Insecticidal bacterial proteins identify the midgut epithelium as a source of novel target sites for insect control.код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 22:357-371 (1 993) Insecticidal Bacterial Proteins Identify the Midgut Epithelium as a Source of Novel Target Sites for Insect Control Brian A. Federici Department of Entomology and Interdepartmental Graduate Program in Genetics, University of California, Riverside, California The spore-form ing bacterium, Bacillus thuringiensis (Bt), produces a crystal Iine parasporal body during sporulation, which i n many subspecies contains one or more proteins selectively toxic to insect midgut epithelia. Most of these proteins are protoxins with a molecular mass of about 133,000. When ingested by a susceptible insect, the inclusions dissolve in the midgut juices and are activated by proteolytic cleavage, which releases a toxic peptide of about 65,000. In susceptible insects, this peptide binds to sites on the microvillar membrane, causing cytolysis, apparently by forming transmembrane pores. The cytolysis of midgut cells results i n paralysis and subsequent death of the insect. Though less common, naturally truncated protein toxins with masses of about 70,000 also occur. Three major pathotypes of Bt proteins are known: Cry1 and Cryll proteins toxic primarily to lepidopterous insects, Crylll proteins toxic to coleopterous insects, and CrylV proteins toxic to dipterous insects. The genes encoding more than 50 Bt crystal proteins have been cloned and sequenced and several of these have already been used to construct recornbinant microbial insecticides and transgenic organisms including viruses, algae, and insect-resistant plants. Analysis of Bt genes indicate that the protein toxins they encode consist of at least two functional domains: a seriesoffive blocks of conserved amino acids that comprise the structural core of the molecule including the putative toxic moiety, and a hypervariable region thought to define the spectrum of activity. Determination of the mode of action at the molecular level and the genetic basis for insect specificity should enable recombinant DNA technology to be used to expand the insect host range of Bt as well as increase its toxicity against insects. Moreover, the high toxicity and specificity of the insecticidal proteins produced by Bt and their action on midgut microvilli suggest that other types of peptides or simple organic molecules could be designed as selective insecticides to attack other midgut functions such as ion channels, peptide and amino acid transport systems, and midgut proteases. o 1993 WiIey-Liss, Inc. Key words: Bacillus thuringiensis, insecticidal proteins, midgut epithelium Received December 16, 1991; accepted March 26, 1992. Address reprint requests to Dr. Brian A. Federici, Department of Entomology, University of California, Riverside, CA 92521. 0 1993 Wiley-Liss, Inc. 358 Federici INTRODUCTION The critical need for safe and effective alternatives to chemical insecticides has stimulated considerable interest in using pathogens as biological control agents for insects of agricultural and medical importance. Although interest is now greater than ever, the actual use of pathogens in control programs is low, and accounts for less than 1%of the insect control agents used annually around the world. The pathogen that has been the most successful and holds considerable potential for further development is the insecticidal bacterium Bacillus thuringiensis. The species Bacillus thuringiensis (Bt) Berliner is actually a complex of bacterial subspecies, all of which are characterized by the production of a parasporal body during sporulation. This parasporal body contains one or more proteins, typically in a crystalline form, and many of these are highly toxic to certain species of insects. The crystal proteins of insecticidal isolates are known as endotoxins and typically occur in the parasporal body as protoxins, which after ingestion are activated by proteolysis in the midgut lumen. The activated toxins destroy midgut epithelial cells, paralyzing the insect and eventually causing its death. Over the past decade, research has intensified on Bt as a result of several factors, the most important of which are continued pressure to reduce or eliminate synthetic chemical insecticides from pest control programs, the safety of the bacterium and its protein toxins to vertebrates and the environment, its unique mode of action, its chemical properties as a protein insecticide, and the development of recombinant DNA technology. The latter two are particularly important because our knowledge of Bt protein molecular biology is advancing rapidly, and this knowledge will facilitate the use of recombinant DNA techniques to develop a much wider range of insect control agents and strategies. These will likely include recombinant insecticides as well as transgenic plants protected against insects such as grasshoppers, aphids, and others for which we currently have no effective Bts or other control methods. As a result of the enormous potential that recombinant DNA technology provides for advances in plant protection and other future insect control strategies based on Bt, almost every major chemical company, and many emerging biotechnology companies worldwide, are investing heavily in this area of research. Aside from this specific focus on Bt, the novel mode of action of Bt proteins, i.e., the use of midgut microvillar proteins as targets to attack and cause insect death, suggests that alternative midgut functions or proteins might prove just as useful as nove1 target sites. In particular, these would