Construction of a smart glutathione peroxidase mimic with temperature responsive activity based on block copolymer

Article abstract of DOI:10.1039/c0sm01081b

To construct a smart artificial antioxidative enzyme on a nano-scaffold, a novel method for designing glutathione peroxidase (GPx) active sites on block copolymer vesicles was developed by simple blending predesigned temperature-sensitive block copolymers with the main catalytic units of GPx. A series of functional block copolymers, poly(N-isopropylacrylamide)-b- polyacrylamides loaded with recognition and catalytic sites, were synthesized via ATRP and click chemistry. Through altering the molar ratio of the functional copolymers, the optimum GPx mimic based on copolymer vesicles was obtained by self-assembly of temperature-sensitive block copolymers through a blending process. Significantly, the catalytic activity of the optimum GPx mimic can be well modulated by changing the temperature. It was proved that the change in self-assembly structure of the block copolymer played an important role in the modulation of the catalytic activity. This method not only bodes well for designing smart antioxidative enzyme mimics that could be used in cosmetics as antioxidative additives but also highlights the construction of other regulatory biologically related functional biomaterials.

Full text of DOI:10.1039/c0sm01081b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Soft Matter
Cite this: Soft Matter, 2011, 7, 2521
www.rsc.org/softmatter
PAPER
Construction of a smart glutathione peroxidase mimic with temperature
responsive activity based on block copolymer†
Yanzhen Yin, Liang Wang, Haiyan Jin, Chunyan Lv, Shuangjiang Yu, Xin Huang, Quan Luo, Jiayun Xu
*
and Junqiu Liu
Received 30th September 2010, Accepted 6th December 2010
DOI: 10.1039/c0sm01081b
To construct a smart artificial antioxidative enzyme on a nano-scaffold, a novel method for designing
glutathione peroxidase (GPx) active sites on block copolymer vesicles was developed by simple
blending predesigned temperature-sensitive block copolymers with the main catalytic units of GPx. A
series of functional block copolymers, poly(N-isopropylacrylamide)-b-polyacrylamides loaded with
recognition and catalytic sites, were synthesized via ATRP and click chemistry. Through altering the
molar ratio of the functional copolymers, the optimum GPx mimic based on copolymer vesicles was
obtained by self-assembly of temperature-sensitive block copolymers through a blending process.
Significantly, the catalytic activity of the optimum GPx mimic can be well modulated by changing the
temperature. It was proved that the change in self-assembly structure of the block copolymer played an
important role in the modulation of the catalytic activity. This method not only bodes well for
designing smart antioxidative enzyme mimics that could be used in cosmetics as antioxidative additives
but also highlights the construction of other regulatory biologically related functional biomaterials.
in which a catalytic center based on selenium/tellurium was
Introduction
introduced into existing or artificially generated substrate
binding scaffolds by chemical or genetic strategies.11–22
Reactive oxygen species (ROS), like hydrogen peroxides, organic
hydroperoxides and phospholipid hydroperoxides, are generated
as byproducts of cellular metabolism. Usually, many human
oxidative stress-related diseases such as reperfusion injury,
inflammatory process, neuronal apoptosis and cancer are the
result of the overproduction of ROS. Generally, such oxidative
stress-related diseases are controlled by the antioxidative defense
system, especially by the antioxidative enzyme system.1,2 As
a member of the family of antioxidative enzymes, glutathione
peroxidase (GPx, Ec.1.11.1.9), is an important selenium-con-
taining enzyme which functions to protect various living
organism from aerobic oxidative stresses by catalyzing the
reduction of hydroperoxides (ROOHs) using glutathione (GSH)
as a reducing substrate.3 The catalytic moiety of GPx has been
well studied.3,4 And three important catalytic factors contribute
mainly to maintaining efficient GPx catalytic activity: the cata-
lytic selenium center, the recognition site, and the hydrophobic
cavity. Owing to its biologically crucial role, considerable efforts
have been devoted to produce organoselenium/tellurium
compounds that mimic the properties of GPx in recent years.5–10
Recently, in our group, based on the understanding of the
structure of GPx, some artificial GPx models have been designed
Although many human oxidative stress-related diseases are the
result of the overproduction of ROS, it is well known that ROS
are normally friendly to humans and play a role as signal mole-
cules in the metabolism. Commonly, ROS are continuously
produced and they keep balanced under the strict control of enzymes
and antioxidants within the cells. They will not result in various
illnesses unless excess ROS are produced in the human body.23 So
a desirable GPx mimic should be endowed with some smart char-
acteristics: it should not remove the ROS completely and its catalytic
efficiency could be regulated by some environmental stimuli, such as
temperature, pH, ion strength etc. This smart GPx mimic is possible
to control the concentration of ROS to an appropriate degree, which
means that ROS can both play an important role in the metabolism
and not result in various stress-related illnesses. Our previous work
suggests that either the microgel GPx mimic16 or the block copol-
ymer GPx mimic13,17, based on thermally sensitive PNIPAM,
display tunable catalytic activity. The catalytic activity of the smart
GPx mimics can be turned reversibly by changing the temperature.
However, these GPx mimics were just endowed with a catalytic
center, which is only one of the three catalytic factors employed to
maintain the high catalytic activity. Consequently, how to combine
three catalytic factors into one smart GPx mimic to obtain the
optimum GPx mimic is still a great challenge.
State Key Laboratory of Supramolecular Structure and Materials, College
of Chemistry, Jilin University, Changchun, 130012, People’s Republic of
China. E-mail: junqiuliu@jlu.edu.cn; Fax: +86–431–85193421
† Electronic Supplementary Information (ESI) available: Synthetic
procedures. See DOI: 10.1039/c0sm01081b/
Although GPx mimics constructed previous have demon-
strated high catalytic activity, some disadvantages still remain in
such efficient GPx models. Firstly, considering that the
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 2521–2529 | 2521

View Article Online
limitations of a complicated synthetic route, high cost, low
productivity exist in the preparation of the GPx models, the
development of a simple and efficient method to construct GPx
mimics is still a significant goal. Secondly, as the molar ratio of
the catalytic factors is fixed in one scaffold, the achievement of
the optimum GPx mimic via altering the molar ratio of the
catalytic factors is more difficult and the catalytic activity can not
be easily controlled through changing the composition of the
catalytic factors. To resolve these problems, a blending process,
as a versatile strategy exploited for the development of new
polymeric materials with property profiles superior to those of
the individual components24–33 was employed to construct
blended GPx mimics in our previous work.19 The blended GPx
mimic designed via this versatile strategy not only combines the
three catalytic factors into one scaffold simply, but also exhibits
high catalytic activity by easily altering the molar ratio of the
functional copolymers. Subsequently, if the blending process can
be applied to construct a GPx mimic using smart materials, it will
be easier to meet the challenge of achieving the optimum GPx
mimic with the combination of the three catalytic factors into one
GPx model. Meanwhile, this GPx mimic can be endowed with
the advantages of simple and efficient preparation and high
catalytic activity obtained by altering the molar ratio of catalytic
factors. Significantly, the blended GPx mimic based on the smart
PNIPAM polymer could potentially be used in some aspects such
as cosmetics as antioxidative additive. Additionally, efforts have
be made in exploring the smart blended GPx mimic based on
a smart PNIPAM polymer for constructing the smart biode-
gradable GPx mimic which could potentially be used in medical
applications. Now, the construction of the smart biodegradable
GPx mimic based on PNIPAM and biodegradable polymers is
being researched.
