Báo cáo khoa học: Purification and characterization of cathepsin B-like cysteine protease from cotyledons of daikon radish, Raphanus sativus docx

Purification and characterization of cathepsin B-like
cysteine protease from cotyledons of daikon radish,
Raphanus sativus
Akihiko Tsuji, Yayoi Kikuchi, Kentaro Ogawa, Hiroko Saika, Keizo Yuasa and Masami Nagahama
Department of Biological Science and Technology, University of Tokushima Graduate School, Japan
Genome analysis has indicated that plants, like
animals, possess a variety of protease genes. The
Arabidopsis thaliana genome has 680 protease (known
and putative peptidase) sequences representing all
six catalytic types: aspartic, cysteine, glutamic, metallo,
serine and threonine peptidases (MEROPS peptidase
database; reflecting the
diverse functions of plant proteases. In A. thaliana,
cDNAs encoding 10 members [vignain, brassicain,
RD19A, pseudotzain, aleurain, cathepsin B-like cyste-
ine protease (CBCP), RD21A, XCP1, XCP2 and
SAG12] of the papain-like cysteine protease (C1A)
family were identified [1]. Indeed, recent genetic studies
have indicated the involvement of papain-like cysteine
Keywords
cathepsin B; cotyledon; cysteine protease;
germination; occluding loop
Correspondence
A. Tsuji, Department of Biological Science
and Technology, University of Tokushima
Graduate School, 2-1 Minamijosanjima,
Tokushima 770-8506, Japan
Fax: +81 88 655 3161
Tel: +81 88 656 7526
E-mail:
(Received 29 July 2008, revised 1

September 2008, accepted 5 September
2008)
doi:10.1111/j.1742-4658.2008.06674.x
Plant cathepsin B-like cysteine protease (CBCP) plays a role in disease
resistance and in protein remobilization during germination. The ability of
animal cathepsin B to function as a dipeptidyl carboxypeptidase has been
attributed to the presence of a dihistidine (His110-His111) motif in the
occluding loop, which represents a unique structure of cathepsin B. How-
ever, a dihistidine motif is not present in the predicted sequence of the
occluding loop of plant CBCP, as determined from cDNA sequence analy-
sis, and the loop is shorter. In an effort to investigate the enzymatic prop-
erties of plant CBCP, which possesses the unusual occluding loop, we have
purified CBCP from the cotyledons of daikon radish (Raphanus sativus)by
chromatography through Sephacryl S-200, DEAE–cellulose, hydroxyapatite
and organomercurial–Sepharose. The molecular mass of the enzyme was
estimated to be 28 kDa by SDS ⁄ PAGE under reducing conditions. The
best synthetic substrate for CBCP was t-butyloxycarbonyl Leu-Arg-Arg-4-
methylcoumaryl 7-amide, as is the case with human cathepsin B. However,
the endopeptidase activity of CBCP towards glucagon and adrenocortico-
tropic hormone showed broad cleavage specificity. Human cathepsin B
preferentially cleaves model peptides via its dipeptidyl carboxypeptidase
activity, whereas daikon CBCP displays both endopeptidase and exopepti-
dase activities. In addition, CBCP was found to display carboxymonopepti-
dase activity against the substrate o-aminobenzoyl-Phe-Arg-Phe(4-NO
2
).
Daikon CBCP is less sensitive (1 ⁄ 7000) to CA-074 than human
cathepsin B. Expression analysis of CBCP at the protein and RNA levels
indicated that daikon CBCP activity in cotyledons is regulated by
post-transcriptional events during germination.

Abbreviations
Abz, o-aminobenzoyl; ACTH, adrenocorticotropic hormone; Boc, t-butyloxycarbonyl; CA074, N-(
L-3-trans-propylcarbamoyl-oxirane-2-carbonyl)-
L-isoleucyl-L-proline; CBCP, cathepsin B-like cysteine protease; E-64, trans-epoxysuccinyl-L-leucylamide-(4-guanidino)butane; MCA,
4-methylcoumaryl 7-amide; PVDF, poly(vinylidene difluoride); pyr,
L-pyroglutamyl; TLCK, Na-p-tosyl-L-lysine chloromethyl ketone; Z,
carbobenzoxy.
FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS 5429
proteases in various physiological processes, such as
programmed cell death and disease resistance [2].
However, with the exception of vignain, aleurain and
XCP, the enzymatic properties of these enzymes have
not been determined. Vignain was purified from germi-
nating cotyledons of soybean [3], aleurain was purified
from barley aleurone layers [4] and caster bean
endosperm [5], and XCP from white clover was
expressed in Escherichia coli and characterized [6].
CBCP was first identified as a gibberellin-regulated
gene during wheat germination and showed highest
homology with the mammalian cysteine protease,
cathepsin B (EC 3.4.22.1) [7]. During germination,
CBCP is expressed in aleurone cells surrounding
the endosperm and plays a role in the mobilization of
seed storage proteins and the support of seedling
growth. CBCP has been cloned from various plants,
including A. thaliana [1], barley [8] and Nicotiana rus-
tica [9], and CBCP transcript expression profiles have
been investigated. Recently, Gilroy et al. reported
the involvement of CBCP in the plant disease resis-
tance hypersensitive response [10]. The amino acid

sequences predicted from cDNA sequences of the
CBCPs display a high degree of identity in plants.
Unlike most other cysteine proteases of the papain
superfamily, cathepsin B displays both endopeptidase
and exopeptidase (dipeptidyl carboxypeptidase) activi-
ties [11–13]. A unique sequence of cathepsin B referred
to as the occluding loop between Ile105 and Pro126
(IPPCEHHVNGSRPPCTGEGDTP, human mature
cathepsin B), which is not present in papain, was
shown to play an important role in the exopeptidase
activity of cathepsin B [12–14]. This loop is highly con-
served in cathepsin B from vertebrates and inverte-
brates. Two histidine residues (His110 and His111)
within this occluding loop provide positively charged
anchors for the C-terminal carboxylate group of the
substrate [15,16]. In contrast, plant CBCP possesses an
occluding loop that differs from that of animal cathep-
sin B, although amino acid sequence alignment showed
that the cysteine residues involved in disulfide bridge
formation and substrate binding, in addition to active
site residues, are well conserved [7–9]. Plant CBCP has
a shorter loop than animal cathepsin B and contains
only one histidine residue. Deletion of either His110 or
His111 in human cathepsin B markedly affects both
cleavage specificity and sensitivity to endogenous
inhibitors such as cystatin c and the propeptide of
cathepsin B [14]. Differences between the cleavage
specificity of plant CBCP and animal cathepsin B have
yet to be determined. The enzymatic characterization
of plant CBCP is a necessary first step towards identi-

fying the physiological function of this protease.
In this study, we purified CBCP from daikon radish
(Raphanus sativus), in an effort to compare its enzy-
matic properties with those of mammalian cathepsin B.
Daikon radish and A. thaliana are in the same Brassic-
aceae family. Therefore, it was expected that the amino
acid sequence and developmental regulation of CBCP
would be similar, although expression analyses of
proteases in daikon radish have hitherto not been
performed. Furthermore, it is easier to prepare suffi-
cient amounts of cotyledons from daikon radish for
the purification of CBCP than from A. thaliana.We
investigated CBCP cleavage specificity and sensitivity
to protease inhibitors. This article represents the first
report detailing the enzymatic characterization of a
plant CBCP.
Results
Purification of cysteine protease from the
cotyledons of daikon radish
The trans-epoxysuccinyl-l-leucylamide-(4-guanidino)
butane (E-64)-sensitive enzyme expressed in cotyledons
of daikon radish was purified 3200-fold by employing
a series of column chromatographic procedures utiliz-
ing Sephacryl S-200, DEAE–cellulose, hydroxyapatite
and organomercurial–Sepharose, as described in
Experimental procedures. The specific activity against
Boc-Leu-Arg-Arg-4-methylcoumaryl 7-amide (MCA)
of the final preparation was 22.9 lmolÆmin
)1
Æmg