include the numerous enzymatic and transport proteins that occur in the midgut microvillar membrane and are so important to insect growth and development. In the context of this meeting, which deals primarily with the nervous system as a source of new target sites for insect control, the purpose of this paper is to show that what we have learned about the insecticidal proteins of B . thuringiensis over the past ten years identifies the midgut epithelium as a potential source of new target sites. This will be done by briefly reviewing our knowledge of the basic biology of Bt and the insecticidal Insecticidal Bacterial Proteins 359 proteins it produces, and then suggesting examples of midgut functions that might serve as target sites. SYSTEMATICS AND GENERAL BIOLOGY From the standpoint of bacterial systematics, B. thuringiensis is a very close relative of B. cereus, and in fact the only reliable character that differentiates the two is the occurrence of the parasporal body in the former. As a result, the validity of maintaining the two as separate species has legitimately been brought into question [l].It also now appears that in almost all known isolates of B. thuringiensis, the proteins that make up the parasporal body are encoded on plasmids borne by the bacteria . These plasmids can be lost during growth and reproduction, and when this happens, the B. thuringiensis strain “becomes” B. cereus, using accepted methods of bacterial identification and classification. Nevertheless, because the name is so entrenched in the literature, and the bacterium is becoming so important in pest control, the name B. thuringiensis is likely to stand, and with the subspecific epithets, will be used here. As of 1991, at least several thousand isolates of Bt have been obtained from a variety of sources such as living and dead insects, soil, grass, grain dust, and water. These have been divided into groups, of which there are now nearly 30 , that are differentiated from one another on the basis of the H antigen, i.e., flagellar serotype, which is indicated by a number (e.g., H 1)or a number and letter combination (e.g., H 3a3b) as well as a serovar name (e.g., H 1 is thuringiensis; H 3a3b is kurstukz). Despite Bt’s growing importance as a pesticide, little is actually known about its biology and the role of the insecticidal parasporal body in nature. As noted above, Bt can be isolated from a wide variety of habitats, but it does not grow well in many of these because it is not a major or dominant species, Moreover, it is interesting in that, unlike many viruses, fungi, and protozoa, Bt has never been reported as the cause of large-scale epizootics. Bt is, however, quite commonly isolated from grain dust, and the original description by Berliner was based on an isolate from the Mediterranean flour moth, Anugustu kuehniellu (Zeller). In addition, the cadavers of insects killed by Bt provide very suitable substrates for Bt’s reproduction and sporulation [ 5 ] .It would appear, therefore, that bacteria which harbor plasmids encoding insecticidal bacterial proteins have a selective advantage when the spores and parasporal bodies occur together, and the number of parasporal bodies ingested by an insect are sufficient to cause death. Despite Bt’s widespread occurrence in nature, optimal reproduction probably occurs in insects, and its biology, or perhaps more appropriately that of the plasmids which bear the insecticidal proteins, may be more meaningfully understood in this context. MAJOR PATHOTYPES The numerous isolates of Bt that have been screened for insecticidal activity can be divided among three major pathotypes: those which exhibit toxicity to either (a) lepidopterous, (b) dipterous, or (c) coleopterous insects (Table 1). 360 Federici TABLE 1. Nomenclature and Properties of Insecticidal Proteins and Their Encoding Genes From Bacillis thurinD.ensis* Gene taxon Protein name Spectrum of activity cylA(a) cylA(b) cyIA(c) cylB cyIC cyID cyIE cyIlA cylIB cy1IlA cy1lIB cylVA cylVB cyrvc cyIVD CryIA(a) CryIA(b) CryIA(c) CryIB CryIC CryID CryIE CryIIA CryIIB CryIIIA CryIIIB CryIVA CryIVB CryIVC CryIVD CVtA Lepidoptera Lepidoptera Lepidoptera Lepidoptera Lepidoptera Lepidoptera Lepidoptera Lepidoptera/Diptera Lepidoptera Coleoptera Coleoptera Dipterab Diptera Diptera Diptera Diptera/other CytA Amino acidsa Mass (kDa)= 1,176 1,155 1,178 1,207 1.189 1,165 1,171 633 633 644 659 1,180 1,136 675 643 248 133.2 131.0 133.3 138.0 134.8 132.5 132.0 70.9 70.8 73.1 74.2 134.4 127.8 77.8 72.4 27.4 *From Hofte and Whiteley . aThese proteins are protoxins which are activated in vivo by proteolytic cleavage in the insect midgut after ingestion. The Cry proteins are cleaved to form activated toxins in the range of 60-65 kDa, with most of the protein (ca 600 amino acids) in the 130 kDa size range being cleaved from the c-terminus. A small amount of cleavage also occurs (25-30 amino acids) from the n-terminus of all Cry proteins. bCryIVproteins are only of significant toxicity to dipterous insects of the suborder nematocera, e.g.,insects such as mosquitoes, blackflies, chironomid midges, psychodid flies, and crane flies. ‘In vitro, the CytA protein has been shown to be cytolyticfor a wide range of cell types including those from invertebrates, as well as vertebrates. Proteolysis yields a broadly cytolytic protein of 25 kDa. The first successful commercial product for insect control was registered well over 20 years ago and had as its active ingredient the HD 1isolate of B. thuringiensis subspecies kurstuki (serotype H 3a3b), an isolate highly and selectively toxic to lepidopterous insects . Early studies of this isolate