Patrick Brown
Professor
College of Agriculture & Environmental Science
University of California
Davis, CA 95616

Crop cultivars developed using recombinant-DNA technologies (rDNA crops) have been rapidly adopted by agricultural producers in the United States; and until recently, foods derived from these crops have been tacitly accepted by US consumers. In contrast, many European consumers have shown a marked resistance to these technologies which, in turn, has resulted in the passage of trade restrictions and of laws that limit the import, growth or use of rDNA crops throughout much of Europe. The public uproar in Europe, and the protests surrounding the World Trade Organization meeting in Seattle, has now raised the awareness of many in the USA and given birth to a vocal and growing group of concerned consumers.

The intensity of the current debate has surprised many in the scientific community and has escalated into a highly polarized and increasingly antagonistic debate. Scientists, and the professional organizations that represent them, have been publicly supportive of this technology and often dismissive of public concerns. Most scientific comment suggests that 'education' is the key to gaining the needed acceptance, while almost no comment has recognized or adressed the fears of the public. Those who oppose rDNA technology interpret the apparent willingness of the US scientific community to embrace this new technology, while failing to adequately address the potential risks, as a betrayal of public trust.

Public uncertainty has resulted in the loss of markets, and will increasingly do so, for the current generation of rDNA crops and foods. Though this is clearly of substantial economic concern, by far the most significant consequence of public concern is the threat that this conflict poses for the entire field of plant biotechnology which holds far greater promise of human benefit than that offered by any existing rDNA crop. The loss of this technology through careless and premature implementation would be truly devastating to the goal of developing more abundant and nutritious foods in an environmentally sensitive fashion.

This issue requires immediate and thoughtful attention from plant scientists. We must recognize that our knowledge of the processes that regulate gene incorporation and expression are in their infancy and that our capacity to manipulate the plant genome is crude. Given this current lack of understanding it is certainly possible that the current regulatory safeguards are inadequate and may not be offering sufficient protection
against inadvertent creation of health and ecological problems.

Since the public education and research system is based upon a foundation of public trust, it is essential that we recognize and admit the unknowns associated with molecular biology and act with caution and integrity.

The following text describes some of the uncertainties associated with rDNA technology and illustrates how the scientific community's defense of the current generation of rDNA crops represents a substantial threat to the future of this promising new technology.

Are the Current Generation of rDNA Crops, and the Regulatory System that approved them, Scientifically Defensible?

In 1989 the National Research Council, following extensive scientific review, publicly concluded that crops derived from rDNA techniques do not differ substantially from those derived using traditional techniques. This conclusion forms the basis for current FDA policy that regulates the production and use of rDNA crops and foods. This conclusion is based upon the principle of "substantial equivalence" which states that the introduction of a gene of known and safe function into a crop of known characteristics is technologically neutral, hence the resulting crop can be presumed to be safe and is not subject to mandatory testing prior to release or use in foods. As this principle is central to the scientific and regulatory acceptance of this technology it deserves careful examination.

Is There Equivalence between rDNA and 'Traditional' Sexual Gene Transfer?

To adequately compare these technologies it is essential that each is well characterized and understood. The molecular processes that control gene incorporation and expression following a normal sexual crossing event, however, are only poorly understood and the extent of our ignorance is further revealed weekly as new processes involved in the regulation of gene expression in plants are determined. The inadequacy of our understanding is well illustrated by the host of genetic phenomena (such as co-suppression, intron-mediated enhancement, transcriptional regulation, protein-gene interactions etc) for which we have essentially no mechanistic understanding. Our knowledge of these processes is clearly in its infancy and few would claim that we understand more than a small percentage of the processes regulating sexual reproduction in plants.

Further, most of what is known of gene transfer using traditional and rDNA techniques illustrates the profound manner in which they differ. Traditional crossing involves the movement of clusters of functionally linked genes, primarily between homologous chromosomes, and including the relevant promoters, regulatory sequences and associated genes involved in the coordinated expression of the character of interest in the plant. The molecular regulation of this process and the biochemical and evolutionary significance of these controls is poorly understood.

In contrast to traditional techniques, current rDNA technologies (those used in all currently approved rDNA crops) involve the random insertion of genes in the absence of normal promoter sequences and associated regulatory genes. As there are very few examples of plant traits in which we have identified the associated regulatory genes, the introduction of a fully 'functional' gene using rDNA techniques is currently not possible. R-DNA techniques also involve the simultaneous insertion of viral promoters and selectable markers and facilitates the introduction of genes from incompatible species. These genetic transformations cannot occur using traditional approaches - which further illustrates the profound manner in which these processes differ.

Genetic material can be moved within and between species by the poorly understood processes of gene transposition. Though the occurrence of this phenomenon in traditionally bred plants is superficially equivalent to rDNA techniques (which involve the random insertion of "artificial transposons"), the mechanisms governing this process and the significance of transposition in traditional gene transfer are unknown. Given our profound lack of understanding of these processes it is impossible to compare sensibly the two processes. Indeed, it can be argued that gene transfer via rDNA techniques resembles the process of viral infection far more closely than it resembles traditional breeding.

