Soluble artificial metalloproteases with broad substrate selectivity, high reactivity, and high thermal and chemical stabilities

Article abstract of DOI:10.1007/s00775-010-0662-x

To design soluble artificial proteases that cleave peptide backbones of a wide range of proteins with high reactivity, artificial active sites comprising the Cu(II) complex of 1-oxa-4,7,10-triazacyclodedecane (oxacyclen) and the aldehyde group were synthesized. The aldehyde group was employed as the binding site in view of its ability to reversibly form imine bonds with ammonium groups exposed on the surfaces of proteins, and Cu(II) oxacyclen was exploited as the catalytic group for peptide hydrolysis. The artificial metalloproteases synthesized in the present study cleaved all of the protein substrates examined (albumin, γ-globulin, myoglobin, and lysozyme). In addition, the activity of the best soluble artificial protease was enhanced by up to 190-fold in terms of k _cat/K _m. When the temperature was raised to 80 °C, the activities of the artificial proteases were significantly enhanced. The activity of the artificial protease was not greatly affected by surfactants, including sodium dodecyl sulfate. The intermediacy of the imine complex formed between the artificial protease and the protein substrate was supported by an experiment using sodium cyanoborohydride. Soluble artificial metalloproteases with broad substrate selectivity, high reactivity, high thermal and chemical stabilities, and small molecular weights were thus synthesized by positioning the aldehyde group in proximity to Cu(II) oxacyclen.

Full text of DOI:10.1007/s00775-010-0662-x

J Biol Inorg Chem (2010) 15:1023–1031
DOI 10.1007/s00775-010-0662-x
ORIGINAL PAPER
Soluble artificial metalloproteases with broad substrate selectivity,
high reactivity, and high thermal and chemical stabilities
•
Min Gyum Kim Sang Ho Yoo Woo Suk Chei
•
•
•
Tae Yeon Lee Hye Mi Kim Junghun Suh
•
Received: 3 February 2010 / Accepted: 12 April 2010 / Published online: 28 April 2010
Ó SBIC 2010
Abstract To design soluble artificial proteases that cleave
peptide backbones of a wide range of proteins with high
reactivity, artificial active sites comprising the Cu(II)
complex of 1-oxa-4,7,10-triazacyclodedecane (oxacyclen)
and the aldehyde group were synthesized. The aldehyde
group was employed as the binding site in view of its ability
to reversibly form imine bonds with ammonium groups
exposed on the surfaces of proteins, and Cu(II) oxacyclen
was exploited as the catalytic group for peptide hydrolysis.
The artificial metalloproteases synthesized in the present
study cleaved all of the protein substrates examined (albu-
min, c-globulin, myoglobin, and lysozyme). In addition, the
activity of the best soluble artificial protease was enhanced
by up to 190-fold in terms of kcat/Km. When the temperature
was raised to 80 °C, the activities of the artificial proteases
were significantly enhanced. The activity of the artificial
protease was not greatly affected by surfactants, including
sodium dodecyl sulfate. The intermediacy of the imine
complex formed between the artificial protease and the
protein substrate was supported by an experiment using
sodium cyanoborohydride. Soluble artificial metallopro-
teases with broad substrate selectivity, high reactivity, high
thermal and chemical stabilities, and small molecular
weights were thus synthesized by positioning the aldehyde
group in proximity to Cu(II) oxacyclen.
Keywords Proteolysis ꢀ Cu(II) oxacyclen ꢀ
Artificial metalloprotease ꢀ Peptide-cleaving catalyst ꢀ
Proteins
Abbreviations
Cyclen
MALDI
MS
1,4,7,10-Tetraazacyclododecane
Matrix-assisted laser desorption/ionization
Mass spectrometry
Oxacyclen 1-Oxa-4,7,10-triazacyclododecane
PAGE
SDS
Polyacrylamide gel electrophoresis
Sodium dodecyl sulfate
Time-of-flight
TOF
Introduction
Synthesizing enzyme-like catalysts for the hydrolysis of
peptide bonds in proteins is challenging in view of the high
stability [1, 2] of peptide bonds as well as the importance
of proteases in biology [3] and industry. At pH 7 and
25 °C, the half-life for spontaneous hydrolysis of a peptide
bond is about 500 years [1, 2]. The importance of proteases
in industry is demonstrated by the fact that proteases
comprise about 60% of all industrial enzyme sales globally
[4]. By using artificial proteases, it is possible to overcome
the intrinsic limitations of natural proteases, such as their
low thermal and chemical stabilities.
Electronic supplementary material The online version of this
article (doi:10.1007/s00775-010-0662-x) contains supplementary
material, which is available to authorized users.
The first step towards artificial proteases with high
reactivity and high thermal and chemical stabilities is to
obtain an effective catalytic center for peptide hydrolysis.
