Share Button

Bt Toxin in Food Crops is Inadequately Tested

Citizens Oversight (2016-01-10) Ray Lutz

This Page:


You will learn (with evidence) that
  1. Cry proteins which are generated from GM crops differ from those generated by Bt bacteria.
  2. The US FDA and EPA do not mandate any animal whole-food feeding studies.
  3. Studies show that Cry-related toxins do bind to the human gut.
  4. The big difference between GMO food and Bt sprays is that they Bt spray is on the surface and is easily denatured by stomach acid whereas Bt toxins in plant tissue are not necessarily denatured and digested at that point.
  5. most studies consider whether Cry protein fragments will be absorbed and perhaps cause allergic reactions, rather than whether they will cause "leaky gut" like they do in butterflies and moths.

Clearly, there is a need for more study here, but scientists go out of their way to reason that no more information can be gained by studying further. Such conclusions are an outrage to science.

Although this is still a work in progress, it is useful to start the conversation so others may want to contribute to it.

The Issue

We hear that GMO foods are all inherently safe, due to the 1992 U.S. policy declaring that such foods are "Generally Recognized as Safe" (GRAS). GMO advocates state that an incredible amount of testing has been performed and in no case has any danger been found. A big question that must be asked then is whether certain testing is being avoided altogether. We believe this is the case with regard to Bt toxin which is now being embedded in GMO food crops such as corn, soy, and others.

[This article, which must be viewed as a working document rather than a final statement, outlines the issue and provides related documents which support this concern. Quotes will be from the most recently cited article if not separately cited]

What is Bt Toxin?

The Bt trait which has been included in the DNA of these crops using genetic engineering, produces causes each cell in the plant to produce a toxin which kills many pests, including the corn borer.
The insecticidal activity of Bacillus thuringiensis (Bt) was discovered in 1901 in Japan, where the bacterium was isolated from infected silkworms, and was later (1911) rediscovered in Germany in infected flour moth chrysalids (reviewed in Sanchis, 2011). For over 50 years, Bt strains and their insecticidal proteins have been used as commercial biological pesticides (Betz et al., 2000; Sanchis, 2011). The first U.S. registration of a Bt microbial product was in 1961; by 1998, there were approximately 180 products registered in the U.S. Environmental Protection Agency (EPA, 1998a,b). There are reported to be over 120 microbial products in the European Union Hammond and Koch, 2012) and approximately 276 Bt microbial formulations registered in China (Huang et al., 2007). In China, 10s of 1000s of tons of Bt microbial formulations are applied to food crops, forests, and potable water, the latter as a means of controlling mosquitoes and other insect vectors of human disease (WHO/IPCS, 1999; Ziwen, 2010). The extensive use of Bt microbial pesticides worldwide is likely due to their specificity against a limited number of target insect species that greatly limits the potential for impacts to beneficial and non-target organisms (NTOs; Receptor-Mediated Selectivity of Bt Cry Proteins) and lack of environmental persistence of Cry proteins (WHO/IPCS, 1999; Betz et al., 2000; OECD, 2007; Federici and Siegel, 2008). (1)

Bacillus thuringiensis (Bt) are gram-positive spore-forming bacteria with entomopathogenic properties. Bt produce insecticidal proteins during the sporulation phase as parasporal crystals. These crystals are predominantly comprised of one or more proteins (Cry and Cyt toxins), also called δ-endotoxins. Cry proteins are parasporal inclusion (Crystal) proteins from Bacillus thuringiensis that exhibit experimentally verifiable toxic effect to a target organism or have significant sequence similarity to a known Cry protein. Similarly, Cyt proteins are parasporal inclusion proteins from Bacillus thuringiensis that exhibits hemolytic (Cytolitic) activity or has obvious sequence similarity to a known Cyt protein. These toxins are highly specific to their target insect, are innocuous to humans, vertebrates and plants, and are completely biodegradable. Therefore, Bt is a viable alternative for the control of insect pests in agriculture and of important human disease vectors. (2)

We note in the description above that they are "innocuous to humans" but that statement is not very scientific. What should be said instead is that "direct ingestion in humans results in dissociation of Cry and Cryt proteins due to the highly acid environment of the human stomach, and by the time they make it to the more alkaline intestines, the are denatured." The term "innocuous" is a very non-scientific way to state this.

