Preparation and characterization of galactosylated alginate-chitosan oligomer microcapsule for hepatocytes microencapsulation

Article abstract of DOI:10.1016/j.carbpol.2014.06.025

Galactosylated alginate (GA)-chitosan oligomer microcapsule was prepared to provide a sufficient mechanical stability, a selective permeability and an appropriate three-dimensional (3D) microenvironment for hepatocytes microencapsulation. The microcapsule has a unique asymmetric membrane structure, with a dense layer located in the inner surface and gradually decreasing toward the outside surface. The stable microcapsule was obtained when GA lower than 50%, while the permeability was increased with increasing of GA. A balance between mechanical stability and permeability was achieved through modulating membrane porosity and thickness. The optimal microcapsule displays a selective permeability allowing efficient transport of human serum albumin while effectively blocking immunoglobulin G. Hepatocytes exhibited high and long term viability (>92%), proliferability, multicellular spheroid morphology, and enhancement of liver-specific functions in the microcapsule wherein galactose moieties present chemical cues to support cell-matrix interactions while the 3D structure of the microcapsule behaves physical cues to facilitate cell-cell interactions.

Full text of DOI:10.1016/j.carbpol.2014.06.025

Carbohydrate Polymers 112 (2014) 502–511
Contents lists available at ScienceDirect
Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol
Preparation and characterization of galactosylated alginate–chitosan
oligomer microcapsule for hepatocytes microencapsulation
Meng Tiana, Bo Hanb,c, Hong Tand, Chao Youa,∗
a Department of Neurosurgery, West China Hospital, Sichuan University, Chengdu 610041, Sichuan, PR China
b Department of Surgery, Keck School of Medicine, University of Southern California, Los Angeles, CA 90032, USA
c Department of Biomedical Engineering, University of Southern California, Los Angeles, CA 90032, USA
d College of Polymer Science and Engineering, State Key Laboratory of Polymer Materials Engineering, Sichuan University, Chengdu 610065, Sichuan,
PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Galactosylated alginate (GA)–chitosan oligomer microcapsule was prepared to provide a sufficient
mechanical stability, a selective permeability and an appropriate three-dimensional (3D) microenvi-
ronment for hepatocytes microencapsulation. The microcapsule has a unique asymmetric membrane
structure, with a dense layer located in the inner surface and gradually decreasing toward the out-
side surface. The stable microcapsule was obtained when GA lower than 50%, while the permeability
was increased with increasing of GA. A balance between mechanical stability and permeability was
achieved through modulating membrane porosity and thickness. The optimal microcapsule displays a
selective permeability allowing efficient transport of human serum albumin while effectively blocking
immunoglobulin G. Hepatocytes exhibited high and long term viability (>92%), proliferability, multicel-
lular spheroid morphology, and enhancement of liver-specific functions in the microcapsule wherein
galactose moieties present chemical cues to support cell–matrix interactions while the 3D structure of
the microcapsule behaves physical cues to facilitate cell–cell interactions.
Received 19 February 2014
Received in revised form 10 June 2014
Accepted 11 June 2014
Available online 19 June 2014
Keywords:
Galactosylated alginate
Microcapsule
Mechanical stability
Selective permeability
Microenvironment
Hepatocytes
© 2014 Elsevier Ltd. All rights reserved.
1. Introduction
three-dimensional (3D) microenvironment for encapsulated cells
to form tissue-like spheroids and maintain their differentiated
Cell microencapsulation, a promising technology which is gen-
erally based on the immobilization of cells within a microcapsule
surrounded by a semipermeable membrane, has been considered
potentially therapeutic strategy for the treatment of a variety
of diseases including diabetes, liver failure, cancer, as well as
central nervous system and cardiovascular diseases (Hernández,
Orive, Murua, & Pedraz, 2010). The membrane of the microcap-
sule is expected to provide not only sufficient mechanical stability
that facilitates their production, handling, and use in applications
involving shear force such as in in vitro perfusion bioreactors and
in vivo implantation conditions, but also selective permeability that
permits the diffusion of oxygen, nutrients, waste metabolites, and
the therapeutic molecules of interest while excluding host anti-
bodies and T-cells, thus protecting the encapsulated cells from the
host immune system. Besides mechanical and selective permeable
considerations, an ideal microcapsule should also be biocompat-
ible, prepared in mild conditions, and provide an appropriate
functions. However, none of the existing microcapsules can satisfy
all of these requirements.
Alginate based microcapsules are nowadays the most widely
studied for cell microencapsulation due to their apparent biocom-
patibility and mild conditions of preparation (Goh, Heng, & Chan,
2012). These microcapsules are formed through a polyelectrolyte
complexation reaction between an alginate core and a surrounded
polycation layer such as poly-l-lysine, poly-l-ornithine, chitosan,
lactose modified chitosan, and so on (Lim & Sun, 1980; Murua et al.,
2008; Muzzarelli, 2009a). Nevertheless, the resulting microcap-
sules exhibit low mechanical resistance and poor stability, which
shows the limited success in industrial and medical applications
(Breguet, Gugerli, Pernetti, Stockar, & Marison, 2005). Instead of
the traditional high molecular weight polycations, the use of chi-
tosan oligomer permits the formation of microcapsules having
good mechanical stability, and in particular the preparation is car-
ried out in a one-step procedure under physiological conditions
(Bartkowiak & Hunkeler, 1999; Muzzarelli, 2009b), which repre-
sents a strong advantage over the other microcapsules. However,
this microcapsule has a permeability lower than 30% when diffu-
sion of dextran with a molecular weight of 70 kDa, a molecular size
∗
Corresponding author. Tel.: +86 28 85422972; fax: +86 28 85422136.
