Báo cáo khoa học: Analysis of the region for receptor binding and triggering of oligomerization on Bacillus thuringiensis Cry1Aa toxin potx

Analysis of the region for receptor binding and
triggering of oligomerization on Bacillus thuringiensis
Cry1Aa toxin
Fumiaki Obata, Madoka Kitami, Yukino Inoue, Shogo Atsumi, Yasutaka Yoshizawa and Ryoichi Sato
Graduate School of Bio-Applications and Systems Engineering, Tokyo University of Agriculture and Technology, Japan
Introduction
Bacillus thuringiensis (Bt) is a Gram-positive soil bacte-
rium that produces insecticidal proteins, called Cry
toxins, during sporulation. Various Bt formulations
have been used as pesticides, and genetically modified
crops carrying Cry genes have been developed [1,2].
However, several problems are associated with the
Keywords
Bacillus thuringiensis; Bombyx mori;
BtR175; Cry1Aa; oligomerization
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 4 June 2009, revised 3 August
2009, accepted 12 August 2009)
doi:10.1111/j.1742-4658.2009.07275.x
The determination of the receptor-binding region of Cry toxins produced
by Bacillus thuringiensis is expected to facilitate an improvement in their
insecticidal ability through protein engineering. We analyzed the region on
Cry1Aa molecules involved in interactions with the cadherin-like protein
receptor BtR175 using cysteine-substituted mutant toxins and several syn-

thetic peptides corresponding to the loops in domain 2. In addition, the
region necessary to trigger oligomerization was analyzed using these
mutant toxins. The mutant toxins were modified by two types of molecule,
i.e. digested fragments of the Cry1Aa precursor with an average molecular
mass of 2 kDa and 5-iodoacetamidofluorescein, which has a molecular
mass of 515 kDa. We examined whether these modifications interfere with
the toxin–BtR175 interaction as a result of steric hindrance. 5-Iodoacetami-
dofluorescein modification of R311C, N376C and G442C revealed steric
hindrance effects, indicating that R311 on loop 1, N376 on loop 2 and
G442 on loop 3 are on the contact face of the toxin–BtR175 interface when
Cry1Aa binds to BtR175. Loop 2 is thought to interact with BtR175
directly, as a peptide corresponding to the N-terminal half of loop 2,
(365)LYRRIILG(372), has the potential to bind to BtR175 fragments.
Meanwhile, mutant toxins with cysteine substitutions in loops 1 and 2 were
oligomerized by the binding of digested fragments in the activation process
without receptor interaction, and the wild-type toxin formed oligomers by
interaction with BtR175 fragments. These observations suggest that loops 1
and 2 form both a binding region and a sensor region, which triggers toxin
oligomer formation.
Structured digital abstract
l
MINT-7259673, MINT-7259722, MINT-7259737, MINT-7259757, MINT-7259774, MINT-
7259791, MINT-7259808, MINT-7259685, MINT-7259707, MINT-7259830: btr175
(uniprotkb:
Q9XY09) binds (MI:0407)tocry1Aa (uniprotkb:P0A366)bysurface plasmon
resonance (
MI:0107)
Abbreviations
APN, aminopeptidase N; BBMVs, brush border membrane vesicles; Bt, Bacillus thuringiensis; BtR175, 175 kDa cadherin-like
protein from Bombyx mori; DEAE, diethylaminoethyl; DF, digested fragments of Cry1Aa precursor; GST, glutathione S-transferase;

IAF, 5-iodoacetamidofluorescein.
FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS 5949
practical use of Cry toxins, such as the emergence of
resistant insects with continuous implementation and
the difficulty of screening new Bt strains for pests to
which no identified Cry toxin has sufficient toxicity
[3–5]. To overcome these problems, attempts have
been made to improve Cry toxins artificially [6].
The mechanism of insecticidal activity of Cry toxins
has not been elucidated fully and several conflicting
hypotheses have been proposed [7,8]. One hypothesis
suggests that Cry toxins form a pore on the target cell
membrane, as described below. Precursors of Cry tox-
ins are activated as a result of partial digestion by pro-
teases in the gut digestive juice of susceptible larvae
[9]. Activated Cry1A binds to the cadherin-like protein
receptor expressed on the brush border membrane
(BBM) of epithelial cells of the midgut [10,11]. Cry1A
toxins undergo a conformational change with a1-helix
cleavage and oligomerization [12,13]. Cry1A toxin olig-
omers bind to GPI-anchored receptors [e.g. amino-
peptidase N (APN) or alkaline phosphatase (ALP)],
followed by pore formation on the cell membrane
[14,15]. This induces osmotic swelling and bursting of
the cell, leading to disruption of the midgut tissue and
death of the insect [16].
The three-dimensional structure of Cry1A toxins has
been revealed by X-ray crystallography [17]. Cry1Aa
toxin is a 60 kDa protein having three domains:
domain 1 consists of a bundle of seven helices (a2is

separated into two helices, a2a and a2b, by a short
loop); domain 2 has a b-prism structure composed of
antiparallel b-sheets and loops; and domain 3 has a
b-sandwich structure composed of two antiparallel
b-sheets. Domain 2, especially some of its loops, has
been shown to play a critical role in receptor binding
[18,19].
Many analyses of the receptor-binding region have
been performed using various Cry toxin mutants. For
example, mutations of loops 1, 2 and 3 of Cry1Aa
have been found to affect significantly the binding
activity to BBM vesicles (BBMVs) and the toxicity to
Bombyx mori [19–21]. Similarly, such loops in
Cry1Ab, Cry1Ac and Cry1C are also important in
binding to the BBMVs of target insects [20,22,23].
Moreover, the binding regions for individual receptor
proteins included in BBMVs have been reported. For
example, the two arginines, 368 and 369, conserved in
loop 2 of Cry1Ab and Cry1Ac, are associated with
binding to APN [24]. Two arginine residues localized
in a8 of Cry1Ac have also been indicated to be impor-
tant in binding to each APN from Manduca sexta and
Lymantria dispar [25]. In addition, alanine mutations
of amino acid residues in loop 2 of Cry1Aa and
Cry1Ab reduced binding to cadherin-like protein BtR1
and APN from M. sexta and Heliothis virescens [22].
However, we cannot exclude the possibility that the
loss of binding ability is derived from disruption of the
toxin conformation caused by the mutation introduced
distant from the receptor-binding site. Indeed, muta-

tions in the region distant from the binding site cause
unsuspected changes in binding ability in the case of
antigen–antibody interactions [26].
To determine the binding region directly without
amino acid substitutions, loop regions were produced as
synthetic peptides and their binding abilities were inves-
tigated. Synthetic peptides corresponding to loops 2
and 3 inhibited the interaction of Cry1Aa toxins and
BtR1 from M. sexta, whereas synthetic peptides corre-
sponding to loops a8 and 2 blocked Cry1Ab toxins
from binding to BtR1 [27,28]. In addition, synthetic
peptides corresponding to loop 3 interrupted the bind-
ing of Cry1Ac toxins to the H. virescens cadherin-like
protein, HevCaLP [29]. These experimental data suggest
that loops a8, 2 and 3 may be associated with receptor
binding, but the interaction between Cry1Aa toxins and
BtR175 has not been examined to date.
Recently, we have reported the establishment of a
system to improve Cry toxins in vitro using phage dis-
play, which is a well-known tool for directed molecular
evolution [30]. In this system, random mutations are
introduced into the toxins and mutants are displayed
on phages to prepare a phage library. Then, mutant
toxins with high affinities to the receptor are screened
using high-throughput selection systems, such as
panning. As the affinity of Cry toxins for receptors is
correlated with toxicity [31], some of the mutants with
high affinities to the receptor should also show strong
insecticidal activity. Studies on molecular evolutionary
engineering of antibodies have suggested that the

induction of mutations in or close to the region inter-
acting with the receptor is effective in generating high-
affinity mutant toxins [26]. Therefore, the analysis of
the receptor-binding region in more detail is needed to
determine which region(s) should be mutated. In this
experiment, we adopted three approaches: (a) the con-
struction of cysteine-substituted mutant toxins and the
analysis of their binding activity to the receptor to
investigate the role of mutated amino acid residues; (b)
the analysis of the influence of the modification of the
cysteine residues of mutants with 5-iodoacetamidofluo-
rescein (IAF), which induces steric hindrance; and (c)
the analysis of synthetic peptides that can mimic the
binding function of Cry1Aa toxin loop regions to the
receptor. Using these analyses, we can estimate a
receptor-interacting region and a region that is in close
proximity to the receptor at the toxin–receptor-binding
interface.
Receptor binding ⁄ oligomerization of Cry1Aa mutants F. Obata et al.
5950 FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS
In addition, we found that a proportion of the cyste-
ine-substituted mutant toxins oligomerized spontane-
ously during the process of activation by trypsin.
Oligomerization is considered to be an important step
in the insecticidal activity of Cry toxins [32]. Several
studies using Cry1Ab have indicated that it forms a
pre-pore structure before insertion, and various Cry
toxin oligomers have been analyzed [13,32–34]. For
Cry1Aa, the oligomer toxins triggered by BBMVs from
M. sexta and B. mori have been analyzed [35,36], but

the type of molecule and the region(s) on the Cry1Aa
toxin that is involved in toxin oligomerization remain
unclear. Therefore, we dissected the mechanism by
which Cry1Aa toxins undergo oligomerization using
these mutant toxins.
Results
Production and activation of cysteine-substituted
mutant toxins and their modification
Precursors of cysteine-substituted mutant toxins,
R281C, Q293C, R311C, S373C, N376C, G442C and
Y445C, were all produced in Escherichia coli as gluta-
thione transferase (GST)-fusion recombinant proteins.
These toxins were activated by trypsin in the anion
exchange column and eluted as purified activated
toxins, as reported by Nagamatsu et al. [37]. The acti-
vated mutant toxins were then analyzed by SDS-
PAGE under nonreducing conditions, and a band
slightly larger than the wild-type toxin and an addi-
tional smear were observed (Fig. 1A). This larger
band and smear shifted to the same size as the wild-
type toxin on reduction with dithiothreitol (Fig. 1A),
indicating that certain peptides were bound to the
cysteine-substituted mutant toxins via the disulfide
bonds of cysteine residues. In addition, the molecular
masses of the peptides isolated from a mutant toxin
treated with dithiothreitol were analyzed using Tricine
SDS-PAGE, and the results indicated a broad band
spreading from 1 to 4 kDa (data not shown). As
described in the Discussion section, these peptides
were suggested to be digested fragments (DFs)

derived from protoxin, and presumably convenient to
use as sources of steric hindrance. Therefore, we used
these DF-bound toxins, as well as IAF-modified
toxins, to analyze the site-specific effects of steric
hindrance. Conversely, nonmodified mutant toxins
were prepared by dithiothreitol reduction of DF-
bound purified toxins. After the reduction and
removal of the DFs, IAF modification was performed
and checked by western blotting using an anti-fluores-
cein IgG (Fig. 1B).
Quantitative binding analysis of
cysteine-substituted mutant toxins
First, we performed binding analyses of nonmodified
mutant toxins to BtR175 fragments immobilized on
the cuvette of an IAsys optical biosensor. Figure 2
shows the overlaid binding curves of the wild-type
toxin and all nonmodified mutant toxins. The
mutant toxins R281C, Q293C and G442C showed
almost the same levels of binding to BtR175,
whereas Y445C exhibited lower binding activity than
the wild-type toxin. In contrast, R311C (localized on
loop 1), S373C and N376C (both localized on
loop 2) showed much higher binding activity than
the wild-type toxin.
Oligomerization of cysteine-substituted mutant
toxins
Although the affinity of Cry1Aa toxins to BtR175 may
be improved by cysteine substitution, the generation of
mutants with improved affinity at such high rates
seems unlikely. Therefore, we postulated that these

mutants had oligomerized during preparation. If
60 kDa
Wt
R281C S373CR311C
DTT
+–
+–
+–+–
A
B
R281C
Wt
a
b
1
2
31
2
3
Fig. 1. DF and IAF modifications of cysteine-substituted mutant
toxins. (A) Cysteine-substituted mutant toxins were subjected to
nonreducing SDS-PAGE with (+) or without ()) dithiothreitol (DTT)
treatment. Wt, wild-type toxin. (B) Coomassie brilliant blue staining
(a) and western blotting using an anti-fluorescein IgG (b) were per-
formed after SDS-PAGE of R281C and the wild-type toxin, which
were reduced with dithiothreitol and then modified with IAF. IAF
modification was specific for the mutant toxins. Lane 1, no treat-
ment (DF-bound toxin); lane 2, dithiothreitol reduction (nonmodified
toxin); 3, IAF reaction (IAF-modified toxin).
F. Obata et al. Receptor binding ⁄ oligomerization of Cry1Aa mutants

FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS 5951
oligomerization occurs, the apparent amount of bound
toxin increases because several toxin molecules can
bind to one receptor.
First, we characterized the wild-type Cry1Aa toxin
oligomer. After incubation of the toxin with BBMVs,
the band with a molecular mass exceeding 200 kDa
was detected with the 60 kDa monomer by western
blotting using an anti-Cry1Aa serum (Fig. 3A). From
the size, we speculated that the protein correspond-
ing to the 200-kDa band had an oligomeric (perhaps
tetrameric) structure, formed by the assembly of
Cry1Aa toxin monomers. The oligomeric structure
increased between 15 min and 2 h of incubation with
BBMVs, and dissociated to the monomer with treat-
ment at 70 °C or above in SDS-PAGE sample buffer
(Fig. 3A). Furthermore, we showed that oligomeriza-
tion was induced by the BtR175 fragments used for
the binding analysis to an extent similar to BBMVs
(Fig. 3B).
Next, we analyzed the mutant toxin oligomers that
showed higher binding to BtR175 than the wild-type
toxin (Fig. 2). The wild-type toxin and R311C were
analyzed before and after incubation with BBMVs.
For the wild-type toxin, the oligomeric band was
observed only after incubation, whereas it was
observed even before triggering by BBMVs for the
R311C mutant toxin (Fig. 3C). This band disappeared
almost entirely on 5 min of incubation at 95 °Cin
SDS-PAGE sample buffer, similar to the wild-type

oligomeric band (data not shown). These data clearly
show that R311C forms oligomers without any interac-
tion with BBMVs. Next, we analyzed all the mutant
toxins to determine whether oligomerization had
already occurred before interaction with BBMVs, as in
R311C (Fig. 3D). S373C and N376C mutants demon-
strated strong oligomeric bands. In contrast, although
R281C, G442C and Y445C mutants showed oligomeri-
zation, their ratios were quite small. We then examined
the N-terminal sequence of the oligomeric band using
S373C, and obtained the two sequences GIFGP and
GPSQW. These are sequences from residues 66 and
69, respectively, both of which are located in the loop
A
B
200
116
97
66
12 3
CD
200
116
97
66
3412 567
123456
200
116
97

66
Temp.(°C)
50 50 30 50 70 100
Wt R311C
200
116
97
66
BBMV
+
–+
–
Fig. 3. Oligomerization of Cry1Aa wild-type and cysteine-substi-
tuted mutant toxins. (A) Oligomerization of Cry1Aa wild-type toxin
triggered by BBMVs. Aliquots of 20 lL of 100 n
M of wild-type
Cry1Aa were mixed with 5 lg of BBMVs and incubated for 15 min
or 2 h at 37 °C. Samples were then mixed with SDS-PAGE sample
buffer and incubated for 5 min at various temperatures, as
indicated above the figure. Each sample was subjected to 7.5%
SDS-PAGE and transferred onto nitrocellulose membranes, and the
toxins were visualized by western blotting using anti-Cry1Aa serum.
Lane 1, without BBMVs; lane 2, 15 min of incubation with BBMVs;
lanes 3–6, 2 h of incubation with BBMVs. (B) Oligomerization of
Cry1Aa wild-type toxin triggered by BtR175 as well as BBMVs.
Aliquots of 20 lLof10n
M Cry1Aa were mixed with 0.5 lgof
BBMVs, BtR175 or BSA, and incubated for 2 h at 37 °C. Then, the
samples were mixed with SDS-PAGE sample buffer and incubated
for 5 min at 50 °C. The oligomeric structure was detected by wes-

tern blotting as in (A). Lane 1, BBMVs; lane 2, BtR175 fragments;
lane 3, BSA. (C) Oligomerization of R311C. Oligomerization of
R311C was analyzed with or without BBMV triggering under the
same conditions as in (B). Wt, wild-type toxin. (D) Western blotting
of each cysteine-substituted mutant toxin after HPLC purification
without BBMV triggering using anti-Cry1Aa serum. Lane 1, R281C;
lane 2, Q293C; lane 3, R311C; lane 4, S373C; lane 5, N376C; lane
6, G442C; lane 7, Y445C.
–50
50
150
250
–50 150 350 550 750
Time (s)
Resonance (arc s)
Wild type
Y445C
G442C
R281CQ293C
R311C
S373C
N376C
Fig. 2. Binding analysis of cysteine-substituted mutant toxins to
the BtR175 fragment using an IAsys optical sensor. Each cysteine-
substituted mutant toxin was added at 150 n
M to the cuvette on
which the BtR175 fragment was immobilized, and the binding
curve was analyzed for 10 min. After changing the solution to buf-
fer, the dissociation curve was analyzed for 4 min. For wild-type
Cry1Aa, the binding and dissociation curves were analyzed for 5

and 1 min, respectively.
Receptor binding ⁄ oligomerization of Cry1Aa mutants F. Obata et al.
5952 FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS
between a-helices 2a and 2b. Therefore, at least for
S373C, the oligomerized toxin was cleaved at this posi-
tion, and a-helices 1 and 2a were ablated.
Binding analysis of DF-bound toxins and
IAF-modified toxins
Regardless of whether the toxin does or does not oli-
gomerize, modification at cysteine residues can affect
the binding affinity of the BtR175 fragment as a result
of steric hindrance. Therefore, we next analyzed the
effects of DF and IAF modifications, which show
greater and smaller hindrance, respectively, for the
mutant toxins (Fig. 4).
First, we focused on R281C, Q293C and G442C, all
of which showed almost the same binding curve as the
wild-type toxin (Fig. 2). For R281C, a decrease in
binding was seen in the DF-bound form, but not in
the IAF-modified form. Although even the DF-bound
form of Q293C demonstrated high binding ability,
G442C showed reduction of binding in both DF-
bound and IAF-modified forms. Next, we analyzed
R311C, S373C and N376C, all of which showed
increased binding compared with the wild-type toxin
(Fig. 2). Unexpectedly, increased binding was observed
in the DF-bound form of R311C, although the IAF-
modified form showed reduced binding capability.
S373C displayed decreased binding only in the DF-
bound form and not in the IAF-modified form, similar

to R281C, whereas N376C showed decreased binding
in both DF-bound and IAF-modified forms, similar to
G442C. In addition, the binding curve of DF-bound
Y445C was the same as that of the nonmodified form,
which showed lower affinity (data not shown).
Analysis of the binding ability of synthetic
peptides
To evaluate the binding capability of the loop regions
of Cry1Aa toxins directly, we used eight synthetic pep-
tides corresponding to seven loops (Fig. 5A, Table 1):
loop a8 [(276)FDGSFRGM(283)], loop 1 [(311)
RG(312)], loop 2 [(368)RIILGSGPNNQ(378)], loop 3
[(439)QAAGAVY(445)], loop a8–b2 [(290)NIR-
QPHLM(297)], loop b4–b5 [(336)PLFGNAGNAAPP
(347)] and loop b8–b9 [(404)QRG(406)]. The first four
of these loops have been reported to be associated with
receptor binding. Loop 2 was divided into two halves
because it is longer than the others. The abilities of
these eight peptides to bind to BtR175 fragments were
analyzed by IAsys, and only P5-1, corresponding to
the first half of loop 2, showed a higher binding activ-
ity (Fig. 5B).
Discussion
DFs bind spontaneously to cysteine-substituted
mutant toxins in the activation process
We used cysteine-substituted mutant toxins to analyze
the receptor-binding region. Cysteine was chosen
because of the possibility of directed modification by
reaction with the thiol group in a protein with no
other free thiol groups. By substitution of a certain

residue for cysteine, we can obtain a mutant toxin with
only one cysteine, as activated wild-type Cry1Aa
–5
10
25
40
55
–50 200 700450
–50 200 700450
–50 200 700450
R281C
DF
IAF
Non–labelled
–5
10
25
40
55
–50 200 450 700
–50 200 450 700
–50 200 450 700
Q293C
DF
Non–labelled
R311C
–30
120
270
420

