by Institute of Science in Society (Prof. Joe Cummins, Dr. Mae-Wan Ho and Lim Li Ching)

Monsanto has applied to import its GM oilseed rape GT73 into Europe for use in animal feed and processing. The Scientific Panel on GMOs of the European Food Safety Authority has given it a favourable opinion, and there will soon be a vote on it at the Council of Ministers. Here's a description of what it is and why it should be rejected.

Oilseed rape is a major crop for oil and animal feed

Oilseed rape (Brassica napus) is grown as a commercial crop in 50 countries with a combined harvest of over 40 million metric tonnes. The major producers of rapeseed in 2000 were China, Canada, India, Germany, France, Australia, and the United Kingdom. Canola is a genetic variation of B. napus with low levels of the natural rapeseed toxins glucosinolate and erucic acid. Canola is grown for its seed, which represents a major source of edible vegetable oil and pressed cake from oil extraction is also used in livestock feeds [1]. Oilseed rape is called canola in North America because the commercial oil-producing varieties were developed in Saskatchewan, Canada.

Monsanto's canola GT73 was released commercially in 1995 in Canada [2] and the same strain, designated RT73, was released commercially in the United States in 1999 [3]. Japan approved the release of GT73 in 1995 [1] and Australia in 2003 [1]. Approval of all releases was based on essentially the same data sets.

GT73 in the EU

GT73 was notified for food use (as rapeseed oil) in the European Union (EU) in November 1997, under the simplified procedure of the Novel Foods Regulation. This means that rapeseed oil from GT73 was considered `substantially equivalent' to its conventional counterpart and only required notification by the company, with no risk assessment or explicit approval process. Products made from rapeseed oil may include fried foods, baked foods and snacks.

An application for the import and use of GT73, excluding cultivation, was submitted in 1998 to the competent authority of the Netherlands. It gave this application a favourable opinion, and in January 2003 recommended that GT73 be approved. Several member States raised questions, including the UK, via its Advisory Committee on Releases to the Environment (ACRE) [4]. One of the concerns related to increased liver weights in rats fed GT73, compared with controls (see later).

The European Food Safety Authority's (EFSA) Scientific Panel on GMOs was requested to give its opinion on GT73 to resolve the uncertainties. In February 2004, EFSA gave its verdict that "GT73 oilseed rape is as safe as conventional oilseed rape and therefore the placing on the market of GT73 oilseed rape for processing and feed use is unlikely to have an adverse effect on human or animal health or, in the context of its proposed use, on the environment" [5].

Despite EFSA's positive assessment of GT73 for feed and processing, the regulatory committee could not reach a qualified majority to authorize GT73 in June 2004. There were 43 votes in favour of approving GT73 (Belgium, Czech Republic, Finland, France, Netherlands, Latvia, Portugal, Slovakia, Sweden), 57 votes against (Austria, Cyprus, Denmark, Estonia, Greece, Hungary, Italy, Malta, Lithuania, Luxembourg, Poland, UK), and 24 abstentions (Germany, Ireland, Spain, Slovenia) [6].
The application now passes to the Council of Ministers, which will make its decision very soon. If the Council cannot decide, the decision will rest with the European Commission, which has shown every sign of being in favour of approving GT73.

No event-specific characterization provided of transgene insert

GT73 (RT73) oilseed rape has been made tolerant to the herbicide glyphosate. Two transgenes were used. The first is the epsps gene coding for the enzyme 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS), isolated from the common soil bacterium Agrobacterium tumefaciens, and is a glyphosate tolerant form of EPSPS. The EPSPS enzyme is part of the important shikimate pathway involved in the production of aromatic amino acids. When conventional canola plants are treated with glyphosate, the plants cannot produce the aromatic amino acids and die, but the enzyme encoded by the transgene is insensitive to glyphosate.

The second transgene in GT73 codes for a modified version of glyphosate oxidase (GOX) enzyme. The gox gene inserted into GT73 was isolated from the bacterium Ochrobactrum anthropi. The GOX enzyme accelerates the normal breakdown of the herbicide glyphosate into two compounds, aminomethylphosphonic acid (AMPA) and glyoxylate [1-3, 7]. In the absence of GOX, unacceptable levels of the herbicide may accumulate in the canola cake in animal feed.

The two transgenes were introduced into GT73 in a plasmid using the bacterium, Agrobacterium tumefaciens. The epsps and gox genes were each driven by the 35S promoter from a modified figwort mosaic virus and terminated with the 3' (terminal) end of the pea rbcS E9 gene. The shikimate pathway is located in the chloroplast, so the chloroplast transit signal peptide sequence from the ribulose-biphosphate carboxylase and EPSPS of Arabidopsis is used to target the transgene products to the chloroplast. According to the company, only the primary genes and the sequences necessary for their activity in the plant cell were inserted into the canola cells while sequences from the plasmid such as the plasmid origin of replication and a gene for streptomycin resistance were lost from the commercial strain. Monsanto claimed that only one transgene insert is present [5], but the exact site of insertion was not reported [3, 5].

