Science in Society Archive

Scientists Confirm Failures of Bt-Crops

Ineffective against insect pests, harmful to health and biodiversity, yield drag, pest resistance. Dr. Mae-Wan Ho

Farmers were first

Scientific studies from many countries have now backed up what farmers have known for years, that Bt crops – genetically engineered with Bt toxin proteins from the soil bacterium Bacillus thuringiensis targeted at insect pests - often failed to protect against pest attacks, and have other problems as well.

Scientists in India, China and the United States found that the levels of Bt toxin produced by Bt crops vary substantially in different parts of the plant and in the course of the growing season, and are often insufficient to kill the targeted pests. This could lead to greater use of pesticides, and accelerate the evolution of pest resistance to the Bt toxin. Pest resistance to a Bt toxin has indeed arisen in the field in Australia.

The Bt toxins are a family of similar Cry proteins identified by numbers and letters. Each Cry protein differs somewhat in amino acid sequence and targets specific pests.


Scientists at the Central Institute of Cotton Research studied Bt cotton hybrids approved for commercial planting in India: Bollgard-MECH-12, Bollgard-MECH-162, Bollgard-MECH-184, Bollgard-RCH-2, Bollgard-RCH-20, Bollgard-RCH-134, Bollgard-RCH-138 and Bollgard-RCH-144. All the varieties were created by using Indian parent-varieties to which the crylAc gene was introduced from the Bt-cotton variety, Coker 312, ultimately derived from transformation event MON531 (Monsanto). 

The researchers found that the amount of Cry1Ac protein varied across the varieties and between different plant parts. The leaves had the highest levels; whereas the levels in the boll-rind, square bud and ovary of flowers were clearly inadequate to fully protect the fruiting parts producing the cotton bolls. Increasing numbers of armyworm (Helicopverpa armigera) larvae survived as toxin levels went below 1.8 mg /g wet weight of the plant parts. Thus, a critical level of 1.9 mg/g was needed to kill all the pests. Regardless of plant varieties, the level of toxin decreased with the age of the plant, though the decrease was more rapid in some hybrids than in others. By 110 days, Cry1Ac expression decreased to less than 0.47mg/g in all hybrids. 

In a separate study, scientists at the same institute tested the susceptibility of an insect pest from different regions in India to Bt toxin [2].  They took samples of larvae of the spotted bollworm, Earias vitella from 27 sites in 19 cotton-growing districts of North, Central and South India during the 2002 and 2003 cropping seasons and tested their susceptibility to Cry 1Ac toxin protein purified from E. coli strains expressing the recombinant protein. The LC50 - the concentration killing 50 percent of the larvae – of Cry1Ac ranged from 0.006 to 0.105 mg/ml. There was a 17.5 fold overall variability in susceptibility among the districts. The highest variability of 17.5 fold was recorded from districts of South India. The variability in pest susceptibility, like the variable expression of the Cry1A proteins in Bt crops, will reduce the efficacy of Bt pest control.

However, using recombinant CrylA proteins from bacteria to test for susceptibility in pests can be entirely misleading (see below).


A study was carried out in the Institute of Plant Protection, Chinese Academy of Agricultural Sciences in Beijing on two Bt cotton varieties: GK19, with a Cry1Ac/Cry1Ab fused gene, developed by the Biotechnology Research Institute of Chinese Academy of Agricultural Sciences, and BG1560, with a Cry1Ac gene, supplied by Monsanto [3]. The test site was in Tianmen County, Hubei Province, an intensive planting area in the middle of the Yantze River valley. The results showed that the toxin content in the Bt cotton varieties changed significantly over time, depending on the part of the plant, the growth stage and the variety. Generally, the toxin protein was expressed at high levels during the early stages of growth, declined in mid-season, and rebounded late in the season. In line with the study in India, the scientists found that the toxin content in leaf, square, petal and stamens were generally much high than those in the ovule and the boll. The researchers pointed out that such variability in toxin expression could accelerate the development of pest resistance to the toxin.


Scientists at the Southern Insect Management Research Unit of the United States Department of Agriculture (USDA) studied both Bt maize hybrids expressing Cry1Ab (such as event MON810) and Bt cotton varieties expressing Cry1Ac (such as event MON531) [4].

