Aminopeptidase N isoforms from the midgut of Bombyx mori and
Plutella xylostella ^ their classi¢cation and the factors that determine
their binding speci¢city to Bacillus thuringiensis Cry1A toxin
Kazuko Nakanishia, Katsuro Yaoib, Yasushi Naginoa, Hirotaka Haraa, Madoka Kitamia,
Shogo Atsumia, Nami Miuraa, Ryoichi Satoa;?
aGraduate School of Bio-Applications and Systems Engineering, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan
bResearch Institute of Biological Resources, National Institute of Advanced Industrial Science and Technology (AIST), AIST Tsukuba Central 6,
1-1-1 Higashi, Tsukuba, Ibaraki 305-8566, Japan
Received 27 December 2001; revised 9 April 2002; accepted 9 April 2002
First published online 23 April 2002
Edited by Takashi Gojobori
BmAPN3 and PxAPN3, from the midguts of Bombyx mori and
Plutella xylostella, respectively, were cloned, and a total of
eight APN isoforms cloned from B. mori and P. xylostella were
classified into four classes. Bacillus thuringiensis Cry1Aa and
Cry1Ab toxins were found to bind to specific APN isoforms
from the midguts of B. mori and P. xylostella, and binding
occurred with fragments that corresponded to the BmAPN1
Cry1Aa toxin-binding region of each APN isoform. The results
suggest that APN isoforms have a common toxin-binding
region, and that the apparent specificity of Cry1Aa toxin
binding to each intact APN isoform seen in SDS^PAGE is
determined by factors such as expression level in conjunction
with differences in binding affinity.
European Biochemical Societies. Published by Elsevier Science
B.V. All rights reserved.
Novel aminopeptidase N (APN) isoform cDNAs,
? 2002 Federation of
Key words: Aminopeptidase N; Cry1A toxin; Receptor;
Bombyx mori; Plutella xylostella; Bacillus thuringiensis
Aminopeptidase N (APN) (EC 22.214.171.124) is an enzyme that
preferentially cleaves neutral amino acids from the amino-ter-
minus of proteins or oligopeptides. This enzyme is widely
distributed within the plant and animal kingdoms. In animals,
APN is most abundant in the brush border membrane of the
intestine, and is an enzyme involved in the digestion of protein
in food. In addition, APN is a major Bacillus thuringiensis Cry
toxin receptor candidate in the midgut of insects.
The Gram-positive bacterium B. thuringiensis produces sev-
eral insecticidal proteins called Cry toxins. Cry toxins possess
insecticidal speci¢city, but do not a¡ect mammals. Conse-
quently, Cry toxins have been used as microbial insecticides.
Recently, Cry toxin genes have been used in the genetic devel-
opment of insect-resistant plants. However, it is not clear why
each Cry toxin kills only speci¢c insects.
Cry toxins produced as protoxins are solubilized and acti-
vated proteolytically in the midgut of a susceptible insect [1^
3]. The activated toxin binds to speci¢c receptors on epithelial
cells in the midgut [4,5], subsequently creating a pore in the
cell membrane that eventually leads to cell death . There-
fore, the toxin-speci¢c binding to the receptor is one of the
key factors determining insecticidal speci¢city. Several APNs
[7^13] as well as cadherin-like molecules [14,15] have been
identi¢ed as Cry toxin receptor candidates. Experiments in-
volving membranes reconstituted with APN suggest that APN
promotes the insertion of Cry toxins into the cell membrane,
and is also involved in pore formation initiated by three toxins
Eighteen APN isoform cDNAs from eight lepidopterans
have been cloned and registered in databases: from Bombyx
mori, Heliothis virescens, Plutella xylostella, Helicoverpa punc-
tigera, Manduca sexta, Limantria dispar, Plodia interpunctella,
and Epiphyas postvittana. Oltean et al.  found that APN
isoforms from insects were grouped into at least four classes
according to dendrogram analyses of APN sequences. How-
ever, no insect has been reported to have isoforms from all
four classes. There are uncloned APN isoforms, such as M.
sexta Cry1C toxin-binding 106-kDa APN  and L. dispar
APN2  and the fourth APN isoform in B. mori . Not
all APN isoforms in lepidopterans have been cloned, and it is
not known how many classes of APN are present in lepidop-
terans or whether lepidopterans possess APN isoforms from
all four classes.
