Colloids and Surfaces B: Biointerfaces 128 (2015) 515–521

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces

journal homepage: www.elsevier.com/locate/colsurfb

Systemic modulation of the stability of pluronic hydrogel by a small

amount of graphene oxide
Da-Ae Won, Manse Kim, Giyoong Tae ∗
School of Materials Science and Engineering, Gwangju Institute of Science and Technology, 123 Cheomdan-gwagiro, Oryong-dong, Buk-gu, Gwangju
500-712, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: Thermo-sensitive and injectable hydrogels have been widely investigated for drug delivery, tissue engi-
Received 22 August 2014 neering, and other biomedical applications. Pluronic copolymers can form thermo-sensitive physical gel
Received in revised form 26 February 2015 state, thus applicable for injectable hydrogels. However, they are not stable in vivo, showing a very fast
Accepted 2 March 2015
dissolution, which limits their applications. We propose a novel Pluronic-based physical hydrogel with
Available online 7 March 2015
enhanced stability by simply adding a small quantity of graphene oxide (GO) which has a large surface
area and can make strong interactions with Pluronic. Further carboxylated GO could act as a more efficient
Keywords:
additive. The addition of GO increased the moduli of hydrogels, but more importantly, it enhanced the
Pluronic
Nanogel
stability of Pluronic gel dramatically. The in vitro dissolution rate of Pluronic hydrogel could be system-
Graphene oxide atically modulated by increasing GO content. Upon subcutaneous injection at a sol state, GO-containing
Degradation hydrogel induced a stable gel state, and was maintained over several weeks whereas very fast degra-
Release dation was observed without the addition of GO. Furthermore, histological analyses demonstrated that
Injectable the GO-containing Pluronic hydrogel was biocompatible and showed no severe inflammatory response.
Similarly, GO-containing hydrogel resulting from the packing of Pluronic-based nanogel also showed
the more enhanced stability by the addition of GO both in vitro and in vivo. In both systems, hydrogels
with remarkably enhanced stability by the addition of GO were also effective for the sustained release
of loaded protein, and the release rates were mainly determined by the degradation rates of hydrogels.
Thus, these GO-containing Pluronic systems can be used as a thermo-sensitive injectable system with a
sufficient stability in vivo.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction Pluronic, biocompatible triblock copolymers composed of

Hydrogels have many advantages such as biocompatibility,
ability for absorbing water and low inflammatory responses
[1], applicable as a reservoir for slow elution of drugs or pro-
teins [2]. Particularly, injectable hydrogels are more preferred for
drug/protein delivery and tissue engineering. Injectable hydrogels
can be made by using responses to environmental stimuli such as
temperature [3–6] or pH [7]. For example, these materials are in a
flow state before administration, but once injected, they rapidly
become a gel under physiological conditions. Such systems can
provide several advantages including delivery of large amount in
a minimally invasive manner, gelation at the desired tissue sites,
and minimized scar formation, thus reducing the risk of infection
[6,8]. In addition, bioactive molecules or cells can be incorporated
by simple mixing before injection [9].
∗ Corresponding author. Tel.: +82 62 715 2305; fax: +82 62 715 2304.
E-mail address: gytae@gist.ac.kr (G. Tae).
hydrophobic poly(propylene oxide) and hydrophilic poly(ethylene
oxide), are a good example of thermo-sensitive polymers as
an injectable hydrogel systems [10]. Pluronic can self-assemble
into micelles in aqueous solutions above critical micelle con-
centration (CMC) and critical micelle temperature (CMT), and
exhibit temperature-responsive sol–gel transition behaviors [5].
Above CMT and at high concentration of Pluronic, gelation
occurs due to micelle packing, resulting in a dramatic change in
their rheological properties [11,12]. Using this thermo-reversible
gelation, Pluronic F127 (PF127) has been studied widely, espe-
cially as an injectable system for local drug delivery [13].
However, the application of PF127 hydrogel has been limited
because of a very fast dissolution, thus fast drug release as
well as weak mechanical properties in vivo. Therefore, various
reports have attempted to enhance the stability of PF127 gel
in vivo. Chen et al. showed a thermo-responsive nanocompos-
ite hydrogel composed of hexamethylene diisocyanate, PF127
and hyaluronic acid [6]. Our group also reported an injectable
and photo-polymerizable Pluronic based hydrogel by using a

http://dx.doi.org/10.1016/j.colsurfb.2015.03.002
0927-7765/© 2015 Elsevier B.V. All rights reserved.
516 D.-A. Won et al. / Colloids and Surfaces B: Biointerfaces 128 (2015) 515–521
sufficiently long induction period during photo-polymerization
[14].
Recently, there has been a remarkably increased attention
to biological applications of graphene and graphene oxide (GO).