Non–labelled
DF
IAF
–10
20
50
80
110
DF
IAF
Non–labelled
S373C
N376C
–30
45
120
195
270
DF
IAF
Non–labelled
G442C
–5
10
25
40
DF
IAF
Non–labelled
Time (s)

Time (s)
Resonance (arc s)
Resonance (arc s)Resonance (arc s)
Fig. 4. Influence of DF and IAF modifications of cysteine-substi-
tuted mutant toxins on binding to BtR175. Nonmodified, DF-bound
or IAF-modified toxin was added at a concentration of 150 n
M to
the cuvette on which BtR175 fragments were immobilized, and the
binding curve was analyzed for 10 min. IAF-modified Q293C was
not analyzed because DF-bound toxin showed no difference from
nonmodified toxin.
F. Obata et al. Receptor binding ⁄ oligomerization of Cry1Aa mutants
FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS 5953
contains no cysteine. Mutant toxins formed a broad
band slightly larger than that of the wild-type toxin
after activation and purification by anion-exchange di-
ethylaminoethyl (DEAE)-HPLC (Fig. 1A). In addition,
reduction of these toxins with dithiothreitol resulted in
the elution of peptides of 1–4 kDa (data not shown).
The proteins present in a DEAE column at the time of
activation by trypsin were only the precursors of
mutant toxins and trypsin. Trypsin is a protease that
cleaves specifically at the C-terminal end of basic
amino acids. From the amino acid sequence of
Cry1Aa, various types of DFs with cysteine and a
molecular mass of 1–4 kDa would be expected to be
produced from the C-terminal half of Cry1Aa precur-
sors by tryptic digestion. Therefore, peptides bound to
cysteine-substituted mutant toxins were inferred to be
DFs derived from the C-terminal half of the toxin pre-

cursor. In addition, we analyzed the N-terminal
sequences of R281C and Q293C after HPLC purifica-
tion, and obtained the sequence of the activated toxin,
which begins from amino acid 29. However, the
sequences of peptides expected to bind to the cysteine-
substituted mutant toxins were not identified. These
data suggest that various DFs were bound to the
mutant toxins. Toxin-binding DFs seem to range in
size from 1 to 4 kDa, as observed on SDS-PAGE
(Fig. 1A, data not shown), suggesting that they can
exert wider steric hindrance effects than IAF. That is,
in comparison with DF, the inhibitory effect of IAF
should be restricted to a narrower region.
The region on Cry1Aa toxins for contact and
interaction with BtR175
From the results of the IAsys binding analysis, the cys-
teine substitutions at R281, Q293 and G442 did not
seem to influence the conformation of Cry1Aa toxins,
as the binding abilities of these mutants to BtR175
were no different from that of the wild-type toxin
(Fig. 2). Three cysteine-substituted mutant toxins,
R311C, S373C and N376C, which showed apparent
increases on BtR175 binding, were demonstrated to
have adopted an oligomeric structure during the acti-
vation process (Fig. 3D). However, analyses of the
contact region or the region in close proximity to
BtR175 at the toxin–receptor-binding phase were
expected to be implemented without a negative influ-
ence of oligomerization by analyzing the effects of
modification of cysteine residues. Therefore, we exam-

ined the regions that influenced the binding ability to
BtR175 by DF and IAF modifications. N376 located
on loop 2 and G442 located on loop 3 were indicated
to be in the contact face in the BtR175-binding phase,
P1
P7
P4
P3
P2
P6
P5-1
P3
P5-1
P5-2
P6
P7
P1
P4
P2
P5-2
–50
Resonance (arc s)
0
75
150
225
B
A
100 400250 550
Time (s)

Fig. 5. Binding analysis of synthetic peptides to BtR175 fragments.
(A) Locations of the synthetic peptides. P5-1, which is indicated high-
est binding property binding to BtR175 fragments in (B), is shown as
a dot model. (B) Binding analysis of synthetic peptides by IAsys.
Each synthetic peptide was added to the cuvette at 1 m
M, and the
binding curve was analyzed until it reached equilibrium.
Table 1. Sequences of synthetic peptides and corresponding loop
regions.
Synthetic peptide Loop Sequence
P1 a8 (276)FDGSFRM(283)
P2 a8–b2 (290)NIRQPHLM(297)
P3 1 (309)VHRGFN(314)
P4 b4–b5 (338)FGNAGNAAP(346)
P5-1 2 (365)LYRRIILG(372)
P5-2 2 (373)SGPNNQEL(380)
P6 b8–b9 (401)IYRQRGTV(408)
P7 3 (439)QAAGAVY(445)
Receptor binding ⁄ oligomerization of Cry1Aa mutants F. Obata et al.
5954 FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS
as both DF and IAF modifications showed significant
steric hindrance (Fig. 4). Unlike these two mutants,
the DF modification of R311C increased the apparent
binding ability to BtR175 (Fig. 4). Although the actual
reason is still unknown, bound DFs may have assisted
in R311C binding to BtR175. In contrast, IAF modifi-
cation inhibited the binding of R311C (Fig. 4). Both
phenomena may indicate that R311 is in close proxim-
ity to the BtR175 molecule in the toxin–receptor-bind-
ing phase. In addition, R281 located on loop a8 and

