Aryl thiol substrate 3-carboxy-4-nitrobenzenethiol strongly stimulating thiol peroxidase activity of glutathione peroxidase mimic 2, 2′-ditellurobis(2-deoxy-β-cyclodextrin)

Article abstract of DOI:10.1021/ja045964v

Artificial glutathione peroxidase (GPx) model 2, 2′-ditellurobis(2- deoxy-β-cyclodextrin) (2-TeCD) which has the desirable properties exhibited high substrate specificity and remarkably catalytic efficiency when 3-carboxy-4-nitrobenzenethiol (ArSH) was used as a preferential thiol substrate. The complexation of ArSH with β-cyclodextrin was investigated through UV spectral titrations, fluorescence spectroscopy, ¹H NMR and molecular simulation, and these results indicated that ArSH fits well to the size of the cavity of β-cyclodextrin. Furthermore, 2-TeCD was found to catalyze the reduction of cumene peroxide (CuOOH) by ArSH 200 000-fold more efficiently than diphenyl diselenide (PhSeSePh). Its steady-state kinetics was studied and the second rate constant k_max/K_ArSH was found to be 1.05 × 10⁷ M^-1 min^-1 and similar to that of natural GPx. Moreover, the kinetic data revealed that the catalytic efficiency of 2-TeCD depended strongly upon the competitive recognition of both substrates for 2-TeCD. The catalytic mechanism of 2-TeCD catalysis agreed well with a ping-pong mechanism, in analogy with natural GPx, and might exert its thiol peroxidase activity via tellurol, tellurenic acid, and tellurosulfide.

Full text of DOI:10.1021/ja045964v

Published on Web 11/24/2004
Aryl Thiol Substrate 3-Carboxy-4-Nitrobenzenethiol Strongly
Stimulating Thiol Peroxidase Activity of Glutathione
Peroxidase Mimic 2, 2’-Ditellurobis(2-Deoxy-â-Cyclodextrin)
Zeyuan Dong, Junqiu Liu,* Shizhong Mao, Xin Huang, Bing Yang, Xiaojun Ren,
Guimin Luo, and Jiacong Shen
Key Laboratory for Supramolecular Structure and Materials of Ministry of Education,
Jilin UniVersity, Changchun 130012, Peoples Republic of China
Received July 7, 2004; E-mail: jqliu2000@yahoo.com
Abstract: Artificial glutathione peroxidase (GPx) model 2, 2’-ditellurobis(2-deoxy-â-cyclodextrin) (2-TeCD)
which has the desirable properties exhibited high substrate specificity and remarkably catalytic efficiency
when 3-carboxy-4-nitrobenzenethiol (ArSH) was used as a preferential thiol substrate. The complexation
1
of ArSH with â-cyclodextrin was investigated through UV spectral titrations, fluorescence spectroscopy, H
NMR and molecular simulation, and these results indicated that ArSH fits well to the size of the cavity of
â-cyclodextrin. Furthermore, 2-TeCD was found to catalyze the reduction of cumene peroxide (CuOOH)
by ArSH 200 000-fold more efficiently than diphenyl diselenide (PhSeSePh). Its steady-state kinetics was
7
-1
-1
studied and the second rate constant kmax/KArSH was found to be 1.05 × 10 M min and similar to that
of natural GPx. Moreover, the kinetic data revealed that the catalytic efficiency of 2-TeCD depended strongly
upon the competitive recognition of both substrates for 2-TeCD. The catalytic mechanism of 2-TeCD catalysis
agreed well with a ping-pong mechanism, in analogy with natural GPx, and might exert its thiol peroxidase
activity via tellurol, tellurenic acid, and tellurosulfide.
Introduction
and hydrophobic amino acid residues (Phe, Trp, Asp) form a
3
d
hydrophobic cavity for thiol GSH binding. The selenium
undergoes a redox cycle involving the selenol (ESeH) as the
active form. The selenol is first oxidized to selenenic acid
It was well established that a variety of human diseases have
been generated by oxidative stress of reactive oxygen species
(ROS). ROS were generated as byproducts of cellular metabo-
(ESeOH), which reacts with reduced glutathione (GSH, 1) to
lism and were mainly controlled by antioxidative defense
system, especially by antioxidant enzymes.1 In biological
organism the antioxidant enzymes, superoxide dismutase (SOD),
catalase (CAT), and glutathione peroxidase (GPx) contribute
dominatingly to enhance cellular antioxidative defense against
form selenenylsulfide (ESeSG). A second glutathione then
regenerates the active form of the enzyme by attacking the
ESeSG to form the oxidized glutathione (GSSG) (Scheme 1).4
In contrast to the cytosolic GPx, which uses GSH exclusively
as cosubstrate, other enzymes such as plasma GPx or phospho-
lipid hydroperoxide GPx readily accept many thiols as sub-
2
oxidative stress. GPx (EC 1.11.1.9) is a well-known selenoen-
zyme which catalyzes the reduction of harmful hydrogen
peroxides and organic peroxides by glutathione (GSH, 1) and
protects the biomembranes and other cellular components from
oxidative damage. The enzyme active site includes a seleno-
cysteine residue which forms a catalytic triad with Trp148 and
Gln70 in a depression on the protein’s surface, and some charged
5
strates.
In recent years, there were increasing interests in mimicking
the functions of this important antioxidant enzyme not only for
elucidating catalytic mechanism but also for potential pharma-
ceutical application, and several attempts had been made to
produce synthetic selenium/tellurium compounds which mimic
3
4
,6
the properties of glutathione peroxidase. Ebselen (2-phenyl-
1,2-benzoisoselenazol-3(2H)-one), ebselen homologues, sele-
nenamides, diselenides, R-phenylselenoketones, selenium-
containing enzymes, antibodies, and cyclodextrins and their
(
1) (a) Spector, A. FASEB J. 1995, 9, 1173. (b) Salvemini, D.; Wang, Z. Q.;
Zweier, J. L.; Samouilov, A.; Macarthur, H.; Misko, T. P.; Currie, M. G.;
Cuzzocrea, S.; Sikorski, J. A.; Riley, D. P. Science 1999, 286, 304.
2) (a) Fridovich, I. J. Biol. Chem. 1997, 272, 18515. (b) Berlett, B. S.;
Stadtman, E. R. J. Biol. Chem. 1997, 272, 20313. (c) Beckman, K. B.;
Ames, B. N. J. Biol. Chem. 1997, 272, 19095. (d) Steinberg, D. J. Biol.
Chem. 1997, 272, 20963. (e) Floh e´ , L. Curr. Top. Cell Regul. 1985, 27,
(
4
73.
(4) Mugesh, G.; du Mont, W. W.; Sies, H.; Chem. ReV. 2001, 101, 2125.
(5) Takebe, G.; Yarimizu, J.; Saito, Y.; Hayashi, T.; Nakamura, H.; Yodoi, J.;
Nagasawa, S.; Takahashi K. J. Bio. Chem. 2002, 277, 43, 41254.
(6) (a) Mugesh, G.; Singh, H. B. Acc. Chem. Res. 2002, 35, 226. (b) Mugesh,
G.; Singh, H. B. Chem. Soc. ReV. 2000, 29, 347. (c) Mugesh, G.; du Mont,
W. W. Chem. Eur. J. 2001, 7, 1365. (d) Luo, G.; Ren, X.; Liu, J.; Mu, Y.;
Shen, J. Curr. Med. Chem. 2003, 10, 1151. (e) Back, T. G.; Moussa, Z. J.
Am. Chem. Soc. 2003, 125, 13455. (f) You, Y.; Ahsan, K.; Detty, M. R. J.
Am. Chem. Soc. 2003, 125, 4918. (g) McNaughton, M.; Engman, L.;
Birmingham, A.; Powis, G.; Cotgreave, I. A. J. Med. Chem. 2004, 47, 233.
(
3) (a) Floh e´ , L.; Loschen, G.; G u¨ nzler, W. A.; Eichele, E. Hoppe-Seyler’s Z.
Physiol. Chem. 1972, 353, 987. (b) Selenium in Biology and Human Health;
Burk, R. F., Ed.; Springer-Verlag: New York, 1994. (c) Tappel, A. L.