E-mail address: Youchao@vip.126.com (C. You).
http://dx.doi.org/10.1016/j.carbpol.2014.06.025
0144-8617/© 2014 Elsevier Ltd. All rights reserved.

M. Tian et al. / Carbohydrate Polymers 112 (2014) 502–511
503
similar to that of albumin, indicating that the permeability of the
microcapsule is required to be improved for cell microencapsula-
tion.
oligomer (Mn < 3000) was purchased from Zhejiang Aoxing
Biotechnology Co., Ltd. (China). Lactobionic acid and 1-ethyl-3-
(3-dimethy-lamino-propyl) carbodiimide (EDC) were supplied by
Sigma-Aldrich (USA). N-hydroxylsuccinimide (NHS) was purchased
from Pierce (USA). RPMI-1640 and phenol red-free RPMI-1640
media were obtained from Gibco (USA). Fetal bovine serum (FBS)
was purchased from HyClone (USA). Fluorescein isothiocyanate
labeled human serum albumin (FITC-HAS), fluorescein isothio-
cyanate labeled human immunoglobulin G (FITC-IgG), and urea
nitrogen kit were from Nanjing Jiancheng Bioengineering Institute
(China). Live/dead assay and Alamer blue assay kits were purchased
from Invitrogen (USA). Human albumin ELISA quantitation kit was
obtained from Bethyl Laboratories (USA).
Microcapsule can provide a 3D microenvironment for encapsu-
lated cells to mimic their in vivo environment in vitro, since it has a
liquid core, in which cells can grow freely surrounded by a semiper-
meable membrane. The presence of a 3D microenvironment that
recapitulates various aspects of the in vivo microenvironment
such as cell–matrix and cell–cell interactions has been reported to
greatly improve the cell functions in vitro (Cho et al., 2006; Huang,
Zhang, Wang, Wang, & Tang, 2012; Yang, Gotob, Ise, Cho, & Akaike,
2002). Alginate derived from brown algae as an important natu-
rally occurring carbohydrate polymer has been widely used in a
variety of biomedical applications due to its biocompatibility and
nontoxicity (Lee & Mooney, 2012; Xu et al., 2013). However, its
lack of specific interactions with cells limited its use as an appro-
priate 3D microenvironment for cells. For example, hepatocytes
encapsulated in alginate/Ca2+ microcapsules exhibited low viabil-
ity and liver functions due to the lack of cell–matrix and cell–cell
interactions (Yang et al., 2002).
Hepatocytes are anchorage-dependent cells and have the
tendency to form aggregates spontaneously in vitro. When enzy-
matically isolated from the liver and cultured in monolayer or
suspensions, they rapidly lose adult liver morphology and differ-
entiation functions. Recent study on maintenance of liver-specific
functions of hepatocytes was focused on introduction of galac-
tose moieties as asialoglycoprotein-receptor (ASGP-R) ligands to
improve the hepatocytes anchorage and the interaction of hepa-
tocytes with scaffold materials (Cho et al., 2006; Chu et al., 2009;
Chung et al., 2002; Fan, Shang, Yuan, & Yang, 2010; Gotoh, Ishizuka,
Matsuura, & Niimi, 2011; Seo, Choi, Akaike, Higuchi, & Cho, 2006a;
Seo, Kim, Choi, Akaike, & Cho, 2006b; Wang et al., 2010; Yang
et al., 2012). It was reported that hepatocytes attached onto galac-
tosylated surface changed from spreading to round shapes with
an increase of galactose moieties concentration and maintained
the differentiated functions with promoting spheroids formation,
which was shown to maintain a tissue-like cytological structure
and sustain higher levels of many differentiated functions than
hepatocytes cultured as monolayers (Yang et al., 2002). Therefore,
it has been known that galactose moieties can guide hepatocyte
adhesion through receptor-mediated interaction and enhance cell
functions such as albumin secretion and urea synthesis.
In this work, a combination of chitosan oligomer and galac-
tose moieties modification was investigated in order to provide
a microcapsule with mechanical stability, selective permeability,
and appropriate 3D microenvironment for hepatocytes microen-
capsulation. The use of chitosan oligomer permits the formation of
microcapsules in a one-step procedure under physiological condi-
tions, which will improve the viability of the encapsulated cells. The
modification with galactose moieties is used to provide cell–matrix
interactions for hepatocytes adhesion and enhance cell functions.
Moreover, the introduction of galactose moieties to the backbone
of alginate will enhance the permeability of the microcapsule, since
part of carboxylic groups along the chains of alginate was replaced.
Therefore, the mechanical stability and permeability can be mod-
ulated to achieve a balance. Finally, cell viability, proliferation,
morphology and functions such as albumin secretion and urea syn-
thesis of the hepatocytes within the microcapsule were studied.
2.2. Synthesis of galactosylated alginate (GA)
As shown in Fig. 1, the synthesis of GA was carried out by
coupling alginate with galactose moiety in the presence of EDC
and NHS in N,N,N-tetra-methylethylenediamine (TEMED) buffer
as described elsewhere with a minor modification (Chiari, Cretich,
Riva, & Casali, 2001; Fan et al., 2010; Wang et al., 2010; Yang et al.,
2002). Briefly, 10 g of lactobionic acid was dissolved in 65 ml boil-
ing 2-methoxyethanol, followed by the addition of 40 ml toluene,
which caused the precipitation of a white solid. Lactobionic lactone
was obtained by distillation of the toluene and white solid mix-
ture at 124 ◦C for three times. Then, 5 g of lactobionic lactone was
refluxed with 30-fold excess ethylenediamine (29.4 ml) dissolved
in 50 ml anhydrous methanol at 70 ◦C for 2 h. The monoamine
terminated lactobionic lactone (l-NH2) was precipitated with chlo-
roform and the obtained precipitate was vacuum-dried. Finally, 1 g
of l-NH2 was added to a stirred solution of alginate sodium (1 g) in
100 ml of 50 mM TEMED (0.3 M NaCl, pH 6.5) containing EDC and
NHS (EDC = 2.5 g, molar ratio NHS/EDC = 0.2) for 24 h. The resulting
solution was dialyzed against distilled water for 7 days at room
temperature and then lyophilized to obtain GA. The GA was char-
acterized by Fourier transform infrared spectroscopy (FTIR, Nicolet
670, USA), 1H nuclear magnetic resonance spectroscopy (1H NMR,
Bruker AV II—400 MHz, Switzerland), and elemental analysis (Vario
EL III, Germany).