S373 located on loop 2 were shown to be in relatively
close proximity to the BtR175 molecule, but not criti-
cally in the contact face of the binding phase, because
only the DF modification inhibited the interaction
(Fig. 4). By mapping cysteine-substituted amino acid
residues on the Cry1Aa toxin three-dimensional struc-
ture, R311, N376 and G442 were located relatively
close to each other and R281 and S373 were not far
from these residues (Fig. 6). Previously, we have
reported that the anti-Cry1Aa toxin monoclonal anti-
body 2A11, which inhibits the binding of Cry1Aa
toxin to BtR175, appears to have an epitope in the
region consisting of G372, S373, G439, A440, G442,
Y445 and T445 [38], which is consistent with the
results of the present study.
As the Y445C mutant loses its binding ability as a
result of a cysteine-substituted mutation (Fig. 2) and is
located in the putative binding region between Cry1Aa
toxin and BtR175 (Fig. 6), Y445 may have a critical
role in binding to BtR175. However, the observation
that synthetic peptide P7, which includes Y445, has
little affinity to BtR175 (Fig. 5) suggests that Y445
substitution for cysteine may alter the conformation of
the toxin. Moreover, P5-1 showed higher binding
ability compared with the other peptides (Fig. 5). This
observation strongly suggests that a region involved
in interactions with BtR175 exists in (365)LYR-
RIILG(372) of loop 2. In a previous study, an alanine
substitution or deletion of 365–372 in loop 2 of
Cry1Aa toxins resulted in 1000-fold less toxicity than

the wild-type toxin to B. mori [19]. In addition, alanine
or glutamic acid substitution of two arginine residues
at positions 368 and 369 of Cry1Ab and Cry1Ac tox-
ins decreased the toxicity to L. dispar and M. sexta
[24]. As lysine substitution of these two arginine resi-
dues showed no effect on toxicity or receptor-binding
ability, a basic amino acid at this site seems to be
important [24]. Only a few residues are conserved
within loop 2 of Cry1Aa, Cry1Ab and Cry1Ac,
although two arginine residues are conserved. This
observation also indicates that the N-terminal half of
loop 2 plays a key role in the interaction between
Cry1Aa toxins and BtR175.
Oligomerization of cysteine-substituted mutant
toxins without receptor binding
Previously, the oligomer of a Cry toxin has been
reported to show tolerance to heating at 100 °C [13].
However, we could not detect oligomers under such
temperature conditions, and thus oligomers were
thought to degrade in the SDS-PAGE sample buffer at
high temperatures (data not shown). We therefore
optimized the conditions to observe the oligomers by
SDS-PAGE. The oligomer was shown to decompose
above 70 °C in the sample buffer (Fig. 3A), which was
consistent with the results of Ihara and Himeno [36].
We analyzed the oligomeric form of the Cry1Aa wild-
type toxin using these conditions, and showed that
almost all of the toxin molecules existed in monomeric
form after HPLC purification, and oligomerization
R281C

S373C
Y445C
G442C
R311C
Q293C
N376C
R281C
S373C
Y445C
G442C
R311C
Q293C
N376C
Loop α8
Loop 3
Loop 2
Loop 1
Fig. 6. Mapping of the mutation site. Red indicates the amino acid
residues that showed inhibition of BtR175 binding by DF binding.
Yellow indicates the amino acid residues that showed inhibition of
BtR175 binding by both DF binding and IAF modification. Blue indi-
cates the amino acids that showed inhibition of BtR175 binding
without modification. Light gray indicates the amino acids that did
not inhibit BtR175 binding even with DF binding. Top, cartoon
model; bottom, surface model.
F. Obata et al. Receptor binding ⁄ oligomerization of Cry1Aa mutants
FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS 5955
was triggered by cadherin-like protein BtR175 or
BBMVs from B. mori (Fig. 3B).
R311C, as well as S373C and N376C, spontaneously

oligomerized during the activation process in DEAE-
HPLC without interaction with BtR175 or BBMVs, in
contrast with the wild-type (Fig. 3C, D). Cry1Aa tox-
ins contain two helices, a2a and a2b, the C-termini of
which are located at residues 63 and 71, respectively
[17]. Although only a1 was cleaved when Cry1Ab was
oligomerized and the N-terminal 60 residues were
removed when Cry1Ac was oligomerized [13,39], a1
and a2a were ablated from the N-terminus, resulting
in the removal of more than 65 residues when Cry1Aa
toxins were oligomerized (this study).
Considering the hypothesis of Go
´
mez et al. [13],
the oligomerization of mutant toxins requires inter-
action with the receptor or a receptor-like molecule
at the time of tryptic digestion in the HPLC column.
However, no such receptor-like molecule was present
in the column for wild-type toxins, as the wild-type
Cry1Aa toxin did not oligomerize in this process. In
addition, if cysteine-substituted mutant toxins were
linked to each other via disulfide bonds, the toxins
should have formed dimers. The oligomers, however,
were suggested to be tetramers (Fig. 3), indicating a
different mechanism of oligomerization. In addition,
DFs bound to the mutant toxins via thiol in the
process of activation (Fig. 1A). Therefore, we
concluded that DFs mimicked the receptor and
triggered the oligomerization of cysteine-substituted
mutant toxins. Cysteine-substituted mutant toxins

may expose helices a1 and a2a of domain 1 on cova-
lent binding to DFs on their cysteine residues, and
the two helices may be subsequently ablated by tryp-
sin, followed by the aggregation of toxin monomers.
Indeed, the triggering of oligomerization by molecules
other than the original receptor has been reported
previously: for example, the recombinant monoclonal
antibody scFV73, which inhibits Cry1Aa toxin
binding to BtR1, and Cyt1Aa, which binds to the
receptor-binding region of Cry11Aa [13,40]. These
observations indicate that a conformational change
and subsequent oligomerization can be triggered if
some proteins bind to the receptor-binding region of
Cry toxin. Thus, R311, S373 and N376, or loops 1
and 2 of the Cry1Aa toxin, may be located on a
receptor-sensing region that promotes oligomerization.
That is, these amino acid residues or structures may
be present in the receptor-binding region. Indeed,
even IAF-modified toxins R311C and N376C showed
decreased binding ability (Fig. 4). In contrast, cyste-
ine-substituted mutant toxins of loop a8, loop 3 and
other parts, R281C, Q293C, G442C and Y445C,
showed little oligomer formation without receptor
interaction (Fig. 3D).
Application to protein engineering
This research has yielded two important findings
related to the improvement of Cry toxin activity. First,
the results of the present study reveal that the region
on the binding interface of the Cry1Aa molecule may
be responsible for binding. Our data suggest that the