)1
.
About 100 lg of purified enzyme was obtained from
1 kg of cotyledons. The purified enzyme did not bind
concanavalin A–Sepharose gel, suggesting that the
enzyme does not contain high-mannose oligosaccha-
rides. Purified enzyme eluted as a single symmetrical
peak on HPLC with a Superose 12PC3.2 ⁄ 30 gel
filtration column, and its molecular mass was esti-
mated to be 21 kDa (data not shown). As shown in
Figs 1A,B, the final preparation yielded a single pro-
tein band on native PAGE in the absence of SDS,
whereas four bands (28, 15, 13 and 11 kDa) were
detected on SDS ⁄ PAGE. N-terminal sequences of
these bands were examined and compared with
sequences of other cysteine proteases. As shown in
Fig. 1C, the N-terminal sequence of the 28 kDa
protein (LPKSFDARTHWPQXT) showed the highest
degree of identity with CBCP. The 15 and 11 kDa
proteins had the same N-terminal sequence, suggesting
that both the 15 and 11 kDa proteins represent pro-
cessed or degraded forms of the 28 kDa enzyme. In
contrast, the sequence of the 13 kDa protein (QLW-
SESKHYS) differed from these sequences, although a
similar sequence is present in the middle region of the
Characterization of daikon CBCP A. Tsuji et al.
5430 FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS
catalytic domain of CBCP from other plants, such as
Arabidopsis (KLWSESKHYS, DDBJ accession num-
ber AF370193) [1] and wheat (QVWEEKKHFS,

DDBJ accession number X66012) [7]. Both sequences
were located at position 123–132 starting from the
N-terminus of the mature enzyme. As mentioned
below, the sequence of the 13 kDa protein was present
in the deduced amino acid sequence derived from dai-
kon radish CBCP cDNA. Recently, Van der Hoon
et al. labeled cysteine proteases in Arabidopsis leaves
using biotinylated E-64, and a 30-kDa labeled protein
was identified by MS as a CBCP [17]. It is highly likely
that the 28 kDa protein represents a single-chain form
of CBCP. The 15, 13 and 11 kDa fragments may be
degradation products of CBCP produced in the course
of the purification.
Isolation of cDNA encoding the catalytic domain
of daikon radish CBCP
In order to classify the purified cysteine protease,
cDNAs encoding the propeptide and catalytic domains
of the enzyme were isolated by PCR as described in
Experimental procedures. Figure 2 shows the compos-
ite cDNA and predicted amino acid sequences of two
PCR products. The N-terminal sequences of band a
(LPKSFDARTHWPQCT) and band c (QLWSESK-
HYS) of the purified enzyme match perfectly with the
deduced sequence, residue numbers 104–118 and 226–
232, respectively. A signal peptide cleavage site was
predicted to be located between Ala31 and Glu32,
using the signalP3.0 server ( />services/SignalP/). Two possible glycosylation sites
(Asn152 and Asn311) were also identified. The
deduced sequences of the propeptide and catalytic
domains showed high amino acid identity with CBCP

from A. thaliana (90%, DDBJ accession number
AF370193) [1], potato (72%, DDBJ accession number
AY450641) [8] and tobacco (72%, DDBJ accession
number DQ492287) [9]. The sequence of the daikon
enzyme showed highest amino acid identity
with human cathepsin B (47%, DDBJ accession
number M14221) [17] among the animal cysteine pro-
teases. Alignment of the cysteine protease sequence
identified the putative active site residues (Cys132,
His287 and Asn308), as shown in Fig. 3. The cysteine
residues involved in disulfide bridge formation
are completely conserved (Cys14–Cys43, Cys26–
Cys71, Cys62–Cys128, Cys63–Cys67, Cys100–Cys132
and Cys108–Cys119 in human mature cathepsin B).
Residues in the S1, S2 and S1¢ sites (S1, Gln23 and
Gly74; S2, Glu245 and Gly198; S1¢, Val176) are also
conserved. Ala77 and Ala173 in the S2 site of human
mature cathepsin B are replaced by isoleucine and ser-
ine, respectively. Thus, amino acid sequence alignment
indicates that the purified enzyme from daikon radish
is a CBCP. One striking difference between plant
CBCP and animal cathepsin B occurs in the occluding
loop region, which plays a critical role in the exopepti-
dase activity of cathepsin B (Fig. 4). All animal
cathepsin B enzymes, including those from human
[18], cattle [19], mouse [18], chicken [20], Atlantic hali-
but (DDBJ acession number DQ993253), yellow meal-
worm [21] and Clonorchis sinensis (DDBJ accession
number EF102086), possess similar occluding loop
structures. The occluding loop structure is also con-

served in invertebrates; however, the loop is either
partially or entirely absent in CBCPs from some
parasites and nematodes [22–25]. Two types of CBCP
that have a loop containing a dihistidine motif
A
C
B
Fig. 1. Gel electrophoresis and N-terminal sequence of the purified
enzyme. (A) Native PAGE of the purified enzyme (0.5 lg) in the
absence of SDS. (B) SDS ⁄ PAGE of the purified enzyme (2 lg) and
N-terminal sequence of bands a–d identified by Edman degradation.
The marker proteins were as follows: b-galactosidase (116 kDa),
phosphorylase b (97 kDa), BSA (67 kDa), ovalbumin (45 kDa), glyc-
eraldehyde-3-phosphate dehydrogenase (36 kDa), trypsin inhibitor
(20 kDa) and egg cystatin (12.8 kDa). (C) Similarities of N-terminal
sequences between the purified enzyme and other plant cysteine
proteases. The cysteine proteases are: CBCP (DDBJ accession
numbers: Arabidopsis, AF370193; wheat, X66012; potato,
AY450641), RD21A (DDBJ accession number AK221689), XCP2
(DDBJ accession number AF191028), F9P14.12 (DDBJ accession
number AC025290) and aleurain (DDBJ accession number
AF233883). Identical amino acid residues are shown in bold.
A. Tsuji et al. Characterization of daikon CBCP
FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS 5431
Fig. 2. Nucleotide sequence of the cDNA encoding the purified enzyme and its deduced amino acid sequence. The deduced amino acid
sequence is displayed below the nucleotide sequence in a one-letter code. The nucleotide sequences of primers corresponding to the N-ter-
minal (Met1–Cys9) and C-terminal (Ala337–Lys343) sequences were derived from Arabidopsis CBCP (DDBJ accession number AF370193).
The positions of primers used for PCR amplification are marked with bold lines. The amino acid sequences identified are underlined. Amino
acids 1–31 and 32–103 comprise a putative signal peptide and propeptide, respectively. The essential active site residues Cys132, His287
and Asn308 are shown in bold. Possible glycosylation sites (Asn152 and Asn311) are double-underlined. The accession number of the

sequence in the DDBJ ⁄ EMBL ⁄ GenBank nucleotide sequence databases is AB377273.
Characterization of daikon CBCP A. Tsuji et al.
5432 FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS
(Trypanosoma congolense-1) or a single histidine
(T. congolense-6) were present in Trypanosoma [25], as
shown in Fig. 4. Rehman and Jasmer classified nema-
tode CBCPs into four groups on the basis of charac-
teristics of the occluding loop region [22]. Group I
members possess a similar occluding loop structure
containing a dihistidine motif. Group II members lack
a dihistidine motif but have a single histidine in the
loop. Group III and IV members lack histidine in the
loop. In Group IV members, loop-forming cysteines
appear to be present, although seven or eight amino
acids that would ordinarily form the terminal loop
upon disulfide bridge formation are absent. In con-
trast, all plant CBCPs possess a shorter occluding
loop containing a single histidine residue. In an effort
to identify the role of this short occluding loop in
plant CBCP function, the cleavage specificity of dai-
kon CBCP was analyzed.
Fig. 3. Sequence similarities between plant CBCP and animal cathepsin B. The five sequences were aligned for maximum homology. The
proteases are: daikon CBCP, Arabidopsis CBCP (DDBJ accession number AF370193, amino acids 1–344), tobacco (Nicotiana benthamiana)
CBCP (DDBJ accession number DQ492887, amino acids 1–341), human cathepsin B (DDBJ accession number M14331, amino acids 1–335)
and mouse cathepsin B (DDBJ accession number M14222, amino acids 1–335). Identical amino acid residues of the aligned sequences are
marked with asterisks. Active site residues are indicated by arrows. The occluding loops of human and mouse cathepsin B are underlined.
A. Tsuji et al. Characterization of daikon CBCP
FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS 5433
Cleavage specificity of CBCP towards synthetic
substrates

