Synthesis and characterization of in situ cross-linked hydrogel based on

self-assembly of thiol-modied chitosan with PEG diacrylate using Michael

type addition
Da-yong Teng
a,1
, Zhong-ming Wu
b,1
, Xin-ge Zhang
a,
*
, Yan-xia Wang
a
, Chao Zheng
a
,
Zhen Wang
a
, Chao-xing Li
a,
*
a
The Key Laboratory of Functional Polymer Materials of Ministry Education, Institute of Polymer Chemistry, Nankai University, Tianjin 300071, China
b
Metabolic Diseases Hospital, Tianjin Medical University, Tianjin 300070, China
a r t i c l e i n f o
Article history:
Received 24 February 2009
Received in revised form
16 November 2009
Accepted 7 December 2009
Available online 4 January 2010
Keywords:
Chitosan
In situ hydrogel
Michael type addition
a b s t r a c t
Anovel injectable in situ cross-linked hydrogel has beendesigned via Michael type additionbetweenthiol-
modied chitosan (CSNAC) and PEG diacrylate (PEGDA). Hydrogel was rapidly formed in situ under
physiological conditions. The gelation time depended on the content of free thiols in CSNAC, temperature,
and concentration of CSNAC and PEGDA. Thermogravimetric analysis showed the thermal stabilities of
hydrogels. SEM observation results conrmed a porous 3D structure. Rheological studies showed that the
cross-linking density and elasticity of hydrogel had a correlation to the content of CSNAC and PEGDA.
Swelling studies revealed that these hydrogels had a high initial swelling and were degradable under
physiological conditions. And swelling was highly temperature-dependent and was directly related to the
amount of cross-linking. Biological activities of the hydrogels were evaluated by in vitro cell compatibility
on HDFs and A549 cells and the results indicated that the hydrogel was biocompatible.
2009 Elsevier Ltd. All rights reserved.
1. Introduction
Hydrogels are three-dimensional cross-linked polymer
networks, which may absorb from 10% to 20% up to thousands of
times their dry weight in water, giving them physical character-
istics similar to soft tissues. Due to their excellent biocompati-
bility, causing minimal inammatory responses, thrombosis and
tissue damage, many hydrogels have been studied extensively for
biomedical applications [1,2], such as drug delivery [3] and tissue
engineering [4,5]. Hydrogels that are injectable and can be formed
in situ under physiological conditions have received much atten-
tion recently because of their many favorable characteristics. For
instance, injectable materials can conform to complex shapes at
the site of injury, and can strongly adhere to the tissue during
gelation. Furthermore, in situ gelation allows minimize the inva-
siveness of the surgical technique by permitting the use of injec-
tion or laparoscopic methods. During the last decade, many
potential applications have been explored for ideal injectable in
situ crosslinkable hydrogels [610].
In situ gelation can be obtained after UV-polymerization [11],
introducing non-reversible covalent bonds, or via self-assembly by
either reversible interactions or non-reversible chemical reactions
[1219]. Among these in situ gelling systems, self-assembling
hydrogels which can be formed in time or in response to a certain
stimulus are of great interest. Both physical interactions, such as
electrostatic [1214] or hydrophobic interactions [1517], as well as
end-group-specic chemical reactions, such as Michael addition
[18,19], can be exploited for the design of self-assembly polymeric
networks. However, physically cross-linked hydrogels are generally
mechanically weaker than chemically cross-linked hydrogels.
Therefore, in situ gelation hydrogels prepared via Michael addition
between thiols and acrylates have received much attention
recently. The reaction can be carried out under physiological
conditions and biomimetic scaffolds can be easily obtained by
incorporation of thiol-bearing molecules. Hubbell et al. reported
the hydrogels by Michael type addition between small molecules
bearing several thiol groups and multi-arm star PEG acrylates or
vinyl sulfones [2022]. The experiments suggest that the cells can
