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

Location of the Bombyx mori 175 kDa cadherin-like
protein-binding site on Bacillus thuringiensis Cry1Aa toxin
Shogo Atsumi
1,2
, Yukino Inoue
1
, Takahisa Ishizaka
1
, Eri Mizuno
1
, Yasutaka Yoshizawa
1
,
Madoka Kitami
1
and Ryoichi Sato
1
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:
(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.
FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS 4913
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-R
1
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-R
1
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-R
1
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-R
1
[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-R
1
[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 being mapped. Three Cry1A
toxin-binding sites have been mapped in BT-R
1
. 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
BtR175 binding site on the Cry1Aa toxin S. Atsumi et al.
4914 FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS
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 protein (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.
S. Atsumi et al. BtR175 binding site on the Cry1Aa toxin
FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS 4915
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
A B C D E F G
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.
BtR175 binding site on the Cry1Aa toxin S. Atsumi et al.
4916 FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS
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
a b c
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).
S. Atsumi et al. BtR175 binding site on the Cry1Aa toxin
FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS 4917
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.
BtR175 binding site on the Cry1Aa toxin S. Atsumi et al.
4918 FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS
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 B

max
, k
diss
and K
D
of
wild-type Cry1Aa toxin and the N340C mutant
(Table 1). There was almost no difference in the B
max
,
k
ass
, k
diss
and K
D
of the N340C and N340C–NAM
mutants (Table 1). These results indicate that substitu-
tion of Asn340 for Cys and the introduction of NAM
on the N340C mutant have rather subtle effects on
binding. In contrast, the K
D
of 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 LC
50
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. B
max
, maximum binding amount; k
ass
,
association rate constant; k
dis
, dissociation rate constant; K

D
,
dissociation equilibrium constant; NAM, single Cys-substitution
mutant-bound NAM.
Toxins
B
max
(pgÆmm
)2
Æ5 min
)1
)
k
ass
(· 10
3
M
)1
Æs
)1
)
k
dis
(· 10
3
Æs
)1
)
K
D

(nM)
Cry1Aa 21.10 87.5 8.20 93.7
N340C 20.05 15.9 2.92 184
N340C–NAM 15.19 7.42 2.57 346
Y445C 3.320 1.19 2.70 2260
Y445C–NAM 1.880 0.546 1.39 2550
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
)1
artificial diet. The assays were repeated
three times, and mean values are shown.
Treatment Mortality (%)
– 0.0 ± 0.0
Cry1Aa 63.3 ± 2.4
Cry1Aa + 2C2 66.7 ± 2.4
Cry1Aa + 1B10 73.3 ± 4.7
Cry1Aa + 2F9 13.3 ± 4.7
Cry1Aa + 2A11 0.0 ± 0.0
S. Atsumi et al. BtR175 binding site on the Cry1Aa toxin
FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS 4919
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 LD
50
was determined by a probit analysis for
each toxin. As a result, the LD
50
± 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 LD
50
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-R
1
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-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
BtR175 binding site on the Cry1Aa toxin S. Atsumi et al.
4920 FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS
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-binding 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-binding 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-
giensis strains 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 ⁄ P
i
(100 lL per well) at 37 °C for
2 h. The wells were blocked by incubating the plates with
NaCl ⁄ P
i
containing 2% BSA (150 lL per well) at 37 °C 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 ⁄ P
i
. 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 37 °C 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 ⁄ B
max
, 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 B
max
is the enzymatic activity in the absence of these proteins.

The assays were repeated three times, and mean values were
plotted.
S. Atsumi et al. BtR175 binding site on the Cry1Aa toxin
FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS 4921
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 ⁄ P