Cry proteins are specifically toxic to the insect orders Lepidoptera, Coleoptera, Hymenoptera and Diptera, and also to nematodes. In contrast, Cyt toxins are mostly found in Bt strains active against Diptera. The Cry proteins comprise at least 50 subgroups with more than 200 members. Cry proteins are defined as: a parasporal inclusion protein from Bt that exhibits toxic effects to a target organism, or any protein that has obvious sequence similarity to a known Cry protein (Crickmore et al., 1998). Cyt toxins are included in this definition but it was agreed that proteins that are structurally related to Cyt toxins retain the mnemonic Cyt (Crickmore et al., 1998). Primary sequence identity among different gene sequences is the bases of the nomenclature of Cry and Cyt proteins. Additionally, other insecticidal proteins that are not related phylogenetically to the three-domain Cry family have been identified. Among these, are binary-like toxins and Mtx-like toxins related to B. sphaericus toxins, and parasporins produced by B. thuringiensis (Crickmore et al., 1998).

We note from the passage above that there are many variations on the theme which are still considered Cry and others Cyt. Still other insecticidal proteins are produced by Bt bacteria. This is why this article discusses Bt toxin rather than focusing on a specific member of the 50 subgroups and 200 members, or the other similar toxins that are mentioned above. This is why testing a specific Bt-generated toxin is not as good as testing the group of all toxins generated by the Bt bacteria in an actual real-life situation.

The mode of action of Cry toxins has been characterized principally in lepidopteran insects. As mentioned previously, it is widely accepted that the primary action of Cry toxins is to lyse midgut epithelial cells in the target insect by forming pores in the apical microvilli membrane of the cells (Aronson and Shai, 2001; de Maagd et al., 2001, Bravo et al., 2005). Nevertheless, it has been recently suggested that toxicity could be related to G-protein mediated apoptosis following receptor binding (Zhang et al., 2006). Cry proteins pass from crystal inclusion protoxins into membrane-inserted oligomers that cause ion leakage and cell lysis. The crystal inclusions ingested by susceptible larvae dissolve in the alkaline environment of the gut, and the solubilized inactive protoxins are cleaved by midgut proteases yielding 60–70 kDa protease resistant proteins (Bravo et al., 2005).

Note that Bt toxins also are active against mosquito larvae:

It is proposed that Cry toxins bind to specific protein receptors in the microvilli of the mosquito midgut cells. In contrast, Cyt toxins do not bind to protein receptors but directly interact with membrane lipids inserting into the membrane and forming pores (Thomas and Ellar, 1983; Gill et al., 1987; Li et al., 1996; Promdonkoy and Ellar 2003) or destroying the membrane by a detergent like interaction (Butko, 2003).

Three major applications of Bt toxins have been achieved: (i) in the control of defoliator pests in forestry, (ii) in the control of mosquitoes that are vectors of human diseases, and (iii) in the development of transgenic insect resistant plants.
The development of transgenic crops that produce Bt Cry proteins has been a major break through in the substitution of chemical insecticides by environmental friendly alternatives. In transgenic plants the Cry toxin is produced continuously, protecting the toxin from degradation and making it reachable to chewing and boring insects. Cry protein production in plants has been improved by engineering cry genes with a plant biased codon usage, by removal of putative splicing signal sequences and deletion of the carboxy-terminal region of the protoxin (Schuler et al., 1998). The use of insect resistant crops has diminished considerably the use of chemical pesticides in areas where these transgenic crops are planted (Qaim and Zilberman, 2003). Interestingly, the use of Bt-cotton in countries like China, Mexico and India showed that the use of this Bt-crop had a significant positive effect on the final yield and a reduction in the use of chemical pesticides, since in these countries the yield loss is mainly due to technical and economical constrains which are overcome in part by the use of insect resistant crops (Qaim and Zilberman, 2003; Toenniessen et al., 2003).

