Although I have never worked myself in plant biotechnology, I would like to take the opportunity to respond to your call for contributions to the GM Science Review. I have followed this topic with some interest since I chaired the Royal Society's first report on this topic in 1998.

In 1999, I wrote a discussion paper for the Academy of Medical Sciences called "Diet and Diseases Facts and Fantasies" comparing the situation with regard to BSE with that concerning GM foods. Although this field has moved on in the last 2½ years, I think this paper still essentially contains my views on the topic. It can be accessed from (<www.acmedsci.ac.uk/f_pubs.htm>.)

Just in the last week or two there has been renewed interest in the BMA's "provisional report on the impact of genetic modification on agriculture, food and health" issued in 1999; and the reported decision of Sir David Carter, the Chairman of their Board of Science, to revisit the topic this year. One reason that this report is again giving rise to interest is that there are claims from Zambia that this report was one reason why the Zambians refused to import GM maize to feed their population. If this is indeed true, it should give the BMA and other influential organisations pause before they issue firm recommendations on subjects which they have no particular expertise and which indeed fall out with their remit.

Bearing this in mind myself, I will restrict myself to making a few further comments on the question of GM foods and human health.

The idea that the process of genetic modification of plants rather than any particular modification carries a risk to human health has no serious scientific basis that I am aware of. The only mechanism suggested was that the cauliflower mosaic virus promoter which is frequently used might recombine with human viruses such as HIV and Hepatitis B to form super viruses that could kill us all (see Ho et al 1999 Microbial Ecology in Health and Disease 11 194-197). This gave rise to a certain amount of correspondence. I find it difficult to take this suggestion seriously. In order to get recombination of a virus with the promoter with which it is supposed to recombine these would have to be present in the same cell and in the same cellular compartment and exactly how this is envisaged was not discussed. While some hepatitis viruses can infect enterocytes DNA from food does not. Furthermore, it has been pointed out that the amount of cauliflower mosaic virus which we eat is barely affected by GM food. This virus is a common infective agent for Brassica species and in our lifetimes we eat large quantities of it and always have done so.

The question of leaving antibiotic resistance markers in products was discussed in the original Royal Society report and it was felt that it was better that they should be removed. I concur with this but it should be pointed out that as a factor in generating antibiotic resistance in bacteria that infect humans this is a trivial cause. It would be much more profitable to concentrate on ways of influencing prescribing habits and the prevention of world-wide antibiotic abuse. The antibiotics used in biotechnology are hardly used in humans anymore and I am not sure that any transfer from transgenic plants to bacteria has ever been demonstrated.

The other topics of concern are all related to individual transgenic products and apply equally well to all novel foods whether generated by conventional plant breeding or genetic modification. To this extent I think that the section on "Topics: GM food safety" on the Website of the GM Science Review is perhaps marginally unfortunate since it applies the safety assessment entirely to GM food, whereas surely exactly the same process is needed when a novel food is generated by conventional plant breeding techniques.

Food allergy is a significant problem in human health but this is not obviously influenced by genetic modification or indeed to any large extent by plant breeding. The problem that people have in avoiding food allergens comes from much further down the food chain. Many processed foods contain ingredients that consumers do not necessarily expect in them. It is now well known that small amounts of nut protein occur very widely in foods and are a problem to people who are allergic to nuts. There used to be a similar problem with very small amounts of penicillin occurring in products from penicillin treated farm animals who were highly allergic to penicillin. This was, at one time, even considered a problem with lactose used as a propellant for administering drugs by inhalation if the cows from whose milk the lactose was made had been treated with penicillin.

The potential health benefits of some genetically modified products are also overlooked. There is published evidence that one mycotoxin (fumonisin) occurs in smaller quantities in genetically modified maize than its non transgenic counterpart (Munksvold et al 1999 Plant Disease 83 130-138) and perhaps too little is made of this type of evidence.

Finally, in relation to the furore surrounding the claims of Dr Pusztai it needs to be realised that when comparisons are made between a conventional food and its genetically modified counterpart these comparisons do have to be made with foods that are otherwise congenic. As Dr Gatehouse pointed out in the case of the transgenic potatoes (which he himself had made) this was not the case and there were likely to be differences between the two types of potato being compared, which had nothing to do with the introduced gene. Dr Gatehouse's letter to the Lancet making this point was never published (although it was put onto his Website (<http://silver-server.dur.ac.uk/GM_Plants_Pages/Lancet.html>) and his objections were largely ignored throughout the media.

I greatly welcome your GM Science Review not least because this form of distorted presentation, where only arguments that the campaigning media and anti-GM NGOs regard as grist to their mill are presented, will be countered.

Professor Sir Peter J. Lachmann FRS FMedSci
Centre for Veterinary Science
Madingley Road
Cambridge
CB3 0ES