Materials Science and Engineering C 70 (2017) 287–295

Contents lists available at ScienceDirect

Materials Science and Engineering C

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

Covalently antibacterial alginate-chitosan hydrogel dressing integrated

gelatin microspheres containing tetracycline hydrochloride for
wound healing
Huinan Chen a, Xiaodong Xing a,⁎, Huaping Tan b,⁎⁎, Yang Jia b, Tianle Zhou b, Yong Chen c,
Zhonghua Ling c, Xiaohong Hu d
a
School of Chemical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China
b
School of Materials Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, China
c
Department of Orthopaedics, Jinling Hospital, Nanjing 210002, China
d
School of Material Engineering, Jinling Institute of Technology, Nanjing 211169, China

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

Article history: An antibacterial and biodegradable composite hydrogel dressing integrated with microspheres is developed for
Received 2 April 2016 drug delivery and wound healing. The mechanism of gelation is attributed to the Schiff-base reaction between
Received in revised form 12 August 2016 aldehyde and amino groups of oxidized alginate (OAlg) and carboxymethyl chitosan (CMCS). To enhance anti-
Accepted 31 August 2016
bacterial and mechanical properties, tetracycline hydrochloride (TH) loaded gelatin microspheres (GMs) were
Available online 3 September 2016
fabricated by an emulsion cross-linking method, followed by integrating into the OAlg-CMCS hydrogel to pro-
Keywords:
duce a composite gel dressing. In vitro gelation time, swelling, degradation, compressive modulus and rheological
Bacterial infection properties of the gel dressing were investigated as the function of microsphere ratios. With increasing ratios of
Hydrogel microspheres from 10 to 40 mg/mL, the composite dressing manifested shorter gelation time and lower swelling
Microspheres ratios, as well as higher mechanical strength. Comparing to other formulations, the gel dressing with 30 mg/mL
Wound dressing microspheres showed more suitable stabilities and mechanical properties for wound healing. Also, in vitro drug
Antibacterial activity release results showed that the loaded TH could be sustained release from the composite gel dressing by contrast
with pure hydrogels and microspheres. Furthermore, powerful bacteria growth inhibition effects against
Escherichia coli and Staphylococcus aureus suggested that the composite gel dressing, especially the one with
30 mg/mL GMs containing TH, has a promising future in treatment of bacterial infection.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction
In burn care, Escherichia coli (E. coli) and Staphylococcus aureus (S.
aureus) are the common bacteria and proliferate rapidly after infection
[1,2]. A suitable and effective dressing material with antibiotics can
not only overcome this shortcoming but reduce the occurrence of anti-
biotic resistance of bacteria by controlling the local concentration of an-
tibiotics [2–4]. Recently, biodegradable hydrogels have been widely
used in tissue engineering and drug delivery, which can function as in-
jectable scaffolds for osteogenesis, cartilage repair, adipose regeneration
and wound healing [5–14]. Dressings cast from hydrogels are developed
with the advantages of moist environment, good biocompatibility, the
ability of absorbing wound exudates and no adherence with wound-
⁎ Corresponding author.
⁎⁎ Correspondence to: School of Materials Science and Engineering, Nanjing University
of Science and Technology, 200 Xiao Ling Wei St, Nanjing 210094, China.
E-mail addresses: xingxiaodong07@njust.edu.cn (X. Xing), hptan@njust.edu.cn
(H. Tan).
tissue [1,15]. Hydrogel dressings are preferred over other dressing ma-
terials because they can avoid the secondary damage when dressing
changed. Many natural polymer dressings, developed with chitosan, al-
ginate, collagen and hyaluronic acid, have already been studied in
wound healing and tissue regeneration [15–19].
Many methods have been employed for the preparation of natural
hydrogels, including chemical cross-linking with agents such as
carbodiimide, glutaraldehyde, genipin and adipic dihydrazide [20–24].
However, the chemical cross-linking agents are the major obstacle in
the use as injectable in situ forming polymer scaffolds due to their tox-
icity to cells [25,26]. By contrast with chemical methods in the synthesis
of hydrogel, the Schiff-base reaction proves to be a non-cytotoxic cross-
linking reaction, without the participation of any chemical cross-linking
agents [11,27]. Herein, a covalent hydrogel dressing is presented by the
cross-linking of amine groups from water-soluble carboxymethyl chito-
san (CMCS) and aldehyde groups from oxidized alginate (OAlg) via the
Schiff-base reaction. Chitosan and alginate are natural polysaccharides
with good biocompatibilities and biodegradabilities, which have been
widely applied in drug delivery, gene therapy and tissue engineering

http://dx.doi.org/10.1016/j.msec.2016.08.086
0928-4931/© 2016 Elsevier B.V. All rights reserved.
288 H. Chen et al. / Materials Science and Engineering C 70 (2017) 287–295

Fig. 1. Reaction schematic illustration of the composite OAlg-CMCS hydrogel (Gel) integrated with gelatin microspheres (GMs) containing tetracycline hydrochloride (TH) via the Schiff-
base reaction.