Graphene is a single layer of sp2 -hybridized carbon atoms in a 2-d
crystalline lattice [15]. GO is an oxidized derivative of graphene to
make it water dispersible. The planar structure and ultra-high sur-
face area of graphene and GO make them suitable for loading a large
number of substances including metal, drugs, biomolecules, and
even cells [16–18]. Graphene is very hydrophobic, thus graphene
can interact with hydrophobic molecules based on ␲–␲ stacking
and van der waals interaction, whereas GO is negatively charged
with carboxyl and hydroxyl groups, so hydrogen bonding and
charge interaction as well as hydrophobic association can con-
tribute to the interaction with other molecules. Therefore, GO can
associate with variety of molecules including amphiphilic poly-
mers. Based on the non-covalent interaction of GO with variety of
molecules, GO composite was shown to significantly increase the
mechanical and thermal properties of host materials [18–20]. For
e.g., Zhang et al. added GO as a nano-filler into a polyvinyl alcohol
(PVA) matrix, and showed the increase in tensile strength, break-
ing elongation, and compressive strength [18]. There have also
been several studies reporting the formation of hydrogel by mix-
ing GO with polymers that can provide sufficient interaction with
GO [21–25], including PVA [21,22], DNA [23], chitosan [24], and
Pluronic [25] in certain concentration ranges. Also, GO modified
with cyclodextrin (CD) was used to form complex hydrogel with
azo-functionalized polymer [26], and Pluronic-coated GO was used
to make a complex hydrogel with CD [27]. In all of these reports,
both GO or modified GO and polymer were necessary for hydrogel
formation. Without GO, no gelation occurs, thus GO functions as a
main gelator. In contrast, there has been no study reporting that
the addition of a small amount of GO can increase and modulate
the thermodynamic stability of pre-existing physical gel systems.
GO interacts strongly with Pluronic, mainly by the hydrophobic
association between PPO part of Pluronic and GO, so the Pluronic
coating of GO nanosheet (∼40 nm) could increase the colloidal sta-
bility of GO in physiological media and neutralize the surface charge
of carboxylated GO [28]. In addition, the presence of carboxyl group
and hydroxyl group of carboxylated GO can induce the hydrogen
bonding with Pluronic, evidenced by the effect of pH in the inter-
action between carboxylated GO and Pluronic [25], similar to the
interaction between PVA and GO [22].
In this paper, we propose GO as an effective additive to
increase and modulate the stability of physical hydrogel systems,
here hydrogels formed by Pluronic or Pluronic-based nanogel, in
addition to its contribution to the mechanical properties of the
hydrogels. We investigate the physical gel system at high concen-
tration of Pluronic (17 wt%) or Pluronic-based nanogel (11 wt%),
where they can form hydrogel by themselves by the packing of
micelle or nanogel. However, since these gels are achieved by
lyotropic transition, the hydrogels formed by themselves are not
stable but quickly disappear when excess water is added. We report
here that by adding a very small amount GO to Pluronic gel systems,
the stability of physical gels already formed without GO can be
greatly enhanced and modulated in an excess water environment
like physiological situation.
2. Materials and methods
2.1. Materials
Pluronic F127 (PEO100 PPO65 PEO100, MW 12.6 kDa) was
donated from BASF Corp. (Seoul, Korea). Acryloyl chloride, tri-
ethylamine, anhydrous toluene, potassium phosphate monobasic,
sodium phosphate dibasic, potassium chloride, sodium chloride,
and sodium azide were purchased from Sigma-Aldrich (St. Louis,
MO, USA). Anhydrous diethyl ether was purchased from Fisher
Scientific Inc. (Pittsburgh, PA, USA). 4-(2-Hydroxyethoxy) phenyl-
(2-hydroxyl-2-propyl) ketone (Irgaure 2959) was obtained from
Ciba Specialty Chemicals Inc. (Basel, Switzerland). Single layer
graphene oxide (X, Y dimension < 500 nm) was obtained from
Angstrom Materials Inc. (Dayton, OH, USA) as a 0.5% (w/v) solu-
tion. Chloroacetic acid and sodium hydroxide were purchased from
Aldrich (Milwaukee, WI, USA). Dialysis membrane [MWCO 50,000
& 3500] was the product of Spectrum Laboratories, Inc. (Houston,
TX, USA). 0.2 ␮m cellulose sterilization syringe filters and 0.8 ␮m
cellulose sterilization syringe filters were purchased from Toyo
Roshi Kaisha. Ltd. (Tokyo, Japan). 0.2 ␮m nylon syringe filter was
purchased from Whatman (Florham Park, NJ, USA). Human VEGF
ELISA Development Kit (900-K10) was purchased from Peprotech
(Rocky Hill, NJ, USA).
