Science in Society Archive

GM Cotton that People Forgot

GM cotton has aroused relatively little resistance outside the Third World for the simple reason that it is wrongly perceived to be a non-food crop. Prof. Joe Cummins and Dr. Mae-Wan Ho report

GM cotton a triple-threat

Cotton is a triple-treat (or threat) crop because it produces fibre, food and feed. Fibre is recovered from the flower bolls, while the seeds are pressed to yield oil for the kitchen and cake for animal feed. Monsanto Corporation has been a major source of genetically modified (GM) cotton lines.

Bollgard cotton

A line called Bollgard was first marketed in the United States in 1995, followed in later years by Canada, Australia, China, Argentina, Japan, Mexico, South Africa, India and the Philippines. In 2002, an enhanced line called Bollgard II was approved in the United States, Canada, Australia, Japan and the Philippines.

Bollgard II was made from Bollgard simply by inserting into the plant cells a gene cassette containing a Bacillus thuringiensis (Bt) toxin different from the one in the original Bollgard. From the transformed cells, a line containing two different Bt toxin genes were selected. Two toxin genes were more than twice as effective than the original Bollgard and theoretically, far less likely to allow insect resistant mutants to evolve.

The DNA segment used for transformation - propagated on the plasmid vector PV-GHBK11 vector - contains the cry2Ab and uidA gene cassettes. The cry2Ab gene cassette consists of the enhanced 35S promoter from the cauliflower mosaic virus (CaMV), the leader sequence from petunia heat shock 70 gene, the synthetic cry2Ab coding region based on the sequence of the cry2Ab gene from B thuringiensis (Bt) subsp. kurstaki, and the terminator of the nos gene from Agrobacterium tumefaciens that provides the transcription termination signal. The uidA gene cassette contains the E. coli uidA gene flanked by the same regulatory sequences as the cry2Ab gene in the cry2Ab gene cassette. The segment was cut out with a restriction enzyme, purified, precipitated onto gold particles, and introduced into the meristems of the recipient variety DP50B with a gene gun. The transformed tissue was identified and selected by histochemical staining that allows visual detection of the uidA gene product, the GUS protein. Based on this screening, the transformant, referred to as the cotton event 15985 was selected for commercial development into Bolgard II. Bollgard II has as its principal gene a synthetic approximation of a bacterial toxin gene because the original bacterial gene does not function well in the plant [1-5]. As the synthetic gene has a unique DNA sequence, it has not been subject to evolution and its recombinational and other properties relevant to safety are unknown and untested.

The original Bollgard (BollgardTM also called IngardTM in Australia) comprises three lepidopteran- resistant cotton lines 531, 757, and 1076 expressing the cry1Ac gene coding for Cry1Ac toxin protein from the soil bacterium Bacillus thuringiensis var. kurstaki that confers resistance to lepidopteran-insects in general, and cotton bollworm, tobacco budworm, and pink bollworm, in particular. Upon ingestion of this protein by susceptible insects, feeding is inhibited, eventually resulting in death. The protein-coding region of the gene was modified for optimal expression in plants. To express the gene, this region is fused to the promoter derived from the 35S gene of cauliflower mosaic virus (CaMV) with a duplicated enhancer region and to the terminator of the soybean alpha subunit of the beta-conglycinin gene. The lines also express the nptII gene from Escherichia coli coding for the neomycin phosphotransferase enzyme, which confers resistance to the antibiotic kanamycin [6]. The three lines differ in their insertion site of the transgenes on the cotton chromosomes. Line (event) 531 was selected as the parent for Bolgard II.

Thus Bolgard II has two separate transgene insertions with some regions of DNA homology. Such regions could act as recombination signals for somatic or meiotic recombination, leading to drastic chromosome rearrangements. The claim to genetic stability reported in the governmental reviews is simply the finding that the insertions segregate according to Mendelian ratios in a few crosses and does not consider molecular and chromosomal instability associated with inter- and intra-chromosomal recombination at sites of DNA homology. Signs of instability and other failures have been observed in the field (see "Australia adopts GM cotton" and "GM cotton fiascos around the world" this series).

