Natural biflavones as?novel inhibitors of?cathepsin B and?K

Article abstract of DOI:10.1016/j.ejmech.2006.06.002

Cathepsin B and K, two important members in lysosomal proteases, involve in many serious human diseases, such as tumor and osteoporosis. In order to find their novel inhibitors, we performed the inhibition assays of cathepsin B and cathepsin K in vitro, randomly screened compounds from plants, and found six biflavones, named AMF1-5 and HIF, can potently inhibit cathepsin B and cathepsin K, especially AMF4 and HIF with IC₅₀ of 0.62 and 0.58?μM against cathepsin B. They are novel inhibitors for cathepsin B and K. Inhibition and flexible docking studies indicated that these biflavones are reversible inhibitors of cathepsin B, and their binding patterns and interaction modes with cathepsin B made them more specific to cathepsin B endopeptidase.

Full text of DOI:10.1016/j.ejmech.2006.06.002

European Journal of Medicinal Chemistry 41 (2006) 1247–1252
http://france.elsevier.com/direct/ejmech
Original article
Natural biflavones as novel inhibitors of cathepsin B and K
a,b
a,b
a,
a
a
*
G.-Z. Zeng , X.-L. Pan , N.-H. Tan , J. Xiong , Y.-M. Zhang
a State Key Laboratory of Phytochemistry and Plant Resources in West China,
Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650204, China
b Graduate School of the Chinese Academy of Sciences, Beijing 100039, China
Received in revised form 10 May 2006; accepted 2 June 2006
Available online 07 July 2006
Abstract
Cathepsin B and K, two important members in lysosomal proteases, involve in many serious human diseases, such as tumor and osteoporosis.
In order to find their novel inhibitors, we performed the inhibition assays of cathepsin B and cathepsin K in vitro, randomly screened compounds
from plants, and found six biflavones, named AMF1-5 and HIF, can potently inhibit cathepsin B and cathepsin K, especially AMF4 and HIF
with IC50 of 0.62 and 0.58 μM against cathepsin B. They are novel inhibitors for cathepsin B and K. Inhibition and flexible docking studies
indicated that these biflavones are reversible inhibitors of cathepsin B, and their binding patterns and interaction modes with cathepsin B made
them more specific to cathepsin B endopeptidase.
© 2006 Elsevier Masson SAS. All rights reserved.
Keywords: Biflavones; Inhibitors; Cathepsin B; Cathepsin K; Docking
1. Introduction
atom, and followed by proton donation from His199 (His162)
[2–4]. CatB is a unique member in cathepsins for its dual roles
as both endopeptidase and exopeptidase (dipeptidyl carboxy-
peptidase). The occluding loop (Ser104–Pro126) of CatB is
responsible for the latter activity. Because His110 and
His111, two important residues, face the prime side of the
substrate-binding cleft to stabilize the terminal carboxylate of
a peptide, this arrangement accounts for its dipeptidyl carbox-
ypeptidase activity [2–5]. There is a pH-dependent transition
for CatB from exo- to endo-peptidase. Below pH 5.5, CatB
mainly acts as an exopeptidase for its proteolytic degrading
activity in lysosomal. Above pH 5.5, a mixture of endo- and
exo-proteolytic CatB activities presents with a slightly greater
percent activity as an endopeptidase. CatB is active in neutral
condition, and it has maximum endopeptidase activity around
pH 7.4. In tumor invasion and metastasis, CatB was found out-
side the lysosomal compartment; it acts mainly as endopepti-
dase and plays an important role in degrading extracellular
matrix (ECM) and basement membrane [6–8]. Moreover,
increased expression, activity and secretion of cathepsin B
have been observed in more than one type of tumors, such as
breast, gastric, lung, prostate carcinomas and so on [9]. So
inhibitors specifically targeted endopeptidase of CatB will be
potentially more therapeutic. CatK, expressed in bone-
Cathepsins, also known as lysosomal proteases, are the
main members of the papain-like protease family, which are
implicated in many pathological situations and have been stu-
died intensely as potential drug targets [1].
