Biotechnology Project
Center for Science in the Public Interest
. Comments on EPA Docket No. OPP-30509B, Application by Monsanto for the Registration of Mon 863 Corn Containing Cry3Bb1

Center for Science in the Public Interest(*1) (CSPI) appreciates the opportunity to comment on Monsanto Co.’s application to register MON 863 corn. Corn rootworms (CRW), the target of MON 863, are extremely destructive pests. More chemical insecticide is used to control CRW then any other insect pests of corn in the U.S. The insecticides that control CRW include several organophosphates, which can have serious adverse effects on humans and the environment. In addition, rotation of corn with soybeans, which reduces insecticide use, is no longer controlling CRW in some regions. As those regions expand, chemical insecticide use can be expected to increase.

MON 863 may benefit farmers and the environment if it is found safe and replaces those chemical pesticides. Even if MON 863 is safe, it will likely be lost as an effective CRW control unless an adequate insect resistance management (IRM) plan is developed and adopted. The demonstrated ability of CRW to develop resistance to insecticides, and overcome the barriers of crop rotation, show the need for a rigorous and cautious IRM program. Without such a program, any benefits from the use of MON 863 will be fleeting.

CSPI’s comments below identify flaws in Monsanto’s safety assessment of MON 863 and proposed interim IRM plan. Those deficiencies must be corrected to assure that MON 863 is safe and that its value is conserved.


CSPI identifies for EPA the following deficiencies in Monsanto Co.’s application to register corn containing event MON 863, a version of the Bacillus thuringiensis delta endotoxin Cry3Bb1. EPA should direct Monsanto to correct each of the following inadequacies prior to registering MON 863:

  1. Monsanto has not evaluated potential allergenicity or toxicity of Cry3Bb1 with the most appropriate methods for determining digestive stability.
  2. The protocols used to determine in vitro gastric stability of Cry3Bb1 were not the protocols international experts agree are most useful. Monsanto should repeat gastric stability tests at pH 2.0, and follow other guidelines elaborated by Food and Agriculture Organization/World Health Organization (FAO/WHO) of the United Nations for conducting those tests (1).

  3. Monsanto did not adequately compare the amino acid sequence of Cry3Bb1 with known allergen protein sequences.
  4. Current protocols recommend testing for homology of six identical contiguous amino acids as a possible indicator of allergenicity. Monsanto should repeat their analysis since international experts (1) now agree that Monsanto’s use of eight contiguous amino acid homology may miss shorter matches that indicate allergenicity.

  5. Monsanto did not conduct non-target organism toxicity testing according to EPA guidelines.
  6. Several of the non-target organism tests used to predict environmental risks were not conducted with sufficiently high doses of Cry3Bb1. Monsanto should repeat non-target testing at maximum hazard doses (at least 10 fold higher than the maximum expected environmental exposure) where those doses were not used.

  7. Monsanto did not conduct maximum hazard dose tests of important non-target beetle and nematode species.

    Additional beetles species, such as ground beetles, should be tested because beetles are most closely related to CRW and therefore most likely to be harmed by Cry3Bb1. Monsanto only tested two ladybird beetle species, and they were not tested at a maximum hazard dose. In addition, further testing of free-living nematodes should be conducted since preliminary tests showed significant population reduction of two nematode species when exposed to MON 863 corn.

  8. Monsanto has not conducted field testing long enough to acquire adequate data.

    Current field data are not sufficient to determine the environmental safety of MON 863. Furthermore, Monsanto indicates that it will conduct field tests for only two years. Due to variability of environmental conditions and data, field studies should be conducted for as long as needed to acquire sufficient data, which may be for more than two years. Additional field experiments should include several free-living nematode species and monitoring of total nematode populations, since preliminary tests show significantly decreased numbers of two nematode species exposed to MON 863.

  9. Monsanto has not conducted studies of the persistence of Cry3Bb1 in the soil under field conditions.

    The recent Bt reassessment Scientific Advisory Panel (SAP) stated that it is important to determine the persistence of Bt protein in the soil under field conditions. Monsanto should conduct soil persistence experiments for Cry3Bb1 under field conditions and measure Cry3Bb1 concentrations both close to and distant from MON 863 corn roots.

  10. Monsanto does not provide adequate data or rationale to support its proposed interim insect resistance management (IRM) plan for MON 863.

    Data are unavailable for many aspects of CRW biology necessary to devise an IRM plan that can confidently prevent resistance. Therefore, Monsanto should be required to develop a more protective IRM plan rather than an interim plan. EPA should consult with independent experts to determine the most effective means to prevent resistance to MON 863, determine important data gaps, and identify research to address those gaps. Such consultation should also address other current means of protecting corn against CRW that would contribute to IRM for MON 863, such as rotation with soybeans. If MON 863 supplants corn/soybean rotation for controlling CRW, increased pressure for resistance to Cry3Bb1 will occur.