i
), and the dissociation of binding
was observed for 5 min (dissociation response). At the end
of this cycle, the NaCl ⁄ P
i
in 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 Y315C atagaggctttaat
tGttggtcagggcatc
Y315C gatgccctgacca
aCaattaaagcctctat
N340C N340C tccctttatttggg
TGtgcggggaatgcag
N340C ctgcattccccgc
aCAcccaaataaaggga
V444C V444C aagcagctggagca
TGttacaccttgagag
V444C ctctcaaggtgta
aCAtgctccagctgctt
Y445C Y445C agctggagcagtt
tGTaccttgagagctcc
Y445C ggagctctcaaggt
ACaaactgctccagct
R448C R448C cagtttacaccttg
TgTgctccaacgtttt
R448C aaaacgttggagc
AcAcaaggtgtaaactg
BtR175 binding site on the Cry1Aa toxin S. Atsumi et al.
4922 FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS
Laboratories, Tokyo, Japan) and dialyzed as described by
Atsumi et al. [38].
Immobilization of BtR175 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). The well of each cuvette was coated with GST
27 kDa BtR175 immobilized on the carboxylate surface via
its amino group using succinimide ester chemistry, as
described below. The immobilization buffer was 10 mm
sodium acetate buffer (pH 5.0). For ligand coupling, the
cuvette was washed with NaCl ⁄ P
i
containing 0.05% Tween-
20 (pH 7.4) for 10 min prior to activation of the surface
using N-hydroxysuccinimide and 1-ethyl-3-(3-dimethyla-
minopropyl)-carbodiimide. The cuvette was then equili-
brated using sodium acetate buffer (pH 5.0), and
540 lgÆmL
)1
GST 27 kDa BtR175 was added to the well.
Eventually, 2.73 ngÆmm
)2
of GST 27 kDa BtR175 was
immobilized on the sensing surface via amino groups. The
noncoupled ligand was removed by washing with 20 mm
HCl, and the unreacted surface was blocked with BSA.
Binding of Cry1Aa toxin constructs to GST
27 kDa BtR175
After covalent immobilization of GST 27 kDa BtR175 on
the carboxylate sensor surface, the binding of five Cry1Aa
toxin constructs (i.e. Cry1Aa, N340C, Y445C, N340C–
NAM and Y445C–NAM) to GST 27 kDa BtR175 was
determined using an IAsys resonant mirror optical biosen-
sor. To measure the binding of each toxin construct with

GST 27 kDa BtR175, each toxin construct was added to
the GST 27 kDa BtR175 immobilized surface at a concen-
tration of 1 lm, and the association was observed for 5 min
(binding response). Subsequently, each toxin construct solu-
tion was replaced with protein-free binding buffer
(NaCl ⁄ P
i
), and the dissociation of binding was observed for
5 min (dissociation response). At the end of this cycle, the
NaCl ⁄ P
i
in the cuvette was replaced with 20 mm HCl to
regenerate the sensor surface. Kinetic constants were evalu-
ated from profiles for the binding of each Cry1Aa toxin
construct to GST 27 kDa BtR175 using the kinetic analysis
program fast fit (Affinity Sensors). The apparent associa-
tion rate constant, k
on
(s
)1
), was calculated from a single
exponential fit of the binding response curve, and the
apparent dissociation rate constant, k
off
(s
)1
), was calcu-
lated from the dissociation response curve. The dissociation
constant, k
diss

(m
)1
Æs
)1
), was determined using the exp-
ression k
off
= k
diss
, and the association rate constant, k
ass
(m
)1
Æs
)1
), was determined on the basis of the equa-
tion k
ass
=(k
on
⁄ k
off
) ⁄ [Ligate] [53]. The equilibrium disso-
ciation constant was determined on the basis of the
equation K
D
= k
diss
⁄ k
ass

.
Insects
Kinshu · Showa, a hybrid strain of Bo. mori, was pur-
chased from Ueda-Sanshu Co. (Ueda, Japan) and reared
on an artificial diet (Silkmate 3M; Nihon-Nosanko Co.,
Yokohama, Japan) at 25 °C.
Bioassay
Activated Cry1Aa toxins (2.00 lg) were preincubated for
1 h with or without 15.0 lg of each antibody in 100 lLof
MilliQ-treated water. 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, and mortal-
ity was recorded after 3 days. The assays were repeated
three times.
To determine the LD
50
, toxin construct solutions serially
diluted with mulberry leaf juice were fed to each of 20 day
1 third-instar larvae with a micropipette, and then the lar-
vae were reared on an artificial diet. The number of dead
larvae was recorded after 24 h, and the data were analyzed
using the probit method.
Acknowledgement
This work was supported by a Grant-in-Aid for Scien-
tific Research (B) (18310053) from the Ministry of
Education, Culture, Sports, Science, and Technology
of Japan.
References
1 Aronson AI, Beckman W & Dunn P (1986) Bacillus
thuringiensis and related insect pathogens. Microbiol

