Science in Society Archive

GM Food & Feed Not Fit for "Man or Beast"

Dr. Mae-Wan Ho and Prof. Joe Cummins review some of the scientific evidence behind a series of recent scandals involving the safety of GM food and feed. They expose fatal flaws in the regulatory process and highlight how Europe is in danger of approving GM varieties that are genetically unstable and hence illegal as well as unsafe. They demand a full enquiry into the abuse of science that has allowed GM crops not fit for human or animal consumption to enter our food chain.

Based on a paper presented at an ISP Briefing to Parliament, House of Commons, 29 April 2004.

Latest incidents to cast doubt on the safety of GM food

The European Food Safety Authority (EFSA) has given Monsanto's GM maize Mon863, containing the biopesticide Cry3Bb1 against the corn rootworm, a positive assessment. However, French newspaper Le Monde [1] has seen secret documents revealing health impacts of the GM maize, described as "very disturbing" by scientists of the French commission for genetic engineering (CBG), including kidney malformations and increases in white blood cells in male rats and high blood sugar and reduced immature red blood cells in female rats.

Last year, up to 100 villagers in the south of the Philippines living near GM maize plots suffered debilitating illnesses when the GM maize came into flower [2]. Prof. Terje Traavik of the Norwegian Institute of Gene Ecology in Tromsø found antibodies to Cry1Ab produced by the GM maize against the corn borer in the blood of 39 villagers [3]. The maize variety was Dekalb 818 YG, a hybrid between Monsanto's Mon 810 and a locally adapted variety (Dekalb 818). Report has come in of the same illnesses recurring this year [4].

Bt toxins known to be harmful

The Cry proteins, dozens of them, are also called Bt toxins because they are produced by different strains of the soil bacterium Bacillus thuringiensis [5, 6]. Reports in the scientific literature have documented that bacterial spores of B. thuringiensis, containing a mixture of different toxins, can cause allergic reactions in farm workers; that some toxins are immunogenic in animals, Cry1Ac in particular, has been identified as a potent immunogen, as potent as cholera toxin; that cells in the lining of the small intestine in rats have proteins that bind to the toxins [7], and further, Cry1Ab protein is 92% indigestible in pigs [8].

Regulatory sham over Bt crops

The findings on Bt toxins have been completely ignored in a regulatory process that can only be described as a sham [5].

Worse still, Bt genes in crops are synthetic or hybrid constructions, with important changes from the naturally occurring bacterial genes. Yet, toxicity tests are routinely done using the natural toxins, and not the toxin produced in the GM crop plants, with the result that the Bt toxins in GM crops are almost completely unknown and untested for toxicity [5, 6].

There’s evidence that the natural toxin is not the same as, or “substantially equivalent” to, the GM toxin. Green lacewings suffer significantly reduced survival and delayed development when fed an insect pest (lepidopteran) that has eaten GM maize containing the Bt toxin Cry1Ab, but not when fed the same pest treated with much higher levels of the natural toxin [9, 10]. This is an extremely important effect passed on through the food chain; and has been documented in several laboratories. Unfortunately, the researchers misrepresented the results to mean that Cry1Ab does not harm beneficial insect predators [11].

All GM genes differ from natural genes

All foreign genes inserted into GM organisms are different from their natural counterparts. The minimum construct consists of a promoter, a gene-switch that says to the cell, "copy the following message (the gene or coding sequence) for making a protein", and another signal, the terminator, to say, "stop here, end of message". All three parts are often from different sources. The gene itself could also be a composite of different DNA, often made artificially in the laboratory [12].

It is generally not easy to get the foreign gene to work, so a very aggressive promoter is needed, literally to force the cell to make the protein. The cauliflower mosaic virus (CaMV) 35S promoter is the most popular one used, and is often accompanied by other 'boosters' from a variety of sources.