Eight of 18 cloned APN isoforms, B. mori APN1 , M.
sexta APN1 , APN2 , H. virescens 120-kDa APN ,
170-kDa APN , 110-kDa APN , L. dispar APN1 
and E. postvittana APN  are reported to bind to Cry toxin.
There are reports that indicate that di¡erent Cry toxins bind
to distinct APN isoforms in the brush border membrane
vesicle (BBMV). In M. sexta, both Cry1Aa and Cry1Ac tox-
ins bind to APN1 , Cry1Ab toxin binds to APN2 , and
Cry1C toxin binds to 106-kDa APN . In H. virescens,
Cry1Aa and Cry1Ab toxins bind to 170-kDa APN only, while
Cry1Ac toxin binds to 120-kDa, 110-kDa, and 170-kDa APN
[11,12,18,24]. In L. dispar, Cry1Ac toxin binds to APN1, but
Cry1Aa, Cry1Ab, Cry1C, Cry2A, and Cry3A toxin do not
bind to either APN1 or APN2 [9,10]. Therefore, it is not
known whether each APN isoform binds to Cry toxin and
whether each toxin binds to a speci¢c class of APN isoform
from various lepidopterans.
0014-5793/02/$22.00 ? 2002 Federation of European Biochemical Societies. Published by Elsevier Science B.V. All rights reserved.
PII: S0 014-57 93(02)027 08-4
*Corresponding author. Fax: (81)-42-388 7277.
E-mail address: email@example.com (R. Sato).
FEBS 26077FEBS Letters 519 (2002) 215^220
The receptor domains involved in the binding of Cry toxin
have been studied. In the APN isoform derived from B. mori,
Cry1Aa toxin binds to a Cry1Aa toxin-binding region be-
tween Ile135 and Pro198 of BmAPN1 . Moreover, the
same region of PxAPN3 binds to the Cry1Aa toxin . By
contrast, in cadherin-like protein, only eight amino acid resi-
dues are identi¢ed as Cry1Aa and Cry1Ab toxin-binding epi-
This study sought to clarify the binding speci¢city of Cry
toxin to the classes of APN, and investigated the molecular
mechanisms involved in this process. We found a novel APN
cDNA sequence, derived from the midgut of both B. mori and
P. xylostella, which suggests that at least four classes of APN
are present in the midgut of lepidopterans. Moreover, we
showed that Cry1Aa and Cry1Ab toxins bind to speci¢c
APN isoforms when intact APN molecules are used, although
every APN isoform from all four classes had a common
Cry1Aa and Cry1Ab toxin-binding region. We predicted the
factors determining the Cry1A toxin-binding speci¢city to
each APN isoform. All APN isoforms seem to have a com-
mon partial structure that both Cry1Aa and Cry1Ab toxins
can bind to, and this structure determines the binding ability
of Cry1Aa and Cry1Ab toxins with APN. Other factors might
also have an e¡ect on the binding of Cry1A toxin.
2. Materials and methods
2.1. Preparation of Cry1Aa and Cry1Ab toxins
Cry1Aa and Cry1Ab crystals were puri¢ed from B. thuringiensis
serovar sotto strain T84A1 and recombinant Escherichia coli express-
ing Cry1Ab toxin  (provided by Sumitomo Chemical), respec-
tively. The puri¢ed crystals were solubilized and then activated as
described elsewhere .
2.2. Cloning of APN isoform cDNAs from the B. mori and
P. xylostella midgut
Total RNA, from the midgut of ¢fth instar B. mori (KinshuU
Showa) and fourth instar P. xylostella, was isolated and cDNAs
were synthesized according to previously described methods . De-
generate primers used for PCR ampli¢cation were designed from three
conserved peptide sequences that are found in APNs from a number
of organisms: GAMEWG (S1, 5P-GIGCNAYNGARAAYTGGG),
YRVNYD (R1, 5P-RTCRTAVTTIACDCKGTA) and WLNEGFA
primer (S2, 5P-TITWYMGIYTICCNACNACNAC) was designed
from the peptide sequence YRLPTTT, which is conserved in B. mori
APN1, H. virescens 120-kDa APN, and M. sexta APN1. PCR was
performed with template cDNA derived from the midgut of B. mori
and P. xylostella. PCR products were cloned into the T-overhang
vector p123T (MoBi Tec) and sequenced. Following sequence analy-
sis, a partial cDNA of a novel APN-like protein derived from B. mori,
named BmAPN3, was identi¢ed.