Curr. Top. Cell. Regul. 1984, 24, 87. (d) Epp, O.; Ladenstein, R.; Wendel,
A. Eur. J. Biochem. 1983, 133, 51. (e) Reich, H. J.; Jasperse, C. P. J. Am.
Chem. Soc. 1987, 109, 5549. (f) Wendel, A. In The Enzymatic Basis of
Detoxification; Jakoby, W. B., Ed.; Academic Press: New York, 1980;
Vol. I, p 333. (g) Ganther, H. E.; Kraus, R. J. Methods Enzymol. 1984,
1
07, 593. (h) Ladenstein, R. Pept. Protein ReV. 1984, 4, 173.
10.1021/ja045964v CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 16395-16404
9
16395

A R T I C L E S
Dong et al.
Scheme 1. Catalytic Cycle for GPx
tellurium analogues had all been demonstrated to catalyze the
4,6
reduction of hydroperoxides (ROOH) in the presence of thiols.
In the early effects to design the enzyme model with GPx-
like activity, the main ideas were focused on the imitation of
Se‚‚‚N interaction which was demonstrated important in stabi-
lization and activation of reactive selenol moiety in catalytic
had been reported to act as a GPx mimic by us13 and its catalytic
efficiency was 24-fold than that of PhSeSePh when thiol GSH
(1) was used as substrate. In our artificial enzyme model 2-TeCD
the active site designed had been placed on the secondary side
of cyclodextrin since it was established that a bound substrate
would have its prosthetic groups at the secondary side of
cyclodextrin. Cyclodextrins were known to favor aryl groups
7
cycle. Although some GPx mimics had some increase in
activity, most of them displayed only limited catalytic enhance-
ment. On the basis of the structural understanding for GPx, the
nature of enzyme for molecular recognition and catalysis as well
as the early work,6 we conceived that the generation of specific
binding ability for thiol substrate and correct incorporation of
the functional selenium/tellurium groups should be critical
approaches for the construction of an effective GPx model. The
previous studies from our group in preparing GPx models by
1
2,14
in the cavities.
The aryl thiol ArSH (2) may be able to take
d
some advantage of the binding cavities of 2-TeCD.
In present paper, 2-TeCD catalyzed the reduction of ROOH
by ArSH (2) was studied in detail. The complexation of ArSH
(
2) with â-cyclodextrin was investigated through UV spectral
1
titrations, fluorescence spectroscopy, H NMR and molecular
simulation. It was clearly shown that 2-TeCD which has strong
hydrophobic interaction produced large rate accelerations when
ArSH (2) acted as a thiol substrate. The large difference in the
activities of 2-TeCD with thiol ArSH (2) was ascribed to the
role of the binding ability as compared with thiol GSH (1), and
the recognition of substrates in the enzyme model could be
delineated from the catalytically kinetic data.
8
9
monoclonal antibody technique, bioimprinting, and transferring
1
0
nature enzymes had demonstrated this hypothesis. The re-
markable activity enhancements had been obtained when
chemical incorporation of catalytic selenocysteine groups into
existing or artificially generated thiol binding scaffolds. Hilvert’s
11
work in preparing semisynthetic selenoprotein, selenosubtilisin,
also supported this speculation. The selenosubtilisin which had
evolved to bind specific substrate, was 70 000 times more
efficient than diphenyl diselenide (PhSeSePh) in catalyzing the
reduction of tert-butyl hydroperoxide (t-BuOOH) by thiol
substrate 3-carboxy-4-nitrobenzenethiol (ArSH, 2).11 However,
the studies of structure-functional relationship had been blocked
by the complicated nature of macromolecular proteins. Con-
struction of small molecular GPx models becomes a good
alternative for elucidating the origin of substrate recognition in
enzyme catalysis and for potentially therapeutic applications.
In this respect, cyclodextrins were an attractive species for
enzyme model design. This molecule has the ability to accom-
modate various substrates to their cavities, and the two rims of
Results
Catalytic Activity. The catalytic activity was studied ac-
11,15
cording to a modified method reported by Hilvert et al.
using
3-carboxy-4-nitrobenzenethiol (ArSH, 2) as a glutathione (GSH,
1
) alternative. The initial rates (ν0) for the reduction of ROOH
(250 µM) by ArSH (2) (100 µM) in the presence of various
catalysts (eq 1) were determined at 25 °C and pH 7.0 (50 mM
phosphate buffer, 1 mM EDTA) by monitoring the UV
absorption at 410 nm due to the disappearance of the thiolate
absorption.
1
2
hydroxyl groups can be used to introduce prosthetic groups.
catalyst
ROOH + 2ArSH
8 ROH + ArSSAr + H O (1)
2
By coorperating of both recognition by cyclodextrin and
catalysis by ditelluride moiety, cyclodextrin-based enzyme
model 2, 2′-ditellurobis(2-deoxy-â-cyclodextrin) (2-TeCD, 3)
The relative activities of the compounds were summarized
in Table 1. For the peroxidase activity, the enzymatic rates were
corrected for the background (nonenzymic) reaction between
hydroperoxide and thiol. The initial concentration of ArSH (2)
(
7) (a) Iwaoka, M.; Tomoda, S. J. Am. Chem. Soc. 1994, 116, 2557 (b) Mugesh,
G.; Panda, A.; Singh, H. B.; Punekar, N. S.; Butcher, R. J. J. Am. Chem.
Soc. 2001, 123, 839.
-1
was measured from the 410-nm absorbance (ꢀ ) 13 600 M
(
8) (a) Ren, X.; Gao, S.; You, D.; Huang, H.; Liu, Z.; Mu, Y.; Liu, J.; Zhang,
Y.; Yan, G.; Luo, G.; Yang, T.; Shen, J. Biochem. J. 2001, 359, 369. (b)
Ding, L.; Liu, Z.; Zhu, Z.; Luo, G.; Zhao, D.; Ni, J. Biochem. J. 1998,
-1
cm , pH 7.0). The initial rate of the background (nonenzymic)
reaction between H2O2 and ArSH (2) was very slow (ν0 ) 0.507
3
32, 251.
-
1
(9) Liu, J.; Luo, G.; Gao, S.; Zhang, K.; Chen, X.; Shen, J. Chem. Commun.
µM min ), but a slight enhancement in the rate was observed
1
999, 199.
-1
when PhSeSePh (100 µM) was added (ν0 ) 0.012 µM min ).
(
10) Ren, X.; Jemth P.; Board, P. G.; Luo, G.; Mannervik, B.; Liu, J.; Zhang
K.; Shen, J. Chem. Biol. 2002, 9, 789.
(
11) Wu, Z. P.; Hilvert, D. J. Am. Chem. Soc. 1990, 112, 5647.
12) (a) Breslow, R.; Dong, S. D. Chem. ReV. 1998, 98, 1997. (b) Murakami,
Y., Kikuchi, J. I.; Hisaeda, Y.; Hayashida, O. Chem. ReV. 1996, 96, 721.
(13) (a) Ren, X.; Xue, Y.; Liu, J.; Zhang, K.; Zheng, J.; Luo, G.; Guo, C.; Mu,
Y.; Shen, J. ChemBioChem 2002, 3, 356. (b) Ren, X.; Xue, Y.; Zhang. K.;
Liu, J.; Luo, G.; Zheng, J.; Mu, Y.; Shen, J. FEBS Lett. 2001, 507, 377.
(14) Van Etten, R. C.; Sebastian, J. F.; Clowes, G. A.; Bender, M. L. J. Am.
Chem. Soc. 1967, 89, 3242.
(15) (a) Bell, I. M.; Fisher, M. L.; Wu, Z. P.; Hilvert, D. Biochemistry 1993,
32, 3754. (b) Bell, I. M.; Hilvert, D. Biochemistry 1993, 32, 13969.
(
(
c) Kirby, A. J. Angew. Chem. 1996, 108, 770; Angew. Chem., Int. Ed.