Intramolecular catalysis of the hydrolysis of unactivated
peptides by organic functional groups [5] or metal ions
[6] as well as by the cooperative action [7] of a metal ion
and an organic functional group has been intensively
M. G. Kim ꢀ S. H. Yoo ꢀ W. S. Chei ꢀ T. Y. Lee ꢀ
H. M. Kim ꢀ J. Suh (&)
Department of Chemistry,
Seoul National University,
Seoul 151-747, Korea
e-mail: jhsuh@snu.ac.kr
123

1024
J Biol Inorg Chem (2010) 15:1023–1031
investigated since the 1970s. Intermolecular catalysts with
protease activity, however, have been synthesized since the
1990s, using either metal complexes [8–16] or organic
functionalities [17–21] as the catalytic groups. Intramo-
lecular catalytic systems can be considered models of
enzyme–substrate complexes, whereas intermolecular ones
can be viewed as models of enzymes.
Although Cu(II)A and Cu(II)B achieved high proteo-
lytic rates for a variety of common proteins, they are not
applicable to processes which require homogeneous cata-
lysts. Most commercial proteases are used in water-soluble
form. To obtain effective homogeneous artificial metallo-
proteases, the intrinsic proteolytic activity of the metal ion
should be enhanced without anchoring the metal ion to an
insoluble support. In a previous study, we discovered that
the proteolytic activity of Cu(II) cyclen is considerable
enhanced by slightly modifying the chelating ligand. When
one NH of cyclen was replaced with O, to obtain the Cu(II)
complex of 1-oxa-4,7,10-triazacyclodedecane (oxacyclen),
the proteolytic activity of the Cu(II) center is increased
40–80-fold for various protein substrates in terms of
Metal complexes are more promising than organic func-
tional groups as the catalytic centers for artificial proteases
[22–24]. This is because a single metal complex can hydro-
lyze peptide bonds by playing various catalytic roles [25],
whereas cooperation between two or more catalytic groups is
needed when organic functional groups are exploited as the
catalytic centers. Early studies employed metal complexes in
combination with oxidoreductive additives to achieve either
hydrolytic [26] or oxidative [27–29] cleavage of the target
protein. Hydrolytic cleavage [30–37] of peptides or proteins
without the oxidoreductive additives requires highly effective
catalytic centers for the hydrolysis of peptide bonds. Several
kinds of metal complexes have been examined for their
ability to hydrolyze peptide bonds [7–16], but most of them
exhibited low catalytic rates unless they were tethered to
peptide substrates, especially under physiological conditions.
One way to increase the intrinsic proteolytic activity of a
metal center is to alter the polarity of its microenvironment.
For example, it was found that the proteolytic activity [13]
of the Cu(II) complex of 1,4,7,10-tetraazacyclododecane
(cyclen) is enhanced significantly when attached to cross-
linked polystyrene. Subsequently, artificial metallopro-
teases were designed by combining various auxiliary
functional groups with Cu(II) cyclen supported on cross-
linked polystyrene [32, 38]. The most effective artificial
metalloproteases based on Cu(II) cyclen reported to date
are the Cu(II) complexes (Cu(II)A and Cu(II)B) of the
polystyrene derivatives A and B [38]. Here, the aldehyde
group recognizes ammonium groups exposed on the sur-
faces of target proteins by forming imine bonds reversibly.
k
cat/Km [39].
In the present study, we designed effective homoge-
neous artificial metalloproteases by attaching the aldehyde
group acting as the substrate-binding site to the Cu(II)
complex of oxacyclen. In this article, the preparation of
soluble artificial metalloproteases and their proteolytic
features are described.
Materials and methods
Preparation of catalysts
The construction of the chemical library and the synthesis of
the Cu(II) complexes of C–I (Fig. 1) are described in the
‘‘Electronic supplementary material.’’ The Cu(II) complexes
of C–I showed similar visible spectra to that of Cu(II) oxa-
cyclen (kmax = 736 nm, emax° = 140 cm-1 M-1 at pH 9.5),
which was unaffected by incubation under the conditions
applied during kinetic measurements. The stability of the
Cu(II) complexes under the conditions used for the kinetic
measurements was also confirmed by matrix-assisted laser
desorption/ionization (MALDI) time-of-flight (TOF) mass
spectrometry (MS).