Cathepsin B (CatB, EC 3.4.22.1) and Cathepsin K (CatK,
EC 3.4.22.38), two important members of cathepsins, are 30
and 27 kDa bilobal proteins, respectively. The active site and
substrate-binding cleft of CatB are located at the interface
between the two lobes with Cys29 on the left lobe and
His199 on the right lobe, while for CatK are Cys25 and
His162 residues. At optimal pH, a Cys29-S–/His199-ImH+
(for CatK is Cys25-S–/His162-ImH+) ion pair is formed by
the side chains of Cys29 (Cys25) and His199 (His162), sub-
strate peptide bond cleavage is mediated by the nucleophilic
attack by S- from Cys29 (Cys25) on the carbonyl carbon
Abbreviations: Abz-, ortho-aminobenzoic acid; AMC, [7-amino-4-methyl]
coumarin; CA030, EtO-(2S,3S)-tEps-Ile-Pro-OH; DMSO, dimethyl sulfoxide;
Dnp, 2,4-dinitrophenyl; E-64, 1-[[(L-trans-epoxysuccinyl)-L-leucyl] amino]-
4-guanidinobutane; EDTA, ethylenediaminetetraacetic acid; Leupeptin,
acetyl-leucyl-leucyl-arginal; Z-, carbobenzyloxy-.
* Corresponding author.
E-mail address: nhtan@mail.kib.ac.cn (N.-H. Tan).
0223-5234/$ - see front matter © 2006 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2006.06.002

1248
G.-Z. Zeng et al. / European Journal of Medicinal Chemistry 41 (2006) 1247–1252
resorbing osteoclasts selectively and abundantly, can degrade
type I collagen, which is the predominant component of the
extracellular matrix of bone. Thus, CatK plays a pivotal role in
osteoclast-mediated bone resorption and is a potential target in
osteoporosis, and some of CatK inhibitors efficiently block
bone resorption in vivo [4,10].
Many chemotypes have been described as inhibitors of
cathepsins, such as cystatins, leupeptin, E-64, CA030. Most
of inhibitors are peptide-based molecules. Though with potent
inhibition, they generally possess poor pharmacokinetic prop-
erties [11]. Therefore, searching for new inhibitors of cathe-
psins, especially selective reversible nonpeptidic inhibitors,
should be given increased emphasis [3]. For this purpose, we
performed the inhibition assays of CatB and CatK in vitro and
found six biflavones can inhibit CatB and CatK, which are a
novel class of cathepsin inhibitors from plants. This paper
describes the inhibition of these six nonpeptidic compounds
against CatB and CatK.
3. Pharmacology
Natural compounds AMF1-5 and HIF were evaluated for
their ability to inhibit CatB and CatK by fluorescence assays.
Moreover, the specificities of six biflavones against CatB
endopeptidase were assayed with different substrates and dif-
ferent assay buffers (Table 1). Enzyme inhibition data were
expressed as IC50 values (50% inhibitory concentration), and
IC50 values of each inhibitor were calculated by dose-
response curves with four concentrations (dilution ratio = 1/2)
and the highest tested concentration is 2.5 μg/ml. Results are
expressed as mean IC50 values standard deviation. Leupeptin
and E-64 were used as reference compounds.
4. Results and discussion
By inhibition assays of CatB and CatK, we found six bifla-
vones can potently inhibit cathepsin B and cathepsin K and
with some degree of selectivity for CatB (Table 1). In kinetic
studies of the hydrolysis conversion of Z-FR-AMC into AMC,
inhibition of AMF4 and HIF against CatB activity was found
to be time-dependent: Concentrations of AMC were less than
the standard level in the initial time, but as time went on, they
were equal to the standard level. Their inhibition were similar
to leupeptin recognized as a reversible inhibitor, but did not
like E-64, an irreversible inhibitor (Fig. 2). The reversible inhi-
bition of other four biflavones, AMF1-3 and AMF5, against
CatB were similar to that of AMF4 and HIF in our experi-
ments. It demonstrated that these biflavones were reversible
inhibitors of CatB.
