Fabrication and evaluation of reduction-sensitive supramolecular hydrogel based on cyclodextrin/polymer inclusion for injectable drug-carrier application

Article abstract of DOI:10.1039/c1sm05426k

Supramolecular hydrogels based on cyclodextrin/polymer inclusion are an emerging injectable biomaterial for drug controlled-release and cell capsulation. Although the pH- and temperature-sensitivity has been focused on contributing to intelligence, the system sensitive to physiological reduction condition caused by glutathione tripepetide (GSH) has not been reported so far. In this work, novel reduction-sensitive supramolecular hydrogels were, for the first time, fabricated by the inclusion of [poly(ethylene glycol) monomethyl ether]-graft-[disulfide-linked poly(amido amine)] (mPEG-g-SS-PAA) with α-cyclodextrin (α-CD) in aqueous solution. The reduction-sensitivity was ascribed to the disulfide linker in the SS-PAA main chain while various physical conjugations contributed to a reversible gel-sol transition under shearing as a key of injectable function. The drug release from such a supramolecular hydrogel showed a prominent sustained release profile, and the release rate could further be regulated depending upon the reduction condition. It is worth noting that incorporating a low loading-level of reducing agent did not inhibit the formation of hydrogel. As a result, it became possible to use the reduction-sensitivity to regulate the drug release profile in extracellular milieus and normal tissue. Combined with acceptable cytotoxicity, this kind of reduction-sensitive supramolecular hydrogel based on cyclodextrin/polymer inclusion showed a great potential as an injectable smart biomaterial for the application of drug controlled-release.

Full text of DOI:10.1039/c1sm05426k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Soft Matter
Cite this: Soft Matter, 2011, 7, 7386
www.rsc.org/softmatter
PAPER
Fabrication and evaluation of reduction-sensitive supramolecular hydrogel
based on cyclodextrin/polymer inclusion for injectable drug-carrier
application
a
Jiahui Yu, Honglei Fan, Jin Huang
b
ab
and Jinghua Chenbc
*
Received 11th March 2011, Accepted 22nd May 2011
DOI: 10.1039/c1sm05426k
Supramolecular hydrogels based on cyclodextrin/polymer inclusion are an emerging injectable
biomaterial for drug controlled-release and cell capsulation. Although the pH- and temperature-
sensitivity has been focused on contributing to intelligence, the system sensitive to physiological
reduction condition caused by glutathione tripepetide (GSH) has not been reported so far. In this work,
novel reduction-sensitive supramolecular hydrogels were, for the first time, fabricated by the inclusion
of [poly(ethylene glycol) monomethyl ether]-graft-[disulfide-linked poly(amido amine)] (mPEG-g-SS–
PAA) with a-cyclodextrin (a-CD) in aqueous solution. The reduction-sensitivity was ascribed to the
disulfide linker in the SS–PAA main chain while various physical conjugations contributed to
a reversible gel–sol transition under shearing as a key of injectable function. The drug release from such
a supramolecular hydrogel showed a prominent sustained release profile, and the release rate could
further be regulated depending upon the reduction condition. It is worth noting that incorporating
a low loading-level of reducing agent did not inhibit the formation of hydrogel. As a result, it became
possible to use the reduction-sensitivity to regulate the drug release profile in extracellular milieus and
normal tissue. Combined with acceptable cytotoxicity, this kind of reduction-sensitive supramolecular
hydrogel based on cyclodextrin/polymer inclusion showed a great potential as an injectable smart
biomaterial for the application of drug controlled-release.
framework. These physical interactions mainly consist of the
hydrophobic interaction between the cyclodextrin inner and the
polymer, as well as non-covalent aggregation of (pseudo)poly-
rotaxanes formed by the threading of cyclodextrin onto the
polymer chain. Furthermore, such physical interactions result in
the sensitivity functioned as temperature5,9,18–23 and pH23
depending upon the structure of the included polymer chains. In
addition, the structural character of the polymer guest, such as
the cationic or anionic nature of the uncovered polymer
segments, also contributes to the additional stimulus-response
properties, such as the sensitivity to temperature,16 pH5,9,24 and so
on. Such stimulus-response properties of supramolecular
hydrogels also provide the essential intelligence required by drug
delivery and release. However, this kind of supramolecular
hydrogel sensitive to physiological reduction conditions caused
by glutathione tripepetide (GSH) has not been reported so far.
The reduction-sensitive polymers have emerged as a fasci-
nating class of intelligent biomaterials, and are especially applied
to encapsulate drugs in the forms of nano/micro-gel and micelles
for their delivery and controlled-release.25 With respect to the
molecular level, the disulfide linkage is a typical character of
reduction-sensitive polymers, and usually locates in main chains,
side chains or cross-linkers.25 A notable fact has been found that
the cleavage of the disulfide linkage, due to the thiol–disulfide
1
Introduction
Supramolecular hydrogels based on polymer/cyclodextrin
inclusion have attracted great interest as promising injectable
biomaterials,1–3 for applications as drug carriers with a sustained
release function,4–13 or as a cell-encapsulation vehicle with a good
biocompatibility.10,14 In such supramolecular hydrogels, drug
loading can simultaneously occur during the gelation process.
This not only enhances drug loading-levels, but also avoids
structural changes of the drug. In particular, physical encapsu-
lation shows prominent advantages for macromolecular drugs,
such as nucleic acids, peptides, proteins and so on. Furthermore,
a reversible sol–gel transition functioned as shearing force mainly
contributes to an injectable function,5–13,15–17 and breaks through
the limits of the conventional implantable strategy for hydrogels.