Lancaster. Triethylamine and tetrahydrofuran were rigorously
dried with sodium.
Instruments
The characterization of the structure of the compounds was
performed with a Bruker 300MHz spectrometer using a TMS
proton signal as the internal standard. UV-vis spectra were
obtained using a Shimadzu 2450 UV-vis-NIR spectrophotom-
eter. Scanning electron microscopy (SEM) observations were
carried out on a JEOL JSM-6700F scanning electron microscope
with primary electron energy of 3 kV. Transmission electron
microscopy (TEM) observations were carried out on a JEOL
JEM 3010 scanning electron microscope. The buffer pH values
were determined with a METTLER TOLEDO 320 pH meter.
Static light scattering experiments were performed at Malven
ZETAS12-ERNANOSERIES instrument. Molecular weights
and molecular weight distributions were determined by GPC
using THF as the eluent at a flow rate of 1.0 mL minꢀ1
.
Determination of GPx activity17
The catalytic activity was assayed according to a modified
method reported by Hilvert et al.35 The reaction was carried out
ꢁ
at 25 C in a 1 mL quartz cuvette, 700 mL of phosphate buffer
(pH ¼ 7.0. 50 mM) and 100 mL of the catalyst (1 mM) were added,
and then 100 mL of the TNB (3-carboxyl-4-nitrobenzenethiol)
solution (1.5 mM) was added. The mixture in the quartz cuvette
ꢁ
was pre-incubated at the 25 C for 3 min. Finally, the reaction
was initiated by the addition of 100 mL of cumene hydroperoxide
(CUOOH) (2.5 mM), and the absorption decrease of TNB at 410
nm (3410 ¼ 13600 Mꢀ1 cmꢀ1, pH ¼ 7.0) was monitored using
a Shimadzu 2450 UV-vis-NIR spectrophotometer. Appropriate
control of the non-enzymatic reaction was performed and was
subtracted from the catalyzed reaction.
To the best of our knowledge, this is the first example of the
construction of a smart GPx mimic using a blending process.
Herein, based on a series of functional block copolymers, the
smart GPx mimic was constructed by a blending process. The
self-assembly behavior of the blended GPx mimic was observed
above the LCST of the functional block copolymers by static
light scattering, SEM and TEM, and hollow vesicles were
obtained. The optimum blended GPx mimic was obtained by
altering the molar ratio of the functional block copolymers. And
the catalytic activity of the optimum GPx mimic can be easily
modulated by changing the temperature.
Determination of the content of the catalytic center tellurium in
PPAM–Te
Hydrogen peroxide (H2O2) was added to the solution of PPAM–
Te, and after stirring for 10 h at room temperature, the solution
was dialyzed against water for 1 day to remove H2O2. The solid
PPAM–Te polymer in which Te was in the form of Te(IV) was
obtained via freeze-drying. According to the report by Detty
et al.36 one molecule oxidized state Te(IV) can be reduced by two
molecules of TNB, thus the content of catalytic center tellurium
was obtained by monitoring the decrease in UV absorption of
TNB at 410 nm.
Materials and methods
Materials
The functional monomer 1 and monomer 2 were synthesized
as described previously.12
Tris(2-aminoethyl)amine (TREN, Acros) was used as received.
Tris(2-dimethylaminoethyl)amine (Me6TREN) was synthesized
as described previously.34 N-Isopropylacrylamide (NIPAM)
(Aldrich) was recrystallized from hexane and toluene, and dried
under vacuum prior to use. Sodium borohydride, and 3-bromo-
1-propanol were purchased from Fluka and were used without
further purification. Acrylamide, tellurium powder, b-cyclodex-
trin, phenyl methanol and 4-toluene sulfonyl chloride were
purchased from the Shanghai Reagent Co. Acryloyl chloride and
propargyl alcohol were purchased from the Anhui Wotu
Reagent Co. 2-Bromopropanoly bromide was purchased from
And the functional monomer 3 was synthesized according to
the previous procedure.19
Synthesis of benzyl 2-bromopropanoate
2-Bromopropanoyl bromide (2.06 mL, 0.020 mol) was dissolved
in 20 mL anhydrous tetrahydrofuran and added dropwise to
a stirred solution of phenyl methanol (2.09 mL, 0.020 mol) and
triethylamine (2.77 mL, 0.020 mol) in anhydrous tetrahydro-
furan (60 mL) at 0 ꢁC. After complete addition, the mixture was
2522 | Soft Matter, 2011, 7, 2521–2529
This journal is ª The Royal Society of Chemistry 2011

View Article Online
stirred for 3 h at room temperature. The precipitate triethylamine
hydrobromide was filtered and the filtrate was concentrated
under vacuum. The product was purified by silica gel flash
chromatography (petroleum ether/ethyl acetate, 40 : 1). After the
removal of the solvents, the product was isolated as a colorless oil
(yield, 81%).
1H NMR (300 MHz, CDCl3) d (ppm): 7.37 (s, 5H, –C6H5), 5.21
(s, 2H, –CH2C6H5), 4.41 (q, 1H, –CH(CH3)Br), 1.84 (d, 3H, –CH3).
