadequate it’s corrected quickly or it holds back the progress of science for too long.