2.2. Preparation of Pluronic-based nanogel
Chemically crosslinked, Pluronic-based nanogel was prepared
by simple photo-crosslinking of diacylated Pluronic [3]. First,
diacrylated Pluronic F127 (DA-PF 127) was synthesized as previ-
ously reported [3,29,30]. Shortly, dried Pluronic F127 were reacted
with 10 molar excess of triethyamine and acryloyl chloride in
anhydrous toluene with stirring under argon over 17 h, then it
was precipitated in anhydrous diethyl ether in an ice-water bath,
filtered, and dried under vacuum for a few days. The degree of acry-
lation of Pluronic was over 98%, as determined by comparing the
acryl protons ( CH2 , 5.7–6.4 ppm) and methyl protons in propyl-
ene oxide units ( CH3 , 1.1 ppm) in 40 MHz 1 H NMR spectroscopy
(D2 O, JNM-ECX-400P, JEOL, Japan). After that, 0.77 wt% of DA-PF
127 in de-ionized water (DIW) was photo-crosslinked by adding
a photoinitiator Irgacure 2959 in 70% (v/v) ethanol and DIW and
UV-irradiation using an unfiltered UV lamp (VL-4.LC, 8 W, Vilber
Lourmat, France) for 15 min with 1.3 mW cm−2 intensity. Finally,
the unreacted precursors were removed by dialysis (MWCO 50,000)
for 1 day.
2.3. Preparation of carboxylated graphene oxide
The suspension of carboxylated GO sheet was prepared by
chemical treatment and ultrasonication, according to the reference
[31]. To 10 ml of GO (0.1 w/v%) dispersion in DIW, chloroacetic
acid (150 mg) and sodium hydroxide (200 mg) were added. The
solution was stirred at 45 ◦ C for overnight. Then, the solution was
sonicated by an ultrasonic probe with 750 W, 30% intensity (Vibra-
cell VCX 500, Sonics & Materials Inc., Newtown, CT, USA) for 2 h.
The solution was neutralized by dialysis (dialysis bag MWCO 3500)
against distilled water for a few days. After dialysis, the suspen-
sion was sonicated for 30 min and then filtered through a 0.8 ␮m
cellulose acetated filter. The size of carboxylated GO was ∼400 nm,
analyzed by using an electrophoretic light-scattering spectropho-
tometer (632.8 nm, ELS-8000, Otsuka Electronics Co., Toyko, Japan).
2.4. Preparation of the sol state of injectable hydrogels
Two kinds of injectable state hydrogels were prepared by dis-
solving PF 127 itself, or Pluronic-based nanogel in GO solution at
4 ◦ C for 48 h by using a rotary shaker after vortexing. The final con-
centration of PF 127 was 17 wt%, and that of nanogel was 11 wt%.
Three different concentration of GO was used: 0.04, 0.08, and
0.1 wt%.
 D.-A. Won et al. / Colloids and Surfaces B: Biointerfaces 128 (2015) 515–521
2.5. Sol–gel transition phase diagram
The sol–gel transition phase diagrams of both bare hydrogels
and GO-containing hydrogels were determined by a vial tilting
method. Each sample in PBS solution was prepared in 1.5 ml tube,
and samples were placed in a water-bath. Temperature was started
at 4 ◦ C and gradually increased by 3 ◦ C. The temperature of gel point
was decided when the fluidity was not observed by inverting the
test tube for 1 min. The concentration of GO was set at 0.1 wt%
(n = 3).
2.6. In vitro degradation rates of hydrogels
To analyze the stability of various hydrogels, the degradation
rates of hydrogels in vitro were measured. Hydrogels made of
Pluronic itself, Pluronic with GO or carboxylated GO, nanogel itself,
or nanogel with carboxylated GO were placed at the bottom of 15 ml
tubes. The volume of each hydrogel was 200 ␮l. After gelation by
increasing temperature to 37 ◦ C, 5 ml of release buffer (PBS with
2 mM% sodium azide) was added to each tube. Samples were kept
in a shaking incubator (SI-600R, Lab. Companion, JEIO TECH, city,
Korea) at 37 ◦ C and 100 rpm. The whole release buffer was replaced
with a fresh one daily, and the remaining volumes of hydrogels
were measured at each time points (n = 3).
2.7. In vitro release rates of a growth factor from hydrogels
The release rates of vascular endothelial growth factor (VEGF)
from hydrogels containing carboxylated GO were characterized.
100 ng of VEGF was loaded into 200 ␮l of Pluronic (17 wt%) or
Pluronic-based nanogel (11 wt%) with 0.1 wt% of carboxylated GO
and placed at the bottom of a 15 ml tubes. After gelation by increas-
ing temperature to 37 ◦ C, 5 ml of release buffer (PBS with 2 mM%
sodium azide and 0.01 wt% BSA) was added to each tube and incu-
bated at 37 ◦ C under shaking at 100 rpm. The whole release buffer
was replaced with a fresh one daily, and the released amounts of
VEGF from hydrogels were calculated by using an ELISA kit (n = 3).