The APN cDNA fragment (PXfrg1) from P. xylostella has been
cloned previously . The BmAPN3 and PXfrg1 sequences were
used to design oligonucleotide primers speci¢c to the respective
APNs. Using these primers, the complete nucleotide sequences of
BmAPN3 and PXfrg1 were determined utilizing 5P- and 3P-RACE.
2.3. Preparation of speci¢c antisera to four APN isoforms derived
from B. mori
cDNA fragments encoding the non-conserved regions of four
B. mori APN isoforms (Fig. 1), amino acid residues Thr57^Thr173
of BmAPN1 and the homologous parts of another BmAPN, were
ampli¢ed by PCR and cloned into the GST fusion protein expression
vector, pGEX-4T-3 (Amersham Pharmacia Biotech) and then trans-
fected into E. coli BL21. Recombinant APN fragments were produced
using a previously described method . Recombinant APN frag-
ments, puri¢ed by SDS^PAGE and electroelution, were used to raise
class-speci¢c antisera in mice.
2.4. Immunoblot and ligand blot analyses of B. mori and
P. xylostella BBMV
BBMV from the midgut of B. mori and P. xylostella was prepared
according to the method described by Worfersberger et al. .
BBMV proteins were separated by SDS^PAGE and transferred to a
nitrocellulose membrane. The membrane was blocked with 2% bovine
serum albumin (BSA) in TBST (10 mM Tris^HCl pH 8.0, 150 mM
NaCl, 0.05% Tween 20). For the immunoblot, the membrane was
incubated with mouse anti-APN antiserum, followed by goat anti-
mouse IgG^HRP conjugate (Bio-Rad). The bound antibody was de-
tected using ECL Western blotting detection systems (Amersham
Pharmacia Biotech). For the ligand blot, the blocked membrane
was incubated in 10 nM Cry1Aa or Cry1Ab toxin in TBST containing
2% BSA, followed by mouse anti-Cry1Aa toxin antiserum that cross-
reacted with both Cry1Aa and Cry1Ab toxins. The membrane was
incubated in goat anti-mouse IgG^HRP conjugate and the bound
antibody was detected as described above.
2.5. Expression and toxin-binding ability of the Cry1Aa toxin-binding
region in APN classes from B. mori and P. xylostella
The toxin-binding regions from each APN class of B. mori and
P. xylostella were expressed for GST fusion proteins as described
above. The binding ability of APNs to Cry1Aa and Cry1Ab toxins
was assessed using ligand blot analysis.
3.1. cDNA cloning of APN isoforms from the midgut of
RT-PCR strategy was used to clone APN isoform cDNAs
from the midgut of B. mori. Primer pair S1/R1 ampli¢ed an
approximately 900-bp fragment. This fragment was cloned
into the p123T vector and sequenced. Three di¡erent sequen-
ces were identi¢ed. Sequence analysis showed that the frag-
ments were identical to the sequences of B. mori APN1 ,
APN2 , and APN4 (GenBank AB013400). Another primer
pair, S2/R2, produced a 1000-bp product that was identi¢ed
as a novel APN-like sequence, named BmAPN3, after se-
quencing and aligning. Full-length BmAPN3 was obtained
by 5P- and 3P-RACE and sequencing. The sequence data of
BmAPN3 has been deposited in GenBank (accession number
BmAPN3 is 3240 bp long and contains a 3015-bp open
reading frame encoding a putative 1005-amino acid protein
with a theoretical molecular mass of 113.6 kDa. The predicted
start codon is embedded in a consensus Kozak translation
initiation sequence (AAGATGG) . There is a putative
polyadenylation signal sequence (AATAAA). Analysis of the
N-terminal region with the program SignalP (http://www.