showed that it produced a bipyramidal parasporal body that consisted of a protein protoxin with a molecular mass of approximately 133,000. When ingested by a caterpillar, this protein was solubilized in the alkaline midgut and activated through cleavage by midgut proteases, releasing a toxic core peptide of about 65,000. Later it was shown that HD 1also produced a protein with a mass of 65,000 toxic to mosquito larvae and caterpillars, and that this crystallized separately into a smaller cuboidal inclusion associated with the bipyramidal parasporal body . Until the mid-l970s, B. thuringiensis was thought only to produce proteins that were insecticidal to lepidopterous insects. Then in 1976, Goldberg and Margalit  isolated in the Negev desert of Israel what turned out to be a new subspecies, i.e., isruelensis (H 14), that proved highly toxic to the larvae of mosquitoes and blackflies. Studies of the parasporal body of this isolate showed it to be substantially different from the classic bipyramidal crystal. It contained at least four major proteins with molecular masses of 27,000,72,000, Insecticidal Bacterial Proteins 361 128,000, and 135,000, respectively, and these were packaged into different inclusion types and assembled into a spherical parasporal body held together by a fibrous, lamellar envelope 191-Studies of these proteins showed that those with a mass of 72,000, 128,000, and 134,000 were similar to the proteins of lepidoptern-activeisolates, but that the 27,000 protein was markedly different, being highly cytolytic to a wide range of vertebrate and invertebrate cells in vitro [lo]. In addition, several studies have provided evidence that the high toxicity of the B. thuringiensis subspecies isruelensis parasporal body is due to synergistic interactions among two or more of its proteins [9, 11, 121. The discovery of B. thuringiensis subspecies isruelensis was followed in the mid-1980s by the discovery of yet another pathotype of Bt, B. thuringiensis subspecies tenebrionis (H 8a8b), which was shown to be toxic to certain coleopterous insects . This new pathotype contains a single toxin gene, which expresses at different phases of growth two proteins with molecular masses of about 70,000, which apparently differ from each other by only slight differences in size. PROTEIN AND GENE NOMENCLATURE The first gene encoding an insecticidal Bt protein (133,000) was cloned from the HD 1isolate of B . thuringiensis subspecies kurstaki . Subsequent studies of HD 1 employing cloned toxin genes showed that the plasmids it harbored encoded at least five insecticidal proteins, four of which occurred in its parasporal bodies, i.e., three slightly different protoxins with molecular masses of about 133,000 that cocrystallized in the bipyramidal crystal and a fourth protein of 65,000 that crystallized to form the associated cuboidal inclusion. Interestingly, when these cloned genes were used as probes, it was found that the plasmid complements of other isolates of subspecies kurstuki (H 3a3b) varied, as did the number of toxin encoding genes they harbored, with some isolates such as HD 73 containing plasmids encoding and expressing only a single protein toxin. These studies were followed by many others over the past decade, showing that in all subspecies of Bt, the genes encoding insecticidal proteins are located primarily on large transmissible plasmids (see  for a recent review). The existence of three different pathotypes, along with the finding that different proteins expressed by the same pathotype could differ significantly in toxicity to the same insect species, led several investigators to begin computer-assisted analyses of the gene sequences to compare the relatedness of the various Bt genes as well as to attempt to identify important functional domains of Bt proteins. These studies revealed that with the exception of the 27,000 protein of B. thuringiensis subspecies isruelensis, all Bt protein toxins have a significant level of nucleotide and amino acid sequence identity, despite the various names and numbers by which they had been referred to, which were typically based upon the subspecies from which they were cloned. To standardize the terminology and provide a nomenclature that contained informationbased on the gene and protein rather that the bacterial host, Hofte and Whiteley proposed a system in which the standard types of Bt proteins are referred to as “Cry” proteins, and the genes “cry” genes (Table 1).This is 362 Federici followed by a roman numeral which indicates pathotype (I and I1 for toxicity to lepidopterans, I11 for coleopterans, and lV for dipterans), followed by an upper case letter indicating the chronological order in which genes with significant differences were described. The I and II for lepidopteran-toxic proteins also indicate size differences, with the I referring to proteins with molecular masses in the 135,000 range, and the II to those in the 70,000 range. Some epithets also include a lower case letter in parentheses, which indicates minor differences in the nucleotide sequence within a gene type. Thus, CryIA refers to a 130,000 protein toxic to lepidopterous insects for which the first gene (cryZA)was sequenced, whereas CryIVD refers to a 72,000 protein with mosquitocidal activity for which the encoding gene was the fourth from this pathotype sequenced. Though perhaps not perfect, this system is preferred to the chaos that existed before it was proposed. The 27,000 CytA protein first isolated from B. thuringiensis subspecies isruelensis differs from other Bt proteins not only in its smaller size, but also in that it is highly cytolytic to a wide range