  11. Monsanto’s IRM proposal assumes that grower adoption of MON 863 will occur slowly, yet quicker adoption could rapidly lead to resistance by CRW.

    Slow adoption of Bt CRW-protected corn could delay CRW resistance, but Monsanto’s assumption does not consider development of CRW-protected Bt corn by other companies, or the multiple reasons that farmers may plant MON 863 corn. In addition, Monsanto relies on arguments about national adoption rates of MON 863, although local market penetration is also important and will likely occur more quickly than national adoption. EPA should mandate local sales limits to assure that market penetration will low enough to prevent resistance.

  12. Monsanto’s IRM provisions to ensure farmer compliance with IRM requirements are not adequate.

    Monsanto relies on grower education to encourage grower compliance. Recent poor compliance rates for other Bt corn crops suggest that additional actions are necessary to assure compliance, including clear penalties for unjustified non-compliance for growers and Monsanto. EPA should require such compliance and enforcement provisions in the MON 863 IRM plan.

  13. Monsanto’s proposal for resistance monitoring is based upon inappropriate definitions of resistance.

    Monsanto defines CRW resistance based on the LC50 for monitored CRW populations exceeding the upper 95% confidence interval. This definition misses resistance where the slope of the dose-response curve has changed, but where the LC50 remains within the confidence limits. Further, Monsanto’s second definition of resistance, based on root pruning, is also too restrictive. Resistance should be defined using slope and LC50, as well as increased CRW survival on MON 863 corn compared to non-resistant CRW.

  14. Monsanto has not considered whether stacking MON 863 with Cry1 or herbicide tolerance genes will increase the acreage of corn containing Cry3Bb1 or Cry1 compared to unstacked varieties alone.

    The greater the acreage containing Cry3Bb1 or Cry1 genes, the greater the likelihood that insect resistance will develop to either gene. EPA should carefully consider the potential impacts of gene stacking of Cry3Bb1 with other GE products on IRM for CRW protected Bt corn and Cry1 corn borer protected corn. Cry gene stacking should be restricted if it may jeopardize IRM efficacy.

I. Human Dietary Safety

Several safety tests performed by Monsanto did not use protocols that could best assure detection of potential allergenicity or toxicity. In particular, Monsanto incorrectly performed in vitro simulated gastric digestion tests (hereafter “gastric stability”) and sequence comparisons of Cry3Bb1 with allergens and toxins. EPA should require Monsanto to repeat the in vitro digestibility and sequence matching tests using currently recommended protocols.

    A. In Vitro Digestibility Studies

Simulated gastric stability is important because stability is a characteristic common to many food allergens and anti-nutrients, especially the most prevalent and serious food allergens. By contrast, lack of stability means that intact protein is not exposed to the intestines where it can be absorbed and act systemically. Therefore, gastric stability helps predict the likelihood of allergenicity or toxicity. While Monsanto’s tests show that the bacterially produced MON 863 does not possess gastric stability, those tests were not performed with the most reliable protocols.

Monsanto notes that its gastric stability study was performed using the protocols described by Astwood et al. in 1996 (2). In 2001, however, the FAO/WHO convened an international conference of experts to reconsider genetically engineered food allergy assessment, including gastric stability assays (1). That conference recommended that pH 2.0 be used rather than pH 1.2 used in the Astwood et al. protocols (2), since the former is closer to the pH of the stomach when it contains food, while the latter is more indicative of “fasting” pH. That difference in pH can be critical in determining gastric stability. Studies with previous Cry proteins have found different stabilities associated with differences in pH (3,4,5).

While several aspects of the FAO/WHO gastric stability protocols have not been validated against known food allergens or non-allergens, the rationale for the pH change is sound. Therefore, EPA should require Monsanto to repeat the gastric stability assay using pH 2.0 and should also consider requiring the use of other aspects of the updated FAO/WHO protocols.

    B. Protein Sequence Comparisons Between MON 863 and Known Allergens

Monsanto searched protein sequence databases for matches of eight or more contiguous amino acids to indicate possible allergenicity, while the recent FAO/WHO protocols (1) call for determining sequence matches of six contiguous amino acids. Thus, Monsanto’s test might not detect similarity between MON 863 and a known allergen.

Monsanto argues against using less than eight contiguous amino acids to determine potential allergenicity. Monsanto acknowledges that cross-reactive allergy epitopes as short as five amino acids have been found in other proteins, but that they do not bind antibody with affinity as high as the original allergen. Thus Monsanto suggests that antibody affinity is an adequate measure of allergenicity, and that even if Cry3Bb1 had six consecutive amino acids in common with an allergen epitope, that would not be enough to cause an allergic reaction.