The occluding loop of CatB is the major structural determi-
nant of the exopeptidase specificity and relatively lower endo-
peptidase activity [14]. We have used one endopeptidase sub-
strate Z-FR-AMC and two carboxyldipeptidase substrates Abz-
GIVRAK(Dnp) and Abz-FRF(4NO2)A in our experiment to
determine the specificity of these inhibitors against CatB, and
2. Chemistry
Natural biflavone compounds amentoflavone (AMF1), 4′″-
methylamentoflavone (podocarpusflavone A, AMF2), and
7′′,4′′′-dimethylamentoflavone (AMF3) were isolated and iden-
tified from acetone extracts of leaves and branches of Taxo-
dium mucronatum (Taxodiaceae) by chromatographic and
spectral methods [12]; 2,3-dihydroamentoflavone (AMF4), 4′-
methylamentoflavone (bilobetin, AMF5), and hinokiflavone
(HIF) were isolated and identified from methanol extracts of
leaves of Cycas guizhouensis (Cycadaceae) by chromato-
graphic and spectral methods [13] (Fig. 1). Their purities are
> 95%. Both Taxodium mucronatum and Cycas guizhouensis
were collected from Kunming Botany Garden (Kunming Insti-
tute of Botany, Chinese Academy of Sciences, China).
Detailed purifications and identifications of these six bifla-
vones were described before by us [12,13].
Fig. 1. Chemical structures of AFM1, 2, 3, 4, 5 and HIF.

G.-Z. Zeng et al. / European Journal of Medicinal Chemistry 41 (2006) 1247–1252
1249
Table 1
Inhibitory activities of inhibitors against CatB and CatK
Inhibitor
IC50 (μM)
Cathepsin B
Abz-GIVRAK(Dnp) Abz-FRF(4NO2)A
Cathepsin K
Z-GPR-AMC
Z-FR-AMC
pH 7.4
pH 5.5
AMF1
AMF2
AMF3
AMF4
AMF5
HIF
1.17 0.25
1.03 0.32
0.67 0.16
0.62 0.13
0.81 0.11
0.58 0.21
0.051 0.014
0.014 0.001
0.49 0.15
0.98 0.23
0.52 0.12
0.23 0.03
0.68 0.22
0.52 0.08
0.021 0.005
0.031 0.007
49.63 4.99
> 90.50
> 88.26
8.54 2.63
76.78 5.19
36.65 12.84
0.130 0.011
0.016 0.002
41.74 9.13
75.19 0.76
80.84 1.22
11.10 5.37
> 92.51
32.90 14.38
0.160 0.009
0.022 0.006
1.88 0.38
2.51 0.28
1.57 0.41
1.39 0.15
1.55 0.25
1.54 0.37
0.039 0.013
0.003 0.001
LP
E-64
Averages were calculated from at least three independent experimental data. “> μM” meant that 50% inhibitory concentration of the compound is beyond the
sample stock solution in our compound library.
Fig. 3. The pH dependence of CatB activity with the substrate Z-FR-AMC.
Fig. 2. Time-dependent inhibition and reversibility of biflavones. Relationship
between [AMC] and reaction time was monitored (Z-FR-AMC hydrolyzed by
CatB with buffer pH 5.5): Standard (without inhibitor), LP (0.13 μM), E-64
(0.016 μM), AMF4 (0.6 μM) and HIF (1.2 μM), respectively (Km is 0.504 mM
for Z-FR-AMC in the reaction system).
found that inhibitory activities of six biflavones against CatB
endopeptidase were about 14–132 times higher than their exo-
peptidase ones, which for leupeptin was only three times, and
for E-64 was almost the same (Table 1).