The shearing-response is originated from physical interactions as
the driving forces of gelation and the supporting of hydrogel
aInstitute of Drug Discovery and Development, Institutes for Advanced
Interdisciplinary Research, East China Normal University, Shanghai,
200062, China. E-mail: huangjin@iccas.ac.cn
bCollege of Chemical Engineering, Wuhan University of Technology,
Wuhan, 430070, China
cSchool of Medicine and Pharmaceutics, Jiangnan University, Wuxi,
214122, P. R. China
7386 | Soft Matter, 2011, 7, 7386–7394
This journal is ª The Royal Society of Chemistry 2011

View Article Online
exchange reaction, is sensitive to the reduction conditions in
the human body, namely the disulfide bonds show enough
stability to a low concentration of glutathione tripeptide (GSH),
ca. 2–20 mM in body circulation and extracellular milieus, while
the disulfide linkage quickly cleaves under high GSH concen-
tration, ca. 0.5–10 mM of intracellular condition and 4-fold over
normal tissue in tumors.
were purchased from Aldrich. All other reagents and solvents
were obtained commercially and used as received, unless noted
otherwise.
2.2 Cell line and culture
Mouse muscular cell line L929 was supplied from Institute of
Biochemistry & Cell Biology, Chinese Academy of Sciences.
Cells were cultured in RPMI 1640 (Gibco BRL, Paris, France),
supplemented with 10% fetal bovine serum (FBS, HyClone,
Logan, Utah), streptomycin at 100 mg mLꢀ1, and penicillin at 100
U mLꢀ1. All cells were incubated at 37 ꢁC in a humidified 5% CO2
atmosphere. Cells were split by using trypsin/EDTA solution
when almost confluent.
In this study, the disulfide-linked poly(amido amine) (SS–
PAA) containing alkyne focal groups as the main chain was
designed to become the origin of reduction-sensitivity. More-
over, considering that a branched structure, such as a grafted
comb-type,20–23 a high-densely double-grafted brush19 and so on,
could improve gelation of the polymers containing a low
molecular weight of poly(ethylene glycol) (PEG) segments via
cyclodextrin inclusion, azide-terminated poly(ethylene glycol)
monomethyl ether (mPEG–N3) was grafted onto SS–PAA by
‘‘click’’ chemistry to produce the guest of comb-like polymer
(mPEG-g-SS–PAA) for the cyclodextrin inclusion. Subse-
quently, a novel reduction-sensitive supramolecular hydrogel
was fabricated by the inclusion of a-cyclodextrin (a-CD) onto
mPEG-g-SS–PAA, and the elastic modulus and shear-thinning
behavior were investigated for verifying the injection function
and evaluating mechanical strength. Based on the reduction-
degradation behavior of mPEG-g-SS–PAA traced by gel
permeation chromatography, the drug-release profiles were
measured with bovine serum albumin (BSA) as a model protein
drug. Hence, the incorporation of dithiothreitol (DTT) as
a reducing agent model and a substitute of GSH was expected to
regulate the drug release behavior even at a low reduction
concentration, such as extracellular milieus in the normal tissue.
Finally, the in vitro cytotoxicity of such reduction-sensitive
supramolecular hydrogels was evaluated by the MTT method in
order to explore their possibility as injectable biomaterials.
2.3 Synthesis of N,N0-bis(acryloyl) cystamine (BAC)
The N,N0-bis(acryloyl) cystamine (BAC) was synthesized
according to the previous literature.26 Briefly, cystamine dihy-
drochloride (10 mmol) was dissolved in a mixture of 15 mL of
3.5ꢁM NaOH and 10 mL chloroform. This solution was heated to
50 C, and 5 mL of chloroform containing 20 mmol of acryloyl
chloride was added dropwise under constant stirring over 15 min
while the reaction temperature was maintained at ca. 50 ꢁC. After
separating the phases while still warm, the aqueous phase was
discarded. The remaining organic phase was cooled down to
room temperature, and the product precipitated directly from the
solution. The white crystal product was recovered by filtration
and recrystallized from chloroform to give a yield of 50%.
2.4 Synthesis of disulfide-linked poly(amido amine) (SS–PAA)
As shown in Fig. 1, the disulfide-linked poly(amido amine) (SS–
PAA) was synthesized by the Michael addition polymerization
between propargyl amine and BAC.27 Firstly, BAC (2.60 g,
10 mmol) and propargyl amine (550.8 mg, 10 mmol) were added
into a flask and then dissolved in 20 mL MeOH. The solution was
mechanically stirring at room temperature in the dark under
a nitrogen atmosphere for 7 days. Then, excess propargyl amine
(10 mol%) was added into the reaction solution to consume any
unreacted acrylamide groups, and the reaction was performed at
room temperature for at least an additional one week. Finally,
the reaction solution was evaporated, and then the residue was
dried in vacuum at 40 ꢁC to give the SS–PAA containing alkyne
focal groups.
2
Experimental
2.1 Materials
Cysteamine hydrochloride, propargyl amine, acryloyl chloride
and poly(ethylene glycol) monomethyl ether (mPEG) with Mn of
1900 Da were purchased from Alfa Aesar and used as received.
Dithiothreitol (DTT), sodium ascorbate and copper (II) sulfate
pentahydrate were purchased from Aladdin. Pristine a-cyclo-
dextrin (a-CD), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenylte-
trazolium bromide (MTT) and bovine serum albumin (BSA)
Fig. 1 Synthesis scheme of poly(ethylene glycol) monomethyl ether]-graft-[disulfide-linked poly(amido amine)] (mPEG-g-SS–PAA).