The enzymatic properties of daikon CBCP were inves-
tigated using synthetic substrates. When the effect of
pH on the activity towards t-butyloxycarbonyl (Boc)-
Leu-Arg-Arg-MCA was examined, the enzyme dis-
played the highest activity at pH 5.5 and the enzyme
was stable at acidic pH (pH 5–6.5). When the enzyme
was incubated at 25 °C for 10 min at pH 8.0, most of
the activity was lost. CBCP had optimal activity at
40 °C. The activity of the purified enzyme was also
examined against various synthetic substrates and
compared with that of human cathepsin B (Table 1).
Of the substrates examined, Boc-Leu-Arg-Arg-MCA
and l-pyroglutamyl (pyr)-Arg-Thr-Lys-Arg-MCA were
shown to be good substrates for daikon CBCP. The
enzyme hydrolyzed substrates with a basic amino acid
at the P1 position, and showed preference for a
positively charged amino acid at the P2 position.
The k
cat
⁄ K
m
values of Boc-Leu-Arg-Arg-MCA and
pyr-Arg-Thr-Lys-Arg-MCA were 2.8-fold and 1.6-fold
higher, respectively, than that of carbobenzoxy
(Z)-Phe-Arg-MCA, the latter substrate being com-
monly used for cathepsin B assays. For tripeptide sub-
strates with dibasic amino acids (Arg-Arg) at the P1
Fig. 4. Alignment of occluding loop sequences of plant CBCP from daikon (DDBJ accession number AB377273), A. thaliana (DDBJ acces-
sion number AF370193), rice (DDBJ accession number AY916493), barley (DDBJ accession number AJ310426), wheat (DDBJ accession
number X66012), cowpea (DDBJ accession number AM748426), barrel medic (DDBJ accession number ABD149038), garden pea (DDBJ

accession number AJ251536), potato (DDBJ accession number AY450641), sweet potato (DDBJ accession number AAK69541), tobacco
(DDBJ accession number X81995), grape (DDBJ accession number CAO040249), coast spruce (DDBJ accession number EF083997), moss
(DDBJ accession number EDQ82213), human (DDBJ accession number M14221), cow (DDBJ accession number L06075), mouse (DDBJ
accession number M14222), chicken (DDBJ accession number U18083), Atlantic halibut (DDBJ accession number DQ993253), yellow meal-
worm (DDBJ accession number DQ356051), Clonorchis sinensis (DDBJ accession number EF102086), Caenorhabditis elegans (cpr-6; DDBJ
accession number L39894), T. congolense-1 (TcoCB1; DDBJ accession number EU233643) and T. congolense-6 (TcoCB6; DDBJ accession
number EU233648). His110 and His111 (human cathepsin B numbering), which are essential for the exopeptidase activity of cathepsin B,
are arrowed.
Characterization of daikon CBCP A. Tsuji et al.
5434 FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS
and P2 positions, the enzyme activity increased with
the order Leu > Gln at the P3 position. Substrates
with glycine at the P2 position, such as Boc-Ile-Glu-
Gly-Arg-MCA and Boc-Gln-Gly-Arg-MCA, and sub-
strates with a single amino acid, such as Z-Arg-MCA,
were only marginally hydrolyzed by daikon CBCP
(data not shown). The enzyme did not hydrolyze
substrates for chymotrypsin-like enzyme (glutaryl-Ala-
Ala-Phe-MCA) and aminopeptidase (Leu-MCA, Arg-
MCA). In contrast, human cathepsin B is most active
against Boc-Leu-Arg-Arg-MCA and Z-Phe-Arg-MCA.
Both enzymes exhibit preference for leucine at the P3
position for tripeptide substrates with Arg-Arg at the
P1 and P2 positions. The Boc-Leu-Arg-Arg-MCA to
Boc-Gln-Arg-Arg-MCA k
cat
⁄ K
m
ratios for daikon
CBCP (18%) and human cathepsin B (20%) are simi-

lar. The most remarkable difference in cleavage speci-
ficity between daikon CBCP and human cathepsin B is
reflected in the absolute value of k
cat
⁄ K
m
. The k
cat
⁄ K
m
values of Boc-Leu-Arg-Arg-MCA (10.8-fold), pyr-
Arg-Thr-Lys-Arg-MCA (31.1-fold), Z-Phe-Arg-MCA
(4.3-fold), Z-Arg-Arg-MCA (16.4-fold) and Boc-Gln-
Arg-Arg-MCA (9.9-fold) for daikon CBCP were mark-
edly higher than those for human cathepsin B. The
difference appears to be largely due to an increase in
k
cat
. These results suggest that daikon CBCP possesses
higher endopeptidase activity than human cathepsin B.
Hydrolysis of glucagon and adrenocorticotropic
hormone (ACTH)
The ability of the enzyme to hydrolyze model peptides
was examined. Human glucagon and ACTH 1–24 were
digested using the purified enzyme at 37 °C for 20 h.
Cleavage products were separated by RP-HPLC, and
peptide amino acid sequences were determined. As
shown in Fig. 5A, three dipeptides (Leu26-Met27,
Phe22-Val23 and Tyr13-Leu14), one tripeptide (Phe22-
Val-Gln24) and larger fragments (His1-Lys12 and

His1-Arg18) represented the hydrolysis products fol-
lowing glucagon digestion. These results indicated that
the Lys12-Tyr13, Leu14-Asp15, Arg18-Ala19, Asp21-
Phe22, Val23-Gln24, Gln24-Trp25 and Met27-Asn28
bonds were cleaved by the purified enzyme. These
results strongly suggested that Asn28-Thr29, Leu26-
Met27, Gln24-Trp25 and Phe22-Val23 dipeptides were
released sequentially by the dipeptidyl carboxypepti-
dase activity of CBCP. Peaks 13, 23 and 29 were
applied to the sequencer; however, no N-terminal
amino acids were identified. It is highly likely that
hydrophilic dipeptides dissociate from the glass fiber
disk coated with polybrene used in the sequence analy-
sis. ACTH was also digested, and sequences of the
generated peptides were analyzed (Fig. 5B). Peak 31
eluted faster than control ACTH; however, its total
sequence could not be determined. Digestion of ACTH
resulted in cleavage of the His6-Phe7, Phe7-Arg8,
Arg8-Trp9, Gly14-Lys15, Val20-Lys21, Lys21-Val22
and Val22-Tyr23 bonds. Although the C-terminal tri-
peptide Val-Tyr-Pro24 was identified as a major cleav-
age product in peak 11, the C-terminal dipeptide
Tyr23-Pro24 was identified as a minor cleavage
product in peak 7. The internal peptide fragments
Arg8-Gly14 and Lys15-Val20 were generated by the
digestion. It is highly likely that the Phe7-Arg8,
Gly14-Lys15 and Lys21-Val22 bonds were cleaved by
the endopeptidase activity of the enzyme. These results
indicated that daikon CBCP possesses both endopepti-
dase and dipeptidyl carboxypeptidase activities. In

order to draw a comparison with the cleavage specific-
ity of human cathepsin B, ACTH was digested with
human cathepsin B under the same conditions for
20 h, and its cleavage products were analyzed. Two
major products, the C-terminal peptide Tyr23-Pro24
and N-terminal peptide Ser1-Lys21 were identified,
suggesting that dipeptide Tyr23-Pro24 was generated
by the dipeptidyl carboxypeptidase action of cathep-
sin B. The N-terminal peptide Ser1–Val22 was detected
as a minor product. Therefore, daikon CBCP seemed
to exhibit a stronger preference for hydrolyzing sub-
strates via its endopeptidase activity as compared with
human cathepsin B.
Hydrolysis of b-casein
Bovine b-casein was digested with the purified enzyme
at pH 5.5, and SDS ⁄ PAGE was used to analyze the
Table 1. Kinetic constants for the hydrolysis of peptide MCA sub-
strates and Abz-Phe-Arg-Phe(4-NO
2
)-OH by CBCP and human liver
cathepsin B. Enzyme activity was determined in 0.1
M acetate buf-
fer (pH 5.5) at 37 °C. The concentration of daikon CBCP and human
cathepsin B were determined by active site titration with E-64.
Standard errors for the determination of K
m
and k
cat
were lower
than 7%.