[28–32]. Due to their excellent gel-forming properties, CMCS and algi-
nate also show promise in biomedically-relevant hydrogel systems.
In this work, we intend to enhance the performances of natural
hydrogels when applied as antimicrobial dressings, especially mechan-
ical properties and drug delivery. We hypothesized that the incorpora-
tion of antibiotic-loaded microspheres into gel scaffolds would show
expected mechanical properties and sustained drug release. During
the past decade, many efforts have been devoted towards fabricating
gelatin-based microspheres (GMs) for delivery of cell growth factors
and genes to induce tissue regeneration, either in localized or targeted
manners [33,34]. Due to the ability of controlled release, GMs have
been widely adopted as carriers to immobilize or encapsulate cell
growth factors for enhancing cell proliferation and differentiation. Cell
growth factors and bioactive drugs can be not only directly encapsulat-
ed into GMs during preparation, but also grafted or coated on surface of
GMs. Therefore, tetracycline hydrochloride (TH) loaded GMs were fab-
ricated by an emulsion cross-linking method, followed by integrating
into OAlg-CMCS hydrogels to produce a composite gel dressing. The ef-
fects of varying the ratio of microspheres on gelation time, microstruc-
ture, morphology, equilibrium swelling, compressive modulus and
degradation in vitro were examined. Furthermore, the antibacterial
properties against E. coli and S. aureus were also evaluated in order to
examine the positive effects in treatment of bacterial infection.
2. Experimental section
2.1. Materials
Sodium alginate and carboxymethyl chitosan (CMCS) were supplied
by Sinopharm Chemical Reagent Co., Ltd. (China). Gelatin was pur-
chased from Tianjin Kemiou Chemical Reagent Co., Ltd. (China). Tetra-
cycline hydrochloride (TH), glutaraldehyde, ninhydrin, t-butyl
carbazate and sodium periodate were purchased from Aladdin Industri-
al Corporation (Shanghai, China). All chemicals and reagents were used
as received.
2.2. Synthesis of oxidized alginate (OAlg)
Sodium alginate (3 g) was dissolved to 150 mL ultrapure water
under stirring to obtain a solution of 2% (w/v) concentration. An aque-
ous solution of sodium periodate (3 g, 10 mL) was added dropwise
into the alginate solution and stirred for 24 h at room temperature in

Fig. 2. (a) FTIR spectra of a: gelatin, b: TH, c: GMs, d: TH-loaded GMs (TH/GMs). (b) FTIR spectra of a: alginate, b: OAlg, c: CMCS, d: Gel.
 H. Chen et al. / Materials Science and Engineering C 70 (2017) 287–295
Fig. 3. SEM images showing surface morphologies of GMs and TH/GMs. (a, b) GMs, (c, d) TH/GMs. (e) Size distribution of GMs and TH/GMs at room temperature. The average
hydrodynamic diameters were 72.94 μm and 87.78 μm, respectively.
the dark. Ethylene glycol (2 mL) was added into the solution to stop the
oxidation reaction. 1 g sodium chloride was dissolved into the reaction
product and the solution was further purified by dialysis (MWCO
14,000) against ultrapure water for 3 days. The purified solution was ly-
ophilized at −50 °C and collected finally. The percentage oxidation of
OAlg was quantified by measuring the number of aldehydes in the poly-
mer using t-butyl carbazate [11,12].
2.3. Preparation of microspheres
TH loaded gelatin microspheres (TH/GMs) were prepared by the
emulsion cross-linking method. Briefly, 1 g gelatin and 0.1 g TH were
dissolved in 10 mL ultrapure water. The solution was added dropwise
into 60 mL soybean oil containing 1 mL span-80 at 60 °C under stirring
by a mechanical stirrer. After emulsification for 10–15 min, the emul-
sion was cooled to 5 °C by an ice bath. Glutaraldehyde (1 mL, 50% v/v)
was added into the solution and the reaction was stirred for 60 min. Ac-
etone (30 mL) was poured in order to dehydrate and flocculate the
coacervated droplets for 40 min [35]. The produced microspheres
were centrifuged at 2500 rpm for 5 min, washed with isopropanol
and ethanol to remove the oil phase, and then freeze-dried at
− 50 °C. The average hydrodynamic diameters of the microspheres
were determined using a Malvern Mastersizer 3000 at room temper-
ature. To determine loading capacity, 20 mg microspheres were
completely dismissed in 20 mL PBS (pH = 7.4) using an ultrasonic
cell crusher. After centrifuged, the TH concentration of residues
were measured by an ultraviolet spectrophotometer (BioMate 3S,
Thermo Fisher Scientific). Cross-linking degrees of gelatin in micro-
spheres were quantified by measuring the number of amino groups
by the ninhydrin assay [11,12].
Fig. 4. (a) Degradation of GMs and TH/GMs with respect to weight loss after incubation in PBS for 7 days and 14 days at 37 °C. (b, c) SEM images showing GMs after degradation for 7 and
14 days, respectively. (d, e) SEM images showing TH/GMs after degradation for 7 and 14 days, respectively.
289
2.4. Preparation of hydrogels
OAlg and CMCS were dissolved in PBS separately at concentrations
of 15 wt% and 4 wt%, respectively. For cross-linking of the hydrogel,
OAlg and CMCS solutions were completely mixed with the volume
ratio of 1:4. For preparation of the microsphere embed hydrogel, TH/
GMs were thoroughly dispersed in CMCS solution at room temperature.
The TH/GMs/CMCS mixture was compounded with OAlg solution to
cross-link hydrogels. The gelation time of hydrogels was monitored at
37 °C. Cross-linking degree of hydrogels was quantified by measuring
the number of amino groups by the ninhydrin assay [11,12].
2.5. Morphological characterization
Morphological analysis of microspheres and hydrogels was charac-
terized after the samples were freeze-dried at −50 °C and gold-coated
using a Cressington 108 Auto (Cressington, Watford, UK) for about
90 s. Morphologies were viewed using a scanning electron microscopy
(SEM) (Quanta 250FEG, FEI,USA) operated at 3.5 kV accelerating
voltage.
2.6. Fourier transform infrared (FTIR) spectrum analysis
Fourier transformed infrared (FTIR) spectra of microspheres, poly-
saccharides and formed hydrogels were measured to confirm the ex-
pected pendant functionalities. Various samples were recorded with a
FTIR spectrometer (Nicolet iS 10, Thermo Fisher Scientific, USA) by
collecting 32 accumulative scan in 500 cm−1 to 4000 cm− 1 region
against a blank KBr pellet background. The spectral resolution of FTIR
spectrometer was superior to 0.4 cm−1.
290 H. Chen et al. / Materials Science and Engineering C 70 (2017) 287–295

Fig. 5. (a) Gelation time of GMs/Gel as a function of GMs concentrations at 37 °C. Values reported are an average of n = 5, ±standard deviation. (b) Photographs of GMs/Gel as a function of
GMs concentrations after gelation. (c, d) Photographs of GMs/Gel before and after gelation with 30 mg/mL GMs.