of cell types in vitro, including those of vertebrates (see  for a review). In addition, it shares no apparent relatedness with Cry proteins. Owing to these differences and its broad cytolytic activity, Hofte and Whiteley 121 referred to this as the CytA protein encoded by the CytA gene. PARASPORAL BODY SHAPE AND COMPOSITION In general, the shape of the parasporal body is a good but not absolute indication of an isolate’s pathotype. For example, most isolates of Bt produce a large bipyramidal parasporal crystal (0.5 x 1 pm) that is almost always only toxic to lepidopterous insects. In isolates of some subspecies, such as B. thuringiensis subspecies kurstuki, the bipyramidal crystal may be accompanied by a smaller cuboidal crystal toxic to lepidopterans and mosquitoes. Others such as B. thuringiensis subspecies isruelensis (H 14) and the PG-14 isolate of B. thuringiensis subspecies mowisoni (H 8a8b) produce spherical parasporal bodies (0.7-1 Fm) that are toxic primarily to nematocerous dipterans (e.g., mosquito and blackfly larvae), whereas the ”tenebrionis” strain of B. fhuringiensis subspecies mowisoni (H 8a8b) produces a thin square crystal that is toxic only to certain species of coleopterans. The degree of protein complexity within these parasporal bodies can vary considerably. Single crystals can be composed of a single type of protein molecule or a mixture of as many as three. In addition, a single parasporal body may be composed of from one (e.g.,certain isolates of subspecies kurstuki) to four (e.g., the mosquitocidal PG-14 isolate of subspecies mowisoni) crystals [2,15]. An example of a simple crystal is that of the HD-73 isolate of B. thuringimsis subspecies kurstuki. This isolate contains one cry gene and encodes only the CryM(c) protein which forms a typical bipyramidal crystal during sporulation. The related HD 1 isolate of the same subspecies, however, carries five cry genes [CryIA(u), (b), (c); cryZIA; q U B ] and produces at least four of these during sporulation. The three CryA proteins cocrystallizein a single bipyramidal crystal, whereas the smaller CryIIA protein forms the associated cuboidal inclusion. The most complexparasporalbody occursin B. thuringiensis subspecies morrisoni (Fig. 1)where 3 CryIV proteins, one Cry1 protein, and the CytA protein crystallize Insecticidal Bacterial Proteins 363 ,144 -135 ‘128 - 65 - 27 Fig. 1 . Illustration of some of the key properties of the rnosquitocidal parasporal body of 6acillus thuringiensis subspecies morrisoni (PC-14). a: Scanning electron micrograph of the basically spherical parasporal body and a dislodged bipyrarnidal crystal. x 30,000. b: Transmission electron micrograph illustrating 4 different inclusions of the parasporal body bound together by a fibrous envelope. x 80,000. c: Comparison of the protein composition of the parasporal bodies from 6. thuringiensis subspecies israelensis (1) and 6. thuringiensis subspecies morrisoni (2). The numbers to the right indicate approximate masses of the proteins in kilodaltons. into four different inclusion types that are bound together in a fibrous envelope [151. GENETIC AND STRUCTURALANALYSES OF CRY PROTEINS Since the cloning and sequencing of the first Bt insecticidal protein gene , more than fifty other genes have been cloned and sequenced. Analysis 364 Federici of the numerous nucleotide and deduced amino acid sequences available now for cry genes/Cry proteins has identified two key features common to the activated portions of most Bt proteins regardless of pathotype [Z].The first is that most proteins share in common five different blocks of amino acids, in each of which the sequence from one toxin to another is highly conserved. These blocks are distributed over the molecule from amino acid position 153 to about 680, and are thought to constitute segments of the molecule critical to maintaining its tertiary structure as well as forming the key components that account for the toxicity of Cry proteins. With respect to the latter, the first two conserved blocks are located in the n-terminus of the activated toxin, and are likely hydrophobic alpha helices involved in inserting the toxin into the microvillar membrane and possibly spanning it to form a pore .The second feature of the molecule is that a highly variable block exists in the C-terminal portion of the activated toxin, spanning the region from about amino acid 280 through 460. The sequence of amino acids in the variable block is thought to define the spectrum of activity or host range, probably by attaching the toxin to specific protein binding sites, possibly glycoproteins, on the microvillar membrane, and contributing to the insertion of the toxin into the membrane. Evidence in support of the putative functions of these regions has been provided recently by site-directed mutagenesis of Bt proteins and resolution of the crystal structure of the CryIIIA protein. In the site-directed mutagenesis studies, Ge et al.  exchanged the highly variable block of amino acids between the CryIA(a)and CryIA(c)proteins. CryIA(a)is normally highly toxic for larvae of the silkworm, Bombyx mori, whereas CryIA(c) is virtually nontoxic. Interchanging the variable amino acid block between these two Cry toxins reversed their toxicity for B. mori. By deletion analysis, these investigators were able to identify the region from amino acid 332 to 450 as the CryIA(a) segment that determined specificity (and likely binding) of this endotoxin to B. mori microvilli. Thus, each molecule had the capacity to be highly toxic for B. mori, provided it had an appropriate variable region, in this case amino acids 332 to 450 of the CryIA(a)toxin. In