binding region is located on or around (365)LYR-
RIILG(372), R311, N376, G442 and Y445 (Figs 2, 4
and 5), and therefore loops 1, 2 and 3 (especially
loop 2) are good candidate regions for the introduction
of mutations to obtain mutant toxins with a high affin-
ity to the cadherin-like receptor BtR175 of B. mori.
Indeed, improvement of these loops by mutation has
been performed for several Cry toxins. For example,
loop 2 mutants of Cry1Ab and loop 1 mutants of
Cry3Aa have higher insecticidal activity than the wild-
type [41,42]. Moreover, a Cry4Ba loop 3 mutant,
which mimics loop 3 of Cry4Aa, showed toxicity for
Culex pipiens, which is not susceptible to Cry4Ba but
to Cry4Aa, and this mutant showed binding ability to
a protein considered to be a receptor [43]. Likewise,
changing loops 1 and 2 of the lepidopteran-specific
Cry1Aa toxin to those of the Cry4 toxin resulted in
mosquitocidal toxicity [44]. In this way, mutations in
loops of the putative binding region have been shown
to enable improvement or alteration of the toxicity of
Cry toxins. Therefore, we expect the application of this
knowledge in the phage display system of Cry toxins
described in the Introduction. To date, a mutant with
high affinity to BtR175 has been isolated from a loop
2 random mutant library [30].
The other important result of this study is that tox-
ins oligomerize spontaneously without receptor inter-
action. Sobero
´
n et al. [45] have generated a deletion

mutant of a1 of Cry1Ab that oligomerizes spontane-
ously. This mutant toxin showed toxicity to Cry1Ab-
resistant Pectinophora gossypiella with deletion of the
cadherin receptor, and to BtR1-silenced M. sexta [45].
These results indicate that a cadherin-like protein is
necessary for the oligomerization of toxins, and if the
oligomeric structure is constructed spontaneously,
resistance by cadherin-like protein deletion can also
be overcome. Cysteine-substituted mutant toxins in
this study also oligomerized spontaneously without
any interaction with cadherin-like proteins, and there-
fore these toxins are expected to be applicable as
insecticides or proteins expressed in transgenic plants
for the control of insects with cadherin-like protein
mutations.
Receptor binding ⁄ oligomerization of Cry1Aa mutants F. Obata et al.
5956 FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS
Materials and methods
Preparation of Cry1Aa cysteine-substituted
mutant toxins and BtR175 fragment
Wild-type Cry1Aa protoxin was expressed as described pre-
viously [38]. Single cysteine mutants were constructed by
site-directed mutagenesis using primers designed to substi-
tute R281, Q293, R311, S373, N376, G442 and Y445 for
cysteine (R281C-S, 5¢-TTGATGGTAGTTTTTGTGGAAT
GGCTCAGA-3¢; R281C-AS, 5¢-TCTGAGCCATTCCACA
AAAACTACCATCAA-3¢; Q293C-S, 5¢-ACCAGAATATT
AGGTGTCCACATCTTATGGA-3¢; Q293C-AS, 5¢-TCC
ATAAGATGTGGACACCTAATATTCTGT-3¢; R311C-S,
5¢-ATACTGATGTGCATTGTGGCTTTAATTATT-3¢;R3

11C-AS, 5¢-AATAATTAAAGCCACAATGCACATCAGT
AT-3¢; S373C-S, 5¢-AATTATACTTGGTTGTGGCCCAA
ATAATCA-3¢; S373C-AS, 5 ¢-TGATTATTTGGGCCACA
ACCAAGTATAATT-3¢; N376C-S, 5¢-TTGGTTCAGGCC
CATGTAATCAGGAACTGT-3¢; N376C-AS, 5¢-ACAGTT
CCTGATTACATGGGCCTGAACCAA-3¢; G442C-S, 5¢-
TGAGCCAAGCAGCTTGTGCAGTTTACACCT-3¢; G44
2C-AS, 5¢-AGGTGTAAACTGCACAAGCTGCTTGGCT
CA-3¢; Y445C-S, 5¢-AGCTGGAGCAGTTTGTACCTTGA
GAGCTCC-3¢; Y445C-AS, 5¢-GGAGCTCTCAAGGTACA
AACTGCTCCAGCT-3¢). The mutated DNAs were cloned
into a GST-tagged expression vector, pGEX-4T3 (GE
Healthcare, Little Chalfont, UK), and used for the transfor-
mation of the BL21 strain of E. coli. Each mutant toxin was
expressed, activated and purified as described previously [38].
Briefly, E. coli, which produces each toxin, was cultured in
MMI broth with ampicillin at 37 °C, and gene expression
was induced by isopropyl thio-b- d-galactoside. The inclusion
body of protoxin was harvested and solubilized, and then
protoxin was applied to a DEAE column (Shodex IEC
DEAE-825, Showa Denko, Tokyo, Japan) connected to an
HPLC system (WatersÔ 600, Milford, MA, USA), and toxin
was activated by 0.5 mgÆmL
)1
trypsin for 2 h at 37 °C in the
column. Activated toxin was eluted using a linear gradient of
Tris ⁄ HCl buffer, and the purity was checked by SDS-PAGE.
The concentration of purified toxin was determined by a
Coomassie protein assay kit (Pierce, Rockford, IL, USA)
using bovine serum albumin as a standard.

A 27 kDa fragment of Glu1108–Val1464, which contains
the toxin-binding region of cadherin-like protein, BtR175
[37], from B. mori, was expressed as a GST-fusion protein
in E. coli and solubilized as described previously [46]. The
BtR175 fragment was then purified using a MagneGST Ô
Protein Purification System (Promega, Madison, WI, USA).
Preparation of BBMV from B. mori
BBMVs from the midgut of fifth instar larvae of B. mori
were prepared according to the method described by
Wolfersberger et al. [47].
Amino acid sequence of mutant toxins and DFs
Mutant toxins with or without DFs were subjected to 10%
SDS-PAGE under nonreducing conditions and transferred
onto a poly(vinylidene difluoride) (PerkinElmer Life Sci-
ences, Boston, MA, USA) membrane. The N-terminal
sequence of each sample was determined by a Procise Ò
cLC protein sequencer (Applied Biosystems, Foster City,
CA, USA).
Analysis of DF binding and IAF modification of
mutant toxins
To identify the binding of DFs after purification, cysteine
mutant toxins were subjected to SDS-PAGE under reduc-
ing and nonreducing conditions. The reducing condition
was created by the addition of 2-mercaptoetanol at a con-
centration of 1% to the SDS-PAGE sample buffer. DFs
were removed from mutant toxins by reduction for 30 min
at 4 °C with 10 mm dithiothreitol in NaCl ⁄ P
i
buffer, and
the reduced toxins were dialyzed in NaCl ⁄ P

