Abstract EFSA responded to our perspective article on the safe use of the insect pathogen Bacillus thrurigiensis (Bt). In doing so they admitted that there is no direct evidence to suggest that B. thuringiensis can cause diarrhoea. They nevertheless continue to repeat the assertion that Bt cannot be distinguished from Bacillus cereus, even though nearly all Bt strains, and certainly all biopesticide strains, can be distinguished from B. cereus using multi-locus sequencing typing. EFSA also continue to repeat the unsupported and speculative hypothesis that Bt strains could be capable of causing cryptic infections in humans. This hypothesis is very much against the weight of all available safety and epidemiological data. Moreover, genotyping schemes of B. cereus group clinical infections also show that biopesticide strains have never been associated with human infections. Our position that Bt biopesticides and Bt isolates from the clade dominated by invertebrate pathogens are incapable of causing infections in humans is well supported by the international community of scientists familiar with the data on the safety of Bt after more than four decades of extensive use in agriculture and forestry. microbial pesticide, risk assessment, Bacillus thuringiensis, biological safety In our perspective piece on safety and biology of Bacillus thuringiensisis (Bt), we sought to bring to bear both the weight of historical evidence and recently available molecular data to ongoing discussions on the licensing and regulation of Bt-based microbial products in the European Union (Raymond and Federici 2017). We welcome the acknowledgement by EFSA that there are no clear epidemiological or aetiological data to link Bt to food poisoning (da Silva Felicio 2017). Regrettably, their response to much of the data we presented and summarized is to continue to equivocate over the safety of this important microbial pesticide (da Silva Felicio 2017). We are aware of EFSA’s role as a body responsible for risk assessment, rather than for determining policy. It is precisely their fundamental risk assessment philosophy that we would like to address here. Neither of the present authors is a stranger to risk assessment: BR sits on a national committee advising the UK government on the risks of release of genetically modified organisms and biocontrol agents; BF advised the World Health Organization on the widespread use of Bt to help control onchocerciasis in West Africa, and has served on Scientific Advisory Panels for the US Environmental Protection Agency for more than 10 years (our scientific opinions are our own and, of course, do not necessarily reflect those of these bodies). Notably the inclusion of Bt in the onchocerciasis control programme was critical for the sustainability management of the blackfly vector and ultimately helped preserve the health of millions of people in West Africa (World Health Organization 1995). We still disagree with a number of points raised in the response by EFSA. First, the original review commissioned by EFSA and their response letter do not acknowledge that the genetic heterogeneity of the Bacillus cereus group can be a very useful guide to risk assessment. In the most extreme example, Bacillus anthracis is, by one estimate, a million times more dangerous to vertebrates than entomopathogenic B. thuringiensis (Siegel 2001). While we admit that the taxonomic position of Bt is confused and polyphyletic, it is also clear that while Bt and Bc isolates that are more closely related to anthracis are more dangerous to vertebrates, isolates that belong to the group dominated by invertebrate pathogens are much less so (Guinebretière et al.2010; Raymond et al.2010; Raymond and Bonsall 2013). This separation of most B. cereus group pathogens into these two groups (anthracis like strains and invertebrate specialists) is now supported by whole genome data and detailed analyses of the distribution of plasmids and virulence genes (Zheng et al.2017). The statement that ‘the possibility that some cases of foodborne illness attributed to B. cereus group are actually caused by B. thuringiensis cannot be excluded’ reveals much about the risk assessment philosophy in EFSA. This statement undermines the basis of acute oral toxicity testing that EFSA, and a wide range of risk-assessing bodies, enforce for chemical and biological pesticides. Toxicity and infection tests are there precisely to assess these infection risks. In science it is