the studies of the CryIIIA crystal structure, Li et al.  have shown that the molecule is wedge-shaped and consists of three domains. Domain I makes up almost half of the molecule, and stretches from the n-terminus up to and including amino acid 290. This domain, which contains all of the first conserved block and a major portion of the second conserved block noted above, is composed of a seven-helix amphipathic bundle, with six helices surrounding a central helix, a structure suitable for spanning the lipid portion of the microvillar membrane. Domain I1 consists of three antiparallel beta sheets around a hydrophobic core, and stretches from amino acid 291 to 500, and thus includes the hypervariable block of amino acids that apparently define the spectrum of activity. Within this domain are portions of conserved blocks three and four. Domain 111 is comprised of amino acids 501 to 644 and consists of two antiparallel beta sheets, within which are found the remainder of conserved block number four along with blocks four and five. The three-dimensional structure resolved by Li et al.  also shows that the five conserved blocks comprise an important part of the interfacing structure among the three domains, thereby forming the structural core of the molecule. Insecticidal Bacterial Proteins 365 The structural and experimental genetic analyses in combination with analyses of the gene sequences are providing important insights into the molecular biology and toxicology of Cry proteins. Based on our current understanding, Cry proteins consist of three domains, with Domain I, and to a lesser extent Domain 111, containing the structural components that enable the toxin to insert into and possibly span the microvillar membrane, forming lethal pores, whereas Domain I1 contains amino acid sequences that initially bind the toxin to the microvillar membrane, and may also be involved in its insertion. MODE OF ACTION Numerous studies have now shown that though the spore can play a role in the pathogenicity of B. thuringiensis, it is the parasporal body that causes the rapid paralysis which ultimately leads to death of the insect (for reviews see ). In the typical Bt, B. fhuringiensis subspecies kurstaski, for example, the parasporal body dissolves after ingestion upon encountering the alkaline (pH 8-10) juices of the midgut. As noted above, many of the known Bt toxins are actually protoxins of about 133,000 daltons (e.g., CryI, CryIVA, and CryIVB) from which active toxins in the range of 60,000-70,000 daltons are cleaved by gut proteases. These activated Cry toxin molecules pass through the peritrophic membrane and bind to specific receptors [19-211, apparently glycoproteins , on the microvilli of midgut epithelial cells. Binding is an essential step in the intoxication process, and in susceptible insects the toxicity of a particular Bt protein is often correlated with the number of specificbinding sites on the microvillar membrane and the affinity of the Bt molecules for these sites [17,22,23]. However, binding by itself does not, in all cases, lead to toxicity, indicating that insertion and likely some sort of processing in the microvillar membrane is required to obtain toxicity [21,24]. Within minutes of binding to a cell, the microvilli lose their characteristic structure. The microvillar membrane slackens and the microtubules degenerate. Subsequently, the cells and organelles such as mitochondria become vacuolated and begin to swell [25,26]. This swelling continues until the cells lyse and slough from the basement membrane of the midgut epithelium. Both columnar and goblet cells are destroyed as intoxication advances. As more and more cells slough, the alkaline gut juices begin to leak into the hemocoel where as a result, the hemolymph pH rises by a half unit or more. This causes the paralysis and eventual death of the insect . Though this general picture of the mode of action has been known for some time, the actual process of intoxication at the molecular level is not well understood, particularly the series of events that occur after the toxin binds to the microvillar membrane. There are now a considerable number of studies on the physiological events that accompany intoxication (see  for other recent key references), but it remains to be determined more accurately which are causes as opposed to effects. At this point there is general agreement that binding and insertion of an activated Cry toxin into the microvillar membrane leads quickly (minutes if not seconds) to an increase in cell membrane cation permeability and an influx of cations, especially potassium. To balance the 366 Federici cationic influx, the cell takes in water, which leads to hypertrophy and eventual lysis. Two major hypotheses have been proposed to explain the cationic influx. The first is that Cry toxins act directly on the microvillar K+-ATPase pump , either intefering with the export of potassium from the cell or causing a passive leakage inward from the midgut lumen. The second major hypothesis is that Cry toxins insert in the membrane, forming nonspecific cation pores [16,30], with each pore consisting of several Cry molecules, e.g., a hexameric membrane-spanning tubular cluster of molecules with the channel in the center . The latter is arrived at by a reconfiguration and aggregation of the Cry molecules after insertion in the membrane in such a way that two of the alpha helices from each Cry molecule associate to form the actual channel . In both of these hypotheses, the toxic action takes place at the microvillar membrane; i.e., no internalization of the toxin occurs nor is there involvement of a second