Science in Society Archive

Reply to Questionnaire Codex Guideline for the Conduct of Food Safety Assessment of Foods Derived from Recombinant-DNA Plants

Professor Joe Cummins and Dr. Mae-Wan Ho

For Independent Science Panel*

Question # 1: In your view, what would be important factors in determining if a recombinant-DNA plant is to be considered a A Recombinant-DNA Plant Modified for Nutritional or Health Benefit@, and as such within the scope of the proposed annex?

Answer: First, it is essential to exclude the evaluation of pharmaceutical products such as oral vaccines, immune modulating proteins, hormones, antimicrobials and growth factors  from the present considerations. Such products should be considered separately and reviewed independently from the current annex; they are inherently hazardous and require special consideration under a separate annex.

 Plants modified for nutritional or health benefit should include only those with enhanced vitamins, minerals, antioxidants (for example plant phenolics (flavonoids) known to fight cancer) and enhanced primary metabolites such as essential amino acids and health promoting reduced linolenic fats. Recombinant genes derived from a plant used to modify another plant species should not automatically be deemed safe for humans. This is highlighted in a recent investigation in which a protein from bean was found immunogenic when expressed in pea [1]. Specifically, immunological assessments carried out for the first time on a transgenic protein revealed that post-translational processing subsequent to gene transfer into an alien species introduced new antigenicities that turned a previously harmless protein into a strong immunogen. In addition, the transgenic protein promoted immune reactions against multiple other proteins in the diet.

As practically all the transgenic proteins involve cross-species gene transfer, they will be subjected to different post-translational processing, and hence they too, will have the potential to become immunogenic. And yet, none of the transgenic proteins that have been commercially approved has been tested. This omission is a most serious public health issue, and the Independent Science Panel has already called for an immediate ban on all GM food and feed until proper assessment on the immunogenicity of the transgenic proteins has been carried out [2]. This should apply all the more so to GM crops coming to the market that are meant for human consumption.

Nutritional enhancement and traits for improved health may be developed using conventional breeding or marker assisted selection then those traits combined with recombinant traits such as herbicide tolerance and/or insect resistance. Monsanto’s Vistive soybean, for example, combines natural genes reducing the oil content of linolenic acid, but also has transgenes conferring the Roundup Ready trait. The transgene activity is known to affect the precursor pools leading to formation of trans fatty acids [3]. The point is that when nutritional or health traits are combined with transgenic ones, the interaction of the two should be fully evaluated. Interactions between transgenic nutrition and health traits and other genetic modifications within a cultivar should be carefully evaluated, as should interactions between conventional nutritional or health traits with transgenes in recombinant strains with which they are combined.

A number of nutritional- and health-related transgenic plants are being developed (see Box). We must ask if these developments provide real benefits for people, and whether there are safer, cheaper and more effective alternatives for producing the nutritional/health benefit. More importantly, in every single case, it is necessary to assess the GM plant, the transgene and protein for toxicity and immunogenicity, for reasons given above.

Nutritional and health-related GM crops under development

· Cassava is being genetically modified to enhance activity of an enzyme that destroys the toxic cyanogenic glycosides [4]. But these compounds, which release cyanide when eaten, are normally destroyed if the cassava is adequately processed.

· A sweet protein, brazzien, was produced in GM maize by introducing a gene from an African plant. The sweetener was proposed for use in the food industry [5].

· A synthetic gene for porcine alpha-lactalbumin was introduced into maize along with a signal peptide from maize for localizing the protein to the endoplasmic reticulum. The pig-corn was supposed to provide a more digestible plant protein for humans [6].

· Human milk proteins genes were used to modify maize, supposed to improve human nutrition [7].

· Enhanced seed phytoesterol was achieved by transferring a gene from the rubber tree to tobacco [8].  The compounds would be used as cholesterol-lowering foods when food crops are modified.

· Canola was modified as a source of omega-3 fatty acid using fungal genes [9].

· Very long fatty acids were produced in Arabidopsis using genes from a tiny alga, a protozoa and a fungus [10].

· Soybean naturally deficient in methionine is being engineering to remedy the amino acid deficiency in free amino acids and in storage proteins [11].

· Soybean was modified with a gene from maize, delta-zein, which is rich in methionine [12].

· Maize was modified with a bacterial gene that provided increased free lysine in cellular pools [13].