Synthesis of PNIPAM–Br
PNIPAM–Br was synthesized according to the polymerization
procedure reported by Masci et al.37 The typical procedure was
as follows: NIPAM (2260 mg, 20 mmol), Me6TREN (115.1 mg,
0.5 mmol), water (5.0 mL) and DMF (5.0 mL) were introduced
into a schlenk tube equipped with a magnetic stirring bar, fol-
lowed by two freeze-pump-thaw cycles. Then, CuBr (71.7 mg, 0.5
mmol) was added under nitrogen and followed by two freeze-
pump-thaw cycles. Finally, the initiator benzyl-2-bromopropa-
noate (60.78 mg, 0.25 mmol) was added to the schlenk tube via
a syringe to start the polymerization. The mixture was stirred for
10 h at room temperature. Then the mixture was exposed to air to
terminate the polymerization and was dialyzed against water for
2 d, a colorless polymer was obtained after the aqueous solution
was freeze-dried. GPC analysis of the polymer revealed a Mn of
8036 and a polydispersity, Mw/Mn, of 1.18. The degree of poly-
merization (DP) of PNIPAM was estimated to be 69. The mac-
roinitiator was denoted as PNIPAM–Br.
Fig. 1 The structure of the functional monomers 1–3 and substrates
(NBT, TNB).
ATRP, which is one of the significant controlled/living radical
polymerizations employed to design versatile functional mate-
rials and polymers with complex architectures and composi-
tions.38–44 Recently, the smart materials which respond to
environmental stimuli have been well developed; thermally
sensitive polymers in particular have a broad application in drug
delivery, optical filtering and controlled biomolecule recovery.45–49
The thermally sensitive PNIPAM attracts particular attention
since it will undergo a reversible volume phase transition at near-
physiological temperatures.50–52 Due to this characteristic of
PNIPAM, some smart functional block copolymers incorpo-
rated with catalytic factors were synthesized via ATRP. By
combining cyclodextrins and smart polymers such as PNIPAM
via ATRP and click chemistry, new materials can be prepared.
Ritter et al. synthesized supramolecular CD containing copoly-
mers54 and supramolecular polymers using cyclodextrin–cucur-
bit[6]uril (CD-click-CB)55 and calixarene-click-cyclodextrin56 as
combi-receptors. And Liu et al. synthesized stimuli-responsive
watersoluble Janus-type heteroarm star copolymers using ATRP
and click chemistry.57 Recently, in our group, a modular
bifunctional artificial enzyme with both superoxide dismutase
(SOD) and GPx activities was constructed via ATRP and click
chemistry.13 According to the synthetic routes reported in such
previous research, the functional polymers were prepared and the
synthetic routes of macroinitiator PNIPAM–Br and PPAM,
PPAM–Te, PPAM–N, and PPAM–CD are given in Scheme 1
and Scheme 2. The characterizations of these block copolymers
and the contents of catalytic factors in the corresponding
copolymers are given in the ESI.†
The preparation procedures of the functional block copoly-
mers, PPAM, PPAM–Te, PPAM–N, PPAM–CD were the same
as that of PNIPAM–Br except that PNIPAM–Br was used as the
initiator and the corresponding functional monomer was added.
The detailed preparation procedures and the characterizations of
the functional block copolymers were given in the ESI.†
Results and discussion
Preparation and characterization of the smart functional block
copolymers
The crystal structure of bovine erythrocyte glutathione peroxi-
dase was reported by Epp et al. in 1983.4 The catalytic active site
of GPx has been well studied. It was indicated that three
important catalytic factors contribute to maintaining efficient
GPx catalytic activity: the catalytic selenium center, the recog-
nition site, and the hydrophobic cavity. Furthermore, based on
our previous studies, tellurium, a quaternary ammonium salt,
and cyclodextrin were selected as efficient and novel alternative
catalytic factors to the selenium moiety, arginines in the binding
sites, and the hydrophobic cavity, respectively. Herein, based on
previous studies, the corresponding functional monomers 1–3
were synthesized (Fig. 1). Monomer 1, with a similar function of
selenocysteine, was prepared as a catalytic center. Monomer 2
was responsible for the complexation of carboxyl groups of the
substrates. And monomer 3 was included to provide a hydro-
phobic microenvironment being similar to the hydrophobic
cavity in the native GPx.
As a thermally sensitive polymer, PNIPAM with a lower
critical solution temperature (LCST) of 32 ꢁC undergoes
a reversible volume phase transition at near-physiological
temperatures. Typically, a great deal of hydrogen bonds are
Subsequently, some thermally sensitive functional block
copolymers modified with the three catalytic factors: catalytic
center, binding site, hydrophobic cavity, were synthesized via
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 2521–2529 | 2523

View Article Online
Scheme 1 Synthesis of PNIPAM macroinitiator PNIPAM–Br
Fig. 2 Temperature dependence of optical transmittance at 600 nm
obtained for pH 7.0, 50 mM PBS of (a) PPAM–Te, (b) PPAM–N, (c)
PPAM–CD and (d) PPAM–N–CD–Temax at concentrations of 1 mg
formed between the amide groups and surrounding water
molecules below the LCST, but above the LCST the cleavage of
the hydrogen bonds result in the PNIAPAM block shifting from
being hydrophilic to hydrophobic. This phenomenon is also
observed in the smart functional block copolymers in this work.
Fig. 2 shows the temperature dependence of the optical trans-
mittance at 600 nm obtained in pH 7.0, 50 mM PBS for func-
tional block copolymers at concentrations of 1 mg mLꢀ1. The
LCSTs of PPAM–Te, PPAM–N, PPAM–CD are 32.6, 33.0, 33.3
ꢁC respectively. Also, the LCST of the optimum blended GPx
mLꢀ1
.
Fig.
3 Determination of GPx activity using 3-carboxyl-4-nitro-
benzenethiol (TNB) as a substrate.
ꢁ
mimic PPAM–N–CD–Temax is 33.1 C. We can clearly see that
the optical transmittance of either the individual functional block
copolymer or the optimum blended GPx mimic trend to have
a fixed value when the temperature above 36 ꢁC. Accordingly, the
thermally sensitive copolymer can self-assemble to form vesicles
when the hydrophilic PNIPAM block becomes hydrophobic
with increasing temperature. Considering that the self-assembled
structure largely affects the catalytic activity of a GPx mimic, the
catalytic activities of optimization of the blended GPx mimics
have been deꢁtermined when the stable self-assembled vesicles
formed at 36 C.
