Dextran based photodegradable hydrogels formed via a Michael addition

Article abstract of DOI:10.1039/c1sm05291h

A photodegradable, covalently crosslinked hydrogel system has been constructed from the biocompatible polymers dextran and poly(ethylene glycol) using the acrylate-thiol Michael addition as the crosslinking method. Light sensitivity of the hydrogel was in

Full text of DOI:10.1039/c1sm05291h

View Article Online / Journal Homepage / Table of Contents for this issue
Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Dynamic Article Links <
C
Soft Matter
Cite this: Soft Matter, 2011, 7, 4881
www.rsc.org/softmatter
PAPER
Dextran based photodegradable hydrogels formed via a Michael addition†
Ke Peng,a Itsuro Tomatsu,a Bram van den Broek,b Chao Cui,c Alexander V. Korobko,d John van Noort,b
c
Annemarie H. Meijer, Herman P. Spaink and Alexander Kros
c
a
*
Received 18th February 2011, Accepted 14th March 2011
DOI: 10.1039/c1sm05291h
A photodegradable, covalently crosslinked hydrogel system has been constructed from the
biocompatible polymers dextran and poly(ethylene glycol) using the acrylate–thiol Michael addition as
the crosslinking method. Light sensitivity of the hydrogel was introduced by placing a non-toxic
photolabile o-nitrobenzyl moiety in between dextran backbone and acrylate group. Hydrogels were
prepared under physiological conditions without the need of any additional reagents by mixing
solutions of dextran functionalized with acrylate-modified o-nitrobenzyl moieties (Dex-AN) and
dithiolated poly(ethylene glycol) (DSPEG). The degradation of the hydrogels due to UV irradiation
was investigated with scanning electron microscopy, infrared and UV-vis spectroscopy. Using green
fluorescent protein (GFP) as a model protein, light triggered protein release from the obtained gel
matrices was investigated in different forms. Furthermore, photodegradation of the hydrogel via two
photon excitation was also examined using focused pulsed near infrared (NIR) laser beam as a light
source.
which is a remote stimulus that can be controlled spatially and
Introduction
temporally with great ease and convenience.13–15 Moreover, the
light irradiation except deep-UV does not have a harmful effect
on various bioactive compounds including most of the
proteins.16–18 Therefore, light sensitive hydrogels are attracting
great attention and a variety of systems are being developed.19–27
Recently, Anseth et al. developed a series of covalently
crosslinked photodegradable hydrogels and demonstrated that
the migration of stem cells in the hydrogel network can be
controlled via photodegradation.28,29 For this they used free
radical co-polymerizations of water-soluble acrylate monomers
and photodegradable difunctionalized crosslinkers. Inspired by
their work, we have introduced light sensitivity to our in situ
forming hydrogel systems.4,5 With our method, hydrogels having
similar properties can be obtained from biocompatible polymers
and prepared under physiological conditions without the need of
any additional reagents (e.g. radical initiators or catalysts, which
are potentially toxic) and no side-products are formed. There-
fore, we believe that this strategy for fabrication of covalently
crosslinked hydrogels is promising for future biomedical appli-
cations (Scheme 1).30–32
Hydrogel materials are widely used in a diverse range of appli-
cations,1 especially for drug delivery, biosensing, and tissue
engineering, due to their excellent biocompatibility and capa-
bility as a container for fragile bioactive molecules such as
proteins.2,3 Recently we have developed hydrogel systems
composed of the biocompatible polymers, dextran and poly
(ethylene glycol) (PEG), which were crosslinked in situ to form
hydrogels.4,5 The excellent biocompatibility of the obtained gel
was demonstrated in vivo using a zebrafish embryo assay.5,6
One of the main issues in current biomaterial science is to
control the properties of hydrogels (e.g. swelling rate, perme-
ability and mechanical strength) with external stimuli such as
pH, temperature, and light.7–11 For example, these stimuli-
responsive hydrogels are beneficial for an efficient drug delivery
system.12 Light is an ideal stimulus to manipulate the hydrogel,
aLeiden Institute of Chemistry, Leiden University, P.O. Box 9502, 2300 RA
Leiden, The Netherlands. E-mail: a.kros@chem.leidenuniv.nl; Fax: +31
71527 4397; Tel: +31 71 527 4234
bPhysics of Life Processes, Leiden Institute of Physics, Leiden University,
In the current system, a hydrogel was formed by mixing
two solutions: dextran functionalized with acrylate-modified
o-nitrobenzyl moieties (Dex-AN) and dithiolated poly(ethylene
glycol) (DSPEG) via a Michael addition between the acrylate and
thiol groups. The photolabile o-nitrobenzyl moiety was intro-
duced between the dextran and PEG backbones, thus other
than unconnected polymers, no small-molecule fragments are
generated after photodegradation, which decreases the risk of
toxic side effects. Furthermore, the biocompatibility of the
Niels Bohrweg 2, 2333 CA Leiden, the Netherlands
cInstitute of Biology, Leiden University, P.O. Box 9502, 2300 RA Leiden,
The Netherlands
dNano Structured Materials, TU Delft, Julianalaan 136, 2628 BL Delft,
The Netherlands
† Electronic supplementary information (ESI) available: Movies of the
microscopic hydrogel degradation under UV irradiation, and
movements of quantum dots in the hydrogel matrix before and after
photodegradation
10.1039/c1sm05291h
via
two-photon
excitation.
See
DOI:
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 4881–4887 | 4881

