Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 History of Bacillus thuringiensis Research

Bt was first isolated in Japan in 1901 by Ishiwata, but first fully described by the German Berliner and named by him after the province in Germany (Beegle and Yamamoto 1992) . Because of its origin (silkworm), Bt was long considered a risk for the silk industry, but a first product was launched in France in 1938 under the name Sporeine. In the 1950s, several studies focussed on the parasporal body or crystal, which had been noted before but was now characterized in terms of solubility, content, and insecticidal activity. Several European countries, the USSR, as well as the USA started commercial production again in the same period. Research and development received a further boost from formation of large collections of natural Bt isolates. The common belief that Bt was only active against Lepidoptera (the larvae of butterflies and moths) was refuted by the discovery of Bt israelensis, with activity against Diptera (larvae of blackflies and mosquitos) in 1977. Helped by the growth of large strain collections, held all over the world and thought to run in the tens of thousands of strains, Bt’s known activity range was further increased with the discovery of Bt tenebrionis, with activity against Coleoptera (beetles) in 1983. Strains with further new (claimed) activities followed, such as against Hymenoptera (bees and wasps), Hemiptera, Mallophaga, Nematoda and Protozoa, although these have not all been confirmed. With the discovery of methods for genetic modification of plants in 1982, the idea of using genes encoding Bt toxins in transgenic plants to render them resistant to insect pests was rapidly put into practice, with the first report appearing in 1987, and the first transgenic crops in the field by 1996 (de Maagd et al. 1999) .

2 Phylogeny and Ecology of Bacillus thuringiensis

Bt is a member of the Bacillus cereus group of gram-positive bacteria, and typically it will form so-called endospores when it senses adverse environmental conditions. Such spores form a though protective capsule and are dormant, non-reproductive and very hardy structures, that may “germinate”, i.e. resume vegetative growth and division, when conditions are favourable again (Fig. 20.1a). Genetically, Bt is very similar to B. cereus, a causative agent of food poisoning, and to B. anthracis, the causative agent of anthrax in mammals. The distinction is in the details: only Bt forms parasporal crystals during sporulation, which make it a (facultative) insect pathogen, while it lacks the distinctive mammalian-active toxins of B. anthracis. Toxins are encoded (mostly) by genes on mobile genetic elements, plasmids, so that Bt strains that have lost those plasmids are indistinguishable from B. cereus. Although originally associated with insect populations, both Bt as well as B. cereus are found in a large diversity of habitats, such as the phylloplane , soil, and stored grain products. Many isolated Bt strains have no demonstrable toxicity for any tested insect species, and with Bt occurring together with B. cereus, the role of parasporal crystal production in the bacterium’s ecological niche, at a considerable cost of resources, remains somewhat a mystery. The consensus is to call Bt a facultative entomopathogen (Glare and O’Callaghan 2000) .

Fig. 20.1
figure 1

a The life cycle of Bacillus thuringiensis. b Primary structure of three-domain protoxin proteins. The white parts are removed upon activation. Colors of the three domains correspond to those in the next panel. c Cartoon representation of the three-dimensional structure of the 3-domain toxins. d Schematic representation of the mode of action of Bt toxins in the insect gut

Full size image

There are several other bacterial species that produce insecticidal, and even parasporal crystal proteins related to those of Bt, such as Lysibacillus sphaericus, which is used against mosquito larvae, several Paenibacillus species, which cause diseases in honeybee larvae and in the white grubs of Japanese beetles, Brevibacillus laterosporus (toxic to Diptera), and Clostridium bifermentans (toxic to mosquitos) (de Maagd et al. 2003) .

3 Bt Toxin Diversity, Mode of Action, and Specificity

Bacillus thuringiensis, as a species, has a vast armoury of different toxins, both parasporal crystal proteins as well as insecticidal and more general toxins produced during vegetative growth (de Maagd et al. 2003) . The parasporal crystal proteins (and some vegetative, soluble proteins) make Bt an insect pathogen. In a natural setting the toxin’s action on the insect gut may help the spore to get access to the body cavity, where it can germinate and cause septicaemia. In general, toxicity of a strain can be explained by the activity of the particular toxins that the strain produces. Each strain can produce one or more toxins, together determining the host specificity of the strain. By far the most common insecticidal proteins of Bt are the delta-endotoxins , which together form the crystal. Two major groups, Cry (crystal) and Cyt (cytolyic) delta-endotoxins, can be distinguished. The largest group are the Cry proteins, with over 120 different main types known today, and many more minor variants. While initially Cry proteins were classified according to their general insect order activity (CryI against Lepidoptera, CryII against Leps and Diptera, CryIII against Coleoptera, etc.), this classification was found to be untenable in the face of the growing number of holotypes. A new classification based on amino acid sequence homology was later accepted (see: http://www.btnomenclature.info/). Although all Cry1 proteins are still generally active against some Lepidoptera species, and Cry3 proteins against Coleoptera etc., the more recent higher rank number toxins have more varying activity. In general, each toxin has activity against only one or a few species within the same order, some toxins have a wider activity spectrum within an order (particularly so for Lepidoptera) and some have been found to be active against species from different orders. All this also depends on the number of target species that have actually been tested (see also: The Bacillus thuringiensis toxin specificity database at http://www.glfc.cfs.nrcan.gc.ca/bacillus/). Thus, it is important to keep in mind that to refer to “Bt toxin” is meaningless without knowing the specific type of toxin and its activity spectrum.