Optimizing the structure of the smart blended GPx mimic
The catalytic activity of the smart blended GPx mimic for the
reduction of cumene hydroperoxide (CUOOH) by 3-carboxyl-4-
nitrobenzenethiol (TNB) was evaluated according to the
modified method reported by Hilvert et al.35 using TNB as
a glutathione (GSH) alternative (see Fig. 3). The activity was
given assuming one molecule catalytic center (Te–monomer) in
the enzyme model as one active site of the enzyme. Reaction was
initiated by the subsequent addition of hydroperoxide, and the
absorbance at 410 nm (3410 ¼ 13600 Mꢀ1 cmꢀ1, pH ¼ 7.0) was
recorded for a few minutes to calculate the reaction rate. The
relative activities are summarized in Table 1. By altering the
molar ratio of PPAM–Te, PPAM–N, PPAM–CD, the optimum
GPx mimic PPAM–N–CD–Temax was obtained by a blending
process. Considering that the self-assembly structure could affect
the catalytic activity of the blended GPx mimic, the detailed
structure of the optimum GPx mimic PPAM–N–CD–Temax was
investigated by static light scattering, SEM and TEM. As shown
in Fig. 4, the hydrodynamic diameters of PPAM–N–CD–Temax
in aqueous solution are typically 8 nm when the temperature is
Table 1 The initial rates (v0) and activities for the reduction of hydro-
peroxides (250 mM) by TNB (150 mM) in the presence of the GPx mimics
at pH 7.0 (50 mM PBS) and 36 ꢁC
a
v0/mM minꢀ1
CUOOH
Catalysts
Thiol
PPAM–Te
TNB
TNB
TNB
TNB
2.29
4.94
7.01
13.19
PPAM–N–Temax
PPAM–CD–Temax
PPAM–N–CD-Temax
a
the initial rate of reaction was corrected for spontaneous oxidation. The
concentration of catalyst is 1 ꢂ 10ꢀ3 mM, assuming a one molecule
catalytic center (tellurium moiety) is equivalent to one active site of
enzyme.
Scheme 2 Synthesis of functional poly(N-isopropylacrylamide)-b-poly-
acrylamide (PPAM, PPAM–Te, PPAM–N, PPAM–CD) modified with
catalytic factors.
2524 | Soft Matter, 2011, 7, 2521–2529
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Fig. 4 Hydrodynamic diameters of PPAM–N–CD–Temax at varying
temperature (a, 25 ꢁC; b, 32 ꢁC; c, 36 ꢁC; d, 40 ꢁC) determined using
a Malvern ZETAS12-ERNANOSERIES instrument.
ꢁ
lower than the LCST (curve a, 25 ꢁC; curve b, 32 C). Never-
theless, the corresponding hydrodynamic diameters shift to
larger than 100 nm when the temperature is higher than the
LCST (curve c 36 ꢁC; curve d, 40 ꢁC). The difference in the
diameters at the corresponding temperature is endowed from
the different assembly behavior of PPAM–N–CD–Temax when
the hydrophilic PNIPAM block becomes hydrophobic with
increasing temperature. Furthermore, the actual morphology of
the optimum GPx mimic when the temperature is higher than the
LCST of PPAM–N–CD–Temax was observed by SEM. From the
SEM image in Fig. 5, we can clearly see that the presence of
spherical nanoparticles of 90 nm in diameter. As the SEM shows,
the dimensions of the nanoparticles in the dry state are smaller
than those detected by a zetasizer nano instrument, since the
zetasizer nano instrument reports the average hydrophobic
diameter of nanoparticles in solution which includes the contri-
bution from the swollen corona. To further investigate the
structure of the spherical nanoparticles, their detailed
morphology was observed by TEM (see Fig. 6). From the image
obtained by TEM, we can see that the nanopariticles are hollow
vesicle-like nanoparticles with a thinner region surrounded by
a thicker region and clearly confirm the boundary between the
thinner and the thicker ones. Similar behavior has been reported
by Chen and coworkers.53 Consequently, based on the above
measurements, we can draw a conclusion that the blended GPx
mimic is soluble in water when the temperature lower than LCST
of PPAM–N–CD–Temax, and it can self-assemble to from hollow
vesicles when the temperature is higher than LCST of PPAM–N–
CD–Temax. A graphical representation of the GPx mimic con-
structed by the blending process is displayed in Scheme 3.
Subsequently, the optimization of the blended GPx mimic
constructed by the functional copolymers with quaternary
ammonium salt or cyclodextrin was accomplished. Considering
that the catalytic activity is crucially important for the success of
Fig. 6 TEM image for PPAM–N–CD–Temax
.
a GPx mimic, we chose the data of the catalytic activity as an
effective value to reflect the optimization of the GPx mimic by
the blending process. On the one hand, the blended GPx mimic,
PPAM–N–Te, was constructed from PPAM–Te modified with
a catalytic center and PPAM–N modified with a quaternary
ammonium salt using the blending process. By plotting the
catalytic reaction rate against the molar ratio of PPAM–N to
PPAM–Te as shown in Fig. 7A, we can note that the catalytic
reaction rate of PPAM–N–Te increases to some extent with the
increase in molar ratio. It reached the highest value (PPAM–N–
Temax ¼ 4.94 mM minꢀ1) when the molar ratio is 5 : 1. However,
the catalytic reaction rate goes down when the molar ratio
increases further. We speculate that the reason for this interesting
phenomenon is that a slight change of the structure would result
in dramatic change in catalytic activity for a native enzyme. So
for this smart blended GPx mimic, a better match between the
catalytic center tellurium and quaternary ammonium salt would
exhibit strong substrate binding ability and result in the higher
catalytic activity. Thus, the GPx mimic exhibits the highest
catalytic activity when the best match is achieved with the molar
ratio is 5 : 1 However, the catalytic reaction rate goes down when
the molar ratio increases further despite the binding ability being
enhanced, which proves that stronger binding does not imply
higher catalytic activity. The possible reason for this is that the
binding ability endowed from the quaternary ammonium salt
increases to the higher degree when the molar ratio is high, which
could bind many more substrates in a distribution away from the
catalytic center and make the completion of the catalytic cycle
inefficient. Similar results were also reported in our previous
work.13,14 On the other hand, using the same protocol as
Scheme 3 A graphical representation of the GPx mimic constructed by
Fig. 5 SEM image for PPAM-N-CD-Temax
.
a blend process
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 2521–2529 | 2525

View Article Online
Fig. 7 Plots of catalytic reaction rates v0 against molar ratio of func-
tional block copolymer: (A), PPAM–N to PPAM–Te; (B) PPAM–CD to
PPAM–Te.