2.8. Measurement of rheological properties
The rheological properties of various hydrogels were measured
by using a rheometer (Gemini, Malvern Instruments, Malvern,
Worcestershire, UK) with parallel plate geometry of 15 mm diam-
eter and a roughened surface. The gap size was set to 0.45 ␮m and
the sample volume was 100 ␮l. The temperature of plate was low-
ered to 4 ◦ C, and then the gelation kinetics was measured at 37 ◦ C.
A single frequency at 1 rad/s was used to get gelation kinetics with
different concentrations of GO, and then a frequency sweep with
0.1% strain was performed at 37 ◦ C.
2.9. In vivo animal experiments
In vivo stability and biocompatibility of various hydrogels were
studied using male BALB/c mice (8 weeks old). All animals were
obtained from Orient Bio Inc. (Seoul, Korea) and were handled
in accordance with the guidelines of the Animal Care and Use
Committee of Gwangju Institute of Science and Technology (GIST).
Hydrogels were prepared as a sol state by lowering temperature
at 4 ◦ C and injected subcutaneously at the back of mice using a
syringe with a 22G needle. To characterize the in vivo stability
of injected hydrogels, the sizes of injected hydrogels were mon-
itored at determined time points: 1day, 1 week, 2 weeks, and
8 weeks.
For analyzing biocompatibility of the GO-containing hydrogel,
histochemical staining of injected sites was done. Animals were
sacrificed and the skin parts containing injected gels were cut out.
517
These skins were fixed in 10% formalin, dehydrated in alcohols, and
embedded in paraffin. The paraffin was sectioned by using a micro-
tome at 5 ␮m thickness, and hematoxylin and eosin (H&E) staining
was applied. The photographs of stained sections were taken by
optical microscopy.
3. Results and discussions
To increase and modulate the stability of physical gel systems,
GO was used as an effective additive in this study. Our group pre-
viously reported that the addition of a small amount of Pluronic
into GO suspension could induce an insoluble, but rather fragile,
gel-like state in a limited concentration range [25]. In this case, GO
acts as a main gelator. No gel-like state was formed without GO
or below 0.3 wt% of GO. And, the concentration range of Pluronic
was limited to below 1 wt%. Above 1 wt% of Pluronic, the gel-like
state disappeared. So, both GO and Pluronic were necessary for gel
formation, and the excess amount of Pluronic prevented the forma-
tion of a gel state. In contrast, in this study, the lyotropic physical
hydrogels formed by the packing of Pluronic micelle or Pluronic-
based nanogel at high concentration of Pluronic were the systems
of interest. The addition of a very small amount of GO (up to 0.1 wt%)
could greatly improve and systematically modulate the stability of
hydrogels formed by Pluronic micelle (17 wt%) or Pluronic-based
nanogel (11 wt%) in an open environment both in vitro and in vivo.
Thus, the role of GO is an additive, not a gelator, in this study.
3.1. In vitro stability of Pluronic hydrogel
By the micelle packing, Pluronic solution shows a thermo-
reversible, lyotropic sol–gel phase transition. To characterize the
effect of the addition of GO on the stability of Pluronic hydrogels,
the degradation rate of Pluronic hydrogel at 17 wt% solution in PBS
in vitro was measured in a sink condition under shaking at 37 ◦ C.
The remaining volume of hydrogel in the presence of excess vol-
ume of buffer was monitored and the buffer was exchanged with
a fresh one daily. As expected, without GO, Pluronic gel itself was
degraded very fast within 1 day in an excess water environment.
In contrast, by adding 0.1 wt% GO, the stability of Pluronic hydrogel
was greatly improved (Fig. 1a).
Since hydrogen bonding, in addition to hydrophobic associa-
tion, can contribute to the interaction between Pluronic and GO,
the difference between normal GO and further carboxylated GO
for improving the stability of Pluronic hydrogel (Fig. 1a) was com-
pared. Further carboxlyated GO was prepared by reacting GO with
chloroacetic acid (ClCH2 COOH) in an alkaline condition, which con-
verts hydroxyl groups (–OH) of GO into carboxyl group (–COOH)
[31]. After chemical FTIR data (Fig. 1b) shows the increased C O
(1670 cm−1 ) and C O (1100 cm−1 ) bond compared to –OH group
(3400 cm−1 ) after treatment, which supports the carboxylation of
GO [32]. The difference between carboxylated GO and normal GO
for improving the stability of Pluronic hydrogel was also large; most
of gel was degraded in day 5 with normal GO, whereas more than
40% of hydrogel was remained with carboxylated GO at the same
time. This large difference might result from the more increased
hydrogen bonding that carboxylated GO could provide as well as
the better dispersion of carboxylated GO in physiological buffer
due to an increased negative charge, thus providing more uniform
distribution in the hydrogel compared to normal GO. Therefore,
carboxylated GO was used instead of normal GO for all other exper-
iments.