secrete extracellular matrix that can adhere to the hydrogels,
although the PEG materials resist protein adsorption. Prestwich
et al. synthesized hydrogels by Michael type addition between
* Corresponding authors. Tel.: 86 22 2350 1645; fax: 86 22 2350 5598.
E-mail addresses: [email protected] (X.-g. Zhang), [email protected]
(C.-x. Li).
1
Da-yong Teng contributed equally with Zhong-ming Wu, and they are the
co-rst authors for this paper.
Contents lists available at ScienceDirect
Polymer
j ournal homepage: www. el sevi er. com/ l ocat e/ pol ymer
0032-3861/$ see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2009.12.003
Polymer 51 (2010) 639646
thiol-modied hyaluronic acid (HA), a major nonsulfated glucosa-
minoglycan constituent of extracellular matrix (ECM), and PEG
diacrylate [2325]. These injectable synthetic ECM (sECM) hydro-
gels had very high water content (>96%), biodegradability,
mechanical properties, and gelation times (w26 min) suitable for
potential clinical applications. In addition, Hiemstra et al. prepared
hydrogels via Michael type addition between thiol-modied
dextran and PEG tetra-acrylate or dextran vinyl sulfone conjugate
[2628]. The results indicate that these hydrogels are very prom-
ising for use in biomedical applications, since they can be rapidly
formed in situ and are biodegradable with adjustable degradation
times to match a particular application.
It is well known that chitosan (CS) is a natural polysaccharide
which has been widely used in pharmaceutical areas thanks to its
attractive biocompatibility, biodegradability, antimicrobial activity,
retention of growth factors, release of glucosamine and N-acetyl-
glucosamine monomers and oligomers, and stimulation of cellular
activities [2931]. Modied CS has been administered to humans in
the form of dressings for wounded soft and bone tissues [32,33],
such as articular cartilage [34]. In particular, PEG-modifying CS with
thermo-sensitive properties improves blood compatibility of CS,
and also decreases the enzyme degradation, which is of great
interest in drug delivery and tissue engineering [3538]. Insup et al.
synthesized a novel chitosanPEO in situ hydrogel via grafting of 2-
carboxyethyl acrylate to CS and Michael addition reaction was
processed between acrylate modied CS and PEO hexa-thiols [39].
However, the thermo-sensitive properties of the hydrogel were not
presented. In this paper, a novel in situ cross-linked hydrogel that
was formed under physiological conditions by Michael type addi-
tion between thiol-modied CS and PEG diacrylate was prepared.
Moreover, gelation time, mechanical properties, thermally swelling
properties and in vitro cell compatibility of the hydrogels were
characterized.
2. Experimental
2.1. Materials
Chitosan (degree of deacetylation 86%, with molecular mass of
300 kDa) was purchased from the Zhejiang Yuhuan Ocean
Biochemical Co., Ltd. (Zhejiang, China) andusedas received. N-acetyl-
L-cysteine (NAC), 1-ethy-3-(3-dimethyl- aminopropyl-carbodiimide)
hydrochloride (EDAC$HCl), 1-hydroxybenzotrizole (HOBt), 3-[4,5-
dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) and
Ellmans reagent were obtained from the J&K China Chemical Ltd.
Poly(ethylene glycol) diacrylate (PEGDA, M
w
2 kDa, degree of
substitution 87%, as determined by
1
HNMR) was synthesized by PEG
diol and Acryloyl chloride according to the literature [40]. All other
chemicals were of analytical grade and were used without further
purication.
2.2. Synthesis of chitosanNAC
ChitosanNAC (CSNAC) conjugates were prepared as described
by Kraulandet al. [41]. Typically, 500 mgof CS was dispersed inwater
(46 mL), andaddedHOBt (348.6mg, 2.58mmol) to above suspension
under stirring until a clear solution obtained. Afterwards, NAC
(842.0 mg, 5.16 mmol) was added to the solution followed by the
addition of a solution of EDAC$HCl (1978.5 mg, 10.32 mmol, 4 mL).
The pH of the reaction mixture was adjusted to 5 by the addition of
1 M HCl. The reaction mixture was incubated for 3 h at room
temperature under stirring. The resultant mixture was dialyzed (Mw
10000 cut) rst against 5 mMHCl containing 2 mMEDTAfor 3 days at
4