Fig. 9 Plots of initial rates at different concentrations of CUOOH in the
presence of the PPAM–N–CD–Temax. The initial concentration of TNB
was fixed at 0.15 mM, The concentrations of CUOOH used were 0.05,
0.10, 0.25, 0.5, 1, 2.5 and 5mM.
mentioned above, the blended GPx mimic, PPAM–CD–Te, was
constructed from PPAM–Te and PPAM–CD modified with
catalytic center and cyclodextrin (Fig. 7B). The blended GPx
mimic exhibits the highest catalytic reaction rate (PPAM–CD–
Temax ¼ 7.01 mM minꢀ1) when the molar ratio of PPAM–Te to
PPAM–CD is 6 : 1.
rate enhancement (PPAM–N–CD–Temax ¼ 13.19 mM minꢀ1) was
observed under identical conditions when the optimum GPx
mimic was added. Seen from Fig. 9, the saturation kinetics of the
optimum GPx mimic for the peroxidase reaction is studied at the
individual concentrations of CUOOH, which indicates that
PPAM–N–CD–Temax exhibits typical saturation kinetics and
acts as a real catalyst for the peroxidase reaction. In the TNB
assay system, the apparent kinetic parameters were obtained
Finally, the optimum blended GPx mimic was constructed
using the blending process of three block copolymers, PPAM–
Te, PPAM–N, and PPAM–CD. To obtain the optimum blended
GPx mimic, PPAM–N–CD–Te, we keep the molar ratio of
PPAM–N to PPAM–Te constant (5 : 1) and change the molar
ratio of PPAM–CD to PPAM–Te, and curve a in Fig. 8 was
obtained. As the presence of three catalytic factors in PPAM–N–
CD–Te and the good match among them, the optimum blended
GPx mimic (PPAM–N–CD–Temax ¼ 13.19 mM minꢀ1) with the
highest catalytic reaction rate was achieved successfully when the
molar ratio is 5 : 6 : 1. Although the three catalytic factors in
the vesicles of the blended GPx mimic distribute in a randomly
manner, the good match of the catalytic factors still could be
achieved by changing the molar ratio of the three functional
copolymers. The catalytic behavior of the optimum GPx mimic
has directly confirmed this conclusion (Fig. 8, curve a). Now, we
can draw a conclusion that the construction of a simple and
efficient smart GPx mimic has been achieved through altering the
molar ratio of three functional copolymers by the blending
process based on the experiments mentioned above.
Vmax ¼ 49.82 mM minꢀ1, kcat ¼ 49.82 minꢀ1, KmCUOOH ¼ 508.2
app
mM, kcatapp/KmCUOOH ¼ 9.80 ꢂ 104 Mꢀ1 minꢀ1, and the turnover
number per catalytic tellurium center was calculated to be 50
minꢀ1
.
Generally, the recognition of substrates is one the most
important characteristics for an efficient enzyme mimic. For
native GPx, the binding site formed between two arginines and
carboxylic groups of GSH and the hydrophobic cavity composed
of some hydrophobic amino residues play an important role in
binding substrates and maintaining a high catalytic activity. For
the blended GPx mimic, a similar binding site and hydrophobic
cavity were incorporated, which was endowed with quaternary
ammonium salt being analogous to the binding site, and cyclo-
dextrin which can effectively recognize aromatic thiol substrates
being similar to the hydrophobic cavity in native GPx. Conse-
quently, both the catalytic tellurium centers and the recognition
ability derived from the binding sites and the hydrophobic
cavities play an important role in maintaining such high catalytic
activity.
Catalytic behavior of the optimum smart GPx mimic
To confirm the interaction between the substrate TNB and the
blended GPx mimic, UV spectra were employed to check the
substrate binding ability. It is known that the maximum absor-
bance wavelength of TNB is 410 nm in aqueous solution (pH ¼
7.0), and it will appear as a red shift when it is in a hydrophobic
The catalytic activity of the optimum GPx mimic, PPAM–N–
CD–Temax, was studied using TNB as a GSH alternative. Nor-
mally, the reaction between CUOOH and TNB is very slow
under spontaneous oxidation conditions. However, a remarkable
Fig. 8 Plots of catalytic reaction rates v0 against molar ratio of func-
tional copolymer. (a) PPAM–N–CD–Te: ratio of PPAM–CD to PPAM–
Te (the ratio of PPAM–N to PPAM–Te was fixed to 5 : 1); (b) PPAM–
AM–Te: the condition is as same as (a) except that PPAM–N and
PPAM–CD are replaced by PPAM.
Fig. 10 (A): UV spectra with varying concentration of PPAM–N–CD–
Temax in water; the concentration of TNB is 88 mM, and from a to e the
concentrations of PPAM–N–CD–Temax are 0, 0.2, 0.4, 1, and 1.5 mM,
respectively. (B) PPAM–AM–Te: the conditions are as same as in (A)
except PPAM–N and PPAM–CD are replaced by PPAM.
2526 | Soft Matter, 2011, 7, 2521–2529
This journal is ª The Royal Society of Chemistry 2011

View Article Online
hydrophobic interaction equally existed between TNB and
PPAM–AM–Te. The possible reason is that the hydrophobic
microenvironment endowed from the hydrophobic PNIPAM
block in self-assembled vesicles, when the temperature is higher
than the LCST of PNIAM it can bind the substrate TNB to some
degree. Although PPAM–AM–Te has some substrate binding
ability, the catalytic activity of it is striking lower than that of
PPAM–N–CD–Temax under various molar ratios of the func-
tional copolymers as shown in Fig. 8. The absence of catalytic
factors in PPAM–AM–Te and the better match among the
catalytic factors in PPAM–N–CD–Temax possibly give rise to the
remarkable difference of the two blended GPx mimics. In short,
the above evidences draw a conclusion that not only the strong
substrate binding but also the good match among the catalytic
factors play important roles in constructing an efficient GPx
mimic. Successfully, PPAM–N–CD–Temax, constructed by the
blending process in this study, exhibits both strong substrate
binding ability and high catalytic activity.