2.7. Compressive properties 2.10. Degradation in vitro

Hydrogel dressings were prepared in an injector using the mixture of
solutions above to form columned hydrogels (15 mm diameter, 10 mm
height). The stress-strain curves of hydrogels were determined using a
Universal Testing Machine (Instron 5943, USA) at a constant displace-
ment rate loading of 1 mm/min to the stress reached the yield point. Com-
pressive moduli of samples were calculated from the stress-strain curves.
2.8. Rheological properties
Hydrogel dressings (1 mm height) were prepared by mixing the re-
action solutions and placed onto the center of the bottom plate. The
storage modulus (G′) and loss modulus (G″) varied with angular fre-
quency were investigated using a Modular Compact Rheometer
(MCR102, Anton Paar, Austria) with the parallel plate geometry
(25 mm diameter) over a frequency range (0.1–100 rad/s) at a constant
amplitude (γ = 1%) at 37 °C.
2.9. Swelling kinetics
Swelling kinetics of hydrogel dressings were characterized by moni-
toring the mass changes during incubation. The hydrogels were im-
mersed in PBS and kept at 37 °C. The swollen hydrogels were removed
at specific checkpoints and immediately weighed (Ws) after the excess
of fluid lying on the surfaces was absorbed with a filter paper. The swollen
hydrogels were lyophilized at −50 °C and weighted (Wd). The swelling
ratio was calculated using the formula of SR = (Ws − Wd) / Wd.
Degradation behavior of hydrogel dressings was characterized
by monitoring the weight loss. The initially weighed hydrogels
(W0) were immersed in PBS and put into a constant temperature
shaking table (37 °C, 70 rpm). At time points of measurement,
hydrogels were removed and weighed (Wt) after wiping the resid-
ual PBS. The weight remaining ratio was defined as 100% × Wt / W0.
The weight loss ratio was defined as 100% × (W0 − Wt) / W0. The
degradation behavior of microspheres was measured using the
similar method by putting 20 mg microspheres in 8 mL PBS.
Weight loss ratio was tested after the microspheres were washed
and freeze-dried.
2.11. TH release in vitro
For in vitro release experiment, TH release from GMs, TH-loaded
hydrogels (TH/Gel), and TH/GMs embed hydrogels (TH/GMs/Gel)
were analyzed. Group of TH release from the microspheres was deter-
mined alone by placing 10 mg microspheres in a sample vial containing
8 mL of PBS. 1 mL TH/Gel and 1 mL GMs/Gel with same drug loading ca-
pacity of TH were suspended in 10 mL of PBS. All the three group sam-
ples were put into a shaking table (37 °C, 70 rpm). A 2 mL sample of
release medium was soaked up from the sample vials periodically
and replenished with 2 mL fresh PBS to maintain a constant volume.
The TH in medium was measured at 357 nm using the BioMate 3S ul-
traviolet spectrophotometer. A cuvette filled with fresh PBS was
scanned as a blank to provide a spectral background in the ultraviolet
determination.