fundamentally difficult to exclude the possibility of a whole range of notions, including the absence of all risk. Invisible gremlins could be defecating in the food of honest EU citizens and giving them diarrhoea; this possibility cannot be completely excluded. We can, however proceed by hypothesis and evaluation, and here it is crucial that there is no evidential support for these gremlins. What the authors of the EFSA opinion are doing is proposing a novel hypothesis, which claims that Bt is capable of causing cryptic infections in humans. While risk assessment does involve some imaginative hypothesis construction, the cryptic infection hypothesis also does not have any evidential support, despite testing above and beyond the normal levels of proof required for licensing of plant protectants in the EU or any other jurisdiction. In addition, as we described in our perspective, it is possible to distinguish between Bt based microbial pesticides and the rest of the B. cereus group using multi-locus sequencing typing (MLST). Licensed biopesticides are derived from specialized invertebrate pathogenic clades and have characteristic genotypes that are not shared by B. cereus sensu stricto. Since thousands of strains have been genotyped by MLST we can use MLST databases to test this hypothesis of cryptic infection. Bt kurskaki, or Sequence Type 8 in the oldest MLST scheme (Priest et al.2004), is the most widely used biopesticide genotype (as DiPel) and also probably the most common genotype in the environment. Yet not one of any of the hundreds of B. cereus group strains isolated from humans has this genotype—data which fail to support, once again, the idea of cryptic infections. The notion that we require a better animal model for the safety testing of Bt is surprising. On the one hand it is absurd to try to identify an animal model for Bt, when it cannot infect vertebrates. Moreover, we only require an animal model when it is impossible to conduct experiments on humans. For Bt, there have been human volunteer infection studies, as well as substantial epidemiological data from the application of Bt to large cities. All these confirm that Bt is safe, and the evidence is overwhelming that it does not infect humans or other vertebrates when ingested or inhaled (Federici and Siegel 2007). The authors are not alone in holding this opinion. In order to determine whether there is a broad scientific consensus on the safety of Bt we canvassed the members of the Society for Invertebrate Pathology, the leading international society promoting the study of invertebrate pathogens as well as the Chinese Society for Microbiology and China Society of Plant Protection. Thus far, nearly over 200 scientists familiar with the data on the safety of Bt, after more than four decades of extensive use in agriculture and forestry, agree with the position stated in our previous perspective piece, as well as this letter (available at https://docs.google.com/spreadsheets/d/1oH-qLfVxgk9XjyEiDWZ89jPKxVOqznKZDKN8RBZpuBs/edit?usp=sharing). The philosophy that we must attempt to exclude all risk is of course embedded in the ‘precautionary principle’. While the precautionary principle has value, particularly when there is an absence of good data, we believe that sensible risk assessment should proceed from evidence-based assessment of hazards and of the probability of their occurrence. Bacillus cereus food poisoning typically presents as mild and transient diarrhoea (Stenfors Arnesen et al.2008), and in the worst case scenario the hazard from Bt is therefore mild. Truly dangerous infections require either Bacillus cytotoxicus producing Cyt K or specific B. cereus genotypes that produce the cereulide toxin (Lund, De Buyser and Granum 2000; Guinebretière et al.2010); no one has ever suggested that Bt is ever likely to pose that kind of hazard. As for probabilities, the balance of the evidence suggests that the probability that insect pathogenic Bt can cause food poisoning is very remote indeed. Its ability to produce enterotoxins in vitro is largely irrelevant since preformed enterotoxins produced by Bt or B. cereus are not expected to survive transit in the gut (Ceuppens et al.2012); in order to pose a risk to humans Bt would have to replicate in the intestinal tract, and there is no evidence that this occurs in any vertebrate. However, EFSA appear to embrace the precautionary principle to the extent that they seem to disregard the evidence of well-established