The safety of B. thuringiensis to humans, other vertebrates, and nontarget invertebrates has been the subject of numerous studies over the past 50 years. These studies began early during Bt's development as an insecticide.33 Because this bacterium was one of the first living organisms registered as an insecticide, many different types of tests were used to determine whether it had any infectious activity to nontarget organisms. Moreover, because prior to registration it was known that most of its toxic activity was due to the proteins that comprised the parasporal body, toxicological studies were also undertaken to determine whether the bacterium and formulated products were toxic to many different types of organisms, including humans. In addition, after Bt's use in agriculture and forestry was well under way, epidemio-logical studies of the these populations were carried out to determine whether there were any health effects because human populations living in suburban areas were periodically subjected to intensive aerial spraying to eliminate invasive species of highly destructive lepidopteran pests. These studies showed that Bt had little if any effect on human health or most nontarget organisms, especially in comparison to many commonly and extensively used synthetic chemical insecticides.
Then, during the 1990s, new concerns emerged about the safety of Bt due to its close relationship to B. cereus, which by that time was known to produce protein toxins during vegetative growth that could cause vomiting (emetic toxins) or diarrhea (enterotoxins). Additionally, the development of genetically engineered insect-resistant crops based on Cry proteins became a controversial new technology that triggered a new round of concerns about the safety of Bt and its Cry proteins to humans and nontarget organisms. These concerns resulted in a wide variety of new studies, many still ongoing, that have reiterated the safety of Bts used as insecticides and have shown that the novel crops based on Bt Cry proteins were remarkably safe for vertebrates and nontarget organisms. It was determined, for example, that Bt strains used in commercial products were capable of producing emetic and entero-toxins during vegetative growth. However, no evidence was found that these were present in commercial products at levels that could cause illness, or that these caused outbreaks of gastrointestinal disease as a result of proper or even improper use of Bt products. Moreover, detailed epidemiological studies carried out in the late 1990s in Canada and New Zealand found no confirmed health impacts on human populations in suburban areas that were treated aerially with commercial Bt formulations to control invasive or natural lepidopteran pests. These studies are further discussed in Section 188.8.131.52.
More recently, owing to an unusual level of concern by the public over the use of genetic engineering techniques to produce food crops (fanned in large part by the public press), extensive studies were undertaken to determine the effect of Bt crops on nontarget organisms in the laboratory and in the field under commercial growing conditions. In the latter case, many of these studies have been long-term, taking place over periods from two to six years. To date, none of these studies has shown any significant impact of human health or on the various nontarget populations studied, again especially when compared to the known detrimental nontarget impacts of many chemical insecticides still used in agriculture, forestry, and vector control. After a brief history of tests to evaluate Bt safety, the most critical of these studies are summarized below.
In addition to insecticidal efficacy, the major impetus for using Cry proteins in Bt crops was their long history of safety to nontarget organisms, especially to vertebrates. The most important levels of Bt Cry protein specificity described above
(Section 184.108.40.206), i.e., activation, binding, and membrane insertion, apply equally to evaluating the safety of Cry proteins whether used in Bt crops or bacterial insecticides. Therefore, data that demonstrate the safety of bacterial insecticides containing Cry proteins are relevant to assessing Bt crop safety. Extensive testing has been and remains required to meet the rigorous safety requirements established by governmental agencies such as the EPA (see also Chapter 4). Many of these studies have their origin in tests developed to evaluate synthetic chemical insecticides but were modified to evaluate properties such as infectivity, mutagenicity, and teratogenicity. However, because hundreds of safety tests were conducted over several decades to register numerous bacterial insecticides based on different subspecies of Bt, governmental agencies considered it valid to use the results of these tests as part of the background information and data used to register Bt crops based on Cry proteins. This strategy has been criticized on the basis that Cry proteins produced by Bt crops are not identical to those used for safety testing that are produced in Bt or surrogate hosts, such as Escherichia coli. In an absolute sense, this is correct because Cry proteins produced in Bt crops are often truncated (in some crops, significantly) compared to protoxins produced in Bt or E. coli. They therefore differ from the latter in mass and exact amino acid sequence.