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
Fig. 1 1H NMR spectrum of Dex-AN (DMSO-d6 containing 5% D2O).
The chemical structure shows the substitution of AN at C-3 position of
glucose unit as a possible structure.
further confirmed by the presence of small peaks around 5.0 ppm
(peak g) due to the peak shift of glucosidic protons of the
anhydroglucose units upon esterification. The degree of substi-
tution (the number of substitutes per 100 anhydroglucose units)
was determined to be 10 according to the ratio between the
signal at 7.1 ppm corresponding to AC and the dextran
anomeric proton at 4.7 ppm. The DSPEG was synthesized as
reported previously and the Ellman test showed a thiol func-
tionality of 90%.38
Scheme 1 Chemical structures of the components of in situ forming
photodegradable hydrogel: dextran functionalized with acrylate-modi-
fied o-nitrobenzyl moieties (Dex-AN) and dithiolated poly(ethylene
glycol) (DSPEG). Upon mixing, these underwent Michael addition
between acrylate and thiol groups to form a light sensitive hydrogel.
Upon the irradiation, the hydrogel was decomposed due to the cleavage
of the photolabile moiety, which triggered the release of the protein
(GFP) entrapped in the polymer network.
Macroscale hydrogel formation
Hydrogels were obtained by mixing the same amounts of solu-
tions of Dex-AN (200 mg mLꢀ1) and DSPEG (120 mg mLꢀ1) in
phosphate buffered saline (PBS). Upon mixing the solutions,
gelation occurred within 60 min at 37 ꢁC. The ratio between
acrylate and thiol groups was 1 : 1 and the concentrations of the
polymers were 9 wt% and 5 wt% for Dex-AN and DSPEG,
respectively.
o-nitrobenzyl moiety has been proven; both the residues before
and after photodegradation and the degradation process are
inert for fragile biomacromolecules including proteins.33–36
Herein we report the preparation of both macroscale and
microscale hydrogels and hydrogel degradations upon UV irra-
diation (365 nm). As a possible application of these hydrogels,
the capability of protein storage and light triggered release using
green fluorescent protein (GFP) as a model protein were also
examined. Upon the light-induced degradation of the hydrogel
matrix, the entrapped proteins were released from the matrix and
migrated into the media. Furthermore, besides with UV light,
photodegradation by two-photon excitation using near infrared
(NIR) light was tested to broaden the applicable field of the
obtained hydrogel system in future.
The mechanical property of the obtained hydrogel was char-
acterized with rheometry; the dependences of storage (G0) and
loss (G00) moduli on angular frequency (u) were measured. As
shown in Fig. 2, after mixing of Dex-AN and DSPEG, it showed
a dominant of G0 than G00 in the frequency range measured. This
observation and nearly parallel behavior of G0 and G00 at low
Results and discussion
Synthesis of Dex-AN and DSPEG
Dex-AN was prepared by a conventional carbodiimide mediated
esterification of the hydroxyl groups of dextran (20 kDa) with
4-(4-(1-(acryloyloxy)ethyl)-2-methoxy-5-nitro-phenoxy)butanoic
1
acid (AN).28,37 The obtained Dex-AN was characterized by H
NMR, as shown in Fig. 1, besides signals attributed to dextran,
signals at 7.6, 7.1, 6.4–6.0, 4.1 and 3.9 ppm due to the protons
from the AN group were observed which indicated the func-
tionalization of dextran with AN. The AN derivatization was
Fig. 2 Dynamic moduli of the hydrogel obtained by mixing of solutions
of Dex-AN (30 mg in 150 mL PBS) and DSPEG (18 mg in 150 mL PBS).
4882 | Soft Matter, 2011, 7, 4881–4887
This journal is ª The Royal Society of Chemistry 2011