However, Monsanto does not give adequate support to the use of affinity as an accurate predictor of clinical symptoms. In fact, in some cases of oral allergy syndrome where there is evidence that the cross reacting pollen allergen is the sensitizing agent, the homologous cross reacting food allergen can cause clinically significant symptoms, despite having no more than six identical or similar contiguous amino acids in common with the pollen (6,7). In addition, only a relatively small number of cross-reacting epitopes have been mapped, so it is premature to dismiss conclusions about the clinical importance of matches at a six amino acid level based on Monsanto’s limited examples.

Monsanto mentions concern that matches of six amino acids may occur randomly at relatively high rate. It is possible that a six (or eight) amino acid match could occur with a known allergen that does not correspond to an IgE epitope, and therefore is not directly involved in the allergic response. In addition, because relatively few allergen epitopes have been determined, sequence data may not be sufficient to determine the allergenic importance of such a match. The FAO/WHO protocol (1) recognizes that possibility, but also recognizes that it can be addressed with an assay using pooled serum from individuals known to be allergic to the sequence-matched allergen. EPA should require Monsanto to repeat the search for protein sequence matching at a six amino acid level, which can be done easily since the databases are already available.

II. Environmental Safety

Non-target organism toxicity tests conducted by Monsanto are deficient, according to both EPA guidelines (8) and a recent SAP of expert scientists convened to evaluate those guidelines (9). Several of those tests did not use a maximum hazard dose, as recommended in current pesticide guidelines (8). The inadequacy of dosage levels is even greater for soil-dwelling organisms, because Monsanto’s test dosage is based on an underestimate of actual exposure to Cry3Bb1 in MON 863 fields.

In addition, Monsanto does not subject enough beetle species to maximum hazard dose tests, improperly minimizes dramatic population reductions for two of three nematode species exposed to MON 863 (10), and has not completed important field trials that examine the environmental impact of MON 863 corn. Taken together, those deficiencies cast doubt upon Monsanto’s conclusion that MON 863 is environmentally safe.

    A. High Dose Non-Target Organism Acute Toxicity Tests

These tests were not performed in accordance with EPA’s microbial test guidelines that are also used to evaluate genetically engineered plant-incorporated protectants (8), or in accordance with the recommendations of an EPA convened SAP that evaluated those guidelines (9). Monsanto did not use maximum hazard doses in several of its non-target organism tests.

Acute maximum hazard dose toxicity tests should be performed at a high multiplicity of expected exposure to increase the possibility that any adverse effects would be detected in those short term tests. Also, use of a high dose increases the chance that adverse effects on a less sensitive test species may be detected. Different levels of sensitivity between species have commonly been recognized with Cry proteins, such as between Colorado potato beetle and CRW in response to Cry3Bb1.

Instead of using the accepted definition of maximum hazard dose, which is typically at least 10 - 100 times the estimated environmental concentration (EEC), Monsanto’s tests are based on the absolute concentration of Cry3Bb1 in the plant for non-soil organisms (11). Monsanto characterizes this as &#quot;...very high dietary concentrations (Table 7)” (12). While Table 7 of Monsanto’s safety assessment document (12) confirms that high concentrations of Cry3Bb1 were used compared to some tests for previous Bt toxins, the table fails to disclose that those other Bt toxins were expressed at many times lower concentrations in crop plants than Cry3Bb1 in MON 863. The absence of comparative expression data in Table 7 gives a false impression of greater margins of exposure for MON 863 in toxicity tests than actually occur.

EPA’s SAP stated the importance of maximum hazard dose testing, saying “...the higher the maximum dose the better.” The Panel recognized “...that it is not always practical to use the 100X [EEC] level” but that the maximum hazard dose should remain at least 10 times the EEC (9).

While Monsanto’s use of MON 863 corn tissue in several non-target tests is useful for detecting possible unintended deleterious changes in the plant, the dose of Cry3Bb1 delivered in those tests is not high enough for acute maximum hazard testing. In tests with some organisms, Monsanto used purified Cry3Bb1 protein at higher concentrations than occur in MON 863, but those concentrations remain below the recommendations of both EPA testing guidelines and the SAP. EPA should require Monsanto to repeat maximum hazard toxicity tests at 100 times higher than the EEC for the following invertebrates where the test dose was less that 10 times higher than the EEC: adult honeybee, parasitic hymenoptera, earthworm, and the ladybird beetle (Coleomegilla maculata). The SAP recognized that a 100X dose may sometimes not be achievable. If that is the case with some of the above organisms, a dose between 10X and 100X should be used and justified.

In addition, Monsanto should be required to repeat those tests for other non-target insects that were conducted at less than 100 times EEC, unless the dose used by Monsanto can be justified. Those organisms include: collembola, larval honey bee, ladybird beetle (Hippodamia covergens), and green lacewing.

The method used by Monsanto to determine the soil EEC of 13.3 g Cry3Bb1/g soil underestimates the exposure of soil organisms (see section II.D. “Persistance of MON 863 in the Soil”). The EEC should be based, as a minimum, on the concentration of Cry3Bb1 in the roots of MON 863 corn, or 58 g Cry3Bb1/g (13), since root herbivores, detritus feeders, or their predators and parasites could be exposed to those levels of Cry3Bb1. On this basis, the tests performed by Monsanto on collembola were only about 15 times the more realistic EEC, while the earthworm test was done at EEC levels.