messenger system leading to inhibition of metabolic processes inside the cell. Although the pore-forming hypothesis has gained considerable acceptance over the past few years, it remains to be clearly validated experimentally. In this regard, recent evidence that Cry toxins may act inside the cell suggest that the mode of action may be more complex than currently thought. In patch-clamp experiments with lepidopteran cells cultured in vitro, Schwartz et al.  have shown that the most immediate effects of Cry intoxication observed are on the activation of calcium and then chloride ion channels. They have suggested that binding of the toxin to the cell membrane results in an activation of calcium channels and an influx of calcium into the cell. This is followed by internalization of the receptor-toxin complex or endocytosis of the toxin. In either case, once within the cell, the toxin either unmasks anionic channels or inserts into the plasma membrane, forming chloride channels. Subsequently, these events lead to an influx of water and cations, followed by cell hypertrophy and lysis. These recent observations and discrepancies in interpretations of the data from toxicological and physiological experiments at the cellular level demonstrate the need for more research on the molecular mode of action of Cry toxins. With respect to the broadly cytolytic CytA toxin, this is also thought to form pores in the microvillar membrane, but it differs from the Cry toxins in that it apparently does not require a receptor protein for insertion into the membrane. Instead, it has a high affinity for unsaturated fatty acids in the lipid bilayer of the cell plasma membranes . GENETIC ENGINEERING OF BACILLUS THUMNGZENSIS In addition to the development of Bt isolates as conventional insecticides, transconjugate and genetically engineered strains with novel properties are now corning to market. For example, Ecogen, Inc. of Langhorne, Pennsylvania, has combined three different plasmids, two encoding proteins toxic to lepidopterous insects (CryI), the other a protein toxic to coleopterans (CryIII), into a single strain of B. thuringiensis subspecies kurstaki. This transconjugate strain has an expanded host range that includes both lepidopterous and coleopterous insects. And recently Mycogen Corporation of San Diego, Insecticidal Bacterial Proteins 367 California, has received registration for the first genetically engineered bacterial insecticide, MVP, a killed strain of Pseudomonas fluorescens that contains a Bt toxin active against lepidopterous insects . The apparent advantage of this strain over conventional Bts is greater residual activity in the field, which is obtained by fixing the Pseudomonas cell wall around the Bt protein. Because Bt genes can be manipulated quite easily using recombinant DNA technology, it is likely that a variety of different approaches to controlling insects based on Bt’s insecticidal proteins will be developed over the next few years. For example, fusion proteins with an expanded host range have already been made by fusing two Bt genes together . Another exciting yet controversial development already at hand is the successful engineering of Bt genes into plants such as cotton, tomato, and tobacco to produce insect-resistant transgenic plants [35-381. In all of these cases, only a single Bt gene was transformed into the plant species, and it is already known that insects are prone to develop resistance quickly to engineered bacteria expressing a single gene . This work is controversial primarily from the standpoint of the high risk that the release of such plants would pose for the development of resistance to both conventional and engineered Bts used as periodically applied microbial insecticides. In addition, although there is no evidence whatsoever that transgenic plants are unsafe to eat, the question of their safety has been raised recently . The concern over the potential for the development resistance for conventional and engineered uses of Bt is valid, and the recent organization of an industry-based resistant management group hopefully means the threat of resistance is being taken seriously. Despite potential problems with the use of Bt, it should be noted that though Bt has come a very long way in the last five years, it is still at an early stage of development. In the future, the use of recombinant DNA technology and an expanded knowledge of Bt’s molecular biology, particularly its mode of action at the molecular level, should lead to a greater range of novel and safe insecticides including some for insects for which we currently have few cost-effectivecontrol methods. For example, if the hypotheses concerning the specificity and toxin domains are correct, then it should be possible to extend Bt‘s host range by altering the highly variable region so that the proteins bind to the midguts of a wider range of insect pests. Another strategy being investigated is to fuse the Bt toxin domain to proteins capable of binding to insect midgut cells. Examples of such proteins include the binding domains of antibodies made against surface proteins on insect gut cells, viral surface proteins responsible for attaching insect viruses to midgut cells, and other Bt gut-binding proteins. If successful, it is possible that at least some of these recombinant proteins may be useful in transgenic plants, providing an opportunity for controlling even sucking insects. POTENTIAL TARGETS FOR NOVEL INSECTICIDES IDENTIFIED BY Bt RESEARCH Perhaps even more importantly, especially with respect to the topic of this meeting, is that Bt identifies the midgut epithelium as a new source of 368 Federici