We see that the entirety of this article quoted above had to do with the efficacy of the Bt toxin in killing insects of certain varieties and really nothing regarding why they are not also toxic to humans.

Human Toxicity

Traditional Bt pesticide preparations are applied as a spray to the outside surfaces of the plant. This takes time and effort and can be washed off with water.

Although Bt microbial preparations are safe and efficacious, they are limited in their duration of effectiveness because they can be washed off the plant (e.g., by rain) or inactivated by sunlight within days after application (Federici and Siegel, 2008), and they require considerable water, heat, and feedstock to produce, and must be manually applied, either by hand sprayer on small plots or by machine if applied to large tracts. (3)

This is in contrast with GM Bt varieties where each cell of the plant produces toxins similar to those produced by Bt. However, we note also that toxins produced by GM plants are not identical to those produced by Bt, either intentionally by combining with other sequences to produce novel protein forms, or accidentally, if the DNA sequence is modified through the engineering process. One possible accidental modification may result from injecting the DNA sequence near another sequence which may result in a combination of the two sequences when the proteins are produced.

Below, we learn of the former modification:
Another type of modification that has been used when developing a Bt Cry-containing GM crop is swapping portions or whole domains from one Cry protein with portions or whole domains from another Cry protein (Höfte and Whiteley, 1989; Nakamura et al., 1990; Ge et al., 1991; Honée et al., 1991). Domain swapping has been shown to be an effective way to change the spectrum of activity of a native Cry protein to include a new target pest.

Minor changes in these proteins are not considered a reason to repeat all the studies as they find the results can be "bridged" to the other varieties. We must note that this also means that if mistakes were made in those initial tests, then bad results will be propagated rather than checked by additional tests. Additionally, if those tests are not sufficient to catch human toxicity, then depending on those studies in the future may assume that everything is well when it is not.

There are several important, well-established examples that demonstrate that small changes in Bt amino acid sequence do not change the safety profile for NTOs. Regulatory authorities have required functional studies with sensitive insect bioassays to demonstrate that these small changes do not impact biological activity. If these assays indicate biological activity equivalence of the two protein forms, any other properties of the proteins are considered to be equivalent as well. Consequently, it is not necessary to repeat all of the assays performed for the original safety assessment; regulators will consider “bridging” to the form of the protein that was used for the environmental safety testing.

The literature does admit that changes have been made to the toxins produced by GM plants when compared with toxins produced by Bt in nature, but state that these changes are not significant if the action of the toxin is about the same and because safety has already been demonstrated with traditional Bt sprays.

Thus, despite the changes introduced, the extensive history of safe human consumption of native Cry proteins can be applied to the safety assessment of these modified proteins (Hammond et al., 2013).

This same article then brings in a non-scientific rationale for concluding that they are safe:
Second, Bt Cry proteins in either their microbial or plant-incorporated protectant (i.e., GM crops) form are classified as biopesticides by the U.S. EPA (2015). This is an important designation because biopesticides are generally considered be inherently less toxic and have a narrower spectrum of activity (i.e., affect only the target pest and closely related organisms) than conventional pesticides (EPA, 2015).

How something is designated by the EPA is certainly out of place here. Science fact does not depend on the designation of an authority figure. This is how religion works, not science.

This article uses the "History of safe use" of Bt sprays as reason to think that GM Bt food are also safe, while incorporating the change in the protein which does exist when comparing naturally occurring Bt toxins and those in GM crops.
One of the steps in initial (“Tier I”) evaluation of a protein is to examine its history of safe use (HOSU). As discussed in Section “The Use of Bt Cry Proteins in GM Crops,” some Bt proteins are modified from their native form for use in GM crops; thus, it is important to consider whether the HOSU of one protein can be applied to related proteins (Hammond et al., 2013).