To investigate the pH preference of CatB against the peptide
Z-FR-AMC, we measured kcat/Km values under pseudo first-
order condition. The pH dependency is characterized by a gra-
dual increase in activity when pH is raised from 3.0 to 7.4, and
the kcat/Km value at pH 7.4 is much higher than that at pH 5.5,
which meant that, in weak neutral condition, CatB can hydro-
lyze the endopeptidase substrate Z-FR-AMC more efficiently,
Fig. 4. Docking results of the complexes between AMF1-5, HIF and CA030
ligands and human CatB. The binding conformations of CA030, AMF1,
AMF2, AMF3, AMF4, AMF5, and HIF are displayed in stick model
and it is with greater endopeptidase activity than exopeptidase
one. Though Z-FR-AMC is not a suitable substrate to study the
occluding loop function, the pH-dependent profile also showed
that there is a better affinity between CatB and Z-FR-AMC at
pH 7.4 than pH 5.5 (Fig. 3). We tested the inhibitory capacities
of six compounds against CatB with Z-FR-AMC in acidic
(pH 5.5) and weak neutral (pH 7.4) conditions, IC50 results
also indicated that under the weak neutral condition, the inhi-
bitory activities of six biflavones are slightly more potent than
those under acidic condition. While IC50 for E-64 is contrary,
IC50 at pH 7.4 is twice as many as it at pH 5.5 (Table 1). These
results correlated with that from different substrates. All results
representation with different colors for each compound: CA030 in red,
AMF1 in blue, AMF2 in gray, AMF3 in orange, AMF4 in tan, AMF5 in silver,
and HIF in green. The hydrophobic surface is displayed in a schematic
representation in pink. This was prepared using VMD.
indicated that these biflavone inhibitors were more specific to
CatB endopeptidase activity.
The binding conformations (Fig. 4) and binding energies
(Table 2) of six biflavones with CatB were performed and pre-
dicted by molecular docking (Fig. 5). Six biflavones potently
inhibited CatB with the order

1250
G.-Z. Zeng et al. / European Journal of Medicinal Chemistry 41 (2006) 1247–1252
Table 2
Docking energies (kJ/mol) of six biflavones
Ligand
AMF1
AMF2
AMF3
AMF4
AMF5
HIF
Total_score
G_score
–169.22
–154.28
–145.69
–158.67
–148.54
–144.35
PMF_score
–38.65
–31.21
–31.88
–0.29
D_score
–133.96
–149.67
–145.78
–143.31
–145.97
–151.20
Chem_score
–32.47
–30.74
–23.40
–24.30
–49.83
–50.11
–61.18
–58.54
–57.88
–63.14
–16.22
–17.67
–24.50
–28.90
Docking scoring values for complexes of the human cathepsin B receptor with optimal structure of six biflavones ligands. G_score, PMF_score, D_score and
Chem_score are based on hydrogen-bonding interaction, statistical ligand–receptor atom-pair interaction potentials, electrostatic and hydrophobic contributions to
binding energy, a diverse training set of 82 receptor-ligand complexes, respectively. Lower scores indicate more favorable binding.
HIF > AMF4 > AMF3 > AMF5 > AMF2 > AMF1 (Table 1),
which correlated well with the calculation results (Table 2).
Results indicated that there is a good correlation between the
binding energies and the experimental inhibitory potency pIC50
(–logIC50) with the r2 (correlation coefficient) value of 0.903
(Fig. 6). For AMF1-3 and AMF5, more methoxy groups the
molecule contained, more potent inhibitory activity it had. It
mainly attributed that methoxy group is electron donor which
can increase the electron density, HOMO energy and hydro-
phobic group. Moreover, AMF5 and AMF2, which have one
OCH3 substitute except in different positions, have different
inhibitory activities for the results of steric and electric effects
which we proposed previously [15]. However the structures of
HIF and AMF4 are different from the others. HIF, in which 4′
and 6″ substitutes were connected by an oxygen atom, was the
most potent inhibitor in these biflavones, because it occupied
almost the entire active site of CatB (Fig. 4), and formed many
hydrophobic contacts with the active site of CatB (Fig. 5).