Engl. 1996, 35, 707. (d) Bender, M. L.; Komiyama, M. Cyclodextrin
Chemistry; Springer-Verlag: New York, 1978. (e) D’Souza, V. T.; Bender,
M. L. Acc. Chem. Res. 1987, 20, 146.
16396 J. AM. CHEM. SOC.
9
VOL. 126, NO. 50, 2004

Aryl Thiol Substrate 3-Carboxy-4-Nitrobenzenethiol
A R T I C L E S
a
Table 1. Initial Rate (ν0) and Activity for the Reduction of ROOH
(
250 µM) by Thiol GSH (1) (100 µM) and ArSH (2) (100 µM) in the
Presence of Various Catalysts at pH 7.0 (50 mM PBS, 1 mM
EDTA) and 25 °C
ν
0
b(M
‚
min-1)
activity
ArSHc GSHd
catalyst
hydroperoxide
ArSH
PhSeSePhe
ebselen
PhTeTePh
H2O2
H2O2
H2O2
(0.12 ( 0.01) × 10-7
1
1
0.48
67 0.90
0
e
-7
â-cyclodextrin H2O2
(0 8.03 ( 0.17) × 10
0.00052
-
-
-
6
6
5
2
-TeCD
H2O2
(2.55 ( 0.16) × 10
t-BuOOH (5.45 ( 0.27) × 10
CuOOH
(2.45 ( 0.09) × 10
21250 23.87
45417 32.53
204167 44.62
a
The initial rate of reaction was corrected for the spontaneous oxidation
b
in the absence of catalyst. All values are means of at least five times and
calculated from the first 5-10% of the reaction, and ν0 value ) means (
c
S. D. The concentration of catalyst: [PhSeSePh] ) 100 µM, [ebselen] )
1
)
0 µM, [PhTeTePh] ) 100 µM, [â-cyclodextrin] ) 100 µM and [2-TeCD]
1 µM in ArSH assay system. Calculated based upon GPx activity of
PhSeSePh equal to 1, assuming the rate has a first-order dependence on
d
the concentration of catalyst. The concentration of catalyst: [PhSeSePh]
)
10 µM, [ebselen] ) 5 µM, [PhTeTePh] ) 10 µM, [â-cyclodextrin] )
1
00 µM and [2-TeCD] ) 1 µM in coupled reductase assay system.
e
Calculated based upon GPx activity of PhSeSePh equal to 1. The reaction
solution contains 10% methanol (v/v), and methanol has no effect on the
activity.
The initial rate in the presence of diphenyl ditelluride (Ph-
-
1
TeTePh) (100 µM) was also slow (ν0 ) 0.803 µM min ). The
â-cyclodextrin itself was found to be inactive in this assay
system. When ebselen (10 µM) was tested, the initial rate had
some extent decrease (ca. 10%) compared to the spontaneous
oxidation. Ebselen had a too slow turnover in catalytic process,
and the stoichiometric reaction of ebselen and ArSH (2) maybe
resulted in this decrease. Under the identical conditions the
cyclodextrin-based ditelluride (2-TeCD) (1 µM), however,
exhibited a remarkable rate enhancement (2.546 µM min ).
Assuming that the rate had a first-order dependence on the
concentration of catalyst catalyzed the reduction of ROOH by
ArSH (2), these data (Table 1) suggested that the 2-TeCD was
at least 200 000-fold more efficient than PhSeSePh.
Figure 1. (A) [E] /ν (min) vs 1/[t-BuOOH] (M-1) for 1 µM 2-TeCD in
0
0
PBS, pH 5.5 and 25 °C, at [ArSH] ) 18 µM (9), 83 µM (O), 145 µM (2).
-
1
(
B) [E]0/ν0 (min) vs 1/[ArSH] (mM ) for 1 µM 2-TeCD in PBS, pH 5.5
and 25 °C, at [t-BuOOH] ) 5.1 mM (9), 10.1 mM (b), 50.0 mM (2),
100.0 mM (1).
-
1
specificity for reduction of aryl cumene hydroperoxide (CuOOH).
In the presence of 2-TeCD, CuOOH was reduced by ArSH (2)
at least 10 times faster than hydrogen peroxide (Table 1), similar
6
g
result had been reported recently. It therefore seemed that the
hydrophobic cavity provided by the cyclodextrin of 2-TeCD
acts as a binding site for the hydroperoxide substrate.
The activities of these compounds were also assessed using
GSH (1) as a thiol substrate in the classical coupled reductase
1
6
Kinetics. To probe the mechanism of 2-TeCD promoting the
peroxidase reaction, detailed kinetic studies were undertaken.
assay system under identical experimental conditions. The
relative activities of the compounds were also listed in Table
2
-TeCD catalyzed the reduction of a variety of structurally
1
. These data indicated that 2-TeCD was only 24 and 49 times
distinct hydroperoxides (ROOH), from the hydrophilic H2O2
via the tert-butyl hydroperoxide (t-BuOOH) to the bulky aryl
cumene hydroperoxide (CuOOH). Double-reciprocal plots (Fig-
ure 1, 2, and see Supporting Information) of initial rate versus
substrate concentration at all the individual concentration
revealed the characteristic parallel lines of a ping-pong mech-
more efficient than PhSeSePh and ebselen, respectively.
These results should not be surprising, since the enzyme
model had been designed to bind specific substrates, and this
specificity was one of its most desirable properties. For
structurally distinct thiol substrates, 2-TeCD exhibited a large
difference in thiol peroxidase activity. Cyclodextrins seem to
be a preferential scaffold for the compound ArSH (2) (the aryl
group in ArSH) rather than the hydrophilic compound GSH
18
anism with at least one covalent intermediate, in analogy with
3
a
natural GPx. Saturation kinetics were observed for each of
the enzymatic peroxidase reactions at all the individual con-
centrations of ArSH (2) and ROOH investigated. The kinetic
parameters for the enzymatic reactions between ArSH (2) and
the hydroperoxide substrates H2O2, t-BuOOH, and CuOOH were
shown in Table 2. These values were deduced from fitting the
experimental data to a ping-pong kinetic mechanism. The data
did not fit well to other models, such as sequential or
1
7
(
1). This observation allowed the difference between the
activities seen in the two assay systems to be elucidated and
suggested that the attempts to enhance the catalytic activity of
2
-TeCD should focus on recognition for the thiol substrate.
Furthermore, the notable result from this assay system was the
(
16) Wilson, S. R.; Zucker, P. A.; Huang, R.-R. C.; Spector, A. J. Am. Chem.
Soc. 1989, 111, 5936.
(17) Wenz, G. Angew. Chem. 1994, 106, 851; Wenz, G. Angew. Chem., Int.
Ed. Engl. 1994, 33, 803.
(18) Dalziel, K. Biochem. J. 1969, 114, 547.
J. AM. CHEM. SOC.
9
VOL. 126, NO. 50, 2004 16397

A R T I C L E S
Dong et al.
Table 2. Kinetic Parameters for the Peroxidase Activity of 2-TeCDa
k
max/KROOH
1
k
max/KArSH
1
hydroperoxide
k
max (min-1)
K
ROOH
K
ArSH
(µm)
(M- min-1)
(M- min-1)
3
4
5
7
H2O2
t-BuOOH
CuOOH
300 ( 17
900 ( 59
3710 ( 118
48 ( 3
55 ( 7
23 ( 1
28 ( 3
120 ( 18
627 ( 28
(6.00 ( 0.05) × 10
(1.64 ( 0.14) × 10
(1.61 ( 0.11) × 10
(1.05 ( 0.06) × 10
(7.50 ( 0.31) × 10
(5.92 ( 0.12) × 10
6
6
a
Values ) means ( S. D..
Scheme 2. Proposed Mechanism of the Thiol Peroxidase
Reaction of Ditellurides
equilibrium-ordered mechanisms. The relevant steady-state
equation (eq 2) for the enzymatic peroxidase reaction is
ν0
)
[
E]0
kmax[ArSH][ROOH]
(2)
KROOH[ArSH] + KArSH[ROOH] + [ArSH][ROOH]
where ν0 is the initial reaction rate, [E]0 is the initial enzyme
mimic concentration, kmax is a pseudo-first-order rate constant
and KROOH and KArSH are the Michaelis-Menten constants for
the hydroperoxide and ArSH (2), respectively.
The rate constants of the background reaction between ArSH
(
2) and hydroperoxide had been reported to vary in the order
k+l(RSH) above. It was apparent from this efficiency that the
size of ArSH (2) fits well to the cavity of â-cyclodextrin to
facilitate the binding of the thiol substrate in â-cyclodextrin
cavity and produce large rate accelerations. The semisynthetic
enzyme selenosubtilisin which had evolved to bind specific
substrate was in detail investigated and acted as an excellent
15b
k(H2O2) > k(CuOOH) > k(t-BuOOH).