We must underline here that they are not considering the fact that there is a big difference in how these are processed by our digestive system. Bt sprays are easy to wash off and any that persist on the surface of the plants is quickly denatured by the acidic environment of the stomach. In contrast, Bt toxins in GM crops are not on the surface, but exist in every single cell throughout the plant. They are not easy to wash off and may not be denatured by stomach acids.

The Koch article considers whether CRY proteins will be digested (denatured) by the stomach, and they consider if the stomach will be acidic enough if the protein is mixed with other foodstuffs, concluding that "Cry proteins are readily degraded in this assay":
Impact of Protein Digestibility Most ingested dietary proteins undergo hydrolytic digestion and/or degradation (Delaney et al., 2008). To approximate the effects of protein exposure to conditions in the mammalian GI tract, a validated in vitro assay to assess the potential stability of proteins to pepsin digestion has been developed. This reliable and reproducible assay uses a fixed pepsin:protein ratio and low pH (pH 1.2 and 2.0; Thomas et al., 2004). Cry proteins are readily degraded in this assay (EPA, 2001; Okunuki et al., 2002; Herman et al., 2003; Thomas et al., 2004; Cao et al., 2010; Guimaraes et al., 2010). Under conditions of higher pH and lower ratios of pepsin to Cry protein, Cry 1 Ab protein is more slowly degraded, as is expected since pepsin becomes less active at a greater pH (Guimaraes et al., 2010). Although Guimaraes et al. (2010) suggest that the current low-pH test may need to be revisited, Ofori-Anti et al. (2008) reported that, as anticipated based on classic enzymology, varying pH, and pepsin concentration had only small effects on digestion of proteins of intermediate stability to pepsin and no effects on proteins that are either stable in the presence of pepsin or rapidly digested by pepsin.

They go on to consider the implications of cry protein fragments, but not intact cry proteins to mammalian gut:
In pigs and calves, Cry protein fragments are detectable but are progressively reduced in size as they travel down the GI tract. None were detected in the liver, spleen, or lymph nodes (Chowdhury et al., 2003a,b) indicating they were too large to be systemically absorbed from the GI tract.
But absorption was never the active mechanism of these proteins which normally remain in the gut and cause it to "leak" -- to use layman's terms. So studying whether it is absorbed is hardly that important, especially for Cry protein fragments. A great deal of the article deals with general digestibility of proteins and absorption rather than dealing with the active mechanism of Cry and related toxins.

This article confirms that the FDA and EPA do NOT require animal feeding studies of whole foods. What is missing from historical testing is the fact that with Cry protein toxins embedded in every cell of the plant, it may persist through the stomach and become active in the intestine, where it may have similar action as that found in Lepidoptera order insects, either directly on the human gut or indirectly by harming microbiota bacteria that exist in our GI tract.
Following review of relevant data in submitted dossiers from registrants, the U.S. FDA and EPA have not, to date, considered additional animal toxicology studies with whole foods (i.e., GM corn grain or soy meal) as necessary to confirm safety. Rather, they have considered the weight of evidence comprised in part by HOSU, the demonstrated safety of the trait in mammals (i.e., acute and subchronic toxicity testing results), and compositional and agronomic tests to come to address unintended effects and come to the conclusion that Bt crops are as safe as their conventional comparators. (4)
This then proves the assertion that "The FDA does not mandate testing of Bt toxins in the human gut" since they don't even require animal testing.

The article discloses the fact that other countries do require 90-day rodent studies
However, 90-day rodent subchronic feeding studies with whole foods were often required by some countries in the EU to confirm the safety of the first generation of Bt crops (Table ​Table44). Additional repeat-dose toxicology studies that have been conducted for other purposes are also summarized in Table ​Table44 including reproduction and chronic studies on Bt crops as well as feeding studies on commercial Bt microbial formulations (see also Bartholomaeus et al., 2013).