H
N
Measurements
NH
NH
N H
X
X
II
Cu
N
H
Distilled and deionized water was used for the preparation
of buffer solutions. The buffers (0.05 M) used in this study
were acetate (pH 5.0), 2-(N-morpholine)ethanesulfonate
(pH 6.0), N-2-hydroxyethylpiperazine-N0-ethansulfonate
(pH 7.0–8.0), and borate (pH 9.0–9.5). All buffer solutions
were filtered with 0.45 lm Millipore microfilter and
autoclaved before use in the kinetic measurements. pH
measurements were carried out with a Dongwoo Medical
DP-880 pH/ion meter. Bovine serum albumin, bovine
serum c-globulin, chicken egg white lysozyme, and horse
heart myoglobin were purchased from Sigma and used
without further purification. In kinetic measurements, the
NH
OH2
cyclen: X = NH
oxacyclen: X = O
Cu(II)cyclen: X = NH
Cu(II)oxacyclen: X = O
H
N
O
H
N
O
O
NH
H
H
N
PS
O
O
O
H
Br
PS
N
H
N
N
N
O
H
HN
N H
N
N
O
O
N
H
O
O
HN
HN
N
O
H
H
N
N
O
N
H
A
B
123

J Biol Inorg Chem (2010) 15:1023–1031
1025
temperature was controlled with a Vortemp manufactured
by Labnet. The degree of cleavage of proteins was mea-
sured by sodium dodecyl sulfate (SDS) polyacrylamide gel
electrophoresis (PAGE) with a Mighty Small II SE 250
model. The densities of the electrophoretic bands were
analyzed with an AlphaImager 2200 model and an Al-
phaEase model. MALDI–TOF MS analysis was performed
with sample solutions that were prepared as follows. After
incubation of the reaction solution (200 lL) containing the
catalyst and the substrate for 24 h, the mixture was evap-
orated to dryness with a centrifugal vacuum concentrator
(Hanil Model 3180C), and then 10 lL water were added to
the residue. A 5 lL aliquot of the mixture was desalted and
purified with a ZipTip C18 (P10, Millipore) using the
matrix solution as the eluent. MALDI–TOF MS measure-
ments were carried out with a Bruker Daltonics Autoflex II
LIFT–TOF/TOF mass spectrometer in the positive ion
reflectron mode. NMR spectra were recorded with a Bruker
DPX 300 MHz model.
attaching various substituents to 1,3,5-triazine. As sum-
marized in the ‘‘Electronic supplementary material,’’ two
derivatives of oxacyclen were used as R1-NH2 and 30
amines as R2-NH2, in addition to two aldehyde precursors
(Y-NH2).
The proteolytic activities of the members of the library
were examined by measuring the degree of cleavage of
myoglobin (1.0 9 10-5 M) after incubation for 24 h at
pH 9.5 and 50 °C. Among the 60 Cu(II) oxacyclen deriv-
atives containing the aliphatic aldehyde group, the Cu(II)
complex of C (Fig. 1) was found to have the highest pro-
teolytic activity. Among the 60 Cu(II) oxacyclen deriva-
tives containing the aromatic aldehyde group, the Cu(II)
complexes of D and E were chosen as the optimum
structures. F and G were synthesized as controls of C. H
and I were prepared as controls of D and E, respectively.
The proteolytic activities of the Cu(II) complexes of C–I
were examined using bovine serum albumin, bovine serum
c-globulin, chicken egg white lysozyme, and horse heart
myoglobin as the substrates. Myoglobin is oxidized to
metmyoglobin in the presence of oxygen. The myoglobin
purchased from a commercial source and used in the
present study was also in the met form, as confirmed by its
visible spectrum [40].
Results
To obtain effective soluble artificial metalloproteases, a
chemical library was constructed as summarized in
Scheme 1. Here, a group of Cu(II) complexes of oxacyclen
with alkyl or aryl aldehyde pendants were synthesized by
While the buffer solution containing a protein substrate
was incubated with a Cu(II) complex, the disappearance of
the protein was observed by SDS-PAGE [41, 42], as
Scheme 1 Synthetic route for a
chemical library of Cu(II)
complexes of oxacyclen
derivatives
R1
N
Cl
N
Cl
NH
N
N
N
N
N
N
Y NH2
DIE A
R1 NH2
DIE A
Y
Y
Cl
N
Cl
N
Cl
Cl
N
H
H
R1
NH
N
N
removal of CuCl2
R2
NH2
a Cu(II)
complex of
oxacyclen
Y
protecting
groups
HN
R2
N
N
DIE A
H
boc
O
boc
O
N
N
N
N
:
:
or
R1 NH2
N
N
boc
boc
NH2
NH2
OMe
H2N
H2N
or
Y NH2
O
OMe
O
123

1026
J Biol Inorg Chem (2010) 15:1023–1031
Fig. 1 Structures of the
derivatives of oxacyclen
synthesized to measure the
kinetics of their Cu(II)
complexes
C: n = 2, X = -CHO, Y = -CH(OH)-CH2OH, Z = -H
O
D: n = 2, X =
E: n = 3, X =
CHO , Y =
(CH )-CH CH , Z = -H
3 2 3
-CH
NH
HN
N
, Y = -CH(CH )-CH CH , Z = -H
CHO
3
2
3
(CH2)n
HN
H
N
F: n = 2, X = -CHO, Y = -CH2CH3, Z = -H
N
X
N
N
Z
G: n = 2, X = -CH3, Y = -CH(OH)-CH2OH, Z = -H
HN
Y
H: n = 2, X =
I: n = 3, X =
, Y = -CH(CH3)-CH2CH3, Z = -H
, Y = -CH(CH3)-CH2CH3, Z = -H
exemplified by the results summarized in Fig. 2. The data
in Fig. 2 indicate the disappearance of the substrate protein
and the formation of various fragments during the reaction.
The rate of cleavage of the substrate protein was measured
by monitoring [13, 17–21, 32, 38] the decrease in the
intensity of the electrophoretic band corresponding to the
substrate protein, as illustrated in Fig. 3. The pseudo-first-
order kinetic constant (k0) thus obtained is related to the
cleavage of the substrate protein molecule itself and does
not provide information on the further fragmentation of the
initial cleavage products. c-Globulin has two subunits with
distinctly different molecular weights (25 and 50 kDa).