Seed distribution is controlled by the licenses of the patentee, and seed lines can, and should be screened at that point for translations, duplications or deficiencies resulting from intra- and inter chromosomal recombination.

Furthermore, in evaluating safety to humans and the environment, the toxin proteins are frequently isolated from liquid culture of the bacteria to avoid having to carry out the more expensive isolation of the toxins from cotton plants. As the toxin transgenes are synthetic approximations of the natural genes and the toxin proteins are not identical, the test results with bacterial proteins do not truly represent the impact of the toxins from the transgenic cotton plants.

Some feeding studies indicated Bollgard II controlled insect pests more effectively [7]. One research group predicted that the need for supplemental insecticides would be reduced or eliminated for lepidopteran pests [8]. Another research group indicated, however, that insect-resistance to Bollgard II could best be controlled with an overspray of chemical insecticide [9]. Further studies showed that resistance to the two Cry toxins seemed to evolve simultaneously, raising considerable doubt over the efficacy of gene stacking in delaying insect resistance [10]. Studies reported by researchers from Monsanto Corporation showed that the Cry1Ac toxin and the Cry2Ab toxin were produced in equivalent amounts in Bollgard II, but that Cry2Ab was the larger contributor to insect toxicity, and they suggested a relatively simple resistance monitoring policy [11]. It seems likely that chemical pesticides will be needed to combat insect resistance arising in Bololgard II after all.

The regulation of Bollgard II has been 'fast and loose'. Bollgard II was supposed to address the major concern of resistance management, but research is already indicating that gene stacking is not a panacea and that chemical pesticide overspray will be required to cope with developing resistance.

Round up Ready Cotton

Roundup Ready cotton, like Bollgard I and II, is also used for fibre, food and feed. Roundup Ready (rr cotton) was first marketed in the United States in 1995, and in later years, in Canada, Japan, Argentina, South Africa, Australia, the Philippines and in 2004, in China.

The herbicide tolerant cotton marketed as rr cotton was originally derived from two different transformation events of a cotton line called Coker 312. These events, designated 1445 and 1698, differed in both gene sequences inserted and insertion sites in the cotton genome. Currently, event 1445 is the primary rr cotton marketed [12-15].

Event 1445 originated from a plasmid containing a synthetic approximation of the glyphosate oxidase (gox) gene driven by a modified figwort mosaic virus promoter and terminated by the nos terminator tnos from Agrobacterium; followed by a synthetic CP4 epsps gene derived from Agrobacterium strain CP4, encoding the enzyme 5-enolpyruvylshikimate-3-phosphate synthase (conferring resistance to glyphosate), also driven by the figwort mosaic virus promoter. Synthetic approximations of the genes were used because the bacterial genes are not very active in plant cells. The synthetic epsps sequence was preceded by a chloroplast targeting sequence from Arabidopsis, and terminated by a terminator from the pea plant.

In addition, two antibiotic resistance marker genes were present: aad from a bacterial transposon,Tn5, conferring resistance to streptomycin and spectomycin, inserted after the epsps gene cassette; followed by kanamycin resistance gene, also from Tn5 driven by the cauliflower mosaic Virus (CaMV) promoter and terminated with the tnos. In the marketed crop, event 1445 appeared to have lost the gox gene but retained aad, which the company claims, is inactive in the cotton plant. However, the rr cotton failed to gain approval from the European Commission in 1999 on account of serious concerns over the aad antibiotic resistance marker [16, 17]. The fact that it is inactive in cotton plants is irrelevant, because it is surely active in bacteria, to which it could be transferred.

Event 1698 is similar except that it has an additional epsps gene [12-15]. The events were described as being "stably inherited", with no molecular genetic evidence.