· A grape stilbene synthetase was used to modify tomato to reduce lipid peroxidation [14].

· Maize modified with prentyltransferase from barley seeds resulted in a large increase in vitamin E [15].

· Golden rice expresses a daffodil gene to increase beta-carotene, which is then converted to vitamin A. This modification has now been extended to Indica rice [16]. The golden rice cultivars do not produce sufficient beta-carotene to fulfill the human requirement for vitamin A, however.

· Fructans are considered important functional foods because they promote the growth of beneficial intestinal bacteria. Fructans enhance mineral re-sorption, decrease cholesterol and hence may help prevent cardiovascular disease, colon cancer and osteoporosis. Onion fructosyl transferase was used to modify sugar beet, a crop that does not normally produce fructans [17].

· Deficiencies of iron and zinc in food crops are widespread, and approaches to increasing trace element uptake or increasing trace element absorption were reviewed including genetic modification [18]. A technique called ion genomics has been developed to elucidate all of the genes involved in mineral nutrition in plants [19]. Iron was fortified in rice seeds by modifying the iron in the seeds using soybean ferritin gene expressed in the seed endosperm [20]. Some plant species can utilize zinc from zinc deficient soils, this ability has been studied to find out whether or not it can be transferred food crops suffering from deficiency [21]. A polyhistidine sequence was fused to a rubisco sub-unit by plastid transformation of tobacco. The polyhistidine sequence specifically binds zinc, which accumulates at low zinc levels in the culture medium [22]. The approach may be applied to food crops.

Question #2: In order to assist with the identification of additional safety and nutritional considerations that the assessment of recombinant-DNA plants modified for nutritional or health benefit may warrant, please consider the elements listed below and for each of them provide comments on the need, added value and relevance of addressing such items given the intent of the modification introduced in such recombinant-DNA plants.

a) Estimation of potential exposure distribution patterns - how to go about determining potential exposure distribution patterns in both target and non-target populations of a country and evaluate the safety of such exposure in vulnerable groups.  Techniques are available using population dietary intake data that permit modelling of usual intakes through simulated inclusion of the modified food in exchange for foods reported to be used the dietary survey.  In this regard, lessons can be learned from modeling of potential intakes resulting from vitamin and mineral addition to foods;

Answer: Modeling is appropriate but there is no substitute for controlled trials and investigations, first in animals, and then in consenting, informed humans. It is essential that the GM foods be clearly labelled in the marketplace to provide a means of identifying the GM foods in epidemiological studies as part of post-release monitoring and risk management.

b) Bioavailability - when bioavailability testing would be advisable and some considerations of the techniques available for determining bioavailability of various types of substances;

Answer: Bioavailability should be incorporated into regulatory reviews of  all of the modified plants purporting to enhance nutrition or health. Bioavailability can be studied using radioactively labeled tracers of the compounds or elements being studied. Gut cell cultures have been employed in such studies; approaches to studying bioavailability of nutrients have been reviewed [23].

c) Upper limits of safe intake - the need to determine upper limits of safe intake for the nutrient or bioactive substance, if they are not already defined, and how to assess the level of exposure according to population sub-group of the substance or substances in question against those upper limits;

Answer: Upper limits of elements such as iron are essential because  iron overload in males leads to a condition called hemochromatosis, resulting in liver dysfunction and cancer. Vitamin A toxicity is linked to birth defects and toxic side effects in adults. Genetic modification to provide for deficiency in some geographic areas may therefore create toxic side effects in areas where the diet for the nutrients is adequate. The supplemented crops must be clearly identified and efforts taken to prevent their use without informed consent.

Upper limits for novel supplements such as human milk protein in maize or pig lactalbumin in maize clearly require full testing. Full tests for immunogenicity and toxicity must be carried out on all novel proteins, as stated above.