C in the dark, and then the same medium but additionally con-
taining 1% of NaCl and nally 1 mM HCl. The dialyzed fragments
were thenlyophilizedandstoredat 4

Cuntil further use. Thiol group

content on thiolated chitosan was determined using the Ellmans
reagent as described by Krauland et al. [41].
2.3. Gelation time and swelling tests
To determine the gelation time, solutions of CSNAC and PEGDA
(molar ratios of thiol to acrylate group were 1:1, 1:2 and 1:5) in
NaH
2
PO
4
/Na
2
HPO
4
buffer (PBS, pH 7.4, 0.1 M) were mixed and incu-
bated at 37

C by vortexing. In comparison, CSNAC was dissolved in

PBS(pH7.4, 0.1M) at 37

Ctoinducehydrogel states. Thegelationtime

was determined by a ow test utilizing a glass test tube inverting
method reported by Jeong et al. [42]. As the solution lost uidity in
1 min, the sample was regarded as a gel. For the swelling test, the
prepared hydrogels were allowed to swell and erode at 25 or 37

C
after applying 10 mL PBS. The swollen and degraded hydrogels were
weighed at various time intervals after removal of PBS. After each
weighing, PBS was refreshed. The swelling ratio of the hydrogels was
calculated from the initial dry hydrogel weight (W
0
) and the swollen
hydrogel weight after exposure to PBS (W
t
): Swelling ratio (%) W
t
/
W
0
.
2.4. Rheological experiments
A Stress Tech rheometer (Reologica Instruments Inc.) with stan-
dard steel parallel-plate geometry of 25 mm diameter was used for
the rheological characterization of all hydrogel samples (after gela-
tion). The test methods employed were oscillatory stress sweep and
frequency sweep. The stress sweep was performed on hydrogels to
determine the linear viscoelastic region (LVR) proles of them and
compare the storage modulus (G
0
) under the same physical condition.
The stress sweep was set up by holding the temperature (25

C) and
frequency (1 Hz) constant while increasing the stress level from0.1 to
100 Pa. We also subjected the hydrogels to a frequency sweep at
a xed shear stress (10 Pa) and temperature (25

C), the oscillatory
frequency was increased from0.01 to 20 Hz and the G
0
was recorded.
The plots of G
0
versus shear stress or frequency from the two sweep
tests were obtained directly from the software controlling the
rheometer.
2.5. Characterization of hydrogels
X-ray diffraction patterns were obtained by using a diffractom-
eter (XRD-6000, Shimadzu) equipped with a qq goniometer. The
operating conditions during the experiment were 40 kV and 50 mA
with Cu Ka radiation. The diffraction patterns were acquired over
a diffraction angle of 2q 1030

.
To examine thermal stability of hydrogels, the hydrogel samples
were measured using thermogravimetric analysis (TGA, Metzsh).
Decomposition proles of TGAwere recorded with a heating rate of
10

C/min in nitrogen between 20 and 650

C.
To observe the interior morphologies of hydrogels, the swollen
hydrogel samples were quickly frozen in liquid nitrogen and further
freeze-dried in a Freeze Drier (Christ alpha2-4 LD plus, Christ)
under vacuum at 90

C for at least 2 d until all the solvent was
sublimed. The freeze-dried hydrogels were then fractured carefully,
and the interior morphologies of the hydrogels were visualized by
using a scanning electron microscope (SEM, SS-550, Shimadzu).
Before the SEM observation, the hydrogel samples were xed on
aluminum stubs and coated with gold.
2.6. In vitro cell compatibility
The viability of HDFs and A549 cells cultured in a mediumwhere
the hydrogel was previously suspended and swelled was evaluated.
D.-y. Teng et al. / Polymer 51 (2010) 639646 640
In particular, the hydrogel was incubated in Dulbeccos modied
Eagles medium (DMEM) at 37

C for 5 d under orbital stirring at
120 rpm. After incubation, the medium conditioned by the
hydrogel, was centrifuged at 11800 rpm, 4

C for 30 min, and then
ltered to remove the hydrogel. The cells were seeded into 96-well
plates at 10,000 cells per well. The plates were then returned to the
incubator and the cells were allowed to growto conuence for 24 h.
Subsequently, the culture medium was replaced with the condi-
tioned medium. After 3 d and 7 d of incubation (37

C, 5% CO
2
),
20 mL of MTT solutions were used to replace the mixture in each
well. The plates were then returned to the incubator and incubated
for a further 4 h in 5% CO
2
at 37

C. Then, the culture medium and
MTT were removed. Isopropanol (150 mL) was then added to each
well to dissolve the formazane crystals. The plate was placed in 5%
CO
2
at 37

C for 10 min and for 15 min at 6

C before measurement.
The optical density was read on a microplate reader at 570 nm. Cell
viability was determined as a percentage of the negative control
(untreated cells). Each experiment was performed in triplicate and
results are reported as mean standard deviation.
3. Results and discussion
3.1. Synthesis and characterization of CSNAC
CSNAC was achieved by the covalent attachment of NAC to the
amine groups of CS. The coupling reaction was mediated by EDAC
and HOBt as we have previously reported [43]. And the amount of
immobilized thiol groups was strongly dependent on the molar
ratio of reagents (CS, NAC, HOBt and EDAC). Particularly, NAC/EDAC
molar ratio signicantly inuenced the amount of thiol group per
gram polymer. In the present research, we used two kinds of NAC/
EDAC molar ratios and Ellmans tests showed the contents of free
thiol group were 361.4 and 221.0 mmol/g, respectively (Table 1).
3.2. In situ formation and characterization of hydrogel
Upon exposure to air, the thiols on CSNAC are oxidized to form
a spontaneous, albeit slow, disulde-crosslinked hydrogel (CN
hydrogel). Then, to enhance the rate of cross-linking, hydrogels were
formed in situ via Michael type addition between CSNAC and
PEGDA in PBS at pH 7.4 and 37