cbs.dtu.dk/) predicted a signal peptide sequence (MANY-
KVIIFLAACVLAQA, residues 1^18) with a cleavage site be-
tween Ala18 and Phe19. The glycosylphosphatidylinositol
(GPI) anchor signal sequence , consisting of three small
amino acids (DAA, residues 982^984) and a stretch of hydro-
phobic residues (PVSTFLSVAVVALVAVVNLIM, residues
985^1005), is found at the C-terminus. Six potential N-glyco-
sylation sites (NXS/T)  and 11 O-glycosylation sites, pre-
cbs.dtu.dk/), were found to be present. BmAPN3 is character-
ized by a consensus zinc-binding/gluzincin motif (HEXX-
HX18E, residues 371^394)  and a gluzincin aminopeptidase
motif (GAMEN, residues 335^339) . BLAST searches of
the GenBank database using the protein sequence revealed
that BmAPN3 was most similar to other lepidopteran
APNs, including H. virescens 120-kDa APN  (64.6% iden-
tical), L. dispar APN1  (65.9% identical), and P. interpunc-
K. Nakanishi et al./FEBS Letters 519 (2002) 215^220
tella APN1  (63.4% identical). The deduced amino acid
sequence of BmAPN3 was aligned with other B. mori APN
isoforms using ClustalX (Fig. 1). BmAPN3 contained almost
all the conserved residues found in three other B. mori APN
3.2. cDNA cloning of APN isoforms from the midgut of
Various APN isoform cDNAs from the midgut of P. xy-
lostella were cloned as above. Three di¡erent cDNA sequen-
ces, obtained using primers S1 and R1, were identical to the
sequences of P. xylostella APN1 , APNA , and APN4
(GenBank AJ222699). The full length of previously cloned
PXfrg1  was sequenced using 5P- and 3P-RACE. The 5P-
end of PXfrg1 was not determined, although a 5P-truncated
cDNA was obtained and renamed PxAPN3. The sequence
data of PxAPN3 has been deposited in GenBank (accession
PxAPN3 contained a 2826-bp open reading frame encoding
a putative 942-amino acid protein. There was a sequence (AA-
TAA) similar to a polyadenylation signal sequence (AA-
TAAA). The PxAPN3 sequence also appeared to have a pos-
Fig. 1. Comparison of the deduced amino acid sequence of BmAPN3 with those of BmAPN1 , BmAPN2 , and BmAPN4 (GenBank
AB013400). Multiple-sequence alignment was performed using the program ClustalX. Highly conserved residues have dark backgrounds. The
large boxed region was used as the antigen to raise speci¢c antiserum. The area surrounded by the bold line is the region corresponding to the
Cry1Aa toxin-binding region of BmAPN1.
K. Nakanishi et al./FEBS Letters 519 (2002) 215^220
sible GPI anchor signal sequence. The sequence contained
four putative N-glycosylation sites (NXS/T) and 22 O-glyco-
sylation sites. PxAPN3 is characterized by a consensus zinc-
binding/gluzincin motif (HEXXHX18E) and a gluzincin ami-
nopeptidase motif (GAMEN). BLAST searches of the Gen-
Bank database using the protein sequence revealed that
PxAPN3 was highly similar to lepidopteran APNs.
3.3. Classi¢cation of lepidopteran APNs
The deduced amino acid sequences encoded by BmAPN3
and PxAPN3 were aligned with 18 other lepidopteran APNs
using ClustalX. The lepidopteran APNs were grouped into
four classes on the phylogenetic tree derived from a ClustalX
alignment (Fig. 2). The group composed of BmAPN1 ,
MsAPN1 , Hv170kDaAPN , HpAPN1 , and
PxAPNA  was tentatively named class 1. The group com-
posed of BmAPN2 , MsAPN2 , PxAPN1 , and Ld
V APN2  was categorized as class 2. The group composed
of Hv120kDaAPN , LdAPN1 , PiAPN1 , HpAPN3
, EpAPN , BmAPN3, and PxAPN3 was named class 3.
The group composed of Hv110kDaAPN , BmAPN4
(AB013400), HpAPN2 , and PxAPN4 (AJ222699) was
named class 4. In both B. mori and P. xylostella, class 1, 2,
and 4 APNs were evident. In addition, class 3 APNs were
evident in B. mori and P. xylostella.
3.4. Immunoblot and ligand blot analyses of B. mori and
P. xylostella BBMV
To identify the four APN isoforms in B. mori BBMV, an
immunoblot analysis was performed using class-speci¢c anti-
sera (Fig. 3A). The anti-BmAPN1, -BmAPN2, -BmAPN3,
and -BmAPN4 antisera recognized 115-, 90-, 110-, and 100-
kDa proteins in BBMV, respectively (Fig. 3A, lanes 3^6).
After immunoblot with anti-BmAPN3 antiserum, immuno-
blots with anti-BmAPN1 or -BmAPN4 antiserum were per-
formed using the same membrane. The result of this double
immunoblot analysis con¢rmed that each antiserum recog-
nized the di¡erent protein (data not shown). On the other
hand, anti-APN antiserum, which was raised using the puri-
¢ed intact BmAPN1 as an antigen, recognized the 120- and
230-kDa proteins in BBMV, as well as the proteins recognized
by anti-BmAPN1, -BmAPN2 and -BmAPN3 antisera (Fig.