This journal is ª The Royal Society of Chemistry 2011 Soft Matter, 2011, 7, 7386–7394 | 7387

View Article Online
2.7 Reduction-degradation of mPEG-g-SS–PAA
2.5 Synthesis of mPEG-g-SS–PAA copolymer
The mPEG-g-SS–PAA copolymer was synthesized by a click
reaction between alkyne focal groups in SS–PAA and azide-
terminated mPEG (mPEG–N3).28 Herein, the mPEG–N3 was
prepared from poly(ethylene glycol) methyl ether according to
the previous literatures.29 A representative procedure for the
click conjugation is as follows: SS–PAA (315.46 mg, 1 mmol
acetylene) and mPEG–N3 (2.33 g, 1.2 mmol) were firstly dis-
solved in MeOH; subsequently, sodium ascorbate (95.10 mg,
0.48 mmol) and copper (II) sulfate pentahydrate (120.0 mg,
0.48 mmol) were added into the reaction vessel, and then was
mechanically stirred at room temperature for 48 h; the crude
reaction mixture was dialyzed against distilled water for 7 days
using a dialysis membrane (MWCO: 3500 Da) to remove excess
mPEG–N3 and other by-products followed by lyophilizing. To
adjust the polydispersity index (PDI) of the graft copolymer, the
lyophilized graft copolymer was further purified by reprecipita-
tion 5 times. Methyl alcohol and ethyl ether were used as solvent
and precipitant, respectively. The resultant graft copolymer
was dried in vacuum to yield 2.08 g of mPEG-g-SS–PAA
(yield: 92%).
The degradation behavior of mPEG-g-SS–PAA was investigated
in the presence of water-soluble reducing agent, such as dithio-
threitol (DTT).30 The concentrations of DTT in initial polymer
solutions were given as 40, 20 and 10 mM, respectively. Firstly,
the mPEG-g-SS–PAA was treated with DTT in PBS buffer under
a nitrogen atmosphere at 37 ꢁC. Aliquots were periodically
withdrawn, and analyzed by GPC to determine the molecular
weight and its distribution of degraded fragments after dilution
with PBS buffer. Before measurement, the solution after reduc-
tion degradation was dialyzed against distilled water for 7 days
using a dialysis membrane (MWCO: 1000 Da) to remove the
excess DTT. At the same time, the structure of degraded
fragment was determined by 1H NMR while the transition from
the mPEG-g-SS–PAA copolymer to small molecules, named as
the degradation rate in this work, was obtained from the
decreasing percentage of the peak area assigned to the original
copolymers.
2.8 Preparation of supramolecular hydrogels
Supramolecular hydrogels based on mPEG-g-SS–PAA and a-
CD were prepared according to formulations listed in Table 1,
where some systems contained various loading-levels of DTT as
reduction reagent. Generally, 200 mg a-CD was firstly dissolved
into 1 mL of deionized water, and then mixed with 1 mL aqueous
solution containing mPEG-g-SS–PAA (40 mg, equal to 0.018
mmol disulfide bonds). The ultimate concentrations of a-CD and
mPEG-g-SS–PAA in the mixture were fixed to be of 100 mg
mLꢀ1 and 20 mg mLꢀ1, respectively. The resulting mixture was
vigorously stirred for a while followed by sonication for 10 min,
and finally conditioned at room temperature. The gelation time
was settled at the time point when the mixing solution became an
immobile gel in the inverted cuvettes.
2.6 Characterizations of polymers
1H nuclear magnetic resonance (1H NMR) spectra were recorded
on a Bruker Avarice ꢀ 500 NMR spectrometer, and the sample
concentration was 35 mg mLꢀ1 in DMSO-d6 or CDCl3. Fourier
transform infrared (FT-IR) spectra were recorded on a Nicolet
Nexus 670 spectrometer in the range of 4000–400 cmꢀ1 with use
of a KBr pellet.
The molecular weight and its distribution of polymers were
measured with Agilent 1200 gel permeation chromatography
(GPC) (Agilent Technologies Inc. Shanghai Branch). Agilent
1200 refractive index detector and aqueous SEC start-up kit were
used. Chromatography columns (PL aquagel-OH MIXED
columns, Polymer Laboratories Ltd. Amherst, MA, USA) were
calibrated with poly(ethylene glycol) kit and polystyrene kit for
different mobile phase, respectively. The column temperature
Considering that such supramolecular hydrogels can be used
to regulate the drug release behavior under the condition of low
GSH concentration, such as extracellular milieus in the normal
tissue of the human body, the DTT, as a reducing agent model
and a substitute of GSH, was designatedly introduced into this
kind of reduction-sensitive supramolecular hydrogels at the stage
of mixing a-CD with mPEG-g-SS–PAA. The addition amount of
DTT was given to produce the molar ratio of DTT vs. disulfide
ꢁ
was maintained at 25 C. Mobile phases were phosphate buffer
(PBS) for mPEG-g-SS–PAA and tetrahydrofuran (THF) for SS–
PAA, respectively, and the flow rate was 1.0 mL minꢀ1
.
Table 1 Compositions and gelation time of the mixing aqueous solution containing mPEG-g-SS–PAA, a-CD and DTT
Compositions
DTT/disulfide bond
molar ratio
Sample code
a-CD (mg mLꢀ1
)
mPEG-g-SS–PAA (mg mLꢀ1
)
Gelation time (min)a
Gel/DTT-0
Gel/DTT-1
Gel/DTT-5
Gel/DTT-10
Sol/DTT-25
Sol/DTT-50
Sol/DTT-100
100
100
100
100
100
100
100
20
20
20
20
20
20
20
No DTT
1 : 1
5 : 1
10 : 1
25 : 1
50 : 1
100 : 1
10
15
25
40
No gelation
No gelation
No gelation
a
The gelation time was settled at the time point when the mixture solution became an immobile gel in the inverted cuvettes.
7388 | Soft Matter, 2011, 7, 7386–7394
This journal is ª The Royal Society of Chemistry 2011

View Article Online
2.11 Cell viability assays of supramolecular hydrogels
bond as100 : 1, 50 : 1, 25 : 1, 10 : 1, 5 : 1 and 1 : 1, respectively.