In contrast, the
analogous rate constants of 2-TeCD and hydroperoxide (k )
kmax/KROOH) vary as k(CuOOH) > k(t-BuOOH) > k(H2O2). It
was possible that the first series reflected the intrinsic rate of
reaction between the hydroperoxides and a thiolate in the
absence of any significant binding effects, while the latter series
indicated that CuOOH and t-BuOOH were able to take binding
advantage of the cyclodextrin template and hence raised its kmax/
KROOH above that of hydrogen peroxide by increasing kmax and
lowering KROOH. It was worth noting that the Michaelis-Menten
contant (KArSH) values listed in Table 2 vary in magnitude in
the order KArSH (H2O2) < KArSH (t-BuOOH) < KArSH (CuOOH),
for the enzymatic reactions with H2O2, t-BuOOH, and CuOOH.
This clearly indicated that the competitive binding affinity of
thiol and hydroperoxide substrates for GPx mimic 2-TeCD did
exist in the enzymatic reaction. Furthermore importantly, Table
1
5
GPx mimic. Surprisingly, 2-TeCD catalyzed t-BuOOH by
ArSH (2) with higher second rate constant
a
4
-l
-l
(
k+l(t-BuOOH) ) 1.64 × 10 M min ) compared to seleno-
3 -l -l
subtilisin (k+l(t-BuOOH) ) 4 × l0 M min ). For structurally
distinct thiol substrates, the second rate constant of 2-TeCD-
4
-l
catalyzed reduction of ROOH by GSH (1) (k+l(GSH) ) l0 M
min ) was lower as compared with ArSH (2) (k+l(ArSH) )
l0 M min ). When GSH (1) acted as a thiol substrate 2-TeCD
was clearly far less efficient than with ArSH (2). It was
obviously shown that the capacity to bind the thiol substrate is
essential for the enzymatic catalytic efficiency.
-
l 13a
7
-l
-l
2
revealed that kmax/KROOH was not identical and increased upon
Mechanism. The kinetic data were presented in support of a
ping-pong mechanism with at least one covalent intermediate.
Further experiments were needed to characterize each of the
intermediates in the catalytic cycle. Engman and co-workers
had recently reported that diaryl ditelluride which mimic the
properties of GPx carried out the reaction mechanism as shown
in turn changing from H2O2 via t-BuOOH to CuOOH. This
revealed that the GPx mimic 2-TeCD has substrate specificity
for substrate hydroperoxide.
For GPx, acting on ROOH and GSH, at physiological pH
and 37 °C, the bimolecular rate constants for the reactions were
as follow: k+l(ROOH) ) 10 M min and k+l(GSH) ) l0
M
parameters for 2-TeCD, acting on ROOH and ArSH (2), at pH
8
-1
-l
7
2
0
in Scheme 2.
-l
-l
3a,h,5,19
min as determined previously.
The equivalent
It was known that the rate of the enzymatic reaction depends
on the concentration of intermediate ESeSG in Scheme 1.
Mugesh and co-workers had confirmed that the catalyst-substrate
complex (RSeSPh) does exist during the catalytic cycle through
5
-1
-l
5
.5 and 25 °C, were k+l(CuOOH) ) 1.61 × 10 M min and
k+l(ArSH) ) 5.92 × l0 M min ; k+l(t-BuOOH) ) 1.64 ×
l0 M min and k+l(ArSH) ) 7.50 × l0 M min ; k+l(H2O2)
6
-l
-l
4
-l
-l
-l
6
-l
-l
7b
a detailed kinetic studies. Herein, we did similar kinetic studies
to characterize the catalyst-substrate complex. Double-reciprocal
plots (Lineweaver-Burk plots) of initial rate versus substrate
concentration yielded families of linear lines (Figures 2 and 3).
The parallel lines corresponded to different concentrations of
the catalyst and indicated that the rates increase linearly with
the concentration of 2-TeCD. When the concentration of
3
-l
7
-l
-l
)
6 × l0 M min and k+l(ArSH) ) 1.05 × l0 M min .
However, under the similar conditions, the equivalent parameters
for selenosubtilisin, acting on t-BuOOH and ArSH (2), were
3
-l
-l
k+l(t-BuOOH) ) 4 × l0 M min and k+l(ArSH) ) 1.5 ×
7
-l
-l 11
l0 M min . Although these parameters were measured at
different pH and temperature for different thiol substrates, and
could not be directly equated, they did allow an approximate
comparison of these systems. The bimolecular reactions between
GPx, selenosubtilisin, and 2-TeCD and their respective thiol
substrates were very similar, as evidenced by the values of
(19) G u¨ nzler, W. A.; Vergin, H.; Muller, I.; Floh e´ , L. Hoppe-Seyler’s Z. Physiol.
Chem. 1972, 353, 1001.
(
20) Engman, L.; Stern, D.; Cotgreave, I. A.; Andersson, C. M. J. Am. Chem.
Soc. 1992, 114, 9737.
16398 J. AM. CHEM. SOC.
9
VOL. 126, NO. 50, 2004

Aryl Thiol Substrate 3-Carboxy-4-Nitrobenzenethiol
A R T I C L E S
Figure 4. (A) UV spectra of isolated CDTeSAr in phosphate buffer, pH
-
1
-1
5
.5 (ꢀ336 ) 9000 M cm ). (B) After addition of excess DTT (for ArSH,
-
1
-1
pH 5.5, ꢀ410 ) 12 600 M cm ).
Scheme 3
peroxide, diaryl ditelluride reacted with thiols to give telluro-
2
0
sulfide (eq 3). The tellurenyl sulfides derived from reactions
ArTe-TeAr + 2RSH + H O f 2ArTeSR + 2H O (3)
2
2
2
Figure 2. (A) [E]0/ν0 (min) vs 1/[CuOOH] (M-1) for 1 µM 2-TeCD in
PBS, pH 5.5 and 25 °C, at [ArSH] ) 83 µM (9), 188 µM (O), 260 µM
between ditellurides and thiol may disproportionate to the
corresponding dichalcogenides. Moreover, the tellurenyl sul-
21
-
1
(
2). (B) [E]0/ν0 (min) vs 1/[ArSH] (mM ) for 1 µM 2-TeCD in PBS, pH
fides may react with the corresponding tellurols to produce the
5.5 and 25 °C, at [CuOOH] ) 3.1 mM (9), 6.3 mM (b), 20.0 mM (2),
0.0 mM (1).
2
2
5
dichalcogenides as shown in Scheme 3. However, in the
absence of hydroperoxide, the reaction of 2-TeCD with ArSH
(2) (eq 4) could be readily monitored spectroscopically by the
2
-TeCD + ArSH h CDTeH + CDTeSAr (4)
disappearance of thiolate at 410 nm. The CDTeSAr (5) could
be isolated via the reaction of 2-TeCD with 2 equiv of ArSH
(2) under air by gel filtration on Sephadex G-15 and was
characterized (vide infra). The treatment of CDTeSAr (5) with
excess 1, 4-dithio-DL-threitol (DTT) led to the release of ArSH
(
(
2) (Figure 4). The dependence of the yield of isolated CDTeSAr
5) on the thiol concentration suggested to us that the form of
CDTeSAr (5) was in equilibrium (eq 5) with the corresponding
tellurolate:
-
-
Figure 3. Lineweaver-Burk plots obtained for the model reaction in the
presence of different enzyme mimic concentration [2-TeCD] at pH 5.5 (PBS)
and 25 °C. The initial H2O2 concentration was fixed to 250 µM.
CDTeSAr + ArS h CDTe + ArSSAr
(5)
The position of equilibrium lay far to the left. In this case, the
enzymatic activity was observed to decrease with increasing
pH (Figure 5). To study the nature of this variation in enzymatic
catalysis, the complexation of â-cyclodextrin and ArSH (2) was
carried out by UV spectral titrations at various pH. It was found
that the complex stability constant of â-cyclodextrin and ArSH
2) decreased with increasing pH (see Supporting Information).
The deprotonated nature of ArSH (2) in different pH was
responsible for this decrease, and the polar substituents at high
2-TeCD was maintained constant while substrate concentration
[ArSH] was increased, a rapid increase of rate was observed in
the initial stages; however, when the substrate concentration
was increased further, the rate became constant (see Supporting
Information). At the same time, when the concentration of
(
2-TeCD was increased, the rates become very high for higher
concentration of ArSH (2). From this observation, it was clearly
shown that the intermediate (CDTeSAr (5)) did exist during
the catalytic cycle.