After evaluating the initial application submitted by Monsanto, some member States had requested additional information on the molecular aspects of the dossier. However, it is clear from the EFSA opinion that no independent tests were carried out, and the favourable opinion was based solely on information supplied by the company. Worryingly, the EFSA opinion [5] stated: "Comments raised by the Member States on specific molecular detection methodologies are presently not within the scope of the GMO Panel remit." In other words, there is no event-specific characterization, and therefore, no unique method for detecting this GMO for the purpose of identification or traceability, nor for addressing safety and liability issues that may arise.

The same EFSA dossier revealed that there are molecular changes at the insertion site, specifically 40 bp of the host genome is missing from GT73 while 22 bp of extraneous DNA of unknown origin is present at the 5' junction of the insert. Nevertheless, these are considered not to pose a safety risk, based solely on the lack of homology to known toxins and allergens.

No molecular evidence of transgene stability

The transgenes were claimed by the company to be inserted in a stable and Mendelian fashion. ISIS has pointed out more than once that this claim of genetic stability - based on a failure to depart from `Mendelian ratios' in the offpring generation - is not an acceptable criterion of genetic stability in the absence of independent ascertainment of the parental genotypes [8-13]. But EFSA has accepted the same criterion of transgene stability. It stated [5]: "The inserted DNA is inherited in a stable fashion in a nuclear chromosome as indicated by a number of parameters, e.g. predicted Mendelian segregation ratios (over several generations) from crosses between GT73 and conventional oilseed rape."

Extensive changes in the codons of transgenes from native genes ignored

Few of the regulatory documents have dealt with extensive alterations in the genetic codes of the native genes in the transgenes inserted into GT73, but all of them acknowledge that the codes were altered to enhance production of the bacterial gene products in the plant. The United States Food and Drug Administration consultation on canola GT73 provided a somewhat fuller description of the alterations in the bacterial DNA [14] while the patent for the EPSPS used in canola GT73 provides an extensive description of the code alteration [15]. Native genes from bacteria or humans do not function very well in crop plants because gene _expression is influenced by codon bias specific to plants, mammals or bacteria. For that reason, the genetic code is altered by genetic engineers to achieve optimum gene _expression. The optimized transgenes used in modified crops are mainly synthetic approximations of the real bacterial gene [16]. The synthetic genes are very different from the genes that evolved in bacteria and for that reason their characteristic recombination and mutation deserves special attention, as does its toxicology and allergenic potential. However, these factors have been largely ignored by the regulators.

Toxicology & allergenicity tests invalid

Even though the transgenes were altered in DNA sequence from the native bacterial genes, the proteins actually tested for mammalian toxicity and environmental safety were not isolated from GT73 but from the bacteria [5]. The bacterial surrogate enzymes were assumed to be identical to the enzymes produced in GT73 by cursory observations using techniques such as gel electrophoresis, N terminal analysis and enzyme activity, even though the presence of four anomalous amino acids were noted in the bacterial GOX [7]. Digestibility and degradability were tested with the bacterial proteins in simulated gastric fluid. And acute toxicity tests in mice were similarly done with the bacterial proteins.

Allergenicity tests were even less reliable, as they depended on theoretical evaluations based on assumptions that have been extensively questioned. For example, the Austrian government, based on an analysis of a number of applications for GMO approval in the EU, has concluded that no direct testing of potentially allergenic properties of GM corps and their products has been carried out [17].

Instead, conclusion that the protein in question is unlikely to exhibit allergenic properties is largely based on the following theoretical considerations: the newly introduced protein originates from a non-allergenic source; there is no significant sequence homology to known allergens; the protein will be rapidly digested in the intestine; the protein is not glycosylated; the _expression level of protein in the GM crop is low; and the protein is not new to the human diet. The Austrian government has questioned each of these arguments and their underlying assumptions in the light of recent scientific data.

Consequently, these tests were neither meaningful nor valid. Empirical tests should have been conducted at the very least, on the real proteins isolated from GT73, not the bacterial surrogates.
The EFSA did include the warning that, "Since cross-reactivity between GOX and tropomyosin is not ruled out completely, persons allergic to shrimp meal should be aware of the possibility of hypersensitivity reaction when working with GT73 oilseed rape."

Inadequate inappropriate feeding trials with unexplained adverse effect

According to the EFSA opinion [5], "A satisfactory explanation was sought for the potentially adverse effect observed in one of the three rat feeding studies." We believe that this refers to the concerns expressed in regard to a confidential Monsanto feeding study that showed that rats fed GT73 experienced a 15% increase in their liver weights.

The UK's ACRE and ACAF (Advisory Committee on Animal Feedingstuffs) had first raised concerns in March 2003 that the difference in the rats' liver weights could not be explained, as volunteered by Monsanto, by higher glucosinolate concentration in the GM diets compared with the corresponding control diets [4]. Subsequently, Monsanto provided further information on this. But both ACRE and ACAF were "not satisfied" that Monsanto had supported their hypothesis. They demanded a satisfactory explanation for this potentially adverse response.