Safe upper limits of ingestion should be established using pure nutrients or enzymes and the final foodstuffs, first in animals then in human volunteers.

d) Stability - what is involved in stability testing, and why that is a component of risk assessment;

Answer: Stability of primary nutrients such as vitamins and minerals are well established, however the stability of novel proteins such as the enzymes introduced into the modified crop should be undertaken because the novel products may create unexpected toxic by-products. The behavior of these products during food processing and storage must be studied, and altered products should be subject to toxicity testing.

e) Risk/benefit consideration - what consideration should be given to the benefit intended to be provided by the modified food in drawing conclusions from the risk assessment.  Benefits may accrue to certain target groups while at the same time, health risks may be a concern for others, but also there may be benefits at lower intakes and adverse effects at higher intakes.  Additionally, how strong the evidence is for the benefits compared to potential adverse effects may need to be assessed in these circumstances;

Answer: The precautionary principle must prevail in every case. Consideration must be given as to whether the benefit really exists, and whether cheaper, safer alternatives exist, given the known risks of genetic modification.

f) Animal feeding studies - when should animal feeding studies be considered and what types of studies might be useful depending on the question;

Answer: Animal studies should be undertaken with every modified crop or in crosses between modified crops and  crops bearing  genes for nutrients or health products that have been enhanced using conventional breeding or marker assisted breeding. It has become common practice to used crude measures of size to evaluate the outcome of feeding experiments. That is not acceptable. Full tissue and organ necropsy is required in every case to detect cell damage and to identify pre-cancerous lesions. In addition, immunogenic and toxicity tests must also be carried out.

g) Study Design - the design and conduct of studies to obtain reliable, repeatable data on composition of the modified food with respect to the intended and potential unintended changes, at the appropriate stages of crop and food production.  In particular, whereas in evaluating unintended effects comparative data in the raw product is usually quite acceptable, where an intended change has been introduced, there will need to be data on the variation in tissue concentration relevant to the parts of the plant that will be used for food production and data to show that the content of the substance remains stable with time, processing and storage.  The impact of factors known to affect crop composition from year to year and by geographical location, soil type and fertility, etc. may need special attention in the case of intended changes.

Answer: Common stress factors in crops such as water deprivation or water logging, nitrogen deprivation or over use and temperature stress all contribute markedly to nutritional value and health related products. We understand that Codex has pointed to the need for evaluation of stress in the approval of GM crops. Codex should work to formalize the most significant stressors and to insure that those significant stressors are evaluated correctly in the approval of GM crops slated for the global market. It is certainly clear that GM crops approved under optimum environmental conditions cannot presume to be substantially equivalent to GM crops produced under conditions of extreme stress.

h) Any other considerations? Please specify.

Answer: Two major flaws are present in the regulation of GM crops in North America. The first of these is secretive field tests of GM crops prior to their commercialization. Bystanders and residents near the test areas are provided little or no information as to the nature of the crops being tested. These individuals are exposed to pollen, plant debris from broken and decayed plant material in dust and from transgenic products in surface and ground water. The other major flaw is that the GM food products are not labelled in the marketplace and those suffering ill effects from consuming the novel materials have no way of knowing what has injured them. Proper epidemiology of human exposures is impossible without labelling. GM nutrition and health products must be labelled and details on field tests must be revealed to the public in full.

*The Independent Science Panel, launched 10 May 2003 at a public conference in London, UK, consists of dozens of prominent scientists from 11 countries spanning the disciplines of agroecology, agronomy, biomathematics, botany, chemical medicine, ecology, epidemiology, histopathology, microbial ecology, molecular genetics, nutritional biochemistry, physiology, toxicology and virology (