They found that Cry1Ab was variable depending on location in the same leaf as well as between leaves at different stage of growth. The tips of maize leaf at the V7 stage had a higher concentration compared with the middle section of the leaf, and the middle section of the V9 leaf had the lowest concentration. Also, the green tissues richest in chlorophyll had the highest toxin levels, the yellow-green tissues with reduce chlorophyll had less, and the white-yellow tissues poorest in chlorophyll had the least. The weight of fall armyworm larvae measured at day 5 of feeding showed a decrease that was significantly correlated with the amount of toxin present in the plant material, while there was 100 percent mortality in the southwestern corn borer larvae regardless of the level of toxin in the plant tissues.

In the Bt cotton, the level of CrylAc was significantly lower in boll tips where flowers had remained attached, compared with normal boll tips. Boll tips where the flowers remained attached are often the sites at which corn earworms, Helicopverpa zea (Boddie) penetrate Bt cotton bolls. In both Bt maize and Bt cotton, tissues that had low chlorophyll content also had reduced Cry1A proteins.

The US Environment Protection Agency recommends planting a certain percent of crop area with non-Bt varieties to serve as ‘refuge’, in order to ensure that enough susceptible insects are produced to limit the evolution of resistance. An important requirement for the refuge strategy to work effectively is a high level of expression of the toxin, so heterozygous insects (those with one copy of resistance gene) will fail to survive to reproduce. Thus, any reduction from high toxin levels will compromise the refuge strategy and the effectiveness of Cry1A proteins in pest control.

Researchers at the University of Arizona Tucson and the Arizona Cotton Research and Protection Council, Phoenix had found a “surprisingly high” frequency (0.16) of the Cry1Ac resistance gene in field populations of the pink bollworm in Arizona in 1997, which did not appear to increase further as expected in 1998 or 1999 [5]. However, the tests were done with the recombinant Cry1Ac protein produced in the bacterium, Pseudomonas fluroescens, and not from the Bt cotton plant, and could be giving entirely misleading results on the evolution of resistance in the field (“No Bt resistance?” SiS20) [6].

Bt resistance in Australia

A population of the Australian cotton bollworm, Helicoverpa armigera – the most important agricultural pest in Australia as well as China, India and Africa - has developed resistance to Cry1Ac at 275-times the level that would have killed the non-resistant insect [7]. Some 70 percent of the resistant larvae were able to survive on Bt cotton expressing Cry1Ac (Ingard). The resistance is inherited as an autosomal semi-dominant trait (the heterozygote with one copy of the resistance gene is half as resistant as the homozygotes with two copies of the resistance gene).

Bt cotton varieties expressing Cry1Ac (Ingard) have been grown in Australia to control the cotton bollworm since 1996, and a new variety containing both Cry1Ac and Cry2Ab was commercially released in late 2003. Resistance monitoring in Australia and China had suggested that pest susceptibility to Cry1Ac was declining in the field. In 2001, a strain of cotton bollworm was isolated from the survivors in the New South Wales and Queensland monitoring programme that appeared to be resistant to Cry1Ac. The researchers have now confirmed the findings, and attributed the high level of resistance to a 3- to 12-fold over-expression of an enzyme, serine protease, which binds avidly to Cry1Ac toxin, preventing it from acting, and possibly, detoxifying it by breaking it down.

Canadian scientists find yield and economic disadvantage in Bt maize

Researchers at the Eastern Cereal and Oilseed Research Centre, Agriculture and Agri-Food Canada, Ottawa, Ontario, carried out a field experiment over three years to compare commercial corn hybrids with their corresponding Bt-hybrids belonging to the Monsanto and Syngenta [8]. They found that some of the Bt hybrids took 2-3 additional days to reach silking and maturity and produced a similar or up to 12 percent lower grain yields, with 3-5 percent higher grain moisture content at maturity in comparisons with their non-Bt counterparts. Higher grain moisture content increases drying cost. Bt hybrid seeds also have a $25-30 premium per ha.

The economic disadvantages are dwarfed in comparison with impacts on biodiversity and human and animal health that have been known for years, however (see below: also“Bt risks negligible” SiS 2002, 13/14).

Bt maize more woody

It has been known for some time that genetic modification is full of pitfalls, among which are many unintended effects. A paper published in 2001 [9] reported that the content of lignin (woody substances) was high by 33 to 97 percent in the Bt maize varieties tested: Bt11, Bt176 and Mon810. Now, researchers at environmental and agricultural institutes in Leipzig, Aachen and Muncheberg, Germany, and the University of Waterloo in Ontario, Canada, have confirmed increases in lignin in two Bt maize lines, Novelis (event MON00810-6, from Monsanto) and Valmont (event SYN-EV176-9, from Syngenta), compared with their respective isogenic varieties, Nobilis and Prelude, all grown under identical conditions [10]. The increases in lignin are more modest, and are restricted to the stems of the plants: Novelis by 28 percent over Nobilis, and Valmont by 18 percent over Prelude.

Increase in lignin content will impact on the digestibility of the plant for livestock, it also decreases the rate at which the plant material break down, affecting nutrient recycling, the soil microbial community, and soil carbon balance.

Intriguingly, an earlier report has also found increased lignin in Monsanto’s Roundup Ready soya, genetically modified to be tolerant to the herbicide Roundup [11], which caused the stem to split open in hot climate and crop losses of up to 40 percent.

These results suggest that genetic modification per se may be increasing lignin content, perhaps as a response to metabolic stress from the high levels of transgene expression driven by aggressive viral promoters.

Impacts on biodiversity and health

Bt toxins are known to harm beneficial/endangered insect species and soil decomposers [12]:

  • Pollen from Bt-maize was lethal to the larvae of the monarch butterfly.
  • Increased mortality of lacewing larvae fed on artificial diet containing Bt-maize or on corn-borer larvae that had eaten Bt-corn.
  • Bt sprays used to reduce caterpillars in forests led to fewer black-throated blue warbler nests.
  • A parasite of corn-borers, Macrocentris cingulum, was found to be reduced in Bt-cornfields compared with non-Bt corn fields.
  • One preparation of Bt (var. tenebrionis), reported to be specific for Coleoptera, caused significant mortality in domestic bees.
  • Soil-dwelling collembola, Folsomia candida, an important decomposer, suffered significant mortality from transgenic maize with Cry1Ab.
  • Bt not only remains in the soil with Bt-plant debris, it is actively exuded from the plant roots where it binds to soil particles and persists for 180 days or more, so its effects on soil decomposers and other beneficial arthropods may be extensive.

Bt-toxins are actual and potential allergens for human beings. Field workers exposed to Bt spray experienced allergic skin sensitization and induction of IgE and IgG antibodies to the spray [13]. Recombinant Cry1Ac protoxin was found to be a potent mucosal immunogen, as potent as cholera toxin [14]. A Bt strain that caused severe human necrosis (tissue death) killed mice infected through the nose within 8 hours, from clinical toxic-shock syndrome [15]. Both Bt protein and Bt-potato harmed mice in feeding experiments [16]. All Bt-toxins along with many other transgenic proteins exhibit similarities to known allergens and are hence suspected allergens until proven otherwise (“Are transgenic proteins allergenic?” SiS 25) [17-19].

Recently, much publicity has been given to a report from scientists in Portugal published in the house journal of the American Academy of Allergy, Asthma and Immunology, because it claimed “lack of allergenicity of transgenic maize and soya samples” [20].

A careful reading of the report reveals, however, that the researchers had no evidence that the small number of subjects they tested have ever been exposed to transgenic maize and soya. They wrote:  “Bearing in mind that since 1998 all the GM products under testing were approved for commercialisation in the European Union.., we assumed that consumption of maize and soya food-derived products implied a consumption of GM soya and maize.” (emphasis added). Moreover, the tests performed were limited to skin pricks and IgE antibodies, both known to be limited in reliability [21]. Most of all, there are many allergies that do not involve IgE antibodies [22].

Nevertheless, the researchers stated, “In this study we did not obtain any differential positive results, which allows us to conclude that the transgenic products under testing seem to be safe regarding their allergenic potential.” (emphasis added).