Rev 50, 1–24.
2 Hofmann C, Lu
¨
thy P, Hu
¨
tter R & Pliska V (1988)
Binding of the delta-endotoxin from Bacillus thuringi-
ensis to brush-border membrane vesicles of the cab-
bage butterfly (Pieris brassicae). Eur J Biochem 173,
85–91.
3 Grochulski P, Masson L, Borisova S, Pusztai-Carey
M, Schwartz JL, Brousseau R & Cygler M (1995)
Bacillus thuringiensis CryIA(a) insecticidal toxin: crystal
structure and channel formation. J Mol Biol 254, 447–
464.
4 Chen XJ, Curtiss A, Alcantara E & Dean DH (1995)
Mutations in domain I of Bacillus thuringiensis delta-
endotoxin CryIAb reduce the irreversible binding of
toxin to Manduca sexta brush border membrane vesi-
cles. J Biol Chem 270, 6412–6419.
S. Atsumi et al. BtR175 binding site on the Cry1Aa toxin
FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS 4923
5 Gomez I, Miranda-Rios J, Rudino-Pinera E, Oltean DI,
Gill SS, Bravo A & Soberon M (2002) Hydropathic
complementarity determines interaction of epitope
(869)HITDTNNK(876) in Manduca sexta Bt-R(1)
receptor with loop 2 of domain II of Bacillus thuringien-
sis Cry1A toxins. J Biol Chem 277, 30137–30143.
6Go
´

mez I, Dean DH, Bravo A & Sobero
´
n M (2003)
Molecular basis for Bacillus thuringiensis Cry1Ab toxin
specificity: two structural determinants in the Manduca
sexta BT-R
1
receptor interact with loops alpha-8 and 2
in domain II of Cy1Ab toxin. Biochemistry 42, 10482–
10489.
7 Burton SL, Ellar DJ, Li J & Derbyshire DJ (1999)
N-Acetylgalactosamine on the putative insect receptor
aminopeptidase N is recognised by a site on the
domain III lectin-like fold of a Bacillus thuringiensis
insecticidal toxin. J Mol Biol 287, 1011–1022.
8 Knight PJ, Crickmore N & Ellar DJ (1994) The receptor
for Bacillus thuringiensis CryIAc delta-endotoxin in the
brush border membrane of the lepidopteran Manduca
sexta is aminopeptidase N. Mol Microbiol 11, 429–436.
9 Lee MK, You TH, Gould FL & Dean DH (1999) Iden-
tification of residues in domain III of Bacillus thuringi-
ensis Cry1Ac toxin that affect binding and toxicity.
Appl Environ Microbiol 65, 4513–4520.
10 Denolf P, Hendrickx K, Van Damme J, Jansens S,
Peferoen M, Degheele D & Van Rie J (1997) Cloning
and characterization of Manduca sexta and Plutella
xylostella midgut aminopeptidase N enzymes related
to Bacillus thuringiensis toxin binding proteins. Eur J
Biochem 248, 748–761.
11 Gill SS, Cowles EA & Francis V (1995) Identification,