target sites for engineered proteins. Since World War 11, the focus of most insecticide research has been on the development of synthetic organic chemicals that targeted receptors in the nervous, hormonal, or cuticular systems of insects. The midgut has not been viewed as an important source of target sites since the arsenicals. The advent of recombinant DNA technology, however, provides an opportunity to reexamine the midgut as a source of target sites for which inhibitory insecticidal proteins, small peptides, and even organic molecules can be designed that are highly selective for insects. This may turn out to be the most important lesson we learn from Bt research. Needless to say, the midgut in all insect species is an extremely important and highly active tissue metabolically. The rnicrovillar membrane itself contains a wide range of surface and internal proteins involved in transport functions such as maintaining ion gradients, absorption of amino acids and other nutrients, and excretion of enzymes . All of these represent potential target sites to be blocked by peptides that can be delivered either in the form of microbial insecticides, as protein insecticides produced by fermentation, or through transgenic plants. In addition, the developing field of mirnetics indicates that it may be possible to synthesize simple insecticidal organic molecules that specifically bind to and inhibit these target proteins. The rationale behind such strategies has already been developed. For example, monoclonal antibodies can be made that bind to and disrupt the function of important microvillar surface proteins. Short inhibitory peptides can be synthesized that mimic the hypervariable region responsible for binding the antibody to a specific protein. These in turn can be used to synthesize, with the aid of computer modeling, simple organic molecules, if necessary, as insecticides. Though no precedent exists for developing insecticides this way, such a strategy has been used recently to develop peptides and simpler organic molecules that bound to a cellular viral receptor and inhibited its ability to induce cellular proliferation . Just ten years ago, virtually nothing was known about the genetics and molecular biology of insecticidal Bt proteins. The advances over the past decade have been remarkable, and include the cloning and analysis of over 50 cry genes, the development of transgenic insecticidal plants and bacterial insecticides, and more recently, resolution of the crystal structure of the first Cry protein. These developments were made possible largely by the advent of recombinant DNA technology. If the rapid progress over the recent past is any indication, further use of this technology in combination with developing knowledge of insect molecular biology and new chemical technologies such as mimetics should lead to a range of new types of selective and environmentally safe pesticides in the future. And the midgut provides an excellent potential source of targets for these. LITERATURE CITED 1. Baumann LK, Okamato BM, Unterman MJ, Lynch J, Baumann P:Phenotypic characterization of Bacillus thuringiensis and Bacillus cereus. J Invertebr Pathol44, 329 (1984). Insecticidal Bacterial Proteins 369 2. Hofte H, Whiteley HR: Insecticidal crystal proteins of Bacillus thuringiensis. Microbiol Rev 53, 242 (1989). 3. de Barjac H, Frachon E: Classificationof Bacillus thuringiensis strains. Entomophaga 35, 233 (1990). 4. Berliner E: h e r die schlaffsucht der Mehlmottenraupe [ Ephestia kuhniella, Zell.] und ihren Erreger Bacillus thuringiensis, n. sp. Z Angew Entomol2, 29 (1915). 5. Aly C, Mulla MS, Federici BA: Sporulation and toxin production by Bacillus thuringiensis var. isruelensis in cadavers of mosquito larvae (Diptera: Culicidae). J Invertebr Pathol46, 251 (1985). 6. Dulmage HT: Insecticidal activity of isolates of Bacillus thuringiensis and their potential for pest control. In: Microbial Control of Insect Pests and Plant Diseases, 1970-1980. Burges HD, ed. Academic Press, London, pp 19S222 (1981). 7. Yamamoto T, McLaughlin RE: Isolation of a protein from the parasporal crystal of Bacillus thuringiensis var. kurstaki toxic to mosquito larvae, Aedes taeniorhynchus. Biochem Biophys Res Commun 103, 414 (1981). 8. Goldberg LJ, Margalit J: A bacterial spore demonstrating rapid larvicidal activity against Anopheles sergentii, Uranotaenia unguiculata, Culex univitattus, Aedes aegypti, and Culex pipiens. Mosq News 37, 355 (1977). 9. Ibarra JE, Federici BA: Isolation of a relatively non-toxic 65-kilodalton protein inclusion from the parasporal body of Bacillus thuringiensis subsp. isruelensis. J Bacteriol165, 527 (1986). 10. Thomas WE, Ellar DJ: Bacillus thuringiensis var. israelensis crystal delta-endotoxin: Effects on insects and mammalian cells in vitro and in vivo. J Cell Sci 60, 181 (1983). 11. Wu D, Chang FN: Synergism in mosquitocidal activity of 26 and 65 kDa protein from Bacillus thuringiensis subsp. israelensis crystal. FEBS Lett 190, 23 (1985). 12. Chilcott CN, Ellar DJ: Comparative toxicity of Bacillus thuringiensis var. israelensis crystal proteins in vivo and in vitro. J Gen MicrobiolZ34, 2558 (1988). 13. Krieg A, Huger A, Langenbruch G, Schnetter W: Bacillus thuringiensisvar. tenebrionis: A new pathotype effective against larvae of Coleoptera. J Appl Entomol96, 500 (1983). 14. Schnep HE, Whiteley HR: Cloning and expression of the Bacillus thuringiensis crystal protein gene in Escherickia coli. Proc Natl Acad Sci USA 78, 2893 (1981). 