Substrate
Daikon CBCP
Human
cathepsin B
k
cat
(s
)1
)
K
m
(lM)
k
cat
⁄ K
m
(M
)1
Æs
)1
)
k
cat
⁄ K
m
(M
)1
Æs
)1
)

Boc-Leu-Arg-Arg-MCA 14.2 19.2 740 000 68 400
Pyr-Arg-Thr-Lys-Arg-MCA 7.59 18.2 417 000 13 400
Z-Phe-Arg-MCA 5.23 20.0 262 000 60 800
Z-Arg-Arg-MCA 5.70 25.0 228 000 13 900
Boc-Gln-Arg-Arg-MCA 1.32 9.80 135 000 13 600
Abz-Phe-Arg-Phe
(4-NO
2
)-OH
1.02 65.8 15 500 Not
detected
A. Tsuji et al. Characterization of daikon CBCP
FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS 5435
A
B
C
Fig. 5. Cleavage of glucagon, ACTH and
b-casein by daikon CBCP. HPLC profiles of
glucagon (A) and ACTH (B) digests. Peaks
were collected and sequenced. (C) Time-
course of b-casein digestion by daikon
CBCP. The digestion products were sepa-
rated by SDS ⁄ PAGE (15% gel), transferred
to a PVDF membrane, and then sequenced.
The amino acid sequences (one-letter code)
of the peptides and cleavage sites are indi-
cated by arrows. The arrow in parenthesis
indicates a minor cleavage site.
Characterization of daikon CBCP A. Tsuji et al.
5436 FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS

time-course of the digestion. The generated fragments
(bands a–e) separated by SDS ⁄ PAGE were transferred
onto a poly(vinylidene difluoride) (PVDF) membrane
and sequenced (Fig. 5C). Cleavage of the Gln54-
Thr55, Ser57-Leu58, Gly94-Val95 and Lys99-Glu100
bonds had occurred. Although b-casein possesses two
arginines at positions 25 and 207, the Ile26-Arg207
fragment was not found.
Monopeptidyl carboxypeptidase activity of CBCP
Cathepsin X is a lysosomal cysteine protease that
belongs to the papain family and acts primarily as a
monopeptidyl carboxypeptidase [26–28]. Superposition
of the cathepsin X and cathepsin B structures indicates
that His23 of the miniloop of cathepsin X occupies a
region in space that partially overlaps with His110 of
the occluding loop of cathepsin B [29,30]. The occlud-
ing loop of daikon CBCP is shorter than that present
in human cathepsin B and contains only a single
histidine, like cathepsin X. These data suggested
that daikon CBCP, like cathepsin X, possesses carb-
oxymonopeptidase activity. This idea was also sup-
ported by the data from the model peptide digestion
suggesting that Trp25 of glucagon and Lys21 of
ACTH were cleaved by CBCP via carboxymonopepti-
dase activity. In an effort to determine whether daikon
CBCP possesses carboxymonopeptidase activity, the
quenched fluorogenic substrate o-aminobenzoyl (Abz)-
Phe-Arg-Phe(4NO
2
)-OH, which contains a fluorescent

N-terminal Abz group internally quenched by interac-
tion with a nitrophenylalanyl residue at the P1¢ posi-
tion [26], was used. In a control experiment, human
cathepsin B activity towards this substrate was also
assayed. Na
¨
gler et al. reported that the k
cat
⁄ K
m
(m
)1
Æs
)1
) towards this substrate for human cathepsin X
was 1.23 · 10
5
at pH 5.0. The k
cat
⁄ K
m
(m
)1
Æs
)1
) for
daikon CBCP was estimated to be 15 500 (k
cat
,
1.02 s

)1
; K
m
, 65.8 lm) (Table 1). On the other hand,
the monopeptidyl carboxypeptidase activity of human
cathepsin B detected was negligible. These results sug-
gested that daikon CBCP, like cathepsin X, possesses
monopeptidyl carboxypeptidase activity.
Susceptibility to inhibitors
The effects of various protease inhibitors on daikon
CBCP were investigated. Although the enzyme activity
was strongly inhibited by cysteine protease inhibitors
such as leupeptin, antipain, E-64, egg white cystatin,
chymostatin, Na-p-tosyl-l-lysine chloromethyl ketone
(TLCK) and iodoacetamide (Table 2). Phenyl-
methanesulfonyl fluoride, pepstatin, soybean trypsin
inhibitor, o-phenanthroline and EDTA had no inhibi-
tory effects. It should be noted that daikon CBCP,
unlike cathepsin B, was not significantly inhibited
by 1 lm N-(l-3-trans-propylcarbamoyl-oxirane-2-car-
bonyl)-l-isoleucyl-l-proline (CA074). CA074 exhibited
10
4
-fold greater inhibition of cathepsin B than cathep-
sin L and cathepsin H [31]. This specificity of CA074
has been attributed to the presence of two histidines in
the occluding loop of cathepsin B [16,32]. CA074 is an
irreversible inhibitor of cathepsin B, so the second-
order rate constants of inactivation of daikon CBCP
and human cathepsin B by CA074 were compared.

When Boc-Leu-Arg-Arg-MCA was used as a substrate,
product formation by daikon CBCP in the presence of
CA074 approached an asymptote, indicating that dai-
kon CBCP is also inhibited irreversibly by CA074. The
second-order rate constants of daikon CBCP and
human cathepsin B with CA074 were 57.0 m
)1
Æs
)1
and
4.02 · 10
5
m
)1
Æs
)1
, respectively, indicating that daikon
CBCP was 7000-fold less sensitive than human cathep-
sin B to CA074.
Egg cystatin is a competitive reversible inhibitor of
papain-like cysteine proteases. The K
i
value of 9.5 nm
for the human cathepsin B–cystatin interaction was
found to be consistent with values obtained by others
[33]. On the other hand, the K
i
value for daikon CBCP
was estimated to be 38 nm, indicating that daikon
CBCP was slightly less sensitive to cystatin than

human cathepsin B.
Expression of CBCP during germination
The expression of CBCP in the germinating embryo
was analyzed at both the RNA and protein levels. Cot-
yledons of embryos (3, 5 and 8 days after imbibition)
were extracted and subjected to gel filtration on Seph-
acryl S-200 (Fig. 6A). The Boc-Leu-Arg-Arg-MCA-
cleaving activity in the second peak increased following
germination, and the activity reached a maximum at
Table 2. Effects of various cysteine protease inhibitors on the
activity of CBCP. Activities toward Boc-Leu-Arg-Arg-MCA were
assayed at pH 5.5 in the presence of various inhibitors.
Protease inhibitor Concentration % Inhibition
Leupeptin 1 l
M 98.4
Antipain 1 l
M 98.3
E-64 1 l
M 95.4
CA074 1 l
M 21.0
Egg cystatin 1 l
M 92.7
Chymostatin 1 l
M 57.2
TLCK 1 l
M 80.4
Iodoacetamide 2 m
M 90.6
A. Tsuji et al. Characterization of daikon CBCP

FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS 5437
5 days after imbibition. The expression of CBCP was
analyzed at the protein level using cystatin–Sepharose
[34]. As shown in Fig. 6B, a single 28-kDa protein was
clearly detected from cotyledons obtained 8 days after
imbibition. From cotyledons obtained 3 days after
imbibition, a faint 28 kDa band and additional bands
were also detected. A clear 28 kDa band and addi-
tional bands were observed from cotyledons obtained
5 days after imbibition. The identity of the 28 kDa
protein was confirmed by amino acid sequencing, and
the sequence LPXSFDAE, identical to the N-terminal
sequence of CBCP, was obtained. Other sequences
were not present. The time-course of cysteine protease
activity assayed using Boc-Leu-Arg-Arg-MCA and
expression of the 28 kDa protein are shown in
Fig. 6C. Expression of mature CABP at 5 and 8 days
after imbibition increased approximately threefold
relative to that observed at 3 days after imbibition.
Although the enzyme activity reached a maximum
at 5 days and subsequently decreased, the expression
levels of the 28 kDa protein at 5 and 8 days after imbi-
bition were almost the same. The cystatin–Sepharose
protein-binding profiles indicated the presence of
other cysteine proteases that have smaller molecular
masses than CBCP. It is likely that, as the expression
of other cysteine proteases decreased markedly at
8 days, the Boc-Leu-Arg-Arg-MCA-cleaving activity
also decreased. In an effort to determine whether the
increase in CBCP activity results from transcriptional

or post-transcriptional events, the expression of CBCP
transcript was investigated by RT-PCR. As shown in
Fig. 6D, the level of CBCP transcript in cotyledons
was highest at 2 days after imbibition and decreased
gradually during days 3–8. It is therefore highly likely
that the expression of CBCP activity in cotyledons
during germination is controlled by post-transcrip-
tional events.
Discussion
The current investigation highlighted the cleavage spec-
ificity and inhibitor sensitivity of daikon CBCP, which
possesses a short occluding loop lacking a dihistidine
motif. The unique occluding loop structure of cathep-
sin B, which was previously used to identify the
enzyme, also represents the point at which the cathep-
sin B-specific inhibitor CA074 acts [16,32]. The occlud-
ing loop plays an important role in the dipeptidyl
carboxypeptidase activity of cathepsin B and interac-
tion of the propeptide with the active site [13–15].
Deletion of the entire loop abolished exopeptidase
activity while conserving endopeptidase activity [13].
AB
D
C
Fig. 6. Expression analysis of daikon CBCP during germination. (A) Gel filtration of extract derived from daikon cotyledons (25 g) following
imbibition (3, 5 and 8 days). The Boc-Leu-Arg-Arg-MCA-hydrolyzing activity at pH 5.5 was assayed. Fractions indicated by the horizontal bar
were pooled, concentrated and used for cystatin–Sepharose analysis. (B) Identification of cysteine proteases by cystatin–Sepharose analysis.
Proteins bound to cystatin–Sepharose were separated by SDS ⁄ PAGE (12% gel) and sequenced as described in Experimental procedures.
(C) The Boc-Leu-Arg-Arg-MCA-hydrolyzing activity in the second peak in (A) and the intensities of the 28 kDa band in (B) are plotted against
days after imbibition. (D) RT-PCR analysis of CBCP transcript in cotyledons of germinating embryos (0, 1, 2, 3, 4, 5, 6 and 8 days after imbi-

bition). CBCP cDNA encoding the catalytic domain was amplified by PCR as described in Experimental procedures. Actin 2 transcript was
amplified as an internal control.
Characterization of daikon CBCP A. Tsuji et al.
5438 FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS
As shown in Fig. 4, sequences within the occluding
loop of cathepsin B are highly conserved in vertebrates
and invertebrates. However, some parasitic organisms
and nematodes possess a modified loop sequence, com-
prising either amino acid or small peptide inserts or
large deletions [23–25]. In contrast, plant CBCP pos-
sesses only one histidine within the loop. Furthermore,
the segment between the two cysteines is very short
(four residues) as compared with animal cathepsin B
(10 residues). The cleavage specificities of daikon
CBCP and human cathepsin B towards MCA sub-
strates were similar. Boc-Leu-Arg-Arg-MCA was a
good substrate for both enzymes. However, the k
cat
values for daikon CBCP towards all synthetic sub-
strates tested were higher than the values for human
cathepsin B.
The specificity of the endopeptidase activity towards
peptides and protein substrates should be investigated
with caution, given that fragments released by the
endopeptidase can be further digested by the dipeptidyl
carboxypeptidase, thus making identification of the ini-
tial endopeptidase cleavage site difficult. Therefore, the
endopeptidase cleavage site was identified from N-ter-
minal residues of the generated internal peptides
(longer than three residues). Analyses of digestion

products of glucagon, ACTH and b-casein showed
that daikon CBCP possesses both endopeptidase and
dipeptidyl carboxypeptidase activities. Furthermore,
daikon CBCP exhibited monopeptidyl carboxypepti-
dase activity towards Abz-Phe-Arg-Phe(4-NO
2
)-OH.
Although human cathepsin B primarily acts on gluca-
gon [11] and ACTH as a dipeptidyl carboxypeptidase,
daikon CBCP digested these peptides utilizing both
endopeptidase and dipeptidyl carboxypeptidase activi-
ties. Cleavage products from ACTH mainly resulted
from the endopeptidase action of CBCP, whereas it is
likely that Trp25 and Gln24 might have been removed
from the C-terminus by the monopeptidyl carboxypep-
tidase action of CBCP during the digestion of
glucagons. Furthermore, the peptide containing Ala19-
Gln-Asp21 was not detected. This peptide might be
digested by the monopeptidyl carboxypeptidase action
of CBCP. It is highly likely that the Phe7-Arg8,
Gly14-Lys15 and Lys21-Val22 bonds in ACTH, and
the Gln54-Thr55, Ser57-Leu58, Gly94-Val95 and
Lys99-Glu100 bonds in b-casein, were cleaved by the
endopeptidase action of CBCP. A strict preference for
amino acids at the P1, P2, P3, P1¢ and P2¢ substrate
positions was not observed for CBCP.
The sensitivity of cathepsin B and CBCP to CA074
clearly differed. CA074 is a strong, irreversible inhibi-
tor of cathepsin B [16,31,32]. The second-order rate
constant of inactivation for CBCP (57.0 m

)1
Æs
)1
) was
markedly lower (7000-fold) than the constant for
cathepsin B (4.02 · 10
5
m
)1
Æs
)1
), as expected, given the
inhibitory mechanism of CA074. These results con-
firmed that the His110-His111 sequence within the
occluding loop is essential for the strong inhibitory
activity of CA074. High molecular weight inhibitors
provide a way to probe active site accessibility. Illy
et al. reported that the occluding loop deletion mutant
of human cathepsin B showed 40-fold higher affinity
for cystatin c than wild-type cathepsin B [13]. A K
i
value of 9.5 nm for the human cathepsin B–cystatin
interaction was found, whereas the K
i
value for daikon
CBCP was estimated to be 38 nm, indicating that dai-
kon CBCP was slightly less sensitive to cystatin than
human cathepsin B. These results suggested that the
influence of the occluding loop on active site accessibil-
ity of daikon CBCP and human cathepsin B is almost

the same. Krupa et al. showed that His110 is critical
for the exopeptidase activity of cathepsin B, with the
Asp22–His110 ion pair stabilizing the electrostatic
interaction, whereas His111 makes a positive 10-fold
contribution to the exopeptidase activity [14]. Asp22 is
conserved in CBCP. These results suggest that His102
in the occluding loop of CBCP corresponds to His110
of human cathepsin B.
More recently, the cleavage specificities of two types
of recombinant CBCP from T. congolense, TcoCBc1
with His-His and TcoCBc6 with only one His residue
in the occluding loop, were compared [25]. The former
enzyme, like human cathepsin B, exhibited weak endo-
peptidase and strong exopeptidase (carboxypeptidase)
activities. In contrast, the latter enzyme, unlike daikon
CBCP, only showed carboxypeptidase activity. There
are several differences in occluding loop structure and
inhibitor sensitivity between Trypanosoma and daikon
CBCP. First, the segment of 11 amino acids between
the two cysteines in the occluding loop of TcoCBc1
and TcoCBc6 is longer than that of daikon CBCP
(four amino acids), although the amino acid sequences
of CBCPs from T. congolense are homologous with
daikon CBCP (39% amino acid identity). Second, the
inhibitor sensitivities of Trypanosoma CBCP and dai-
kon CBCP differ markedly. TcoCBc1 was not inhib-
ited by leupeptin, chymostatin or cystatin, but was
inhibited by CA074 and E-64. These results indicate
that CBCP from T. congolense is an atypical CBCP in
terms of its enzymatic properties. Therefore, TcoCB6