Fortunately, when the molar ratio of DTT vs. disulfide bond was
lower than 10 : 1, the mixture composed of a-CD, mPEG-g-SS–
PAA and DTT could result in gelation in spite of inevitable
cleavage of part of the disulfide linkage. Meanwhile, a higher
molar ratio of DTT vs. disulfide bond prolonged the gelation
time.
A total of 600 mg of each freeze-dried powder sample of
hydrogel, 100 mg of mPEG-g-SS–PAA and 500 mg of a-CD were
put into 5 mL of RPMI1640 cell culture medium in a 50 mL
centrifuge tube, respectively, and then placed in a shaker incu-
bator (37 ꢁC, 60 rpm) for 2 days. After that, the media were
filtered with 0.22 mm sterile filter into a sterile container and
ꢁ
stored in a refrigerator at 4 C before use.
L929 cells were seeded into 96-well plate at an initial density of
6 ꢃ 103 cells per well in 100 mL growth medium and incubated for
24 h to reach 80% confluency at the time of treatment. Cell
culture medium was replaced by 100 mL of the extracted leached
medium from the corresponding hydrogel sample, mPEG-g-SS–
PAA and a-CD. After 48 h of incubation, the well was washed
twice with PBS solution, and then, a total of 10 mL of MTT
solution (0.5 mg mLꢀ1 in PBS) was added into each well. After
further incubation for 4 h in a CO2 cell incubator, MTT solution
was removed while 100 mL of dimethyl sulfoxide (DMSO) was
added into each well. Finally, the plate was incubated for 10 min
in the incubator, and then for 15 min at room temperature. The
optical density was measured at 570 nm using an automatic BIO-
TEK microplate reader (Powerwave XS, USA) to evaluate the
metabolic activity of the cells, and the cell viability was calculated
from the following equation:
2.9 Characterizations of supramolecular hydrogels
X-ray diffraction (XRD) measurement of the freeze-dried
hydrogel powder was performed on a D/max-250 X-ray
diffractometer (Rigaku Denki, Japan) equipped with a Cu Ka
radiation source (l ¼ 0.154 nm). The diffraction data were
collected in a range of 2q ¼ 5–60ꢁ using a fixed time mode with
a step interval of 0.02ꢁ. At the same time, the systems where the
molar ratio of DTT vs. disulfide bond was higher than 10 : 1,
were also freeze-dried as powders, and then analysed by XRD
measurements.
The rheological behaviors of supramolecular hydrogels were
measured on an AR 2000ex rheometer (TA Instruments, USA)
using parallel plate geometry with a diameter of 60 mm at 25 ꢁC.
The gap distance between two plates was fixed at 1 mm. A
frequency sweep test was conducted on each sample to determine
values of storage modulus (G0) and loss modulus (G00) over
a frequency (u) range of 0.1–200 rad sꢀ1. The steady flow
behaviors were performed over a shearing rate range of 0.01–
50 sꢀ1. Temperature control was established by a Julabo FS 18
cooling/heating bath, and kept within ꢂ0.5 ꢁC over an extended
time.
ODsample
Cell viability ð%Þ ¼
ꢃ 100%
(2)
ODcontrol
where ODsample represents the OD value from a well treated with
sample and OD control from a well treated with PBS buffer only.
Each experiment was carried out in triplicate. Results were
shown as means and corresponding standard deviations (mean ꢂ
SD).
2.10 Drug release profiles from supramolecular hydrogels
3
Results and discussion
Bovine serum albumin (BSA), as a model protein drug, was
used to investigate the effect of DTT on the release profiles
of supramolecular hydrogels. In the process of preparing
supramolecular hydrogels with and without DTT, 6.0 mg of
BSA was dissolved in 0.60 mL of phosphate-buffered saline
(PBS) solution (pH 7.4) containing 90 mg of a-CD, and
then mixed with 0.3 mL of PBS containing 18 mg of
mPEG-g-SS–PAA or mPEG-g-SS–PAA/DTT mixture. The
resulting mixture was placed into a 1 mL cuvette, and
incubated in a 37 ꢁC water bath for 1 h to form a BSA-
loaded hydrogel.
3.1 Structure of mPEG-g-SS–PAA copolymer
As shown in Fig. 1, the reduction-degradable mPEG-g-SS–PAA
was synthesized via a click cycloaddition reaction between azido-
terminated mPEG (PEG–N3) and the alkyne-functionalized
disulfide-linked poly(amido amine) (SS–PAA). As the precursor
of synthesizing alkyne-functionalized SS–PAA, the BAC
monomer was synthesized according to a classic reaction
pathway with acryloyl chloride under conventional Schotten–
Baumann conditions. 1H NMR verified the structure of as-
prepared BAC, and their detail chemical shifts were shown in
Fig. 2a. The FT-IR of BAC showed the characteristic adsorption
bands at 1560 cmꢀ1 (N–H bending, amide II), 1650 cmꢀ1 (C]O,
amide I) and 3250 cmꢀ1 (N–H stretching) (Fig. 3a).31 Subse-
quently, the alkyne-functionalized SS–PAA was synthesized
conveniently by Michael addition of molar equivalents of
propargyl amine vs. BAC. The reaction was carried out under
mechanical stirring for 7–10 days until the solution became
viscous. After that, in order to consume any unreacted acryl-
amide groups, 10% mol excess of propargyl amine was added
followed by reaction for another 7 days to terminate the poly-
merization with amino end-groups. The final structure of alkyne-
functionalized SS–PAA was confirmed by 1H NMR spectrum in
Fig. 2b. The signal located at 2.23 ppm suggested the presence of
For in vitro release studies, the cuvette containing drug-loaded
hydrogel was placed in a test tube with 10 mL of PBS buffer, and
then conditioned in a shaking table at 37 ꢁC. After a given
interval, 0.3 mL of buffer medium was removed to determine the
absorbance at 595 nm on a UV-1900PC spectroscope (Shanghai,
China), and hence calculated the concentration of released BSA.