However, it appears that EFSA has dismissed those concerns. A list of uninformative feeding trials was presented on various animals of extremely short duration in which mostly body weights and sometimes, liver weights were recorded. No histology was carried out. Because there were no apparent gross pathological changes in the rat livers following examination at necropsy, EFSA considered the difference in liver weights an "incidental finding".

Contamination unavoidable

The regulatory reviews leading to commercialization of GT73 oilseed rape without exception discounted the rapid pollution of transgenic crops by wind spread pollen or by seed dispersal by animals or vehicles. This can happen during transport, without planting in the field. Escaped seed can germinate and potentially cross-pollinate with conventional oilseed rape, feral populations and wild relatives. ACRE had also raised concerns regarding seed spill, and was "not convinced that seed spill will not occur and that feral populations will not materialise" [4].

There is clear and growing evidence that widespread deployment of GM oilseed rape will lead to widespread contamination of conventional crops. A 2003 report showed that 95% of certified seed stock in western Canada were polluted to detectable levels with glyphosate tolerance genes and 52% exceeded the allowable contamination of certified seed [18]. The widespread deployment of GM oilseed rape for a variety of herbicides is leading to pyramiding of the genes for herbicide tolerance [19], creating crops that turn into fertile weeds that are difficult to eradicate.

Europe's oilseed rape should keep its GM-free status before it too is contaminated beyond redemption.

References
1. Agbios Data base product description MON-00073-7 (GT73, RT73) pp1-3 http://www.agbios.com/dbase.php?action=Show Prod&data=GT73%2C+RT73
2. Canadian Food Inspection Agency Plant Biosafety Office Decision Document DD95-02:Determination of of environmental safety of Monsanto Canada Inc.'s Roundup Herbicide tolerant Brassica napus canola line GT73 1995, pp1-10 http://www.inspection.gc.ca/english/plaveg/bio/dde.shtml
3. Animal and Plant Health Inspection Service USDA Docket no. 98-089-2 Monsanto co. Determination of Nonregulated status for canola genetically engineered for glyphosate tolerance 1999, pp1-33. http://www.aphis.usda.gov/brs/aphisdocs2/98_21601p_com .pdf
4. ACRE Advisory Committee on Releases to the Environment. Advice on a notification for marketing of herbicide tolerant GM oilseed rape, 24 September 2003, http://www.defra.gov.uk/environment/acre/ad vice/pdf/acre_advice36.pdf
5. Opinion of the Scientific Panel on Genetically Modified Organisms on a request from the Commission related to the Notification (Reference C/NL/98/11) for the placing on the market of herbicide-tolerant oilseed rape GT73, for import and processing, under Part C of Directive 2001/18/EC from Monsanto. The EFSA Journal 2004, 29, 1-19
6. "New Europe" blocks U.S. food import, Friends of the Earth Europe, 16 June 2004, http://www.foeeurope.org/press/2004/GR_16_june_US_food .htm
7. ANZFA Australia New Zealand Food Authority. Draft risk analysis report application A363 Food produced from glyphosate tolerant canola line GT73, 2002, pp1-73 http://www.agbios.com/docroot/decdocs/01-290- 009.pdf
8. Ho MW and Cummins JC. GM food & feed not fit for "man or beast". ISIS Report 29 April 2004
9. Ho MW. GM Science Review deeply flawed. Science in Society 2003, 19, 7-9.
10. Ho MW. GM maize approved on bad science. ISIS Report 25 February 2002; also, Science in Society 2002, 15, 10-25.
11. Ho MW. Questionable stability at JIC. ISIS Report 2 March 2001
12. Ho MW. Letter to the Scottish Parliament Petitions Committee from ISIS. 28 February 2002.
13. Ho MW. GM rice unstable. ISIS News 9/10, July 2001.
14. US Food and Drug Administration Biotechnology Consultation Note to the File BNF No.000020 Monsanto's glyphosate tolerant canola line GT73 1995 pp1-4 http://www .cfsan.fda.gov/~rdb/bnfm020.html
15. Eicholtz D, Gasser D and Kishore G. Glyphosate-tolerant- 5-enopyruvyl-3-phosphoshikimate synthetase, 1999, United States patent 5,866,775.
16. Cummins J. Synthetic genes in food crops, ISIS Press Release 1 September 2004,
17. Spök A, Hofer H, Lehner P, Valenta R, Stirn S, Gaugitsch H. Risk assessment of GMO products in the European Union: Toxicity assessment, allerginicity assessment and substantial equivalence in practice and proposals for improvement and standardization, July 2004, Austrian Federal Environment Agency Monograph.
18. Cummins J. Transgenic contamination of certified seed stocks, ISIS report 2003,
19. Ho M. What lurks behind triple herbicide tolerant oilseed rape? ISIS report 2002,