Monsanto and the regulators seem to agree that direct human exposure to the transgenes and their products will be very limited because cottonseed oil contains very little protein and DNA. Nevertheless, farm animals consume a great deal of seed cake.

Monsanto's safety assessment of rr cotton dismissed the possibility that the epsps gene and the antibiotic resistance marker genes could participate in horizontal gene transfer with soil bacteria [12]. However, the bacterial marker gene for kanamycin reistance in transgenic sugar beet was observed to readily transform soil Pseudomonas [18] while transgenic potato marker gene readily transformed soil Actinobacter [19] through homologous recombination. In both cases, the marker persisted for long periods in the soil bacteria and such bacteria are capable of exchanging genes with animal pathogens. It is very likely that the streptomycin resistance marker gene will readily transform soil bacteria.

Monsanto's claim, that to effectively transform bacteria the marker genes require co-transformation with a bacterial promoter, is not realistic; operator fusions are commonplace in bacteria suggesting that the marker genes can easily become activated. There are also special mobile genetic elements called integrons containing sites with ready-made promoters for insertion of antibiotic resistance coding sequences so they can be expressed [20].

Glyphosate applications can be used to control weeds prior to flowering, but glyphosate application after initiation of flowering in rr cotton reduced pollen viability and seed set with reduced yield [21, 22], while glyphosate application to rr cotton combined with water stress resulted in abscission of young cotton bolls [23]. Use of rr cotton seems to require irrigation technology and considerable technical savvy.

An additional concern related to use of glyphosate on cotton is that the herbicide has been shown to move from cotton fabric into and through human skin [24].

GM cotton not safe

Regulators seem to have taken a relaxed attitude towards many safety issues including antibiotic resistance markers going into GM crops. The potential toxicities of the synthetic genes, their ability to recombine and stability have yet to be documented. Already, all the transgene products, Cry1Ac, Cry2Ab, CP4 EPSPS, as well as the marker gene product, UidA, show stretches of amino-acid sequence identities to known allergens [25], and are hence suspected allergens; at least, until proven otherwise by further studies.