C (CNP hydrogel, Fig. 1). The
concentration is dened as the dry weight of CSNAC per volume of
buffer.
Table 2 shows the gelation time of hydrogels with different
compositions and concentrations. As can be seen, the gelation time
of CNP hydrogel was much shorter than that of CNhydrogel. Indeed,
Michael type addition between thiol group and acrylate group has
a faster rate than the spontaneous oxidization of thiols [44]. For CNP
hydrogel, at the same concentration and molar ratio of thiol to
acrylate, the gelation time decreased from 45 to 30 min with
increasing the content of free thiols from221.0 to 361.4 mmol/g, and
the tendency is the same as CN hydrogel. The results indicate that
Michael type addition or the spontaneous oxidization was easier to
process as the content of free thiols increased. Then, for CNP2
hydrogel at the same concentration, the gelation time decreased
from45 to25 minwithdecreasing the molar ratio of thiol to acrylate
from 1:1 to 1:5, indicating that the increasing amounts of PEGDA
enhanced the rate of cross-linking between thiol and acrylate group
and consequently shortened the gelation time [24]. For CNP2 (1:1)
hydrogel, the gelation time decreased from 55 to 27 min by
increasing the concentration from 5 to 15 mg/mL, which was in
agreement with the results described by Hiemstra et al. [26,27]. For
CNP2 (1:1) hydrogel at the same concentration, the gelation time
was prolonged from 45 to 130 min by decreasing the temperature
from 37 to 25

C, indicating that Michael type addition preferred to
go under physiological temperature. In general, the gelation time of
hydrogels achieved by Michael type addition was short in compar-
ison with those by the spontaneous oxidization and can be modu-
latedby the concentrationof CSNACandPEGDA, the content of free
thiols in CSNAC, and the temperature.
Fig. 2 shows X-ray diffraction proles of CSNAC and CNP
hydrogel. For CS, at least seven polymorphs have been proposed
[4547], including tendonCS, annealed, I-2, L-2, form I,
form II, and noncrystalline. Pocker and Biswas have proposed
three forms: noncrystalline, hydrated (tendon) crystalline, and
anhydrous (annealed) crystalline [48]. The hydrated crystalline
structure gives a reection at 2q 10

(or peaks at around 8

and
12

), and the anhydrous crystalline structure gives one at 2q 15

.
Fig. 2 illustrates CS shows characteristic peaks around 2q 12.6

,
19

, and 26

[49]. The former peak corresponded to the hydrated

crystalline structure, while the broaden peak around 2q 21

for
CSNAC indicated the existence of an amorphous structure, sug-
gesting that the introduction of NAC changed the crystalline struc-
ture of CS. The crystalline PEGshows strong reections at 18.74

and
22.86

and weak reections at 26.77

, 30.5

, 35.9

, and 40

. In the
CNP hydrogel, the 12.6

reection for CS is absent, and there is

a broad peak in the 1826

region, which may have contributions

fromboth CS and PEG. This indicates that there was a decrease in CS
crystallization upon the introduction of PEG. As the PEG content in
CNP hydrogel increases, the intensity of characteristic reections
Table 1
Thiolated chitosan characterization and composition of hydrogel.
Sample CS:NAC:HOBT:EDAC
(molar ratio)
SH (mmol/g) SH:acrylate
(molar ratio)
CSNAC1 1:2:1:4 361.4 15.0 1:1
CSNAC2 1:2:1:8 221.0 12.4 1:1
1:2
1:5
Fig. 1. Demonstration of the in situ hydrogel formation.
D.-y. Teng et al. / Polymer 51 (2010) 639646 641
from the resulting hydrogel increased. It is natural that the high
content of PEGcan lead tolarger amounts of crystalline structures in
the hydrogel [50].
Thermal stabilities of CS, CSNAC and CNP hydrogel were
measured using TGA analysis. Fig. 3 shows the weight loss and
weight loss rate curves recorded with a heating rate of 10

C/min in
nitrogen between 20 and 650

C. CS and CSNAC showed single
stage thermal decomposition at about 250

C, indicating that thiol-
modied did not affect the thermal stability of CS. However, all the
CNP hydrogels showed a two-stage thermal decomposition. The
rst-stage decomposition temperature was also about 250

C,
suggesting the introduction of PEG via Michael type addition did
not decrease thermal stability of CS, compared with the incorpo-
ration of PEG via a traditional epoxyamine reaction [50].
Furthermore, the second-stage decomposition temperature was
approximately 380