Since the density of the electrophoretic band of the light
chain was weak, the kinetic data were collected only for the
degradation of the heavy chain.
The k0 values were measured for each substrate in the
presence of a Cu(II) complex at various pH values and at
fixed C0 (initially added catalyst concentration) and S0
(initially added substrate concentration) under the condi-
tions of C0 ꢁ S0 to identify the optimum pH (see Figs. S1–
S12 of the ‘‘Electronic supplementary material’’). Kinetic
data were obtained up to pH 9.5 to avoid complications due
to background alkaline degradation of proteins at higher
pH values. The pH dependence of k0 is controlled by the
ionization of Cu(II)-bound water (pKa = 8.7 for Cu(II)
oxacyclen [43]) and various functional groups of the pro-
tein substrate. At the optimum pH thus selected, the k0
values were measured at various C0 concentrations under
the conditions of C0 ꢁ S0. As shown by the dependence of
k0 on C0 (Fig. 4, Figs. S13–S24 of the ‘‘Electronic sup-
plementary material’’), saturation kinetic behavior was
observed for the cleavage of the protein substrates by the
Cu(II) complexes.
Kinetic data for the cleavage of the protein substrates
by the Cu(II) complexes of oxacyclen derivatives can be
analyzed in terms of the Michaelis–Menten scheme
Fig. 2 Results of SDS-PAGE performed on albumin (3.0 9 10-6 M)
incubated with Cu(II) C (5.0 9 10-4 M) at pH 9.5 and 50 °C
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0
0
1
2
3
4
5
6
Co, 10-4M
0.0
0.5
1.0
1.5
2.0
2.5
Time, hr
Fig. 4 Plot of k0 against C0 for the cleavage of albumin
(3.0 9 10-6 M) by Cu(II) C at pH 9.5 and 50 °C
Fig. 3 Plot of ln[S]/S0 against time for the data in Fig. 2
123

J Biol Inorg Chem (2010) 15:1023–1031
1027
Table 1 Values of kinetic
Catalyst
Substrate
Albumin
Temp (°C) pH kcat (h-1
)
Km (10-4 M)
k )
cat/Km (h-1 M-1
parameters for the cleavage of
various protein substrates by the
Cu(II) complexes of various
oxacyclen derivatives at the
optimum pH
Cu(II)C
50
9.5
9.5
9.5
9.5
9.5
9.5
9.5
9.5
9.5
9.5
9.5
9.5
9.5
9.5
7.0
9.5
9.5
9.5
9.5
9.5
9.5
9.5
7.0
1.4 0.2
1.4 0.1
1.1 0.3
0.81 0.17
0.77 0.17
1.5 0.2
1.9 0.6
2.8 1.0
2.8 0.7
5.2 1.1
2.5 0.6
1.5 0.4
1.2 0.2
1.7 0.3
0.88 0.15
0.53 0.13
4.6 0.8
1.4 0.4
3.1 0.4
1.8 0.2
2.4 1.0
5.3 1.6
5.4 0.4
1.6 0.3
13,000 3,000
17,000 3,000
9,800 1,900
3,800 500
8,900 1,700
18,000 3,000
34,000 4,000
2,000 400
2,800 600
6,400 1,400
5,900 900
4,100 700
8,900 1,400
13,000 3,000
920 140
c-Globulin 50
Lysozyme 50
Myoglobin 50
Myoglobin 60
Myoglobin 70
Myoglobin 80
0.75 0.07
0.58 0.05
1.7 0.3
5.2 0.9
9.4 1.2
Cu(II)D
Cu(II)E
Cu(II)H
Cu(II)I
Albumin
50
1.0 0.1
c-Globulin 50
Lysozyme 50
Myoglobin 50
0.70 0.07
0.97 0.13
0.73 0.04
0.68 0.06
0.78 0.05
0.68 0.07
0.43 0.03
0.15 0.02
0.18 0.01
0.32 0.02
0.21 0.06
0.17 0.02
0.23 0.01
0.13 0.009
0.20 0.03
Albumin
50
c-Globulin 50
Lysozyme 50
Myoglobin 50
Albumin
50
1,100 300
560 70
c-Globulin 50
Lysozyme 50
Myoglobin 50
1,800 200
880 270
Albumin
50
320 88
c-Globulin 50
Lysozyme 50
Myoglobin 50
420 30
810 160
11
3
180 30
Cu(II) oxacyclena Albumin
50
9.0 0.069 0.01
8.0 0.11 0.01
9.5 0.093 0.012
5.1 0.3
130 10
c-Globulin 50
12
13
3
4
94
2
Myoglobin 50
70 11
a
Taken from [39]
(Eq. 1). Under the conditions of C0 ꢁ [CS], pseudo-first-
order kinetic behavior is expected, with k0 being derived
as in Eq. 2. The saturation kinetic behavior (Fig. 4,
Figs. S13–S24 of the ‘‘Electronic supplementary mate-
rial’’) observed for the plot of k0 against C0 was analyzed
with a nonlinear fitting program according to Eq. 2. The
dependence of k0 on C0 was analyzed to estimate kcat, kcat
/
Km, and their standard deviations, from which Km and its
standard deviation were calculated. The values of kcat, Km,
and kcat/Km thus estimated for the cleavage of the protein
substrates by the Cu(II) complexes of various oxacyclen
derivatives are summarized in Table 1.