Article first published 20/01/05


  1. U.S. Food and Drug Administration Center for Food Safety and Applied NutritionOffice of Food Additive Safety Biotechnology Consultation Note to the File BNF No. 000074 2002 Protected Bollgard II Cotton Line 15985 2002
  2. Monsanto Corporation Safety Assessment of Bollgard II Cotton Event 15985 2003
  3. Department of Agriculture Animal and plant Health Inspection Service Monsanto Company Availability of Determination of Non-regulated Status for Cotton Genetically Engineered for Insect Resistance (cotton event 15985) 2003
  4. Food Standards Australia New Zealand Oil and Linters Derived from Insect-Protected Cotton Containing Event 15985 Application A436 Final Assessment Report 2002
  5. Canadian Food Inspection Agency - Plant Products - Plant Biosafety Office - Determination of the Safety Canadian Food Inspection Agency Plant Products Directorate Plant Biosafety Office Decision Document DD2003-45 Determination of the Safety of Monsanto's Insect Resistant Bollgard II Cotton (Gossypium hirsutum L.) 2003
  6. Department of Agriculture Animal and plant Health Inspection Service Availability of Determination of Non-regulated Status for Cotton Genetically Engineered for Insect Resistance (Monsanto Company cotton events 531,757 and 1076) 1995
  7. Stewart S, Adamczyk J, Knighten K.and Davis F. Impact of Bt cottons expressing one or two insecticidal proteins of Bacillus thuringiensis Berliner on growth and survival of noctuid (Lepidoptera) larvae. J Econ Entomol 2001, 94, 752-60.
  8. Chitkowski R, Turnipseed S, Sullivan M. and Bridges W. Field and laboratory evaluations of transgenic cottons expressing one or two Bacillus thuringiensis var. kurstaki Berliner proteins for management of noctuid (Lepidoptera) pests. J Econ Entomol 2003, 96, 755-62.
  9. Jackson R, Bradley J, Van Duyn J. and Gould F. Comparative production of Helicoverpa zea (Lepidoptera: Noctuidae) from transgenic cotton expressing either one or two Bacillus thuringiensis proteins with and without insecticide oversprays. J Econ Entomol 2004, 97,1719-25.
  10. Jurat-Fuentes J, Gould F, and Adang M. Dual resistance to Bacillus thuringiensis Cry1Ac and Cry2Aa toxins in Heliothis virescens suggests multiple mechanisms of resistance Appl Environ Microbiol 2003, 69, 5898-906.
  11. Greenplate J, Mullins J, Penn S, Dahm A, Reich B, Osborn J, Rahn P, Ruschke L and Shappley Z. Partial characterization of cotton plants expressing two toxin proteins from Bacillus thuringiensis: relative toxin contribution, toxin interaction, and resistance management. J. Appl. Ent. 2003, 127, 340-7.
  12. Monsanto Corporation Safety Assessment of Roundup Ready Cotton Event 1445 2002
  13. United states Food and Drug Administration Monsanto's glyphosate tolerant cotton lines 1445 and 1698 1995
  14. Department of Agriculture Animal and Plant health Inspection Service Non-regulated Status for Genetically Engineered Cotton line 1445 and 1698 1995
  15. Australia New Zealand Food Authority Draft Risk analysis Report Food produced from glyphosate-tolerant cotton line 1445 2000
  16. Ho MW. Monsanto's GM cottons & Gonorrhea. I-SIS News 7/8 February 2001
  17. Ho MW. DNA in GM food and feed. Science in Society 2004, 23, 34-36.
  18. Meier,P. and Wackernagel,W. Monitoring the spread of recombinant DNA from field plots with transgenic sugar beet plants by PCR and natural transformation of Pseudomonas stutzeri. Transgenic Research 2003,12,293-304.
  19. de Vries,J,Heine,M,Harms,K. and Wackernage,W. Spread of Recombinant DNA by Roots and Pollen of Transgenic Potato Plants, Identified by Highly Specific Biomonitoring Using Natural Transformation of an Acinetobacter sp. 2003 Appl Environ Microbiol 69,4455-62.
  20. Rowe-Magnos DA, Gueront AM, Ploncard P, Dychinco B, Davies J and Mazel D. The evolutionary history of chromosomal super-intefrons provides an ancestry for multiresistant integrons. PNAS 2001, 98, 652-7.
  21. Pline W, Edmisten K, Oliver T, Wilcut J, Wells R and Allen N. Use of digital image analysis, viability stains, and germination assays to estimate conventional and glyphosate-resistant cotton pollen viability. Crop Science 2002, 42, 2193-2200.
  22. May O, Culpepper A, Cerny R, Coots C, Corkernd C, Cothrene J, Croonb K, Ferreirab K, Hartb J, Hayes R, Huberb S, Martens A, McCloskey W, Oppenhuizen M, Patterson M, Reynolds D, Shappley Z, Subramanik J, Wittene T, York A. and Mullinix B. Transgenic cotton with improved resistance to glyphosate herbicide. Crop Sci. 2004, 44:234-240.
  23. Plinea W, Wells R, Little G, Edmisten K and Wilcut J. Glyphosate and water-stress effects on fruiting and carbohydrates in glyphosate-resistant cotton. Crop Science 2003, 43, 879-85.
  24. Wester R, Quan D and Maibach H. In vitro percutaneous absorption of model compounds glyphosate and malathion from cotton fabric into and through human skin. Food and Chemical Toxicology 1996, 34,731-5.
  25. Kleter GA and Peijnenburg AdACM. Screening of transgenic proteins expressed in transgenic food crops for the presence of short amino acid sequences identical to potential, IgE-binding linear epitopes of allergens. BMC Structural Biology 2002, 2(8)

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?