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
Fig. 3 Photographs of a mixture of 50 mL of Dex-AN (200 mg mLꢀ1
and 50 mL of DSPEG (120 mg mLꢀ1) in PBS: taken after mixing (a) and
after irradiation with UV light for 72 hours (b).
)
Fig. 5 FT-IR spectra of the freeze-dried samples of the hydrogel (black
line) and the photodegradated hydrogel (red line). In the enlarged spectra
(right), a decrease of a signal ascribable to NO2 asymmetric stretch can be
seen.39
frequencies indicate that DSPEG crosslinked Dex-AN via
a thiol–acrylate reaction resulted in the formation of hydrogel.
with isosbestic points at 301 and 387 nm and the reaction reached
a steady state after 120 min.40,41
Photodegradation of the macroscale hydrogel
Protein release from the macroscale hydrogel
The photodegradation of the freshly prepared hydrogel (100 mL,
1 cm ꢂ 1 cm ꢂ 1 mm) was shown in Fig. 3. After UV irradiation
of 72 hours, the gel was degraded to a fluidic solution due to the
hydrolysis of the ester group in the o-nitrobenzyl moiety resulting
in the dissociation of the polymer network (Scheme 1).
Our previous hydrogel system showed excellent biocompatibility
in vivo,5 and we have now introduced photoresponsivity into the
hydrogels, thus we believe that the current system can be bene-
ficial for the future biomedical applications, for example, as
a photoresponsive protein carrier. Since hydrogels has a struc-
tural similarity with hydrated body tissues and allows to retain
the proteins in the protective 3-D network and prevents them
from decomposition and denaturation, hydrogels can be an ideal
material.2,42 Moreover stimuli responsive hydrogels hold the
potential to transport entrapped protein to the target site and to
trigger the release by a stimulus at the desired point and time are
drawing more attentions.10 Furthermore as a number of attrac-
tive therapeutic proteins have been developed recently against
a broad range of diseases such as cancers, autoimmune diseases
and metabolic disorders,43 development of containers that can
handle proteins becomes more important.
To investigate the photodegradation of the macroscale
hydrogel, lyophilized samples of the freshly prepared and
degraded hydrogels were subjected to scanning electron micros-
copy (SEM). As shown in Fig. 4a freshly prepared hydrogels
showed a regular microporous structure, indicating the existence
of a well extended homogeneous three-dimensional polymer
network before removing the water molecules. In contrast, dense
amorphous solids were observed from the photodegraded
hydrogel (Fig. 4b), which is similar to the one obtained from
a dextran or PEG solution in the absence of crosslinking.
Using the lyophilized samples of the freshly prepared and
degraded hydrogels, infrared spectroscopy (FT-IR) was also
performed (Fig. 5). FT-IR analysis revealed only a marked
difference after irradiation in the band at 1520 cmꢀ1, which is
ascribed to the asymmetric stretch of NO2. This indicates the
reduction of the NO2 to NO groups which is expected upon the
degradation of the o-nitrobenzyl moiety. The integrity of the
dextran and PEG polymers was not affected upon the irradiation
as expected (see Fig. 5).39
The capability of protein storage and light triggered release
from the obtained hydrogels were examined using green fluo-
rescent protein (GFP) as a model protein. The GFP carrying
macroscale gel (ꢃ10 mL) was placed in the bottom of a 1 cm2
cuvette with 1 mL of fresh PBS and the fluorescence intensities of
the solution part were measured as a function of time (Fig. 7).
Under the UV light irradiation, the emission intensities of GFP
in PBS began to increase immediately, indicating a fast start of
The degradation of the o-nitrobenzyl moiety was also moni-
tored by UV-vis spectroscopy at diluted conditions as shown in
Fig. 6. The absorbance maximum at 350 nm decreased upon UV
irradiation, while the intensity around 280 and 430 nm increased
Fig. 4 SEM images of the freeze-dried hydrogel prepared from Dex-AN
and DSPEG in pure water: before (a) and after UV light irradiation (b).
Scale bars represent 5 mm.
Fig. 6 UV-vis spectra measured at diluted condition after different
irradiation times (0–160 min). Inset: the obtained absorbance at 350 nm
was plotted as a function of irradiation time.
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 4881–4887 | 4883