For example, Mon 863 maize is described on the AGBIOS Database as follows [13]:

"The introduced DNA contained the modified cry3Bb1 gene from B. thuringiensis subsp. kumamotoensis under the control of the 4-AS1 promoter (CaMV 35S promoter with 4 repeats of an activating sequence), plus the 5' untranslated leader sequence of the wheat chlorophyll a/b binding protein (wt CAB leader) and the rice actin intron. The transcription termination sequence was provided from the 3' untranslated region of the wheat 17.3 kD heat shock protein (tahsp17). The modified cry3Bb1 gene encodes a protein of 653 amino acids whose amino acid sequence differs from that of the wild-type protein by the addition of an alanine residue at position 2 and by seven amino acid changes."

There are thus 9 bits of DNA from different sources including the coding sequence, which has been quite substantially altered from the natural gene.

The GM process is unreliable and uncontrollable

That's not all. The artificial constructs are further spliced into gene carriers or vectors, and introduced into cells by invasive methods that result in random integration into the genome, giving rise to unpredictable, random effects, including gross abnormalities in animals and further unexpected toxins and allergens in food crops [14].

A transgenic line is essentially regenerated from a single cell in which specific GM DNA integration occurred. Each event will give rise to a different line. In other words, there is no possibility for quality control. This problem is compounded by the overwhelming instability of transgenic lines, because the artificial constructs cobbled together from DNA of different sources tend to have weak joints, especially if they include elements like the CaMV 35S promoter, which is known to have a fragmentation or recombination hotspot (see later).

Transgenic lines are overwhelmingly unstable
We have referred to the instability of transgenic lines as the "best kept open secret", because everybody has known about it for years, but agree to say nothing, while regulators turned a blind eye [15].

(Claims of genetic stability based on the failure to depart from Mendelian ratios have been widely accepted as evidence of Mendelian inheritance, i.e., a sign of genetic stability. But such claims are bogus for a number of reasons. First, a 'Mendelian ratio' refers to the proportion of different classes of offspring predicted from a cross involving different lines. It depends on assuming that Mendelian inheritance is true; so in order to depart from a particular ratio, a sufficiently large number of offspring are needed to obtain the required level of significance (at 5%). Consequently, a failure to depart from the predicted Mendelian ratio does not prove Mendelian inheritance. On the contrary, the real inheritance may be non-Mendelian (a sign of genetic instability), but an insufficient number of offspring has been produced for the statistical test to reach the required level of significance.

More importantly, the precise Mendelian ratio to use in each case depends on the genotype of the parents, and this needs to be independently ascertained, but is almost never done. This makes nonsense of the predicted ratio. Indeed, the Mendelian ratio used is always the one that most closely matches the result obtained!

One of us had argued this very point at a public hearing on T25 maize in the UK, and got the representative from the company Aventis to concede that Mendelian ratios are not evidence of stability [16].)

Instead, we have been pressing, both in international biosafety conferences and in print, for "event specific" molecular characterisation of the structure of the insert(s) and their position(s) in the genome in successive generations, as the only legitimate proof that the transgenic line is stable [14, 15]. This requirement was finally written into the 2001 European Directive (2001/18/EC) on the deliberate release of GMOs into the environment.

But it was not until last year that French government scientists checked the transgenic inserts of five transgenic lines: Monsanto's Mon810 maize, Roundup Ready soya, GA21 maize, Bayer's T25 maize and Syngenta's Bt 176 maize; and in every case, the transgenic insert(s) had rearranged, not just from the construct used, but since characterised by the company [17].

The results revealed that,

  • All GM inserts had rearranged from the structure provided by the company
  • Many of the breakpoints for rearrangement involve the CaMV 35S promoter, as can be predicted from its known recombination hotspot
  • Scrambling of the genome occurred at the site of insertion
  • GM inserts appear to show a preference for mobile genetic elements (retrotransposons)

The last feature is particularly important, as retrotransposons contain strong promoters that could alter gene expression, and also increase the chances that the inserts will move again, resulting in further genome scrambling and horizontal gene transfer.