By far the most common type of Cry proteins appears to have a conserved three-dimensional structure, consisting of three structural domains, and are hence called the 3-domain toxins. These generally occur in crystals as protoxins, with N-terminal and C-terminal parts that after solubilisation from the crystal are removed by proteolytic action (Fig. 20.1b). In the remaining activated toxin, the first N-terminal domain consists of six amphipathic alpha-helices surrounding a central hydrophobic helix, which are thought to become part of the eventual membrane pore (see below). The second and third domain contain mostly beta-sheets, and are involved in binding to specific insect gut receptors (Fig. 20.1c). To add to the complexity of the Bt toxin arsenal, not all Cry proteins have this conserved structure, and other structures have been found (de Maagd et al. 2003) .

The mode of action and its role in host specificity of the three-domain toxins has been extensively studied and is schematically shown in Fig. 20.1). Crystal proteins act on the gut of the insect host, so crystals need to be ingested to become active. Upon ingestion, the protoxins are solubilized from the crystal. This solubilization is dependent on pH. Subsequently protoxins are processed by proteolytic enzymes of the host’s gut, by which the active toxin is produced. The activated toxin subsequently binds to one or several receptors on the surface of the epithelial cells lining the insect gut. The specific toxin/receptor-interaction is a major determinant of host-range, from the toxin side by domains II and III of the activated toxin. In a further not completely characterized, and somewhat contested sequence of events, possibly involving further processing, structural conformation changes and oligomerization of the toxin upon receptor binding, the toxin inserts into the epithelial cell membrane, forming pores. Both the extensive leakage of electrolytes as well as activation of host signal transduction pathways may contribute, but do eventually result in the death of the gut epithelial cells (Vachon et al. 2012) . The insect stops feeding almost immediately, is paralyzed, and dies. From the above description it should be clear that each step of the mode of action contributes to the unique host specificities of the different toxins.

The other relatively common components of many parasporal crystals are the Cyt toxins, so called because in vitro they show general cytolytic activity against many cells, although in vivo they are generally restricted to Diptera, with some exceptions for coleopteran species reported. The protein structure is a three-layered α-β-α fold and it requires proteolytic activation. It acts on target gut epithelial cells by membrane pore formation, although unlike Cry proteins they have no specific receptors but more generally interact with membrane phospholipids (Soberón et al. 2013) .

Other, less common Bt Cry proteins have very different structures from the 3-domain toxins, Cyt, and from each other. Some of these show more similarity to toxins from other pathogens, which may give clues to their mode of action. Except for Cry34/35 in transgenic plants (see below) these do not appear in products. Non-crystal toxins include the Vegetative Insecticidal Proteins (VIPs). Of these, VIP3 is toxic to a range of lepidopterans through pore formation in the insect gut and has been applied in transgenic plants (see below).

The evolution of the great diversity of strains, toxins and specificities, particularly for the 3-domain toxins, may be explained by the organization of the toxin-encoding genes in the bacterium. Most toxin genes are located on large plasmids, which can be mobilized and transferred between bacteria, thus creating new toxin gene combinations. On the plasmids, cry genes are often clustered in groups, and this may lead to intergenic recombination between homologous genes, which leads to reshuffling of gene fragments. This mechanism can be used for creating new domain combinations in the laboratory, and this mechanism probably played an important role in the generation of structural variation (de Maagd et al. 2001) . Furthermore, most toxin genes are located close to sequences related to DNA transposition. Transposition may mobilize individual genes between plasmids and assemble new combinations of genes on plasmids.

4 Applications of Bt in Agriculture and Forestry

The first commercial use of microbial Bt-based insecticides occurred as early as 1938 in France, and in the 1950s in the United States. The history of Bt discovery and use have been extensively reviewed elsewhere (Beegle and Yamamoto 1992; Glare and O’Callaghan 2000) . Commercial viability as an alternative for chemical insecticides arrived in the 1970s with more potent strains. Bt products are particularly used in high-margin crops, such as in vegetables and fruits. On the other hand, Bt has been used against forestry pests such as those caused by gypsy moth and spruce budworm, by means of wide-scale aerial applications. After its discovery, Bt israelensis was developed for use in the US and in Germany’s Rhine valley for control of biting flies and mosquitos, and in Africa for disease vector control. Bt tenebrionis has been used to some extent for Colorado potato beetle control and for control of chrysomelid beetles in Eucalyptus plantations (Entwistle et al. 1993) . Bt is the single most successful biopesticide product type on the market, in the 1990s taking up over 90 % of the global biopesticide market (which by itself is a small part of the total pesticide market). This number has decreased somewhat to an estimated 55 % in 2012, due both to decreased Bt spray use as well as to an increase in the number and volume of other biopesticide products.