and daikon CBCP are distinct from one another in
terms of the role of the one histidine in the occluding
loop.
As shown in Figs 3 and 4, the amino acids around the
subsites of cathepsin B and the unusual occluding
loop structure are highly conserved in plant CBCPs.
A. Tsuji et al. Characterization of daikon CBCP
FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS 5439
Therefore, it is highly likely that plant CBCPs have the
same enzymatic characteristics as daikon CBCP. The
evolutionary difference in the occluding loop sequence
between animal cathepsin B and plant CBCP may be
linked to differences in their physiological roles or their
activation mechanism. In plants, protein mobilization
during germination is initiated by acidification of the
vacuole, followed by activation of cysteine proteases by
removal of the propeptide [35,36]. The autoprocessing
activity of human cathepsin B is impaired following
deletion of the occluding loop [13]. The affinity of the
propeptide of cathepsin B for the loop-deletion mutant
of cathepsin B was higher than for the wild-type
enzyme. As shown in Fig. 6, the increase in CBCP activ-
ity in cotyledons during germination is likely to be regu-
lated by post-transcriptional events. The activity
increased markedly between 3 and 5 days after imbibi-
tion. It would be of interest to examine the role of the
occluding loop in the autoactivation of plant CBCP.
In conclusion, although daikon CBCP has several
characteristics in common with cathepsin B, there are
some important differences in their respective endopep-

tidase and exopeptidase activity profiles. As compared
with animal cathepsin B, it is likely that plant CBCP
has stronger endopeptidase and monopeptidyl car-
boxypeptidase activities.
Experimental procedures
Materials
Peptide MCA substrates, glucagon, ACTH 1–24, leupeptin,
E-64, chymostatin and CA074 were purchased from the
Peptide Institute (Osaka). Abz-Phe-Arg-Phe(4-NO
2
)-OH
was from MP Biomedicals (Solon, OH, USA). Sephacryl
S-200 was purchased from GE Healthcare (Uppsala, Swe-
den). Organomercurial Sepharose was prepared by coupling
4-aminophenyl-mercuric acetate with cyanogen bromide-
activated Sepharose [37]. Egg cystatin was purified by the
method of Anastasi et al. [38], and immobilized on Sepha-
rose 4B [34]. Purified human cathepsin B from liver was
from Sigma (St Louis, MO, USA). All other chemicals used
were of analytical grade.
Plants
Fresh sprouts of daikon radish (R. sativus) grown by hydro-
ponics were purchased from a grocery store and used for
purification of the enzyme. For analysis of expression of the
enzyme during germination, daikon radish seeds were
allowed to imbibe water on filter paper at 20 °C in the dark.
After 2 days, embryos were grown at 20 °C under long-day-
light conditions (16 h light, 8 h dark) until collection.
Enzyme assay
Endopeptidase activity was assayed using Boc-Leu-Arg-

Arg-MCA unless otherwise indicated. The reaction mixture
contained 0.1 m acetate buffer (pH 5.5) and 50 lm
Boc-Leu-Arg-Arg-MCA. The reaction was initiated by
addition of the enzyme solution. Following incubation at
37 °C for 10–20 min, the reaction was terminated and the
MCA liberated was determined fluorometrically as previ-
ously described [39]. Monopeptidyl carboxypeptidase
activity was measured using the synthetic substrate Abz-
Phe-Arg-Phe(4-NO
2
)-OH. Assay mixtures comprised 10 lm
substrate in 0.1 m acetate buffer (pH 5.5) containing 3%
dimethylsulfoxide. The reaction was initiated with enzyme,
and substrate hydrolysis was monitored by excitation at
320 nm and emission at 420 nm at 25 °C using a Hitachi
F-2000 spectrofluorometer. Protein was determined by the
method of Bradford, using BSA as a standard [40].
Purification of cathepsin B-like cysteine protease
from daikon radish
All purification procedures were performed at 4 °C unless
otherwise indicated. Cotyledons of daikon radish (200 g)
were homogenized in 1 L of ice-cold 20 mm acetate buffer
(pH 6.5) containing 0.5 mm EDTA and 1 mm b-mercapto-
ethanol (buffer A), using a Waring blender. The homogenate
was centrifuged at 12 000 g for 10 min at 4 °C. The superna-
tant was fractionated with ammonium sulfate (60% satura-
tion). The precipitate was dissolved in buffer A and dialyzed
against the same buffer. Following centrifugation at 12 000 g
for 10 min at 4 °C, the resultant supernatant was concen-
trated by ultrafiltration (Amicon YM-10 filter). The concen-

trate was applied to a Sephacryl S-200 column
(2.5 · 100 cm) equilibrated with buffer A containing 0.1 m
NaCl. Boc-Leu-Arg-Arg-MCA-cleaving activity was eluted
as two peaks. The activity in the second peak was completely
inhibited by E-64. The fraction containing E-64-sensitive
activity was concentrated, dialyzed against buffer A, and
applied to a DEAE–cellulose column (1.0 · 6.5 cm) equili-
brated with the same buffer. After washing with buffer A,
Boc-Leu-Arg-Arg-MCA-cleaving activity was eluted using a
linear gradient of NaCl (0–0.2 m) in the same buffer. The
fraction possessing activity was eluted as a single peak, con-
centrated by ultrafiltration, and then applied to a hydroxy-
apatite column (0.8 · 1.5 cm) equilibrated with buffer A.
The activity was eluted using a linear gradient of sodium
phosphate (0–0.2 m) in the same buffer. The fraction with
activity was applied to organomercurial–Sepharose
(0.6 · 0.7 cm) and extensively washed with buffer A contain-
ing 0.5 m NaCl. The enzyme activity was eluted with buf-
fer A containing 0.5 m NaCl and 10 m ml-cysteine. PAGE
was performed in Tris ⁄ glycine buffer (pH 8.9) by the method
of Davis, using a 7.5% gel [41]. The gel was then stained with
Imperial Protein Stain (Pierce, Rockford, IL, USA).
Characterization of daikon CBCP A. Tsuji et al.
5440 FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS
Sequence analysis of CBCP
The purified enzyme separated by SDS ⁄ PAGE [42] was
electroblotted onto a PVDF membrane (Immobilon,
0.45 lm; Millipore, Bedford, MA, USA), according to the
manufacturer’s instructions. The protein band was detected
by staining with Ponceau 3R, and then cut into small pieces

and applied to an automated protein sequencer (Shimadzu
PPSQ-10, Kyoto, Japan).
Cleavage specificity of CBCP towards glucagon,
ACTH and b-casein
For the cleavage specificity analysis, 5 nmol of peptide sub-
strates (glucagon and ACTH) were digested with purified
CBCP (40 pmol) at 37 °C for 20 h. The mixture was then
acidified with trifluoroacetic acid and separated by RP-
HPLC on a lRPC C2 ⁄ C18 pc3.2 ⁄ 3 column (GE Healthcare
Bio-Sciences, Uppsala, Sweden), as previously described
[43]. Purified peptide fragments were applied to glass fiber
disks that had been coated with polybrene, and then ana-
lyzed using an automated protein sequencer. Bovine
b-casein (5 nmol) was digested with CBCP (20 pmol) at
37 °C. Digested fragments were separated by SDS ⁄ PAGE
(15%) and transferred onto a PVDF membrane, and N-ter-
minal sequences were analyzed using an automated protein
sequencer.
Determination of rates of inactivation by CA074
Inhibition of daikon CBCP by CA074 was evaluated
using the method of Tian & Tsou [44]. To monitor time-
based inactivation by CA074, the endopeptidase activity
towards Boc-Leu-Arg-Arg-MCA (50 lm) in the presence
of CA074 was continuously recorded at 25 °C using a
Hitachi F-2000 spectrofluorometer (excitation at 380 nm
and emission at 460 nm). Inactivation rate constants
(k
obs
,s
)1