i
buffer at 4 °C.
IAF modification was performed according to the manufac-
turer’s instructions. Briefly, IAF was added to a final con-
centration of 10 lm to 1 lm of toxin in NaCl ⁄ P
i
buffer
with 5 mm EDTA, and the solution was incubated for 2 h
at 4 °C with light shielding. Extra IAF was removed by
dialysis in NaCl ⁄ P
i
buffer. Modification was checked by
western blotting using biotinylated anti-fluorescein IgG and
horseradish peroxidase–streptoavidin conjugate.
Analysis of binding to BtR175 fragment by IAsys
optical sensor
Binding analysis using IAsys was conducted according to the
method of Atsumi et al. [38]. Briefly, a 150 nm mutant toxin
solution was added to the cuvette on which BtR175 fragment
had been immobilized, and the binding curve was recorded
for 10 min. Then, the toxin solution was removed by altering
the solution to NaCl ⁄ P
i
, and the dissociation curve was ana-
lyzed for 4 min. The cuvette was regenerated after treatment
with 20 mm HCl, and the association and dissociation of each
mutant toxin were analyzed sequentially. For synthetic pep-
tides, 1 mm of solution was added to the cuvette and the bind-
ing curve was analyzed until the reaction reached equilibrium.
Oligomerization of Cry1Aa toxin and detection of

oligomerized toxin
For the construction of oligomer, 20 lL of a 10 or 100 nm
toxin solution were incubated for 2 h at 37 °C with 0.5 or
5 lg BBMV or BtR175 fragment, and the sample was sub-
jected to western blotting. As the oligomeric structure is
unstable at high temperature in SDS-PAGE sample buffer,
each sample was incubated in the buffer for 5 min at 50 °C.
Oligomer was detected using anti-Cry1Aa toxin serum.
F. Obata et al. Receptor binding ⁄ oligomerization of Cry1Aa mutants
FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS 5957
Acknowledgements
This work was supported by a Grant-in-Aid for Scien-
tific Research (B) (21310051) from the Ministry of
Education, Culture, Sports, Science, and Technology
of Japan.
References
1 Crickmore N (2006) Beyond the spore – past and future
developments of Bacillus thuringiensis as a biopesticide.
J Appl Microbiol 101, 616–619.
2 Romeis J, Meissle M & Bigler F (2006) Transgenic
crops expressing Bacillus thuringiensis toxins and biolog-
ical control. Nat Biotechnol 24, 63–71.
3 Griffitts JS & Aroian RV (2005) Many roads to resis-
tance: how invertebrates adapt to Bt toxins. BioEssays
27, 614–624.
4 Tabashnik BE, Gassmann AJ, Crowder DW & Carrie
´
re
Y (2008) Insect resistance to Bt crops: evidence versus
theory. Nat Biotechnol 26, 199–202.

5 Bravo A & Sobero
´
n M (2008) How to cope with insect
resistance to Bt toxins? Trends Biotechnol 26, 573–579.
6 Pardo-Lo
´
pez L, Mun
˜
oz-Garay C, Porta H, Rodrı
´
guez-
Almaza
´
n C, Sobero
´
n M & Bravo A (2009) Strategies to
improve the insecticidal activity of Cry toxins from
Bacillus thuringiensis. Peptides 30, 589–595.
7 Sobero
´
n M, Gill SS & Bravo A (2009) Signaling versus
punching hole: how do Bacillus thuringiensis toxins kill
insect midgut cells? Cell Mol Life Sci 66, 1337–1349.
8 Pigott CR & Ellar DJ (2007) Role of receptors in Bacil-
lus thuringiensis crystal toxin activity. Microbiol Mol
Biol Rev 71, 255–281.
9 Aronson AI, Wu D & Zhang C (1995) Mutagenesis of
specificity and toxicity regions of a Bacillus thuringiensis
protoxin gene. J Biol Chem 177, 4059–4065.
10 Hofmann C, Lu

¨
thy P, Hu
¨
tter R & Pliska V (1988)
Binding of the d-endotoxin from Bacillus thuringiensis
to brush-border membrane vesicles of the cabbage
butterfly (Pieris brassicae). Eur J Biochem 173,
85–91.
11 Keeton TP, Francis BR, Maaty WS & Bulla LA Jr
(1998) Effects of midgut-protein-preparative and ligand
binding procedures on the toxin binding characteristics
of BT-R1, a common high-affinity receptor in Manduca
sexta for Cry1A Bacillus thuringiensis toxins. Appl Envi-
ron Microbiol 64 , 2158–2165.
12 Gazit E, La Rocca P, Sansom MS & Shai Y (1998) The
structure and organization within the membrane of the
helices composing the pore-forming domain of Bacillus
thuringiensis d-endotoxin are consistent with an
‘umbrella-like’ structure of the pore. Proc Natl Acad Sci
USA 95, 12289–12294.
13 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.
14 Schwartz JL, Lu YJ, Sohnlein 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.
15 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 microdo-
mains. Biochim Biophys Acta 1667, 38–46.
16 Knowles BH & Ellar DJ (1987) Colloid-osmotic lysis is
a general feature of the mechanisms of action of Bacil-
lus thuringiensis d-endotoxins with different insect speci-
ficities. Biochim Biophys Acta 924, 509–518.
17 Grochulski P, Masson L, Borisova S, Pusztai-Carey M,
Schwartz JL, Brousseau R & Cygler M (1995) Bacillus
thuringiensis CryIA(a) insecticidal toxin: crystal struc-
ture and channel formation. J Mol Biol 254, 447–464.
18 Ge AZ, Rivers D, Milne R & Dean DH (1991)
Functional domains of Bacillus thuringiensis insecticidal

crystal proteins. J Biol Chem 266, 17954–17958.
19 Lu H, Rajamohan F & Dean DF (1994) Identification
of amino acid residues of Bacillus thuringiensis d-endo-
toxin CryIAa associated with membrane binding and
toxicity to Bombyx mori. J Bacteriol 176, 5554–5559.
20 Rajamohan F, Hussain SRA, 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.
21 Rajamohan F, Lee MK & Dean DH (1998) Bacillus
thuringiensis insecticidal proteins: molecular mode of
action. Prog Nucleic Acid Res Mol Biol 60, 1–27.
22 Rajamohan F, Cotrill JA, Gould F & Dean DH (1996)
Role of domain II, loop 2 residues of Bacillus thuringi-
ensis CryIAb d-endotoxin in reversible and irreversible
binding to Manduca sexta and Heliothis virescens. J Biol
Chem 271, 2390–2397.
23 Abdul-Rauf M & Ellar DJ (1999) Mutations of loop 2
and loop 3 residues in domain II of Bacillus thuringien-
sis Cry1C delta-endotoxin affect insecticidal specificity
and initial binding to Spodoptera littoralis and Aedes
aegypti midgut membranes. Curr Microbiol 39, 94–98.
24 Lee MK, Rajamohan F, Jenkins JL, Curtiss AS &
Dean DH (2000) Role of two arginine residues in
domain II, loop 2 of Cry1Ab and Cry1Ac Bacillus
thuringiensis delta-endotoxin in toxicity and binding to
Manduca sexta and Lymantria dispar aminopepti-
dase N. Mol Microbiol 38, 289–298.