Concerning the reaction mechanism of 2-TeCD catalysis, we
had good reasons to believe that CDTeSAr (5) is a key
intermediate in the catalytic cycle. In the presence of hydro-
(
21) (a) Mugesh, G.; Panda, A.; Kumar, S.; Apte, S. D.; Singh, H. B.; Butcher,
R. J. Organometallics 2002, 21, 884. (b) Fischer, H.; Dereu, N. Bull. Soc.
Chim. Belg. 1987, 96, 757.
22) Haene, G. R. M. M.; De Rooij, B. M.; Vermeulen, N. P. E.; Bast, A. Mol.
Pharmacol. 1990, 37, 412.
(
J. AM. CHEM. SOC.
9
VOL. 126, NO. 50, 2004 16399

A R T I C L E S
Dong et al.
Scheme 4. Proposed Catalytic Mechanism of 2-TeCD.
of enzymatic reaction and the turnover reaction may proceed
via the mechanism shown in Scheme 4. 2-TeCD exerted its
peroxidase activity via tellurol, tellurenic acid, and tellurosulfide
in ArSH assay system.
Complexation of ArSH (2) and â-Cyclodextrin. Since
â-cyclodextrins are cavities in the molecular center, they could
accommodate various guest molecules in their cavities, forming
inclusion complexes. Recently, the studies of cyclodextrin
Figure 5. Plots of ν0 vs pH for the reduction of t-BuOOH (250 µM) by
ArSH (100 µM) in the presence of 2-TeCD (1 µM) at 25 °C in phosphate
buffer.
pH (the deprotonated form of ArSH) led to a low binding in
the highly hydrophobic interior of â-cyclodextrin. It was
interesting to note that the pH value of the undulating range in
Figure 5 is remarkably close to pKa (pKa ) 4.4)1 of ArSH
25
26
complexes were reviewed by Inoue and Schneider in detail.
The aryl thiol ArSH (2) may be able to take some advantage of
the binding site of 2-TeCD, since the enzymatic system exhibited
a significant rate advantage for thiol peroxidase activity. The
specific binding for thiol substrate which contributes to high
activity of GPx mimic had been demonstrated by catalytic
5a
(2), and the enzymatic activity had some increasing from pH
4.6 to 5.0. Thiols were much less nucleophilic than their
corresponding thiolate ions. This would be consistent with a
6d
antibody and seleno-glutathione transferase. We thought that
-
requirement for thiolate, ArS , as a nucleophile in the rate-
the high catalytic activity of 2-TeCD was also driven by the
strong binding between 2-TeCD and ArSH (2). To support this
hypothesis, the binding constant of â-cyclodextrin with ArSH
determining step, again suggestive of the conversion of CDTe-
-
SAr to CDTe . The tellurolate would be readily oxidized
aerobically to generate 2-TeCD. We anticipated that the tellurol
might also be irreversibly trapped by a suitable alkylating
agent.1 To test this hypothesis, 2-TeCD was incubated with
a large excess of iodoacetic acid at pH 4.0 in the presence of
ArSH (2). The compound was recovered by gel filtration on
G-15 Sephadex after additions of thiol, and was found to lose
its peroxidase activity completely. The compound was fully
characterized, and the formation of diorganyl telluride (4)
revealed the tellurol was one of the intermediates of the catalytic
cycle. Furthermore, compound 4 could also be obtained from
the reaction of 2-TeCD and iodoacetic acid in the presence of
sodium hydroborate (vide infra). As a control, 2-TeCD was
similarly treated with iodoacetic acid, but in the absence of thiol.
This batch of 2-TeCD was recovered with full activity. These
results showed that the tellurol group, produced by reduction
of 2-TeCD, can be irreversibly modified by iodoacetic acid and
that this modifications of the prosthetic group abolishes per-
oxidase activity. Since these compounds were diselenides or
ditellurides,20 they may obey the catalytic mechanism that had
been proposed for the natural GPx (Scheme 1). In general,
organotellurium compounds can be oxidized to overoxidized
tellurium species. Cyclodextrin-based telluronic acid was ob-
tained by the reaction of 2-TeCD and an excess H2O2, however,
the overoxidized tellurium species catalyzed the reduction of
H2O2 by ArSH (2) at least 50 times less efficiently than 2-TeCD
in catalytic process.24 This effectively rules out any significant
role of the overoxidized tellurium species in catalytic cycle.
Furthermore, a ping-pong mechanism from the kinetic data also
supported this conclusion. As discussed above, overoxidized
tellurium species may not exist in the primary catalytic cycle
(
2) was measured by UV spectral titrations. According to the
27
modified Hildebrand-Benesi equation, the linear plot of the
reciprocal of the absorbance difference ∆A and molar concen-
tration of the host molecule [H] indicated a 1:1 complex between
the host molecule â-cyclodextrin and the guest molecule ArSH
5,23
3
-1
(2) with a binding constant of above 10 M (see Supporting
Information). We also characterized the complexation of â-cy-
clodextrin and ArSH (2) by means of fluorescence spectroscopy
and found that the fluorescence intensity of ArSH (2) obviously
increase after addition of â-cyclodextrin (see Supporting
Information). This experiment also provided the proof for the
complexation between â-cyclodextrin and ArSH (2). To further
confirm the inclusion complexation of â-cyclodextrin and ArSH
(
2), the catalytic activity of 2-TeCD-catalyzed the reduction of
peroxide by ArSH was assessed in the presence of an inhibitor,
-adamantaneethanol. Since the adamantane group is strongly
bound to the cavity of â-cyclodextrin (association constant: Ka
1
7
b
4
-1 25
>
10 M ), the adamantane group could rival with ArSH
(2) in the inclusion complexation of 2-TeCD and could decrease
the catalytic activity. As expected a large decrease of catalytic
activity of 2-TeCD was observed (see Supporting Information).
At the same time, a H NMR study on the binding inhibit was
investigated and indicated that in the presence of 1-adaman-
taneethanol some amount of ArSH (2) did not enter into the
cavity of â-cyclodextrin, although under identical conditions
in absence of any inhibitor all of ArSH (2) entered into the
cavity of â-cyclodextrin (see Supporting Information). These
results strongly supported the inclusion complexation of enzyme
mimic and thiol substrate ArSH (2) in ArSH assay system. In
1
1
this case, H NMR spectra were expected to provide more
(
23) Forstrom, J. W.; Zakowski, J. J.; Tappel, A. L. Biochemistry 1978, 17,
639.
2
(25) Rekharsky, M. V.; Inoue, Y. Chem. ReV. 1998, 98, 1875.
(26) Schneider, H. J.; Hacket, F.; R u¨ diger, V. Chem. ReV. 1998, 98, 1755.
(27) (a) Benesi, H. A.; Hildebrand, J. H.; J. Am. Chem. Soc. 1949, 71, 2703.
(b) Rossotti, F. J. C.; Rossotti, H. The Determination of Stability Constants;
McGraw-Hill Book Co., Inc.: New York, 1961, p 276.
(
24) The GPx-like activity of cyclodextrin-based telluronic acid catalyzed the
reduction of hydrogen peroxide was found to be 420 (calculated based upon
GPx-like activity of PhSeSePh equal to 1) in ArSH assay system under
identical experimental conditions.
16400 J. AM. CHEM. SOC.
9
VOL. 126, NO. 50, 2004

Aryl Thiol Substrate 3-Carboxy-4-Nitrobenzenethiol
A R T I C L E S
Figure 6. 1H NMR spectrum of ArSH (2) in the absence (A) and in the presence (B) of â-cyclodextrin in D2O at ambient temperature.
information about the cyclodextrin complex state, since analyses
of the UV spectral titrations and fluorescence spectrum of the
cyclodextrin complex indicated the ArSH (2) was located in
1
the cavity. The H NMR spectra (Figure 6) of the ArSH (2)
changed significantly upon addition of excess â-cyclodextrin
a
c
at ambient temperature. The H and H protons of aromatic
1
region in H NMR spectra showed significant upfield shift
compared to the spectrum of the ArSH (2) alone. The H proton
b
showed significant downfield shift, however, indicating that a
b
moderate increase in water exposure of H proton of aromatic
a
b
region. The H and H protons of aromatic region showed
remarkable cleavage compared to the spectrum of the ArSH
a
b
(
2) alone, indicating that the H and H protons of aromatic
region were locating in different microenvironment in the
presence of excess â-cyclodextrin in water. In addition, in H
Figure 7. Lowest Etotal structural model of the complexation of â-cyclo-
dextrin corresponding to the accessible orientation of sulfur headgroup of
ArSH (2) toward the secondary side of â-cyclodextrin.