isolation, and cloning of a Bacillus thuringiensis Cry1Ac
toxin-binding protein from the midgut of the lepidop-
teran insect Heliothis virescens. J Biol Chem 270, 27277–
27282.
12 Nakanishi K, Yaoi K, Shimada N, Kadotani T & Sato
R (1999) Bacillus thuringiensis insecticidal Cry1Aa toxin
binds to a highly conserved region of aminopeptidase N
in the host insect leading to its evolutionary success.
Biochim Biophys Acta 1432, 57–63.
13 Nakanishi K, Yaoi K, Nagino Y, Hara H, Kitami M,
Atsumi S, Miura N & Sato R (2002) Aminopeptidase N
isoforms from the midgut of Bombyx mori and Plutella
xylostella – their classification and the factors that
determine their binding specificity to Bacillus thuringien-
sis Cry1A toxin. FEBS Lett 519, 215–220.
14 Yaoi K, Kadotani T, Kuwana H, Shinkawa A, Takah-
ashi T, Iwahana H & Sato R (1997) Aminopeptidase N
from Bombyx mori as a candidate for the receptor of
Bacillus thuringiensis Cry1Aa toxin. Eur J Biochem 246,
652–657.
15 Gill M & Ellar D (2002) Transgenic Drosophila reveals
a functional in vivo receptor for the Bacillus thuringien-
sis toxin Cry1Ac1. Insect Mol Biol 11
, 619–625.
16 Rajagopal R, Sivakumar S, Agrawal N, Malhotra P &
Bhatnagar RK (2002) Silencing of midgut aminopepti-
dase N of Spodoptera litura by dsRNA establishes its
role as Bt toxin receptor. J Biol Chem 277, 46849–
46851.
17 Sivakumar S, Rajagopal R, Venkatesh GR, Srivastava

A & Bhatnagar RK (2007) Knockdown of aminopepti-
dase-N from Helicoverpa armigera larvae and in trans-
fected Sf21 cells by RNA interference reveals its
functional interaction with Bacillus thuringiensis insecti-
cidal protein Cry1Ac. J Biol Chem 282, 7312–7319.
18 Luo K, Sangadala S, Masson L, Mazza A, Brousseau
R & Adang MJ (1997) The Heliothis virescens 170 kDa
aminopeptidase functions as ‘receptor A’ by mediating
specific Bacillus thuringiensis Cry1Ad-endotoxin binding
and pore formation. Insect Biochem Mol Biol 27, 735–
743.
19 Schwartz JL, Lu YJ, So
¨
hnlein P, Brousseau R, Laprade
R, Masson L & Adang MJ (1997) Ion channels formed
in planar lipid bilayers by Bacillus thuringiensis toxins
in the presence of Manduca sexta midgut receptors.
FEBS Lett 412, 270–276.
20 Vadlamudi RK, Weber E, Ji I, Ji TH & Bulla LA Jr
(1995) Cloning and expression of a receptor for an
insecticidal toxin of Bacillus thuringiensis. J Biol Chem
270, 5490–5494.
21 Nagamatsu Y, Toda S, Yamaguchi F, Ogo M, Kogure
M, Nakamura M, Shibata Y & Katsumoto T (1998)
Identification of Bombyx mori midgut receptor for
Bacillus thuringiensis insecticidal CryIA(a) toxin. Biosci
Biotechnol Biochem 62, 718–726.
22 Nagamatsu Y, Toda S, Koike T, Miyoshi Y, Shigema-
tsu S & Kogure M (1998) Cloning, sequencing, and
expression of the Bombyx mori receptor for Bacillus

thuringiensis insecticidal Cry1Aa toxin. Biosci Biotechnol
Biochem 62, 727–734.
23 Xie R, Zhuang M, Ross LS, Gomez I, Oltean DI,
Bravo A, Soberon M & Gill SS (2005) Single amino
acid mutations in the cadherin receptor from heliothis
virescens affect its toxin binding ability to Cry1A tox-
ins. J Biol Chem 280, 8416–8425.
24 Hara H, Atsumi S, Yaoi K, Nakanishi K, Higurashi S,
Miura N, Tabunoki H & Sato R (2003) A cadherin-like
protein functions as a receptor for Bacillus thuringiensis
Cry1Aa and Cry1Ac toxins on midgut epithelial cells of
Bombyx mori larvae. FEBS Lett 538, 29–34.
25 Gahan LJ, Gould F & Heckel DG (2001) Identification
of a gene associated with Bt resistance in Heliothis vires-
cens. Science 293, 857–860.
26 Sobero
´
n M, Pardo-Lo
´
pez L, Lo
´
pez I, Go
´
mez I, Tabash-
nik BE & Bravo A (2007) Engineering modified Bt tox-
ins to counter insect resistance. Science 318, 1640–1642.
27 Sobero
´
nM,Pe
´

