Translation: beating scientific swords into medical ploughshares

This essay was written by John Galloway and was first published in the 2010 Mill Hill Essays.

Science: use or ornament?

What is science for? Sir JJ Thomson is credited with the proof of the existence of the electron; he won the Nobel Prize for it in 1906. It is alleged that at an annual dinner of the Cavendish Laboratory, one of his colleagues proposed a toast. “To the electron – may it never be of any use to anybody!” Even if no more than apocryphal, those few words crystallise the answer to the question. They suggest that science ought to be above mere utility which is a bit ‘trade’. At the same time they strongly imply that there are those who, rather regrettably, think otherwise.

So, perhaps depending on your point of view, on the one hand science is there to find out interesting things about the world: its purpose is to be enlightening. Or, on the other, science ought to be useful: keeping the streets clean, making us live longer, paying the national bills. Five years ago the UK government’s Economic Impact Group asked the Research Councils, the main funders of research in the UK, to:

…provide views on how they could deliver and demonstrate a significant increase in the economic impact of their investments, and also to provide information on each council’s strategy for the allocation of funding.

The Warry Report was the Research Councils’ response. It didn’t hold back:

The economic benefits too will be diverse, extending beyond productivity gains to conceptions such as value created through better healthcare, better public services at national and local level, through law and policy making and cultural benefits.

That’s quite a claim. It is difficult to know what else there is that could benefit!

The two views about the purpose of science appear to be diametrically opposed, even mutually contradictory. One is a romantic adventure to discover the truth about the world and is pursued only for its own sake; the other, chasing the dragon of the mastery and exploitation of nature; ultimately attaining the utopian state of physical, mental and social well being. Scientists have spotted that it would be a good thing for them to argue that you cannot have the second without the first. Scientists of necessity have to be ‘free-range’, able to follow their own, however idiosyncratic, inclinations, if what they find is to be of any real use to the world outside science. Only in this way will they attract ever-greater bounty from government.

Consider this for instance:

Practical and economic benefits arise from scientific discoveries. Science has economic impact precisely because curiosity-driven research reveals patterns and features of the natural world that we did not know and did not expect.

It is taken from a recent website aimed at recruiting disaffected scientists to petition the government for a reversal, or at least a change, of policy:

A policy now being applied by the UK Research Councils… directs funds to projects whose outcomes are specified in advance…” the UK taxpayer should not support investigations with foregone conclusions.

The sentiments it expressed were echoed last year by a letter to the Financial Times signed by any number of UK Nobel Prize winners.

What is the source of this tension between scientists and their government paymasters? In relation to medical research the editor of The Lancet, Richard Horton, wrote of:

…the huge gap between theory and practice, between the burgeoning scientific basis of medicine and the relatively slow progress of clinical practice and preventive public health… the discontinuity is all too well recognised today with the debates about how to translate research into practice and how best to use the mass of information we already have.

In other words, governments that hold science’s purse strings have made it clearer that they wish to see something more than exponentially growing quantities of increasingly recondite and largely incomprehensible information, however interesting to scientists it might be. They want to see some practical benefits coming out of science, preferably benefits that attracted votes. Notice, by the way, that Horton was writing in 1993.

Where and when did the problem start? It is apt as well as convenient to look at the UK Medical Research Committee, forerunner of the Medical Research Council (MRC), set up in 1913. In his history of the MRC Sir Landsborough Thomson wrote:

Thus began a policy of the concerted funding of medical research from the public purse.…it had functions not previously discharged on the public behalf to more than a slight extent; and it had potentially a scope as wide as the farthest limits of medical science.

There was of course a difficulty. It wasn’t seen back in 1913 but it is seen clearly enough now. Whereas more scientific knowledge and better understanding are necessary for progress, it turns out they are far from sufficient. However wonderful the biology being created, it still has to be turned into practical, safe and reliable forms of treatment or diagnosis; worthwhile clinical outcomes in other words. And this is far easier to promise than to deliver. Indeed, the history of the funding of research from the public purse over pretty much the whole of the last century has been one of governments attempting, though largely failing, to direct or channel the money towards the ‘practical’ problems and priorities of government ministries, rather than merely leaving scientists to get on with it. The famous, or infamous, 1971 Rothschild Report proposed that the Department of Health should commission research from the MRC. Years later a member of the MRC Council at that time commented “The effect was nil.”

Spreading the word

Richard Horton used an interesting word, ‘translate’, to describe research being transformed into practice. It is a sense not given by the Oxford English Dictionary, although it has been in the Oxford Dictionary of English since 1998, for example:

the translation of research findings into clinical practice.