The French scientists presented their results in a poster at a conference with the title: “Characterisation of commercial GMO inserts: a source of useful material to study genome fluidity”. Genome fluidity underlies the paradigm shift in genetics that makes genetic modification both futile and hazardous [18].

Belgian government scientists carried out another study, confirming the instability of the transgenic lines analysed by the French, and found that at least one other transgenic line, Syngenta’s Bt 11 maize, had also rearranged, and that it was contaminated with Bt176 [19].

In the case of other transgenic lines studied, it was unclear whether the company has been allowed to submit new data since its first application for approval, which would be irregular, to say the least.

For Roundup Ready soya GTS 40-3-2, for example, the French study found clear evidence that the GM insert was unstable and had undergone rearrangement. The Belgian study merely referred to the UK’s Advisory Committee for Novel Foods and Processes (ACNFP) website, where it appears that the ACNFP had allowed Monsanto to submit new data in 2000, and again in 2002, presumably to ‘correct’ its ‘error’ in the original dossier.

Transgenic instability is a key safety issue

There were small and large discrepancies between the French and Belgian studies, which suggest that the transgenic lines were not only unstable but also non-uniform. Either one of those should make the transgenic lines illegal for Europe. There is every sign, however, that the European Commission will fudge this to lift the de facto moratorium, which will be a criminal offence in our opinion, as it will subject all European citizens to serious health risks.

Transgenic instability is a key safety issue. A GM variety that has changed its identity since characterised by the company, invalidates any safety tests or assessments that may have been done. It also makes it impossible to identify the GM variety for post-release monitoring, for implementing remedial action in case of harm and for assigning liability

Event specific characterisation of the GM inserts has only just begun. It is not clear how many of the GM varieties currently pending approval in Europe have been analysed (see Box 1).

It is also not legitimate to draw conclusions about the hybrids from data on parental GM lines. We have pointed out [20], for example, in the case of NK603xMon810, that both parental lines have rearranged, but no analyses were carried out on the hybrid and seeds set by the hybrid, where further recombinations are expected between the constructs, as they possess similar sequences that are recombination hotspots (see later): CaMV 35S promoter with enhancer (e35S) and the hsp70 intron.

There can be no approval of any GM variety or hybrid for import, either for growing or for food and processing unless and until event-specific analysis has been carried out and the GM variety/hybrid proven to be stable.

Some GMOs pending approval in Europe*

Bt11sweet corninsect resistance Draft decision to authorise**
NK603maizeglyphosate toleranceEFSA favourable opinion***
GT73 oilseed rapeglyphosate toleranceEFSA favourable opinion
Mon863maize insect resistance EFSA favourable opinion
Mon863xMon810hybrid maizeinsect & glyph. res.No decision from EFSA
Ms8xRf3oilseed rapeglufosinate res. Belgian approval (but denied
for cultivation)
LLRice62riceglufosinate tolerancePositive assess. UK ACRE
Bt Cry1F(1507) maizeinsect & glufo. res.Positive assess. Netherlands
NK603xMon810maizeglyphosate tolerance Consent from UK

* For import and/or use as food and/or feed and/or processing, not for growing.
**Ministers of European countries failed to reach agreement on Bt11 for food use, which is closest to final approval; the European Commission will now have to decide.
***NK603 was rejected for animal feed and food use by EU member states; the dossier now goes to the European Council of Ministers.

Major uncertainties over the safety of the GM process

Let us look at the rest of the evidence in brief; apart from the two incidents mentioned.