Fig. 6. SEM photographs of hydrogels after lyophilization. (a, b) Gel; (c, d) GMs/Gel containing 30 mg/mL GMs. (e, f) Gel after degradation of 7 days and 14 days at 37 °C in PBS; (g, h) Gel
containing 30 mg/mL GMs after degradation of 7 and 14 days at 37 °C in PBS.
 H. Chen et al. / Materials Science and Engineering C 70 (2017) 287–295
Fig. 7. (a) Swelling kinetics of Gel with different concentration of GMs as a function of time in PBS at 37 °C. (b) Degradation of Gel with different concentration of GMs as a function of time
in PBS at 37 °C. Values reported are an average of n = 3, ±standard deviation.
2.12. Antibacterial activity assay
Antibacterial properties of TH/GMs/Gel were tested with E. coli
(Gram-negative, ATCC 25922) and S. aureus (Gram-positive, ATCC
25923). The TH/Gel and pure Gel without TH were also determined as
controls. Both of bacteria were incubated in a Luria-Bertani broth at
37 °C overnight and the bacteria suspension with a certain concentra-
tion was obtained. Antimicrobial activities of hydrogel dressings were
first measured by the method of inhibition zone. The bacterial suspen-
sion was diluted with sterilized PBS to a concentration of 106–
107 CFU/mL. A bacterial culture of 100 μL was spread on nutrient
agar plates. Cylindrical sections of hydrogels (1 mm height, 10 mm
diameter) were placed onto the petri dishes. After incubation for 24 h
at 37 °C, the clear zones formed around the hydrogel sections were mea-
sured to assess the inhibitory effect. Growth inhibition effect of the hy-
drogel dressings was examined against E. coli and S. aureus [36,37].
Briefly, all hydrogel sections above were sterilized by UV light for
5 min. A 10 μL of bacterial suspension (~ 106 CFU/mL) was seeded
onto the surface of each hydrogel after the sections were placed flat
on a piece of medical gauze, which soaked with 10 mL of PBS in a culture
dish for 24 h at 37 °C. Samples were removed from the culture dish and
placed in 1 mL of PBS, which vortex stirred for 2 min to transfer the bac-
teria into the solution. After gradient dilution, 100 μL of each solution
was removed and spread onto LB agar plates. The CFU on the plates
was counted after incubation at 37 °C for 24 h. Each test was repeated
three times under the same conditions.
2.13. Statistical analysis
The experimental data from all the studies were analyzed using
analysis of variance (ANOVA). Differences between group means were
assessed using Duncan's New Multiple Range post hoc tests when neces-
sary. The data obtained from ANOVA is reported such that F (degrees of
freedomeffect, degrees of freedomError) = value. Statistical significance
Fig. 8. (a) Stress-strain curve of Gel with different concentrations of GMs at 37 °C. (b) Enlarged image of stress-strain curve for strain ranging from 0.00 to 0.10 mm/mm. (c) Compressive
moduli of Gel with GMs calculated from the stress-strain curves. Values reported are an average n = 3, ±standard deviation.
291
was set to p value ≤ 0.05. Results are presented as mean ± standard
deviation.
3. Results and discussion
3.1. Preparation of microspheres and hydrogels
We speculate that development of GMs integrated hydrogel dress-
ing with high bioactivity and suitable mechanical performance would
greatly broaden the application for wound healing. The antibacterial
Gel dressing was synthesized by mixing the solution of OAlg with
CMCS via the Schiff-base reaction, which was attributed to the amino
and aldehyde groups of CMCS and OAlg, respectively (Fig. 1). To en-
hance antibacterial properties, TH loaded GMs (TH/GMs) were fabricat-
ed by an emulsion cross-linking method, followed by integrating into
the covalent OAlg-CMCS hydrogel (Gel) to produce a composite dress-
ing. In this study, TH/GMs were cross-linked by glutaraldehyde via the
Schiff-base reaction. In this study, loading capacity of TH/GMs was
44.5 ± 0.29 μg of TH per mg of microspheres with an encapsulation
rate of 48.9 ± 0.32%. The cross-linking degree of gelatin in the GMs
and TH/GMs was calculated as 90.1% and 91.5%, respectively.
FTIR spectra of gelatin, GMs and TH were compared together to con-
firm the structure of TH/GMs (Fig. 2a). As for the gelatin (Fig. 2a curve
a), the wide absorption band around 3290 cm− 1 was due to the
stretching vibration of O\\H bonded to N\\H. The characteristic absorp-
tion bonds at 1630 cm−1, 1540 cm−1 and 1240 cm−1 were assigned to
the C_O stretching vibration (amide I), N\\H bending vibration (amide
II) and C\\N stretching vibration, respectively. By comparing with gela-
tin (Fig. 2a curve a) and TH (Fig. 2a curve b), the FTIR spectra of GMs
without TH (Fig. 2a curve c) and TH/GMs (Fig. 2a curve d) showed a
new absorbance at 1750 cm−1, which was due to C_N stretching vibra-
tion from the Schiff-base reaction. While TH/GMs showed slightly less
intense at 1750 cm−1 than GMs, it might be the addition of TH hindered
the formation of C_N bonds and the wrapped TH around the micro-
spheres influenced the FTIR detection. Furthermore, another evidence
292 H. Chen et al. / Materials Science and Engineering C 70 (2017) 287–295

to demonstrate the cross-linking was the increased intensity of the band