AMF4, saturated at 2,3 positions, is more flexible, which
made it interact with CatB active site cleft well with more
potent activity. By flexible docking study, we found that six
biflavones acted on the active site cleft of CatB just like
CA030 did (Fig. 4), but their binding patterns and interaction
modes were different from CA030 analogs that mainly interact
with the residues His110, His111 and Cys29 [15]. Six bifla-
vones interacted little with the occluding loop, which might
be responsible for their specific inhibitory activities against
CatB endopeptidase. AMF4 is exceptional in six biflavones,
for its A-ring could interact with the occluding loop residues
(His111, Gly121, and Cys119) (Fig. 5), which made it more
potent on CatB exopeptidase activity (Table 1).
In conclusion, we found six biflavones potently inhibit
activities of CatB and CatK by random screening. They are
novel natural inhibitors for cathepsins and with some degree
of selectivity for the inhibition of CatB over CatK. They are
reversible inhibitors for CatB and could inhibit CatB endopep-
tidase activity specifically. The inhibitory specificities against
CatB endopeptidase made them may be more potential in
pathological state like tumor progression. Among these bifla-
vones, AMF4 and HIF are two inhibitors with the best activ-
ities against CatB both for endopeptidase and exopeptidase.
But their binding patterns and interaction modes differ from
CA030, their endopeptidase inhibitory activities are 14–60
times higher than their exopeptidase ones. Docking results
also supported these experimental results. Our findings could
benefit the development of new strategies for design and screen
of natural inhibitors of CatB.
5. Experimental protocols
5.1. Materials
Human liver cathepsin B and human recombinant cathepsin
K were from Calbiochem (Cat# 219364, 219461, Darmstadt,
Germany). The fluorogenic substrate Z-FR-AMC and Z-
GPR-AMC were from Bachem (Cat# I-1160, I-1150, King of
Prussia, Pennsylvania, USA). The fluorescent cleavage product
AMC and cysteine protease inhibitors E-64 and Leupeptin (LP)
were from Sigma (Cat# A9891, E3132, L9783, St. Louis, MO,
USA). The fluoregenic substrates Abz-GIVRAK(Dnp) and
Abz-FRF(4NO2)A were generously given by Dr. Luiz Juliano
(Department of Biophysics, Universidade Federal de Sao
Paulo, Brazil). All other reagents used were commercially
available with analytical grade. Substrate hydrolysis was mon-
itored in a cytofluor II fluorescent plate reader (PerSeptive Bio-
systems).
5.2. Inhibition assays of CatB
Assay of CatB endopeptidase activity was referenced by
what Barrett [16] and Melo et al. [17] did, which was deter-
mined spectrofluorometrically using the fluorogenic substrate
Z-FR-AMC. Two different buffers were used. One buffer
(pH 5.5) contained 60 mM sodium acetate, 15 mM L-cysteine
and 5 mM EDTA. The other buffer (pH 7.4) contained 60 mM
sodium phosphate, 15 mM L-cysteine and 5 mM EDTA. Test
compounds were dissolved in DMSO and diluted with buffers.
The compound solution was mixed with diluted enzyme solu-
tion in a well of 96-well black microplate (Costar) and incu-
bated for 5–10 min at 25 °C. The reaction was initiated by
adding 75 μM fluorescence substrate solution to the well.
After incubation for 45 min at 25 °C, in which produced fluor-
escence is sensitive enough to be detected, AMC released from
Z-FR-AMC was monitored on the Cytofluor II fluorescent
plate reader at 460/40 nm after excitation at 360/40 nm. The
final concentration of CatB in the assay mixture was 3.73 nM.
Assay of CatB exopeptidase activity was referenced by the
method Cotrin used [18] with some modifications: The buffer

G.-Z. Zeng et al. / European Journal of Medicinal Chemistry 41 (2006) 1247–1252
1251
Fig. 6. Plots of experimental data versus Total_score energy for six biflavone
inhibitors.
