(This information was sent to the delegates of Codex Alimentarius in March 1997, see note in the end of this document)

The Difference Between Traditional Breeding Methods and Genetic Engineering

Consequences for Safety and Labelling Regulation Policy

Michael Antoniou, Senior Lecturer in Molecular Pathology, London, UK.
Joe Cummins, Professor Emeritus of Genetics, University of Western Ontario Canada
John Fagan, Professor of Molecuar Biology, Maharishi University of Management USA
Mae-Wan Ho, Professor of Biology, Open University, Milton Keynes, UK
Tore Midtvedt, Professor of Molecular and Cell Biology, the Karlolinska Institute of the University of Stockholm, Sweden

Summary

Genetic engineering, as done so far, means the artificial insertion of one or a few genes into the DNA of the recipient organism. This is fundamentally different from what happens in conventional breeding. 1. The insertion may occur in the middle of gene and thereby disrupt the genetic code of the recipient organism. 2. The inserted gene is also very likely to disturb the action of neighbouring genes. 3. The inserted gene will produce a new protein that may often be alien to the recipient organism.

All these three disturbances may disturb the metabolism of the cell more or less seriously so that new, unexpected substances that may be toxic or allergenic may be produced. This has also been found in a number of cases indicating that the consequences may be dangerous.

Unexpected harmful substances may appear in organisms that are not only substantially equivalent with, but completely equal with their natural counterpart but for the harmful substance. Therefore substantial equivalence is irrelevant and useless as a criterion for safety assessment of genetically engineered foods. There is no scientific alternative to rigorous toxicological testing to ensure satisfactory safety for genetically engineered foods.

As none of the genetically engineered foods on the market today has undergone testing that is even close to the rigorous assessment required to ensure safety for the world population, it is necessary to immediately stop the production and trade with these products until they have undergone the necessary testing. But even so a residual risk remains that some harmful substance will pass undetected.

Labelling of all foods and food ingredients, that are genetically engineered, is absolutely necessary in order to make it possible to trace, as early as possible, unexpected harmful effects of genetically engineered foods.

Introduction

Proponents of genetic engineering of foods have been maintaining that there is no important difference between this technology and breeding. Therefore they argue that it is unjustified by authorities to treat these products differently than natural or conventionally bred foods. Consequently it is argued that engineered foods that are "substantially equivalent" to their natural counterparts should be treated in the same way as natural foods by the legislation.

It is therefore most important to critically scrutinise different arguments that are used to justify the above position.

First, however, let us define genetic engineering and breeding:

Genetic engineering means the artificial insertion of one or a few genes into the genome of a host organism. This allows genes from completely alien species to be introduced.

Conventional breeding means the combination of hereditary traits by natural mating procedures.

Arguments forwarded for the opinion that breeding and genetic engineering are not different

ARGUMENT 1. "Manipulative introduction of genes to other species is not a novel thing. This has been done in crossbreeding between different species since thousands of years".

COMMENT: This argument is seriously misleading.

Crossbreeding is not a manipulation of single genes as is the case in genetic engineering. It is just natural mating. Only closely related species, having almost the same genes, with the exception of a few percent, can mate. From a genetic standpoint it is misleading to equate interbreedable species with completely different species. Rather the former could be regarded as genetic variants of the same species than different species as they are encompassed by a common species barrier. Breeding does not introduce new genetic material into the gene pool of the species but merely selects and brings together in one individual certain genes that are already in the gene pool.

So there has not been any "manipulative introduction of genes" to genuinely different species at all in conventional breeding.

ARGUMENT 2. Reshuffling of genes and parts of chromosomes happens frequently in nature. So "reshuffling" through intentional insertion of a gene is not a very different thing.

COMMENT: This argument is again seriously misleading.