The stability of GO-containing Pluronic hydrogel (17 wt%) was
analyzed by varying the amount of added GO. Compared to Pluronic
hydrogel itself, the degradation rate was greatly reduced even at
0.04 wt% GO. And, by increasing GO concentration, the degradation
518 D.-A. Won et al. / Colloids and Surfaces B: Biointerfaces 128 (2015) 515–521
Fig. 1. (a) Degradation of hydrogel composed of 17 wt% Pluronic with 0.1 wt% car-
boxylated or unmodified GO. (b) FT-IR spectra of carboxylated GO and unmodified
GO. Degradation of (c) hydrogel composed of 17 wt% Pluronic and carboxylated GO
with various concentrations of GO and (d) hydrogel composed of 11 wt% Pluronic-
based nanogel and carboxylated GO with various concentrations of GO. PBS with
2 mM sodium azide in a shaking incubator at 37 ◦ C and 100 rpm was used (n = 3).
rate was significantly reduced further systematically; the degrada-
tion rate with 0.08 wt% GO was similar to that of 0.1 wt% GO until
day 3, but after 4 days, a noticeable difference started to be observed
(Fig. 1c). Therefore, time for complete degradation was prolonged
from 5 days to 10 days, and more than 15 days by increasing the con-
centration from 0.04, 0.08 to 0.1 wt%, respectively, in contrast to the
complete dissolution of Pluronic hydrogel itself within 1 day. These
results indicate that the addition of a small amount of GO com-
pared to Pluronic (17 wt%) could greatly improve the stability of
Fig. 2. Sol–gel transition phase diagrams of (a) Plruonic and GO-containing Pluronic
in PBS, and (b) Pluronic-based nanogel and GO-containing, Pluronic-based nanogel
in PBS (GO: 0.1 wt%, n = 3).
physical Pluronic hydrogel in an open environment, suggesting the
possibility for applying as an injectable hydrogel system in an open
environment like in vivo with a practically useful residence time
at the injection site. Furthermore, the degradation rate of Pluronic
hydrogel could be systemically modulated by the concentration of
GO.
Chemically crosslinked, Pluronic-based nanogel was reported to
present a temperature-sensitive large size variation (∼450 nm at
4 ◦ C and ∼60 nm at 37 ◦ C) while maintaining its structural integrity
[3]. This nanogel solution also becomes a gel state by the packing
of nanogels instead of micelles. The effect of GO on the stability
of hydrogel formed by Pluronic-based nanogel was also investi-
gated. Pluronic-based nanogel (11 wt%) itself was degraded within
7 days at 37 ◦ C, so the gel state formed by the nanogel packing was
more stable than the gel state formed by the micelle packing. By
adding GO, the degradation rate of gel formed by the nanogel pack-
ing was also significantly reduced; time for complete degradation
was prolonged from 10 days to 15 days, and more than 20 days by
increasing the GO concentration from 0.04, 0.08 to 0.1 wt% (Fig. 2d).
The choice of different concentration of Pluronic nanogel (11 wt%)
from Pluronic micelle solution (17 wt%) to measure the dissolution
rate of the hydrogels was based on the different sol–gel transition
boundary between two systems; in both cases, the concentrations
of making gel state were slight higher than the sol–gel transition
boundary to form a gel state at 37 ◦ C.
3.2. Sol–gel transition phase diagram
The effect of the addition of GO (0.1 wt%) on the sol–gel transi-
tion phase boundary of Pluronic or Pluronic-based nanogel in PBS
was analyzed by using a vial tilting method with 3 ◦ C interval of
temperature increase. In the case of Pluronic, the phase bound-
ary of sol-to-gel transition by increasing temperature at a fixed
 D.-A. Won et al. / Colloids and Surfaces B: Biointerfaces 128 (2015) 515–521
concentration of Pluronic did not show significant change by adding
GO. No significant change of the lower phase boundary of Pluronic
with the addition of GO implies that the addition of GO did not
significantly affect the micelle geometry or packing state, which is
reasonable considering the very small amount of GO compared to
Pluronic. On the other hand, by adding GO, 13 wt% Pluronic solution
showed a sol-to-gel transition at ∼28 ◦ C, thus formed a gel state,
whereas Pluronic solution at 13 wt% without GO did not form a gel
state at any temperature. More interestingly, the phase boundary
of gel-to-sol transition at high temperature disappeared by adding
GO (Fig. 2a), which means that at a given concentration, once the
solution becomes a gel state by increasing temperature, it does not
become a sol state again by further increasing temperature. The gel-
to-sol transition of Pluronic solution at a fixed concentration results
from the shrinkage of micelles by the shortening of the PEO part of
Pluronic as temperature increases, which induces the reduction of
micelle packing state. The disappearance of gel-to-sol phase bound-
ary by adding GO suggests that GO might strongly interact with
Pluronic micelles in a packed state, so it could prevent the reduc-
tion of micelle packing by further increasing temperature, thus to
maintain the physical gel state. Overall, the results of phase dia-
gram of Pluronic micelle indicates that by adding GO, the sol-to-gel
transition (and the gel state formation) by Pluronic micelle packing
could occur at a lower concentration of Pluronic, and the gel-to-
sol transition at high temperature disappeared, all supporting the
strong interaction between GO and Pluronic micelles.