C and the weight loss for CNP2 (1:1), CNP2
(1:2) and CNP2 (1:5) was 10%, 23% and 50%, respectively. The
weight loss results were similar to the contents of PEG in CNP
hydrogel and indicate that the decomposition stage was caused by
PEG. In particular, as the molar ratio of thiol to acrylate decreasing
from 1:1 to 1:5, the degradation temperature increased from 357.5
to 388.6

C, this is probably due to the high content of PEG
promoting the crystal growth of the hydrogel [51], which had been
demonstrated by X-ray diffraction.
The morphologies of the swollen CN and CNP hydrogel in PBS at
37

C for different time periods were observed by SEM. As shown in

Fig. 4AD, the SEM images of hydrogels exhibited a highly macro-
porous spongelike structure and the average mesh size is about
10 mm. With the increasing of cross-linking density, the decrease of
the pore size was attributed to the decreasing distance between CS
chains with increasing PEGDA concentration. After being dipped in
PBS for 24 h and 7 d, the chains in the network degraded and the
pores of the hydrogel increased (Fig. 4EH). The porous structure of
the hydrogel shows potential as scaffolds for cell inltration and
growth.
3.3. Rheology
Oscillatory stress sweep allows determination of the linear
viscoelastic region (LVR) proles and storage modulus (G
0
) of the
hydrogels as a function of the content of free thiols, concentration
and the ratio of thiol to acrylate (Fig. 5A). As can be seen, with
increasing cross-linking density of the hydrogel, the structure
breakdown occurred at higher shear stress levels. On the basis of
these data, a stress 10 Pa was chosen for comparison of various
hydrogel samples. Indeed, numerous published reports have
Table 2
Gelation time as function of composition and concentration of hydrogel.
CN1
a
CN2
a
CNP1 (1:1)
b
CNP2 (1:1)
b
CNP2 (1:1) 25

C
b
,
c
CNP2 (1:2)
b
CNP2 (1:5)
b
Concentration (mg/mL) 10 10 10 5 10 15 10 10 10
Gelation time (min)
d
145 15 180 16 30 4 55 5 45 4 27 4 130 13 39 4 25 2
a
CN corresponds to the hydrogel achieved by CSNAC.
b
CNP corresponds to the hydrogel achieved by CSNAC and PEGDA.
c
CNP2 (1:1) 25

C corresponds to the CNP2 (1:1) hydrogel which was formed at 25

C. And all the other hydrogels showed in table were formed at 37

C.
d
The gelation time was determined by a ow test utilizing a test tube inverting method.
Fig. 2. X-Ray diffraction patterns for chitosan (A), chitosanNAC (B), PEGDA (C), CNP2
(1:1) hydrogel (D), and CNP2 (1:5) hydrogel (E).
Fig. 3. Thermogravimetric analysis of chitosan, chitosanNAC, and CNP hydrogels.
(A) the weight loss curves, (B) the weight loss rate curves.
D.-y. Teng et al. / Polymer 51 (2010) 639646 642
Fig. 4. SEM photographs of (A) CN hydrogel, (B) CNP1 (1:1) hydrogel, (C) CNP2 (1:1) hydrogel, and (D) CNP2 (1:5) hydrogel dipped in water for 0.5 h at 37

C; (E) CNP1 (1:1)
hydrogel and (F) CNP2 (1:1) hydrogel for 24 h at 37

C; (G) CNP1 (1:1) hydrogel and (H) CNP2 (1:1) hydrogel for 7 days at 37

C.
D.-y. Teng et al. / Polymer 51 (2010) 639646 643
suggested that the measured G
0
has a strong relationship to the
number of effective intermolecular cross-links formed in the
hydrogel network [5254]. Therefore, the evolution of G
0
as a func-
tion of cross-linking density was monitored to assess the extent of
effective intermolecular cross-links formed in the hydrogel
networks. As we can see, all the CNP hydrogels have a higher G
0
than
CNhydrogel, indicating that the introductionof PEGvia Michael type
addition increased the cross-linking density. Then, for the CNP
hydrogel, the higher cross-linking density and elasticity of hydrogel
can be achieved by increasing the contents of free thiols in CSNAC
and decreasing molar ratio of thiol to acrylate. For the CNP hydrogel
with different concentration, the higher concentration of CSNAC
made the achieved hydrogel having a higher cross-linking density,
suggesting that increasing the concentration of CSNAC can get
a higher content of free thiols under the same volume.
Frequency sweep tests are widely used to obtain information
about the stability of three-dimensional cross-linked networks
[52,55,56]. Fig. 5B shows the plot between G
0
and oscillatory
frequency, the data obtained for all the hydrogels is characterized
by G
0
exhibiting a plateau in the range of 0.012 Hz that is indicative
of a stable, cross-linked network. However, at higher frequencies
(220 Hz), all the hydrogels showed an increase in G
0
, and the
higher cross-linking density made higher rate of increase. The
magnitude of viscoelastic response elicited by a polymeric network
is governed primarily by the length of the exible polymer chains
and the nature of the imposed mechanical motion [52,55]. Longer
chains have characteristic longer relaxation times and cause a low
cross-linking density in the hydrogel network. At higher frequen-
cies, long chains fail to rearrange themselves in the time scale of the
imposed motion and therefore stiffen up, which was characterized
by a sharp increase in G
0
.
3.4. Hydrogel swelling
To study the swelling and degradation of hydrogel, the prepared
hydrogel was dried under vacuumat roomtemperature to a constant
weight, and then the dried hydrogel was immerged into PBS and
allowed to swell at 25 or 37

