Brush border membrane binding properties of Bacillus thuringiensis Vip3A toxin to Heliothis virescens and Helicoverpa zea midguts

BBRC Biochemical and Biophysical Research Communications 339 (2006) 1043–1047 www.elsevier.com/locate/ybbrc

Brush border membrane binding properties of Bacillus thuringiensis Vip3A toxin to Heliothis virescens and Helicoverpa zea midguts Mi Kyong Lee *, Paul Miles, Jeng-Shong Chen Syngenta Biotechnology, Inc., 3054 Cornwallis Road, Research Triangle Park, NC 27709, USA Received 1 November 2005 Available online 1 December 2005

Abstract The binding properties of Vip3A, a new family of Bacillus thuringiensis insecticidal toxins, have been examined in the major cotton pests, Heliothis virescens and Helicoverpa zea. Vip3A bound specifically to brush border membrane vesicles (BBMV) prepared from both insect larval midguts. In order to examine the cross-resistance potential of Vip3A to the commercially available Cry1Ac and Cry2Ab2 toxins, the membrane binding site relationship among these toxins was investigated. Competition binding assays demonstrated that Vip3A does not inhibit the binding of either Cry1Ac or Cry2Ab2 and vice versa. BBMV protein blotting experiments showed that Vip3A does not bind to the known Cry1Ac receptors. These distinct binding properties and the unique protein sequence of Vip3A support its use as a novel insecticidal agent. This study indicates a very low cross-resistance potential between Vip3A and currently deployed Cry toxins and hence supports its use in an effective resistance management strategy in cotton. Ó 2005 Elsevier Inc. All rights reserved. Keywords: Vip3A; Cry1Ac; Cry2Ab; Brush border membrane vesicles; Bacillus thuringiensis; Heliothis virescens; Helicoverpa zea; Insect resistant management

Bacillus thuringiensis (Bt) is well known for its ability to produce insecticidal crystal proteins, called d-endotoxins during its sporulation phase. In recent years, a number of secreted insecticidal proteins called vegetative insecticidal proteins (Vip) have been identified during the vegetative growth phase [1–3]. These Vip proteins have a broad insecticidal spectrum that includes a wide variety of lepidopteran and coleopteran species. Vip3A toxin shows a high level of activity against a range of lepidopteran pests including Agrotis ipsilon, Spodoptera fugiperda, Spodoptera exigua, Heliothis virescens, and Helicoverpa zea. One of the most important features of Vip3A is that it shares no sequence homology with known Bt d-endotoxins or other toxin genes [1]. In a previous study, the mode of action of Vip3A was investigated and compared to that of Cry1Ab d-endotoxin [4]. The 88 kDa full-length Vip3A toxin was shown to be

*

Corresponding author. Fax: +1 919 541 8585. E-mail address: [email protected] (M.K. Lee).

0006-291X/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.11.112

proteolytically activated to a 62 kDa toxin core by either trypsin or gut juice extracts in a similar manner to Cry1A toxins. The study showed that activated Vip3A toxin can bind to brush border membrane vesicles (BBMV) prepared from Manduca sexta in a competitive manner and it does not inhibit the binding of Cry1Ab toxin and vice versa. Furthermore, Vip3A toxin does not bind to the well-known Cry1A toxin receptors, 120 kDa M. sexta aminopeptidase N (APN)-like and 250 kDa cadherin-like molecules, whereas Vip3A does bind to distinct 80 and 100 kDa BBMV proteins. Ion channel analysis demonstrated that the activated Vip3A toxin formed stable ion channels which differed considerably in their principal conductance state and cation specificity from those formed as a result of the action of Cry1Ab. The mode of action of Vip3A thus appears entirely unique, and this supports its use as a novel insecticidal agent. Transgenic varieties expressing two classes of Bt d-endotoxins, Cry1Ac and Cry2Ab, have been introduced into cotton for pest management [5,6]. The X-ray crystal structures of Cry1Aa and Cry2Aa, which have a high sequence

1044

M.K. Lee et al. / Biochemical and Biophysical Research Communications 339 (2006) 1043–1047

homology with Cry1Ac and Cry2Ab, respectively, have been resolved [7,8]. The Cry1Aa toxin is composed of three domains. Domain I consists of 7a helices and is responsible for pore formation, domain II consists of three antiparallel b-sheets and is involved in receptor binding, and domain III consists of two twisted, antiparallel b-sheets and is important in receptor binding and pore formation. Cry2Aa toxin maintains this three domain structure but differs from Cry1Aa toxin in the receptor binding epitopes [8]. While direct structural information is lacking for Vip3A, the primary sequence divergence and an examination of the predicted secondary structure give no indication of a similar domain organization or a putative a-helical bundle region within the polypeptide sequence, as it exists for the Cry toxins. Protein folding blasts revealed that Vip3A might be a pore-forming protein that has the structure of b-barrels (unpublished data). This total lack of sequence and structural homology of Vip3A toxin with other Bt toxins, and its unique binding properties strongly support the hypothesis that cross-resistance to other known Cry1A toxins in the target pests is highly unlikely. Binding of the Bt toxins to the receptors located on the epithelial brush border membrane has been identified as a critical step in their action and also modification of this binding has been identified as a likely mechanism of resistance [9,10]. In order to further investigate the cross-resistance potential of Vip3A to the commercially available Bt toxins for the cotton pests, Cry1Ac and Cry2Ab2, we have examined the membrane binding properties and binding site relationships among these toxins to two cotton pests, H. virescens and H. zea. Materials and methods Preparation of toxins. Vip3A toxin was overexpressed in Escherichia coli and purified as described previously [11]. To obtain the truncated Vip3A toxin (ca. 62 kDa), the 88 kDa full-length toxin was incubated with 2% trypsin (w/w) in PBS buffer, pH 7.4, at 37 °C for 1 h. Alternatively, lepidopteran gut juice extracts collected from H. virescens and H. zea were used to obtain a truncated Vip3A toxin. Gut juice was collected by gently inducing regurgitation and then centrifuged at 13,500g for 10 min at 4 °C. The supernatant (gut juice extracts) was diluted 50-fold with PBS, added to the Vip3A-full length toxin, and incubated for 1 h at 29 °C. Complete protease inhibitor cocktail (1 ll, resuspended as per manufacturerÕs instructions, Roche Molecular Biochemicals, Indianapolis, IN) was added to stop the reaction. Cry1Ac toxin in pBluescript KS was expressed in E. coli and the inclusion body protein was solubilized in 50 mM Na2CO3, 10 mM DTT, pH 9.5. The solubilized full-length Cry1Ac protoxin was digested with 2% trypsin (w/w) at 37 °C for 2 h as described previously [12]. Cry2Ab2 toxin in pTrcHisB vector was expressed in E. coli and the soluble fraction containing Cry2Ab protein was further purified using Bio-Rad His tag purification kit as described in the instruction manual. Purity and stability of toxins were assessed on 8–12% gradient Tris–Glycine Novex gels with Mark12 molecular weight marker (Invitrogen, Carlsbad, CA). Biotinylation of toxins. Vip3A-full length, trypsin activated Vip3A, trypsin activated Cry1Ac, Cry2Ab2 full length, and trypsin activated Cry2Ab2 toxins were dialyzed against 0.1 M borate buffer, pH 8.0, and then biotinylated using a Biotin labeling kit (Roche Molecular Biochemicals) as per the manufacturerÕs instructions. Non-reacted biotin-7-NHS reagent was removed by gel filtration on a prepared Sephadex G-25 column as per the manual instructions (Roche Molecular Biochemicals).

BBMV binding assays. BBMV were prepared from fourth instar H. virescens and H. zea larval midguts by the differential magnesium precipitation method as described in Wolfersberger et al. [13]. The final pellet was resuspended in the binding buffer, consisting of 8 mM NaHPO4, 2 mM KH2PO4, and 150 NaCl (pH 7.4), and protein concentration was measured using the Coomassie Protein Assay Reagent (Pierce Biotechnology, Rockford, IL). For qualitative estimation of competitive binding, 2 nM biotinylated toxin was incubated with 5 lg BBMV in the binding buffer with 0.1% BSA in the presence or absence of increasing amounts of the unlabeled toxin (2.5- to 25-fold excess). After 1 h incubation at RT, reaction mixtures were centrifuged at 13,500g for 10 min. The pellets were washed three times with binding buffer containing 0.1% BSA. Final pellets were resuspended in the binding buffer without BSA and samples were run on 8–16% gradient, Tris–Glycine Novex gels (Invitrogen, Carlsbad, CA). Proteins were transferred to Novex polyvinylidene difluoride (PVDF) membrane (Invitrogen) and probed with streptavidin-conjugated peroxidase (Roche Molecular Biochemicals) and the SuperSignal West Pico Chemiluminescence kit (Pierce) for visualization. BBMV ligand competition binding assays. BBMV proteins (5 lg) prepared from H. virescens and H. zea were separated via SDS–PAGE (4– 12% gradient) and transferred to PVDF membrane. Five nanomole of the biotinylated toxin was incubated with the membrane for 2 h at room temperature in the absence or presence of unlabeled toxins. The membrane was washed three times with TBS buffer (20 mM Tris, 500 mM NaCl, pH 7.5) containing 0.05% Tween 20 and Triton X-100. The toxin-binding proteins were visualized with streptavidin-conjugated peroxidase (Roche Molecular Biochemicals) and the SuperSignal West Pico Chemiluminescence kit (Pierce). SeeBlue Plus2 (Invitrogen) pre-stained standard protein marker was used.

Results and discussions One of the key elements in the mode of action of Bt toxin is its binding to specific receptor(s) on the surface of the midgut epithelial brush border membrane. Binding affinity, binding site concentrations, and irreversible binding phenomena have been demonstrated as critical factors in determining insecticidal activity and specificity. In in vitro BBMV binding studies with susceptible and resistant H. virescens strains, alterations in the receptor binding properties have been observed in the resistant strains [14–17], indicating that receptor binding is a critical factor in resistance development. In order to prevent or delay the development of resistance, a multiple toxin approach has been suggested [18,19]. For this to be effective there must be no cross-resistance between the individual toxins. The specificity or cross-reactivity of the Bt toxins and receptors must therefore be clearly understood. Cross-resistance is believed to be less likely to occur when two or more toxins do not share the same receptors. Recently, two stacked insecticidal gene cotton varieties have been introduced. MonsantoÕs Bollgard II cotton expresses the two Bt d-endotoxins, Cry1Ac and Cry2Ab2 [5,6]. Based on the low sequence homology between these toxins (20%) and their different binding properties, it was claimed that they have different modes of action and that they can therefore be used to provide effective resistance management. Similarly, Dow Agrosciences WideStrike cotton expresses two Bt d-endotoxins, Cry1Ac and Cry1F, for a broad range of lepidopteran control [20].

M.K. Lee et al. / Biochemical and Biophysical Research Communications 339 (2006) 1043–1047

Our recent studies [4] on the mode of action of Vip3A toxin demonstrated that Vip3A has a unique amino acid sequence, and receptor binding properties and pore forming characteristics that are distinctly different to those of Cry1Ab toxin [4]. Thus, the unique mode of action of Vip3A suggests that it will have a valuable role as a novel insecticidal agent in IRM strategies. To further support this view, the possibility for cross-resistance between Vip3A and the endotoxins Cry1Ac and Cry2Ab2 in the two major cotton pests, H. virescens and H. zea, has been examined here. A series of in vitro receptor binding assays were performed to demonstrate the potential for cross-resistance among these three toxins. Proteolytic activation of toxins Vip3A full-length toxin (88 kDa) is further processed to a 62 kDa core fragment (N terminus at 199) in in vivo digestion in the lepidopteran midgut or in vitro digestion with either trypsin or gut juice extracts as described previously [4] (Fig. 1A). Cry1Ac protoxin (130 kDa) can be easily activated with either trypsin or lepidopteran gut juice extracts to 65 kDa core protein (Fig. 1B). Trypsin activated Vip3A and Cry1Ac toxins were used in the BBMV binding assays. In the EPAÕs Biopesticide Registration Action Document on Bollgard II Cotton [5], it is claimed that Cry2Ab2 does not have a trypsin- or chymotrypsin-resistant core. In contrast to this, we have found a 50 kDa core fragment of Cry2Ab2 by either trypsin (Fig. 1C, lane 9) or midgut extracts from H. virescens and H. zea (Fig. 1D, lanes 10 and 11, respectively). Both forms of Cry2Ab2 were used for BBMV binding assays. Digestion of full-length Vip3A with gut juice extracts from non-susceptible larvae of Ostrinia nubilalis also produced a resistant core, suggesting that proteolysis alone is not a determining factor for insecticidal activity.

1

2

3

4 200

64

BBMV competition binding properties Homologous and heterologous BBMV competition binding assays with H. virescens and H. zea were performed to demonstrate the binding site relationships among Vip3A, Cry1Ac, and Cry2Ab2 toxins. Biotin labeled Vip3A toxin bound to both BBMVs in a competitive fashion, where an excess of unlabeled Vip3A toxin significantly reduced the signal of the labeled Vip3A toxin (Fig. 2, lanes 1–4). No competitive binding was observed between Vip3A and Cry1Ac toxins. Excess amounts of cold Cry1Ac toxin (25 times) did not inhibit the binding of labeled Vip3A toxin (Fig. 2, lane 5), indicating that Cry1Ac does not bind to the Vip3A binding sites. In addition, no competitive binding was observed between Vip3A toxin and either full length or truncated Cry2Ab2 toxin. Excess amounts of cold Cry2Ab2 full-length toxin (25 times) did not inhibit the binding of labeled Vip3A, indicating that Cry2Ab2 does not bind to the Vip3A receptor (Fig. 2, lane 6).

Fig. 2. Binding of biotinylated trypsin activated Vip3A toxin to H. virescens BBMV. Two nanomolar of biotinylated toxin was incubated with 5 lg BBMV in the absence (lane 1) and the presence of a 2.5-, 10-, and 25-fold of unlabeled Vip3A toxin (lanes 2, 3, and 4, respectively), 25-fold of unlabeled Cry1Ac (lane 5), and 25-fold of unlabeled Cry2Ab2 toxin (lane 6).

6

7

9

10 250

116 97

148

148

66

98

98

64

64

50

50

36

36

51

55

39

36 31

28

8

250

191 97

5

21

1045

11

12

Fig. 1. Proteolytic activation of Vip3A full-length toxin, Cry1Ac protoxin, and Cry2Ab2 protoxin (lanes 2, 5, and 8, respectively) by trypsin digestion. Full-length Vip3A toxin and Cry1Ac protoxin were digested with 2% trypsin (w/w) for 1 h at 37 °C (lanes 3 and 6, respectively). Cry2Ab2 protoxin was digested with either 2% trypsin (w/w) for 1 h at 37 °C (lane 9) or digested with gut juice extracts from H. virescens and H. zea for 1 h at 37 °C (lanes 11 and 12, respectively). Lanes 1, 4, 7, and 10 represent molecular markers (Invitrogen).

1046

M.K. Lee et al. / Biochemical and Biophysical Research Communications 339 (2006) 1043–1047

BBMV ligand blotting assays

Fig. 3. Binding of biotinylated trypsin activated Cry1Ac toxin to H. virescens BBMV. Two nanomolar of biotinylated toxin was incubated with 5 lg BBMV in the absence (lane 1) and the presence of 25-fold of unlabeled Cry1Ac (lane 2) and Vip3A toxin (lane 3).

Biotin labeled Cry1Ac toxin bound to both BBMV competitively as described previously (Fig. 3). Excess amounts of unlabeled Cry1Ac (25 times) significantly reduced binding (lane 2) of the labeled Cry1Ac toxin. However, no competitive binding between labeled Cry1Ac and Vip3A was observed. Excess cold Vip3A toxin (25 times) did not inhibit the binding of labeled Cry1Ac toxin, indicating that Vip3A does not bind to the Cry1Ac binding sites (lane 3). Compared to the Vip3A and Cry1Ac binding to BBMV, labeled Cry2Ab2 toxin showed much less binding to H. virescens and H. zea BBMV and the binding was not specific. Biotin labeled full-length Cry2Ab2 was not chased off by increasing amounts of unlabeled Cry2Ab2 toxin (data not shown). Cry2Aa is highly homologous to Cry2Ab2 and its non-specific binding properties have also been previously demonstrated by English et al. [21] in H. virescens. More recently, unpublished work using BIAcore 2000 also showed non-specific binding of Cry2Ab2 toxin to BBMV proteins [5]. In the present study, a truncated Cry2Ab2 toxin was also used for the BBMV binding study and it also exhibited non-specific binding. This non-specific binding of the Cry2Ab2 toxin was not inhibited by addition of unlabeled Vip3A toxin, indicating Vip3A does not bind to the Cry2Ab2 non-specific binding sites. In summary, BBMV competition binding assays demonstrated that Vip3A toxin does not share binding sites with either Cry1Ac or Cry2Ab2 toxins in the two Heliothine cotton pests.

BBMV ligand blotting assays were performed to identify putative toxin binding proteins (receptors) on BBMV proteins. Previously, several putative Cry1Ac toxin-binding proteins in H. virescens have been reported (80, 120, 150, 170, and >210 kDa) using BBMV ligand blotting technique [14,22,23]. In the present study, we have observed a similar pattern. In H. virescens, two major peptides were identified as Cry1Ac binding proteins (120 and 150–170 kDa) (Fig. 4A). Biotin labeled Vip3A toxin does not show any binding to these peptides (data not shown). In contrast, excess amounts of cold Cry1Ac toxin competed for the binding of labeled Cry1Ac toxin to these peptides, indicating competitive and specific binding. Moreover, excess amounts of cold Vip3A toxin did not inhibit binding of Cry1Ac to these peptides. In H. zea, more complex binding patterns were observed (Fig. 4B). Previously, three major peptides have been identified as Cry1Ac binding proteins (120, 150, and >210 kDa) and a similar pattern has been observed in this study. In addition to these major peptides, several other peptides have shown to interact with Cry1Ac toxin. As observed in the H. virescens ligand blotting experiment, Vip3A does not bind to any of these peptides (data not shown) and unlabeled Vip3A toxin does not chase off the binding of Cry1Ac toxin. These ligand blotting data further confirm the BBMV competition study. In summary, the in vitro binding studies presented here demonstrate that Vip3A toxin does not bind to the same receptors as Cry1A. This lack of binding of Vip3A to all the known Cry1Ac receptors strongly supports the use of Vip3A as a novel insecticidal agent. Furthermore, in a recent study the unique pore-forming properties of Vip3A have been demonstrated using a voltage clamping study with isolated midgut and planar lipid bilayers [4]. Vip3A channels are shown to be structurally and functionally distinct from those arising from Cry1A toxin action. In summary, the unique protein sequence and distinct receptor binding properties and ion channel properties of Vip3A strongly support the view that the likelihood of a target site cross-resistance between Vip3A and the Cry toxins arising as a result of a mutation in the midgut receptors

A: H. virescens 1 150kDa 120kDa

2

B.H. zea 3

210kDa 150kDa 120kDa

150kDa 120kDa

1

2

3

210kDa 150kDa 120kDa

Fig. 4. H. virescens and H. zea BBMV ligand blotting with biotinylated Cry1Ac toxin in the absence (lane 1) and presence of unlabeled Cry1Ac (lane 2) and Vip3A (lane 3) toxin. BBMV proteins (5 lg) were separated onto SDS–PAGE (4–12% gradient) and transferred to a PVDF membrane. Binding proteins were probed with 5 nM biotinylated Cry1Ac toxin.

M.K. Lee et al. / Biochemical and Biophysical Research Communications 339 (2006) 1043–1047

is extremely remote. These findings also highlight the benefits of Vip3A in insecticide resistance management strategies and support the deployment of SyngentaÕs VipCot cotton varieties expressing Vip3A.

[12]

[13]

Acknowledgments We thank Jason Barnett for excellent technical assistance on cloning and protein purification, Fred Walters for dissecting midguts and the critical comments on the manuscript, and Alan McCaffery for the suggestions on experimental designs and his critical comments on the manuscript.

[14]

[15]

References [16] [1] J.J. Estruch, G.W. Warren, M.A. Mullins, G.J. Nye, J.A. Craig, M.G. Koziel, Vip3A, a novel Bacillus thuringiensis vegetative insecticidal protein with a wide spectrum of activities against lepidopteran insects, Proc. Natl. Acad. Sci. USA 93 (1996) 5389–5394. [2] J.J. Estruch, C.-G. Yu, Plant pest control. U.S. patent 6,291,156 B1, 2001. [3] G.W. Warren, Vegetative insecticidal proteins: novel proteins for control of corn pests, in: N. Carozzi, M. Koziel (Eds.), Advances in Insect Control, Taylor & Francis, Bristol, PA, 1997, pp. 109–121. [4] M.K. Lee, F.S. Walters, H. Hart, N.P. Palekar, J.-S. Chen, The mode of action of the Bacillus thuringiensis vegetative insecticidal protein Vip3A differs from that of Cry1Ab d-endotoxin, Appl. Environ. Microbiol. 69 (2003) 4648–4657. [5] U.S. Environmental Protection Agency, 2002. . [6] Office of the Gene Technology Regulator, 2002. . [7] P. Grochulski, L. Masson, S. Borisova, M. Pusztai-Carey, J.-L. Schwartz, R. Brousseau, M. Cygler, Bacillus thuringiensis Cry1A(a) insecticidal toxin: crystal structure and channel formation, J. Mol. Biol. 254 (1995) 447–464. [8] R.J. Morse, T. Yamamoto, R.M. Stroud, Structure of Cry2Aa suggests and unexpected receptor binding epitope, Structure 9 (2001) 409–417. [9] E. Schnepf, N. Crickmore, J. Van Rie, D. Lereclus, J. Baum, J. Feitelson, D.R. Ziegler, D.H. Dean, Bacillus thuringiensis and its pesticidal crystal proteins, Mirobiol. Mol. Biol. Rev. 62 (1998) 775–806. [10] J. Ferre´, J. Van Rie, Biochemistry and genetics of insect resistance to Bacillus thuringiensis, Annu. Rev. Entomol. 47 (2002) 501–533. [11] C.-G. Yu, M.A. Mullins, G.W. Warren, M.G. Koziel, J.J. Estruch, The Bacillus thuringiensis vegetative insecticidal protein Vip3A lyses

[17]

[18]

[19]

[20]

[21]

[22]

[23]

1047

midgut epithelium cells of susceptible insects, Appl. Environ. Microbiol. 63 (1997) 532–536. M.K. Lee, R.E. Milne, A.Z. Ge, D.H. Dean, Location of a Bombyx mori receptor binding region on a Bacillus thuringiensis d-endotoxin, J. Biol. Chem. 267 (1992) 3115–3121. M. Wolfersberger, P. Luthy, A. Maurer, P. Parenti, P.V. Sacchi, B. Giordana, Preparation of brush border membrane vesicles (BBMV) from larval lepidopteran midgut, Comp. Biochem. Physiol. 86A (1987) 301–308. M.K. Lee, F. Rajamohan, F. Gould, D.H. Dean, Resistance to Bacillus thuringiensis Cry1A d-endotoxins in a laboratory-selected Heliothis virescens strain is related to receptor alteration, Appl. Environ. Microbiol. 61 (1995) 3836–3842. J.L. Jurat-Fuentes, F.L. Gould, M.J. Adang, Altered glycosylation of 63- and 68-kilodalton microvillar proteins in Heliothis virescens correlates with reduced Cry1 toxin binding, decreased pore formation, and increased resistance to Bacillus thuringiensis Cry1 toxins, Appl. Environ. Microbiol. 68 (2002) 5711–5717. J.L. Jurat-Fuentes, F.L. Gould, M.J. Adang, Dual resistance to Bacillus thuringiensis Cry1Ac and Cry2Aa toxins in Heliothis virescens suggests multiple mechanisms of resistance, Appl. Environ. Microbiol. 69 (2003) 5898–5906. J.L. Jurat-Fuentes, L.J. Gahan, F.L. Gould, D.G. Heckel, M.J. Adang, The HevCaLP protein mediates binding specificity of the Cry1A class of Bacillus thuringiensis toxins in Heliothis virescens, Biochemistry 43 (2004) 14299–14305. F. Gould, Sustainability of transgenic insecticidal cultivars: integrating pest genetics and ecology, Annu. Rev. Entomol. 43 (1998) 701– 726. R.T. Roush, Two-toxin strategies for management of insecticidal transgenic crops; can pyramiding succeed where pesticide mixtures have not? Phil. Trans. R. Soc. Lond. B 353 (1998) 1777–1786. J.J. Adamczyk, J. Gore, Laboratory and field performance of cotton containing Cry1Ac, Cry1F, and both Cry1Ac and Cry1F (WideStrikeÒ) against beet armyworm and fall armyworm larvae (Lepidoptera: Noctuidae), Fla Entomol. 87 (2004) 427–432. L. English, H.L. Robbins, M.A. Von Tersch, C.A. Kulesza, D. Ave, D. Coyle, C.S. Jany, S. Slatin, Mode of action of CryIIA: a Bacillus thuringiensis delta-endotoxin, Insect. Biochem. Mol. Biol. 24 (1994) 1025–1035. K. Luo, S. Sangadala, L. Masson, A. Mazza, R. Brousseau, M.J. Adang, The Heliothis virescens 170 kDa aminopeptidase functions as ‘‘receptor A’’ by mediating specific Bacillus thuringiensis d-endotoxin binding and pore formation, Insect Biochem. Mol. Biol. 27 (1997) 735–743. D.J. Banks, J.L. Jurat-Fuentes, D.H. Dean, M.J. Adang, Bacillus thuringiensis Cry1Ac and Cry1Fa d-endotoxin binding to a novel 110 kDa aminopeptidase in Heliothis virescens is not N-acetylgalactosamine mediated, Insect Biochem. Mol. Biol. 31 (2001) 909– 918.