1
NMR spectra the peaks of outside protons of â-cyclodextrin
region in the presence of ArSH (2) became obviously wide
compared to the spectrum of â-cyclodextrin alone (see Sup-
porting Information). This experiment revealed that the interac-
tion of â-cyclodextrin and ArSH (2) did exist in the complex-
ation. At the same time, a 2D NMR spectrum also provided a
further proof for the inclusion complexation of â-cyclodextrin
and ArSH (2), in which the interaction of the aryl protons of
ArSH (2) and inside protons of â-cyclodextrin had been
observed (see Supporting Information).
the cavity of â-cyclodextrin, and each of the orientations resulted
in a stable structure, and accordingly produced an Etotal (see
Supporting Information). We found that ArSH (2) can enter into
the cavity of â-cyclodextrin easily and the orientations of “S to
B” and “S to A” were much easier than those of “N to B” and
“
N to A”, as evidenced by the lower Etotal. It was expected that
the orientation of “S to B” was easiest among them and the
complexation structure was shown in Figure 7. When we
optimized the structure of the intermediate CDTeSAr (5), the
aryl group in CDTeSAr (5) could not enter into the cavity of
cyclodextrin but located upon its secondary side. This result
Molecular Simulation of Complexation of â-Cyclodextrin
and Derivative with ArSH (2). The Dreiding 2.21 force field
used for molecular simulation was able to capture much of the
observed experimental trends for the complexation of â-cyclo-
dextrin and derivative with ArSH (2). The structure of â-cy-
clodextrin was kept fixed during the entire simulation. The total
potential energy (Etotal) acted as a criterion of complexation
stability of â-cyclodextrin and derivative with ArSH (2). There
were four accessible orientations (S to A, S to B, N to A, N to
1
was proved experimentally by H NMR spectrum of CDTeSAr
1
(
5). In the H NMR spectrum the chemical shifts of the aryl
and cyclodextryl protons of CDTeSAr (5) had no any change
compared to the spectrum of ArSH (2) or â-cyclodextrin alone,
indicating no self-inclusion of CDTeSAr (5) (see Supporting
Information). It was essential that the intermediate CDTeSAr
(5) could bind another ArSH molecule to facilitate the catalytic
28
cycle of the enzymatic reaction, otherwise it would be self-
inhibited. To insight into the effects of geometric preference
on the catalysis, the tellurium atom was introduced to the
primary side of â-cyclodextrin. 6, 6′-Ditellurobis(6-deoxy-â-
cyclodextrin) (6-TeCD, unpublished) also displayed thiol per-
B) in the simulation of ArSH (2) (size 5.68-5.91 Å) accessed
(
28) The prepositive “S” implied the sulfur headgroup of ArSH (2) and “N”
implied the nitrogen headgroup of ArSH (2); the latter “A” implied the
primary side of â-cyclodextrin and “B” implied the secondary side of
â-cyclodextrin.
J. AM. CHEM. SOC.
9
VOL. 126, NO. 50, 2004 16401

A R T I C L E S
Dong et al.
size-fit relation between a host cavity and a guest molecule plays
an important role in molecular recognition by cyclodextrins,
indicating that the hydrogen bonding, van der Waals forces,
and hydrophobic interactions should depend on how the size
and/or shape of a guest molecule fit into the host cavity.31 The
complexation of ArSH (2) with â-cyclodextrin was investigated
1
through UV spectral titrations, fluorescence spectrum, H NMR
and molecular simulation, and these results indicated that ArSH
(2) fits well to the size of the cavity of â-cyclodextrin.
Consequently, in our miniature enzyme model, 2-TeCD have
cavities that (a) provide maximum hydrophobic interaction with
a substrate to form complexes, (b) fit the aromatic ring of the
bound substrate, and (c) orient the thiolate ion of the bound
substrate toward active site for nucleophilic attack. All these
results indicated that the thiol substrate ArSH (2) takes some
advantage of the binding site of 2-TeCD and improves the
catalytic efficiency of 2-TeCD, but there was no reason to
believe that this activity was optimal. A semisynthetic enzyme
selonosubtilisin retained some of the substrate specificity of its
natural template and favored ArSH (2) as a preferential thiol
substrate, however, for thiol substrate GSH (1), the enzymatic
Figure 8. Lowest Etotal structural model of the complexation of CDTeSAr
(
(
5) corresponding to the accessible orientation of sulfur headgroup of ArSH
2) toward the primary side of â-cyclodextrin.
oxidase activity, but was obviously lower than 2-TeCD (3).29
The difference of catalytic efficiency might be made by
geometric preference. Inoue and co-worker recently reported a
similar system and showed that the type of substituent introduced
to cyclodextrin drastically affects the molecular recognition
30
ability. We did a similar molecular simulation using CDTeSAr
15a
efficiency was even lower. Similarly, 2-TeCD also seemed
(5) as a â-cyclodextrin alternative. The Te-S-Ar moiety of
to have a preference for ArSH (2), although possibly for different
reasons. Cytosolic GPx exhibits a strong specificity for its thiol
substrate GSH (1), with small structural changes in the thiol
CDTeSAr (5) was flexible, and another ArSH (2) molecule was
easy to locate in the cavity of cyclodextrin. However, we found
that the preferential orientation preferred “S to A” to “S to B”,
as evidenced by the lower Etotal, and “S to A” was easiest
orientation among them (Figure 8). Consequently, the cavity
of â-cyclodextrin oriented the thiolate ion of ArSH (2) toward
active site for nucleophilic attack, and this procedure was
necessary for obtaining high catalytic activity in 2-TeCD
catalysis. From molecular simulation we found that ArSH (2)
could easily enter into the cavity of cyclodextrin to form
complex with a favorable orientation for catalysis.
5,32
leading to large reductions in catalytic efficiency. Cytosolic
GPx, selenosubtilisin, and 2-TeCD all exhibited dramatically
different thiol substrate specificity and catalytic efficiency. It
was clear that 2-TeCD which has strong hydrophobic interaction
with ArSH (2) as thiol substrate produced large rate accelera-
tions.
A comparison of kinetic parameters (Table 2) obtained from
kinetic analyses of 2-TeCD using a variety of structurally distinct
ROOH, such as H2O2, t-BuOOH, and CuOOH, as an oxidative
reagent indicated the following conclusions. First, because the
maximal kcat value for the enzymatic reaction (kmax) was altered,
the hydroperoxide substrate must be involved in a step which
is at least partially rate-determining. Second, the variety of kmax/
KROOH values and saturation kinetics together with different
KROOH values suggested that 2-TeCD has substrate specificity
for hydroperoxides. Third, the high kmax/KArSH values and
saturation kinetics together with the low KArSH values indicated
that ArSH (2) is a preferential thiol substrate of 2-TeCD. Finally,
the variational orders of the kmax/KROOH values and kmax/KArSH
values revealed that the 2-TeCD can recognize and bind these
substrates and the competitive binding affinity of these substrates
for 2-TeCD did exist in the enzymatic reaction. In the natural
GPx the selenolate locates in a shallow depression on the
protein’s surface and may essentially react with any approaching
hydroperoxide: the enzyme has no real substrate specificity for
ROOH, provided that steric hindrance does not prevent their
Discussion
2
-TeCD had been shown as a GPx mimic and its thiol
peroxidase activity was assessed in two assay systems: classical
coupled reductase assay and ArSH assay. When GSH (1) was
used as a thiol substrate in a classical coupled reductase assay
system, the thiol peroxidase activity of 2-TeCD was only 24
times than that of PhSeSePh. However, in ArSH (2) assay
system, as shown in Table 1, it was apparent that 2-TeCD
reduces H2O2, t-BuOOH and CuOOH effectively and the thiol
5
peroxidase activity was almost 10 times than that of PhSeSePh.
This rate enhancement was remarkable, and reflected the
recognition action for thiol substratre in 2-TeCD catalysis.