Fig. 11 The initial rates (v0) and activities for the reduction of hydro-
peroxides (250 mM) by thiol TNB and NBT (150 mM) in the presence of
the PPAM–N–CD–Temax at pH 7.0 (50 mM PBS) and 36 ꢁC. (A)
CUOOH, TNB; (B) H2O2, TNB; (C) CUOOH, NBT; (D)H2O2, NBT.
microenvironment. Fig. 10A shows the UV spectra of the
assayed mixture of various amounts of PPAM–N–CD–Temax. As
expected, the presence of a red shift of the maximum absorbance
wavelength indicated that a strong hydrophobic interaction
existed between the aryl moiety of TNB and the hydrophobic
cavity of PPAM–N–CD–Temax. To further confirm the substrate
binding ability of PPAM–N–CD–Temax, the catalytic reaction
rate was measured in different assay systems (shown in Fig. 11):
Catalytic activity modulated by temperature
To investigate the temperature responsive property of PPAM–
N–CD–Temax, the catalytic activity was determined in the TNB
assay system using CUOOH as a substrate at various tempera-
tures from 20 ꢁC to 50 ꢁC. To our knowledge, for the majority of
temperature-activated reactions, the reaction rates will enhance
with rising temperature according to the Arrhenius equation.
However, it is different for the blended GPx mimic, and a ther-
mally-responsive catalytic activity curve was obtained by plot-
ting the catalytic reaction rate against temperature (Fig. 12). It
was observed that the catalytic activity increases slowly with
increasing temperature when temperature is lower than 30 ꢁC,
and it increases largely with increasing temperature from 30 ꢁC to
36 ꢁC. Furthermore, when the temperature increases above 36
ꢁC, the catalytic activity increases slowly again. We speculate
that the temperature responsive behavior of PPAM–N–CD–
Temax is endowed by the change of the self-assembled structure of
the optimum blended GPx mimic with increasing temperature.
To confirm this, the hydrodynamic diameters of the self-assem-
bled structure of PPAM–N–CD–Temax were assayed at various
temperatures and the temperature dependences of the optical
transmittance of PPAM–N–CD–Temax was measured with the
increasing temperature.
CUOOH, TNB assay system (Fig. 11A), v0 ¼ 13.19 mM minꢀ1
H2O2, TNB assay system (Fig. 11B), v0 ¼ 3.42 mM minꢀ1
CUOOH, NBT assay system (Fig. 11C), v0 ¼ 2.54 mM minꢀ1
;
;
;
H2O2, NBT assay system (Fig. 11D), v0 ¼ 1.57 mM minꢀ1. We
could find that much lower catalytic reaction was observed when
NBT was used as the substrate (C < A; D < B). Considering that
the same GPx mimic was used in the two assay systems, the
dramatic change in the catalytic activity should be mainly
endowed from the electrostatic interaction, which was formed
between the carboxyl in TNB and the quaternary ammonium salt
in the blended GPx mimic when the TNB assay system was used.
Furthermore, the catalytic reaction rates were measured under
the same conditions when ROOH was changed. The rate
constants of the spontaneous reaction between hydroperoxide
and thiol vary in magnitude in the order k(H2O2)
>
k(CUOOH).11 We can note that the catalytic activity of PPAM–
N–CD–Temax has a significant enhancement in the TNB assay
system when CUOOH is used as a substrate instead of H2O2 by
comparing the assay system A and B. The difference in the
catalytic activity when hydrophobic CUOOH as an alternative
substrate is mainly because the hydrophobic microenvironment
endowed from the blended GPx mimic makes the hydrophobic
substrate CUOOH approach the GPx mimic more easily and
completes the catalytic cycle preferentially. Consequently, we can
note that the optimum blended GPx mimic has strong substrate
binding ability through both electrostatic interactions and
hydrophobic interactions based the above experiments.
As a thermal sensitive polymer, PNIAPM exhibits the lower
ꢁ
critical solution temperature (LCST) of 32 C. A great deal of
hydrogen bonds were formed between the amide groups and
However, the strong substrate binding ability does not imply
better catalytic activity. Herein, a contrast experiment was
carried out to confirm this conclusion. Another blended GPx
mimic, PPAM–AM–Te, was constructed using the same method
when the functional block copolymers modified with catalytic
factors were replaced by block copolymer modified with no
catalytic factors. Similarly, the interaction between substrate
TNB and PPAM–AM–Te was also determined by UV spec-
troscopy (Fig. 10B). The presence of a slight red shift of the
maximum absorbance wavelength also implied that the
Fig. 12 Plots of the catalytic rate of PPAM–N–CD–Temax versus
temperature during the catalytic reduction of CUOOH (0.25 mM) by
TNB (0.15 mM) and catalytic center (1.00 mM).
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 2521–2529 | 2527

View Article Online
surrounding water molecules when the temperature is lower than
the LCST, which makes the PNIPAM block totally dissolve in
water. However, when the temperature is higher than the LCST,
the hydrogen bonds between the amide groups and surrounding
water molecules begin to cleave, which results in the hydrophilic
PNIAM block becoming a hydrophobic block. And the self-
assembled structure was observed as the hydrophobic PNIAM
block aggregating together in water above LCST. This
phenomenon is also observed in PPAM–N–CD–Temax, which
exhibits the LCST of 33.1 ꢁC. As shown in Fig. 2B, the
temperature dependence of the optical transmittance of PPAM–
N–CD–Temax was measured at various temperatures from 25 ꢁC
to 45 ꢁC (curve d). It was clearly that the optical transmittance of
PPAM–N–CD–Temax was nearly 100% below 33.1 ꢁC, which
indicated that PPAM–N–CD–Temax was dissolved in water and
the PNIAPM block was hydrophilic. However, the sharp
decrease in optical transmittance when the temperature is higher
than LCST of PPAM–N–CD–Temax was observed. With the
temperature increasing, the optical transmittance of PPAM–N–
CD–Temax became much lower, which indicated that the PNI-
PAM block of the optimum blended GPx mimic became
hydrophobic due to the cleavage of hydrogen bonds between the
amide groups and the surrounding water molecules. And the
shift of PNIAPM from hydrophilic to hydrophobic resulted in
the formation of the self-assembled structure. As shown in Fig. 4,
the measurement of the hydrodynamic diameters of PPAM–N–
CD–Temax further confirmed this conclusion. When the
temperature is lower than the LCST, the hydrodynamic diame-
ters of PPAM–N–CD–Temax are smaller than 10 nm, which
indicated that the PNIAPM block was hydrophilic and PPAM–
N–CD–Temax was dissolved in water. Nevertheless, the hydro-
dynamic diameters of PPAM–N–CD–Temax increase to larger
than 100 nm when the temperature is higher than the LCST,
which indicated that the self-assembled vesicles have formed with
the PNIPAM block of the optimum blended GPx mimic
changing from hydrophilic to hydrophobic. This observation is
in close agreement with the data on the temperature dependence
of the optical transmittance of PPAM–N–CD–Temax measured
with increasing temperature.