2.3. Preparation of microcapsule
The microcapsules were prepared by formation of poly-
electrolyte complexes through carboxylic and amino groups of
GA/alginate and chitosan oligomer, respectively. In brief, a mixture
of GA and alginate with a stated ratio r (r = 0, 0.3, 0.5 and 1, where
r is the ratio of GA weight to sum of alginate and GA weight in the
mixture) was dissolved in 0.01 M phosphate-buffered saline (PBS)
at a concentration of 1% (w/v). Then, 1 ml of GA/alginate solution
was introduced into a 1 ml syringe with a 27 G flat-cut needle, and
the solution was extruded into 50 ml of chitosan oligomer solution
(0.8% (w/v), pH = 7.0) under gently agitation with a small magnetic
bar. The extruded droplets were allowed to complex reaction for 5,
10, and 20 min to form membrane. The formed microcapsules were
filtered, rinsed three times with PBS, and stored at 4 ◦C in PBS/0.01%
sodium azide.
2.4. Microcapsule membrane observation
2. Materials and methods
To observe the membrane of the microcapsule, the microcap-
sules were stained in 0.01% eosine/PBS for 30 min and then washed
with PBS. Membrane structure and thickness were observed using
a spinning disk confocal microscopy (Perkin Elmer, USA).
2.1. Materials
Sodium alginate (viscosity: 495.0 cps at 25 ◦C) was obtained
from Zhejiang Jingyan Biotechnology Co. Ltd (China). Chitosan

504
M. Tian et al. / Carbohydrate Polymers 112 (2014) 502–511
Fig. 1. The reaction route of GA.
2.5. Mechanical property of the microcapsule
2.6. Permeability of the microcapsule
Mechanical property of the microcapsules was determined by a
shear force test (Yin et al., 2003). About 100 microcapsules prepared
under the stated condition were added into a 6-well plate (Corn-
ing, USA). The microcapsules were equilibrated in 3 ml of PBS for
15 min. The plate was then immobilized on an orbital shaker (Model
HY-5, Zhongda Instrument, China) and agitated with a frequency
of 250 rpm at 37 ◦C. The percentage of ruptured microcapsules at
various time points was counted with a phase contrast microscope
(CKX41, Olympus, Japan).
Permeability of the microcapsule was investigated by examin-
ing the diffusion of FITC-HSA and FITC-IgG from microcapsules (Yin
et al., 2003; Zhu et al., 2005). FITC-HAS and FITC-IgG were, respec-
tively, dissolved in the stated ratio GA/alginate mixture solution
with a concentration of 1 mg/ml and then the microcapsule was
prepared as above description. For permeability assay, the micro-
capsules were quickly rinsed with PBS before they were placed
into fresh PBS for measuring the release of FITC-HAS or FITC-IgG
from the microcapsules. At the time point of 5, 10, 20, 40, 80, 120,

M. Tian et al. / Carbohydrate Polymers 112 (2014) 502–511
505
and 160 min, FITC-HAS or FITC-IgG released from the microcapsules
into PBS were collected and analyzed on a spectrofluorometer (F-
7000, Hitachi, Japan). The diffusion (%) was calculated based on the
percentage of the amount of released FITC labeled protein over the
initial amount of FITC labeled protein in the microcapsules.
Known quantities of urea nitrogen standards were used to establish
the standard curve.
2.10. Statistical analysis
Results were presented by mean standard deviation
(M S.D.). Each result was statistically analyzed by SPSS 13.0.
A Student’s t-test was performed to determine the statistical
significance between experimental groups. The values of p < 0.05
were considered statistically significant.
2.7. Hepatocytes microencapsulation
Normal human hepatocytes, L02, were cultured as described
previously (Tian et al., 2010). GA and alginate were sterilized with
UV irradiation and dissolved with PBS to prepare GA/alginate mix-
ture solutions. Hepatocytes were suspended in the mixture solution
with a cell density of 1 × 106 cells/ml and extruded into a gently
agitated chitosan oligomer solution sterilized with a 0.2 ␮m filter
(Millipore, USA) to form microcapsules as above description. The
microcapsules were harvested, washed with PBS, and subsequently
cultivated in phenol red-free RPMI-1640 medium supplemented
with 10% FBS, 100 ␮g/ml penicillin and 100 ␮g/ml streptomycin
at 37 ◦C in a 5% CO2 atmosphere, with medium changed every day.
The cells in the microcapsule were observed under a phase contrast
microscope (CKX41, Olympus, Japan).
3. Results and discussion
3.1. Synthesis and characterization of GA
In the FTIR spectra shown in Fig. 2A, lactobionic acid exhibited
characteristic absorption at 1742 cm−1, which was attributed to
the carbonyl stretching (C O) of carboxylic groups. After be dehy-
drated, the absorptions of carbonyl groups increased significantly,
indicating the formation of lactobionic lactone. The characteristic
absorption at 1650 of l-NH2 is assigned to amide I. For the GA, the
absorption at 1621 cm−1 is hard to assign, since the carbonyl groups
of alginate sodium exhibit characteristic absorptions at 1609 cm−1
.