  • Between 2001 and 2002, twelve dairy cows died on a farm in Hesse, Germany, after eating Syngenta’s Bt176 GM maize, and others in the herd had to be slaughtered on account of mysterious illnesses [21]. To-date, there has been no detailed autopsy reports available, even though the company claims the deaths and illnesses were unrelated to Bt176. Nevertheless the Spanish Food Safety Authority has just withdrawn authorisation for Bt176 cultivation in Spain [22] after it had occupied almost all of the 20 000 hectares of GM maize grown in Spain since 1998 [23]. The decision was taken following an EFSA recommendation that GMOs containing antibiotic resistance marker genes such as that found in Bt 176, be restricted to field trials.
  • Arpad Pusztai and colleagues found that GM potatoes with snowdrop lectin adversely affected every organ system of young rats, and the stomach and small intestine lining grew up to twice the thickness of controls [24].
  • Scientists in Egypt found similar results in the gastrointestinal tract of mice fed GM potato with Bt toxin [25].
  • US Food and Drug Administration had data since the early 1990s showing that rats fed GM tomatoes with antisense gene to delay ripening developed small holes in their stomach [24].
  • Aventis (now Bayer) found 100% increase in deaths of broiler chickens fed glufosinate-tolerant GM maize T25 compared to controls [26].
  • Numerous anecdotes from farmers and others indicating that livestock, wildlife and lab animals avoid GM feed, and fail to thrive or die when forced to eat it [26, 27].

Different species of GM food or feed with different GM genes have caused problems in many species of animals. You don't have to be a scientific genius to suspect that there is something wrong with the GM process itself or the GM insert.
All of the GM inserts involved contain the CaMV35S promoter that we have warned against since 1999 [28-31]. This promoter not only has a fragmentation hotspot making transgenic lines extra unstable, it substitutes for the promoter of a wide range of plant and animal viruses, and is also active in animal cells including human cells.

It is high time we ban all environmental releases of GM crops to make way for non-GM sustainable agriculture [32].

The greatest obstacle to a safe and sustainable future is a corrupt and corrupted science that operates on what can only be described as the anti-precautionary principle. There must now be a thorough enquiry into the safety of GM food and feed, and the systematic abuse of science that has allowed GM food and feed to be approved, which had all the signs of being unsafe.