Article first published 19/12/05


  1. Prescott VE, Campbell PM, Moore A, Mattes J, Rothenberg ME, Foster PS, Higgins TJ and  Hogan SP. Transgenic expression of bean alpha-amylase inhibitor in peas results in altered structure and immunogenicity. J Agric Food Chem. 2005 Nov 16, 53(23):9023-9030.
  2. Ho MW. Transgenic pea that made mice ill. ISIS/ISP report, 27 November 2005.
  3. Cummins J. Beware Monsanto’s Vistive soybeans. I-SIS Press Release  2004; also Science in Society 2005, 25, 5.
  4. Siritunga,D,Arias-Garzon,D,White,W. and Sayre,T. Over-expression of hydroxynitrile lyase in transgenic cassava roots accelerates cyanogenesis and food detoxification. Plant Biotechnology Journal 2004, 2, 37-43.
  5. Lamphear B, Barker D, Brooks C, Delaney D, Lane J, Beifuss K, Love R, Thompson K, Mayor J, Clough R, Harkey R, Poage N, Drees C, Horn M, Streatfield S, Nikolov Z, Woodard S, Hood E,  Jilka J and Howard J. Expression of the sweet protein brazzein in maize for production of a new commercial sweetener. Plant Biotechnology Journal 2005, 3, 103–114.
  6. Yang SH, Moran DL, Jia HW, Bicar EH, Lee M and  Scott MP. Expression of a synthetic porcine alpha-lactalbumin gene in the kernels of transgenic maize. Transgenic Res. 2002, 11(1), 11-20.
  7.  Arakawa T, Chong DK, Slattery CW and  Langridge WH. Improvements in human health through production of human milk proteins in transgenic food plants. Adv Exp Med Biol. 1999, 464, 149-59.
  8. Harker M, Holmberg N, Clayton J, Gibbard C, Wallace A, Rawlins S, Hellyer S, Lanot A and Safford R. Enhancement of seed phytosterol levels by expression of an N-terminal truncated Hevea brasiliensis (rubber tree) 3-hydroxy-3-methylglutaryl-CoA reductase. Plant Biotechnology Journal 2003, 1, 113–121.
  9. Ursin VM. Modification of plant lipids for human health: development of functional land-based omega-3 fatty acids. J Nutr. 2003,133(12), 4271-4.
  10. Qi B, Fraser T, Mugford S, Dobson G, Sayanova O, Butler J, Napier JA, Stobart AK and Lazarus CM.  Production of very long chain polyunsaturated omega-3 and omega-6 fatty acids in plants. Nat Biotechnol. 2004, 22(6):739-45.
  11. Galili G and Hofgen R. Metabolic engineering of amino acids and storage proteins in plants. Metab Eng. 2002, 4(1), 3-11.
  12. KimW and Krishnan HB. Expression of an 11 kDa methionine-rich delta-zein in transgenic soybean results in the formation of two types of novel protein bodies in transitional cells situated between the vascular tissue and storage parenchyma cells. Plant Biotechnology Journal 2004, 2 199–210,
  13. Lucas D. Monsanto Petition for non-regulated status  for lysine maize LY38  Monsanto Petition #04-CR-114U 2004
  14. Giovinazzo G,  D'Amico L, Paradiso A, Bollini R, Sparvoli F and DeGara L. Antioxidant metabolite profiles in tomato fruit constitutively expressing the grapevine stilbene synthase gene. Plant Biotechnology  2005, 3, 57-69.
  15. Dormann P. Corn with enhanced antioxidant potential. Nature Biotechnology 2003, 21,1015-16.
  16. Datta K, Baisakh N, Oliva N, Torrizo L, Abrigo E, Tan J, Rai M, Rehana S,  Al-Babili S,  Beyer P, Potrykus I and Datta S. Bioengineered 'golden' indica rice cultivars with β-carotene metabolism in the endosperm with hygromycin and mannose selection systems Plant Biotechnology 2003, 1,81-90.
  17. Weyens G, Ritsema T, Van Dun K, Meyer D, Lommel M, Lathouwers J,  Rosquin I,   Denys P,  Tossens A,  Nijs M, Turk S, Gerrits N,  Bink S, Walraven B,  Lefèbvre M and Smeekens S. Production of tailor-made fructans in sugar beet by expression of onion fructosyltransferase genes  Plant Biotechnology Journal 2004,  2, 321-327.
  18. Lonnerdal B. Genetically modified plants to improve trace element nutrition.  J. Nutr. 2003, 133,1490S-1493S.
  19. Rea P. Ion genomics. Nature Biotechnology 2003, 21,1149-50.
  20. Goto F, Yoshihara T, Shigemoto N, Toki S and Takaiwa F. Iron fortification of rice seed by the soybean ferritin gene. Nat Biotechnol. 1999, 17(3), 282-6.
  21. Hacisalihoglu G and Kochian L. How do some plants tolerate low levels of soil zinc? Mechanisms of zinc efficiency in crop plants. New Phytologist 2003, 159, 341-50.
  22. Rumeau D, Bécuwe-Linka N, Beyly A, Carrier P, Cuiné S, Genty B, Medgyesy P, Horvath E and Peltier G. Increased zinc content in transplastomic tobacco plants expressing a polyhistidine-tagged Rubisco large subunit.  Plant Biotechnology Journal 2004 2, 389-99.
  23. Wood RJ and Tamura T. Methodological issues in assessing bioavailability of nutrients and other bioactive substances in dietary supplements: summary of workshop discussion. J Nutr. 2001, 131(4 Suppl),1396S-8S.

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a cat?