k1
k2
C þ S ꢀ CSꢂ!C þ P;
ð1Þ
k
ꢂ1
Fig. 5 MALDI–TOF mass spectrum of the product solution obtained
after incubation of myoglobin (1.0 9 10-6 M) with Cu(II)C
(1.0 9 10-5 M) at pH 9.5 and 50 °C for 24 h
k0 ¼ kcatC0=ðKm þ C0Þ ¼ kcatC0=½kcat=ðkcat=KmÞ þ C0ꢃ;
ð2Þ
ð3Þ
ð4Þ
kcat ¼ k2;
Km ¼ ðkꢂ1 þ k2Þ=k1:
illustrated in Fig. 5. The structures of fragments with m/z
of 2,212 and 3,663 in Fig. 5 were determined as Leu(135)-
Gly(153) and Gly(1)-Phe(33), respectively, by MALDI
LIFT TOF–TOF MS [44], as described in the ‘‘Electronic
supplementary material.’’ Fragment Leu(135)-Gly(153) is
A typical MALDI–TOF mass spectrum of the product
solutions obtained by cleavage of the protein substrates by
the Cu(II) complexes of the oxacyclen derivatives is
123

1028
J Biol Inorg Chem (2010) 15:1023–1031
the C-terminal portion obtained by the hydrolytic cleavage
of myoglobin between Ala(134) and Leu(135), and frag-
ment Gly(1)-Phe(33) is the N-terminal portion obtained by
the hydrolytic cleavage between Phe(33) and Thr(34).
To compare the thermal stabilities of the Cu(II) com-
plexes of oxacyclen derivatives with thermostable natural
proteases, the activity of Cu(II)C and that of the protease
contained in a commercial household laundry detergent
were measured at various temperatures. To obtain infor-
mation on the effects of detergent surfactants and other
additives on the Cu(II) complexes, the activity of Cu(II)C
was also measured in the presence of the detergent at
various temperatures. The results are summarized in Fig. 6.
To obtain further information on the chemical stabilities
of the Cu(II) complexes of oxacyclen derivatives, the effect
of SDS was examined. The activity of Cu(II)C (2.0 9
10-4 M) for the cleavage of myoglobin at pH 9.5 and 50 °C
was reduced by 40% upon the addition of 1.0% SDS.
In order to gain insights into the mechanism of the pro-
teolytic action of the metal complexes, the rate of cleavage
of myoglobin (1.0 9 10-5 M) by Cu(II)C (2.0 9 10-4 M)
was examined in the presence of sodium cyanoborohydride.
At pH 6.0 and 50 °C, k0 was 1.0 9 10-2 h-1 in the absence
of sodium cyanoborohydride, and it was 2.4 h-1 in the
presence of 5.0 mM sodium cyanoborohydride.
Discussion
It is well established that the Cu(II) complexes of various
aza ligands catalyze the cleavage of peptide bonds by
hydrolysis [30–32, 38, 45, 46]. The results obtained upon
the structural determination of product fragments by
MALDI LIFT TOF–TOF MS reveal that the cleavage of
peptide bonds of myoglobin by Cu(II)C produces carboxyl
and amino groups. This is consistent with the hydrolytic
cleavage of protein substrates by the Cu(II) complex of the
oxacyclen derivative.
Screening of the chemical library obtained by the route
summarized in Scheme 1 indicated that Cu(II)C, Cu(II)D,
and Cu(II)E are effective homogeneous proteolytic rea-
gents. As summarized in Table 1, Cu(II)C, Cu(II)D, and
Cu(II)E cleaved all four of the proteins examined in the
present study, demonstrating the broad substrate selectivity
of the homogeneous artificial proteases.
The proteolytic activity of Cu(II)C toward albumin
measured at the optimum pH is compared with those of
Cu(II)A, Cu(II) cyclen, and Cu(II) oxacyclen in Fig. 7. As
reported previously [13], the proteolytic activity of Cu(II)
cyclen is enhanced greatly upon its attachment to the sur-
face of crosslinked polystyrene, presumably due to the
hydrophobic microdomains created on the polymer. Intro-
ducing an aldehyde group in proximity to the Cu(II) cyclen
unit positioned on the polystyrene surface to obtain
Cu(II)A further increased the proteolytic activity signifi-
cantly. The high proteolytic activity of Cu(II)A (a hetero-
geneous artificial metalloprotease), as illustrated in Fig. 7,
is due to both its attachment to the polystyrene surface and
the introduction of the aldehyde group.