25 Lee MK, Jenkins JL, You TH, Curtiss A, Son JJ,
Adang MJ & Dean DH (2001) Mutations at the argi-
Receptor binding ⁄ oligomerization of Cry1Aa mutants F. Obata et al.
5958 FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS
nine residues in alpha8 loop of Bacillus thuringiensis
delta-endotoxin Cry1Ac affect toxicity and binding to
Manduca sexta and Lymantria dispar aminopeptidase
N. FEBS Lett 497, 108–112.
26 Hawkins RE, Russell SJ, Baier M & Winter G (1993)
The contribution of contact and non-contact residues of
antibody in the affinity of binding to antigen. The inter-
action of mutant D1.3 antibodies with lysozyme. J Mol
Biol 234, 958–964.
27 Gomez I, Miranda-Rios J, Rudin
˜
o-Pin
˜
era E, Oltean
DI, Gill SS, Bravo A & Sobero
´
n M (2002) Hydro-
pathic complementarity determines interaction of
epitope (869)HITDTNNK(876) in Manduca sexta
Bt-R(1) receptor with loop 2 of domain II of Bacillus
thuringiensis Cry1A toxins. J Biol Chem 277, 30137–
30143.
28 Go
´
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-R1 receptor interact with loops alpha-8 and 2
in domain II of Cy1Ab toxin. Biochemistry 42, 10482–
10489.r
29 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 toxins.
J Biol Chem 280, 8416–8425.
30 Ishikawa H, Hoshino Y, Motoki Y, Kawahara T,
Kitajima M, Kitami M, Watanabe A, Bravo A, Sober-
on M, Honda A et al. (2007) A system for the directed
evolution of the insecticidal protein from Bacillus
thuringiensis. Mol Biotechnol 36, 90–101.
31 Van Rie J, Jansens S, Ho
¨
fte H, Degheele D & Van
Mellaert H (1990) Receptors on the brush border
membrane of the insect midgut as determinants of the
specificity of Bacillus thuringiensis delta-endotoxins.
Appl Environ Microbiol 56, 1378–1385.
32 Jime
´
nez-Jua
´
rez N, Mun
˜
oz-Garay C, Go

´
mez I, Saab-
Rincon G, Damian-Almazo JY, Gill SS, Sobero
´
nM&
Bravo A (2007) Bacillus thuringiensis Cry1Ab mutants
affecting oligomer formation are non-toxic to Manduca
sexta larvae. J Biol Chem
282, 21222–21229.
33 Rausell C, Pardo-Lo
´
pez L, Sa
´
nchez J, Mun
˜
oz-Garay C,
Morera C, Sobero
´
n M & Bravo A (2004) Unfolding
events in the water-soluble monomeric Cry1Ab toxin
during transition to oligomeric pre-pore and membrane-
inserted pore channel. J Biol Chem 279, 55168–55175.
34 Pardo-Lo
´
pez L, Go
´
mez I, Mun
˜
oz-Garay C, Jime
´

nez-
Juarez N, Sobero
´
n M & Bravo A (2006) Structural and
functional analysis of the pre-pore and membrane-
inserted pore of Cry1Ab toxin. J Invertebr Pathol 92,
172–177.
35 Mun
˜
oz-Garay C, Sa
´
nchez J, Darszon A, de Maagd
RA, Bakker P, Sobero
´
n M & Bravo A (2006) Perme-
ability changes of Manduca sexta midgut brush border
membranes induced by oligomeric structures of different
cry toxins. J Membr Biol 212, 61–68.
36 Ihara H & Himeno M (2008) Study of the irreversible
binding of Bacillus thuringiensis Cry1Aa to brush bor-
der membrane vesicles from Bombyx mori midgut.
J Invertebr Pathol 98, 177–183.
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, Inoue Y, Ishizaka T, Mizuno E, Yoshizawa
Y, Kitami M & Sato R (2008) Location of the Bombyx
mori 175 kDa cadherin-like protein-binding site on

Bacillus thuringiensis Cry1Aa toxin. FEBS J 275, 4913–
4926.
39 Aronson AI, Geng C & Wu L (1999) Aggregation of
Bacillus thuringiensis Cry1A toxins upon binding to tar-
get insect larval midgut vesicles. Appl Environ Microbiol
65, 2503–2507.
40 Pe
´
rez C, Mun
˜
oz-Garay C, Portugal LC, Sa
´
nchez J, Gill
SS, Sobero
´
n M & Bravo A (2007) Bacillus thuringiensis
ssp. israelensis Cyt1Aa enhances activity of Cry11Aa
toxin by facilitating the formation of a pre-pore oligo-
meric structure. Cell Microbiol 9
, 2931–2937.
41 Rajamohan F, Alzate O, Cotrill JA, Curtiss A & Dean
DH (1996) Protein engineering of Bacillus thuringiensis
d-endotoxin: mutations at domain II of Cry1Ab
enhance receptor affinity and toxicity towards gypsy
moth larvae. Proc Natl Acad Sci USA 93, 14338–14343.
42 Wu SJ, Koller CN, Miller DL, Bauer LS & Dean DH
(2000) Enhanced toxicity of Bacillus thuringiensis Cry3A
delta-endotoxin in coleopterans by mutagenesis in a
receptor binding loop. FEBS Lett 473, 227–232.
43 Abdullah MA, Alzate O, Mohammad M, McNall RJ,

Adang MJ & Dean DH (2003) Introduction of Culex
toxicity into Bacillus thuringiensis Cry4Ba by protein
engineering. Appl Environ Microbiol 69, 5343–5353.
44 Liu XS & Dean DH (2006) Redesigning Bacillus thurin-
giensis Cry1Aa toxin into a mosquito toxin. Protein
Eng Des Sel 19, 107–111.
45 Sobero
´
n M, Pardo-Lo
´
pez L, Lo
´
pez I, Go
´
mez I, Tabash-
nik BE & Bravo A (2008) Engineering modified Bt tox-
ins to counter insect resistance. Science 318, 1561–1562.
46 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.
47 Wolfersberger MG, Luthy P, Mauer A, Parenti P,
Sacchi V, Giordana B & Hanozet G (1987) Preparation
and partial characterization of amino acid transporting
brush border membrane vesicles from the larval midgut
of the cabbage butterfly (Pieris brassicae). Comp
Biochem Physiol 86, 301–308.
F. Obata et al. Receptor binding ⁄ oligomerization of Cry1Aa mutants
FEBS Journal 276 (2009) 5949–5959 ª 2009 The Authors Journal compilation ª 2009 FEBS 5959