In summary, the only tested non-target invertebrate where the dose met or exceeded the desired 100 times EEC was daphnia. Even if doses between 10 times and 100 times EEC are accepted, only four terrestrial insect species were adequately tested, including only a single beetle species. While EPA microbial pesticide test guidelines only recommend testing of three non-target insect species, those guidelines are not always appropriate for plant pesticides. The SAP (9) recognized that plant pesticides differ significantly from microbial pesticides and may require testing of more than three non-target insect species. It is important to test additional Coleoptera species since important non-target species related to the target corn rootworm (Coleoptera) are often common in corn fields (see section III. B.).

    B. Selection of Appropriate Non-target Insect Species

According to EPA’s Microbial Test Guidelines(8), EPA can require the testing of different non-target organisms than are specified if the agency believes that those other organisms would be more appropriate. EPA should require Monsanto to perform maximum hazard dose tests with several additional Coleoptera, such as ground beetles and other species, to gain confidence that MON 863 will not harm beneficial insects found in corn fields.

Only two species of Coleoptera, both ladybird beetles, have been tested for sensitivity to MON 863, and in at least one case, the concentration of Cry3Bb1 was too low. Since Coleoptera are more closely related to the target organisms than most of the indicator species currently used, and therefore have a higher likelihood of being susceptible to Cry3Bb1, EPA should require appropriate maximum hazard dose testing of additional beetle species.

In addition, preliminary results strongly suggest that free-living nematodes may be harmed by Cry3Bb1 (10). The highest concentration used in those tests was about 10 times less than EEC (based on the root concentration of Cry3Bb1 in MON 863), and was therefore 100-1000 fold below the appropriate maximum hazard dose. Several species of free-living nematodes should be included in additional maximum hazard dose tests (see section III. C.).

    C. Field Data on Non-Target Organism Effects

Greater reliance on non-target field data was recommended by a recent SAP, and is a welcome part of the environmental assessment of MON 863. However, only preliminary results of those field tests are currently available, and those results are insufficient to draw conclusions about environmental impacts of MON 863. Field testing typically needs to be conducted for several years before enough data can be gathered, because it involves more uncontrollable variables than laboratory tests. In addition, environmental conditions that may influence results are variable and often sporadic, and may only be encountered after several years. Therefore it is not possible to predict whether Monsanto’s proposed two years of field testing will be sufficient. EPA should require Monsanto to continue its field tests for as long as necessary to acquire acceptable data rather than for a period fixed two years as proposed by Monsanto.

In addition to concern about the duration of the field tests, Monsanto’s experimental design does not adequately address certain limitations of small field tests. For example, edge effects, such as the immigration of certain beneficial organisms, can mask adverse effects occurring in small field tests. Several of the current field tests do not address the issue of possible edge effects. Monsanto should explicitly address such effects to the extent feasible in all of the current field tests, or at least attempt to address how uncontrollable edge effects may influence those tests.

In preliminary experiments, two of three species of nematodes, including a plant pathogen and a soil inhabiting (free-living) species, suffered significantly reduced populations when exposed to MON 863. The free-living species was exposed to a root homogenate (root extract) that was diluted 10 fold compared to the Cry3Bb1 concentration in MON 863 roots (10). The investigator concluded that because the free-living nematode species were not harmed by a MON 863 soil leachate used in another test, free-living species probably were not at risk. This conclusion was based on the investigators’ unsupported assessment that the leachate more accurately represents the likely means of exposure of free-living nematodes in MON 863 fields.

That conclusion does not account for the many untested species of nematodes that occur in soil and occupy many different important ecological niches. For example, species that feed on detritus containing MON 863 roots may be exposed to higher levels of Cry3Bb1 than occur in the leachate or root extracts. Some untested species of nematodes may be more sensitive than the tested species, and therefore susceptible to the concentrations lower than found in root extracts. It should also be noted that some Cry5 and Cry14 proteins are toxic to nematodes. Cry14Aa1 is toxic to nematodes and CRW, so it may not be surprising that Cry3Bb1 is toxic to both CRW and nematodes (14).

Furthermore, the leachate method may greatly underestimate the concentration of Cry3Bb1 in MON 863 corn field soils. That is because the method used to make the leachate - suspension of soil that contained MON 863 corn plants in an aqueous solution followed by centrifugation, where the supernatant is the leachate - would probably leave most of the Cry3Bb1 in the soil based on the tight binding of active Cry protein to soil clay particles (15,16). Previous studies have shown that aqueous solutions similar to that used in the Monsanto submission are leave much of the active Cry protein in the soil (15,16). Notably, the researchers did not indicate that they tested the leachate for the presence or concentration of Cry3Bb1.