The solution of Pluronic-based nanogel also showed the sol–gel
transition by the packing of nanogels. Compared to Pluronic solu-
tion, the sol-to-gel transition was observed at lower concentration
and lower temperature. Besides, the gel state of nanogel did not
become a sol state even at high temperature (no gel-to-sol tran-
sition by further increasing temperature at a fixed concentration);
as a chemically cross-linked state, the packing of nanogel did not
seem to become loose significantly by increasing temperature. By
adding GO (0.1 wt%) to the nanogel solution, the sol-to gel phase
boundary shifted slightly to a lower concentration and tempera-
ture; at a fixed concentration, the sol state became a gel state at a
slightly lower temperature, and at a fixed temperature, the sol-to-
gel transition was observed at a lower concentration by adding GO.
Also, at 7 wt%, nanogel with GO showed a sol-to-gel transition at
∼28 ◦ C, but nanogel itself did not form a gel state at any temperature
(Fig. 1b). Thus, similar to Pluronic micelle solution, Pluronic-based
nanogel solution also showed that the gelation could occur at a
lower concentration of nanogel by adding GO due to the interac-
tion between GO and nanogel. The enhanced interaction among
nanogels mediated by GO might result in the structural stabilization
and also lower the critical gelation concentration and temperature
[6].
3.3. In vitro release profiles of a growth factor from hydrogels
The VEGF release profiles from hydrogels with 0.1 wt% car-
boxylated GO were characterized. By forming hydrogels with
much enhanced stability, hydrogels with GO showed the sustained
release of loaded VEGF (Fig. 3). Similar to the order of degradation
rate, hydrogel formed by Pluronic-based nanogel (11 wt%) showed
a more sustained release of VEGF compared to that formed by
Pluronic (17 wt%). Except the initial burst (∼10% for nanogel and
∼30% for Pluronic), the release patterns of VEGF from hydrogels
were similar to the degradation rates of the hydrogels. And, the
release rates of VEGF were a little faster than the degradation rates
of the hydrogels. The smaller initial burst from hydrogel formed
by the nanogel seemed to be associated with the characteristics
of the nanogel that can efficiently encapsulate proteins and slowly
release then [30]. However, since the degradation rate of hydro-
gel by the dissociation among nanogel is much faster than the
519
Fig. 3. In vitro release profiles of VEGF from Pluronic (17 wt%) or Pluronic-based
nanogel (11 wt%) with carboxylated GO containing (GO: 0.1 wt%, n = 3) at 37 ◦ C in
100 rpm environment (n = 3).
release rate of VEGF from nanogel, the release of VEGF from GO-
containing hydrogel was mainly determined by the degradation of
the hydrogel, which was also valid for hydrogel formed by Pluronic
micelles. In summary, hydrogels with significantly enhanced sta-
bility by adding a small amount of GO were also useful for achieving
sustained release of loaded proteins.
3.4. The effect of GO on the mechanical properties of hydrogel
The gelation kinetics and storage moduli of hydrogels were
observed by using a rheometer. Hydrogels as a sol state were
loaded on a parallel plate geometry holder at 4 ◦ C. The gelation
was achieved instantaneously by increasing temperature to 37 ◦ C.
By adding GO, moduli of hydrogels were increased in all cases.
And, at the same polymer concentration, the moduli of hydrogels
increased by increasing concentration of GO from 0.04 to 0.1 wt%
(Fig. 4). Therefore, the addition of GO was effective for increasing
the mechanical properties of hydrogels formed by Pluronic micelle
or Pluronic-based nanogels, due to making more efficient physi-
cal crosslinks induced by the added GO. The enhanced mechanical
properties of the hydrogels were similar to the case of other
polymers with GO composite [19,20]. The increase in mechani-
cal properties of the hydrogel by adding GO, however, was not
dramatic compared to the increased stability of the hydrogels.
3.5. In vivo stability and biocompatibility of GO-containing
hydrogels
To demonstrate the feasibility of GO-containing hydrogels as an
injectable, long-lasting hydrogel in vivo, hydrogels with or without
GO was injected subcutaneously at the back of the mice. Since both
bare hydrogels and GO-containing hydrogels maintain the charac-
teristic of sol-to-gel transition by temperature change, they were
injected as a sol state at a low temperature. After injection, gel state
was formed within 3 min post-injection in mice. To monitor the sta-
bility of hydrogels in vivo, the size of hydrogel was monitored after
removing the skin of the injected mice at several time points: 1 day,
1 week, 2 weeks, and 4 weeks.