3A, lane 2). The 115-, 110-, and 100-kDa proteins were also
detected in Coomassie-stained SDS^PAGE gels, while the 90-
kDa protein was not (Fig. 3A, lane 1). A ligand blot was
performed to identify the Cry1Aa and Cry1Ab toxin-binding
proteins in B. mori BBMV. Both toxins bound to the 115-kDa
protein at 10 nM (Fig. 3A, lanes 7, 8). After immunoblot
analysis with anti-BmAPN3 antiserum, ligand blot was per-
formed, using the same membrane. The band of 110 kDa
recognized with anti-BmAPN3 antiserum was certainly di¡er-
Fig. 2. Phylogenetic tree derived from a ClustalX alignment of lepi-
dopteran midgut APNs. GenBank accession numbers are as follows:
B. mori APN1 (BmAPN1), AF084257 ;
(MsAPN1), X89081 ; H. virescens 170-kDa APN (Hv170kDa),
AF173552 ; H. postvittana APN1 (HpAPN1), AF217248 ;
P. xylostella APNA (PxAPNA), AF020389 ; B. mori APN2
(BmAPN2), AB011497 ; M. sexta APN2 (MsAPN2), X97877
; P. xylostella APN1 (PxAPN1), X97878 ; L. dispar V APN2
AF352574 (this study); H. virescens 120-kDa APN (Hv120kDa),
U35096 ; P. xylostella APN3 (PxAPN3), AF109692 (this study);
L. dispar APN1 (LdAPN1), AF126442 ; P. interpunctella APN1
AF217250 ; E. postvittana APN (EpAPN), AF276241 ;
B. mori APN (BmAPN4), AB013400; H. virescens 110-kDa APN
AJ222699; H. postvittana APN2 (HpAPN2), AF217249 .
M. sexta APN1
Fig. 3. SDS^PAGE, immunoblot, and ligand blot analyses of
B. mori (A) and P. xylostella (B) BBMV proteins. B. mori and
P. xylostella BBMV proteins were separated by 5% SDS^PAGE
and then transferred to nitrocellulose membranes: Coomassie bril-
liant blue-stained gel (lanes 1). Immunoblot blot analyses were per-
formed with anti-APN antiserum (lanes 2), anti-BmAPN1 antiserum
(lanes 3), anti-BmAPN2 antiserum (lanes 4), anti-BmAPN3 antise-
rum (lanes 5), and anti-BmAPN4 antiserum (lanes 6). The second-
ary antibody was goat anti-mouse IgG^HRP conjugate and ECL
was the method of detection. Ligand blot analyses were performed
with 10 nM Cry1Aa (lanes 7) and Cry1Ab (lanes 8) toxin. Binding
toxins were detected as described in Section 2. The numbers on the
left are molecular masses. The molecular sizes on the right are dis-
cussed in the text.
K. Nakanishi et al./FEBS Letters 519 (2002) 215^220
ent from the band of 115 kDa detected by the ligand blot
(data not shown).
Using the class-speci¢c antisera for B. mori APN isoforms,
we conducted an immunoblot analysis of P. xylostella BBMV
(Fig. 3B). The anti-BmAPN1 and anti-BmAPN3 antisera rec-
ognized the 120- and 110-kDa proteins in BBMV (Fig. 3B,
lanes 3, 5). The anti-BmAPN2 and anti-BmAPN4 antisera
were not bound to any proteins in BBMV (Fig. 3B, lanes 4,
6). Ligand blot analysis showed that both Cry1Aa and
Cry1Ab toxins bound to the 110-kDa protein at 10 nM
(Fig. 3B, lanes 7, 8).
3.5. Ability of Cry1A toxin to bind to the Cry1Aa
BmAPN1 Cry1Aa toxin-binding region  from each APN
class of B. mori and P. xylostella were expressed as GST
fusion proteins in E. coli cells (Fig. 4A). The ability of
Cry1A toxin to bind to the APN fragments was subsequently
analyzed. Both Cry1Aa and Cry1Ab toxins bound to all four
APN classes derived from B. mori and P. xylostella at 10 nM
(Fig. 4B,C). Bands smaller than GST fusion proteins were
visible in the ligand blot ^ these were likely degradation prod-
ucts of GST fusion proteins.
fragmentscorresponding to the
BmAPN3 and PxAPN3 possess the characteristics of lepi-
dopteran APNs, so they appear to belong to the APN family.