From the standpoint of safety, however, the most important question is whether the Cry proteins produced in Bt crops are substantially equivalent to those produced in Bt or E. coli that are used for safety testing. The answer, as far as is known, is "Yes." Regardless of the mass of the protein produced in the plant, if the amino acid sequence of the activated toxin is the same as that of the test material produced in alternate host there is no reason to expect that the plant-produced proteins will act differently or pose significant, unintended risks to nontarget organisms. There is always the possibility that the plant could modify the protein during or after translation, and this might make the protein not substantially equivalent. But there is no evidence this happens, or if it does, that a protein becomes more toxic, or, for example, allergenic as a result of such modifications. It must also be realized that such modifications, if they do occur, could decrease insecticidal activity, and therefore plants with such altered proteins would be screened out during agronomic trials. Thus, the agronomic trials themselves may be acting as positive screens for yielding Bt cultivars in which the Cry proteins are substantially equivalent to those produced in surrogate hosts.
In the course of registering Bts for use as insecticides, the principal subspecies evaluated in these tests over the past several decades have been B. thuringeinsis subsp. kurstaki (Btk) and B. thuringeinsis subsp. aizawai (Bta). They serve as the active ingredients of numerous commercial formulations used in many countries to control lepidopteran pests of agriculture and forestry: B. thuringeinsis subsp. israelensis (Bti) is used to control the larvae of mosquitoes and black flies, and B. thuringeinsis subsp. morrisoni (strain tenebrionis) (Btm-t) is used to control certain species of beetle pests. The materials evaluated have been the active ingredients, i.e., sporulated cultures containing spores and crystals of Cry and Cyt proteins, as well as formulated products. Among the materials tested are all of the Cry proteins used in commercial Bt crops currently on the market, with the exception of a few chimeric proteins constructed by using portions of two different Cry molecules, for example, CrylA and CrylF.
In determining which types of tests should be done to evaluate the safety of bacterial insecticides, early tests were based primarily on those used to evaluate chemical insecticides. However, as noted above, the tests were modified to evaluate the risks of Bt, specifically the infectivity of the bacteria and toxicological properties of proteins used as active ingredients. Representative nontarget vertebrates and invertebrates include mice, rats, rabbits, guinea pigs, various bird species, fish, predatory and parasitic insects, beneficial insects such as the honeybee, aquatic and marine invertebrates, and plants. The tests are grouped into three tiers, I through III.34 Tier I consists of a series of short-term tests aimed primarily at determining whether an isolate of a Bt subspecies, as the unformulated material, poses a hazard if used at high levels, typically at least 100 times the amount recommended for field use, to different classes of nontarget organisms (Table 3.4). The principal vertebrate tests include acute oral, acute pulmonary (inhalation), and acute intraperitoneal evaluations of the material. The tests vary in length from a week to more than a month, the length depending on the organism. In the most critical tests, the mammals are fed, injected with, and forced to inhale millions of Bt cells in a vegetative or sporu-lated form. If infectivity or toxicity clearly results in any of these tests, depending on the dose and route of administration, the candidate bacterium may be rejected. If uncertainty exists, then Tier II tests must be conducted. These tests are similar to those of Tier I but require multiple consecutive exposures, especially to organisms in which there was evidence of toxicity or infectivity in the Tier I tests, as well as tests to determine if and when the bacterium was cleared from nontarget tissues. If infectivity, toxicity, mutagenicity, or teratogenicity is detected in Tier II, then Tier III tests must be undertaken. These consist of tests such as two-year feeding studies and additional testing of teratogenicity and mutagenicity. The tests can be tailored to further evaluate the hazard based on the organisms in which hazards were detected in the Tier I and II tests.
Was this article helpful?