Microbial Bt products usually consist of dried spore/crystal-mixtures that can be produced by fermentation on a large variety of media, together with other formulation ingredients. Most information about Bt products comes from the US market, where 13 different active ingredients in 123 products where registered (Walker et al. 2003) . However, many more less well characterized products have been produced, particularly in the Soviet Union, in China or other countries (Glare and O’Callaghan 2000) . Strain improvement for higher insecticidal protein content or wider target spectrum has been undertaken by conjugation of toxin encoding plasmids and site-specific recombination, while improvement of individual toxins has been studied using mutagenesis or domain swapping (Baum et al. 1999) .

Despite its prominent position as a biopesticide , and advantages for environmental impact and non-target effects, Bt products have never taken a large part of the overall insecticide market. This is partially due to the lack of Bt products with activity against major insect pest orders such as aphids and white flies. Additionally, Bt is a not a systemic protectant and in most cases can only be used aboveground and on the outside of the plant, leaving the plant unprotected for pests attacking roots (such as Corn rootworm) or burrowing into plant tissues (such as European Corn borer). Expressing Bt toxins in transgenic crops, where they are expressed in these otherwise unprotected tissues and are continuously present without the need for repeated application, has proven to be an effective alternative.

Improvements in transgenic plants focussed on the differences between microbial and plant gene function, removing cryptic RNA splice sites and other elements that negatively affect RNA stability, and finally optimization of codon use for the target plants through the construction of synthetic genes. This results in Bt toxin levels of 0.2–1 % of total soluble protein and proper resistance to target insect species (de Maagd et al. 1999) . Many different crop/gene-combinations for various pests have been developed. However, only a handful of these have reached the commercialization stage and are still being cultivated somewhere in the world today. The two most substantial crops containing Bt genes for insect control today are cotton and corn. Potato, tomato, eggplant, rice, and soybean varieties with Bt genes have been approved for cultivation but their application so far has been negligible. Cotton has been commercialized from 1996 with a varying array of Bt genes, mostly for resistance to cotton bollworm (Lepidoptera) in a large number of countries, including USA, Australia, India and China, and for Pink bollworm in the USA. This resulted in a vast majority of cotton in some countries being of a transgenic Bt variety. Bt maize has also been cultivated since 1996, first for resistance to European corn borer and Mediterranean corn borer (Lepidoptera), later also for resistance to cutworms (Lepidoptera) and for Corn rootworm (Coleoptera) (see also: www.isaaa.org). Recent developments concentrate on the so-called “stacking” of genes, combining several Bt genes in one variety by crossing, both for wider target spectrum as well as for resistance management (see below). For an extensive up-to-date overview of approved or pending Bt crops world-wide, the reader is referred to one of several online databases (http://www.cera-gmc.org/?action=gm_crop_database).

5 Safety and Resistance

Safety issues in relation with the use of Bt sprays and, particularly, with the use of Bt toxins in transgenic crops are a constant source of study and discussion. With regard to mammalian and human toxicity or pathogenicity, these issues may be partially overlapping as both applications involve similar toxins, but other issues are distinct: sprays may lead to ingestion or inhalation of live bacteria, albeit in usually small amounts, consumption of transgenic crops mostly leads to ingestion of individual Cry proteins. These issues are discussed in great detail elsewhere (Glare and O’Callaghan 2000) .

Few reports exist of adverse effects of Bt sprays, despite its long history of use. Its close relation to the recognized food-pathogen B. cereus has raised some concerns. Some Bt strains produce heat-stable β-exotoxins with demonstrated activity against many insects and even fish and mammals, and are banned from use as pesticides by regulatory authorities. The safety of Bt toxins in transgenic crops is part of a larger discussion of transgenic crop safety. Regulatory authorities such as the US EPA (Environmental Protection Agency) and EFSA (European Food Safety Agency) deem the currently approved Bt crops safe for human consumption and see no proof of adverse effects on the environment. Study of environmental and non-target effects of Bt toxins has been an integral part of the safety assessment of transgenic crops for environmental release (EFSA Panel on Genetically Modified Organisms 2010) .

As for any insecticide, extensive and uninterrupted use of Bt sprays or toxins as in transgenic crops exerts strong selective pressure on insect pest populations leading to the increased presence of resistant individuals that, given time, may replace the sensitive population. This potential problem was recognized early on and observed in the field for some Bt sprays, and received considerable attention upon the introduction of Bt crops. Resistance alleles, for example genetic variants of genes for Bt receptors, are present in all populations at a (very) low frequency and can be readily selected in controlled laboratory conditions. Emergence of resistance in transgenic fields seems rare, although some has been reported. This delay may be partially due to resistance management strategies aimed at reducing the speed of resistance developing in a population (Tabashnik et al. 2013) .

6 Concluding Remarks

Over a hundred years of discovery and use of Bt and its toxins has given us a vast domain of knowledge on toxin structure and action, as well as a valuable tool for insect pest control in agriculture. Further commercial development of new strains and crop varieties has slowed down since their first introduction, due to more stringent registration demands and high cost of the approval process for crops, in combination with increased weariness from the consumer side. Still, Bacillus thuringiensis will continue to fascinate researchers for a long time to come.