At the same time, the same volume of fresh PBS solution was
added to keep the total solution volume constant.
The release profile of BSA from the supramolecular hydrogel
without DTT was investigated against the PBS buffer containing
0, 10, 20 and 40 mM DTT, respectively. Moreover, the release
profile of BSA from supramolecular hydrogels with molar ratios
of DTT vs. disulfide bond as 10 : 1, 5 : 1 and 1 : 1 were also
investigated against the PBS buffer without DTT.
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 7386–7394 | 7389

View Article Online
Fig. 2 1H NMR spectra of BAC in CDCl3 (a), SS–PAA in DMSO-d6 (b), mPEG-g-SS–PAA in CDCl3 (c), and the degraded mPEG-g-PAA–SH in
CDCl 3(d).
shows a complete disappearance of the azide vibrational peak
located at 2106 cmꢀ1 as shown in Fig. 3b of mPEG–N3, and
a new absorption in the range of about 1600–1640 cmꢀ1 for the
typical character of triazole ring appeared. At the same time, the
intense stretching bands at 2883 cmꢀ1 and 840 cmꢀ1 in Fig. 3c
were assigned to grafted mPEG segments, while a broad band
located in the range of 3200–3600 cmꢀ1 were assigned to the –NH
groups in the SS–PAA main chain.
3.2 Reduction degradation of mPEG-g-SS–PAA
The degradation behavior of mPEG-g-SS–PAA was investigated
by the incubation of copolymers with DTT under physiological
conditions. As previously reported,32,33 the disulfide bonds in the
SS–PAA main chain was cleaved by DTT due to the formation of
more stable six-membered cyclic disulfide after the oxidation of
Fig. 3 FT-IR spectra of the BAC (a), mPEG–N3 (b), and mPEG-g-SS–
such DTT dithiol. 1H NMR and GPC measurements confirmed
PAA (c).
the cleavage of disulfide bonds to yield mPEG-g-PAA–SH. The
new proton signal located at ca. 1.5 ppm (s, –SH) in Fig. 2d of
alkyne focal group (–ChCH). The weight-average (Mw) and
number-average (Mn) molecular weights of alkyne-functional-
ized SS–PAA measured by GPC were 9.8 ꢃ 104 and 2.1 ꢃ 104,
respectively. The inherent nature of the Michael addition resul-
ted in a relatively high polydispersity index (PDI) of 4.67.
The ‘‘click’’ conjugation between mPEG–N3 and alkyne-
functionalized SS–PAA was carried out using CuSO4$5H2O/
sodium ascorbate as catalyst in aqueous media at room
temperature. The Mw, Mn and PDI of the as-prepared mPEG-g-
SS–PAA measured by GPC were 1.14 ꢃ 105, 5.81 ꢃ 104 and 1.98,
mPEG-g-PAA–SH indicated the presence of the thiol end-group.
At the same time, the ethylene protons next to the thiol end-
group (–CH2–SH) were shown as new peaks located at ca. 2.75
ppm while the signal intensity of methylene protons neighboring
the disulfide bond (–CH2–SS–CH2–) located at ca. 2.92 ppm
weakened in Fig. 2d in contrast with those in Fig. 2a, 2b and 2c.
In addition, the changes of the GPC curves shown in Fig. 4a also
verified the degradation process of mPEG-g-SS–PAA under
a DTT concentration of 40 mM. As the incubation time pro-
longed, the peak intensity of small molecules, corresponding to
ca. 9 min of elution time, gradually increased together with the
decrease of peak proportion assigned to polymer or high
molecular weight fraction. It proved that the mPEG-g-SS–PAA
could be cleaved as a short segment, oligomer or small molecule
under high reductant concentration. At the incubation time of 6
days, the GPC result showed that the high molecular-weight
fraction after degradation had a Mw of 3.0 ꢃ 103, Mn of 2.88 ꢃ
103 and a PDI of 1.03. With a decrease of the DTT concentration
in the incubation media, the degradation rate of mPEG-g-SS–
PAA decreased as shown in Fig. 4b. The faster transition from
1
respectively. Moreover, H NMR confirmed the mPEG-g-SS–
PAA structure, and the detailed signals of chemical shifts were
shown in Fig. 2c. The occurrence of the ‘‘click’’ reaction was
confirmed from the appearance of a new proton signal located at
7.64 ppm (i, singlet) and the methylene proton signal adjacent to
triazole ring located at 4.53 ppm of (j, singlet) (Fig. 2c).
According to the integral ratios of proton signal ‘‘i’’ or ‘‘j’’ vs. ‘‘c’’,
the click conjugation between mPEG–N3 and SS–PAA is
quantitative within the error of 1H NMR measurement. In
addition, the FI-IR spectrum of mPEG-g-SS–PAA (Fig. 3c)
7390 | Soft Matter, 2011, 7, 7386–7394
This journal is ª The Royal Society of Chemistry 2011

View Article Online
hydrogels with various molar ratios of DTT vs. disulfide bond
was listed in Table 1. Obviously, the incorporation of DTT
resulted in the prolongation of gelation time. With an increase in
the molar ratios of DTT vs. disulfide bond from 1 : 1 to 10 : 1, the
gelation time increased due to partial cleavage of disulfide bond
in the main chain of mPEG-g-SS–PAA. When the molar ratio of
DTT vs. disulfide bond was higher than 25 : 1, the severe
breakage of the main chain in mPEG-g-SS–PAA inhibited the
formation of hydrogels. Only a turbid or transparent solution
was observed.