Its first use in this sense may have been by McKinney and Stavely in 1966:

Anti-inflammatory screening suffers from the fact that animal models of laboratory inflammation have not provided ideal counterparts of human arthridites and the translation of results from the laboratory to the clinic has not always been fruitful.

Then there seems to have been something of a gap of nearly 30 years until 1994, when Kelley and Randolph wrote:

The opportunities for translation of information gained from molecular biology to health care delivery have never been greater.

It is worth noticing that the House of Lords Select Committee on Science and Technology did not use the word in their 1988 report called ‘Priorities for Medical Research’.

As we move closer to the present day, however, this shift in sense for translation has gained a great deal of ground. It is now the second most common current use of the word. The press have got hold of it:

The report will also highlight the need for more translational research – studies that get from the laboratory to the patient.

This appeared in The Times on 15 March 2010. MRC itself called its 2003/04 annual review, ‘Translating Research’. The 2006 Department of Health strategy paper, ‘Research for Better Health’ set up Biomedical Research Centres specifically to ‘spearhead translation’. The MRC created a ‘Translational Research Overview Group’ covering its four research boards. Dozens (perhaps hundreds) of translational research institutes and centres have been set up world-wide. And, possibly, the final proof that the word and its meaning are now seriously entrenched was the 2009 launch by the publishers of Science magazine of a sister journal called Science Translational Medicine.

The MRC 2003/04 annual review, Translating Research
The MRC 2003/04 annual review, Translating Research

Present day pestilences

Malaria and influenza (flu) are two scourges of which the Horsemen of the Apocalypse would be proud. The World Health Organisation estimates that there are 3-500 million clinical cases of malaria a year, and growing, and one to three million deaths, including 2000 children every day. Even these figures may be gross underestimates and indeed have just been challenged in The Lancet. And then there is influenza.

Just geometry

As ‘seasonal flu’ the threat of relatively mild infection is, like the poor, always with us. It occurs each year here in the UK, as in many places. But there is also the possibility of far more infectious and lethal strains of the virus emerging. The outbreak of ‘Spanish’ flu immediately following WWI killed between 30 and 40 million people world-wide. Though decidedly milder in its effects, ‘Asian’ flu which appeared in 1957, was highly infectious and two million people may have died from it. In the last few years, the threats posed by both avian and swine flu have forced the country and the wider world on to public health ‘red-alerts’. Like terrorism, these threats may oblige governments to engage in expensive and socially dislocating plans of prevention and treatment irrespective of whether they actually materialise or not.

The flu virus was identified at NIMR in 1933 where its changeable tendency was also noticed. Underlying its capricious and potentially deadly behaviour is simply a viral genome with the potential for a lot of variation. The resulting variety in the structures of the proteins that the genome prescribes creates strains of virus that are difficult to vaccinate against or treat. Their potential mildness or lethality as well as their demographic targets (the old and weak, the young and strong, the middle-aged) are impossible to predict reliably.

A flu protein which drew the attention of the drug industry is Neuraminidase (NA). An enzyme, it breaks the virus out of the cells where it has been multiplying. While not in itself sufficient, therefore, NA is certainly necessary for the spread of infection. And whatever the viral strain, NA’s substrate is always the same, sialic acid on the cell surface. It was argued perhaps 40 years ago that a drug whose molecular structure closely mimicked sialic acid would not only be a good bet but also effective against all strains of the virus. Not only that, but any evolving resistance to such a drug would necessarily also render the virus ‘unfit’ in an evolutionary sense. In other words, a drug-resistant virus would, rather conveniently, also be relatively harmless.

But, “Seek simplicity…and mistrust it” is good advice. Two drugs were developed: Tamiflu (Oseltamivir) and Relenza (Zanamivir), both mimicking the structure of sialic acid. Awkwardly, though, for this optimistic outlook, some mutants resistant to Oseltamivir but still infective were soon reported. On the other hand they continued to be blocked by Zanamivir. In other words they were mutants that could distinguish between the two drugs. It took research by Steve Gamblin’s protein crystallography group at NIMR to explain this singular behaviour. The changes in structure shown, contrive to deny Oseltamivir, but not Zanamivir, the necessary close stereoscopic fit with the active site necessary to block its role in breaking down sialic acid.

Neuraminidase < br/>
In the active site of neuraminidase (blue), oseltamivir fits next to the glutamine amino acid in the structure. In the oseltamivir-resistant neuraminidase (red), the mutation of histidine to tyrosine pushes the glutamine towards the oseltamivir, so that the drug can no longer occupy the active site well. This does not occur with zanamivir because it is a different shape, such that the drug is well placed to interact with the glutamine when either histidine or tyrosine are present.