In addition, the a second free-living nematode species unaffected by the MON 863 root extract was reported not to be feeding. Since ingestion is required for Cry protein toxicity, that test was not valid. Therefore, both of the species that were actually exposed to Cry3Bb1 in a realistic manner were harmed. Those results suggest that many other nematode species might also be harmed by Cry3Bb1.

Since nematodes are important and abundant soil organisms, EPA should require Monsanto to perform additional testing of nematodes to determine the impact of Cry3Bb1 prior to registration. Such testing must include the nematode species already examined, field studies on nematode populations, as well as maximum hazard tests and field tests on additional free-living nematode species.

Finally, even after several years of testing, the data from field experiments often result in a substantial amount of statistical variation that can mask adverse effects. In addition, some environmental impacts may take many years to develop. A recently published report by the National Research Council (17) recognized those limitations of small scale field trails and recommended follow-up monitoring programs. EPA should work with USDA/APHIS to develop such programs.

    D. Persistance of MON 863 in the Soil

Monsanto has determined that Cry3Bb1 degrades at a rate of about 50% in less than three days in soil under laboratory conditions after a single incorporation of finely ground MON 863 leaves. However, the Bt reassessment SAP recommended that the rate of degradation of Cry proteins should be determined under field conditions (18). EPA should require Monsanto to determine the rate of degradation of MON 863 in the field for at least one growing season including sufficient time after harvest to determine how quickly Cry3Bb1 degrades.

Furthermore, single incorporation studies to determine the fate of Cry toxins were considered insufficient by the Bt reassessment SAP (18) because actual deposition likely occurs on a continuous basis both from root biomass, as well as through possible exudation of Cry3Bb1 through the roots (18). If exudation occurs at a sufficient rate, it is possible that the concentration of Cry3Bb1 in the soil could increase, at least through the growing season, because breakdown of Cry proteins may be slowed by binding onto clay particles (19).

Conducting field degradation studies is important because Monsanto reported that expression of MON 863 is high in the root tip (20). Corn in particular is noted for its ability to exude proteins through the root cap (18), so there is particular concern that MON 863 corn may exude Cry3Bb1.

In addition, Monsanto calculated the average concentration of Cry3Bb1 in the soil based on even distribution of a single incorporation of corn plant residue to a depth of six inches. That concentration was used as a soil EEC to determine the Cry3Bb1 concentration used in non-target soil organism toxicity tests. Due to root growth and possible exudation, Monsanto likely underestimates the actual concentration of Cry3Bb1 that soil organisms will encounter. While a single incorporation of Bt plant material would expose soil organisms to steadily decreasing amounts of Cry3Bb1, actual exposure during the growing season will be continuous. When the primary exposure to Cry3Bb1 is ingestion of root material or root exudate by herbivores or detritovors, or indirectly by their predators or parasites, exposure to higher concentrations of Cry3Bb1 than estimated by Monsanto can occur.

Finally, while not a current requirement, EPA should require testing of the impact of Cry3Bb1 on soil microbes. Soil microbes are vital for maintaining the properties of soil responsible for crop productivity and such environmentally critical issues as water conservation. Data was generated on microbial impact for currently registered Bt crops, and the Bt reassessment SAP recognized the importance of evaluating possible impact of transgenic pesticidal crops on soil microorganisms and encouraged additional effort in evaluating such impacts. Unfortunately, EPA has not developed guidelines to assess impact on soil microorganisms so it is not clear what testing should be conducted to assess impact on soil microbes.

III. Insect Resistance Management

The high-dose refugia model proposed by Monsanto may not be appropriate because MON 863 does not deliver a high dose as defined for previous Bt corn (21). There are also many gaps in current information about CRW biology that make development of a simple refuge strategy for CRW resistant Bt corn, especially one that is compatible with current high dose resistance management for Cry1Ab corn, difficult to attain (22).

Monsanto also may underestimate the rate of MON 863 adoption by farmers, has not considered the possible impact of gene stacking on the efficacy of its IRM plan, and does not assure that farmers will adequately comply with IRM plan requirements or that insect resistance would be efficiently detected. Given the gaps in the understanding of CRW biology, any IRM plan should err on the side of caution. Instead, Monsanto fills many of those gaps with inadequately supported assumptions.

Finally, any IRM plan should consider all possible means to prevent resistance rather than be limited to a single preconceived approach, such as the refuge strategy. In particular, current soybean/corn rotations reduce populations of CRW, and thereby could reduce selective pressure for CRW resistance by limiting the acreage where MON 863 is planted . Therefore, it is critical that the effect of commercialization of MON 863 on current CRW management practices is carefully considered, and provisions made to continue practices like soybean/corn rotation at high levels. All of the weaknesses in Monsanto’s current IRM proposal need to be corrected to assure an IRM plan that safeguards the efficacy of MON 863.