As well-known, the aggregation of (pseudo)polyrotaxanes that
formed by the threading of cyclodextrin onto polymer chains,
plays a key role as physical crosslinking to construct the hydrogel
network. Furthermore, for those PEG segments with low
molecular weight that cannot become a gelation solely, the
branched structure, such as grafted comb-type20–23 and high-
densely double-grafted brush,19 provide a possibility to gelation
via cyclodextrin inclusion. As a result, in the system of mPEG-g-
SS–PAA and a-CD, the aggregation of the inclusion complex
between a-CD and grafted PEG segment was a driving force to
induce gelation; it was proved by the XRD patterns in Fig. 5.
Different from two diffraction peaks of PEG crystalline located
at 18.9ꢁ and 23.5ꢁ of 2q for mPEG-g-SS–PAA and the diffraction
pattern of pristine a-CD, all the freeze-dried powders, including
the mPEG-g-SS–PAA/a-CD hydrogels with or without DTT and
the mPEG-g-SS–PAA/a-CD mixture containing a high molar
Fig. 4 GPC curves at the given time (A) of mPEG-g-SS–PAA incubated
in PBS buffer (pH ¼ 7.4) containing 40 mM DTT at 37 ꢁC, and the
degradation rate (B) of mPEG-g-SS–PAA incubated in PBS buffer (pH ¼
7.4) containing various DTT concentrations at 37 ꢁC.
polymer to small molecules, named as the biodegradation of
polymer in this work, was ascribed to that more amount of DTT
might provide more chance to cleave the disulfide bonds of the
main chain in the mPEG-g-SS–PAA copolymer. The highest
DTT concentration of 40 mM could give the degradation of over
ca. 90% mPEG-g-SS–PAA in 6 days. Although the degradation
rate of mPEG-g-SS–PAA was obviously slower than those of
other reports,25 such reduction sensitivity still showed a potential
to regulating the hydrogel structure and drug release profile. In
this case, a slower degradation rate might be ascribed to the steric
hindrance of grafted mPEG segments which inhibited the thiol–
disulfide exchange reaction between DTT and disulfide bonds in
the SS–PAA main chain.
3.3 Fabrication of supramolecular hydrogels
Based on the inclusion between a-CD and grafted mPEG
segment, the mixture of mPEG-g-SS–PAA and a-CD in aqueous
solution could form the hydrogel in tens of minutes depending
upon the molar ratios of incorporated DTT vs. disulfide bond
in mPEG-g-SS–PAA. The gelation time of supramolecular
Fig. 5 XRD patterns of the powders freeze-dried from the a-CD/
mPEG-g-SS–PAA hydrogels with and without DTT and the mixing
solution containing high DTT loading-level, and the powders of a-CD
and mPEG-g-SS–PAA as a comparison.
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 7386–7394 | 7391

View Article Online
ratio of DTT vs. disulfide bond (higher than 25 : 1), showed
a typical diffraction peak located at 19.7ꢁ of 2q assigned to the
‘‘channel’’-type crystalline of the PEG/a-CD pseudopolyrotax-
anes.19,20,22,23 However, when the molar ratio of DTT vs. disulfide
bond was higher than 25 : 1, the diffraction intensities of the
peak located at 19.7ꢁ of 2q obviously weakened, suggesting that
the cleavage of the SS–PAA main chain inhibited the positive
effect of origin comb-like structure on improving the aggregation
of pseudopolyrotaxanes and subsequently driving gelation. It is
well consistent with the observation of no gelation for the
mPEG-g-SS–PAA/a-CD system containing more DTT (the
molar ratio of DTT vs. disulfide bond higher than 25 : 1).
Fortunately, the systems with the relatively low molar ratio of
DTT vs. disulfide bond (i.e. lower than 10 : 1) still became gela-
tion, and might facilitate to extending the applications of this
kind of reduction-sensitive supramolecular hydrogel as a drug
carrier with a function of regulating release behavior in the
condition of lower GSH concentration.
strength and shear-thinning behavior. Fig. 6a depicts the
dynamic rheological behaviors of supramolecular hydrogels with
and without DTT. The storage modulus (G0) of all the hydrogels
exhibited a substantial elastic response, and was greater than the
loss modulus (G00) over the entire range of testing frequencies. It
suggested that these supramolecular hydrogels possessed
a permanent network, in spite that the network framework was
supported by the physical interactions, and hence displayed
strength and rigidity. At the same time, with an increase in the
molar ratio of DTT vs. disulfide bond, both the G0 and G00
plateaus slightly shifted downward. It is ascribed to the degra-
dation of the SS–PAA main chain after introducing DTT. At this
time, the chemical joints along the main chain in the hydrogel
network were cleaved, and hence decreased the mechanical
strength to some extent.
Fig. 6b depicts the steady rheological behaviors of supramo-
lecular hydrogels with and without DTT. With an increase in
shearing rate, the viscosity of all the hydrogel diminished greatly
until the viscosity was almost absence at a shearing rate of 1 sꢀ1
,
suggesting that these supramolecular hydrogels were thixotropic.
When the shearing force was removed, the flowable sols could
recover the same as the hydrogel did. Such reversible gel–sol
transition induced by shearing, similar to previous
reports,5–13,15–18 was a key for the application as an injectable
biomaterial. In addition, the loading-level of DTT still showed
the significant effect on the viscosity values of these supramo-
lecular hydrogels. Furthermore, the dependence of the initial
viscosity upon the loading-level of DTT was similar to that of the
G0 value, namely the initial viscosity decrease with an increase in
the molar ratio of DTT vs. disulfide bond.