Article first published 07/05/04


  1. "French experts very disturbed by health effects of Monsanto GM corn" GMWatch 23 April 2004
  2. "Filipino islanders blame GM crop for mystery sickness. Monsanto denies scientist's claim that maize may have caused 100 villagers to fall ill" John Aglionby in Kalyong, southern Philippines, The Guardian, Wednesday 3 March 3, 2004,2763,1160789,00.html
  3. Traavik, T. Lecture to Special Biosafety Genok and TWN Seminar, 22 February, Kuala Lumpur, and personal communication.
  4. "Despite ban, agriculturists can't stop farmers from planting Bt corn", Allen Estabillo, Minda News 23 April 2004
  5. Cummins J. Regulatory sham over Bt-crops. ISIS report 1 December 2003; also Science in Society 2004, 21, 30.
  6. Cummins J. Bt toxins in genetically modified crops: regulation by deceit. Science in Society 2004, 22 (in press).
  7. Vázquez-Padrón RI, Gonzáles-Cabrera J, Garcia-Tovar C, Neri-Bazan L, Lopéz-Revilla R, Hernández M, Moreno-Fierro L and de la Riva GA. CrylAc protoxin from Bacillus thringiensis sp. kurstaki HD73 binds to surface proteins in the mouse small intestine. Biochem Biophys Res Commun 2000, 271, 54-8; Ho MW. Bt toxin binds to mouse intestine. Science in Society 2004, 21, 7.
  8. Ho MW. Transgenic DNA & Bt toxin survive digestion. Science in Society 2004, 21, 11; Chowdhury EH, Kuribara H, Hino A, Sultana P, Mikami O, Shimada N, Guruge KS, Saito M, Nakajima Y. Detection of corn intrinsic and recombinant DNA fragments and CrylAb protein in the gastrointestinal contents of pigs fed genetically modified corn Bt11. J Anim Sci 2003, 81, 2546-51.
  9. Dutton A, Klein H, Romeis J and Bigler F. "Uptake of Bt-toxin by herbivores feeding on transgenic maize and consequences for the predator Chrysoperia carnea", Ecological Entomology 2002, 27, 441-7.
  10. Romeis J, Dutton A and Bigler F. "Bacillus thuringiensis toxin (Cry1Ab) has no direct effect on larvae of the green lacewing Chrysoperla carnea (Stephens) (Neuroptera: Chrysopidae)", Journal of Insect Physiology 2004, in press.
  11. Dutton A, Romeis J and Bigler F. "Assessing the risks of insect resistant transgenic plants on entomophagous arthropods: Bt-maize expressing Cry1Ab as a case study", BioControl 2003, 48, 611"36.
  12. Ho MW. FAQs on genetic engineering. ISIS tutorial
  14. Ho MW. Genetic Engineering Dream or Nightmare? TWN, Gateway, Gill & Macmillan, Continuum, 1998, 2nd ed. 1999, re-issued 2003 in cd
  15. Ho MW. The best kept secret of GM crops. Science in Society 2002, 15, 9.
  16. Ho MW. GM maize approve on bad science in the UK. Science in Society 2002, 15. 10-25.
  17. Collonier C, Berthier G, Boyer F, Duplan M-N, Fernandez S, Kebdani N, Kobilinsky A, Romanuk M, Bertheau Y. Characterization of commercial GMO inserts: a source of useful material to study genome fluidity. Poster presented at ICPMB: International Congress for Plant Molecular Biology (n°VII), Barcelona, 23-28th June 2003. Poster courtesy of Dr. Gilles-Eric Seralini, Président du Conseil Scientifique du CRII-GEN,; also "Transgenic lines proven unstable" by Mae-Wan Ho, ISIS Report, 23 October 2003
  18. Ho MW. Living with the Fluid Genome. TWN & ISIS, 2003.
  19. Ho MW. Unstable transgenic lines illegal. ISIS Report 3 December 2003; also Science in Society 2004, 21, 23.
  20. Ho MW and Cummins J. Comment on Assessment Report C, submitted to UK ACRE and European Food Safety Authority 6 April 2004 on behalf of ISIS and ISP
  21. Ho MW and Burcher S. Cows ate GM maize and died. Science in Society 2004, 21, 4-6.
  22. El Estado espanol retirara un OGM a instancias de la UE. El maiz Bt 176 podria provoca resistencias a los antibioticos, GARA, Spain
  23. Ho MW. Syngenta's Spanish Trojan horse. Science in Society 2004, 21, 8.
  24. Pusztai A, Bardocz S and Ewen SWB. Genetically modified foods: Potential human health effects. In Food Safety: Contaminants and Toxins, (J P F D'Mello ed.),
  25. Scottish Agricultural College, Edinburgh, CAB International, 2003.
  26. Fares NH and El-Sayed AK. Fine structural changes in the ileum of mice fed on dendotoxin-treated potatotes and transgenic potatoes. Natural Toxins, 1998, 6, 219-33; also "Bt is toxic" by Joe Cummins and Mae-Wan Ho, ISIS News 7/8, February 2001, ISSN: 1474-1547 (print), ISSN: 1474-1814 (online)
  27. Novotny E. Animals avoid GM food, for good reasons. Science in Society 2004, 21, 9-11.
  28. Ho MW. Mice prefer non-GM. Science in Society 2002, 13/14, 24.
  29. Ho MW, Ryan A and Cummins J. Cauliflower mosaic viral promoter - a recipe for Disaster? Microbial Ecology in Health and Disease 1999 11, 194-7.
  30. Cummins J, Ho MW and Ryan A. Hazardous CaMV promoter? Nature Biotechnology 2000, 18, 363.
  31. Ho MW, Ryan A and Cummins J. Hazards of transgenic plants with the cauliflower mosaic viral promoter. Microbial Ecology in Health and Disease 2000, 12, 6-11.
  32. Ho MW, Ryan A and Cummins J. CaMV35S promoter fragmentation hotspot confirmed and it is active in animals. Microbial Ecology in Health and Disease 2000, 12, 189.
  33. Ho MW, Lim LC et al. The Case for a GM-Free Sustainable World. Independent Science Panel Report, ISIS & TWN, 2003

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 tripod?