rez RV, Nun
˜
ez-Valde
´
z ME, Lorence A,
Go
´
mez I, Sa
´
nchez J & Bravo A (2000) Evidence for
intermolecular interaction as a necessary step for
BtR175 binding site on the Cry1Aa toxin S. Atsumi et al.
4924 FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS
pore-formation activity and toxicity of Bacillus thuringi-
ensis Cry1Ab toxin. FEMS Microbiol Lett 191, 221–225.
28 Go
´
mez I, Sa
´
nchez J, Miranda R, Bravo A & Sobero
´
n
M (2002) Cadherin-like receptor binding facilitates pro-
teolytic cleavage of helix alpha-1 in domain I and oligo-
mer pre-pore formation of Bacillus thuringiensis Cry1Ab
toxin. FEBS Lett 513, 242–246.
29 Zhuang M, Oltean DI, Go
´
mez I, Pullikuth AK, Sober-
o

´
n M, Bravo A & Gill SS (2002) Heliothis virescens and
Manduca sexta lipid rafts are involved in Cry1A toxin
binding to the midgut epithelium and subsequent pore
formation. J Biol Chem 277, 13863–13872.
30 Bravo A, Go
´
mez I, Conde J, Mun
˜
oz-Garay C, Sa
´
nchez
J, Miranda R, Zhuang M, Gill SS & Sobero
´
n M (2004)
Oligomerization triggers binding of a Bacillus thuringi-
ensis Cry1Ab pore-forming toxin to aminopeptidase N
receptor leading to insertion into membrane micro-
domains. Biochim Biophys Acta 1667, 38–46.
31 Go
´
mez I, Arenas I, Benitez I, Miranda-Rı
´
os J, Becerril
B, Grande R, Almagro JC, Bravo A & Sobero
´
nM
(2006) Specific epitopes of domains II and III of Bacil-
lus thuringiensis Cry1Ab toxin involved in the sequential
interaction with cadherin and aminopeptidase-N recep-

tors in Manduca sexta. J Biol Chem 281, 34032–34039.
32 Zhang X, Candas M, Griko NB, Rose-Young L &
Bulla LA Jr (2005) Cytotoxicity of Bacillus thuringiensis
Cry1Ab toxin depends on specific binding of the toxin
to the cadherin receptor BT-R
1
expressed in insect cells.
Cell Death Differ 12, 1407–1416.
33 Zhang X, Candas M, Griko NB, Taussig R & Bulla LA
Jr (2006) A mechanism of cell death involving an ade-
nylyl cyclase ⁄ PKA signaling pathway is induced by the
Cry1Ab toxin of Bacillus thuringiensis. Proc Natl Acad
Sci USA 103, 9897–9902.
34 Go
´
mez I, Oltean DI, Gill SS, Bravo A & Sobero
´
nM
(2001) Mapping the epitope in cadherin-like receptors
involved in Bacillus thuringiensis Cry1A toxin interac-
tion using phage display. J Biol Chem 276, 28906–
28912.
35 Dorsch JA, Candas M, Griko NB, Maaty WS, Midboe
EG, Vadlamudi RK & Bulla LA Jr (2002) Cry1A toxins
of Bacillus thuringiensis bind specifically to a region
adjacent to the membrane-proximal extracellular
domain of BT-R(1) in Manduca sexta: involvement of a
cadherin in the entomopathogenicity of Bacillus thurin-
giensis. Insect Biochem Mol Biol 32, 1025–1036.
36 Hua G, Jurat-Fuentes JL & Adang MJ (2004) BT-R