6
5
4
3
2
1
0
3.0
2.5
2.0
1.5
1.0
0.5
0.0
50
60
70
80
Temp,oC
0
10
20
30
Fig. 6 Values of k0 measured for the cleavage of myoglobin
(1.0 9 10-5 M) by Cu(II)C (2.0 9 10-4 M) (white bars), by Cu(II)C
(2.0 9 10-4 M) added together with a commercial household laundry
detergent [0.084% (w/v)] (gray bars), and by a commercial household
laundry detergent [0.084% (w/v)] (black bars) at pH 9.5 and various
temperatures. The concentration of the commercial household laundry
detergent agreed with the recommended dosage
Co, 10-4M
Fig. 7 Plot of k0 against C0 for the cleavage of albumin by Cu(II)
cyclen (inverted open triangles, pH 9.0; from [38]), Cu(II) oxacyclen
(inverted filled triangles, pH 9.0; from [39]), Cu(II)A (empty circles,
pH 9.5; from [38]), and Cu(II)C (filled circles, pH 9.5) at 50 °C
123

J Biol Inorg Chem (2010) 15:1023–1031
1029
0.6
0.5
0.4
0.3
0.2
0.1
0.0
the aldehyde groups of the artificial proteases and the
amino groups of the protein substrate, as proposed [38]
previously for the action of Cu(II)A and Cu(II)B. Essen-
tially all globular proteins possess lysine residues which
provide ammonium cations on the protein surfaces. Imine
formation between the lysine-amino group of the protein
substrate and the aldehyde group of the artificial protease in
the noncovalent intermediate (CS) formed between the
catalyst and the substrate would lead to the covalent
intermediate (CS0). Since imine groups are readily hydro-
lyzed, the formation of CS0 from CS is reversible. The
effective molarity [5] of the Cu(II) oxacyclen moiety
toward the scissile peptide group should be greater in CS0
than in CS.
0
2
4
6
8
10
12
Co, 10-4
M
Fig. 8 Plot of k0 against C0 for the cleavage of myoglobin by Cu(II)C
(filled circles), Cu(II)F (empty circles), Cu(II)G (inverted filled
triangles), and Cu(II) oxacyclen (inverted open triangles; [39]) at
pH 9.5 and 50 °C
The proteolytic activity of Cu(II)C, a homogeneous
artificial metalloprotease, is comparable to that of Cu(II)A,
although Cu(II)C does not obtain a rate enhancement due
to its attachment to the polystyrene surface. The activity of
Cu(II)A was attained by using a macromolecular back-
bone, whereas the activity of Cu(II)C was achieved by
using a small molecule (molecular weight of about 500).
The high activity of Cu(II)C originates from two factors:
the proteolytic activity of Cu(II) oxacyclen is consider-
ably higher than that of Cu(II) cyclen, and an aldehyde
group is positioned in proximity to Cu(II) oxacyclen in
Cu(II)C.
The Michaelis–Menten scheme of Eq. 1 can be modified
as Eq. 5 to include CS0. The same saturation kinetic
behavior (Eqs. 2, 6) is predicted for the dependence of k0
on C0 by both Eqs. 1 and 5. By applying the steady state
approximation to CS0, Eqs. 7 and 8 can be derived.
To evaluate the effects of organic functional groups
present in the organic pendent attached to the oxacyclen
moiety in Cu(II)C, rate data for the cleavage of myoglobin
by Cu(II)C, Cu(II)F, and Cu(II)G are compared in Fig. 8.
The results reveal that the aldehyde group contributes
significantly to the high activity of Cu(II)C, and the
hydroxyl groups exert minor effects.
k1
k2
k3
C þ S ꢀ CS ꢀ CSꢂ!C þ P1 þ P2;
ð5Þ
ð6Þ
ð7Þ
ð8Þ
k
k
ꢂ2
ꢂ1
ꢀÀ
Á
k0 ¼ kcaaptpC0 Kmapp þ C0 ;
app
cat
k
¼ k2k3=ðk2 þ kꢂ2 þ k3Þ\k3;
The artificial metalloproteases synthesized here cleaved
all four of the protein substrates with considerably
enhanced rates compared with Cu(II) oxacyclen. Attach-
ment of the aldehyde-containing organic pendants to
Cu(II) oxacyclen resulted in up to 100-, 180-, and 190-fold
increases in kcat/Km for albumin (Cu(II)C), c-globulin
(Cu(II)C), and lysozyme (Cu(II)E), respectively. At 50 °C,
Km was (0.5–5) 9 10-4 M, comparable to those [47] of
natural proteases, and kcat was (0.4–1.4) h-1, correspond-
ing to half-lives of 0.5–1.7 h for the proteolytic action of
Cu(II) complexes of C–E. The kcat values for the artificial
proteases can be considerably enhanced by raising the
temperature, in sharp contrast to those of natural proteases.
The high proteolytic activities of Cu(II)C, Cu(II)D, and
Cu(II)E can be attributed to the imine formation between
Kmapp ¼ ðk2k3 þ kꢂ1k3 þ kꢂ1kꢂ2Þ=k1ðk2 þ kꢂ2 þ k3Þ:
app
The parameter kcat (or k for the scheme of Eq. 5) is
cat
the maximum value of k0 attained at high C0 concentrations
(C0 ꢁ Km), whereas kcat/Km (or kacaptp/Kmapp for the scheme of
Eq. 5) represents the second-order rate constant (k0/C0)
observed at low C0 concentrations (C0 ꢄ Km). Both kcat
and kcat/Km, therefore, characterize the catalytic efficiency.