Article first published 26/09/05


  1. Kranthi KR, Naidu S, Dhawad CS, Tatwawadi A, Mate K, Patil E, Bharose AA,. Behere GT, Wadaskar RM and Kranthi S.  Temporal and intra-plant variability of Cry1Ac expression in Bt-cotton and its influence on the survival of the cotton bollworm, Helicoverpa armigera (Hübner) (Noctuidae: Lepidoptera). Current Science 2005, 89, 291-7.
  2. Kranthi S, Kranthi KR, Siddhabhatti PM and Dhepe VR. Baseline toxicity of CrylAc toxin against spotted bollworm, Earias vitella (Fab) using a diet-based bioessay. Current Science 2004, 87, 1593-7.
  3. Wan P, Zhang Y, Wu K, Huang M. Seasonal expression profiles of insecticidal protein and control efficacy against Helicoverpa armigera for Bt cotton in the Yangtze River valley of China. J Econ Entomol. 2005 98, 195-201.
  4. Abel CA and Adamczyk JJJr. Relative concentration of Cry1A in maize leaves and cotton bolls with diverse chlorophyll content and corresponding larval development of fall armyworm (Lepidoptera: Noctuidae) and southwestern corn borer (Lepidoptera: Crambidae) on maize whorl leaf profiles. J Econ Entomol. 2004 Oct;97(5):1737-44.
  5. Tabashnik BE, Patin AL, Dennehy TJ, Liu YB, Carriere Y, Sims MA and Antilla L. Frequency of resistant to Bacillus thuringiensis in field poulations of pink bollworm. PNAS 2000, 97, 12980-4.
  6. Cummins J. No Bt resistance? Science in Society 2003, 20, 34-35.
  7. Gunning RV, Dang HT, Kemp FC, Nicholson IC and Moores GD. New resistance mechanism in Helicoverpa armigera threatens transgenic crops expressing Bacillus thuringiensis Cry1Ac toxin. Applied and Environmental Microbiology 2005, 71, 2558-63.
  8. Ma BL and Subedi KD. Development, yield, grain moisture and nitrogen uptake of Bt corn hybrids and their conventional near-isolines. Field Crops Research 2005, 93, 199-211.
  9. Saxena D and Stotzky G. Bt corn has ahigher lignin cotent than on-Bt-corn. Am J Bot 2001, 88, 1704-6.
  10. Obrycki JJ, Losey JE, Taylor OR and Jesse LCH. Transgenic insecticidal corn: beyond insecticidal toxicity to ecological complexity. BioScience 2001, 51, 353-61.
  11. “Splitting headache”, Andy Coghlan, New Scientist, 20 November 1999,
  12. Ho MW and Cummins J. Bt risks negligible? Science in Society 2002, 13/14, 32.
  13. Bernstein I, Bernstein J, Miller M, Tiewzieva S, Bernstein D, Lummus Z, Selgrade M, Doerfler D and Seligy V.  Immune responses in farm workers after exposure to Bacillus thuringiensis pesticides. Environ Health Perspect 1999, 107,575-82.
  14. Vázquez-Padrón RI, Moreno-Fierros L, Neri-Bazán L, de la Riva G and López-Revilla R.  Intragastric and intraperitoneal administration of Cry1Ac protoxin from Bacillus thuringiensis induce systemic and mucosal antibody responses in mice. Life Sciences 1999, 64, 1897–1912.
  15. Hernandez E, Ramisse F, Cruel T, le Vagueresse R and  Cavallo JD. Bacillus thuringiensis serotype H34 isolated from human and insecticidal strains serotypes 3a3b and H14 can lead to death of immunocompetent mice after pulmonary infection. FEMS Immunol Med Microbiol 1999, 24,43-7. 
  16. 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.
  17. Ho MW, Pusztai A, Bardocz S and Cummins J. Are transgenic proteins allergenic? Science in Society 2005, 25, 4-5.
  18. Kleter GA and Peijnenburg Ad ACM. 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
  19. Fiers MWEJ, Kleter GA, Nijland H, Peijnenburg Ad ACM, Nap JP and van Ham R CHJ. Allermatch TM, a webtool for the prediction of potential allergenicity according to current FAO/WHO Codex alimentarius guidelines. BMC Bioinformatics 2004, 5:133
  20. Batista R., Nunes B, Carmo M, et al. Lack of detectable allergenicity of transgenic maize and soya samples. J Allergy Clin Immunol 2005, 116, 403-10.
  21. Aas K. The diagnosis of hypersensitivity to ingested foods. Reliability of skin prick testing and the radioallergosorbent test with different materials. Clin Allergy 1978, 8, 39-50.
  22. Sabra A, Bellanti JA, Rais JM, Castro HJ, MendezeInocencio J, Sabra S. IgE and non-IgE food allergy. Annals of Allergy, Asthma and Immunology 2003, 90, 71-76.

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?