C. All the hydrogels swelled rapidly and

reached equilibrium within 0.5 h at 37

C, but within 4 h at 25

C
(Supporting Information). Fig. 6 shows the equilibriumswelling ratio
of CN and CNP hydrogels. For CNP2 (1:1) hydrogel at different
concentrations, the equilibriumswelling ratio decreased from1336%
to 888% when increasing the concentration from 5 to 15 mg/mL
(Fig. 6A), indicating that the hydrogel prepared at a high concentra-
tion has a tighter structure and a lower water-holding capacity than
at a lowconcentration. For CNP2 hydrogel at different molar ratio of
thiol to acrylate, the equilibriumswelling ratio decreased from1195%
to 568% by decreasing the molar ratio of thiol to acrylate from1:1 to
1:5 (Fig. 6B), suggesting that the construction of the hydrogel with
a higher content of PEGwas too tight to hold water. For CNP hydrogel
at same condition, the equilibrium swelling ratio decreased from
1195% to 907% when increasing the content of free thiols from 221.0
to 361.4 mmol/g (Fig. 6C), implying the high content of free thiols
caused a tight network structure of the hydrogel. For CN hydrogel
with different contents of free thiols, the equilibrium swelling ratios
were similar and lower than all the CNP hydrogels (Fig. 6C). The
results indicate that the thiol oxidation in the CN hydrogel was
similar, as well as the construction of hydrogel. Moreover, there was
a longer distance between the CS chains in CNP hydrogel than CN
hydrogel due to the introduction of PEGmoieties, so CNhydrogel has
a tight network structure compared to CNP hydrogel.
Furthermore, the equilibrium swelling ratios of CNP hydrogels
were higher at 37