Fig. 2 shows that BmAPN3 and PxAPN3 belong to class 3,
and it is clear that all four APN classes are present in B. mori
and P. xylostella. The third APN, a 106-kDa Cry1C-binding
APN, was reported in M. sexta . A 105-kDa APN was
reported in the midgut of L. dispar . Hence, unidenti¢ed
APNs may exist in the midguts of lepidopterans. In addition
to B. mori and P. xylostella, all lepidopterans may possess all
four APN classes. Moreover, in B. mori, the 120- and 230-
kDa proteins that were recognized by anti-APN antiserum
(Fig. 3A, lane 2) might be unidenti¢ed APN isoforms. On
the other hand, in P. xylostella, two novel cDNA fragments,
which were di¡erent from the four APN isoforms cloned and
had high similarity to lepidopteran class 3 and class 4 APNs,
were cloned by RT-PCR using primers S1 and R1 (data not
shown). Therefore, two more unidenti¢ed APN isoforms
should be present in the P. xylostella midgut. Although
PxAPN4 was tentatively grouped in the same class as
BmAPN4, HpAPN2, and Hv110kDaAPN (Fig. 2), the simi-
larity between these isoforms is not high. Thus, the class 4
APNs, which we identi¢ed as one group, might be subdivided
into two or more classes. In conclusion, lepidopteran APN
isoforms are grouped into at least four classes.
Using class-speci¢c antisera,
BmAPN3, and BmAPN4 were identi¢ed as 115-, 90-, 110-,
and 100-kDa proteins, respectively (Fig. 3A, lanes 3^6).
BmAPN1, which was reported to be a 120-kDa protein
[13,21], was a 115-kDa protein in this study, which used 5%
SDS^PAGE instead of 10% SDS^PAGE to separate APN
isoforms. The molecular size of BmAPN2, identi¢ed by im-
munoblot, was consistent with a previous report . Several
partial amino acid sequences of a 100-kDa B. mori APN 
were identical to the deduced amino acid sequence of
BmAPN4. On the other hand, in P. xylostella, PxAPNA
and PxAPN3 were considered to be 120- and 110-kDa pro-
teins, respectively (Fig. 3B, lanes 3^6).
In B. mori, both Cry1Aa and Cry1Ab toxins were proposed
to bind to BmAPN1 (Fig. 3A, lanes 3, 7, 8). In P. xylostella,
both toxins bound to a 110-kDa protein the same size as the
protein recognized by anti-BmAPN3 antiserum (Fig. 3B, lanes
5, 7, 8). Since PxAPN1 and PxAPN4 were not identi¢ed,
Cry1Aa and Cry1Ab toxins were not concluded to bind to
PxAPN3. However, it is certain that neither toxin binds to
PxAPNA, as PxAPNA is a 120-kDa protein. Cry1Aa and
Cry1Ab toxins bind to MsAPN1 and Hv170kDaAPN, which
belong to class 1 [18,40]. Our results showed that both toxins
Fig. 4. Ligand blot analyses of Cry1Aa toxin-binding regions of B.
mori and P. xylostella APN isoforms. Recombinant APN fragments
corresponding to the Cry1Aa toxin-binding region of BmAPN1
were expressed as GST fusion proteins in E. coli cells. The recombi-
nant APN fragments were separated by 12.5% SDS^PAGE and the
Cry1Aa and Cry1Ab toxin-binding ability was analyzed using ligand
blotting. Coomassie brilliant blue-stained gel (A), ligand blot with
10 nM Cry1Aa toxin (B) and Cry1Ab toxin (C). Lanes 1, GST^
BmAPN1 fusion protein; lanes 2, GST^BmAPN2 fusion protein;
lanes 3, GST^BmAPN3 fusion protein; lanes 4, GST^BmAPN4 fu-
sion protein; lanes 5, GST^PxAPNA fusion protein; lanes 6, GST^
PxAPN1 fusion protein; lanes 7, GST^PxAPN3 fusion protein;
lanes 8, GST^PxAPN4 fusion protein; lanes 9, GST alone. The as-
terisks indicate recombinant proteins. The bands larger and smaller
than the recombinant proteins are unidenti¢ed reactive proteins de-
rived from the inclusion bodies only from the APN-transformed
K. Nakanishi et al./FEBS Letters 519 (2002) 215^220
bind to BmAPN1, but not to PxAPNA, although BmAPN1
and PxAPNA also belong to class 1. Binding of Cry1Ab toxin
to class 2 APNs has been reported only in the case of
MsAPN2 , but binding of Cry1Ab toxin to class 2
BmAPN2 was not observed in this study (Fig. 3A, lanes 4,
7, 8). Therefore, it is suggested that each toxin does not nec-
essarily bind to a speci¢c class of APN isoform.