However, a new peak at 1546 cm−1 was observed, which was con-
tributed to amide II, indicating amide bond formation.
2.8. Cell viability and proliferation
In the 1H NMR spectrum of lactobionic acid (Fig. 2B), the signals
at 4.46, 4.36, 4.27, 4.1, 3.9, 3.8, 3.75, 3.70, 3.66, 3.6, and 3.4 ppm
are assigned to H-9, H-1, H-5, H-2, H-4, H-6, H-7, H-8, H-10, H-
11, and H-3 of the lactobionic acid, respectively (Chen et al., 2012;
Kang et al., 2005; Matute et al., 2013; Villa et al., 2013; Zhang, Tang,
& Yin, 2013; Zheng et al., 2012; Zhou et al., 2013). In comparison
with the 1H NMR spectrum of lactobionic acid, the signals assigned
to H-1 and H-4 of l-NH2, respectively, shifted to 4.63 and 1.25 ppm.
The signals at 3.5 and 3.25 ppm are, respectively, assigned to H-12
and H-13 of l-NH2, which evident conjugation of ethylenediamine.
In the 1H NMR spectrum of GA, the signals assigned to H-12 can
be identified. However, the most of the characteristic peaks of l-
NH2 and alginate both appeared in the range of 3.5–4.5 ppm and
overlapped, which resulted in the difficulty to assign all galactose
residue in the GA (Fan et al., 2010; Wang et al., 2010; Yang et al.,
2002). In addition, there are more peaks appeared in the GA. To
examine the possibility that these peaks result from the incom-
plete replacement of the activated ester, the 1H NMR spectrum of
the EDC was also recorded. Indeed, two characteristic peaks could
be easily identified in the GA for EDC. The one is the triplet located in
∼1.0 ppm, which belongs to CH3 group marked H-6 in the EDC due
to a CH2 group nearby. The other is the quartet located in ∼3.0 ppm
which was assigned to H-3 of the CH2 in the EDC due to four neigh-
bor protons (Davidovich-Pinhas, Harari, & Bianco-Peled, 2009).
From the FTIR and 1H NMR results, it can be concluded that
besides the coupling of galactose moieties some EDC molecules
remain attached to the alginate backbone. This phenomenon is sim-
ilar to that reported by Davidovich-Pinhas et al. (2009). To estimate
the content of galacotose moieties and EDC, the element analy-
sis was carried out (Supplementary data). The content of galactose
moieties in the GA was evaluated by element analysis of carbon,
hydrogen and nitrogen content and it showed that 20.0% of car-
boxylic groups in alginate was replaced by galactose moieties.
Accordingly, 8% of EDC was found to remain attached to alginate
backbone. Together, a total of 28% of carboxylic groups in alginate
was replaced.
The viability of the encapsulated cells was determined using
the live/dead assay kit from Invitrogen. Briefly, 1 ␮l of ethidium
homodimer-1 and 0.25 ␮l of calcein acetoxymethyl ester from the
kit were diluted with 500 ␮l RPMI-1640 without phenol red. The
microcapsules were stained for 30 min at room temperature in the
dark and imaged with a spinning disk confocal microscopy (Perkin
Elmer, USA). Green (live cells) and red (dead cells) fluorescence
images were collected separately and merged to determine cell
viability as the ratio of viable cells to total cells counted.
The proliferation of the hepatocytes encapsulated in the micro-
capsule was determined using the Alamar Blue assay following
the manufacturer’s instructions. Briefly, phenol red free RPMI-
1640 medium containing 10% Alamar Blue was added to each
well containing microencapsulated cells and incubated for 4 h, and
then 100 ␮l of the incubation medium was transferred into a 96
well plate. The absorbance of each well was determined using a
microplate reader at a wavelength of 570 nm. The proliferation
rates are presented as fold increase over the value of the absorbance
obtained after the first day of culture (Tian et al., 2012).
2.9. Functional analysis of the microencapsulated hepatocytes
Albumin and urea synthesis were determined to evaluate the
function of the microencapsulated hepatocytes. The medium was
refreshed daily and the collected medium was centrifuged in
14,000 rpm for 10 min. The supernatant was stored at −20 ◦C for
albumin assay. The amount of albumin secreted into the medium
was quantified using a human albumin ELISA quantitation kit
(Bethyl Laboratories, USA) under the conditions recommended by
the manufacturer. To assess the urea synthesis function of the cells
microencapsulated, the culture medium was replaced with fresh
phenol red-free RPMI-1640 medium containing 5 mm NH4Cl. The
cells were cultured in this medium for 120 min, before the medium
was again replaced with the normal medium. The collected medium
was tested for urea production using a Urea Nitrogen Kit (Tian et al.,
2010). Briefly, each collected medium (20 ␮l) was mixed in a tube
with 1 ml of blood urea nitrogen (BUN) acid reagent and 1 ml of BUN
color reagent, and then heated in boiling water for exactly 15 min.
The tubes were allowed to cool in a water bath for about 5 min. The
optical density of each sample in the tube was determined using an
ultraviolet-visible spectrophotometer at a wavelength of 520 nm.
3.2. Microcapsule preparation
The microcapsule was formed by polyelectrolyte complexes
between GA/alginate and chitosan oligomer. The GA/alginate and

506
M. Tian et al. / Carbohydrate Polymers 112 (2014) 502–511
Table 1
Microcapsules prepared with different GA ratios (r is the ratio of GA weight to sum
of alginate and GA weight) and reaction time.