In summary, it is clear that gene transfer using rDNA techniques is substantially different from the processes that govern gene transfer in traditional breeding. The extent to which these processes differ will become increasingly clear as as we gain a better understanding of the processes governing gene movement, expression and regulation.

The presumption of "Substantial Equivalence" - the basis for current regulatory principles - is profoundly flawed and scientifically insupportable.

Do rDNA Techniques offer Greater Precision?

One of the much-touted benefits of rDNA techniques is the capablity to introduce only a discrete and well defined number of genes into the new cultivar whereas a traditional crossing event introduces thousands of genes. This ability to control the types and numbers of genes introduced speeds the introduction of a gene of interest by eliminating the need for extensive backcrossing to the elite parent. Many have suggested that this approach is fundamentally more "precise" than traditional breeding techniques and have argued that the technique is consequently "safer".

The ability to introduce a precisely defined compliment of genes using rDNA techniques, however, is not equivalent to the introduction of a precisely defined and biologically integrated character. Whereas the incorporation of a new character using traditional techniques occurs in a fully functional and appropriately regulated manner, rDNA gene introduction is more or less random, and does not involve introduction of the regulatory sequences normally associated with that gene. Traditional techniques, therefore, result in greater "biological precision" than random gene insertion using rDNA techniques.

The FDA policy statement further suggests that it is highly unlikely that rDNA techniques will result in the inadvertent production of allergens or toxic compounds and that once incorporated into the genome, the introduced gene functions like all other genes in the genome. These statements are offered in support of the premise that rDNA experiments are more predictable than traditional breeding approaches. This presumption is, however, clearly contradicted by a large volume of scientific literature and experimental experience that illustrates the propensity of rDNA techniques to produce unexpected and often lethal perturbations. Indeed metabolic and phenological perturbations are very frequently observed following transformation events and a high percentage of transformants show profound growth aberrations. Indeed the propensity of random gene introduction to cause metabolic disruption is well documented and actively used to probe gene function.

While extreme aberrations can be easily selected out, it is also highly likely that undetected biochemical perturbations remain following essentially all transformation events. Since it is not standard practice to screen transformants there is clearly a potential for biochemically abnormal trangenic plants to persist. This is further exacerbated through the use of tissue culture and embryo rescue etc. which can be used to "rescue" metabolically altered transgenic plants that might otherwise have been eliminated during early plant growth. Whether or not these same perturbations occur following traditional breeding is unknown. Lack of knowledge, however, is not proof of safety.

The metabolic perturbations caused by rDNA gene introduction may result in production of toxic compounds. Many plant species have the capacity to produce toxic compounds which under natural conditions serve to protect against animal and insect predation as well as contributing to disease resistance mechanisms. In certain species, such as those in the Solanum family, there are many well characterized and highly unpalatable or toxic compounds. It is very likely that the majority of the genes involved in the formation of these toxic and unpalatable compounds are still present (though not expressed) in modern tomato and potato. Given the random nature of rDNA gene insertion, and the use of a promiscuous viral promoter sequence, the potential clearly exists that tomato could be induced to produce a toxin as a result of a rDNA gene transfer. Whether this would occur with the same frequency following traditional sexual breeding is unknown. The presumption that it cannot occur is clearly invalid.

Clearly the assumption that a transformed crop is exactly the sum of the original crop and the introduced gene is not acceptable. rDNA techniques are profoundly different from traditional breeding methods and are well known to cause unexpected metabolic perturbations. The principle of substantial equivalence is not scientifically justifiable; hence we can make no a priori assumption of the safety of any rDNA manipulation.

Do rDNA Techniques Provide an Acceptable Level of Risk?

The preceeding discussion clearly demonstrates that the risks associated with rDNA technology cannot be determined given current understanding of gene expression. Nevertheless it has been argued that risk is a normal part of technological advancement and that acceptance of this risk is warranted in the instance of rDNA crops.

While it is true that we accept risks as a normal part of life, most of the risks we accept are defined by experience and are understood before they are taken. Some risks are also taken because the rewards are perceived to outweigh the risks. Traditional breeding has on the whole been an acceptable risk with 10,000 years of experience, and a trust in the motives of those producing the new cultivars.

Many, however, are not yet prepared to accept the risks of rDNA
technologies. This is in part due to a lack of understanding of the risks, the minimal benefit of the current crop of GMOs, and a mistrust of the motives of those selling the technology. Given the current state of our knowledge of this technology and the nature of the GMOs currently available, this lack of public trust is entirely reasonable. Public acceptance will require convincing demonstration of safety and the development of crops with a more direct benefit to the consumers.