    A. Rate of Commercial Adoption of MON 863

Monsanto argues that selection for resistant CRW will be minimal for several years after commercialization due to a slow grower adoption rate of MON 863 corn. Monsanto uses this reasoning, in part, to justify its IRM provisions, which are based on an insufficient understanding of CRW biology. Monsanto argues that several years will be required for it to develop sufficient hybrids with desirable agronomic properties containing MON 863, and that licensing agreements with several major seed companies to develop and market additional varieties containing MON 863 do not exist, unlike the situation with previous Bt corn. However, current licensing agreements should not be relied on to predict future licensing.

Furthermore, based on current knowledge of CRW movement, local adoption may be critical for resistance development. High local adoption may occur much more quickly than national adoption, since varieties containing MON 863 well suited to some localities will likely be available at the time of registration.

In addition, Monsanto’s assessment apparently does not consider that at least one other Bt corn variety aimed at CRW is currently being considered for registration, and may accelerate the availability of Bt corn resistant to CRW. It is not clear whether the other CRW Bt Cry protein utilizes the same insect gut receptor as Cry3Bb1, but the ability to develop cross resistance should be assumed unless data suggest otherwise.

Monsanto also argues that adoption of MON 863 varieties may occur slowly compared to Cry1 corn because, unlike the case with Bt Cry1 targeted to European Corn Borer, relatively effective chemical alternatives exist for CRW. However, adoption of MON 863 may be motivated by other reasons. For example, other reported reasons for the widespread rapid adoption of glyphosate resistant soybeans are grower convenience and time saving and less restrictions on crop rotation (23). The same may be true for MON 863 corn.

For all of the above mentioned reasons, Monsanto’s argument that there is reduced concern for resistance in the next three years is seriously flawed. Therefore, EPA should require restricted sales so that local market penetration remains low enough to minimize the possibility of resistance. EPA should consult with appropriate experts to determine how much to limit local sales.

    B. The Need for Additional Research on CRW Resistance Management

Monsanto proposes continued study of resistance management while MON 863 corn adoption is increasing. It is critical to acquire biological data needed to assess the potential effectiveness of the proposed refuge strategy, and to determine whether other approaches to IRM would be more effective. Monsanto does not offer many specific commitments regarding the types of data that it will develop.

Better data are needed in a number of areas that would impinge on the effectiveness of the proposed refuge strategy, and would facilitate models for resistance development. Those data include better information on the movement and mating behavior of adult CRW. In addition, mechanisms of resistance, initial frequency of resistance alleles, levels of resistance of hetero- and homozygous resistant adults and larvae, and whether there are significant fitness costs associated with resistance are critical parameters that EPA must obtain or require Monsanto to develop.

There is even less known about critical aspects of CRW biology for species other than western CRW, in particular northern CRW, so considerable effort needs to be applied to those species as well. EPA should convene with recognized experts on Bt, insect resistance and CRW to assess what data should be collected to facilitate IRM for MON 863 and other CRW pest management strategies.

In addition, EPA should work with experts, such as the USDA NRC-46 committee and others, to determine how the introduction of CRW resistant Bt corn may effect current CRW management strategies. Of foremost importance is how introduction of products like MON 863 may effect rotation of corn and soybeans.

Currently, corn is rotated with soybeans in many areas, greatly reducing CRW impact because CRW cannot grow on soybean plants. That a practice reduces the population of CRW in most corn fields planted after soybeans, and would therefore reduce the population of CRW selected for Cry3Bb1 resistance. If MON 863 discouraged soybean/corn rotation, or if commodity prices or other factors change so that continuous corn, with MON 863, is favored, unanticipated additional resistance selection pressure may occur.

Using multiple means of controlling insect pests typically reduces the selection pressure on any one of those means, reducing the likelihood that resistance will develop. EPA and Monsanto, with the help of independent experts, should consider how best to integrate IRM for CRW resistant Bt corn with other desirable means of managing CRW to sustain the effectiveness of all CRW management strategies.

    C. Compliance Monitoring and Enforcement

Monsanto’s proposal to monitor and enforce compliance with its interim IRM plan relies heavily on grower education and general surveys. Recent data for previous Bt corn varieties shows compliance with IRM requirements of only around 80% (24), demonstrating the need for more effective methods. To be effective, compliance methods must be able to identify non-compliers and penalize non-compliance. We have previously supplied EPA with recommendations for improved compliance that would also be appropriate for the MON 863 IRM program (25). Monsanto’s IRM plan should include those recommendations.

If resistance is discovered, adequate mitigation is critical to eradicate or at least prevent or delay the spread of the resistant insects. Monsanto makes several proposals for mitigation of resistant CRW should they develop. However, Monsanto limits its action plan to the locality where resistance is discovered, while the development of resistant CRW locally may indicate general inadequacies of the national IRM plan. Therefore, any confirmation of resistance should immediately trigger a comprehensive reassessment of the national IRM plan for MON 863.

