To the editor

I have several questions about and comments on David Schubert's commentary “A different perspective on GM food” published in the October issue (Nat. Biotechnol. 20, 969; 2002). Schubert raises three scenarios in which agricultural biotechnology could lead to the “biosynthesis of molecules that are toxic, allergenic, or carcinogenic.” These include differential post-translational protein processing, unexpected changes in gene expression, and disruption of endogenous enzymatic pathways.

Schubert writes that the only “reasonable solution is to require that all GM plant products destined for human consumption be tested for long-term toxicity and carcinogenicity before being brought to market. These safety criteria must be met for many chemicals and all drugs.”

My first question is why these concerns don't apply equally to new crop varieties that are developed through intensive radiation or chemical mutagenesis? All of these unknowns are as relevant to mutagenesis-derived varieties, or to varieties that are created by wide-crosses, as they are to crops developed through recombinant genetic technologies. It is also worth noting that current US regulatory rules already account for instances where the genetic modification results in a significant change in the composition of the foodstuff.

The issue of allergenicity is tricky because there is currently a double standard. Although there appears to be a zero-tolerance policy for novel foods, allergenic foodstuffs that have been on the market for decades are permitted. Yes, they require clear labeling, but they are not banned or removed from the market. Peanuts, kiwi, corn, wheat, and many other foods are allergenic to some people, sometimes lethally so.

In fact, biotechnology is a powerful tool to reduce these natural risks as well as any potential allergy risks in novel genetically modified (GM) foods. This has now been clearly demonstrated by the recent removal of the major allergenic protein from soybeans. Our predictive capabilities and knowledge of what makes specific proteins allergenic is increasing monthly. Our regulatory authorities already take such risks into full account, as demonstrated by the Starlink episode. Human food approval of Starlink corn was rejected based on purely hypothetical risks. This example proves the already heavy precautionary stance of current biotechnology food regulations.

As to Schubert's suggestion for long-term safety testing of novel biotechnology foods: there are fundamental differences between foodstuffs and products like pharmaceuticals. First, it is simply not possible to conduct the kind of “high-dose” carcinogenicity testing (or whatever else Schubert means by “long-term toxicity” testing) on food products, which is conducted for individual chemicals or pharmaceuticals. To do so requires high-dose testing, and consuming high doses of single foodstuffs of-ten significantly affects health because of a fundamental lack of balance in the diet. Second, if a food is toxic, long-term testing isn't necessary. Short-term (that is, acute toxicity) testing will be fully sufficient.

Moreover, with the current debate about acrylamide formation in cooked foods and the presence of natural carcinogens in our foods, a strict implementation of Schubert's testing scenario would result in most of our totally natural foodstuffs being pulled from the market. What would be the regulatory enforcement standard for carcinogens or “long-term toxicity” in novel foods? Worse than a cup of coffee? Worse than lettuce, fried bacon, or mushrooms? All contain natural carcinogens and chemicals that are toxic at higher doses.

Finally, Schubert contends that the Shawadenko tryptophan toxicity episode was the result of genetic engineering of the bacteria. Having researched this topic considerably, I was under the strong impression that the toxic contaminant in the dietary supplements was found to result from the elimination of a key purification step from product processing, rather than from the bacteria having been genetically engineered.

See Divergent perspectives on GM food by Parrott et al. and the Reply to 'Divergent perspectives on GM food' by D. Schubert.