Cry1Aa toxin binds only to the Cry1Aa-binding region on
BmAPN1 (Ile135^Pro198)  (Fig. 1). Although Cry1Aa and
Cry1Ab toxins bound only to intact BmAPN1 in B. mori,
both toxins bound in a similar way to all the fragments of
the toxin-binding regions of the four classes of APN isoforms,
from both B. mori and P. xylostella (Fig. 4B,C). Since the
Cry1Aa toxin-binding regions have many conserved amino
acid residues such as RXXFPXXDEP in eight APN isoforms
(Fig. 5), Cry1Aa and Cry1Ab toxins may recognize and bind
to a common structure in these regions. Most of the conserved
amino acids within the B. mori and P. xylostella APNs are
also conserved in other lepidopteran APNs, suggesting that
both toxins might bind to this region of APN in insects other
than the Bombycidae and Plutellidae.
BmAPN2, BmAPN3, and BmAPN4 had Cry1Aa toxin-
binding regions whose Cry1Aa toxin-binding abilities were
similar to that of BmAPN1. However, Cry1Aa toxin did
not bind to intact BmAPN2, BmAPN3, or BmAPN4 in
BBMV. Apparent di¡erences in the Cry1Aa toxin-binding
ability between APN isoforms in BBMV could be due to ex-
pression levels in conjunction with di¡erences in binding af-
Acknowledgements: We thank Sumitomo Chemical Co., Ltd. for
kindly providing the recombinant E. coli expressing Cry1Ab toxin.
This work was supported by a Grant-in-Aid of Science Research
(12558068) from the Ministry of Education, Culture, Sports, Science
and Technology of Japan.
 Ho «fte, H. and Whiteley, H.R. (1989) Microbiol. Rev. 53, 242^
 Gill, S.S., Cowles, E.A. and Pietrantonio, P.V. (1992) Annu. Rev.
Entomol. 37, 615^636.
 Knowles, B.H. (1994) Adv. Insect Physiol. 24, 275^308.
 Hofmann, C., Vanderbruggen, H., Ho «fte, H., Van Rie, J., Jan-
sens, S. and Van Mellaert, H. (1988) Proc. Natl. Acad. Sci. USA
 Van Rie, J., Jansens, S., Ho «fte, H., Degheele, D. and Van Mel-
laert, H. (1989) Eur. J. Biochem. 186, 239^247.
 Knowles, B.H. and Dow, J.A.T. (1993) BioEssays 15, 469^476.
 Knight, P.J.K., Crickmore, N. and Ellar, D.J. (1994) Mol. Mi-
crobiol. 11, 429^436.
 Luo, K., Lu, Y.-J. and Adang, M.J. (1996) Insect Biochem. Mol.
Biol. 26, 783^791.
 Valaitis, A.P., Lee, M.K., Rajamohan, F. and Dean, D.H. (1995)
Insect Biochem. Mol. Biol. 25, 1143^1151.
 Lee, M.K., You, T.H., Young, B.A., Cotrill, J.A., Valaitis, A.P.
and Dean, D.H. (1996) Appl. Environ. Microbiol. 62, 2845^2849.
 Gill, S.S., Cowles, E.A. and Francis, V. (1995) J. Biol. Chem.
 Luo, K., Sangadala, S., Masson, L., Mazza, A., Brousseau, R.
and Adang, M.J. (1997) Insect Biochem. Mol. Biol. 27, 735^743.
 Yaoi, K., Kadotani, T., Kuwana, H., Shinkawa, A., Takahashi,
T., Iwahana, H. and Sato, R. (1997) Eur. J. Biochem. 246, 652^
 Vadlamudi, R.K., Weber, E., Ji, I., Ji, T.H. and Bulla, L.A.
(1995) J. Biol. Chem. 270, 5490^5494.
 Nagamatsu, Y., Toda, S., Koike, T., Miyoshi, Y., Shigematsu, S.
and Kogure, M. (1998) Biosci. Biotechnol. Biochem. 62, 727^734.