Bare Pluronic hydrogel disappeared within 1 day (Fig. 5b) in
accordance with the result of in vitro degradation. However, GO-
containing Pluronic gel was degraded very slowly; 84% of the
injected hydrogel was remained after 7 days, and 50% of the hydro-
gel stably existed even after 28 days (Fig. 5a). In the case of
Pluronic-based nanogel, it disappeared within 7 days (Fig. 5d), but
GO-containing nanogel was even more slowly degraded than GO-
containing Pluronic; 91% of the injected hydrogel was remained
after 7 days, and 71% of the hydrogel stably existed after 28 days
(Fig. 5c). The degradation rates of GO-containing hydrogels in vivo
were slower than those in vitro, presumably due to the lack of
520 D.-A. Won et al. / Colloids and Surfaces B: Biointerfaces 128 (2015) 515–521

Fig. 4. Gelation of hydrogel composed of (a) 17 wt% Pluronic and (b) 11 wt%
Pluronic-based nanogel with different concentration of GO from 0 to 0.1 wt%.

shaking and the limited supply of physiological fluid in vivo com-

pared to the in vitro experimental condition. In any case, hydrogels
without GO were degraded too fast to be useful in vivo as well as
in vitro. The formation and long existence of GO-containing hydro-
gels verified that the instability of Pluronic-based hydrogel, which
prevented the practical biomedical application of Pluronic hydro-
gels, could be efficiently overcome by adding a small amount of GO
both in vitro and in vivo.
In addition to the remarkably enhanced in vivo stability, bio-
compatibility of these hydrogels needs to be characterized for
biomedical applications. The safety and toxicity of graphene-
based materials have not been completely resolved [33]. However,
graphene oxide is regarded to be relatively biocompatible com-
pared to other nano-structured carbon like CNT, and there
have been many studies reporting the biomedical application of
graphene oxide with apparently no (or significant) short term tox-
icity issue [34]. For example, PEG-coated graphene oxide showed
non-toxicity for 3 months via several biological analyses in mice
with 20 mg/kg [35]. We also verified no acute cytotoxicity of
graphene oxide coated with Pluronic in vitro [28] and in vivo [25].
Fig. 5. (a) In vivo degradation of hydrogels composed of 17 wt% Pluronic and 0.1 wt%
The histological images of tissues at the injection sites for bare
GO (n = 3). (b) Photos of remaining hydrogel at the injected skin: Pluronic after 1
hydrogels and GO-containing hydrogels were obtained after 3 day (left), and GO-containing Pluronic after 4 weeks (right). (c) In vivo degradation
weeks of injection (Suppl. Fig. 1). Pluronic and nanogel without of 11 wt% Pluronic-based nanogel and 0.1 wt% GO (n = 3). (d) Photos of remaining
GO were already completely degraded, so no sign of polymer was hydrogel at the injected skin: nanogel (upper) and GO-containing nanogel (lower)
after 4 weeks.
found. In the case of GO-containing hydrogels, dark spots were
found due to the presence of GO. However, neither the intense
localization of macrophages or noticeable inflammation response 4. Conclusion
was observed in all cases of GO-containing hydrogels by H&E stain-
ing. Thus, GO-containing hydrogels did not show any sign of acute We report a long-lasting, thermo-sensitive and injectable
biocompatibility issue in vivo. Since, a very small quantity of GO hydrogel formed by Pluronic or Pluronic-based nanogel by adding
(0.1 wt%) with abundant Pluronic, the present formulation may not a small amount (0.1 wt%) of graphene oxide (GO). Simple addition
cause significant toxicity without complete metabolism in vivo. of a small quantity of GO to Pluronic-based gel systems was very
 D.-A. Won et al. / Colloids and Surfaces B: Biointerfaces 128 (2015) 515–521
efficient to increase the stability of the physical gel in an open
environment. Further carboxylated GO was more effective than
bare GO for increasing the stability of the physical hydrogels. The
in vitro dissolution rate of Pluronic-based hydrogels could be sys-
tematically lowered by increasing the content of GO. The stable and
injectable formulations of GO-containing Pluronic-based hydro-
gels were demonstrated in vivo by subcutaneous injection into the
mice. Existence of the injected GO-containing hydrogels over sev-
eral weeks without noticeable acute inflammation response was
confirmed, in contrast to the complete disappearance of Pluronic-
based hydrogels without GO in a short time. Therefore, the present
approach could significantly increase the potential of practical
application of Pluronic systems, and we anticipate that the method
of adding GO could also be applied to enhance the stability of other
physical hydrogel systems.
Acknowledgements
Partial financial supports were provided by the Basic Science
Research Program, NRF, MEST (2013R1A2A2A03068802), Korea,
and by the “GIST-Caltech Research Collaboration Project” through
a grant provided by GIST, Korea.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.colsurfb.