1. Natural reshuffling occurs by recombination mechanisms that are very precisely guided by sequence homology. In effect they replace a gene or genes with allies of the same gene that differ from the original by only a few bases ("code syllables"). They replace one module with another almost identical one. In addition, there are species protection barriers that disallow reshufflings that are incompatible within the species. 2. While natural reshuffling means a minor reorganisation of the genetic code sequence within a gene while it remains in its usual place, genetic engineering represents a random insertion of an often completely alien code sequence into the code sequence of the host. Thereby the natural sequence will be interrupted, which never happens in natural reshuffling.

In addition, to be successful, the insertion has to occur in those parts of the DNA that are active. Otherwise the inserted gene (transgene) will remain silent. However, the integration of the transgene into an active DNA region will always disturb the functioning of the DNA. This will not only take place by the transgene disrupting the sequence of a host gene, but we now know that the mere presence of another foreign, active gene in the vicinity of other active genes will be highly perturbing to host gene function(s).

These effects therefore combine to actually maximise the chances of causing disruptions to host and transgene function in a totally unpredictable manner. This in turn maximises the degree of biochemical disturbance resulting from the disrupted gene function.

Therefore, GE of animals and especially plants, always results in a loss, to a lesser or greater degree, of the tight genetic control and balanced functioning which is retained through conventional cross breeding. With GE, host genes can be silenced (rendered inactive) or inappropriately activated resulting in either a deficiency in a given protein(s) or the presence of the wrong protein(s)in the wrong place or in the wrong quantity or all these combined! So there is an unpredictable creation of new molecules that may be toxic, allergenic or may disturb the metabolism of the cell so as to create unexpected toxic or allergenic molecules (see eg 1, 2,3,4,5).

These phenomena which are technically called "position effects", complicate the production of every GE crop or animal.

So the insertion of an alien gene is something profoundly different from the minor modifications resulting from the reshuffling that occurs naturally (otherwise there would of course not have been any stable species).

The most important consequence of this difference is the appearance of unexpected substances that may be harmful. The serious problem with this is that there does not exist any completely reliable method to identify unexpected harmful substances even with the most rigorous safety assessment methodology.

ARGUMENT 3. In modern laboratory breeding procedures, insertion of parts of chromosomes, so called bridges, has been done without any serious problems.

COMMENT: There is not much experience from commercial exploitation of this technique. In addition, to the extent that this kind of insertion occurs randomly as in genetic engineering, it runs the same kind of risk of creating unexpected substances as described under argument two.

CONCLUSION: There are fundamental differences between breeding and genetic engineering. Contrary to breeding, genetic engineering means random insertion of a sequence of genetic codes into active DNA, disrupting the precise sequence of genetic codes of the host and disturbing the functioning of neighbouring genes. This may give rise to unexpected and potentially toxic or allergenic molecules. Or the nutritional value may be altered.

Most importantly, genetic engineering is capable of introducing into an organism genetic information from very distant species. There is no body of existing information upon which to base prediction regarding how such divergent genetic information will affect the functioning of the recipient organism. This increases tremendously the risk of unanticipated side-effects from such manipulations.

Since many of the genes now being introduced into food-producing organisms are derived from organisms that have never been part of the human food supply we have no way of knowing how humans will respond to the effects of these genes on the food and whether such food is appropriate for our species. They may be allergenic or toxic to humans or they may disturb the metabolism, giving rise to toxic or allergenic molecules.

Extensive experience from toxicology shows that even the most rigorous safety assessment methods may fail to detect unexpected harmful molecules.

There is no significant experience from the field of conventional breeding that can be used to justify equal treatment of genetically engineered products and their natural counterparts.

Furthermore, there is a special kind of risk that is unique to genetic engineering. It is the use of viral genes and vectors in genetic engineering. Research has shown that this genetic material is unstable and may recombine with infecting viruses, giving rise to new viruses that may become potentially dangerous pathogens for plants, animals and human beings.

Because of the fundamental difference between these technologies and especially because the unique aspects of genetic engineering brings with it the risk of emergence of new, unexpected molecules, some of which may be harmful and difficult to detect, it is scientifically unjustified to equate foods produced by genetic engineering with natural or conventionally bred foods in legislation concerning safety aspects.