It is a finding that opens up a rational way to design some new drugs, or at least sensible variations on the existing ones, with potential benefits for both health and the economy. Over to Big Pharma! It also suggests a good theoretical basis for a UK government strategy for stockpiling antivirals against future epidemics:

It would be prudent for pandemic stockpiles of oseltamivir (tamiflu) to be augmented by additional antiviral drugs, including zanamavir (relenza).

The two anti-flu drugs (and the research to improve them) illustrate very nicely the goal that translation ought to aim for, what might be called a ‘technological fix’. In other words they are a concrete expression of the cause-effect relationship which links the problem to its solution. The aim may seem self-evident, but a lot of innovation in medicine falls far short of achieving it.

Telling the one from the other

Whereas flu poses an ever-present threat of a major public health problem in the UK, these days malaria certainly does not; though that does not mean it poses no problems at all. Ironically, one of them arises from our possession of a successful health service. Like some other diseases, malaria can be transmitted by blood transfusion, something first recorded in 1911. Hundreds of thousands of people give blood. Hundreds of thousands travel to places where malaria is endemic. Some tens of thousands do both. Nevertheless, although several hundred cases of malaria are ‘imported’ into Britain every year, only 5 cases of transmission by transfusion were recorded in the 20 years to 2004. Success followed a policy of simply refusing (or deferring) donations from travellers exposed to infection. Success had to be paid for of course. Its cost was counted in safe donations that were lost.

What was needed was a ‘marker’ to discriminate between those infected and those not. The successful candidate presented itself in a nice instance of ‘serendipity’. The vehicle for this happy chance was research into the possibility of creating a malaria vaccine started by Tony Holder some years before he came to NIMR’s Division of Parasitology. Vaccines provide the paradigm for ‘technological fixes’ in their embodiment of the cause-effect relationship. It is a relationship relying on the remarkable discriminatory power of antigens in selecting for antibodies. It follows that considerable research effort goes into finding antigens that provoke a measurable harvest of antibodies. The exclusivity of the antibody-antigen relationship also provides a basis for the extremely reliable identification of proteins as markers for any number of medical conditions through technology like ELISA (Enzyme Linked Immunosorbent Assay).

Knowing what the antigen actually does is not necessary for its value as a marker any more than its utility as the basis of a vaccine. It needs only a unique relationship with its source; preferably that there is a lot of it and that it is accessible. Around 1980 Tony Holder found a protein that satisfied all three needs: Merozoite Surface Protein 1 (MSP1). MSP1 is a protein common on the surface of the merozoite, one of the forms the malarial parasite assumes in blood. It was a reasonable bet as the basis of a vaccine and trials are underway. Medical Research Council Technology (MRCT), the Council’s technology transfer arm, negotiated MSP1’s use as a marker to test blood donors for the results of their possible exposure to malaria. In harness with a close molecular relation, MSP2, it now underpins an ELISA-based kit marketed by Lab 21, a company who supply it to blood transfusion services. Last year NHS Blood and Transplant used it to test 66,500 potential donors for malaria. 1425 tested positive for antibody with only 346 ultimately confirmed, 0.61% of those at risk. For NHS Blood and Transplant, at least, something has been found in translation.

Methods and motives

Science is characterised not by the ‘scientific method’ which seems to be a myth, but as the few examples here illustrate, by its methods. As Barnes and Dupré’s book Genomes and What to Make of Them, published earlier this year reminds us, any science is not just a body of knowledge, made up of models and theories, concepts and ideas, maps, graphs and diagrams, but also a methodology embodied in technology of some sort. Knowing is not separable from doing. The thing being studied can only be conceptualised in terms of the technology. Protein crystallography is a vivid illustration of this dependency. It seems that science grows first by devising new methods and then by expanding their scope into new territory. That’s one important lesson about translation.

In principle, protein crystallography can answer any question that involves the three-dimensional fit between a protein and another molecule. But its medical and social usefulness lies in the particular causes to which it is harnessed, the questions it is chosen to answer, and the way society chooses to use the answers. Bob Edwards has just received the Nobel Prize for research begun 50 years ago at NIMR into the treatment of human infertility. But the translation of that research into producing babies has turned on two very different sides to the human character: the willingness of some women to let other women have their eggs for nothing; and the readiness of some doctors to make huge profits out of those free eggs. The success of translation does not lie with scientists, however good their science, but with human motives, individual and corporate, selfish and unselfish, which underpin health care markets and drive market traders. That is the other important lesson. It is one that scientists and governments alike should learn.


I could not have attempted this essay without a lot of help, particularly from: Margot Charlton, Steve Gamblin, Tony Holder, Frank Norman, Lois Reynolds, Tilli Tansey. I am grateful to all of them.

Leave a comment


email* (not published)