2
-TeCD seemed to be a preferential scaffold for the compound
ArSH (2) (the aryl group in ArSH) rather than the hydrophilic
compound GSH (1). It was generally known that cyclodextrins
have been extensively exploited in the past as enzyme models
and molecular receptors because of their capacity to accom-
modate various guest molecules in their hydrophobic cavities
through host-guest chemistry, and partly because of large
33
reaction. However, the thiol peroxidase activity of 2-TeCD
depended upon the nature of ROOH. In the presence of 2-TeCD,
CuOOH was at least 10 times faster than hydrogen peroxide.
numbers of hydroxyl groups in all directions around the
cavity.1
2a,b
Inoue and co-workers had recently shown that the
(
31) (a) Inoue, Y.; Hakushi, T.; Liu, Y.; Tong, L.; Shen, B.; Jin, D. J. Am.
Chem. Soc. 1993, 115, 475. (b) Inoue, Y.; Liu, Y.; Tong, L.; Shen, B. J.
Am. Chem. Soc. 1993, 115, 10637.
(
29) The GPx-like activity of 6, 6′-ditellurobis(6-deoxy-â-cyclodextrin) (6-TeCD)
catalyzed the reduction of hydrogen peroxide by ArSH or GSH was found
to be 4083 and 1.98 respectively (calculated based upon GPx-like activity
of PhSeSePh equal to 1) in two assay systems under identical experimental
conditions.
(32) Floh e´ , L.; G u¨ nzler, W. A.; Jung, G.; Schaich, E.; Schneider, F. Hoppe-
Seyler’s Z. Physiol. Chem. 1971, 352, 159.
(33) Floh e´ , L. In Glutathione: Chemical, Biochemical, and Medical Aspects;
Dolphin, D., Poulson, R., Avamovic, O., Eds.; Wiley: New York, 1989, p
644.
(30) Liu, Y.; You, C. C.; Wada, T.; Inoue, Y. J. Org. Chem. 1999, 64, 3630.
16402 J. AM. CHEM. SOC.
9
VOL. 126, NO. 50, 2004

Aryl Thiol Substrate 3-Carboxy-4-Nitrobenzenethiol
A R T I C L E S
Our kinetic data also suggested that the attack of ArS- on
CDTeSAr may be rate-determining to some degree and hence
the CDTeSAr would accumulate, while the concentration of
FT66V infrared spectrometer. H NMR and 13C NMR were
measured on a Bruker AM-500 spectrometer. Elemental Analy-
ses were determined on a Perkin-Elmer 240 DS elemental
analyzer. Molecular weight was obtained from a LDI-1700
MALDI-TOF-MS (Linear Scientific Inc., USA). Fluorescence
spectral measurements were performed on a Shimadzu RF-
5301PC Spectrofluorophotometer. The spectrometric measure-
ments were carried out with a Shimadzu 3100 UV-vis-near-
IR Recording Spectrophotometer or Lambda 800 Spectrophoto-
meter interfaced with a personal computer. Data were acquired
and analyzed by using ultraviolet spectroscopy software. The
temperature for UV time course studies was controlled within
(() 0.5 °C by use of a LAUDA compact low-temperature
thermostat RC6 CP. Phosphate buffer (PBS) was used in the
all experiments unless otherwise noted. The buffer pH values
were determined with a METTLER TOLEDO 320 pH Meter.
The concentrations of the hydroperoxide stock solutions were
determined by titration with potassium permanganate.
1
-
CDTe would be low, under steady-state conditions. Such a
low concentration of the tellurolate may be responsible for the
slow rate of consumption of hydroperoxide, which reacts with
this enzymatic intermediate. The natural GPx was believed to
have evolved to near optimal efficiency for the decomposition
of ROOH. All of the GPx family contained an active triad
consisting of selenocysteine, glutamine, and tryptophan resi-
3
4
dues, and the cytosolic GPx contained a GSH binding site
3d
consisting of one lysine and four arginine residues. Our small
molecular enzyme model 2-TeCD lacked these features of
binding site and was relatively simpler than the natural GPx.
However, in the GPx mimic 2-TeCD the hydrophobic cavities
could accommodate accurately ArSH (2) to enhance the catalytic
efficiency. 2-TeCD displayed a higher catalytic efficiency than
that of the semisynthetic enzyme selenosubtilisin which had
evolved to bind specific substrate. It was very interesting that
a small molecular enzyme model exhibits such a remarkable
thiol peroxidase activity. This case successfully corroborated
our strategy of imitating GPx in small molecular enzyme model.
The turnover reaction of 2-TeCD-catalyzed reduction of
ROOH by ArSH (2) may proceed via the mechanism shown in
Scheme 4, in analogy with the natural GPx, and exerted its thiol
peroxidase activity via tellurol, tellurenic acid, and tellurosulfide
in ArSH assay system.
Synthesis of 3-Carboxy-4-Nitrobenzenethiol (ArSH, 2).
3-carboxy-4-nitrobenzenethiol was prepared by reduction of the
corresponding disulfide 5, 5’-dithiolbis(2-nitrobenzoic acid)
following the procedure of Silver and was characterized by
H NMR: (500 MHz, D2O) δ 7.91 (d, 1 H, J ) 8.5 Hz), 7.44-
3
5
1
7.42 (m, 2 H, J ) 8.5 Hz); UV/vis (PBS, pH 7.0): λmax (ꢀ) )
-
1
3
-1
410 nm (13600 mol dm cm ).
Synthesis of Compound 4. Method 1. 2-TeCD (100 mg, 0.04
mmol) and iodoacetate acid (149 mg, 0.8 mmol) were dissolved
in 30 mL of phosphate buffer (pH 4.0) under nitrogen, and large
excess amount of 3-carboxy-4-nitrobenzenethiol was added
dropwise. After the addition was complete, the mixture was
allowed to stir at room-temperature overnight. The mixture was
purified by centrifugation. The resulting solution was freeze-
dried and the lyophilized powder was washed with acetone three
times. The residue was purified on a column of Sephadex G-15
with distilled water as the eluent. The resulting solution was
again freeze-dried and a pure sample was obtained in 27% yield
In conclusion, we had shown that 2-TeCD catalyzes the
reduction of ROOH by an aryl thiol ArSH (2) with remarkable
catalytic efficiency. Studies of the kinetics of the 2-TeCD-
catalyzed reduction of ROOH by ArSH (2) suggested that
binding substrate was essential for the thiol peroxidase activity
of GPx mimics. The high catalytic efficiency and selectivity,
together with water-soluble and thermal stability gave 2-TeCD
a real advantage compared to other GPx mimics. 2-TeCD
represented an excellent alternative for the study of enzymatic
specificity and potential pharmaceutical application.
1
as a colorless solid. H NMR (500 MHz, D2O): δ 3.23-4.08
1
3
(
3
m, 44 H), 4.94-5.27 (m, 7 H); C NMR (500 MHz, D2O): δ
Exceprimental Section
-1
9.6, 60.4, 72.4, 72.6, 73.1, 81.4, 101.7, 161.0; IR (cm ,
KBr): νj ) 3340 (OH), 2960, 2928, 2855 (CH, CH2), 1680
COOH), 1620, 1110, 1080, 1030 (-O-); MALDI-MS: calcd.
1303.6 found 1304.0; Anal. Calcd. for C44H72O36Te‚7H2O: C,
General Procedures. â-cyclodextrin was purchased from
(
Tianjin Chemical Plant, recrystallized three times from distilled
1
3
water, and dried for 12 h at 120 °C in vacuo. 2-TeCD was
prepared as described previously and characterized in detail.
p-Toluenesulfonyl chloride and iodoacetate acid were also
purchased from Tianjin Chemical Plant. Tert-butyl hydroper-
oxide (t-BuOOH) and reduced glutathione (GSH) were pur-
chased from Merck. 1, 4-Dithio-DL-threitol (DTT) was obtained
from Bebco. Cumene hydroperoxide (CuOOH) and 1-adaman-
taneethanol were purchased from Fluka. Diphenyl ditelluride
3
6.91; H, 6.01. Found: C, 36.55; H, 5.65.
Method 2. Under a nitrogen atmosphere, 6.1 mg (0.16 mmol)
of NaBH4 was added to 15 mL (0.04 mmol) of the stock solution
of 2-TeCD and the colorless mixture stirred at room temperature
for 15 min. Iodoacetate acid (25 mg, 0.4 mmol) was then added
dropwise via syringe. After the addition was complete, the
colorless mixture was stirred at room temperature for 1 h.