Fig. 13 Changes of the catalytic activity of the PPAM–N–CD–Temax
during heating-cooling cycles between 25 ꢁC and 43 ꢁC.
Considering that the vesicles have not be formed and a good
match between the catalytic factors has not been achieved when
the temperature is lower than 30 ꢁC, the catalytic activity
increases slowly with increasing temperature. However, when the
vesicles have formed and a better match between the catalytic
factors has been achieved with a temperature increase above the
LCST of the optimum blended GPx mimic, a large enhancement
of the catalytic activity has been observed with increasing the
temperature from 30 ꢁC to 36 ꢁC. The highest catalytic reaction
ꢁ
rate was obtained when the temperature was 43 C (v0 ¼ 14.07
mM minꢀ1).
Now we can draw a conclusion that the temperature respon-
sive catalytic behavior of PPAM–N–CD–Temax is endowed from
the change of the self-assembled structure of the blended GPx
mimic with increasing temperature.
Additionally, as a smart blended GPx mimic, the reversibility
of the catalytic behavior has been determined between 25 ꢁC and
43 ꢁC. As shown in Fig. 13, the catalytic activity of the optimum
blended GPx mimic,PPAM–N–CD–Temax, is still maintained
after several cycles of heating and cooling.
Conclusions
In this work, a simple blending process has been employed to
construct a smart GPx mimic based on block copolymer vesicles
carrying GPx active sites for the first time. As the functional
block copolymers loaded with recognition and catalytic sites,
PPAM, PPAM–Te, PPAM–N, PPAM–CD were synthesized via
ATRP and click chemistry. Using the blending process, the
optimum GPx mimic (PPAM–N–CD–Temax) was obtained by
the self-assembly of temperature-sensitive block copolymers
through altering the molar ratio of the functional copolymers.
Furthermore, the self-assembled structure of PPAM–N–CD–
Temax was investigated using static light scattering, SEM and
TEM and hollow vesicles were obtained when the temperature
was higher than the LCST of the optimum GPx mimic. By
exploring the catalytic behavior, it is noted that not only the
specific substrate binding ability but also the better match among
the catalytic factors play an important role in designing a desir-
able GPx mimic. Significantly, as a smart blended GPx mimic,
the catalytic activity of the optimum GPx mimic can be well
modulated by changing the temperature. It was proved that
a change in the self-assembly structure of the block copolymer at
different temperatures plays an important role in the modulation
of catalytic activity. We anticipate that this study will open up
a new field in designing a smart antioxidative artificial enzyme,
It is well known that a minor change of the structure of the
enzyme will result in a dramatic change in activity for naturally
occurring enzymes. Herein, for PPAM–N–CD–Temax, the cata-
lytic factors are incorporated into the PAM block of the block
copolymer. When the temperature is lower than the LCST, the
PNIPAM block is hydrophilic and the PPAM–N–CD–Temax was
dissolved in water. The catalytic factors of the functional block
copolymers are distributed in a randomly manner in water.
Consequently, lower catalytic activity may be obtained because
a good match of the catalytic factors could not easily be achieved.
When the temperature is higher than the LCST, the PNIPAM
block is hydrophobic and self-assembled vesicles are formed.
Consequently, the catalytic factors of the optimum blended GPx
mimic are concentrated in the vesicle, which results in the cata-
lytic factors in vesicle being much closer than before the vesicle
formed. Accordingly, a better match of the catalytic factors
could easily be achieved and made the completion of the catalytic
cycle efficient. A thermally-responsive catalytic activity curve
obtained by plotting the catalytic reaction rate against temper-
ature in Fig. 12 has confirmed the hypothesis mentioned above.
2528 | Soft Matter, 2011, 7, 2521–2529
This journal is ª The Royal Society of Chemistry 2011

View Article Online
25 L. Yu, K. Dean and L. Li, Prog. Polym. Sci., 2006, 31, 576.
26 M. M. Coleman and P. C. Painter, Prog. Polym. Sci., 1995, 20, 1.
27 E. S. Gil, D. J. Frankowski, M. K. Bowman, A. O. Gozen,
S. M. Hudson and R. J. Spontak, Biomacromolecules, 2006, 7, 728.
28 E. R. Kenawy, G. L. Bowlin and K. Mansfield, J. Controlled Release,
2002, 81, 57.
and we also hope this process could be used in the preparation of
other biologically related functional nanomaterials.
Acknowledgements
29 Y. He, B. Zhu and Y. Inoue, Prog. Polym. Sci., 2004, 29, 1021.
30 K. G. Marra, J. W. Szem, P. N. Kumta, P. A. DiMilla and
L. E. Weiss, J. Biomed. Mater. Res., 1999, 47, 324.
31 J. C. Meredith, A. Karim and E. J. Amis, Macromolecules, 2000, 33,
5760.
Financial support from the Natural Science Foundation of
China (No: 20874036, 20921003), the NSFC for Outstanding
Younger Scientist (No. 20725415), the 111 project (B06009) and
the National Basic Research Program (2007CB808006).
32 S. E. Shaheen, C. J. Brabec, N. S. Sariciftci, F. Padinger, T. Fromherz
and J. C. Hummelen, Appl. Phys. Lett., 2001, 78, 841.
33 W. Y. Chuang, T. H. Young, C. H. Yao and W. Y. Chiu,
Biomaterials, 1999, 20, 1479.
34 M. Ciampolini and N. Nardi, Inorg. Chem., 1966, 5, 41.
35 Z. P. Wu and D. Hilvert, J. Am. Chem. Soc., 1990, 112, 5647.
36 Y. You, K. Ahsan and M. R. Detty, J. Am. Chem. Soc., 2003, 125,
4918.
37 G. Masci, L. Giacomelli and V. Crescenzi, Macromol. Rapid
Commun., 2004, 25, 559.
38 V. Coessens, T. Pintauer and K. Matyjaszewski, Prog. Polym. Sci.,
2001, 26, 337.
Notes and references
1 H. Sies, ‘‘Oxidative Stress: Introductory Remarks’’, in Oxidative
Stress; Sies, H., Ed.; Academic Press:London, 1985, p. 1.