) were determined from the slopes of semi-log
plots of enzyme activity versus time. Apparent second-
order rate constants of inactivation were determined from
replots of k
obs
versus [I], using six values of [I] (CBCP,
5–40 lm; cathepsin B, 20–200 nm).
Isolation of cDNA encoding the catalytic domain
of CBCP
To isolate cDNA encoding the catalytic domain of daikon
radish CBCP, we designed PCR primers corresponding
to the N-terminal sequence (LPKSFDART) of purified
enzyme and a conserved sequence (AGLPSSKN) in
the C-terminal region of the catalytic domain of plant
CBCP. The sequences of the sense and antisense primers
were 5¢-CTACCTAAATCTTTTGATGCTAGAAC-3¢ and
5¢-TTCTTGCTTGAAGGCAAACCAGC-3¢, respectively.
Total RNA was isolated from the cotyledons of daikon
radish 5 days after imbibition using Trizol (Invitrogen,
Carlsbad, CA, USA) according to the manufacturer’s
protocol. One microgram of total RNA was reverse-tran-
scribed with Moloney murine leukemia virus reverse trans-
criptase (Wako Chemicals, Osaka, Japan), using a random
hexamer. PCR was carried out by 30 cycles of denaturation
at 95 °C for 1 min and annealing at 53 °C for 30 s,
followed by extension at 72 ° C for 1 min. The DNA
sequence of the PCR product of expected size (720 bp)
was determined. To isolate cDNA encoding the propeptide
of CBCP, PCR was performed using the sense
primer 5¢-ATGGCTGTTTACAATACCAAACTCTG-3¢,

corresponding to the N-terminal sequence (Met1–
Cys9) of Arabidopsis CBCP, and the antisense primer
5¢-CAGCACCAAATGCCCAGCAAGAACC-3¢, corres-
ponding to the internal sequence of the first PCR product
(nucleotide number 79–103), as described above. The com-
posite cDNA sequence of the two PCR products included
1029 nucleotides, and overlapping nucleotides were identi-
cal. The sequence of the sense primer of the first PCR
product was replaced with the corresponding sequence of
the second PCR product in the composite sequence.
Expression analysis of CBCP following
germination
Cotyledons (25 g) from germinating embryo were homo-
genized with 250 mL of 20 mm acetate buffer (pH 6.5) con-
taining 1 mm b-mercaptoethanol and 0.5 mm EDTA,
fractionated by ammonium sulfate precipitation (60% satu-
ration), and then subjected to gel filtration on Sephacryl
S-200 (2.0 · 105 cm). The protease activity was assayed
using Boc-Leu-Arg-Arg-MCA. The fraction possessing
E-64-sensitive activity was concentrated to 1 mL and incu-
bated with 50 lL of cystatin–Sepharose at 4 °C. After 16 h,
the aliquots were centrifuged at 1500 g for 5 min and the
supernatants removed. More than 90% of Boc-Leu-Arg-
Arg-MCA-hydrolyzing activity was absorbed by the cysta-
tin gel. The gel was washed five times with 20 mm acetate
buffer (pH 6.5) containing 1 mm b-mercaptoethanol, 1 mm
EDTA, 0.1 m NaCl and 0.5% Triton X-100 by centrifuga-
tion at 1500 g for 5 min. Finally, the gel was suspended in
50 lL of loading buffer and treated at 95 °C for 5 min.
Ten microliters of sample was subjected to SDS ⁄ PAGE,

and the gel was stained with Imperial Protein Stain. The
remaining sample was used for sequence analysis. The
28 kDa protein band stained with Ponceau 3R was applied
to an automated peptide sequencer. The intensity of the
28 kDa band was quantified using nih image software. The
expression of CBCP transcripts in cotyledons (3, 4, 5, 6 and
8 days after imbibition) were analyzed by RT-PCR as
described above. Actin 2 transcript was amplified by PCR
using 5¢-GCTGTTCTCTCCCTGTACGCCAGTG-3¢ as the
A. Tsuji et al. Characterization of daikon CBCP
FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS 5441
sense primer and 5¢-CCAGCAGCTTCCATTCCCAC
GAACG-3¢ as the antisense primer. PCR was carried out
for 28 cycles (95 °C for 30 s; 53 °C for 30 s; 72 °C for
30 s), and identification of the PCR products was con-
firmed by sequencing.
Acknowledgements
The authors thank Tsuyoshi Yuasa, Tsuyoshi Kakim-
oto, Tohru Kawaguchi, Takashi Suzuki, Koudai
Katayama and Ayako Hayashi for their technical
assistance. We also thank Yuka Sasaki for determina-
tion of amino acid sequences.
References
1 Seki M, Narusaka M, Kamiya A, Ishida J, Satou M,
Sakurai T, Nakajima M, Enju A, Akiyama K, Oono Y
et al. (2002) Functional annotation of a full-length
Arabidopsis cDNA collection. Science 296, 141–145.
2 Schaller A (2004) A cut above the rest: regulatory func-
tion of plant proteases. Planta 220, 183–197.
3 Asano M, Suzuki S, Kawai M, Miwa T & Shibai H

(1999) Characterization of novel cysteine proteases from
germinating cotyledons of soybean [Glycine max (L.)
Merrill]. J Biochem 126, 296–301.
4 Koehler S & Ho T-H D (1988) Purification and
characterization of gibberellic acid-induced cysteine
endoproteases in barley aleurone layers. Plant
Physiol 87, 95–103.
5 Schmid M, Simpson D, Kalousek F & Gietl C (1998) A
cysteine endopeptidase with a C-terminal KDEL motif
isolated from caster bean endosperm is a marker
enzyme for the ricinosome, a putative lytic compart-
ment. Planta 206, 466–475.
6 Asp T, Bowra S, Borg S & Holm PB (2004) Molecular
cloning, functional expression in Escherichia coli and
enzymatic characterization of a cysteine protease from
white clover (Trifolium repens). Biochem Biophys Acta
1699, 111–122.
7 Cejudo FJ, Murphy G, Chinoy C & Baulcombe DC
(1992) A gibberellin-regulated gene from wheat with
sequence homology to cathepsin B of mammalian cells.
Plant J 2, 937–948.
8 Avrova AO, Taleb N, Rokka V-M, Heilbronn J, Camp-
bell E, Hein I, Gilroy EM, Cardle L, Bradshaw JE,
Stewart HE et al. (2004) Potato oxysterol binding
protein and cathepsin B are rapidly up-regulated in
independent defence pathways that distinguish R
gene-mediated and field resistances to Phytophthora
infestans. Mol Plant Pathol 5, 45–56.
9 Lidett AJ, Moran M, Wong KAL, Furze J, Rhodes
MJC & Hamill JD (1995) Isolation and expression

pattern of a cDNA encoding a cathepsin B-like
protease from Nicotiana rustica. Plant Mol Biol 29,
379–384.
10 Giroy EM, Hein I, Van der Hoorn R, Boevink PC,
Ventar E, McLellan H, Kaffarnik F, Hrubikova K,
Shaw J, Holeva M et al. (2007) Involvement of cathep-
sin B in the plant disease resistance hypersensitive
response. Plant J 52, 1–13.
11 Aronson NN Jr & Barrett AJ (1978) The specificity of
cathepsin B. Hydrolysis of glucagons at the C-terminus
by a peptidyldipeptidase mechanism. Biochem J 171,
759–765.
12 Na
¨
gler DK, Storer AC, Portano FCV, Camona E,
Juliano L & Me
´
nard R (1997) Major increase in endo-
peptidase activity of human cathepsin B upon removal of
occluding loop contacts. Biochemistry 36, 12608–12615.
13 Illy C, Quraishi O, Wang J, Purisima E, Vernet T &
Mort JS (1997) Role of the occluding loop in cathepsin
B activity. J Biol Chem 272, 1197–1202.
14 Krupa JC, Hasnain S, Na
¨
gler DJ, Me
´
nard R & Mort
JS (2002) S¢2 substrate specificity and the role of His
110