C than at 25

C compared to CN hydrogel. The
phenomenon was in good agreement with the results described by
El-Sherbiny et al. [57] and Khurma et al. [58], indicating the
temperature-responsive behavior of CNP hydrogel. This behavior is
due to the dissociation of the hydrogen bond in chitosan within the
hydrogel matrices at higher temperature and thus allowing more
water to move into the matrix of hydrogel [59].
The CN and CNP hydrogels were degradable under physiological
conditions. There are two kinds of degradation in CNP hydrogels.
One is the degradation through hydrolysis of the ester bond
between the thioether and PEG, as the hydrogels via Michael type
addition [2328]; the other is the biodegradation of CS. Indeed, CS
can be easily depolymerized by a variety of hydrolases including
lysozyme, pectinase, cellulases, hemicellulases, lipases, and
amylases [29]. After reaching swelling equilibrium, all the hydro-
gels were rapidly degraded in the rst 24 h, and then the degra-
dation became slow and the persistence of hydrogel remnants was
also observed within six weeks (Supporting Information).
3.5. In vitro cell compatibility
The biocompatibility of CN and CNP hydrogels was evaluated in
vitro on HDFs and A549 cells by indirect methods. In this method,
the growth medium was rst conditioned for 5 d with the
Fig. 5. Rheological behaviors of various hydrogel samples. (A) Oscillatory shear stress
sweep, (B) oscillatory frequency sweep. HCNP2 (1:1): CNP2 (1:1) hydrogel at high
concentration (15 mg/mL); LCNP2 (1:1): CNP2 (1:1) hydrogel at low concentration
(5 mg/mL).
D.-y. Teng et al. / Polymer 51 (2010) 639646 644
hydrogel and then used to incubate the cells. After 3 and 7 d of
incubation, cell viability was determined by the MTT assay and
expressed as a percentage ratio between viable cell number incu-
bated in the growth medium conditioned and not conditioned
by the hydrogel.
Fig. 7 shows cell viability (MTT assay) of CN2 and CNP2 hydro-
gels with different cells and incubated time. As can be seen, for two
different cell types, the cell viability of CNP hydrogel was a little
lower than that of CN hydrogel after 3 d incubated, which may be
due to the rapider degradation of CNP hydrogel, and the degraded
segment of hydrogel may contain some substances that interfere
with the cell viability. And the cells appeared to be recovered by 7 d,
all the cell viabilities were maintained over 80%. These properties
suggest that the CNP hydrogel has many potential clinical appli-
cations in drug delivery and tissue engineering.
4. Conclusion
Thiol-modied chitosan with the contents of free thiol groups
361.4 and 221.0 mmol/g was conveniently prepared by a one-pot
synthesis procedure at room temperature. Hydrogels were rapidly
Fig. 6. Equilibrium swelling ratio of CN and CNP hydrogels under different conditions.
(A) CNP2 (1:1) hydrogels under different concentration, (B) CNP2 hydrogels under
different temperature and molar ratio of acrylate to thiol, (C) CN and CNP hydrogel
under different temperatures.
Fig. 7. Cell viability (MTT assays) for HDFs (A) and A549 (B) cells grown on CN2 and
CNP2 (1:1) hydrogels.
D.-y. Teng et al. / Polymer 51 (2010) 639646 645
formed in situ under physiological conditions by Michael type
addition upon mixing aqueous solutions of CSNAC and (PEGDA) at
different concentrations and ratios of thiol to acrylate. Their
mechanical and swelling properties are readily controlled by the
content of free thiols in CSNAC, temperature, and concentration of
CSNAC and PEGDA. The hydrogels were degradable under physio-
logical conditions. Invitro cell compatibility study on HDFs and A549
cells indicates CNP hydrogel has good compatibility. These hydrogels
present great potential in biomedical applications including drug
delivery, as well as injectable tissue engineered scaffolds.
Acknowledgments
This work was supported in part by the National Natural Science
Foundation of China (Grant No. 20804021), the Natural Science
Foundation of Tianjin (Grant No. 08JCYBJC00300), the Ph.D.
Programs Foundation for New Teachers of Ministry of Education of
China (Grant No. 200800551030), and Tianjin Health Bureau
Scientic Research Found (07kz47) are gratefully acknowledged.
Appendix. Supplementary data
Supplementary information associated with this article could be
found on-line, at doi:10.1016/j.polymer.2009.12.003.
References
[1] Hoffman AS. Adv Drug Deliv Rev 2002;43:312.