    D. Gene Stacking and Resistance Management

Finally, no consideration has been presented concerning the possible impact of stacked transgenes (that is, several transgenes in one plant) on resistance management. Stacking of MON 863 with MON 810 or herbicide tolerance genes may increase the total acreage of corn containing those genes either alone or in combination compared with the acreage containing those genes if they were not stacked.

Of particular concern is the increased possibility of resistance to MON 810. In particular, MON 810 does not provide a sufficiently high dose for corn earworm/cotton bollworm (Helicoverpa zea) for the refuge strategy to be effective if a high percentage of corn acres contain Cry1Ab. Previous evaluation suggested that as long as the percentage of corn in a region containing Cry1Ab remained low enough (below about 50%), selection pressure may be low enough to minimize resistance concerns (18). If stacking of MON 810 with MON 863 significantly increases the acreage of corn containing Cry1 genes, the likelihood of corn earworm resistance may increase substantially.

How widely stacked-gene varieties are grown depends on a number of factors, including how Monsanto and other companies develop and market CRW-protected Bt varieties. EPA must address the issue of stacking to assure that it does not contribute significantly to insect resistance to Bt products. EPA should restrict gene stacking unless it can be assured that stacking will not threaten IRM.

    E. Miscellaneous Deficiencies Underlying Monsanto’s Proposed IRM Plan

Monsanto argues that CRW resistance has developed relatively slowly to chemical insecticides, suggesting that resistance may also develop slowly to Cry3Bb1. However, selection may differ for chemical insecticides and MON 863. Not all CRW larvae may be exposed to chemical insecticides due to uneven distribution in the soil, while virtually all CRW larvae will be exposed to Cry3Bb1 through feeding on MON 863 roots. Therefore, there may be considerably higher selection pressure on CRW by MON 863 compared to chemical insecticides.

It is also critical that MON 863 fields are monitored for CRW resistant to Cry3Bb1. Effective detection of MON 863-resistant CRW depends on sensitive methods and adequate surveillance. The criteria Monsanto provides for defining resistance are not sensitive enough, and could miss cases of CRW resistance. In particular, Monsanto’s definition of resistance requires that CRW exceed the upper 95% confidence interval of the LC50 of non-resistant CRW. This definition would not encompass possible cases of resistance where the slope of the dose-response curve has changed but where the LC50 remains within the confidence limits. EPA should require a definition of resistance using slope as well as LC50. Secondly, resistance should not depend on corn root damage, as proposed by Monsanto, but rather on increased CRW survival, which is a more direct and sensitive measure of resistance.

Refuge corn and MON 863 must have the same crop rotation history to ensure that the refuge produces enough CRW. For example, if a refuge is planted on first year corn while MON 863 is planted on continuously planted corn, the refuge will typically produce relatively fewer CRW, and therefore will not supply the needed numbers of Cry3Bb1-susceptable adults. In addition, farmers must use the same methods to grow MON 863 and refuge corn.

It may be tempting for growers to plant refugia on fields previously planted to soybean because of reduced CRW control costs. Monsanto must develop specific ways to assure that pest management practices on the refuge are compatible with the IRM plan, including disclosure of the prior crop history of fields planted with MON 863 and refuge corn as well as crop management practices.

If MON 863 is as safe and beneficial as Monsanto believes, then it is important to implement an effective IRM plan. EPA must recognize that effective resistance management is pertinent to the purpose of FIFRA and FFDCA, since the loss of a safe product to resistance may mean that growers would resort to use of remaining chemical insecticides.


Doug Gurian-Sherman, Ph.D.
Science Director, Biotechnology Project
Center for Science in the Public Interest
1875 Connecticut Ave., N.W.
Suite 300
Washington, D.C. 20009


1) Food and Agriculture Organization of the United Nations/World Health Organization. “Evaluation of Allergenicity of Genetically Modified Foods: Report of a Joint FAO/WHO Expert Consultation on Allergenicity of Foods Derived from Biotechnology, 22-25 January 2001. Food and Agriculture Organization of the United Nations (FAO), Rome, Italy, 2001.

2) Astwood et al. (1996) Stability of Food Allergens to Digestion in vitro. Nature Biotechnology, 14: 269-273

3) EPA MRID No. 434392-01 (1994) “Assessment of the In vitro Digestive Fate of Bacillus thuringiensis subsp. kurstaki HD-1 Protein”, Joel Ream for Monsanto Co..

4) EPA MRID No. 447343-05 (1998) “Assessment of the Stability to Digestion and Bioavailability of the LYS Mutant Cry9C Protein from Bacillus thruingiensis serovar tolwothi”, H. Noteborn for AgrEvo Co.

5) H.P.J.M. Noteborn et al. (1995) Safety assessment of the Bacillus thuringiensis insecticidal crystal protein CRY1A(b) expressed in transgenic tomatoes, in Genetically Modified Foods - Safety Aspects, American Chemical Society Symposium Series 605, ed. K. Engel et al., Washington, D.C.