On the other hand, 1/Km (or 1/Kampp for the scheme of Eq. 5)
is related to the concentration of CS (or CS0 for the scheme
of Eq. 5) in the steady state.
The greater effective molarity of the Cu(II) center in CS0
would raise the experimentally observed value of kcat. If the
concentration of CS0 is smaller than CS in the steady state,
however, the efficiency of the catalytic action is reduced.
123

1030
J Biol Inorg Chem (2010) 15:1023–1031
To enhance the catalytic efficiency, therefore, it is most
desirable to both increase the effective molarity of the
Cu(II) center in the covalent imine complex and raise the
concentration of the covalent imine complex in the steady
state. By optimizing the structure of the spacer group
connecting the Cu(II) oxacyclen unit and the aldehyde
group, the efficiency of the artificial protease would be
optimized.
40%, which can be ascribed to the unfolded conformation
of myoglobin caused by denaturation.
Artificial proteases are also prepared by the structural
modification of proteins such as immunoglobulin or natural
proteases [50–52]. Biotic artificial proteases based on
protein backbones, however, inevitably have limitations
due to their low thermal and chemical stabilities.
The homogeneous artificial metalloproteases synthe-
sized in the present study are characterized by broad sub-
strate selectivity, high catalytic activity, high thermal
stability, high chemical stability, and small molecular
weights. The activity of the best soluble artificial protease
was enhanced 190-fold in the present study. Further opti-
mization of the structures of the organic pendants con-
nected to Cu(II) oxacyclen may produce soluble artificial
metalloproteases that can replace the natural proteases used
for various industrial applications.
For the proteolytic action of Cu(II) cyclen, k0 was lin-
early related to C0 (up to 7 mM), and the linearity constant
can be taken as kcat/Km (C0 ꢄ Km in Eq. 2) [39]. On the
other hand, Cu(II) oxacyclen manifested saturation kinetic
behavior [39], indicating that the substitution of a nitrogen
atom of cyclen for oxygen facilitated the complexation of
the Cu(II) complex to a protein substrate. Comparison of
the values of various kinetic parameters summarized in
Table 1 reveals that the high proteolytic activities of
Cu(II)C, Cu(II)D, and Cu(II)E involve both the increased
effective molarity of the Cu(II) center in the covalent imine
complex (increased kcat) and the increased concentration of
the covalent imine complex in the steady state (reduced
Km) in most cases.
Acknowledgments This work was supported by
a National
Research Foundation of Korea (NRF) grant funded by the Korean
government (MEST) (No. 2009-0072151). This paper is dedicated to
the memory of the late Prof. Chi Sun Hahn.
The intermediacy of the imine complex is further sup-
ported by the experiment using sodium cyanoborohydride,
which selectively reduces [48] imines to amines. The rate
for the disappearance of myoglobin in the presence of
Cu(II)C was raised by 240 times upon the addition of
sodium cyanoborohydride under the conditions described
in the ‘‘Results’’ section. The enhanced rate of disappear-
ance of the protein substrate is consistent with the trapping
of the imine intermediate with the borohydride reagent.
The reduction of the imine group in CS0 would block
conversion to CS and lead to an increased rate of protein
cleavage by the Cu(II) oxacyclen moiety covalently linked
to the protein.
References
1. Radzicka A, Wolfenden R (1996) J Am Chem Soc 118:6105–
6109
2. Bryant RAR, Hansen DA (1998) J Am Chem Soc 120:8910–8913
3. Overall CM, Blobel CP (2007) Nat Rev Mol Cell Biol 8:245–257
4. Rao MB, Tanksale AP, Ghatge MS, Deshpande VV (1998)
Microbiol Mol Biol Rev 62:597–635
5. Kirby AJ (1980) Adv Phys Org Chem 17:183–278
6. Sutton PA, Buckingham DA (1987) Acc Chem Res 20:357–364
7. Suh J, Park TH, Hwang BK (1992) J Am Chem Soc 114:5141–
5146
8. Chin J (1991) Acc Chem Res 24:145–152
9. Suh J (2003) Acc Chem Res 36:562–570
To demonstrate the thermal stability of Cu(II)C, the rate
data for the cleavage of myoglobin were collected at
50–80 °C. As summarized in Table 1, the proteolytic
activity of Cu(II)C is significantly enhanced at higher
temperatures. Household laundry detergents contain ther-
mostable natural proteases. As revealed by the data of Fig. 6,
Cu(II)C and the protease in the detergent manifest similar
activities at 50 °C. As the temperature is raised, the natural
protease is inactivated due to its lower thermal stability.
To demonstrate the chemical stability of Cu(II)C, the
rate of cleavage of myoglobin was examined in the pres-
ence of 1% (35 mM) SDS as well as the commercial
detergent. The results summarized in Fig. 6 reveal that the
activity of Cu(II)C is not affected by surfactants and other
additives present in the detergent, even at high tempera-
tures. In the presence of 1% SDS, proteins are readily
denatured [49] and natural proteases should be inactivated.