around 1630 cm−1, which was assigned to both C_N and C_O
stretching vibration [38,39]. Compared with the spectra of GMs (Fig.
2a curve c), the characteristic absorption bonds were viewed from the
spectra of TH/GMs (Fig. 2a curve d) due to the existence of TH.
Structurally, OAlg was obtained by breaking the carbon\\carbon
bonds of cis-diol groups to generate active aldehyde groups in molecu-
lar chain of alginate. Determination of the actual aldehyde content of
OAlg revealed an extent of oxidation of 44.5%. As a water-soluble deriv-
ative of chitosan, CMCS was synthesized by introducing carboxymethyl
groups onto the polymer backbone. To confirm the hydrogel formation,
polysaccharide derivatives and Gel were characterized by FTIR (Fig. 2b).
The absorption bonds at 3310 cm− 1, 2930 cm−1, 1580 cm− 1,
1410 cm−1 and 1020 cm−1 were the characteristic peaks of polysaccha-
ride derivatives which were due to the O\\H stretching vibration, C\\H
stretching vibration, C_O stretching vibration, collective C\\H bend-
ing/C\\N stretching vibration and C\\O stretching vibration, respective-
ly (Fig. 2b curve a,b,c). Comparing to the spectrum of alginate (Fig. 2b
curve a), the new absorption peak around 1732 cm−1 of OAlg was con-
sistent with the aldehyde group (Fig. 2b curve b). After cross-linking, it Fig. 9. Frequency dependence of storage modulus G′ and loss modulus G″ of Gel with
different concentration of GMs at 37 °C.
was difficult to detect the absorption peak of C_N stretching vibration
because of the similar location of absorption peak with C_O stretching
vibration. Meanwhile, the characteristic peak of aldehyde groups was which had the gelation time around 103 and 70 s, respectively. When
disappeared and the absorption peak at 1580 cm− 1 became broader GMs were mixed into the solution of CMCS, CMCS may be wrapped
(Fig. 2b curve d). These results suggested the Gel was successfully around the GMs. The water absorption of GMs could increase concen-
cross-linked through the Schiff-base reaction between OAlg and trations of CMCS and OAlg, which promoted the Schiff-base reaction.
CMCS. The cross-linking degree of the hydrogel was calculated as Since these biopolymer microspheres showed intrinsically hydrophilic,
82.1% by the ninhydrin method. the addition of GMs or TH/GMs would increase the concentration of
the polymers during gelation. Subsequently, the cross-linking opportu-
3.2. Characterization of microspheres nity of CMCS and OAlg was increased and the gelation time was de-
creased. In addition, GMs well distributed in hydrogel scaffolds to
Morphologies of freeze-dried GMs and TH/GMs were characterized form centers of cross-linking system, thus decreased the direct contact
by SEM (Fig. 3). Both GMs (Fig. 3a, b) and TH/GMs (Fig. 3c, d) displayed of CMCS and OAlg.
spherical morphologies with uniform particle sizes. As shown in Fig. 3b Photographs of hydrogels containing GMs with different concentra-
and d, the surface of TH/GMs showed much rough than the GMs. As de- tion were shown in Fig. 5b. With higher concentration of GMs, e.g.
termined by the Mastersizer in the aqueous environment, the hydrody- 40 mg/mL, the formed hydrogels showed the fastest gelatin rate. How-
namic diameter of GMs was 72.94 μm (Fig. 3e), while TH/GMs showed a ever, higher concentration of GMs would result in a more heteroge-
slightly larger average hydrodynamic diameter of 87.78 μm. We specu- neous mixture, which might be a problem in the clinical operation.
late that the addition of TH might modify the hydrophobic-hydrophilic Basing our results, we find that excess GMs (40 mg/mL) in gel scaffold
properties of gelatin in the TH/GMs. Also, the TH would increase hydro- led to more heterogeneous system with bubbles. Therefore, the com-
philic properties of microspheres and the inter-molecular distance of posite hydrogel with 30 mg/mL GMs was chosen as the optimized for-
the gelatin. As a kind of water soluble antibiotic, TH made TH/GMs per- mulation for further investigation. Before the gelation, the mixture
formed water swelling in water-based test environment. The hydrody- with 30 mg/mL GMs was flowing liquid, and gradually transformed
namic diameter was significantly larger than SEM results since into solid phase (Fig. 5c, d).
microspheres were readily hydrated due to hydrophilic effects of the
biopolymer [40].
Degradation behavior of GMs and TH/GMs with respect to weight
loss was measured in PBS at 37 °C (Fig. 4). As depicted in Fig. 4a, the
blank GMs showed a marked effect of weight loss, which indicated a
faster rate of degradation than that of TH/GMs (p b 0.05) during degra-
dation. After incubation for 14 days, the weight remaining ratio of GMs
and TH/GMs was 52.0% and 72.4%, respectively. SEM images showed the
surface morphologies of GMs (Fig. 4b, c) and TH/GMs (Fig. 4d, e) during
degradation. There no obvious changes were observed after degrada-
tion of 7 days between GMs (Fig. 4b) and TH/GMs (Fig. 4d). After de-
graded in PBS over 14 days, both GMs and TH/GMs became much
deformed and rough, as well as more conglutination and agglomeration.