3.4 Rheological studies of supramolecular hydrogels
Rheology measurements can evaluate two fundamental proper-
ties for this category of injectable hydrogel, i.e. mechanical
3.5 Drug release profiles of supramolecular hydrogels
Bovine serum albumin (BSA) as a model protein drug was
encapsulated into the hydrogels to investigate in vitro release
kinetics. Fig. 7a shows the effects of the DTT concentrations in
testing reduction media (PBS buffer with pH 7.4) on the BSA
release from the selected mPEG-g-SS–PAA/a-CD hydrogels
without DTT. In the PBS buffer with pH 7.4, the release profile
of BSA-loaded supramolecular hydrogel without DTT showed
a prominent sustained release characteristic, and the accumula-
tive release ratio was only ca. 70% until 30 days. With an increase
of DTT concentrations in testing reduction media, the release
rate of drug obviously increased. Even when the DTT concen-
trations were 20 and 40 mM, the test was carried out only in 22
and 18 days because the severe degradation of mPEG-g-SS–PAA
while the accumulative release ratio was close to 100%.
As mentioned in the introduction, the disulfide bonds show
enough stability in the body circulation and in the extracellular
milieus due to the low concentration of reducing agent, i.e.
glutathione tripepetide (GSH). As a result, the DTT, a reducing
agent model and a substitute of GSH, was designatedly incor-
porated into the supramolecular hydrogel to regulate the drug
release behavior. The release profiles of BSA from the supra-
molecular hydrogels with various molar ratios of DTT vs.
disulfide bonds were investigated against the PBS buffer with pH
7.4. Fig. 7b depicted the effects of the molar ratios of DTT vs.
disulfide bonds on the release process of BSA from supramo-
lecular hydrogel containing DTT. Remarkably, a higher molar
Fig. 6 Dynamic (A) and steady (B) rheological behaviors of the a-CD/
mPEG-g-SS–PAA hydrogels with various molar ratios of DTT vs.
disulfide bonds.
7392 | Soft Matter, 2011, 7, 7386–7394
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Fig. 7 In vitro release profiles of BSA from the a-CD/mPEG-g-SS–PAA
hydrogels against PBS buffer of pH ¼ 7.4 containing various DTT
concentrations (A) and from the a-CD/mPEG-g-SS–PAA hydrogels with
various the molar ratio of DTT vs. disulfide bond against PBS buffer of
pH ¼ 7.4 (B).
ratio of DTT vs. disulfide bond, corresponding to more DTT in
the supramolecular hydrogel, resulted in a higher release rate of
BSA. When the molar ratio of DTT vs. disulfide bond was 10 : 1,
over 90% BSA was released only in 15 days. It was ascribed to the
rapid erosion of the supramolecular hydrogel due to the severe
cleavage of mPEG-g-SS–PAA under high reductant concentra-
tion. Since the incorporation of proper DTT loading-level could
still produce supramolecular hydrogel and show the ability to
regulating the drug release as expected, this strategy might extend
the application of such reduction-sensitive supramolecular
hydrogels.
Fig. 8 Cell viability of the freeze-dried powder of a-CD/mPEG-g-SS–
PAA hydrogel (C) at various concentrations as well as a-CD (A) and
mPEG-g-SS–PAA (B) at equivalent concentrations against L929 cell line
for 48 h.
3.6 Cytotoxicity of supramolecular hydrogels
Combined with injection potential and good sustained release
profile, this kind of reduction-sensitive supramolecular hydrogel
may be considered as an injectable smart biomaterial. As a result,
to know their cytotoxicity as the elementary evaluation of
biocompatibility, the in vitro cell viability of the extracted leached
media from supramolecular hydrogel as well as mPEG-g-SS–
PAA and a-CD was evaluated by the MTT assay using L-929 cell
line. Fig. 8 depicts the dependence of cell viability upon the
testing concentrations for mPEG-g-SS–PAA, a-CD and the
freeze powder of supramolecular hydrogel without DTT, and it
was shown as representative concentration–growth inhibition
curves. When the testing concentration of the freeze-dried
powder of supramolecular hydrogel was lower than 60 mg mLꢀ1
,
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 7386–7394 | 7393

View Article Online
the freeze-dried powder of supramolecular hydrogel without
DTT as well as the mPEG-g-SS–PAA and a-CD with the
equivalent concentrations inhibited cell growth in a dose-
dependent manner, and showed equivalent cell viability at the
same equivalent concentrations. In particular, the cell viability
for the freeze-dried powder of supramolecular hydrogel
without DTT was close to 100% for low concentrations of 1.2
and 3.0 mg mLꢀ1. However, when the testing concentration for
the freeze-dried powder of supramolecular hydrogel without
DTT was higher than 90 mg mLꢀ1, the supramolecular hydrogel
without DTT showed the cell viability of less than 10% due to the
severe cytotoxicity of high a-CD concentration.34 The severe
cytotoxicity of high equivalent concentration of a-CD was also
depicted by the results in Fig. 8a. In virtue of infinite diluted
condition in body circulation, such reduction-sensitive supra-
molecular hydrogel might meet the requirement of low cyto-
toxicity as an injectable biomaterial.
References
1 J. Li, Adv. Polym. Sci., 2009, 222, 79–113.
2 F. van de Manakker, T. Vermonden, C. F. van Nostrum and
W. E. Hennink, Biomacromolecules, 2009, 10, 3157–3175.
3 J. Li and X. J. Loh, Adv. Drug Delivery Rev., 2008, 60, 1000–1007.
4 I. I. Abu Hashim, T. Higashi, T. Anno, K. Motoyama, A. E. H. Abd-
Elgawad, M. H. El-Shabouri, T. M. Borg and H. Arima, Int. J.
Pharm., 2010, 392, 83–91.
5 Y. Chen, X. H. Pang and C. M. Dong, Adv. Funct. Mater., 2010, 20,
579–586.
6 X. L. Zhang, J. Huang, P. R. Chang, J. L. Li, Y. M. Chen,
D. X. Wang, J. H. Yu and J. H. Chen, Polymer, 2010, 51, 4398–4407.
7 X. P. Ni, A. Cheng and J. Li, J. Biomed. Mater. Res., Part A, 2009,
88A, 1031–1036.