and His
111
in the exopeptidase activity of human
cathepsin B. Biochem J 361, 613–619.
15 Musil D, Zucic D, Turk D, Engh RA, Mayr I, Huber
R, Popovic T, Turk V, Towarari T, Katunuma N et al.
(1991) The refined 2.15A
˚
X-ray crystal structure of
human liver cathepsin B: the structural basis for its
specificity. EMBO J 10, 2321–2330.
16 Turk D, Podobnik M, Popovic T, Katunuma N, Bode
W, Huber R & Turk V (1995) Crystal structure of
cathepsin B inhibited with CA030 at 2.0-A
˚
resolution: a
basis for the design of specific epoxysuccinyl inhibitors.
Biochemistry 34, 4791–4797.
17 Van der Hoorn R, Leeuwenburgh MA, Bogyo M, Joos-
ten MHA & Peck SC (2004) Activity profiling of
papain-like cysteine proteases in plants. Plant Physiol
135, 1170–1178.
18 Chan SJ, San Segundo B, McCormick MB & Steiner
DF (1986) Nucleotide and predicted amino acid
sequences of cloned human and mouse preprocathepsin
B cDNAs. Proc Natl Acad Sci USA 83, 7721–7725.
19 Mordier S, Bechet D, Roux MP, Obled A & Ferrara M
(1993) Nucleotide sequence of bovine preprocathepsin
B. A study of polymorphism in the protein coding
region. Biochim Biophys Acta 1174, 305–311.

20 Dong SS, Stransky GI, Whitaker CH, Jordan SE,
Schlesinger PH, Edwards JC & Blair HC (1995) Avian
cathepsin B cDNA: sequence and demonstration that
mRNAs of two sizes are produced in cell types produc-
ing large quantities of the enzyme. Biochim Biophys
Acta 1251, 69–73.
21 Prabhakar S, Chen MS, Elpidina EN, Vinokurov KS,
Smith CM, Marshall J & Oppert B (2007) Sequence
analysis and molecular characterization of larval midgut
cDNA transcripts encoding peptidases from the yellow
Characterization of daikon CBCP A. Tsuji et al.
5442 FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS
mealworm, Tenebrio molitor L. Insect Mol Biol 16, 455–
468.
22 Rehman A & Jasmer DP (1999) Defined characteristics
of cathepsin B-like proteins from nematodes: inferred
functional diversity and phylogenetic relationship. Mol
Biochem Parasitol 102, 297–310.
23 Sajid M & McKerrow JH (2002) Cysteine proteases of
parasitic organisms. Mol Biol Parasitol 120, 1–21.
24 Dacks JB, Kuru T, Liapounova NA & Gedamu L (2008)
Phylogenetic and primary sequence characterization of
cathepsin B cysteine proteases from the oxymonad flagel-
late Monocercomonides. J Eukaryot Microbiol 55, 9–17.
25 Mendoza-Palomares C, Biteau N, Giroud C, Coustou
V, Coetzer T, Authie
´
E, Boulange
´
A & Baltz T (2008)

Molecular and biochemical characterization of a
cathepsin B-like protease family unique to Trypanosoma
congolense. Eukaryot Cell 7, 684–697.
26 Na
¨
gler DK, Zhang R, Tam W, Sulea T, Purisima EO
&Me
´
nard R (1999) Human cathepsin X: a cysteine
protease with unique carboxypeptidase activity.
Biochemistry 38, 12648–12654.
27 Klemenc
ˇ
ic
ˇ
V, Carmona AK, Cezari MHS, Juliano MA,
Juliano L, Gunc
ˇ
ar G, Turk D, Krizˇ aj I, Turk V & Turk
B (2000) Biochemical characterization of human cathep-
sin X revealed that the enzyme is an exopeptidase, act-
ing as carboxymonopeptidase or carboxydipeptidase.
Eur J Biochem 267, 5404–5412.
28 Therrien C, Lachance P, Sulea T, Purisima EO, Qi H,
Ziomek E, Alvarez-Hernandez A, Roush WR &
Me
´
nard R (2001) Cathepsins X and B can be differenti-
ated through their respective mono- and dipeptidyl car-
boxypeptidase activities. Biochemistry 40, 2702–2711.

29 Gunc
ˇ
ar G, Klemenc
ˇ
ic
ˇ
I, Turk B, Turk V, Karaoglanovic-
Carmona A, Juliano L & Turk D (2000) Crystal structure
of cathepsin X: a flip-flop of the ring of H23 allows car-
boxy-monopeptidase and carboxy-dipeptidase activity of
the protease. Structure 8, 305–313.
30 Sivaraman J, Na
¨
gler DK, Zhang R, Me
´
nard R &
Cygler M (2000) Crystal structure of human procathep-
sin X: a cysteine protease with the proregion covalently
linked to the active site cysteine. J Mol Biol 295, 939–951.
31 Towatari T, Nikawa T, Murata M, Yokoo C, Tamai
M, Hanada K & Katunuma N (1991) Novel epox-
ysuccinyl peptides: a selective inhibitor of cathepsin B,
in vivo. FEBS Lett 280, 311–315.
32 Yamamoto A, Hara T, Tomoo K, Ishida T, Fujii T,
Hata Y, Murata M & Kitamura K (1997) Binding
mode of CA074, a specific irreversible inhibitor, to
bovine cathepsin B as determined by X-ray crystal anal-
ysis of the complex. J Biochem 121
, 974–977.
33 Auerswald E, Genenger G, Assfalg-Machleidt I, Kos J

& Bode W (1989) Synthesis of a (desSer1 Ile29 Leu89)
chicken cystatin gene. Expression in E. coli as fusion
protein and its isolation. FEBS Lett 243, 186–192.
34 Tombaccini D, Mocali A, Weber E & Paoletti F (2001)
A cystatin-based affinity procedure for the isolation and
analysis of papain-like cysteine proteinases from tissue
extracts. Anal Biochem 289, 231–238.
35 Otegui M, Herder R, Schulze J, Jung R & Staehelin LA
(2006) The proteolytic processing of seed storage pro-
teins in Arabidopsis embryo cells starts in the multi-
vesicular bodies. Plant Cell 18, 2567–2581.
36 He F, Huang F, Wilson KA & Tan-Wilson A (2007)
Protein storage vacuole acidification as a control of
storage protein mobilization in soybeans. J Exp Bot 58,
1059–1070.
37 Barrett A (1973) Human cathepsin B1. Purification and
some properties of the enzyme. Biochem J 131, 809–822.
38 Anastasi A, Brown MA, Kembhavi AA, Nicklin MJH,
Sayers CA, Sunter DC & Barrett AJ (1983) Cystatin, a
peptide inhibitor of cysteine proteinases. Biochem J 211,
129–138.
39 Tsuji A & Kurachi K (1989) Isolation and characteriza-
tion of a novel large protease accumulated in mamma-
lian cells in the presence of inhibitors. J Biol Chem 264,
16093–16099.
40 Bradford MM (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein
utilizing the principle of protein-dye binding. Anal
Biochem 72, 248–254.
41 Davis BJ (1964) Disc electrophoresis. II. Method and

application to human serum proteins. Ann NY Acad Sci
121, 404–427.
42 Laemmli UK (1970) Cleavage of structure proteins dur-
ing the assembly of the head of bateriophage T4. Natute
227, 680–685.
43 Tsuji A, Yuasa K & Matsuda Y (2004) Identification of
oligopeptidase B in higher plants. Purification and char-
acterization of oligopeptidase B from quiescent wheat
embryo, Triticum aestivum. J Biochem 136, 673–681.
44 Tian W-X & Tsou C-L (1982) Determination of the
rate constant of enzyme modification by measuring the
substrate reaction in the presence of the modifier.
Biochemistry 21, 1028–1032.
A. Tsuji et al. Characterization of daikon CBCP
FEBS Journal 275 (2008) 5429–5443 ª 2008 The Authors Journal compilation ª 2008 FEBS 5443