3.3. Gelation of dressing

The gelation time of hydrogel dressings was significantly influenced

by the concentrations of GMs. As shown in Fig. 5a, the gelation time was
decreased with increasing concentrations of GMs. There were no signif-
icant differences on gelation time between 0 and 10 mg/mL GMs/Gel
(p N 0.05), and both of them were N140 s. The gelation rate of 30 and Fig. 10. Cumulative release profiles of TH from GMs, Gel, and GMs/Gel as a function of time
40 mg/mL GMs/Gel was significantly faster than others (p b 0.05), in PBS at 37 °C. Values reported are an average of n = 3, ±standard deviation.
 H. Chen et al. / Materials Science and Engineering C 70 (2017) 287–295
Fig. 11. Photographic images of inhibition zones of Gel without TH, TH/Gel and TH/GMs/Gel against (a) E. coli and (b) S. aureus after incubating at 37 °C for 24 h.
3.4. Degradation and swelling of dressing
The SEM images of freeze-dried Gel and 30 mg/mL GMs/Gel before
and after degradation were shown in Fig. 6. Before degradation, the
freeze-dried Gel (Fig. 6a, b) and GMs/Gel (Fig. 6c, d) displayed continu-
ous and porous structures, which were similar with other natural mac-
romolecular hydrogel scaffolds [12]. In addition, the participation of
GMs did not make differences in the hydrogel structure of uniform po-
rosity. With a concentration of 30 mg/mL, GMs were found to be widely
distributed onto the walls of the pores under higher magnification (Fig.
6d). Compared with the morphologies of Gel (Fig. 6b), 30 mg/mL GMs/
Gel was possessed of relatively smooth and flat surfaces (Fig. 6d). The
morphologies of Gel and 30 mg/mL GMs/Gel were also observed after
incubation in PBS (Fig. 6e–h). After degradation for 7 and 14 days, the
freeze-dried Gel (Fig. 6e, f) and GMs/Gel (Fig. 6g, h) remained continu-
ous and porous structures. Cracks and fragments from the porous struc-
ture of Gel (Fig. 6e) were clearly visible by contrast with those of GMs/
Gel (Fig. 6f).
Swelling kinetics of hydrogel dressings with different concentrations
of GMs containing TH were shown in Fig. 7a. The swelling ratios of all
hydrogels were increased along with the incubation time in PBS at
37 °C. Generally, high concentrations of GMs resulted in decrease of
swelling ratio on composite dressings. We speculate that the GMs occu-
pied the space structure of hydrogel system with limited water absorp-
tion capacity, which would like to reduce the amounts of water-uptake
in the super absorbent hydrogel network. The swelling ratio of the Gel
without GMs increased to 41.14 after 14 days, which was significantly
higher than the other hydrogels (p b 0.05). Compared with 0–10 mg/
mL GMs/Gel, the 20–40 mg/mL GMs/Gel were more stable in the pro-
cess of swelling. Furthermore, 10–40 mg/mL GMs/Gel indicated a strong
ability of water uptake in the initial 12 h, which might be attributed to
the swelling ability of both hydrogels and GMs.
Weight loss of dressings was monitored as a function of time in PBS
at 37 °C (Fig. 7b). The dressing mass significantly increased during initial
12 h, which was consistent with the swelling result (Fig. 7a). Basically,
the dressing containing less GMs was more readily degradable due to
Table 1
Growth inhibition effects of hydrogel dressings against E. coli and S. aureus after 24 h of
contact culture at 37 °C. Pure Gel to serve as the blank control. Concentration of GMs as
30 mg/mL in Gel. Concentration of TH as 1.335 mg/mL in TH/Gel and TH/GMs/Gel.
Groups Number of recovered E. coli Number of recovered S. aureus
Gel (3.93 ± 0.31) × 106 (7.80 ± 1.59) × 105
TH/Gel (3.60 ± 0.87) × 103⁎ (1.60 ± 0.62) × 103⁎
TH/GMs/Gel (2.60 ± 0.53) × 104⁎ (4.20 ± 1.80) × 103⁎
⁎ Statistical significance was set to p value ≤0.05.
293
the higher swelling ratio [27]. Compared with the 20–40 mg/mL GMs/
Gel, the 0–10 mg/mL GMs/Gel showed significant rapid weight loss
(p b 0.05). Especially as for the blank Gel without GMs, its weight re-
maining ratio was only 30.9% after 14 days. At the final stage of degrada-
tion, the 20–30 GMs/Gel demonstrated better stability than the others.
By addition of GMs or TH/GMs, the concentration of polymers was in-
creased in the gelation system. As a result, the formed dressing became
more solid, which would diminish the swelling and degradation rate.
3.5. Compressive and rheological properties
Compressive properties of GMs/Gel at various ratios were investi-
gated at 37 °C (Fig. 8a). Stress-strain curves were obtained (Fig. 8b),
and compressive modulus were calculated as Fig. 8c. With incorporation
of more microspheres, the compressive modulus of the composite
hydrogels was improved significantly (p b 0.05). The 40 mg/mL GMs/
Gel showed the best compressive performance, with the maximum
modulus of 9.21 kPa. With an excess of GMs (N40 mg/mL) filled into
hydrogels, the GMs could not be evenly dispersed. As a result, the ag-
glomeration of GMs would aggravate the emergence of the crack.
Since the concentration of polymers was increased by addition of GMs
or TH/GMs, the formed gels became more solid and showed the higher
mechanical properties. Synchronously, the amino groups of GMs or TH/
GMs would partially generate Schiff-bases with OAlg and further in-
crease the gel crosslinking, although amino groups of CMCS would
react more easily with aldehyde groups of OAlg in solution.
The rheological properties of composite hydrogels were measured
by monitoring the storage modulus (G′) and loss modulus (G″) as a
function of angular frequency at 37 °C (Fig. 9). Generally, G′ of the
hydrogels was nearly constant with the changing frequency, which
were much larger than the corresponding G″. The elastic behavior of
the hydrogels occupied a leading position rather than the viscous be-
havior, indicating characteristic of a “strong hydrogel” [41–43]. Both G′
and G″ gradually increased along with the concentration of micro-
spheres. There was no significant difference between the 30 mg/mL
and 40 mg/mL GMs/Gel (p N 0.05), and their final G′ values were both
N10 kPa. Such rheological behaviors of hydrogel dressings indicated
the appearance of a suitable material for applications in wound healing
with good elastic and solid properties.
3.6. Antibacterial activities