GA ratio (r)
Membrane-forming
time (min)
Gross
observation
Microcapsule
size (mm)
5
10
20
Non-forming
Non-forming
Non-forming
–
–
–
1
5
10
20
Soft, easily
broken
Soft, easily
broken
Soft, easily
broken
–
–
–
0.5
5
10
20
Stable
Stable
Stable
1.1 0.2
1.0 0.2
1.1 0.2
0.3
0
5
10
20
Stable
Stable
Stable
1.1 0.2
1.1 0.2
1.0 0.2
Microcapsule formation is influenced by various factors, includ-
ing polyanion and polycation concentration, the pH, and the ionic
strength (Rosas-Flores, Ramos-Ramírez, & Salazar-Montoya, 2013).
For cell encapsulation, physiological conditions were required to
maintain the viability of living cells. Thus, an additional prepara-
tion condition was set at pH 7.0 in PBS, since chitosan oligomer is
soluble due to a reduction of hydrogen-bond interactions (Domard,
2011). On the other hand, it was reported that the alginate/chitsoan
oligomer microcapsule has a permeability lower than 30% when
diffusion of dextran with a molecular weight of 70 ka (Bartkowiak
& Hunkeler, 1999). Therefore, the permeability of the microcapsule
is required to be improved for hepatocytes microencapsulation. In
the present study, 1% alginate or GA/alginate concentration was
chosen based on previous reports in which higher alginate concen-
tration resulted in higher charge density, denser membrane, and
lower permeability. It was also important to form microcapsules
using low concentrations of chitosan oligomer because the droplets
of alginate or GA/alginate would float on the surface of the chitosan
oligomer solution with high concentration due to high viscosity.
The chitosan oligomer solution with 0.8% concentration was found
to form microcapsule reliably when using 1% of alginate solution.
Although the microcapsule can be prepared with the pure
alginate solution, the GA/alginate mixture solution with high GA
ratio cannot form microcapsule. The reasons might be due to
a decrease of negative charge density on the droplet surface,
since 28% of carboxylic groups along the chains of alginate were
replaced by galactose moieties or EDC as mentioned above, which
resulted in weak polyelectrolyte complexes between GA and chi-
tosan oligomer. This phenomenon was also present in preparation
of other microcapsules, for instance, Yang et al. (2002) prepared a
Ca2+ crosslinked microcapsule using a mixture of GA and alginate
with a weight ratio of 1:1.
Fig. 2. (A) FTIR spectra of lactobionic acid (a), lactobionic lactone (b), l-NH2 (c),
sodium alginate (d), and GA (e). (B) 1H NMR spectra of lactobionic acid (a), l-NH2
(b), EDC (c), sodium alginate (d), and GA (e).
3.3. Membrane characterization
The confocal micrographs of the microcapsules are shown in
Fig. 3B. All microcapsules exhibit an asymmetric membrane struc-
ture and coarse outside surfaces. A dense layer is located in the
inner surface, and gradually decreases toward the outside sur-
face. The membrane thickness of the microcapsule containing 30%
GA is, respectively, 35 6, 50 5, and 61 7 ␮m, which have no
significant differences with that the microcapsule without GA at
various time points. However, with the GA content increases to 50%,
the membrane thickness decreases significantly (p < 0.05), with
chitosan oligomer concentration were 1% and 0.8%, respectively.
The effects of the ratio of GA to alginate and reaction time on
the microcapsule are shown in Table 1. The microcapsule can
not be formed when the mixture containing 100% GA (r = 1) was
extruded into the chitosan oligomer solution. The formed micro-
capsule prepared with r = 0.5 was too soft and easily broken. The
stable microcapsule was obtained when r is 0.3 or 0. The prepared
microcapsules are shown in Fig. 3A, with a diameter of about 1 mm.

M. Tian et al. / Carbohydrate Polymers 112 (2014) 502–511
507
Fig. 4. The mechanical property of the microcapsule. Error bars represent
means SD for n = 3.
asymmetric membrane structure with a dense skin layer at the out-
side surface and decrease toward to the core of microcapsule. In
our present work, it was found for the first time that the microcap-
sules have an asymmetric membrane structure, with a dense layer
located in the inner surface and gradually decreasing toward the
outside surface. This unique structure is different from the previ-
ously reported membrane structure in which skin layer commonly
lacated in the outside surface (Bartkowiak, 2001; Bartkowiak &
Hunkeler, 1999). Although the mechanisms underlying this phe-
nomenon are not completely known, some reasons can be deduced
from the process of membrane formation. First, after initial skin
layer construction, the membrane grows slowly with time not only
in the direction of alginate but also in the direction of chitosan
oligomer, which was observed by using the analytical ultracentrifu-
gation technique that alginate diffusion to the outside surface of the
microcapsule through the large pores on the skin layer (Bartkowiak,
2001). Second, the microcapsule prepared from a pair of oppositely
charged polysaccharides with relative low concentrations resulted
in increase of the membrane porosity which has the benefit to
increase the diffusion of alginate (Zhu et al., 2005). The membrane
thickness of the microcapsule increased with the increase in reac-
tion time, while at the same reaction time, the thickness decreased
with the increase of GA content, which resulted from weak poly-
electrolyte complexes between GA and chitosan oligomer due to
part of carboxylic groups along the chains of alginate were replaced.
Fig. 3. (A) The appearance of galactosylated alginate–chitosan oligomer microcap-
sule (r = 0.3, t = 10 min, microcapsules were stained with methylene blue). (B) The
confocal micrographs of the microcapsules prepared with r = 0, 0.3 and 0.5 after
reaction time of 5, 10, and 20 min, respectively, scale bar, 50 ␮m.
3.4. Properties of the microcapsule
thickness of 9 3, 24 5, and 32 8 ␮m after reaction time of 5,
10, and 20 min, respectively.
