ngin - Norfolk Genetic Information Network

5 January 2003


1.Schubert replies to Avery and Parrott et al on GM food safety


1.Schubert replies to Avery and Parrott et al

The October issue of Nature Biotechnology featured a piece by David Schubert, professor of neurobiology at the Salk Institute for Biological Studies in San Diego, in which he argued that "GM food is not a safe option, given our current lack of understanding of the consequences of recombinant technology [genetic engineering]", noting in particular that there is no way of predicting where this technology may give rise to toxic, allergenic, or carcinogenic effects.

Predictably Schubert's piece has been followed by letters of criticism from the usual sources - one from Alex Avery of the Hudson Institute and the other with Prakash-supporter Wayne Parrott as the main author and with multiple signatories, including 3 scientists from the John Innes Centre in the UK. Below Schubert responds to these letters.

For the original piece, A different perspective on GM food, DAVID SCHUBERT / Nature Biotechnology v. 20 p.969 Oct02

The Avery and Parrott letters, published under the title "Divergent perspectives on GM foods" can be seen by Nature subscribers here:
Non-subscribers can find them here:
December 2002
Nature Biotechnology Volume 20 Number 12 p 1197
David Schubert replies

It is indeed a fortunate facet of science that the quantitative aspect of authorship does not dictate the validity of the conclusions! The impetus for my commentary was my repeated laboratory observation that the slightest genetic modification of a cell leads to completely unpredicted phenotypes. A careful reading of the plant literature supported my conclusions from animal cells. The commentary intended to raise this concern in the context of generating further impetus for labeling and stringent testing of GM food. I had no a priori commitment to a particular technology, no pre-existing political ideology, and I did not deliberately ignore any pertinent published material. I do believe, however, that both replies suffer from these problems. They make claims about technology and safety-testing requirements that do not exist, use my statements out of context, and are rather adroit at the use of tenses to skirt questions regarding testing.

The issue of traditional breeding versus genetic modification has been extensively discussed. I have, however, always wondered how it can be claimed that they are the same when genes from completely different species (sometimes kingdoms!) are expressed in genetically modified (GM) food products. Although the biological world works by variation and selection, this is generally accomplished in the context of a normal complement of endogenous genes that, though perhaps different, are allelic. This is quite distinct from GM plants where many copies of a gene are introduced and integrate randomly1. It is my experience and that of others that the response to (trans) genes is completely unpredictable, not specific and predictable as argued. Genetic modification is also distinct from breeding two strains that have been safely consumed for extended periods of time.

Parrot and colleagues cite multiple gene deletion in varieties of maize to illustrate their argument that "unintentional consequences" are much more likely to occur in breeding than in biotechnology. However, maize is unique because its genome comprises 80% retrotransposons. It has evolved to deal with redundancy; most other species have not. The fact that plants are capable of producing toxins is a very good argument for testing.

I stated that the illness caused by Shawa Denko KK (Tokyo, Japan) GM tryptophan "was highly correlated with contaminants," not that it definitively caused the disease. I used this as an argument for labeling GM material. There is certainly no good evidence in reviewed journals that it was the purification procedure that caused the problems or that only people who used this brand took larger amounts and therefore became ill. The company has destroyed the bacteria and has paid large out of court settlements.

The most important issue to me is rigorous safety testing. The statements made in the replies that address this issue are completely misleading. Although it is true that in Europe laws are being formulated that require stringent testing of GM foods, in the United States, the FDA has no mandatory safety approval regulation for GM foods and no specific testing requirements2. There are no all-inclusive mandatory food-safety testing requirements in the United States.

The cited testing protocols are only suggestions for producers. There is, however, an effort by a consumer advocacy group, the Center for Science in the Public Interest3, to require GM food products to obtain FDA safety approval. With respect to testing technology, Parrot and colleagues claim that "the protein produced in the new host is subjected to extensive biochemical characterization to confirm that the protein produced is the one and only one intended." However, there is no technique that can assay all cellular proteins. The best to date is 2,528 out of the rice genome of 50,000 genes (a mere 5%)4!

I am very pleased that both letters support rigorous testing of GM food and hopefully all involved will back efforts to hasten mandatory rules through the FDA.

See Divergent perspectives on GM food by Parrott et al. and the Divergent perspectives on GM food - Letter 2 by A. Avery.

1.Svitashev, K. & Somers, D.A. Genome 44, 691-697 (2002). | Article |
4.Koller, A. et al. Proc. Natl. Acad. Sci. USA 99, 11969-11974
(2002). | Article | PubMed |



December 30, 2002
Center for Science in the Public Interest

On December 17, CSPI hosted a panel discussion on pharming at the National Press Club, bringing together scientists, industry
representatives, and policy advocates to discuss the benefits, risks, and regulatory issues surrounding genetically engineered crops that
produce pharmaceutical or industrial chemicals.

A transcript of The Future of Pharming: Can It Be Done Safely can be viewed here:
[includes the interesting suggestion from AgBioWorld co-founder Greg Conko that tolerance levels be introduced for pharma contamination of food]

ngin bulletin archive