infection tests, and in the case of Bt, a wealth of additional data. Bizarrely, they are very quick to exclude the possibility that B. cereus is capable of causing infection simply because it is present at a concentration that they have decided is safe (see comments regarding cheese noodles in da Silva Felicio (2017)), and therefore are ready to ignore uncertainties associated with microbial quantification in food, probabilistic dose response, or genotypic variation in infectivity of a known opportunistic pathogen, in order to defend their position. We believe that requiring more and more data (or more and more restrictions) to address ever more imaginative risk hypotheses for products that are not pathogenic to humans and that pass all safety tests with flying colours is not helpful for farmers and consumers with a stake in environmentally friendly food production. Bt products, aside from their use in integrated pest management programmes, are some of the few insecticides permitted in organic agriculture. People who eat organic vegetables sprayed with Bt products routinely consume viable Bt spores when used just prior to harvest. Bt products have been used in organic agriculture for more than 30 years, without any evidence of causing diarrhoeal disease. Moreover, if restrictions are placed on the use of Bt in agriculture and forestry, it is likely there will be a return to using broad spectrum synthetic chemical insecticides, which we are sure EFSA will agree, have higher risks for non-target organisms, including humans. REFERENCES Ceuppens S, Rajkovic A, Hamelink S et al. Enterotoxin production by bacillus cereusunder gastrointestinal conditions and their immunological detection by commercially available kits. Foodborne Pathog Dis 2012; 9: 1130– 6. Google Scholar CrossRef Search ADS PubMed da Silva Felicio M. Reply to the article ‘In defense of Bacillus thuringiensis, the safest and most successful microbial insecticide available to humanity—a response to EFSA’. FEMS Microbiol Ecol 2017, DOI: 10.1093/femsec/fix171. Federici BA, Siegel JP. Assessment of safety of Bacillus thuringiensis and Bt crops used for insect control. In: Hammond BG (ed.). Safety of Food Proteins in Agricultural Crops . London: Taylor and Francis, 2007, 46– 101. Guinebretière M-H, Velge P, Couvert O et al. Ability of Bacillus cereus group strains to cause food poisoning varies according to phylogenetic affiliation (Groups I to VII) rather than species affiliation. J Clin Microbiol 2010; 48: 3388– 91. Google Scholar CrossRef Search ADS PubMed Lund T, De Buyser ML, Granum PE. A new cytotoxin from Bacillus cereus that may cause necrotic enteritis. Mol Microbiol 2000; 38: 254– 61. Google Scholar CrossRef Search ADS PubMed Priest FG, Barker M, Baillie LWJ et al. Population structure and evolution of the Bacillus cereus group. J Bacteriol 2004; 186: 7959– 70. Google Scholar CrossRef Search ADS PubMed Raymond B, Bonsall MB. Cooperation and the evolutionary ecology of bacterial virulence: the Bacillus cereus group as a novel study system. Bioessays 2013; 35: 706– 16. Google Scholar CrossRef Search ADS PubMed Raymond B, Federici BA. In defense of Bacillus thuringiensis, the safest and most successful microbial insecticide available to humanity—a response to EFSA. 2017; 93: fix084. Raymond B, Wyres KL, Sheppard SK et al. Environmental factors determining the epidemiology and population genetic structure of the Bacillus cereus group in the field. PLoS Pathog 2010; 6: e1000905. Google Scholar CrossRef Search ADS PubMed Siegel JP. The mammalian safety of Bacillus thuringiensis based insecticides. J Invert Path 2001; 77: 13– 21. Google Scholar CrossRef Search ADS Stenfors Arnesen LP, Fagerlund A, Granum PE. From soil to gut: Bacillus cereus and its food poisoning toxins. FEMS Microbiol Rev 2008; 32: 579– 606. Google Scholar CrossRef Search ADS PubMed World Health Organization. Onchocerciasis and its Control: Report of a WHO-Expert Committee on Onchocerciasis Control . WHO Technical Report Series 852. Geneva: World Health Organization, 1995. Zheng J, Gao Q, Liu L et al. Comparative genomics of Bacillus thuringiensis reveals a path to specialized exploitation of multiple invertebrate hosts. MBio 2017; 8: e00822– 17. Google Scholar CrossRef Search ADS PubMed © FEMS 2017. All rights reserved. For permissions, please e-mail: email@example.com
FEMS Microbiology Ecology – Oxford University Press
Published: Jan 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera