Báo cáo Y học: Location of the Bombyx mori 175 kDa cadherin-like protein-binding site on Bacillus thuringiensis Cry1Aa toxin doc

We discuss the location of the BtR175-bind-ing site on Cry1Aa toxin on the basis of location infor-mation obtained with the epitopes of mAbs that block the binding sites for BtR175 and B

protein-binding site on Bacillus thuringiensis Cry1Aa toxin Shogo Atsumi1,2, Yukino Inoue1, Takahisa Ishizaka1, Eri Mizuno1, Yasutaka Yoshizawa1,

Madoka Kitami1and Ryoichi Sato1

1 Graduate School of Bio-Applications and Systems Engineering, Tokyo University of Agriculture and Technology, Japan

2 Division of Insect Sciences, National Institute of Agrobiological Sciences, Owashi, Tsukuba, Ibaraki, Japan

Bacillus thuringiensis, a Gram-positive bacterium,

pro-duces various insecticidal proteins called Cry toxins

These bacteria are used as microbial insecticides and

for the genetic development of insect-resistant plants,

because they are specific to their target insects Cry

toxins are expressed in inclusion bodies as protoxins

(70–140 kDa) during sporulation When a protoxin is

ingested by the target insect, it is solubilized in the insect midgut and digested by proteolytic enzymes [1] After enzymatic activation, the toxic protease-resistant fragment, which is the 60–70 kDa activated toxin, binds to specific receptors located in the columnar cells of the midgut apical brush border membrane [2]

Keywords

Bacillus thuringiensis; BmAPN1;

Bombyx mori; BtR175; Cry1Aa

Correspondence

R Sato, Graduate School of Bio-Applications

and Systems Engineering, Tokyo University

of Agriculture and Technology, Koganei,

Tokyo 184 8588, Japan

Fax: +81 42 388 7277

Tel: +81 42 388 7277

E-mail: ryoichi@cc.tuat.ac.jp

(Received 9 June 2008, revised 25 July

2008, accepted 8 August 2008)

doi:10.1111/j.1742-4658.2008.06634.x

To identify and gain a better understanding of the cadherin-like receptor-binding site on Bacillus thuringiensis Cry toxins, it is advantageous to use Cry1Aa toxin, because its 3D structure is known Therefore, Cry1Aa toxin was used to examine the locations of cadherin-like protein-binding sites Initial experiments examining the binding compatibility for Cry1Aa toxin

of partial fragments of recombinant proteins of a 175 kDa cadherin-like protein from Bombyx mori (BtR175) and another putative receptor for Cry1Aa toxin, aminopeptidase N1, from Bo mori (BmAPN1), suggested that their binding sites are close to each other Of the seven mAbs against Cry1Aa toxin, two mAbs were selected that block the binding site for BtR175 on Cry1Aa toxin: 2A11 and 2F9 Immunoblotting and alignment analyses of four Cry toxins revealed amino acids that included the epitope

of mAb 2A11, and suggested that the area on Cry1Aa toxin blocked by the binding of mAb 2A11 is located in the region consisting of loops 2 and

3 Two Cry1Aa toxin mutants were constructed by substituting a Cys on the area blocked by the binding of mAb 2A11, and the small blocking mol-ecule, N-(9-acridinyl)maleimide, was introduced at each Cys substitution to determine the BtR175-binding site Substitution of Tyr445 for Cys had a crippling effect on binding of Cry1Aa toxin to BtR175, suggesting that Tyr445 may be in or close to the BtR175-binding site Monoclonal anti-bodies that blocked the binding site for BtR175 on Cry1Aa toxin inhibited the toxicity of Cry1Aa toxin against Bo mori, indicating that binding of Cry1Aa toxin to BtR175 is essential for the action of Cry1Aa toxin on the insect

Abbreviations

ACN, aminopeptidase N; BmAPN1, Bombyx mori aminopeptidase N1; BtR175, Bombyx mori 175 kDa cadherin-like protein; GST 27 kDa BtR175, a recombinant fusion protein of glutathione S-transferase and the 27 kDa fragment of Bombyx mori 175 kDa cadherin-like protein; GST 7 kDa BmAPN1, a recombinant fusion protein of glutathione S-transferase and the 7 kDa fragment of Bombyx mori aminopeptidase N; GST, glutathione S-transferase; HRP, horseradish peroxidase; NAM, N-(9-acridinyl)maleimide.

The 3D structures of Cry1Aa toxin have been

deter-mined by X-ray diffraction crystallography [3] This

protein is composed of three domains Domain I is

composed of a seven-a-helix bundle and is involved in

membrane insertion [3,4] Domain II consists of three

antiparallel b-sheets and plays a role in binding to

receptor molecules [3,5,6] Domain III is a b-sandwich

and has several functions, including the recognition of

N-acetylgalactosamine on receptor molecules [3,7–9]

In early studies, various aminopeptidase N (APN)

isoforms from several insect species were identified as

candidate receptors for B thuringiensis Cry toxins [10–

14] Expression of an APN in Drosophila melanogaster

made the larvae sensitive to Cry1Ac toxin [15] RNA

interference experiments have indicated that silencing

of midgut APN in Spodoptera litura and

Heli-coverpa armigera reduces sensitivity to Cry1C and

Cry1Ac toxins [16,17] The 170 kDa APN from

Helio-this virescens plays a role in pore formation in

membrane vesicles [18] The 120 kDa APN from

Manduca sexta mediates channel formation in planar

lipid bilayers [19] These results suggest that APN

functions as a Cry1 receptor and is involved in the

lytic activity of Cry1 toxins In addition, cadherin-like

proteins are selected as candidate receptors [20–23]

Expression of the Bombyx mori 175 kDa cadherin-like

protein (BtR175) on the surface of Sf9 insect cells

made these cells sensitive to Cry1Aa toxin [12] Our

previous work indicated inhibition of the binding of

Cry1Aa and Cry1Ac toxins to BtR175 after

pretreat-ment with antibody against BtR175, as this suppressed

the lytic activity of the toxins on

collagenase-dissoci-ated Bo mori midgut epithelial cells [24] On genetic

mapping analysis, the disruption of a cadherin-like

protein gene (BtR-4) by retrotransposon-mediated

insertion was shown to be linked to high levels of

resis-tance to Cry1Ac toxin in H virescens [25] RNA

inter-ference experiments have indicated that silencing of

midgut BT-R1 in M sexta reduces sensitivity to

Cry1Ab toxin [26] These findings suggest that the

cadherin-like protein plays an important role in Cry

toxin susceptibility and that the cadherin-like protein

is the functional Cry toxin receptor in the insect

midgut

One hypothesis regarding the mode of action of Cry

toxin for M sexta is that binding of monomeric

Cry1Ab toxin to BT-R1 promotes additional

proteo-lytic cleavage in the N-terminal end of the toxin,

facili-tating the formation of a prepore oligomeric structure

that is competent for membrane insertion, and that

oligomer formation is important for toxicity [27,28]

The prepore oligomer has a higher affinity for APN

[29,30] The oligomeric Cry1A toxin structure then

binds to the APN receptor, leading to its insertion into membrane lipid rafts, implying a sequential binding mechanism of Cry1A toxins with BT-R1 and APN receptor molecules [30,31] However, a different mech-anism of action of Cry toxins was recently proposed, based on a study of the effect of Cry1Ab toxin on cul-tured Trichoplusia ni H5 insect cells expressing M

sex-ta BT-R1 [32,33], and was called the signaling model However, modified Cry1A toxins lacking helix 1 formed a prepore oligomeric structure and killed insects without needing intact BT-R1[26] Although no consensus regarding the mode of action of Cry toxins has yet been reached, this result favors the membrane insertion model

The Cry toxin-binding site on cadherin-like protein receptors and the cadherin-like protein receptor-bind-ing site on Cry toxins are bereceptor-bind-ing mapped Three Cry1A toxin-binding sites have been mapped in BT-R1 The first site, NITIHITDTNN(865–875), was mapped using phage display and is involved in binding loop 2 (b6–b7 loop) of Cry1Aa and Cry1Ab toxins [5,34] A second region, 1291–1360, is important for toxin bind-ing [35], and was subsequently narrowed down to a 12 amino acid region, IPLPASILTVTV(1331–1342), which binds loop a8 (a8a–a8b loop) [6] A third region, 1363–1464, is involved in toxin binding and cytotoxicity [36] In the case of the H virescens cadh-erin, this binding region was narrowed down to a 19 amino acid region, 1422–1440, which binds domain II loop 3 (b10–b11 loop) of Cry1Ab and Cry1Ac toxins [23] In Bo mori, a 219 amino acid region (1245–1464)

of BtR175 is responsible for Cry1Aa toxin binding [37] However, the binding site on Cry1Aa toxin for BtR175 is not yet known It is possible to consider the structure of the cadherin-binding site three-dimension-ally, because the 3D structure of Cry1Aa toxin is known Three-dimensional visualization is very useful for understanding the mechanism of selective toxin– receptor interactions

We analyzed the binding site on Cry1Aa toxin for one of the Cry1Aa toxin receptors in Bo mori, a

115 kDa APN type 1 (BmAPN1), using mAbs that block binding between the binding site and the recep-tor [38] In this study, to analyze the binding site on Cry1Aa toxin for the other Cry1Aa toxin receptor, BtR175, using mAbs that block binding between the binding site and the receptor, mAbs that block the binding site for BtR175 on Cry1Aa toxin were selected These antibodies also reduce the toxicity of Cry1Aa toxin against Bo mori Monoclonal antibodies that block the binding site for BmAPN1 on Cry1Aa toxin have no effect on the toxicity of Cry1Aa toxin against Bo mori We found that the binding of

BmAPN1 to Cry1Aa toxin blocked the binding of

BtR175 We discuss the location of the

BtR175-bind-ing site on Cry1Aa toxin on the basis of location

infor-mation obtained with the epitopes of mAbs that block

the binding sites for BtR175 and BmAPN1

Results

Competitive binding of glutathione S-transferase

(GST) 27 kDa BtR175 and GST 7 kDa BmAPN1 to

Cry1Aa toxin

A recombinant fusion protein of GST and 7 kDa

frag-ment of BmAPN1 containing the Cry1Aa

toxin-bind-ing region (GST 7 kDa BmAPN1) [39] and a

recombinant fusion protein of GST and 27 kDa

frag-ment of BtR175 containing the Cry1Aa toxin-binding

region (GST 27 kDa BtR175) [37] were tested for their

ability to block binding of Cry1Aa toxin to GST

27 kDa BtR175 GST 7 kDa BmAPN1 and GST

27 kDa BtR175 blocked the binding of Cry1Aa toxin

to GST 27 kDa BtR175 in a dose-dependent manner

(Fig 1) These results suggest that the surface of

Cry1Aa toxin blocked by GST 7 kDa BmAPN1 is located close to the GST 27 kDa BtR175-binding site

Determination of binding compatibility for GST

27 kDa BtR175, GST 7 kDa BmAPN1, and mAb 1B10, using an IAsys optical sensor

To investigate the relative locations of the binding sites for GST 27 kDa BtR175 and GST 7 kDa BmAPN1

on the surface of Cry1Aa toxin, tests were performed using an IAsys resonant mirror optical biosensor GST

27 kDa BtR175, GST 7 kDa BmAPN1 or mAb 1B10 was bound to Cry1Aa toxin immobilized on a cuvette surface, as demonstrated by the equal heights of the sensorgrams (Fig 2A,B,F) To test for binding compe-tition with the same protein, a solution of GST 7 kDa BmAPN1 was added to the cuvette after GST 7 kDa BmAPN1 had been bound to the immobilized Cry1Aa toxin, and the binding of the added protein to the sur-face was determined (Fig 2C) When additional GST

7 kDa BmAPN1 was added, almost no further binding

to the immobilized Cry1Aa toxin occurred, suggesting that the binding sites for GST 7 kDa BmAPN1 on the surface had been largely blocked by the initial GST

7 kDa BmAPN1 solution This type of binding curve, which demonstrates that there was no additional pro-tein binding, is observed if the second propro-tein (in this case, the same protein) competes with the first protein

To test for binding competition between BmAPN1 and BtR175, GST 27 kDa BtR175 was added to the cuv-ette after GST 7 kDa BmAPN1 had been bound to the immobilized Cry1Aa toxin, and the binding of the added protein to the surface was determined (Fig 2D) When additional GST 27 kDa BtR175 was added, almost no further binding to the immobilized Cry1Aa toxin occurred, suggesting that the binding sites for GST 27 kDa BtR175 had been largely blocked by the initial solution of GST 7 kDa BmAPN1 To test for binding competition in the reverse order, a solution of GST 7 kDa BmAPN1 was added to the cuvette after GST 27 kDa BtR175 had been bound to the immobi-lized Cry1Aa toxin, and the binding of the added pro-tein to the surface was determined (Fig 2E) When GST 7 kDa BmAPN1 was added, almost no further binding to the immobilized Cry1Aa toxin occurred, suggesting that the binding sites for GST 7 kDa BmAPN1 had been largely blocked by the initial solu-tion of GST 27 kDa BtR175 As a control, the com-patibility of binding of BtR175 and mAb 1B10 on the surface of Cry1Aa toxin at the same time was tested Monoclonal antibody 1B10 was raised against Cry1Aa toxin in a BALB⁄ c mouse, and can block the binding

of Cry1Aa toxin to GST 7 kDa BmAPN1 [38] To test

Fig 1 Dose-dependent inhibition of Cry1Aa toxin binding to

BtR175 by GST 27 kDa BtR175 and GST 7 kDa BmAPN1

Biotiny-lated Cry1Aa toxin was preincubated with various concentrations of

GST 27 kDa BtR175 or GST 7 kDa BmAPN1 for 90 min and then

added to wells coated with GST 27 kDa BtR175 It was then

incubated for 90 min with HRP-conjugated streptavidin Bound

streptavidin was detected after incubation with

2,2¢-azinobis(3-ethyl-benzo-6-thiazolinesulfonic acid) solution for 15 min Absorbance at

415 nm was measured using a microtiter plate reader The assays

were repeated three times, and mean values are shown.

for binding competition between BtR175 and mAb

1B10, a solution of mAb 1B10 was added to the

cuv-ette after GST 27 kDa BtR175 had been bound to the

immobilized Cry1Aa toxin, and the binding of the

added protein to the surface was determined (Fig 2G)

When additional mAb 1B10 was added, antibody

binding was the same as that seen with mAb 1B10

alone (Fig 2F), suggesting that the binding sites for

mAb 1B10 had not been blocked by the initial solution

of GST 27 kDa BtR175 This type of binding curve

indicates that mAb 1B10 does not compete for binding

sites with GST 27 kDa BtR175

Monoclonal antibodies that block the binding of

Cry1Aa toxin to GST 27 kDa BtR175

Seven mAbs (1B10, 1E10, 1G10, 2A11, 2C2, 2F9, and

3C7) were raised against Cry1Aa toxin in a BALB⁄ c

mouse [38] These seven mAbs against Cry1Aa toxin

were tested for their ability to block the binding of

Cry1Aa toxin to BtR175 The ability of each mAb to

block the binding of Cry1Aa toxin to GST 27 kDa

BtR175 was investigated by preincubation of each

mAb with biotinylated Cry1Aa toxin before reaction

with GST 27 kDa BtR175 Of the seven mAbs, only

mAbs 2A11 and 2F9 blocked the binding of Cry1Aa

toxin to GST 27 kDa BtR175 The other mAbs (1B10,

1E10, 1G10, 2C2, and 3C7) had little or no effect on

the binding of Cry1Aa toxin to GST 27 kDa BtR175

(Fig 3) The concentration dependence of the blocking

effects of the two blocking mAbs was determined in a

similar experiment Monoclonal antibodies 2A11 and

2F9 blocked the binding of Cry1Aa toxin to GST

27 kDa BtR175 in a dose-dependent manner, but mAbs 1B10 and 2C2 did not block the binding of Cry1Aa toxin to GST 27 kDa BtR175 even at the highest concentration used (data not shown) These results suggest that the surface of Cry1Aa toxin

Fig 2 Determination of binding compatibility for BtR175, BmAPN1 and mAb 1B10, using the IAsys optical sensor Cry1Aa toxin was cova-lently immobilized on the carboxylated surface of a cuvette To measure the binding of receptor molecules to Cry1Aa molecules, BmAPN1 (A) or BtR175 (B) was added to the Cry1Aa-immobilized surface and the association was observed for 5 min Subsequently, each receptor molecule solution was replaced with protein-free binding buffer (NaCl ⁄ P i ), and the dissociation of binding was observed for 5 min At the end of this cycle, the NaCl ⁄ P i in the cuvette was replaced with 6 M guanidine hydrochloride to regenerate the sensor surface Next, we investigated the relative locations of the binding sites for BtR175 and BmAPN1 on the surface of Cry1Aa toxin First, BmAPN1 was added

to block its binding site on the immobilized Cry1Aa toxin Then, BmAPN1 (C) or BtR175 (D) was added as a second molecule, and the addi-tive association was observed for 5 min BtR175 was added to block its binding site on the immobilized Cry1Aa toxin Then, BmAPN1 (E) was added, and the additive association was observed for 5 min As a control, the simultaneous binding of BtR175 and mAb 1B10 on the surface of Cry1Aa toxin was examined (F, G).

Fig 3 Inhibition of Cry1Aa toxin binding to GST 27 kDa BtR175 by mAbs Biotinylated Cry1Aa toxin was preincubated with or without each mAb at 500 n M for 90 min and then added to wells coated with GST 27 kDa BtR175 After washing, it was incubated for

90 min with HRP-conjugated streptavidin Bound streptavidin was detected by incubation with 2,2¢-azinobis(3-ethylbenzo-6-thiazoline-sulfonic acid) solution for 15 min Absorbance at 415 nm was mea-sured using a microtiter plate reader The assays were repeated three times, and mean values are shown.

blocked by mAbs 2A11 and 2F9 overlaps with or is

close to the GST 27 kDa BtR175-binding site

Binding of mAbs 2A11 and 2F9 to Cry toxins

We reported previously that the epitopes of mAbs 2A11

and 2F9 on Cry1Aa toxin are located in domain II [38]

We used two other approaches for epitope mapping of

mAbs 2A11 and 2F9, immunoblotting using deletion

mutants of Cry1Aa toxin, and a binding assay using

synthetic peptides, but it was impossible to determine

details at the amino acid level using these methods [38]

As the epitopes for mAbs 2A11 and 2F9 may consist of

several discontinuous segments of polypeptide chains,

they may be conformational epitopes We looked for

Cry toxins other than Cry1Aa toxin, and used

immuno-blotting analysis to determine the epitope of each mAb

on Cry1Aa toxin We tested the binding of mAbs 2A11

and 2F9 to membrane blots of four Cry toxins

(Cry1Aa1, Cry1Ab8, Cry1Ac1, and Cry9Da2) Three

molecules, Cry1Aa1, Cry1Ab8 and Cry1Ac1 toxins, are

phylogenetically closely related, but Cry9Da2 toxin is not closely related to the others Aliquots of approxi-mately 1 lg of each toxin were subjected to SDS⁄ PAGE, transferred onto nitrocellulose mem-branes, and probed with mAbs 2A11 (Fig 4Bb) and 2F9 (Fig 4Bc) The proteins on the gels were visualized

by staining with Coomassie brilliant blue (Fig 4Ba) With Cry9Da2 toxin, a 55 kDa band (fragment) was equivalent to 1 lg Monoclonal antibody 2F9 recog-nized only Cry1Aa toxin but not Cry1Ab, Cry1Ac or Cry9Da2 toxins (Fig 4Bc) Monoclonal antibody 2A11 recognized Cry1Aa1 toxin and the 55 kDa protein of the activated Cry9Da2 toxin, but not Cry1Ab8 or Cry1Ac1 toxins (Fig 4Bb)

Comparison of the domain II regions of Cry1Aa1, Cry1Ab8, Cry1Ac1 and Cry9Da2 toxins

The sequences of Cry1Aa1 [40], Cry1Ab8 [41], Cry1Ac1 [42] and Cry9Da2 [43] toxins were compared (Fig 4A) Amino acids that were conserved in three or

A

B

Fig 4 Comparison of the domain II regions in Cry1Aa1, Cry1Ab8, Cry1Ac1 and Cry9Da2 toxins (A) and immunoblotting analysis of the reac-tivity of mAbs 2A11 and 2F9 with these toxins (B) (A) The sequences of Cry1Aa1 [40], Cry1Ab8 [41], Cry1Ac1 [42] and Cry9Da2 [43] toxins were aligned using GENETYX-WIN v 5.0.0 software Amino acids conserved in three or four toxins are boxed Amino acids conserved in Cry1Aa1 and Cry9Da2 toxins, but not Cry1Ab8 and Cry1Ac1 toxins, are highlighted (black) The numbers beneath the alignment indicate amino acid numbers in Cry1Aa1 toxin (B) Samples of approximately 1 lg of each toxin [i.e Cry1Aa1 (lane 1), Cry1Ab8 (lane 2), Cry1Ac1 (lane 3), and Cry9Da2 (lane 4)] were subjected to SDS ⁄ PAGE, transferred onto nitrocellulose membranes, and probed with mAb 2A11 (b) or mAb 2F9 (c) The proteins on the gels were visualized by Coomassie brilliant blue staining (a) Arrowheads indicate the 55 kDa activated Cry9Da2 toxin (lane 4).

four toxins were noted Amino acids conserved in

Cry1Aa1 and Cry9Da2 toxins but not Cry1Ab8 and

Cry1Ac1 toxins were highlighted Thirteen amino acids

were conserved in Cry1Aa1 and Cry9Da2 toxins, but

not Cry1Ab8 and Cry1Ac1 toxins: Ile357, Gly372,

Ser373, Phe381, Ser389, Glu404, Arg405, Thr435,

Glu439, Ala440, Gly442, Thr446, and Thr451 Of these

13 amino acids, Glu439, Ala440, Gly442 and Thr446

were located on loop 3 of domain II in the Cry1Aa

toxin 3D structure reported by Grochulski et al [3]

(Fig 5)

Expression and stability of Cry1Aa toxin

Cys-substitution mutants

In Cry1Aa, Cry1Ab, and Cry1Ac toxins, loop 3 of

domain II has been reported to be involved in binding

to cadherin-like proteins or brush border membrane

vesicles [23,44] To determine whether amino acids on

loop 3 are involved in binding to cadherin-like

pro-teins, we introduced five Cys substitutions – Y315C,

V444C, N340C, Y445C, and R448C – in and near

loop 3, and reacted these sites with a small blocking molecule, N-(9-acridinyl)maleimide (NAM) The sin-gle-Cys-substitution mutants of Cry1Aa toxin (amino acids 11–615) were constructed as GST fusion proteins, which were purified from Escherichia coli, solubilized, activated by trypsin, and analyzed by SDS⁄ PAGE Almond & Dean [45] reported that poor expression can be correlated with an unstable or mis-folded protein, but this finding was not applicable to our results, because we found that all mutant proteins (26 kDa GST plus 60 kDa mutant toxins, resulting in

86 kDa proteins) were expressed as strongly as the wild-type Cry1Aa toxin (data not shown) After tryptic digestion, the three single-Cys-substitution mutations

of Cry1Aa toxin (i.e Y315C, V444C, and R448) did not yield a 60 kDa trypsin-resistant core fragment (data not shown) These amino acid substitutions seem

to be involved in a conformational change of Cry1Aa toxin, resulting in the loss of trypsin resistance The two single-Cys-substitution mutations of Cry1Aa toxin (i.e N340C and Y445C) yielded a 60 kDa trypsin-resistant core fragment (data not shown), suggesting that the substitutions did not cause any structural alterations to the protein These two mutant proteins expressed in E coli had four Cys residues (amino acids 84, 137, 168, and 177) in the GST region and only two Cys residues (amino acids 15 and 340 or 445)

in the Cry1Aa toxin region However, the GST region and the N-terminal part of the Cry1Aa toxin region containing five of the six Cys residues were removed

by tryptic digestion Thus, each of the two single-Cys-substitution mutants of Cry1Aa toxin had only one Cys residue (amino acid 340 or 445) after digestion with trypsin We introduced NAM at a region on or near loop 3 of Cry1Aa toxin NAM contains a malei-mide group, which can react specifically with the thiol group of Cys in protein molecules to covalently bind it

to the protein When NAM was reacted with the single-Cys-substitution mutants, it bound specifically

to the Cys residues of the N340C and Y445C mutants

of Cry1Aa toxin NAM molecules introduced on the surface of Cry1Aa toxin at either of these two sites may block cadherin-like protein binding at the site, due to steric hindrance

Binding of Cry1Aa toxin constructs to GST

27 kDa BtR175

To investigate the location of BtR175-binding sites on Cry1Aa toxin, the binding of five Cry1Aa toxin con-structs (wild-type Cry1Aa, N340C, Y445C, N340C– NAM, and Y445C–NAM) to immobilized GST

27 kDa BtR175 molecules was measured using an

Fig 5 Epitope of mAb 2A11 on a 3D model of Cry1Aa toxin The

amino acids in domains I, II and III are colored pink, light green,

and light blue, respectively The amino acids conserved in Cry1Aa

and Cry9Da toxins, but not Cry1Ab and Cry1Ac toxins, are shown

in red Tyr445 and Asn340, which were substituted for Cys in

Fig 6, are shown in blue.

IAsys resonant mirror optical biosensor The profiles

of binding of the toxin constructs with the immobilized

GST 27 kDa BtR175 were monitored (Fig 6), and

kinetic constants were evaluated from profiles for the

binding of each toxin construct to the immobilized

GST 27 kDa BtR175 molecules (Table 1) There were

only subtle differences in the Bmax, kdiss and KD of

wild-type Cry1Aa toxin and the N340C mutant

(Table 1) There was almost no difference in the Bmax,

kass, kdiss and KD of the N340C and N340C–NAM

mutants (Table 1) These results indicate that

substitu-tion of Asn340 for Cys and the introducsubstitu-tion of NAM

on the N340C mutant have rather subtle effects on

binding In contrast, the KDof the Y445C mutant was

significantly lower than that of wild-type Cry1Aa toxin

(Table 1, Fig 6) This result indicates that the

substi-tution of Tyr445 for Cys has a crippling effect on binding

Effects of mAbs that inhibit binding of Cry1Aa toxin to receptors on toxicity

Monoclonal antibodies 1B10 and 2C2 inhibit the binding of Cry1Aa toxin to GST 7 kDa BmAPN1 [38] Monoclonal antibodies 2A11 and 2F9 inhibited the binding of Cry1Aa toxin to GST 27 kDa BtR175 (Fig 3) To investigate the effects of mAb binding on Cry1Aa toxicity, Cry1Aa toxin was preincubated with

or without each antibody Twenty second-instar larvae were fed an artificial diet containing Cry1Aa toxin with or without previous binding to antibodies The LC50 value of Cry1Aa toxin for Bo mori was 1.43 lgÆg)1 artificial diet The mortality rate was 63.3% when 20 second-instar larvae were fed an arti-ficial diet containing Cry1Aa toxin (2.0 lgÆg)1 artifi-cial diet) without any antibodies There was no mortality if the toxin was preincubated with mAb 2A11 (Table 2) This indicated that the toxicity of Cry1Aa toxin against Bo mori was blocked com-pletely when the toxin was preincubated with mAb 2A11 The mortality rate was 13.3% when the toxin was preincubated with mAb 2F9 This indicated that the toxicity of Cry1Aa toxin against Bo mori was reduced significantly when the toxin was preincubated with mAb 2F9 However, the mortality rates were 73.3% and 66.7% when the toxin was preincubated with mAbs 1B10 and 2C2, respectively (Table 2) These results indicate that the toxicity of Cry1Aa toxin against Bo mori was not reduced when the toxin was preincubated with mAbs 1B10 or 2C2, which can block the binding of Cry1Aa toxin to GST

7 kDa BmAPN1 [38]

Fig 6 Sensorgrams for the binding of wild-type and mutant

Cry1Aa toxins to GST 27 kDa BtR175 immobilized on the biosensor

surface The GST 27 kDa BtR175-immobilized cuvette was

equili-brated in NaCl ⁄ P i for 2 min Then, the wild-type toxin or one of the

four mutant Cry1Aa toxins (1 l M ) was added to the cuvette and

allowed to bind for 5 min The toxin was then removed from the

reaction chamber, and NaCl ⁄ P i was added Dissociation occurred

for 5 min after addition of NaCl ⁄ P i

Table 1 Equilibrium and kinetic binding parameters for the binding

of wild-type and mutant Cry1Aa toxins to BtR175 immobilized

on the biosensor surface Bmax, maximum binding amount; kass,

association rate constant; k dis , dissociation rate constant; K D ,

dissociation equilibrium constant; NAM, single Cys-substitution

mutant-bound NAM.

Toxins

Bmax

(pgÆmm)2Æ5 min)1)

kass (· 10 3

M )1Æs)1)

kdis (· 10 3 Æs)1)

KD (n M )

Table 2 Effects of mAbs 2F9 and 2A11 on the toxicity of Cry1Aa toxin against second-instar larvae Cry1Aa toxins were preincubated for 1 h with or without each antibody Twenty second-instar larvae were fed 1.0 g of artificial diet mixed with 2.00 lg of Cry1Aa toxin with or without previous binding to antibodies Mortality was recorded after 3 days The LC 50 value of Cry1Aa toxin against

Bo mori was 1.43 lgÆg)1artificial diet The assays were repeated three times, and mean values are shown.

The effect of substituting Tyr445 for Cys on the

toxicity of Cry1Aa toxin

Twenty third-instar larvae of Bo mori were fed

solu-tions of three serially diluted Cry1Aa toxin constructs

(wild-type Cry1Aa toxin, Y445C, and Y445C–NAM),

and the LD50was determined by a probit analysis for

each toxin As a result, the LD50± 95% confidence

limits of wild-type Cry1Aa toxin, mutant Y445C and

mutant Y445C–NAM were 1.02 (+0.08⁄)0.08), 166.85

(+20.49⁄)13.58), and 154.55 (+11.22 ⁄ )9.38) ng,

respectively The LD50 values of the Y445C and

Y445C–NAM mutants were 164 and 152 times higher

than that of wild-type Cry1Aa toxin, respectively

Discussion

Knowledge of the receptor-binding site on a Cry toxin

would allow improvements in the specificity and

toxic-ity of these toxins In the case of Cry1Ab toxin, several

binding sites for M sexta cadherin BT-R1 have been

identified However, the binding site on Cry1Aa toxin

for the Bo mori cadherin-like protein has not been

reported As the 3D structure of Cry1Aa toxin is

known, it is possible to understand the BtR175-binding

site on Cry1Aa toxin three-dimensionally In this

study, we attempted to determine the binding site on

Cry1Aa toxin for the candidate Cry1Aa toxin receptor

in Bo mori, a cadherin-like protein, BtR175 [37]

(Gen-Bank accession number AB026260.1)

Locational relationship between the

BmAPN1-binding site and the BtR175-BmAPN1-binding site

To locate the BtR175-binding site, the relationship

between the BmAPN1-binding and BtR175-binding

sites on Cry1Aa toxin was investigated using two

methods (Figs 1 and 2) The results suggest that the

BtR175-binding site is close to the BmAPN1-binding

site, and that the binding sites are not located on

opposite surfaces of the Cry1Aa toxin molecule

Monoclonal antibodies 2A11 and 2F9 competed with

BtR175 for binding to Cry1Aa toxin (Fig 3), but did

not compete with BmAPN1 [38] Monoclonal

anti-bodies 1B10 and 2C2 competed with BmAPN1 for

binding to Cry1Aa toxin [38], but did not compete

with BtR175 (data not shown) These results suggest

that the BtR175-binding and BmAPN1-binding sites

are not the same In conclusion, the BtR175-binding

site is on the same face of Cry1Aa toxin as the

BmAPN1-binding site, as the areas occupied by GST

27 kDa BtR175 and GST 7 kDa BmAPN1 overlap or

are close to each other In our previous report, the

BmAPN1-binding site on Cry1Aa toxin was hypothe-sized to consist of conserved amino acids (Gln292, Pro294, His295, Leu296, His433, Leu481, Val444, Val487, Arg500, and Gly505) between Cry1Aa, Cry1Ac and Cry9Da toxins, and nonconserved amino acids surrounding the conserved amino acids [38] In fact, this region is on the same face as loop 3 of domain 2 and is close enough to it to be hidden by the protein that binds to loop 3

The location of the mAb 2A11 epitope that blocks binding of Cry1Aa toxin to GST 27 kDa BtR175 The results of competitive binding assays indicated that the epitope of mAb 2A11 and the BtR175-bind-ing site overlap or are close to each other on Cry1Aa toxin (Fig 3) To analyze the area in which the BtR175-binding site may be located, the epitope

of mAb 2A11 was investigated Immunoblotting anal-ysis using four Cry toxins showed that mAb 2A11 recognized Cry1Aa1 and Cry9Da2 toxins, but not Cry1Ab8 or Cry1Ac1 toxin (Fig 4Bb) These obser-vations suggest that the epitope of mAb 2A11 is localized in the region consisting of the conserved amino acids in Cry1Aa1 and Cry9Da2 toxins but not

in Cry1Ab8 or Cry1Ac1 toxins The epitope of mAb 2A11 on Cry1Aa toxin is in domain II [38] Align-ment of Cry toxin domain II showed that 13 amino acids were conserved in Cry1Aa1 and Cry9Da2 tox-ins, but not in Cry1Ab8 or Cry1Ac1 toxin (Fig 4A) Seven of the 13 amino acids (Gly372, Ser373, Glu439, Ala440, Gly442, Thr435, and Thr446) are located on the same face as the BmAPN1-binding site on the Cry1Aa toxin 3D structure (Fig 5) [38] It seems unlikely that Thr435, which is adjacent to the BmAPN1-binding site, is contained in the epitope of mAb 2A11, because this mAb did not block the bind-ing of Cry1Aa toxin to BmAPN1 The interaction between an antibody and antigen occurs over a large area, with approximate dimensions of 20· 30 A˚ [46], the epitope contains 15–20 amino acids [47], and the epitope is made up of four discontinuous segments of polypeptide chain on average [48] Thus, the epitope

of mAb 2A11 might be located in the region consist-ing of six amino acids (Gly372, Ser373, Glu439, Ala440, Gly442, and Thr446) (Fig 5), because this region fulfills the above conditions

Region in which the amino acid substitution influences the binding of Cry1Aa toxin to BtR175 Substitution for Cys and the steric hindrance by NAM

at Asn340 had subtle effects on the binding of Cry1Aa

toxin to BtR175 (Fig 6, Table 1), suggesting that

Asn340 may be outside of the BtR175-binding site

Meanwhile, substitution of Tyr445 for Cys had a

crip-pling effect on binding of Cry1Aa toxin to BtR175,

indicating that Tyr445 is located in the

BtR175-bind-ing site Otherwise, the mutation might have altered

the conformation of loop 3 enough to reduce the

abil-ity to bind to BtR175 Asn340 and Tyr445 are close to

each other in the Cry1Aa toxin 3D structure

How-ever, they are on different strands and different faces

on the Cry1Aa toxin 3D structure (Fig 5)

The BmAPN1-binding [38] and BtR175-binding

sites seemed to be close to each other in the Cry1Aa

toxin 3D structure (Figs 1 and 2) Glu439, Ala440,

Gly442, and Thr446, which comprise the epitope of

mAb 2A11, are located on loop 3 of Cry1Aa toxin

(Fig 5) On the basis of the above considerations, it

is possible that the BtR175-binding site is located

near loop 3 in domain II (Fig 5) An in vitro binding

assay and an in vivo bioassay using several Ala

substitution mutants of Cry1Ab toxin showed that

Ala substitution of amino acids in loop 3 affects

initial receptor binding and toxicity in M sexta and

H virescens [44] Competition binding assays using

loop region peptides have shown that H virescens

cadherin binds loop 3 of Cry1Ab and Cry1Ac toxins

[23] These results also support the conclusion that

the loop 3 region of Cry1Aa toxin is involved in

binding to BtR175 Recently, it was shown that

sub-stitutions of loops, including loop 3 of Cry1Aa toxin

for the third loop of the complementarity-determining

region of the immunoglobulin heavy chain, do not

affect the binding property of the toxin for H

vires-cens cadherin-like protein [49] This result contradicts

our conclusion mentioned here and other reported

hypotheses [23,36]

Impact of Cry1Aa–BtR175 binding on toxicity

In this study, mAbs 2A11 and 2F9 inhibited the binding

of Cry1Aa toxin to BtR175 (Fig 3) To investigate the

effects of these mAbs on Cry1Aa toxicity, second-instar

larvae were fed Cry1Aa toxin, either alone or together

with a 1000-fold molar excess of each mAb The

mortal-ity rate decreased markedly with mAbs 2A11 and 2F9

(Table 2) Inhibition of the binding of Cry1Aa and

Cry1Ac toxins to BtR175 by pretreatment with antibody

against BtR175 suppresses the lytic activity of toxins on

collagenase-dissociated Bo mori midgut epithelial cells

[24] Substitution of Tyr445 for Cys had a crippling

effect on binding to BtR175 (Table 1) This substitution

caused a 164-fold decrease in toxicity These results

showed that BtR175 is involved in Cry1Aa toxicity

Experimental procedures

Preparation of GST 7 kDa BmAPN1 and GST

27 kDa BtR175

A partial fragment of BmAPN1 (7 kDa BmAPN1) contain-ing the Cry1Aa toxin-bindcontain-ing region (Ile135 to Pro198) was prepared as a GST fusion protein as described by Yaoi et al [39] A partial fragment of BtR175 (27 kDa BtR175) contain-ing the Cry1Aa toxin-bindcontain-ing region (Glu1108 to Val1464, including cadherin repeat-8 and repeat-9, and the N-terminal half of the membrane proximal region) was prepared as a GST fusion protein, as described by Hara et al [24]

Preparation of Cry1Aa, Cry1Ab, Cry1Ac and Cry9Da2 toxins

Cry1Aa1 and Cry1Ab8 toxins were prepared from recombi-nant E coli strains as described by Atsumi et al [38] Cry1Ac and Cry9Da2 toxins were prepared from B thurin-giensisstrains as described by Shinkawa et al [50]

Biotinylation of Cry1Aa toxin Cry1Aa toxin was biotinylated with EZ-LinkSulfo-NHS-LC-LC-biotin (Pierce, Chester, UK) as described by Atsumi

et al.[38]

Competitive binding assay using ELISA plate Ninety-six-well flat-bottomed ELISA plates (Corning Inc., Corning, NY, USA) were coated with 1 lm GST 27 kDa BtR175 solution in NaC⁄ Pi(100 lL per well) at 37C for

2 h The wells were blocked by incubating the plates with NaCl⁄ Picontaining 2% BSA (150 lL per well) at 37C for

2 h and using an avidin⁄ biotin blocking kit in accordance with the manufacturer’s instructions (Vector Laboratories, Burlingame, CA, USA) The wells were washed three times with NaCl⁄ Pi Biotinylated Cry1Aa toxin (20 nm) was preincubated with various concentrations (0, 2, 5, 10, 20,

50, 100, 200 or 500 nm) of GST 7 kDa BmAPN1 or GST

27 kDa BtR175 solution or preincubated with various con-centrations (0, 1, 2, 5, 10, 20, 50, 100, 200, 500, 1000, 2000

or 5000 nm) of each mAb at 37C for 90 min and then added to the wells (100 lL per well) Biotinylated Cry1Aa toxin bound to GST 27 kDa BtR175 was detected with horseradish peroxidase (HRP)-conjugated streptavidin as described by Atsumi et al [38] The results are expressed as

B⁄ Bmax, where B is the enzymatic activity bound to the solid phase measured at various concentrations of GST

7 kDa BmAPN, GST 27 kDa BtR175 or mAbs, and Bmax

is the enzymatic activity in the absence of these proteins The assays were repeated three times, and mean values were plotted

Immobilization of Cry1Aa toxin on the sensing

surface of a biosensor

Single-well cuvettes with carboxylate surfaces were used

with the N-hydroxysuccinimide and

1-ethyl-3-(3-dimethyl-aminopropyl)-carbodiimide coupling system (Affinity

Sensors, Cambridge, UK) The well of each cuvette was

coated with activated Cry1Aa toxin as described by Atsumi

et al.[38]

Determination of binding compatibility for

BtR175, BmAPN1 and mAb 1B10 using the

IAsys optical sensor

After covalent immobilization of Cry1Aa toxin on the

car-boxylate sensor surface, the compatible binding of GST

7 kDa BmAPN1, GST 27 kDa BtR175 and mAb 1B10

against Cry1Aa toxin to the immobilized Cry1Aa toxin was

determined using an IAsys resonant mirror optical

bio-sensor (Affinity Sensors) The concentrations of the GST

7 kDa BmAPN1, GST 27 kDa BtR175 and mAb 1B10

solutions were 55.5, 27.6 and 10 nm, respectively To

mea-sure the binding of individual receptors to Cry1Aa toxin

molecules, GST 7 kDa BmAPN1 or GST 27 kDa BtR175

was added to the Cry1Aa toxin-immobilized surface at the

concentrations given above, and the association was

observed for 5 min (binding response) Subsequently, each

receptor molecule solution was replaced with protein-free

binding buffer (NaCl⁄ Pi), and the dissociation of binding

was observed for 5 min (dissociation response) At the end

of this cycle, the NaCl⁄ Piin the cuvette was replaced with

6 m guanidine hydrochloride solution to regenerate the

sen-sor surface Next, to investigate the relative locations of the

binding sites for GST 27 kDa BtR175 and GST 7 kDa

BmAPN1 on the surface of Cry1Aa toxin, the following

compounds were injected First, GST 7 kDa BmAPN1 was

added to block the binding site on the immobilized Cry1Aa

toxin Then, GST 7 kDa BmAPN1 or GST 27 kDa BtR175

was added as a second molecule, and the additive

associa-tion was observed for 5 min GST 27 kDa BtR175 was

added to block the binding site on the immobilized Cry1Aa

toxin Then, GST 7 kDa BmAPN1 was added, and the

additive association was observed for 5 min As a control,

the compatibility of binding of GST 27 kDa BtR175 and

mAb 1B10 on the surface of Cry1Aa toxin at the same time

was tested The plots during regeneration of the sensor

sur-face were omitted from the sensorgram because the values

exceeded 20 000 arcsec during surface regeneration

Preparation of hybridomas and monoclonal

antibodies against Cry1Aa toxin

A BALB⁄ c mouse was immunized with 20 lg of activated

Cry1Aa toxin, and hybridomas were produced Ascites

were obtained from the hybridomas Seven mAbs were raised against Cry1Aa toxin: mAbs 1B10, 1E10, 1G10, 2A11, 2C2, 2F9 and 3C7 were purified from ascites [38] Aliquots of purified antibodies were conjugated with HRP using the methods of Wilson & Nakane [51]

Immunoblotting Prepared proteins (1 lg per lane) were subjected to SDS⁄ PAGE (10% polyacrylamide gel) according to the method of Laemmli [52] After SDS⁄ PAGE, the proteins were transferred onto Immobilon nitrocellulose membranes (Millipore, Bedford, MA, USA), blocked with 2% BSA in

10 mm Tris⁄ HCl (pH 8.3) with 150 mm NaCl and 0.05% (v⁄ v) Tween-20 (TNT) for 12 h at 4 C, and then incubated for 2 h at room temperature with HRP-conjugated mAb diluted 1 : 10 000 The membranes were washed three times with TNT The bands on the blots were visualized using an ECL detection system (GE Healthcare UK Ltd., Little Chalfont, UK)

Site-directed mutagenesis Five pairs of complementary mutagenic oligonucleotide primers (Table 3) were designed to introduce Cys substitu-tions Site-directed mutations were introduced into the Cry1Aa toxin gene using pN615, a plasmid containing the

B thuringiensis activated Cry1Aa toxin gene as a template

as described by Atsumi et al [38] The mutated DNA was used to transform E coli BL21(DE3) cells Mutations were confirmed by DNA sequencing

Introduction of small blocking molecules into single-Cys-substitution Cry1Aa toxin mutants Two Cys-substitution mutants of Cry1Aa toxin were reacted with a small blocking molecule, NAM (Dojindo

Table 3 Mutagenic primers used for site-directed mutagenesis of Cry1Aa toxin genes The mutated cysteine codons are underlined and the nucleotide point mutations are shown in upper case Mutants Primers Oligonucleotide sequence (5¢- to -3¢)

Y315C gatgccctgaccaaCaattaaagcctctat

N340C ctgcattccccgcaCAcccaaataaaggga

V444C ctctcaaggtgtaaCAtgctccagctgctt

Y445C ggagctctcaaggtACaaactgctccagct

R448C aaaacgttggagcAcAcaaggtgtaaactg