CONSEQUENCES FOR SAFETY AND LABELLING ISSUES:

The use of the principle of substantial equivalence for risk assessment of genetically engineered foods, as it has been applied by FAO, the FDA of USA and EU, is potentially very dangerous as it neglects the potential presence in these foods of unexpected new molecules. A product could not only be substantially equivalent but even be IDENTICAL with its natural counterpart in all respects except for the presence of a single harmful compound. In other words, substantial equivalence is irrelevant and useless, from a scientific standpoint, as a criterion of food safety (6)(7).

To minimise the risk of unexpected harmful effects, it is necessary to require all genetically engineered foods and food ingredients to be submitted to rigorous safety assessment even if no method of comparing them with their natural counterpart has been able to reveal any difference. There is no scientifically tenable alternative to rigorous safety assessment including clinical tests using paid human volunteers for assessment of toxicity as well as allergenicity (8).

As none of the genetically engineered foods on the market today has undergone testing that is even close to the rigorous safety assessment required to ensure safety for the world population, it is necessary to immediately stop the production and trade with these products until they have undergone the necessary testing. But even so, a residual risk remains that some harmful substance will pass undetected.

Labelling of all foods and food ingredients, that are genetically engineered, is absolutely necessary in order to make it possible to trace, as early as possible, unexpected harmful effects of genetically engineered foods.

References with short comments:

1. Violand BN et al. Protein Science. 3:1089-97, 1994. Unexpected appearance of an abnormal aminoacid, epsilon-N-acetyllysine, in the production of bovine growth hormone by a genetically engineered bacterium.

2. Reddy SA, Thomas TL. Nature Biotechnology, vol 14, 639-642, may 1996. Unexpected appearance of a toxic substance, octadecatetraeic acid, at genetic engineering of a tobacco plant for production of gamma-linolenic acid.

3. Inose, T. Murata, K. Int. J. Food Science Tech. 30: 141-146, 1995. Unexpected appearance of a toxic and mutagenic metabolite, methyl-glyoxal, at the genetic engineering of a yeast to increase fermentation rate.

4. Nordlee, J.A. et al.The New England Journal of Medicine 14: 688-728; 1996. Appearance of allergy against nut allergenes at the insertion of a nut gene into a soy bean.

5. Mayeno, A.N. et al Tibtech 12:364, 1994. Appearance of a highly toxic abnormal variety of the aminoacid tryptophane at the genetic engineering of a bacterium for increased produtivity. As the bacterium was destroyed, experimental proof is impossible, but this toxin has never appeared at production of tryptophan by nonengineered bacteria of the same kind.

6. Fagan J. "The failings of the principle of substantial equivalence in regulating transgenic foods". http://www.psrast.org/jfsubeq.htm

7. OECD, DSTI/STP (95)18, Paris, 1995, pages 79-87. Paper presented at an OECD workshop. It concludes that substantial equality is an unreliable basis for safety assessment of genetically engineered food.

8. Fagan J. "Assessing the safety and nutritional quality of genetically engineered foods". http://www.psrast.org/jfassess.htm

______________________________________________________________________________

NOTE: This paper was prepared for and sent to the delegates at the FAO Codex Alimentarius meeting in Ottawa in April 1997. The Codex, is the international forum for establishing the food policy of countries belonging to World Trade Organisation (WTO).

The board of Codex had decided to support the proposition of USA to deregulate labelling of GE-foods entirely (that is, no requirement for any kind of labelling). Fortunately, and to the surprise of the board, the meeting decided not to follow the recommendation of the board, and expressed strong criticism to the board because it had been very late in informing about its position. Only five countries, including USA, voted in favour of complete deregulation of labelling. Unfortunately, the delegates were unable to decide upon an alternative solution as none had been prepared by the board. So the issue was referred back to the board.