The SDS lowered the proteolytic activity of Cu(II)C by
10. Chin J, Jubian V, Mrejen K (1990) J Chem Soc Chem Commun
1326–1328
11. Zhu L, Qin L, Parac TN, Kostic NM (1994) J Am Chem Soc
116:5218–5224
12. Hegg EL, Burstyn JN (1995) J Am Chem Soc 117:7015–7016
13. Jang BB, Lee KP, Min DH, Suh J (1998) J Am Chem Soc
120:12008–12016
14. Kaminskaia NV, Johnson TW, Kostic NM (1999) J Am Chem
Soc 121:8663–8664
15. Saha MK, Bernal I (2003) J Chem Soc Chem Commun 612–613
16. Kasai M, Ravi RG, Shealy SJ, Grant KB (2004) Inorg Chem
43:6130–6132
17. Suh J, Hah SS (1980) J Am Chem Soc 120:10088–10093
18. Suh J, Oh SJ (2000) Org Chem 65:7534–7540
19. Oh S, Chang W, Suh J (2001) Bioorg Med Chem Lett 11:1469–
1472
20. Kim H, Paik H, Kim M, Chung YS, Suh J (2002) Bioorg Med
Chem Lett 12:2557–2560
21. Kim H, Kim M, Paik H, Chung Y-S, Hong IS, Suh J (2002)
Bioorg Med Chem Lett 12:3247–3250
22. Chei WS, Suh J (2007) In: Karlin KD (ed) Progress in inorganic
chemistry, vol 55. Wiley, Hoboken, pp 79–142
123

J Biol Inorg Chem (2010) 15:1023–1031
1031
23. Suh J, Chei WS (2008) Curr Opin Chem Biol 12:207–213
24. Lee TY, Suh J (2009) Chem Soc Rev 38:1949–1957
25. Suh J (1992) Acc Chem Res 25:273–279
26. Rana TM, Meares CF (1991) Proc Natl Acad Sci USA 88:10578–
10582
27. Shepartz A, Cuenoud B (1990) J Am Chem Soc 112:3247–3249
28. Hoyer D, Cho H, Schultz PG (1990) J Am Chem Soc 112:3250–
32449
29. Gallagher J, Zelenko O, Walts AD, Sigman DS (1998) Bio-
chemistry 37:2096–2104
30. Suh J, Moon S-J (2001) Inorg Chem 40:4890–4895
31. Moon SJ, Jeon JW, Kim H, Suh MP, Suh J (2000) J Am Chem
Soc 122:7742–7749
39. Jang SW, Suh J (2008) Org Lett 10:481–484
40. Bodwell CE, McClain PE (1971) In: Price JF, Schweigert BS
(eds) The science of meat and meat products, 2nd edn. WH
Freeman, San Francisco, p 97
41. Lameli UK (1970) Nature 227:680–685
42. Hames BD (1990) Chapter 1. In: Hames, BD, Rickwood D (eds)
Gel electrophoresis of proteins. IRL, New York
´
43. Amorim MTS, Chaves S, Delgado R, Frausto da Silva JJR (1991)
J Chem Soc Dalton Trans 3065–3072
44. Suckau D, Resemann A, Schuerenberg M, Hufnagel P, Franzen J,
Holl A (2003) Anal Bioanal Chem 376:952–965
45. Jeon JW, Son SJ, Yoo CE, Hong IS, Song JB, Suh J (2002) Org
Lett 4:4155–4158
32. Yoo CE, Chae PS, Kim JE, Jeong EJ, Suh J (2003) J Am Chem
Soc 125:14580–14589
46. Jeon JW, Son SJ, Yoo CE, Hong IS, Suh J (2003) Bioorg Med
Chem 11:2901–2910
´
33. Milovic NM, Badjic JD, Kostic NM (2004) J Am Chem Soc
126:696–697
47. Polgar L (1989) Mechanisms of protease action. CRC, Boca
Raton
34. Yang G, Miao R, Li Y, Hong J, Zhao C, Guo Z, Zhu L (2005)
Dalton Trans 1613–1619
48. Borch RF, Bernstein MD, Durst HD (1971) J Am Chem Soc
93:2897–2904
´
49. Michaux C, Pomroy NC, Prive GG (2008) J Mol Biol 375:1477–
1488
35. Jitsukawa K, Mabuchi T, Einaga H, Masuda H (2006) Eur J Inorg
Chem 21:4254–4263
36. Yashiro M, Kawakami Y, Taya J, Arai S, Fujii Y (2009) Chem
Commun 1544–1546
50. Fujie T (2002) Chem Rev 102:4885–4906
51. Taguchi H, Planque S, Sapparapu G, Boivin S, Hara M,
Nishiyama Y, Paul S (2008) J Biol Chem 283:36724–36733
52. Varadarajan N, Rodriguez S, Hwang B-Y, Georgiou G, Iverson
BL (2008) Nat Chem Biol 4:290–294
´
´
´
37. Rajkovic S, Glisic BD, Zivkovic MD, Djuran MI (2009) Bioorg
Chem 37:173–179
38. Yoo SH, Lee BJ, Kim H, Suh J (2005) J Am Chem Soc
127:9593–9602
123