5.3. Kinetics of six biflavones against CatB
Experimental procedure was the same as assay of CatB
except that increasing AMC hydrolyzed from Z-FR-AMC
was monitored at 20–30 min interval in 700 min. The buffer
(pH 5.5) contained 60 mM sodium acetate, 15 mM L-cysteine
and 5 mM EDTA. Furthermore, by measuring initial rates at
various substrate concentrations, kinetic parameter Km was
determined based on the Michaelis–Menten equations.
5.4. pH dependence of CatB activity with the substrate Z-
FR-AMC
The pH profile of CatB activity was obtained by measuring
kcat/Km values on the relationship v = [E][S]kcat/Km, at
[S] << Km.kcat/Km values were determined by measuring initial
rates at various substrate concentrations, when the enzyme
final concentration was 3.73 nM. The reaction buffers were
60 mM sodium acetate (pH 3.0–5.8), 60 mM sodium phos-
phate (pH 5.8–8.0). Both contained 15 mM L-cysteine and
5 mM EDTA. Buffer pH ranged from 3.0 to 8.0 and was
adjusted with 1 M NaOH or HCl.
5.5. Inhibition assay of CatK
Compound inhibitory activity against CatK was measured
as the method Aibe et al. [19] and Barrett [16] used before
with some modifications. Test compounds were diluted with
the buffer (pH 5.0, 100 mM sodium acetate, 20 mM L-
cysteine, and 2 mM EDTA), and were mixed with diluted
enzyme solution for 5–10 min at 25 °C, then the 75 μM sub-
strate Z-GPR-AMC was added to start the reaction. The final
concentration of CatK in the assay mixture was 4.63 nM. After
incubation for 120 min at 37 °C, produced fluorescence was
monitored on the Cytofluor II fluorescent plate reader at
460/40 nm after excitation at 360/40 nm.
Fig. 5. Binding modes of the optional conformation poses of AMF4 and HIF
ligands with the active sites of CatB receptor were drawn using the Ligplot
program. Broken lines in green represent hydrogen bonds and spiked residues
form hydrophobic contacts with the inhibitors (connected by broken lines in
black).
(pH 5.0) contained 60 mM sodium acetate, 10 mM L-cysteine
and 2.5 mM EDTA, and the hydrolysis reaction was initiated
by adding 30 μM fluorescence substrate (Abz-GIVRAK(Dnp)
or Abz-FRF(4NO2)A) solution. After incubated for 45 min at
25 °C, produced fluorescence was monitored on the Cytofluor
II fluorescent plate reader at 460/40 nm after excitation at
360/40 nm.
5.6. Flexible docking
The method is the same as our previous research [15] with
some modifications. The 3D structures of six biflavones were

1252
G.-Z. Zeng et al. / European Journal of Medicinal Chemistry 41 (2006) 1247–1252
[3] G.Z. Zeng, N.H. Tan, R.R. Jia, X.L. Pan, Acta Bot. Yunnan 27 (2005)
337–354.
[4] X.L. Pan, N.H. Tan, G.Z. Zeng, H.J. Han, H.Q. Huang, Bioorg. Med.
Chem. 14 (2006) 2771–2778.
[5] M.E. McGrath, Annu. Rev. Biophys. Biomol. Struct. 28 (1999) 181–204.
[6] B.E. Linebaugh, M. Sameni, N.A. Day, B.F. Sloane, D. Keppler, Eur. J.
Biochem. 264 (1999) 100–109.
[7] L. Polgar, C. Csoma, J. Biol. Chem. 262 (1987) 14448–14453.
[8] B.E. Cathers, C. Barrett, J.T. Palmer, R.M. Rydzewski, Bioorg. Chem.
30 (2002) 264–275.
[9] J.E. Koblonski, M. Ahram, B.F. Sloane, Clin. Chim. Acta 291 (2000)
113–135.