Fig. 4 shows the mechanical stability of the prepared microcap-
sule. The rapture percentage of the microcapsule formed with 50%
GA increases with the shaking time prolonged, with percentage
of 89.0%, 50.0%, and 32.3% after 12 h of shaking, respectively, for
membrane reaction of 5, 10, and 20 min, indicating poor mechan-
ical stability. In contrast, both the microcapsules formed with 0%
and 30% GA exhibited improved mechanical stability. In addition,
it can be observed that the mechanical stability increased with the
membrane thickness, with 4% and 1% rapture percentage, respec-
tively, for the microcapsules formed with 30% GA with membrane
thickness of 50 and 61 ␮m.
Due to the microcapsule formed with 50% GA exhibits poor
mechanical stability, the permeability study focused on the micro-
capsule formed with 30% and 0% GA. As shown in Fig. 5A, the release
profile of FITC-HSA from all the microcapsules exhibited two stages.
The first stage would be one of rapid burst releases, in which about
Membrane structure and thickness are critical as they ultimately
influence the properties of the microcapsule. The membrane struc-
ture can be divided into two major types, i.e., homogeneous and
heterogeneous membrane (or termed symmetric and asymmet-
ric membrane). For the classic alginate-poly-l-lysine microcapsule,
the membrane structure can range from almost homogeneous
to highly heterogeneous depending on the selected gelling ions,
the conditions used for coating alginate microbeads by poly-l-
lysine, and the effects of washing and storing of microcapsules
(Strand, Morch, Espevik, & Skjak-Braek, 2003). In most cases,
however, microcapsules prepared by utilizing a polyelectrolyte
complexation reaction between two oppositely charged polysac-
charides form an asymmetric membrane structure as a result of
the phase separation process. The formed membrane exhibits an

508
M. Tian et al. / Carbohydrate Polymers 112 (2014) 502–511
transport of nutrients, oxygen, and the therapeutic molecules of
interest while blocking the humoral component of the immune sys-
tem such as IgG to induce undesired inflammatory and immune
responses that can cause cell death (Zhang, Zhao, et al., 2013).
The mechanical stability and permeability of the microcapsule are
significantly dependent on the membrane thickness and porosity
(Breguet et al., 2005; Ma et al., 2013; Zhu et al., 2005). A thick and
dense membrane could provide good mechanical stability. Never-
theless, a thin and loose membrane reduces the diffusion distance
and resistance and favors mass transfer. For modulating the prop-
erties of microcapsules, a balance has to be achieved between the
membrane thickness and porosity, both of which varied greatly as
a function of the preparation conditions, including polymer com-
position, concentration, and reaction time. Ma et al. (2013) studied
the effect of membrane thickness (10, 15, 20, and 30 ␮m) on the
strength and permeability of the alginate-poly-l-lysine-alginate
microcapsule with the similar membrane porosity. The integrity
ratio of microcapsules with 15, 20, and 30 ␮m membrane thick-
nesses maintained over 88% during culture time and there was no
significant difference until 21 days. However, the microcapsules
with 10 ␮m membrane thickness were too weak and with a rap-
ture percentage more than 40%. In the aspect of permeability, the
microcapsules with 10 and 20 ␮m membrane thickness showed
higher BSA diffusion than others. So, the membrane thickness of
20 ␮m is optimal in this regard.
Alginate–chitosan oligomer microcapsule has shown good
mechanical stability which mainly results from a significantly
thicker membrane compared to other microcapsules (Bartkowiak,
2001); however, it in turn leads to a limited permeability when
diffusion of dextran with a molecular weight of 70 ka, a molecu-
lar size similar to that of albumin. To modulate the microcapsule’s
permeability without significantly compromising the mechanical
stability, efforts are made to increase the porosity of the mem-
brane in this work. Meanwhile, the membrane thickness can be
adjusted with reaction time. Here, we first prepared the microcap-
sule using a low concentrations of alginate and chitosan oligomer.
The results showed the microcapsule exhibited a unique mem-
brane structure and with a membrane thickness ranging between
30 and 60 ␮m, and the permeability of albumin was obviously
improved. Moreover, the permeability was further improved when
the microcapsule formation with GA, since part of carboxylic groups
along the chains of alginate were replaced which resulted in weak
polyelectrolyte complexes between alginate and chitosan oligomer
and formation loose membranes. On the other hand, to protect
the encapsulated cells from host immune destruction, the mem-
brane of the microcapsule is expected to block the immune related
molecules such as IgG (Zhang et al., 2012). The prepared micro-
capsule showed minor diffusion of IgG at initial stage and then
approaching release equilibrium, suggesting the potential to be
able to function as an immunoprotection barrier. Therefore, it can
be concluded that the microcapsules have a selective permeability
that allows not only effective transport of human serum albumin
but minimized diffusion of the human immune system such as IgG
to protect the encapsulated cells. Given an appropriate mechan-
ical stability, the optimal microcapsule for cell encapsulation was
prepared with 30% GA and 10 min reaction time, which has a mem-
brane thickness of 50 ␮m and an albumin permeability of 70% after
180 min, approximately more than two times higher than that of
previous report (Bartkowiak & Hunkeler, 1999).
Fig. 5. The FITC-HSA (A) and FITC-IgG (B) release curves of the microcapsule. Error
bars represent means SD for n = 3.
20% of FITC-HSA was released during the initial 20 min. The sec-
ond stage would be one of slow releases, which was the extended
release of the protein that was retarded in the microcapsule. The
microcapsules formed with 5 min of membrane reaction showed
faster release than that of microcapsule formed with 10 and 20 min
membrane reaction. About 50% of FITC-HSA was released from the
microcapsule in the initial 50 min, reaching 90% after 180 min. The
microcapsules formed with 30% GA showed faster release than
that of microcapsule formed without GA. After 180 min of release,
both microcapsules containing GA formed with 10 and 20 min of
membrane reaction time showed about 70% FITC-HSA release. In
contrast, the release of the microcapsules formed without GA was
about 60%. The release profile of FITC-IgG is illustrated in Fig. 5B.
All microcapsules displayed similar release tendency, in which the
release percentage was increased with the time increases during
the initial 100 min, followed by approaching release equilibrium,
with about 16% of FITC-IgG released from the microcapsule after
180 min. Considering that the microcapsules were immersed in
the FITC-IgG solution and allowed to equilibrate resulting in the
adsorption of IgG on the surface of the microcapsule before deter-
mination of the release profile, this suggests that the actual release
amounts should be lower than that of the determined results.
For cell microencapsulation, the membrane of microcapsule
should provide, on one hand, a sufficient mechanical resistance to
bear both the forces exerted by the in vitro and in vivo environ-
ment, and on the other hand, a selective permeability to efficient
3.5. Hepatocytes microencapsulation
The optimal microcapsule that prepared with 30% GA and
10 min reaction time was selected to encapsulate hepatocytes
based on the above structure and property tests. The microcap-
sule formed with the same preparation conditions without GA was

M. Tian et al. / Carbohydrate Polymers 112 (2014) 502–511
509
used as a control. In both cases, the microcapsules were found to be
stable and maintain their morphological integrity in the cell culture
medium after 10 days, despite with a slight swelling, suggest-
ing that the membrane of the microcapsule provides a sufficient
mechanical stability for cell culture.
cell proliferation was observed on days 7 and 10 with decreases of
about 12% and 18%, respectively. For the GA-containing microcap-
sule, the cells showed obvious proliferation on day 5 and then no
significant increase was observed on day 7 and 10.
The morphology of hepatocytes encapsulated in the microcap-
sule is shown in Fig. 6D. In the microcapsule containing GA, the
hepatocytes rapidly aggregated to form multicellular spheroids
on day 2. With the time prolonged, the number and size of
the spheroids increased, with diameters enlarged up to 100 ␮m,
indicating a specific interaction of hepatocytes with GA. On the
contrary, although most hepatocytes remained as single cells at
the beginning culture, some of cells formed aggregates after 5 days
within the control microcapsule. This is consistent with the previ-
ous reports (Seo et al., 2005; Yang et al., 2002), indicating that the
3D structure of the microcapsule support cell–cell interactions at
some extent.
The albumin secretion and urea synthesis of the encapsulated
hepatocytes within the GA–chitosan oligomer and the control
microcapsule are shown in Fig. 7A and B. The albumin secretion
of hepatocytes within the control microcapsule decreased dur-
ing 10 days culture whereas the hepatocytes within GA–chitosan
oligomer microcapsule showed increase of albumin secretion and
higher than that of control after 7 days culture (p < 0.05). Simi-
larly, the hepatocytes within GA–chitosan oligomer microcapsule
showed higher urea synthesis than that of control after 5 days cul-
ture (p < 0.05).
The viability of hepatocytes is shown in Fig. 6A and B. During
10 days culture, the cells in the GA-containing microcapsule dis-
played and maintained high viability more than 92%. The cells in
the microcapsule without GA also exhibited high viability at the
initial time, with 92% and 90% on day 2 and 5, respectively, and
then the viability decreased to 86% on day 10. The cell viability
is related with the preparation conditions, microcapsule size, per-
meability as well as 3D microenvironment (Tran et al., 2013). The
preparation conditions have distinct effect on the viability of the
encapsulated cells at the initial time. Although two-step procedure
is widely used to prepare microcapsules such as alginate-poly-
l-lysine and alginate–chitosan microcapsules, the encapsulated
cells are confined in a noncultivating environment and exposed
to many harmful elements (Yang et al., 2002). In contrast, one-
step procedure used in this work is preferred due to operation
in mind conditions and resulted in high viability. Small size and
good permeability is beneficial to diffusion of oxygen, nutrients,
waste metabolites which are essential for cell viability. The 3D
microenvironment that recapitulates various aspects of the in vivo
microenvironment such as cell–matrix and cell–cell interactions
are also essential for hepatocyte viability. Cell–matrix interactions
support cell anchorage and attachment, and cell–cell interactions
result in multicellular aggregates with a structural organization
mimicking the hepatocytes in vivo.
Microcapsule should provide an appropriate 3D microenviron-
ment for the encapsulated cells, not only for cell growth but also
for maintaining and promotion of functional expressions. Hepato-
cytes are anchorage-dependent cells. In 2D cultures, the cells are
unable to attach to the pure alginate matrix and to maintain sur-
vival due to the highly hydrated anionic surface that resists cell
Cell proliferation within microcapsules was assessed using Ala-
mar blue assay. As shown in Fig. 6C, after 10 days of culture, no cell
proliferation was observed in the control group and a reduction of
Fig. 6. (A) Live/dead staining of the hepatocytes in the GA-containing microcapsule (b1 and b2) and control (a1 and a2). Scale bar, 100 ␮m. (B) Viability of the hepatocytes
in the microcapsule. Error bars represent means SD for n = 3. (C) Proliferation of the hepatocytes in the microcapsule. Error bars represent means SD for n = 3, * p < 0.05.
(D) Phase-contrast micrographs of hepatocytes in the GA-containing microcapsule (b1–b3) and control (a1–a3). Scale bar, 100 ␮m.