Acetone was added and the compound was allowed to precipi-
tate. The residue was purified by a similar way above and a
pure sample was obtained in 92% yield as a colorless solid. Its
characteristics agreed well with the results above.
(
PhTeTePh) was obtained from Aldrich. Sodium hydroborate,
diphenyl diselenide (PhSeSePh), 2-phenyl-1, 2-benzoisoselena-
zol-3(2H)-one (ebselen), 5, 5’-dithiolbis(2-nitrobenzoic acid),
â-nicotinamide adenine dinucleatide phosphate reduced form
Synthesis of CDTeSAr (5). 2-TeCD (100 mg, 0.04 mmol)
was dissolved in 10 mL of distilled water, and 2 equiv (16 mg,
(
NADPH), and glutathione reductase were purchased from
Sigma. Sephadex G-15 was purchased from Amersham Phar-
macia Biotech, Uppsala, Sweden. All other chemicals were of
the highest purity commercially available and were used without
further purification. IR spectra were recorded on a Bruker IFS-
0
.08 mmol) of 3-carboxy-4-nitrobenzenethiol was added drop-
wise. After the addition was complete, the mixture was allowed
to stir at room temperature under air for 2 h. The resulting
mixture was purified by centrifugation and Sephadex G-15
(
34) Ursini, F.; Maiorino, M.; Brigelius-Floh e´ , R.; Aumann, K. D.; Roveri, A.;
Schomburg, D.; Floh e´ , L. Methods Enzymol. 1995, 252, 38.
(35) Silver, M. Methods Enzymol. 1979, 62D, 135.
J. AM. CHEM. SOC.
9
VOL. 126, NO. 50, 2004 16403

A R T I C L E S
Dong et al.
column chromatography (Φ5 × A80 cm; λ ) 254 nm) with
distilled water as the eluent. The resulting solution was freeze-
dried and the lyophilized powder was washed with ethyl ether
three times. After residue was dried in vacuo, a pure sample
Kinetic Analysis. The reactions of the enzyme-catalyzed
reduction and the nonenzymatic reduction of hydroperoxides
by ArSH (2) were studied by following the disappearance of
the thiolate absorption at 410 nm, at pH 5.5 (50 mM phosphate
buffer, 1 mM EDTA) and 25 °C. To investigate the dependency
of rate on substrate concentration, the reaction rates were
determined at several concentration of one substrate while
keeping the concentration of the other constant. All kinetic
experiments were performed in a solution containing phosphate
buffer (50 mM, pH 5.5), ethylendiaminetetraacetate (EDTA, 1
mM), and appropriate concentrations of ArSH (2), hydroper-
oxides, and 2-TeCD. The reaction rates were determined on
Shimadzu 3100 UV/vis-near-IR Recording Spectrophotometer.
The reaction was initiated by addition of ROOH. The enzymatic
rates were corrected for the background reactions between
ROOH and ArSH (2). The initial concentration of ArSH (2)
1
was obtained in 39% yield as a light yellow solid. H NMR
(
500 MHz, D2O): δ 3.22-4.18 (m, 42 H), 4.88-5.26 (m, 7
-1
H), 7.15-8.05 (m, 3H); IR (cm , KBr): νj ) 3340 (OH), 2928
(
CH, CH2), 1688, 1562, 1525 (Ar), 1625, 1154, 1080, 1030
-1
(-O-); UV/vis (PBS, pH 5.5): λmax (ꢀ) ) 336 nm (9000 mol
3
-1
dm cm ); MALDI-MS: calcd. 1442.6 found 1443.1; Anal.
Calcd. for C49H73O38NSTe‚6H2O: C, 37.86; H, 5.47; N, 0.90;
S, 2.06. Found: C, 38.01; H, 5.72; N, 1.04; S, 2.21.
Synthesis of Cyclodextrin-Based Telluronic Acid. 2-TeCD
(100 mg, 0.04 mmol) was dissolved in 15 mL of distilled water,
and a large excess amount (30%, 150µL) of H2O2 was added.
After the addition was complete, the mixture was allowed to
stir at room-temperature overnight. The resulting mixture was
purified by Sephadex G-15 column chromatography with
distilled water as the eluent. The resulting solution was freeze-
dried and a pure sample was obtained in 80% yield as a colorless
-
1
was measured from the 410-nm absorbance (ꢀ ) 12 600 M
-
1
cm , pH 5.5). Each initial rate was measured at least 5 times
and calculated from the first 5∼10% of the reaction. Line-
weaver-Burk plots were obtained by using the Origin 7.0
(professional version) program. For each set of experiments a
straight line was drawn with the best-fit method.
1
solid. H NMR (500 MHz, D2O): δ 3.29-4.28 (m, 42 H),
4
.76-5.19 (m, 7 H); 13C NMR (500 MHz, D2O): δ 60.5, 72.3,
-
1
7
2.5, 73.6, 81.5, 102.4; IR (cm , KBr): νj ) 3342 (OH), 2926
Molecular Simulation. We had used the molecular modeling
2
36
(CH, CH2), 1621, 1155, 1080, 1031 (-O-); MALDI-MS:
program CERIUS 4.6 (Accelrys Inc.; San Diego, CA) to carry
out our molecular simulation. The Dreiding 2.21 force field from
calcd. 1295.6 found 1296.3; Anal. Calcd. for C42H70O37Te‚
2
6
H2O: C, 35.96; H, 5.89. Found: C, 36.18; H, 6.03.
CERIUS software package was used in the entire simulation
which had been found to be reliable for many organic system.37
All of the simulations were started from energy-minimized
structures obtained through the Smart Minimizer method and
Convergence Level was set to high. The spline function was
chosed for switching function in nonbond interaction. The partial
charges were assigned with Charge Equilibration method before
each Minimization.
Coupled Reductase Assay. The GPx-like activities of these
1
6
compounds were measured using the Wilson’s method with
minor modification. The assay mixture contained 50 mM
phosphate buffer, pH 7.0, 1 mM EDTA, 100 µM GSH, 250
µM ROOH, 0.25 mM NADPH, 1 unit of glutathione reductase,
and a moderate amount of test compound at 25 °C. Reaction
was initiated by the subsequent addition of ROOH and the
absorbance at 340 nm was recorded for a few minutes to
calculate the rate of NADPH consumption. Methanol (10%, v/v)
was used to increase the solubility of PhSeSePh and PhTeTePh
in the enzymatic reactions.
In the coupled reductase assay system, the consumption of
NADPH in the absence of catalyst was shown as follows:
ROOH [NADPH consumption (µM/min) ( standard deviation],
H2O2 [1.28 ( 0.05], t-BuOOH [0.61 ( 0.02], CuOOH [0.75 (
Acknowledgment. We are grateful to the financial support
of the Major State Basic Research Development Program (973
Project, Grant. No. G2000078102), 863 High-Tech R&D (863
Project, Grant No. 2004AA215250), and the Natural Science
Foundation of China (20174013).
Supporting Information Available: The details of kinetics
1
of 2-TeCD, UV spectral titrations, fluorescence spectrum, H
0
.03].
and 2D NMR measurements, and molecular simulation. This
material is available free of charge via the Internet at
http://pubs.acs.org.
In the enzymatic reactions, the following NADPH consump-
tions corrected for the background reactions were obtained:
catalyst [concentration of catalyst (µM), ROOH, NADPH
consumption (µM/min) ( standard deviation], Ebselen [5, H2O2,
JA045964V
7
[
0
.28 ( 0.43], PhSeSePh [10, H2O2, 30.33 ( 1.69], PhTeTePh
10, H2O2, 27.16 ( 1.13], â-cyclodextrin [100, H2O2, 0.16 (
.02], 2-TeCD [1, H2O2, 72.40 ( 3.13; t-BuOOH, 98.66 ( 4.26;
CuOOH, 135.33 ( 3.82].
(
36) MIS.; Molecular Simulation, Inc.; Currently Accelrys Inc.; San Diego, CA,
1999.
(37) (a) Force Field-Based Simulations, General Theory and Methodology, April,
1997. (b) Comba, P.; Hambley, T. W. Molecular Modeling, Weinheim:
New York, 1995.
16404 J. AM. CHEM. SOC.
9
VOL. 126, NO. 50, 2004