15. Federici BA, Luthy P, Ibarra JE: The parasporal body of BTI: Structure, protein composition, and toxicity. In: Bacterial Control of Mosquitoes and Blackflies; Biochemistry, Genetics, and Applications of Bacillus thuringiensis and Bacillus sphaericus. de Barjac H, Sutherland D, eds. Rutgers University Press, New Brunswick, pp 1W (1990). 16. Ellar DJ: Pathogenicity determinants of entompathogenic bacteria. In: Proceedings Vth International Colloquium on Invertebrate Pathology and Microbial Control. Pinnock D, ed. Society for Invertebrate Pathology, Adelaide, Australia, pp 298-302 (1990). 17. Ge AZ, Shivarova NI, Dean DH: Location of the Bombyx mori specificity domain on a Bacillus thuringiensis delta-endotoxin protein. Proc Natl Acad Sci USA 86, 4037 (1989). 18. Li J, Carroll J, Ellar DJ: Crystal structure of insecticidal delta-endotoxin from Bacillus thuringiensis at 2.5 A resolution. Nature 233, 815 (1991). 370 Federici 19. Hofmann C, Vanderbruggen H, Hofte H, Van Rie J, Jansen S, Van Mallaert H: Specificity of Bacillus thuringiensis delta-endotoxins is correlated with the presence of high affinity binding sites in the brush border membrane of target insect midguts. Proc Natl Acad Sci USA 85, 7844 (1988). 20. Knowles BH, ElIar JD: Characterization and partial purification of a plasma membrane receptor for Bacillus thuringkmis var. krrrstaki lepidopteran-specific delta-endotoxin. J Cell Sci 83, 89 (1986). 21. Garczynski SF, Crim JW, Adang MJ: Identification of putative brush border membrane-binding molecules specific to Bacillus thuringiensis delta-endotoxin by protein blot analysis. Appl Eaviron Microbiol57, 2816 (1991). 22. Van Rie J, Jansens S, Hofte H, Degheele D, Van MellaertH: Specificityof Bacillus thuringiensis delta-endotoxins: Importance of specific receptors on the brush border membrane of the mid-gut of target insects. Eur J Biochem 186, 239 (1989). 23. Van fie J, Jansens S, Hofte H, Deghelle D, Van Mallaert H Receptors on the brush border membrane of the insect midgut as determinants of the specificity of Bacillus thuringiensis delta-endotoxins. Appl Environ Microbiol56, 1378 (1990). 24. Wolfersberger MG: The toxicity of two Bacillus thuringietasis delta endotoxins to gypsy moth larvae is inversely related to the affinity binding sites on the brush border membranes for toxins. Experientia 46, 475 (1990). 25. Huber HE, Luthy P: Bacillus thuringzensis delta-endotoxin: Composition and activation. In: Pathogenesis of Invertebrate Microbial Diseases. Davidson EW, ed. Allanheld, Osman & Co., Totowa, New Jersey, pp 209-234 (1981). 26. Luthy P, Ebersold HR: Bacillus thuringiensis delta-endotoxin: Histopathology and molecular mode of action. In: Pathogenesis of Invertebrate Microbial Diseases. Davidson EW, ed. Allanheld, Osman & Co., Totowa, New Jersey, pp 235-267 (1981). 27. Heimpel AM: A critical review of Bacillus thuringiensis var. thuringiensis Berliner and other crystalliferous bacteria. Annu Rev Ent 12, 287 (1967). 28. Schwartz JL, Garneau L, Masson L, Brousseau R Early response of cultured lepidopteran cells to exposure to delta endotoxin from Bacillus thuringiensis: Involvement of calcium and anionic channels. Biochem Biophys Acta 1065,250 (1991). 29. English LH, Cantley LC: Delta endotoxin is a potent inhibitor of the (Na, K)-ATPase. J Biol Chem 261, 1170 (1986). 30. h o w l e s BH, Ellar DJ: Colloid-osmotic lysis is a general feature of the mechanism of action of Bacillus tlruringiensis delta endotoxins with different insect specificities. Biochem Biophys Acta 924, 509 (1987). 31. Hodgman TC, Ellar DJ: Models for the structure and function of the Bacillus thuringiensis deltalendotoxins determined by compilationalanalysis. DNA Sequen-J DNA Sequen Mapp 1, 97 (1990). 32. Thomas WE, Ellar DJ: Mechanism of action of Bacillus thuringiensis var. israelensis delta-endotoxin. PEBS Lett 154, 362 (1983). 33. Feitelson JS, Quick TS, Gaertner F: Alternate hosts for Bacillus thuringiensis delta-endotoxin genes. In: New Directions in Biological Control. Baker R, Dunn P, eds. Alan R. Liss, Inc., New York, pp 561-571 (1990). Insecticidal Bacterial Proteins 371 34. Honee G, Vriesen W, Visser B: A translation fusion product of two different insecticidal crystal genes of Bacillus thuringiensis exhibits an enlarged insecticidal spectrum. Appl Environ Microbiol56, 823 (1990). 35. Fischhoff DA, Bowdisch KS, Perlak FJ, Marrone PG, McCormick SH, Niedermeyer JG, Dean DA, Kusano-Kretzmer K, Mayer EJ, Rochester DE, Rogers GS, Fraley RT: Insect tolerant transgenic tomato plants. BioiTechnol 5,807 (1987). 36. Vaeck M, Reynearts A, Hofte H, Jansens S, De Beuckeleer M, Dean C, Zabeau M, Van Montagu MM, Leemans J: Transgenic plants protected from insect attack. Nature 328, 33 (1987). 37. Gould F: Evolutionary biology and genetically engineered crops. BioScience 38, 26 (1988). 38. Perlak FJ, Deaton RW, Armstrong TA, Fuchs RL, Sims SR, Greenplate JT, Fischhoff DA: Insect resistant cotton plants. Biomechnol 8, 939 (1990). 39. Stone TB, Sims SR, Marrone PG: Selection of tobaccobudworm for resistance to a genetically engineered Pseudomonas fluorescens containing the delta-endotoxin of Bacillus thuringiensis subsp. kurstaki. J Invert PatHol53, 228 (1989). 40. Goldburg RJ, Tjaden G: Are B.T.K. plants really safe to eat? Bio/Technol8, 1011(1990) 41. Dow JAT: Insect midgut function. Adv Insect Physiol19, 187 (1986). 42. Taylor CW: Calcium regulation in insects. Adv Insect Phys 19, 151 (1986). 43. Giordana B, Sacchi FV, Parenyi P, Hanozet GM: Amino acid transport systems in intestinal brush-boarder membranes from lepidopteran larvae. Am J Phys 257, R494 (1989). 44. Saragovi HU, Fitzpatrick D, Raktabutr A, Nakanishi H, Kahn M, Greene MI: Design and synthesis of a mimetic from an antibody complementarity-determining region. Science 253, 792 (1991).