The cumulative release of TH from GMs, TH-loaded hydrogels (TH/
Gel), and 30 mg/mL GMs/Gel was tested at 37 °C in PBS (Fig. 10). The cu-
mulative release of TH from GMs exhibited an obvious burst release of
89.93% within 12 h, which was significantly higher than TH/Gel and
GMs/Gel (p b 0.05). After 14 days, the cumulative release of GMs
294 H. Chen et al. / Materials Science and Engineering C 70 (2017) 287–295
reached to a value of 94.54%. The reason why the cumulative release of
TH from bare GMs was higher is that part of TH might be attached to the
surface of TH/GMs, which readily released in PBS. Compared with GMs,
TH/Gel and GMs/Gel had better sustained release effect over an initial
period of 4 days. The cumulative release of TH from TH/Gel gradually
came to about 66% after 14 days. One can conclude that the double im-
pediments of microspheres and hydrogel network made the least cu-
mulative release of TH, resulting in the best sustained release effect. As
for the 30 mg/mL GMs/Gel, the cumulative release increased gradually
to 61.4% after 14 days, which was significantly steady by contrast with
the release value of TH/Gel. These results demonstrated that the
30 mg/mL GMs/Gel is capable of drug controlled delivery with good
properties of delayed release due to the double combination obstacle
of microspheres and hydrogels.
The inhibition zones of Gel, TH/Gel and GMs/Gel against E. coli and S.
aureus were measured using the disc diffusion test (Fig. 11). Comparing
with the sample of Gel without TH, clear zones of inhibition were
viewed around the samples of GMs/Gel and TH/Gel, which had little
sign of inhibition zone. The diameters of the inhibition zones for GMs/
Gel dressings against E. coli and S. aureus were 8.00 ± 1.00 mm and
8.04 ± 0.40 mm, respectively. While the diameters for TH/Gel against
E. coli and S. aureus were a little larger, with the values of 10.50 ±
0.50 mm and 10.00 ± 0.35 mm, respectively. This result may be caused
by the slight different release of TH from GMs/Gel and TH/Gel. Never-
theless, one can conclude that the GMs/Gel had similar antimicrobial ef-
fects with TH/Gel resisting to E. coli and S. aureus. The growth inhibition
effects of TH/GMs/Gel and TH/Gel against E. coli and S. aureus were ex-
hibited in Table 1. The number of recovered bacteria on the surface of
Gel was served as blank control. Compared with Gel, both TH/Gel and
TH/GMs/Gel significantly decreased the number of recovered bacteria
by two magnitude at least, a pretty good effect in aspect of inhibiting
the growth of bacteria (p b 0.05).
4. Conclusions
In summary, an antibacterial OAlg-CMCS hydrogel embed with GMs
containing TH was cross-linked as wound dressing via the Schiff-base
reaction. Increasing the concentrations of GMs in hydrogels to an
appropriate value, e.g. 30 mg/mL, the stability, compressive modulus,
G′ and G″ of composite dressing were significantly improved. In vitro
drug release studies of the composite dressing proved a sustained
release effect of TH, which was benefited from a double drug delivery
system of microsphere/hydrogel obstacle. The test of inhibition zones
demonstrated the powerful antibacterial effect of the 30 mg/mL GMs/Gel
dressing against E. coli and S. aureus. This composite dressing with excel-
lent growth inhibition effects would have a great potential treatment
against bacterial infection. It can be anticipated that this antibacterial com-
posite dressing renders a high therapeutic efficiency to enhance wound
healing.
Acknowledgements
We thankfully acknowledge the Scientific Research Foundation for
Returned Scholars (Ministry of Education of China) (HP1201106), the
Jiangsu Province Science and Technology Support Program of China
(BE2013712), the Research Fund for the Department of Orthopedics
Clinical Research Center of Jiangsu Province (BL2012002), the National
Natural Science Foundation of China (81130078, 51203074) and the
Fundamental Research Funds for the Central Universities (AE16001).
References
[1] P.T. Sudheesh Kumar, V.K. Lakshmanan, T.V. Anilkumar, C. Ramya, P. Reshmi, A.G.
Unnikrishnan, S.V. Nair, R. Jayakumar, Flexible and microporous chitosan hydro-
gel/nano ZnO composite bandages for wound dressing: in vitro and in vivo evalua-
tion, ACS Appl. Mater. Interfaces 4 (2012) 2618–2629.
[2] J.J. Elsner, I. Berdicevsky, M. Zilberman, In vitro microbial inhibition and cellular re-
sponse to novel biodegradable composite wound dressings with controlled release
of antibiotics, Acta Biomater. 7 (2011) 325–336.
[3] V.K.M. Poon, A. Burd, In vitro cytotoxity of silver: implication for clinical wound care,
Burns 30 (2004) 140–147.
[4] B.S. Atiyeh, M. Costagliola, S.N. Hayek, S.A. Dibo, Effect of silver on burn wound in-
fection control and healing: review of the literature, Burns 33 (2007) 139–148.
[5] M. Fan, J. Yan, H. Tan, D. Ben, Q. He, Z. Huang, X. Hu, Nanostructured gel scaffolds for
osteogenesis through biological assembly of biopolymers via specific nucleobase
pairing, Macromol. Biosci. 14 (2014) 1521–1527.
[6] M. Fan, Z. Zhang, J. Mao, H. Tan, Injectable multi-arm poly (ethylene glycol)/
hyaluronic acid hydrogels for adipose tissue engineering, J. Macromol. Sci. A 52
(2015) 345–352.
[7] M. Fan, Y. Ma, J. Mao, Z. Zhang, H. Tan, Cytocompatible in situ forming chitosan/
hyaluronan hydrogels via a metal-free click chemistry for soft tissue engineering,
Acta Biomater. 20 (2015) 60–68.
[8] M. Fan, Y. Ma, Z. Zhang, J. Mao, H. Tan, X. Hu, Biodegradable hyaluronic acid
hydrogels to control release of dexamethasone through aqueous Diels-Alder
chemistry for adipose tissue engineering, Mater. Sci. Eng. C 56 (2015)
311–317.
[9] H. Tan, X. Hu, Injectable in situ forming glucose-responsive dextran-based hydrogels
to deliver adipogenic factor for adipose tissue engineering, J. Appl. Polym. Sci. 126
(2012) E180–E187.
[10] H. Tan, C. Xiao, J. Sun, D. Xiong, X. Hu, Biological self-assembly of injectable hydrogel
as cell scaffold via specific nucleobase pairing, Chem. Commun. 48 (2012)
10289–10291.
[11] H. Tan, C.R. Chu, K.A. Payne, K.G. Marra, Injectable in situ forming biodegradable chi-
tosan-hyaluronic acid based hydrogels for cartilage tissue engineering, Biomaterials