 Sangadala, S., Walters, F.S., English, L.H. and Adang, M.J.
(1994) J. Biol. Chem. 269, 10088^10092.
 Cooper, M.A., Carroll, J., Travis, E.R., Williams, D.H. and El-
lar, D.J. (1998) Biochem. J. 333, 677^683.
 Oltean, D.I., Pullikuth, A.K., Lee, H.-K. and Gill, S. (1999)
Appl. Environ. Microbiol. 65, 4760^4766.
 Valaitis, A.P., Mazza, A., Brousseau, R. and Masson, L. (1997)
Insect Biochem. Mol. Biol. 27, 529^539.
 Hua, G., Tsukamoto, K. and Ikezawa, H. (1998) Comp. Bio-
chem. Physiol. B 121, 213^222.
 Yaoi, K., Nakanishi, K., Kadotani, T., Imamura, M., Koizumi,
N., Iwahana, H. and Sato, R. (1999) Biochim. Biophys. Acta
 Knight, P.J.K., Knowles, B.H. and Ellar, D.J. (1995) J. Biol.
Chem. 270, 17765^17770.
 Denolf, P., Hendrickx, K., Van Damme, J., Jansens, S. and Pe-
feroen, M. (1997) Eur. J. Biochem. 248, 748^761.
 Banks, D.J., Jurat-Fuentes, J.L., Dean, D.H. and Adang, M.J.
(2001) Insect Biochem. Mol. Biol. 31, 909^918.
 Garner, K.J., Hiremath, S., Lehtoma, K. and Valaitis, A.P.
(1999) Insect Biochem. Mol. Biol. 29, 527^535.
 Simpson, R.M. and Newcomb, R.D. (2000) Insect Biochem. Mol.
Biol. 30, 1069^1078.
 Yaoi, K., Nakanishi, K., Kadotani, T., Imamura, M., Koizumi,
N., Iwahana, H. and Sato, R. (1999) FEBS Lett. 463, 221^224.
 Nakanishi, K., Yaoi, K., Shimada, N., Kadotani, T. and Sato, R.
(1999) Biochim. Biophys. Acta 1432, 57^63.
 Go ¤mez, I., Oltean, D.I., Gill, S., Bravo, A. and Sobero ¤n, M.
(2001) J. Biol. Chem. 276, 28906^28912.
 Oeda, K., Oshie, K., Shimizu, M., Nakamura, K., Yamamoto,
H., Nakayama, I. and Ohkawa, H. (1987) Gene 53, 113^119.
 Worfersberger, M., Luethy, P., Maurer, A., Parenti, P., Sacchi,
F.V., Giordana, B. and Hanozet, G.M. (1987) Comp. Biochem.
Physiol. 86A, 301^308.
 Kozak, M. (1987) Nucleic Acids Res. 15, 8125^8148.
 Englund, P.T. (1993) Annu. Rev. Biochem. 62, 121^138.
 Gavel, Y. and Von Heijne, G. (1990) Protein Eng. 3, 433^442.
 Hooper, N.M. (1994) FEBS Lett. 354, 1^6.
 Laustsen, P.G., Rasmussen, T.E., Petersen, K., Pedraza-Diaz, S.,
Moestrup, S.K., Gliemann, J., Sottrup-Jensen, L. and Kristensen,
T. (1997) Biochim. Biophys. Acta 1352, 1^7.
 Zhu, Y.-C., Kramer, K.J., Oppert, B. and Dowdy, A.K. (2000)
Insect Biochem. Mol. Biol. 30, 215^224.
 Chang, W.X.Z., Gahan, L.J., Tabashnik, B.E. and Heckel, D.G.
(1999) Insect Mol. Biol. 8, 171^177.
 Emmerling, M., Chandler, D. and Sandeman, M. (2001) Insect
Biochem. Mol. Biol. 31, 899^907.
 Masson, L., Lu, Y.-J., Mazza, A., Brousseau, R. and Adang,
M.J. (1995) J. Biol. Chem. 270, 20309^20315.
Fig. 5. Comparison of the Cry1Aa toxin-binding regions of B. mori and P. xylostella APNs. The sequences of BmAPN1, BmAPN2, BmAPN3,
BmAPN4, PxAPNA, PxAPN1, PxAPN3 and PxAPN4 were compared. Perfectly conserved amino acid residues have black backgrounds. Highly
conserved amino acid residues have gray backgrounds.
K. Nakanishi et al./FEBS Letters 519 (2002) 215^220