1
a
extracellular cadherin repeat 12 mediates Bacillus thurin-
giensis Cry1Ab binding and cytotoxicity. J Biol Chem
279, 28051–28056.
37 Nagamatsu Y, Koike T, Sasaki K, Yoshimoto A &
Furukawa Y (1999) The cadherin-like protein is essen-
tial to specificity determination and cytotoxic action of
the Bacillus thuringiensis insecticidal CryIAa toxin.
FEBS Lett 460, 385–390.
38 Atsumi S, Mizuno E, Hara H, Nakanishi K, Kitami M,
Miura N, Tabunoki H, Watanabe A & Sato R (2005)
Location of the Bombyx mori aminopeptidase N type 1
binding site on Bacillus thuringiensis Cry1Aa toxin. Appl
Environ Microbiol 71, 3966–3977.
39 Yaoi K, Nakanishi K, Kadotani T, Imamura M, Koiz-
umi N, Iwahana H & Sato R (1999) Bacillus thuringien-
sis Cry1Aa toxin-binding region of Bombyx mori
aminopeptidase N. FEBS Lett 463 , 221–224.
40 Schnepf HE, Wong HC & Whiteley HR (1985) The
amino acid sequence of a crystal protein from Bacillus
thuringiensis deduced from the DNA base sequence.
J Biol Chem 260, 6264–6272.
41 Oeda K, Oshie K, Shimizu M, Nakamura K, Yamam-
oto H, Nakayama I & Ohkawa H (1987) Nucleotide
sequence of the insecticidal protein gene of Bacillus
thuringiensis strain aizawai IPL7 and its high-level
expression in Escherichia coli. Gene 53, 113–119.
42 Adang MJ, Staver MJ, Rocheleau TA, Leighton J,
Barker RF & Thompson DV (1985) Characterized

full-length and truncated plasmid clones of the crystal
protein of Bacillus thuringiensis subsp. kurstaki HD-73
and their toxicity to Manduca sexta. Gene 36, 289–300.
43 Wasano N, Ohba M & Miyamoto K (2001) Two d-
endotoxin genes, Cry9Da and a novel related gene,
commonly occurring in Lepidoptera-specific Bacillus
thuringiensis Japanese isolates that produce spherical
parasporal inclusions. Curr Microbiol 42, 129–133.
44 Rajamohan F, Hussain SR, Cotrill JA, Gould F &
Dean DH (1996) Mutations at domain II, loop 3, of
Bacillus thuringiensis CryIAa and CryIAb d-endotoxins
suggest loop 3 is involved in initial binding to lepidop-
teran midguts. J Biol Chem 271, 25220–25226.
45 Almond BD & Dean DH (1993) Structural stability of
Bacillus thuringiensis delta-endotoxin homolog-scanning
mutants determined by susceptibility to proteases. Appl
Environ Microbiol 59, 2442–2448.
46 Amit AG, Mariuzza RA, Phillips SE & Poljak RJ
(1986) Three-dimensional structure of an antigen–anti-
body complex at 2.8 A
˚
resolution. Science
233, 747–
753.
47 Laver WG, Air GM, Webster RG & Smith-Gill SJ
(1990) Epitopes on protein antigens: misconceptions
and realities. Cell 61, 553–556.
48 Jones S & Thornton JM (1996) Principles of protein–
protein interactions. Proc Natl Acad Sci USA 93, 13–
20.

49 Pigott CR, King MS & Ellar DJ (2008) Investigating
the properties of Bacillus thuringiensis Cry proteins with
novel loop replacements created using combinatorial
molecular biology. Appl Environ Microbiol 74, 3497–
3511.
50 Shinkawa A, Yaoi K, Kadotani T, Imamura M,
Koizumi N, Iwahana H & Sato R (1999) Binding of
phylogenetically distant Bacillus thuringiensis cry toxins
S. Atsumi et al. BtR175 binding site on the Cry1Aa toxin
FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS 4925
to a Bombyx mori aminopeptidase N suggests impor-
tance of Cry toxin’s conserved structure in receptor
binding. Curr Microbiol 39, 14–20.
51 Wilson MB & Nakane PK (1976) The covalent coupling
of proteins to periodate-oxidized sephadex: a new
approach to immunoadsorbent preparation. J Immunol
Methods 12, 171–181.
52 Laemmli UK (1970) Cleavage of structural proteins
during the assembly of the head of bacteriophage T4.
Nature 227, 680–685.
53 Nygren H, Werthen M & Stenberg M (1987) Kinetics
of antibody binding to solid-phase-immobilised antigen.
Effect of diffusion rate limitation and steric interaction.
J Immunol Methods 101, 63–71.
BtR175 binding site on the Cry1Aa toxin S. Atsumi et al.
4926 FEBS Journal 275 (2008) 4913–4926 ª 2008 The Authors Journal compilation ª 2008 FEBS