6) Breiteneder H. et al. (1995) Molecular characterization of Api g 1, the major allergen of celery (Apium graveolens), and its immunological and structural relationships to a group of 17-kDa tree pollen allergens, Eur. J. Biochem. 233: 484-489

7) Hoffman-Sommergruber K, et al. (1999) IgE reactivity to Api g 1, a major celery allergen, in a central European population is based on primary sensitization by Bet v 1, J. Allergy Clin. Immunol. 104 (2 pt 1): 478-484

8) Microbial Pesticide Test Guidelines, OPPTS 885.4340, Nontarget Insect Testing, Tier I, EPA 712-C-96-336, February 1996

9) EPA, SAP Report No. 99-06A, February 4, 2000. FIFRA Scientific Advisory Panel Meeting, December 8, 1999. Session I - Characterization and Non-Target Organism Data Requirements for Protein Plant Pesticides.

10) EPA MRID No. 456530-03, Non-target organism trials, Monsanto Study No. 00-CR-032E-7, submitted 2002, 70 p.

11) Ward, D, “Administrative Materials for Amendment of Application to Register a Bacillus thuringiensis Cry3Bb1 Protein and the Genetic Material (Vector ZMIR13L) Necessary for its Production in Corn Event MON 863; EPA File Symbol 524-LEI”, submitted to EPA January 8, 2002...

12) EPA MRID No. 454240-09 (2001) “Safety Assessment of Cry3Bb1 Variants in Corn Rootworm Protected Corn”, J.D. Astwood et al., for Monsanto Co.

13) EPA MRID No. 454240-01 (2001) “Amended Report for MSL-16559: B.t. Cry3Bb1.11098 and NPTII Protein Levels in Sample Tissues Collected from Corn Event MON 863 Grown in 1999 Field Trails”, Y.A. Dudin et al. for Monsanto Co.

14) Zeigler D.R. (1999) Bacillus Genetic Stock Center Catalog of Strains, Seventh Edition, Part 2: Bacillus thuringiensis and Bacillus cereus, 58 p.,

15) Palm C.J. et al. (1994) Quantification in soil of Bacillus thuringiensis var. kurstaki -endotoxin from transgenic plants. Molecular Ecology, 3: 145-151

16) Tapp, H. et al. (1994) Adsorption and binding of the insecticidal proteins from Bacillus thuringiensis subsp. kurstaki and subsp. tenebrionis on clay minerals. Soil Biolo. Biochem. 26:663-679

17) National Research Council, National Academy of Science, “Environmental Effects of Transgenic Plants: The Scope and Adequacy of Regulation”, 2002, p. 320, National Academy Press, Washington, D.C.

18) EPA SAP Report No. 2000-07, March 12, 2001, FIFRA Scientific Advisory Panel Meeting, October 18-20, 2000, “Bt Plant-Pesticides Risk and Benefit Assessments”

19) Tapp, H and Stotzky, G. (1998) Persistance of the insecticidal toxin from Bacillus thuringiensis subsp. kurstaki in soil. Soil Biol. Biochem. 30 (4):471-476

20) EPA MRID No. 455770-01 (2001) “An Interim resistance management plan for corn event MON 863: a transgenic corn rootworm control product”, T. Vaughn et al. for Monsanto Co., see p. 22

21) Meinke, L.J. et al. (2001) Letter from NRC-46 Corn Rootworm Research Committee to EPA

22) Ostlie, K. (2001) Crafting crop resistance to corn rootworm. Nature Biotech. 19: 624-625

23) Carperter, J. and Gianessi, L. (1999) Why U.S. farmers are adopting genetically modified crops. Economic Perspectives

24) Insect Resistance Management Grower Survey, Presentation by the Agricultural Biotechnology Stewardship Technical Committee on 29 January 2002. Calculation based on 92% of the 87% of growers who complied with refuge size requirements also complying with refuge placement requirements.

25) Jaffe, G., “Comments on Compliance Assurance Programs for Bt Crops Submitted by Registrants to Environmental Protection Agency on January, 31, 2002”, letter to Marcia E. Mulkey, Director, Office of Pesticide Programs, Environmental Protection Agency, Washington, D.C., on behalf of Center for Science in the Public Interest, March 1, 2002.

*1. CSPI is a nonprofit education and advocacy organization that focuses on improving the safety and nutritional quality of our food supply and on reducing the carnage caused by alcoholic beverages. CSPI seeks to promote health through educating the public about nutrition and alcohol; it represents citizens’ interests before legislative, regulatory, and judicial bodies; and it works to ensure advances in science are used for the public good. CSPI is supported by the more than 900,000 member-subscribers to its Nutrition Action Healthletter, through foundation grants, and through sales of educational materials.

Frequently-Asked Questions Articles and Speeches Regulatory Actions Reports and Reviews Take Action!