2015.03.002.
References
[1] N.B. Graham, Med. Device Technol. 9 (1998) 18.
[2] T.R. Hoare, D.S. Kohane, Polymer 49 (2008) 1993.
[3] W.I. Choi, G. Tae, Y.H. Kim, J. Mater. Chem. 18 (2008) 2769.
521
[4] K.M. Park, S.Y. Lee, Y.K. Jeong, J.S. Na, M.C. Lee, K.D. Park, Acta Biomater. 5 (2009)
1956.
[5] Y. Lee, H.J. Chung, S. Yeo, C.H. Ahn, H. Lee, P.B. Messersmith, T.G. Park, Soft
Matter 6 (2010) 977.
[6] Y.Y. Chen, H.C. Wu, J.S. Sun, G.C. Dong, T.W. Wang, Langmuir 29 (2013) 3721.
[7] W.S. Shim, J.S. Yoo, Y.H. Bae, D.S. Lee, Biomacromolecules 6 (2005) 2930.
[8] M. Aliaghaie, H. Mirzadeh, E. Dashtimoghadam, S. Taranejoo, Soft Matter 8
(2012) 7128.
[9] L. Yu, J. Ding, Chem. Soc. Rev. 37 (2008) 1473.
[10] K. Mortensen, J.S. Pedersen, Macromolecules 26 (1993) 805.
[11] J. Elisseeff, K. Anseth, D. Sims, W. Mcintosh, M. Randolph, R. Langer, Proc. Natl.
Acad. Sci. U.S.A. 96 (1999) 3104.
[12] D. Cohn, G. Lando, A. Sosnik, S. Garty, A. Levi, Biomaterials 27 (2006) 1718.
[13] M. Morishita, J.M. Barichello, K. Takayama, Y. Chiba, S. Tokiwa, T. Nagai, Int. J.
Pharm. 212 (2001) 289.
[14] S.Y. Lee, G. Tae, Y.H. Kim, J. Biomater. Sci. Polym. Edn. 18 (2007) 1335.
[15] A.K. Geim, K.S. Novoselov, Nat. Mater. 6 (2007) 183.
[16] Y. Wang, Z. Li, J. Wang, J. Li, Y. Lin, Trends Biotechnol. 29 (2011) 205.
[17] L. Feng, L. Wu, X. Qu, Adv. Mater. 25 (2013) 168.
[18] L. Zhang, Z. Wang, C. Xu, Y. Li, J. Gao, W. Wang, Y. Liu, J. Mater. Chem. 21 (2011)
10399.
[19] K. Haraguchi, T. Takehisa, S. Fan, Macromolecules 35 (2002) 10162.
[20] X. Ma, Y. Li, W. Wang, Q. Ji, Y. Xia, EurPolym. J. 49 (2013) 389.
[21] H. Bai, C. Li, X. Wang, G. Shi, J. Phys. Chem. C 115 (2011) 5545.
[22] H. Bai, C. Li, G. Shi, Chem. Commun. 46 (2010) 2376.
[23] Y. Xu, Q. Wu, Y. Sun, H. Bai, G. Shi, ACS Nano. 4 (2010) 7358.
[24] D. Han, L. Yan, ACS Sustainable Chem. Eng. 2 (2014) 296.
[25] A. Sahu, W.I. Choi, G. Tae, Chem. Commun. 48 (2012) 5820.
[26] J. Liu, G. Chen, M. Jiang, Macromolecules 44 (2011) 7682.
[27] S.Z. Zu, B.H. Han, J. Phys. Chem. C 113 (2009) 13651.
[28] A. Sahu, W.I. Choi, J.H. Lee, G. Tae, Biomaterials 34 (2013) 6239.
[29] J.Y. Kim, W.I. Choi, Y.H. Kim, G. Tae, S.Y. Lee, K. Kim, I.C. Kwon, J. Controlled
Release 147 (2010) 109.
[30] W.I. Choi, Y.H. Kim, G. Tae, Macromol. Res. 19 (2011) 639.
[31] W. Zhang, Z. Guo, D. Huang, Z. Liu, X. Guo, H. Zhong, Biomaterials 32 (2011)
8555.
[32] R.M. Silverstein, G.C. Bassler, T.C. Morrill, Spectrometric Identification of
Organic Compounds, 4th ed., John Wiley and Sons, New York, 1981.
[33] K.H. Liao, Y.S. Lin, C.W. Macosko, C.L. Haynes, ACS Appl. Mater. Interfaces 3
(2011) 2607.
[34] C. Chung, Y.K. Kim, D. Shin, S.R. Ryoo, B.H. Hong, D.H. Min, Acc. Chem. Res. 46
(2013) 2211.
[35] K. Yang, J. Wan, S. Zhang, Y. Zhang, S.T. Lee, Z. Liu, ACS Nano 5 (2011) 516.