These studies will need to be reviewed in detail as there is no summary of results here. However, we must note that effects on human health are much more stringent than whether a mouse dies in 90 days, which is the method used to determine the toxicity of true toxins. Whole-food studies require generally higher ratio of GM food and longer time periods to be able to effectively determine if adverse human health effects may be predicted. We note that trans-fats won't kill a rat in 90-days, but is still banned from food due to adverse health impacts.

We note that some studies did produce alarming results but were blown off as outliers that need not be further investigated.
One reproduction study (Kiliç and Akay, 2008) reported minor histologic findings in the rats fed Bt maize, but these findings were not consistent with the weight of evidence of many other subchronic rat studies where no evidence of treatment-related histologic changes have been reported.

Unfortunately, the scientific community is bent on convincing themselves that animal feeding studies and human gut studies need not be performed because they are convinced that no further information can be obtained. This is in stark contrast to the inconsistent results obtained that imply that there is actually a problem.
The issues surrounding whole food testing of GM crops (Bt and non-Bt) have recently been examined by two groups of authors (Bartholomaeus et al., 2013; Kuiper et al., 2013). Considering the limitations of whole food testing, such as low sensitivity and difficulty in defining the test material, both groups concluded that routine whole food testing does not add meaningful information to the risk assessment of GM crops and cannot be scientifically justified.

This article does in fact report that cry proteins DO BIND to the intestinal wall:
Vázquez-Padrón et al. (2000b) reported binding of Cry 1 Ac protein to BBMVs [brush-border membrane vesicle] isolated from mouse small intestine.
And then they go on to explain away the risk factors here. What this does imply, however, is that whole food studies are desperately needed. Certainly, it is unlikely that a person will die from Bt toxin, but health effects may nonetheless be significant and unknown.

The usual testing of CRY proteins cited is pouring pure CRY proteins into the stomach of mice(5):

Six oral gavage studies in mice established the LD50 to be >3,280 mg/kg to >5,200 mg/kg for these proteins. Based on these results there is a safety factor of greater than 50,000 for human dietary exposure to Cry 1 Ab and Cry 1 Ac proteins in corn or cottonseed, greater than one million for Cry 3 A protein in potato, and greater than two million for Cry 1 Ac protein in tomato.

But again, this is clearly different from actually eating GMO Bt food because the CRY proteins are not on the surface of the plant, or just poured in... and therefore, are not necessarily digested.

Another article (6) suggests that oral gavage is not a reliable way to test endocrine disruptors. Bt toxin may be an endocrine disruptor. We also know that glyphosate (Roundup) is an endocrine disruptor.

NOTE: although this review of this topic is incomplete, there is sufficient information above to make it a useful discussion point, and it will be enhanced in the future. All changes to the document can be reviewed in the historical list of changes shown below.

--Ray Lutz



1 , 3 , 4 : Koch, et al, "The food and environmental safety of Bt crops" --

2 : Bravo, et al, "Mode of action of Bacillus thuringiensis Cry and Cyt toxins and their potential for insect control"

5 : SAFETY OF Bacillus thuringiensis Proteins Used to Control Insect Pests in Agricultural Crops --

6 : Should oral gavage be abandoned in toxicity testing of endocrine disruptors? --

Media Form edit

Title Bt Toxin in Food Crops is Inadequately Tested
Publisher Citizens Oversight
Author Ray Lutz
Pub Date 2016-01-10
Media Link
Remote Link
Embed HTML
Forum Link
Keywords Gmo Open Forum, March Against Monsanto
Media Type Article
Media Group News, Blog Entry
Curator Rating Plain
Author Name Sortable
Static Image HTML
Thumbnail Link
Topic revision: r3 - 2016-04-10, RaymondLutz

This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Copswiki? Send feedback