8 S. P. Zhao, J. Lee and W. Xu, Carbohydr. Res., 2009, 344, 2201–2208.
9 L. X. Ren, L. H. He, T. C. Sun, X. Dong, Y. M. Chen, J. Huang and
C. Wang, Macromol. Biosci., 2009, 9, 902–910.
10 D. Wu, T. Wang, B. Lu, Z. D. Xu, S. X. Cheng, X. J. Jiang,
X. Z. Zhang and R. X. Zhuo, Langmuir, 2008, 24, 10306–10312.
11 X. Li and J. Li, J. Biomed. Mater. Res., Part A, 2008, 86A, 1055–1061.
12 J. Li, X. Li, X. Ni, X. Wang, H. Li and K. W. Leong, Biomaterials,
2006, 27, 4132–4140.
13 J. Li, X. Ni and K. W. Leong, J. Biomed. Mater. Res., Part A, 2003,
65A, 196–202.
4
Conclusions
14 T. Wang, X. J. Jiang, Q. Z. Tang, X. Y. Li, T. Lin, D. Q. Wu,
X. Z. Zhang and E. Okello, Acta Biomater., 2009, 5, 2939–2944.
15 K. L. Liu, J. L. Zhu and J. Li, Soft Matter, 2010, 6, 2300–2311.
16 S. P. Zhao, L. M. Zhang and D. Ma, J. Phys. Chem. B, 2006, 110,
16503–16507.
17 S. P. Zhao, L. M. Zhang and D. Ma, J. Phys. Chem. B, 2006, 110,
12225–12229.
18 J. Li, X. Li, Z. Zhou, X. Ni and K. W. Leong, Macromolecules, 2001,
34, 7236–7237.
19 L. H. He, J. Huang, Y. M. Chen, X. J. Xu and L. P. Liu,
Macromolecules, 2005, 38, 3845–3851.
20 K. M. Huh, Y. W. Cho, H. Chung, I. C. Kwon, S. Y. Jeong, T. Ooya,
W. K. Lee, S. Sasaki and N. Yui, Macromol. Biosci., 2004, 4, 92–99.
21 H. S. Choi, K. Kontani, K. M. Huh, S. Sasaki, T. Ooya, W. K. Lee
and N. Yui, Macromol. Biosci., 2002, 2, 298–303.
22 K. M. Huh, T. Ooya, W. K. Lee, S. Sasaki, I. C. Kwon, S. Y. Jeong
and N. Yui, Macromolecules, 2001, 34, 8657–8662.
23 H. S. Choi, K. Yamamoto, T. Ooya and N. Yui, ChemPhysChem,
2005, 6, 1081–1086.
24 R. X. Yuan and X. T. Shuai, J. Polym. Sci., Part B: Polym. Phys.,
2008, 46, 782–790.
Based on the design of molecular structure, the reduction-
sensitive disulfide bond and the included poly(ethylene glycol)
(PEG) segments were combined together to synthesize [PEG
monomethyl ether]-graft-[disulfide-linked poly(amido amine)]
(mPEG-g-SS–PAA). Moreover, the inclusion of a-cyclodextrin
(a-CD) to PEG segments induced the formation of supramo-
lecular hydrogels. Such supramolecular hydrogels inherited the
shear-thinning character and reduction-sensitivity, and showed
an adjustable sustained drug-release characteristic depending
upon the reduction condition. Fortunately, low loading-level of
reducing agent cannot inhibit the formation of hydrogel, and
provided a possibility of regulating drug release profile in
extracellular milieus and normal tissue. This is the first report on
reduction-sensitive supramolecular hydrogel based on cyclo-
dextrin/polymer inclusion. Consideration of their acceptable
cytotoxicity, such reduction-sensitive supramolecular hydrogels
showed a great potential as an injectable smart biomaterial for
the application of drug controlled-release.
25 F. Meng, W. E. Hennink and Z. Zhong, Biomaterials, 2009, 30, 2180–
2198.
26 H. Yu and D. W. Grainger, Macromolecules, 1994, 27, 4554–4560.
27 C. Lin, C. J. Blaauboer, M. M. Timoneda, M. C. Lok, M. van
Steenbergen, W. E. Hennink, Z. Zhong, J. Feijen and
J. F. J. Engbersen, J. Controlled Release, 2008, 126, 166–174.
28 B. Parrish and T. Emrick, Bioconjugate Chem., 2007, 18, 263–267.
29 C. Hua, S. M. Peng and C. M. Dong, Macromolecules, 2008, 41,
6686–6695.
30 Y. Sun, X. L. Yan, T. M. Yuan, J. Liang, Y. J. Fan, Z. W. Gu and
X. D. Zhang, Biomaterials, 2010, 31, 7124–7131.
31 H. Yu and D. W. Grainger, Macromolecules, 1994, 27, 4554–4560.
32 N. V. Tsarevsky and K. Matyjaszewski, Macromolecules, 2002, 35,
9009–9014.
Acknowledgements
The research work was financially supported by 973 Projects of
Chinese Ministry of Science and Technology (2009CB930300);
Foundation
of
Shanghai
Municipality
Commission
(10410710000, 09540709000); Fundamental Research Funds for
the Central Universities (Self-Determined and Innovative
Research Funds of WUT 2010-II-022); and Open Project of
Chinese Academy of Sciences Key Laboratory of Biological
Effects of Nanomaterials and Nanosafety.
33 H. L. Fan, J. Huang, Y. P. Li, J. H. Yu and J. H. Chen, Polymer, 2010,
51, 5107–5114.
34 J. C. Debouzy, F. Fauvelle, S. Crouzy, L. Girault, Y. Chapron,
€
M. Goschl and A. Gadelle, J. Pharm. Sci., 1998, 87, 59–66.
7394 | Soft Matter, 2011, 7, 7386–7394
This journal is ª The Royal Society of Chemistry 2011