[2] Peppas NA, Hilt JZ, Khademhosseini A, Langer R. Adv Mater 2006;18:134560.
[3] Hoare TR, Kohane DS. Polymer 2008;49:19932007.
[4] Lee KY, Mooney DJ. Chem Rev 2001;101(7):186979.
[5] Nuttelman CR, Rice MA, Rydholm AE, Salinas CN, Shah DN, Anseth KS. Prog
Polym Sci 2008;33:16779.
[6] Hate A, Amsden B. J Control Release 2002;80:928.
[7] Packhaeuser CB, Schnieders J, Oster CG, Kissel T. Eur J Pharm Biopharm
2004;58:44555.
[8] Kretlow JD, Klouda L, Mikos AG. Adv Drug Deliv Rev 2007;59:26373.
[9] Van Tomme SR, Storm G, Hennink WE. Int J Pharm 2008;355:118.
[10] He CL, Kim SW, Lee DS. J Control Release 2008;127:189207.
[11] Sawhney AS, Pathak CP, Hubbell JA. Macromolecules 1993;26:5817.
[12] Mi FL, Shyu SS, Wong TB, Jang SF, Lee ST, Lu KT. J Appl Polym Sci 1999;74:
1093107.
[13] Kuo CK, Ma PX. Biomaterials 2001;22:51121.
[14] Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF,
Hennink WE. Biomaterials 2005;26:212935.
[15] Huang X, Lowe TL. Biomacromolecules 2005;6:21319.
[16] Haines-Butterick L, Rajagopal K, Branco M, Salick D, Rughani R, Pilarz M, et al.
Proc Natl Acad Sci U S A 2007;104:77916.
[17] Choi HS, Yui N. Prog Polym Sci 2006;31:12144.
[18] Qiu B, Stefanos S, Ma J, Lalloo A, Perry BA, Leibowitz MJ, et al. Biomaterials
2003;24:118.
[19] Tae G, Kim YJ, Choi WI, Kim M, Stayton PS, Hoffman AS. Biomacromolecules
2007;8:197986.
[20] Metters A, Hubbell JA. Biomacromolecules 2005;6:290301.
[21] Van de Wetering P, Metters AT, Schoenmakers RG, Hubbell JA. J Control
Release 2005;102:61927.
[22] Rizzi SC, Hubbell JA. Biomacromolecules 2005;6:122638.
[23] Shu XZ, Ghosh K, Liu YC, Palumbo FS, Luo Y, Prestwich GD, et al. J Biomed
Mater Res 2004;68A:36575.
[24] Ghosh K, Shu XZ, Mou R, Lombardi J, Prestwich GD, Rafailovich MH, et al.
Biomacromolecules 2005;6:285765.
[25] Shu XZ, Ahmad S, Liu YC, Prestwich GD. J Biomed Mater Res 2006;79A:90212.
[26] Hiemstra C, van der Aa LJ, Zhong Z, Dijkstra PJ, Feijen J. Macromolecules
2007;40:116573.
[27] Hiemstra C, van der Aa LJ, Zhong Z, Dijkstra PJ, Feijen J. Biomacromolecules
2007;8:154856.
[28] Hiemstra C, Zhong Z, van Steenbergen MJ, Hennink WE, Feijen J. J Control
Release 2007;122:718.
[29] Ravi Kumar MNV, Muzzarelli RAA, Muzzarelli C, Sashiwa H, Domb AJ. Chem
Rev 2004;104:601784.
[30] Sashiwa H, Aiba S. Prog Polym Sci 2004;29:887908.
[31] Rinaudo M. Prog Polym Sci 2006;31:60332.
[32] Gingras M, Paradis I, Berthod F. Biomaterials 2003;24:165361.
[33] Wang YC, Lin MC, Wang DM, Hsieh HJ. Biomaterials 2003;24:104757.
[34] Francis Suh JK, Matthew HWT. Biomaterials 2000;21:258998.
[35] Bhattarai N, Ramay HR, Gunn J, Matsen FA, Zhang MQ. J Control Release
2005;103:60924.
[36] Bhattarai N, Matsen FA, Zhang MQ. Macromol Biosci 2005;5:10711.
[37] Wua J, Wei W, Wang LY, Su ZG, Ma GH. Biomaterials 2007;28:222032.
[38] Zhang XG, Zhang HJ, Wu ZM, Wang Z, Niu HM, Li CX. Eur J Pharm Biopharm
2008;68:52634.
[39] Kim M-S, Choi Y-J, Noh I, Tae G. J Biomed Mater Res 2007;83A:67484.
[40] Cruise GM, Scharp DS, Hubbell JA. Biomaterials 1998;19:128794.
[41] Krauland AH, Hoffer MH, Bernkop-Schnu rch A. Drug Dev Ind Pharm
2005;31:88593.
[42] Jeong B, Bae YH, Kim SW. Macromolecules 1999;32:70649.
[43] Wang X, Zheng C, Wu Z, Teng D, Zhang X, Li C, et al. J Biomed Mater Res Part B
Appl Biomater 2009;88B:15061.
[44] ShuXZ, LiuYC, Palumbo FS, LuoY, PrestwichGD. Biomaterials 2004;25:133948.
[45] Ogawa K, Hirano S, Miyanishi T, Yui T, Watanabe T. Macromolecules 1984;17:
9735.
[46] Saito H, Ryoko T, Ogawa K. Macromolecules 1987;20:242430.
[47] Samuels RJ. J Polym Sci Polym Phys Ed 1981;19:1081105.
[48] Pocker Y, Biswas SB. Biochemistry 1980;19:50439.
[49] Radhakumary C, Nair PD, Reghunadhan Nair CP, Mathew S. J Appl Polym Sci
2009;114:287386.
[50] Kiuchi H, Kai W, Inoue Y. J Appl Polym Sci 2008;107:382330.
[51] Zhao W, Yu L, Zhong X, Zhang Y, Sun J. J Macromol Sci Phys 1995;B34(3):2315.
[52] Anseth KS, Bowman CN, Brannon-Peppas L. Biomaterials 1996;17:164757.
[53] Santovena A, Alvarez-Lorenzo C, Concheiro A, Llabres M, Farina JB. Biomate-
rials 2004;25:92531.
[54] Lee KY, Bouhadir KH, Mooney DJ. Biomaterials 2004;25:24616.
[55] Kavanagh GM, Ross-Murphy SB. Prog Polym Sci 1998;23:53362.
[56] Lutolf MP, Hubbell JA. Biomacromolecules 2003;4:71322.
[57] El-Sherbiny IM, Abdel-Bary EM, Harding DRK. J Appl PolymSci 2006;102:97785.
[58] Khurma JR, Nand AV. Polym Bull 2008;59:80512.
[59] Kim SJ, Park SJ, Kim SI. React Funct Polym 2003;55:539.
D.-y. Teng et al. / Polymer 51 (2010) 639646 646