2 H. Sies, Angew. Chem., Int. Ed. Engl., 1986, 25, 1058.
ꢀ
€
3 L. Flohe, G. Loschen, W. A. Gunzler and E. Eichele, Hoppe-Seyler’s
Z. Physiol. Chem., 1972, 353, 987.
4 O. Epp, R. Ladenstein and A. Wendel, Eur. J. Biochem., 1983, 133, 51.
5 G. Mugesh and H. B. Singh, Acc. Chem. Res., 2002, 35, 226.
6 G. Mugesh and H. B. Singh, Chem. Soc. Rev., 2000, 29, 347.
7 T. G. Back and Z. Moussa, J. Am. Chem. Soc., 2003, 125, 13455.
8 Y. You, K. Ahsan and M. R. Detty, J. Am. Chem. Soc., 2003, 125, 4918.
9 M. McNaughton, L. Engman, A. Birmingham, G. Powis and
I. A. Cotgreave, J. Med. Chem., 2004, 47, 233.
10 G. Mugesh, W. W. du Mont and H. Sies, Chem. Rev., 2001, 101, 2125.
11 Z. Y. Dong, J. Q. Liu, S. Z. Mao, X. Huang, B. Yang, G. M. Luo and
J. C. Shen, J. Am. Chem. Soc., 2004, 126, 16395.
12 X. Huang, Z. Y. Dong, J. Q. Liu, S. Z. Mao, G. M. Luo and
J. C. Shen, Macromol. Rapid Commun., 2006, 27, 2101.
13 S. J. Yu, Y. Z. Yin, J. Y. Zhu, X. Huang, Q. Luo, J. Y. Xu, J. C. Shen
and J. Q. Liu, Soft Matter, 2010, 6, 5342.
14 X. Huang, Y. Liu, K. Liang, Y. Tang and J. Q. Liu,
Biomacromolecules, 2008, 9, 1467.
15 X. Huang, Y. Z. Yin, Y. Liu, X. L. Bai, Z. Z. Zhang, J. Y. Xu,
J. C. Shen and J. Q. Liu, Biosens. Bioelectron., 2009, 25, 657.
16 X. Huang, Y. Z. Yin, Y. Tang, X. L. Bai, Z. Z. Zhang, J. Y. Xu,
J. Q. Liu and J. C. Shen, Soft Matter, 2009, 5, 1905.
17 X. Huang, Y. Z. Yin, X. Jiang, Y. Tang, J. Y. Xu, J. Q. Liu and
J. C. Shen, Macromol. Biosci., 2009, 9, 1202.
18 Y. Tang, L. P. Zhou, J. X. Li, Q. Luo, X. Huang, P. Wu, Y. G. Wang,
J. Y. Xu, J. C. Shen and J. Q. Liu, Angew. Chem. Int. Ed., 2010, 49,
3920.
19 Y. Z. Yin, X. Huang, C. Y. Lv, L. Wang, S. J. Yu, Q. Luo, J. Y. Xu
and J. Q. Liu, Macromol. Biosci., 2010, 10, 1505.
20 H. J. Yu, J. Q. Liu, B. August, J. Li, G. M. Luo and J. C. Shen, J. Biol.
Chem., 2005, 280, 11930.
39 W. A. Braunecker and K. Matyjaszewski, Prog. Polym. Sci., 2007, 32,
93.
40 S. Edmondson, V. L. Osborne and W. T. S. Huck, Chem. Soc. Rev.,
2004, 33, 14.
41 H. Kong, C. Gao and D. Y. Yan, J. Am. Chem. Soc., 2004, 126, 412.
42 J. Pyun, T. Kowalewski and K. Matyjaszewski, Macromol. Rapid
Commun., 2003, 24, 1043.
43 K. Matyjaszewski, Chem.–Eur. J., 1999, 5, 3095.
44 S. Y. Liu, J. V. N. Weaver, Y. Q. Tang, N. C. Billingham, S. P. Armes
and K. Tribe, Macromolecules, 2002, 35, 6121.
45 H. Yoshida, K. Klinkhammer, M. Matsusaki, M. Moller, D. Klee
and M. Akashi, Macromol. Biosci., 2009, 9, 568.
46 Y. Lu, Y. Mei, M. Drechsler and M. Ballauff, Angew. Chem., Int. Ed.,
2006, 45, 813.
47 S. Nayak and L. A. Lyon, Angew. Chem., Int. Ed., 2005, 44, 7686.
48 A. S. Hoffman and P. S. Stayton, Prog. Polym. Sci., 2007, 32, 922.
49 J. Stadermann, S. Fleischmann, M. Messerschmidt, H. Komber and
B. Voit, Macromol. Symp., 2009, 35, 275.
50 J. Gao and B. J. Frisken, Langmuir, 2003, 19, 5217.
51 N. Dingenouts, C. Norhausen and M. Ballauff, Macromolecules,
1998, 31, 8912.
52 H. Senff, W. Richtering, C. Norhausen, A. Weiss and M. Ballauff,
Langmuir, 1999, 15, 102.
53 M. Hua, T. Kaneko, X. Y. Liu, M. Q. Chen and M. Akashi, Polym.
J., 2005, 37, 59.
54 S. W. Choi, M. Munteanu and H. Ritter, J. Polym. Res., 2009, 16,
389.
55 M. Munteanu, S. W. Choi and H. Ritter, Macromolecules, 2009, 42,
3887.
21 X. M. Liu, L. A. Silks, C. P. Liu, M. Ollivault-Shiflett, X. Huang,
J. Li, G. M. Luo, Y. M. Hou, J. Q. Liu and J. C. Shen, Angew.
Chem., Int. Ed., 2009, 48, 2020.
22 Y. Ge, Z. H. Qi, Y. Wang, X. M. Liu, J. Li, J. Y. Xu, J. Q. Liu and
J. C. Shen, Int. J. Biochem. Cell Biol., 2009, 41, 900.
23 H. Sies, Exp. Physiol., 1997, 82, 291.
56 B. Garska, M. Tabatabai and H. Ritter, Beilstein J. Org. Chem., 2010,
6, 784.
57 Z. S. Ge, J. Xu, J. M. Hu, Y. F. Zhang and S. Y. Liu, Soft Matter,
2009, 5, 3932.
24 M. Tian, C. D. Qu, Y. X. Feng and L. Q. Zhang, J. Mater. Sci., 2003,
38, 4917.
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 2521–2529 | 2529