The concerns expressed by many are further validated by the current generation of GMOs that have been incorporated into the food system without adequate public consultation and scientific scrutiny. The current generation of GMO crops do not provide any tangible public benefit, have not contributed to reduced food costs, and have no confirmed ecological benefit. This is well illustrated by the two most prevalent types of GMOs
in use in the US.

Insect-resistant crops containing the gene encoding the Bacillus thuringiensis toxin have been planted widely in the US. This transgenic technique promises to reduce the use of pesticides and reduce growers' costs. While reduction in pesticide use is an admirable goal there are significant grounds to question the appropriateness of the current generation of Bt-producing crops and to question the haste with which these crops were released for widespread use.

The current generation of Bt crops utilize a single Bt gene rather than the complex of Bt genes that are available. There is widespread agreement amongst scientists that this use of a single Bt gene will increase the speed with which pest resistance will develop. To help alleviate the development of insect resistance the USDA and Monsanto now advise growers to plant refuge areas to ensure non-resistant insects persist under the premise that this will reduce the rate of resistance development. While this is theoretically sound there is insufficient ecological data to determine optimal size of these refuges or to estimate how effective they will be.

The current generation of Bt crops also utilize antibiotic resistance as the selectable marker and rely upon viral promoters to ensure high degrees of expression. This clearly introduces a risk associated with a promoter designed to be free of regulatory controls, it excites those who see viral and antibiotic-resistance genes as threatening, and it ensures that the Bt protein is distributed uniformly throughout the plant. The uniform presence of the Bt protein enhances the likelihood of resistance development and ensures that the protein is present throughout plant development and is present in the pollen. The death of Monarch larvae was a direct consequence of the presence of active Bt toxin in the pollen. While some have questioned the scientific relevance of this study it did illustrate the inherent flaws in this cultivar.

Methods exist (or will soon exist) that make the use of viral promoters and antibiotic resistance markers unnecessary. There is no justification for the expression of Bt in the pollen, and the release of cultivars with a single Bt gene is certain to hasten resistance development. In the absence of data to support the refugia concept there is very little to prevent the development of widespread insect tolerance of Bt.

Clearly the release of the first generation of Bt-containing crops was premature and based upon flawed scientific principles. Regulatory and scientific support for this cultivar is clearly questionable.

The other dominant type of GMO in use today is the Roundup-Ready varieties of cotton, soyabean and corn. Not only do these cultivars contain many of the same questionable genes as those in Bt crops, but also they have the additional propensity to contribute to the development of herbicide-resistant weed species for which the consequences are poorly understood. Roundup-Ready crops are also of questionable ecological value and build a long-term dependence on the use of the herbicide Glyphosate. Not insignificantly, the overtly 'corporate' nature of these crops and the dependence they build on high cost and ecologically questionable technologies has resulted in widespread suspicion of the motives of those promoting these cultivars.

It is abundantly clear that the current generation of GMO's were developed using an untested and unsophisticated technology and were released prematurely to ensure early returns on corporate investment. Clearly this does not represent a sound jusification for the release and widespread use of these crops.

Perhaps one of the most profoundly flawed justifications of GMOs is illustrated in the often cited refrain "GMO foods have been widely available in the marketplace for the past 5 years and not one incident of harm to public health has been documented". Since every introduced gene is inserted into a different genetic location, and every gene differs in functions and interactions within the genome, and as every species can be expected to 'react' differently to the gene introduction process, it is clear that the safety of one GMO is in no way predictive of the safety of another. In many respects the claim of safety by association is no more valid than the claim that the safety of aspirin predicts the safety of all future drugs.

Conclusion

The real threat to the future of plant biotechnology is the irresponsible and premature releases of the first generation of GMOs that are full of unsound scientific assumptions, rife with careless science, and arrogantly dismissive of valid concerns. The current generation of GMOs provide little real benefit except corporate profit and marginally improved grower returns, while at the same time introducing a host of poorly studied human and ecological risks. Not surprisingly, many have questioned the value of these crops and the integrity of those who support their use.

Given these issues and the overall lack of knowledge of rDNA technology it can only be concluded that the current FDA regulations guiding the release and testing of GMOs is inadequate. It can further be concluded that the technology is inadequately developed to ensure its safety. In the absence of a sound scientific basis to predict the full consequences of rDNA crop development, we must either subject all new crops to a rigorous testing program that considers all potential health, social and environmental concerns or halt further release of rDNA crops until a firm scientific understanding of the biological principles is attained.

As scientists it is our responsibility to recognize that we do not yet have sufficient knowledge of the process to use it safely. We must work towards adressing all of the concerns explicit in the current generation of crops, and must support a rigorous testing program to ensure the safety of all GMO food stuffs in the interim. To date many in the scientific community have been unwilling to rationally consider the concerns surrounding the current GMOs and have wrongly considered that a defense of GMOs is a prerequisite to protect the science of plant biotechnology. Nothing could be further from the truth or more threatening to the future of this technology.