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
solution (100 mL). The photodegradation of the hydrogel was
then studied with optical microscopy. The complete degradation
of a hydrogel particle was observed within 30 minutes of irra-
diation, while in the absence of radiation no changes were
observed (see Movie 1 in the ESI†).
Furthermore, GFP loaded microscale hydrogel blocks were
also prepared by mixing a Dex-AN solution containing GFP and
a DSPEG solution. The release from the hydrogel was monitored
using fluorescence microscopy. When the gel was kept in the
dark, the shape of the fluorescent block was not changing within
30 minutes (Fig. 8), showing that the GFP was firmly encapsu-
lated in the hydrogel matrix. In contrast, UV exposure (365 nm)
of the hydrogel led to a rapid decrease in fluorescence indicative
of the loss of GFP. After 30 min of irradiation, the hydrogel
disappeared completely, suggesting that the polymer network
was eroded and the GFP was liberated into the surrounding
medium. The decrease of fluorescence of the GFP-loaded gels
was plotted as a function of time. Under irradiation (circle), the
fluorescence decreased significantly in the first 10 minutes as
most of the GFP was released concurrently with the erosion of
the polymer network. In the control experiment without UV
irradiation (diamond) only a small decrease was observed in the
first 6 min presumably due to the loss of loosely bound GFP on
or near the surface of the hydrogel, after which the fluorescence
stabilized showing that the encapsulated GFP remained inside
the hydrogel matrix. It should be noted here that 10 minutes of
irradiation with UV light is feasible for experiments in living
zebrafish embryos.45–47 Thus we believe that the current system
can also be applied for in vivo use in zebrafish embryos and other
transparent animals by adjusting the size of the hydrogel.
To exclude photo-bleaching of the GFP as a possible cause for
the observed fluorescence decrease, the stability of GFP under
UV irradiation was examined. As shown in Fig. 9, the fluorescent
intensity of the GFP solution was preserved after 30 min of UV
irradiation. Thus the observed fluorescence decrease in Fig. 8 was
due to the decrease in the concentration of GFP (i.e. degradation
of the hydrogel) and not to photo-bleaching of the GFP itself.
Fig. 7 Release of green fluorescent protein (GFP) from the macroscale
hydrogel. The samples were placed under UV light (circle) and dark
(diamond), respectively.
the release. The release of GFP can be continued; ꢃ50% of the
loaded proteins were released in 60 min under UV irradiation
and reached ca. 70% in 90 min. In contrast, no significant
increase in the release of GFP into the surrounding solution was
observed without UV light irradiation even after 90 minutes (ca.
5%). Therefore we conclude that the release of GFP was in a UV
light responsive manner (Scheme 1b).
Microscale hydrogel preparation and degradation
Since the degradation of the macroscopic hydrogel requires
long time UV exposure, microscale hydrogel particles were
prepared using PDMS mold in order to increase the speed of
degradation.44
Sub-millimetre scale hydrogel blocks with a volume of
approximately 5 nL were prepared and submersed in a PBS
Two-photon hydrogel degradation
It has been shown that o-nitrobenzyl moieties can not only be
photocleaved by 365 nm UV light, but also by two-photon
Fig. 8 Release of GFP from the gel matrices upon the irradiation with
UV light. Pictures show representative states of the gel matrices for both
samples kept in the dark and under UV irradiation taken with fluores-
cence microscope. Scale bars represent 200 mm. The difference of the
luminosities of the gels from the original state is calculated from the
pictures and plotted as a function of time.
Fig. 9 Fluorescence spectra of a GFP solution (2 mg mLꢀ1) in PBS
measured before (black line) and after 30 min of UV-irradiation (red
line). The GFP fluorescence did not change significantly by UV irradia-
tion (ca. 95% of the emission at 509 nm).
4884 | Soft Matter, 2011, 7, 4881–4887
This journal is ª The Royal Society of Chemistry 2011