[10] V. Turk, B. Turk, G. Guancar, D. Turk, J. Kos, Advan. Enzyme Regul.
42 (2002) 285–303.
[11] J.-P. Falgueyret, R.M. Oballa, O. Okamoto, G. Wesolowski, Y. Aubin,
R.M. Rydzewski, P. Prasit, D. Riendeau, S.B. Rodan, M.D. Percival, J.
Med. Chem. 44 (2001) 94–104.
[12] Y.M. Zhang, N.H. Tan, H.Q. Huang, R.R. Jia, G.Z. Zeng, C.J. Ji, Acta
Bot. Yunnan 27 (2005) 107–110.
constructed by using molecular modeling software package
Sybyl6.9 [20]. Partial atomic charges were calculated by Gas-
teiger–Huckel method, and energy minimizations were per-
formed using the Tripos force field with a distance dependent
dielectric and the powell conjugate gradient algorithm (conver-
gence criterion of 0.001 kcal/mol Å). The X-ray crystal struc-
ture of CatB complexed with CA030 (pdb entry: 1csb) was
retrieved from the Protein Data Bank (PDB) [21,22]. The
active site was defined as all the amino acid residues enclosed
within 6.5 Å radius sphere centered by the bound ligand,
CA030. The docking procedure was repeated 100 times for
each inhibitor and subsequent the Cscore program of Sybyl
was employed to score each conformation. The chosen docked
conformation for each inhibitor was selected from the docking
results on the basis of their FlexX_score (Total_score).
[13] J. Xiong, N.H. Tan, K. Gu, Chinese J. Org. Chem. 25 (Suppl.) (2005)
461.
Acknowledgments
[14] D.K. Nagler, A.C. Storer, F.C.V. Portaro, E. Carmona, L. Juliano, R.
Menard, Biochemistry 36 (1997) 12608–12615.
[15] X.L. Pan, N.H. Tan, G.Z. Zeng, Y.M. Zhang, R.R. Jia, Bioorg. Med.
Chem. 13 (2005) 5819–5825.
[16] A.J. Barrett, Biochem. J. 187 (1980) 909–912.
[17] R.L. Melo, L.C. Alves, E.D. Nery, L. Juliano, M.A. Juliano, Anal. Bio-
chem. 293 (2001) 71–77.
[18] S.S. Cotrin, L. Puzer, W.A.S. Judice, L. Juliano, A.K. Carmona, M.A.
Juliano, Anal. Biochem. 335 (2004) 244–252.
[19] K. Aibe, H. Yazawa, K. Abe, K. Teramura, M. Kumegama, H. Kawa-
shima, K. Honda, Biol. Pharm. Bull. 19 (1996) 1026–1031.
[20] Tripos Inc, SYBYL Molecular Modeling Software, Sybyl Version 6.9,
St. Louis; MO, 2001.
[21] D. Turk, M. Podobnik, T. Popovic, N. Katunuma, W. Bode, R. Huber,
V. Turk, Biochemistry 34 (1995) 4791–4797.
We thank Dr. Luiz Juliano (Department of Biophysics, Uni-
versidade Federal de Sao Paulo, Brazil) for generously giving
us substrates Abz-GIVRAK(Dnp) and Abz-FRF(4NO2)A of
Cathepsin B and his expert suggestion for our experiment.
The present work was supported by the Foundation of Chinese
Academy Sciences (West Light Program, KSCX1-09-03-1 and
KSCX-SW-11) and the National Natural Science Foundation
of China (30572258).
References
[1] B. Turk, D. Turk, V. Turk, Biochim. Biophys. Acta 1477 (2000) 98–111.
[2] J.S. Mort, D.J. Buttle, Int. J. Biochem. Cell Biol. 29 (1997) 715–720.
[22] A. Yamamoto, K. Tomoo, K.I. Matsugi, T. Hara, I. Yasuko, M. Murata,
K. Kitamura, T. Ishida, Biochim. Biophys. Acta 1597 (2002) 244–251.