30 (2009) 2499–2506.
[12] H. Tan, J.P. Rubin, K.G. Marra, Injectable in situ forming biodegradable chitosan-
hyaluronic acid based hydrogels for adipose tissue regeneration, Organogenesis 6
(2010) 173–180.
[13] H. Tan, C.M. Ramirez, N. Miljkovic, H. Li, J.P. Rubin, K.G. Marra, Thermosensitive in-
jectable hyaluronic acid hydrogel for adipose tissue engineering, Biomaterials 30
(2009) 6844–6853.
[14] H. Tan, M. Fan, Y. Ma, J. Qiu, X. Li, J. Yan, Injectable gel scaffold based on biopolymer
microspheres via an enzymatic reaction, Adv. Healthc. Mater. 3 (2014) 1769–1775.
[15] H.E. Thu, M.H. Zulfakar, S.F. Ng, Alginate based bilayer hydrocolloid films as potential
slow-release modern wound dressing, Int. J. Pharm. 434 (2012) 375–383.
[16] R. Jayakumar, M. Prabaharan, P.T.S. Kumar, S.V. Nair, H. Tamura, Biomaterials based
on chitin and chitosan in wound dressing applications, Biotechnol. Adv. 29 (2011)
322–337.
[17] J. Stephen-Haynes, E. Gibson, M. Greenwood, Chitosan: a natural solution for wound
healing, J. Community Nurs. 28 (2014) 48–53.
[18] M.G. Albu, M. Ferdes, D.A. Kaya, M.V. Ghica, I. Titorencu, L. Popa, L. Albu, Collagen
wound dressings with anti-inflammatory activity, Mol. Cryst. Liq. Cryst. 555
(2012) 271–279.
[19] S. Kondo, Y. Kuroyanagi, Development of a wound dressing composed of hyaluronic
acid and collagen sponge with epidermal growth factor, J. Biomater. Sci. Polym. Ed.
23 (2012) 629–643.
[20] B. Balakrishnan, A. Jayakrishnan, Self-cross-linking biopolymers as injectable in situ
forming biodegradable scaffolds, Biomaterials 26 (2005) 3941–3951.
[21] T.A. Holland, J.K.V. Tessmar, Y. Tabata, A.G. Mikos, Transforming growth factor-beta1
release from oligo(poly(ethylene glycol) fumarate) hydrogels in conditions that
model the cartilage wound healing environment, J. Control. Release 94 (2004)
101–114.
[22] J.A. Wieland, T.L. Houchin-Ray, L.D. Shea, Non-viral vector delivery from PEG-
hyaluronic acid hydrogels, J. Control. Release 120 (2007) 233–241.
[23] K.L. Moffat, K.G. Marra, Biodegradable poly(ethylene glycol) hydrogels crosslinked
with genipin for tissue engineering applications, J. Biomed. Mater. Res. B 71B
(2004) 181–187.
[24] H. Tan, A.J. DeFail, J.P. Rubin, C.R. Chu, K.G. Marra, Novel multiarm PEG-based
hydrogels for tissue engineering, J. Biomed. Mater. Res. A 92A (2010) 979–987.
[25] H. Tan, K.G. Marra, Injectable, biodegradable hydrogels for tissue engineering appli-
cations, Materials 3 (2010) 1746–1767.
[26] H. Tan, H. Li, J.P. Rubin, K.G. Marra, Controlled gelation and degradation rates of in-
jectable hyaluronic acid-based hydrogels through a double crosslinking strategy, J.
Tissue Eng. Regen. Med. 5 (2011) 790–797.
[27] J. Sun, C. Xiao, H. Tan, X. Hu, Covalently crosslinked hyaluronic acid-chitosan hydro-
gel containing dexamethasone as an injectable scaffold for soft tissue engineering, J.
Appl. Polym. Sci. 129 (2013) 682–688.
[28] H. Tan, Q. Shen, X. Jia, Z. Yuan, D. Xiong, Injectable nano-hybrid scaffold for
biopharmaceuticals delivery and soft tissue engineering, Macromol. Rapid
Commun. 33 (2012) 2015–2022.
[29] H. Tan, X. Gao, J. Sun, C. Xiao, X. Hu, Doubly stimulus-induced stem cell aggregation
during differentiation on biopolymer hydrogel substrate, Chem. Commun. 49
(2013) 11554–11556.
[30] H. Tan, J. Wu, L. Lao, C. Gao, Gelatin/chitosan/hyaluronan scaffold integrated with
PLGA microspheres for cartilage tissue engineering, Acta Biomater. 5 (2009)
328–337.
[31] R. Cortesi, E. Esposito, M. Osti, E. Menegatti, G. Squarzoni, S.S. Davis, C. Nastruzzi,
Dextran cross-linked gelatin microspheres as a drug delivery system, Eur. J.
Pharm. Biopharm. 47 (1999) 153–160.
[32] Y.H. Xiao, S. Ma, J.H. Chen, Z.G. Chai, F. Li, Y.J. Wang, Antibacterial activity and bond-
ing ability of an adhesive incorporating an antibacterial monomer DMAE-CB, J.
Biomed. Mater. Res. B 90B (2009) 813–817.
 H. Chen et al. / Materials Science and Engineering C 70 (2017) 287–295
[33] A.O. Elzoghby, Gelatin-based nanoparticles as drug and gene delivery systems:
reviewing three decades of research, J. Control. Release 172 (2013) 1075–1091.
[34] Y. Lan, W. Li, Y. Jiao, R. Guo, Y. Zhang, W. Xue, Y. Zhang, Therapeutic efficacy of an-
tibiotic-loaded gelatin microsphere/silk fibroin scaffolds in infected full-thickness
burns, Acta Biomater. 10 (2014) 3167–3176.
[35] N. Beyth, I. Yudovin-Farber, R. Bahir, A.J. Domb, E.I. Weiss, Antibacterial activity of
dental composites containing quaternary ammonium polyethylenimine nanoparti-
cles against Streptococcus mutans, Biomaterials 27 (2006) 3995–4002.
[36] B. Balakrishnan, S. Lesieur, D. Labarre, A. Jayakrishnan, Periodate oxidation of sodi-
um alginate in water and in ethanol-water mixture: a comparative study,
Carbohydr. Res. 340 (2005) 1425–1429.
[37] K.H. Bouhadir, K.Y. Lee, E. Alsberg, K.L. Damm, K.W. Anderson, D.J. Mooney, Degrada-
tion of partially oxidized alginate and its potential application for tissue engineering,
Biotechnol. Prog. 17 (2001) 945–950.
[38] H. Yu, C. Xiao, Synthesis and properties of novel hydrogels from oxidized konjac
glucomannan crosslinked gelatin for in vitro drug delivery, Carbohydr. Polym. 72
(2008) 479–489.
295
[39] S. Kim, Y. Kang, C.A. Krueger, M. Sen, J.B. Holcomb, D. Chen, J.C. Wenke, Y. Yang, Se-
quential delivery of BMP-2 and IGF-1 using a chitosan gel with gelatin microspheres
enhances early osteoblastic differentiation, Acta Biomater. 8 (2012) 1768–1777.
[40] W. She, K. Luo, C. Zhang, G. Wang, Y. Geng, L. Li, B. He, Z. Gu, The potential of self-
assembled, pH-responsive nanoparticles of mPEGylated peptide dendron-doxorubi-
cin conjugates for cancer therapy, Biomaterials 34 (2013) 1613–1623.
[41] M.J. Ramazani-Harandi, M.J. Zohuriaan-Mehr, A.A. Yousefi, A. Ershad-Langroudi, K.
Kabiri, Rheological determination of the swollen gel strength of superabsorbent
polymer hydrogels, Polym. Test. 25 (2006) 470–474.
[42] H. Bai, K. Sheng, P. Zhang, C. Li, G. Shi, Graphene oxide/conducting polymer compos-
ite hydrogels, J. Mater. Chem. 21 (2011) 18653–18658.
[43] H. Tan, J.P. Rubin, K.G. Marra, Direct synthesis of biodegradable polysaccharide de-
rivative hydrogels through aqueous Diels-Alder chemistry, Macromol. Rapid
Commun. 32 (2011) 905–911.