510
M. Tian et al. / Carbohydrate Polymers 112 (2014) 502–511
The results showed that the hepatocytes encapsulated in the GA-
containing microcapsule exhibited high and long term viability,
proliferability, multicellular spheroid morphology, and enhance-
ment of liver-specific functions. Therefore, it may be considered
that the galactose moieties in the microcapsule supported hepato-
cytes growth and promoted the formation of hepatocyte spheroids,
and then enhanced the cell functions. Unlike integrin, ASGP-R
does not physiologically function as an adhesion receptor. The
galactose moieties are more likely provide anchorage sites for
hepatocytes, followed by incorporation of integrin mediated inter-
actions involved in the recruitment of ECM proteins secreted by
the hepatocytes during culture time. Ambury, Merry, and Ulijn
(2011) recently revealed hepatocytes attached preferentially on the
surface of polyethylene glycol acrylamide (PEGDA) hydrogel modi-
fied with galactose moieties. Moreover, they observed the secreted
ECM proteins adsorption to the PEGDA and galactose moieties-
PEGDA surfaces was similar. This indicated the proteins may attach
in different orientations due to the sugars presented at the sur-
face of the material. Taken together, GA-containing microcapsule
was indicated to be favorable in providing a 3D microenvironment
for hepatocytes microencapsulation, in which galactose moieties
present chemical cues to support cell–matrix interactions for
hepatocytes, while the 3D structure of the microcapsule behaves
physical cues to facilitate cell–cell interactions.
4. Conclusion
The present study showed the microcapsule prepared with GA
and chitosan oligomer having a GA content lower than 50% in
the liquid core provide a sufficient mechanical stability, a selec-
tive permeability and an appropriate 3D microenvironment for
hepatocytes microencapsulation. The microcapsule has a unique
asymmetric membrane structure, with a dense layer located in
the inner surface and gradually decreasing toward the outside
surface. Hepatocytes in the microcapsule exhibited high and long
term viability, proliferability, multicellular spheroid morphology,
and enhancement of liver-specific functions. The results suggest
that the prepared microcapsule should be a promising system for
hepatocytes microencapsulation, which can be used in a variety of
applications such as bioartificial liver, hepatocyte transplantation
as well as drug assessment.
Fig. 7. The albumin secretion (A) and urea synthesis (B) of the hepatocytes in the
microcapsule. Error bars represent means SD for n = 3, * p < 0.05.
adhesion and spreading (Seo et al., 2005). The situation changes
when cells are exposed to 3D cultures wherein cells tend to aggre-
gate to form cellular spheroids due to cell–cell interactions. In the
native liver tissue, hepatocytes entrapped within a 3D microen-
vironment composed of cells and extracellular matrix (ECM), in
which both cell–cell and cell–matrix interactions are facilitated
to cell viability, proliferation as well as functions. To imitate this
natural microenvironment in alginate microcapsule, cell–matrix
interactions are emphasized in our study since cell–cell interactions
are available in the microcapsules. One of the potential strate-
gies employed to improve cell–matrix interactions is to couple
bioactive molecules that are recognized by cell surface receptors.
For example, many studies have reported that cell adhesion was
regulated through immobilization of molecules containing RGD,
a ligand for integrin cell adhesion receptors (Hersel, Dahmen, &
Kessler, 2003). ASGP-R is abundantly located on the membrane
surface of hepatocytes where it binds and endocytosis galactose-
terminated glycoproteins. The use of coupling galactose moieties
for ASGP-R recognition has been regarded as one of the most effec-
tive approaches to enhance cell–matrix interactions of hepatocytes
(Nugraha et al., 2011; Qiu et al., 2012; Seo et al., 2005; Yang et al.,
2012) and hepatocyte-targeted delivery vehicles (Craparo, Triolo,
Pitarresi, Giammona, & Cavallaro, 2013; Li et al., 2013).
Acknowledgements
This work was partly supported by US NIH grant R41AR056177
and Translational Medicine Fund of West China Hospital of Sichuan
University & Science and Technology Bureau of Chengdu ZH13013.
The authors acknowledge Ernesto Barron, Douglas Hauser, and
Hong Dan, researchers at Doheny Eye Institute of University of
Southern California, for valuable suggestion and specimen char-
acterization.
References
Ambury, R. F., Merry, C. L. R., & Ulijn, R. V. (2011). Sugar functionalised PEGA sur-
faces support metabolically active hepatocytes. Journal of Materials Chemistry,
21, 2901–2908.
Bartkowiak, A. (2001). Optimal conditions of transplantable binary polyelectrolyte
microcapsules. Annals of the New York Academy of Sciences, 944, 120–134.
Bartkowiak, A., & Hunkeler, D. (1999). Alginate–oligochitosan microcapsules: A
mechanistic study relating membrane and capsule properties to reaction con-
ditions. Chemistry of Materials, 11, 2486–2492.
Breguet, V., Gugerli, R., Pernetti, M., Stockar, U., & Marison, I. W. (2005). Formation
of microcapsules from polyelectrolyte and covalent interactions. Langmuir, 21,
9764–9772.
Chen, F., Tian, M., Zhang, D., Wang, J., Wang, Q., Yu, X., et al. (2012). Preparation
and characterization of oxidized alginate covalently cross-linked galactosylated
chitosan scaffold for liver tissue engineering. Materials Science and Engineering
C, 32, 310–320.
Herein, galactose moieties were coupled to the alginate back-
bone and then the microcapsule containing galactose moieties
was prepared to provide a 3D microenvironment for hepato-
cytes that supports both cell–cell and cell–matrix interactions.

M. Tian et al. / Carbohydrate Polymers 112 (2014) 502–511
511
Chiari, M., Cretich, M., Riva, S., & Casali, M. (2001). Performances of new sugar-
bearing poly(acrylamide) copolymers as DNA sieving matrices and capillary
coatings for electrophoresis. Electrophoresis, 22, 699–706.
Qiu, Y., Mao, Z., Zhao, Y., Zhang, J., Guo, Q., Gou, Z., et al. (2012). Polycaprolactone
scaffold modified with galactosylated chitosan for hepatocyte culture. Macro-
molecular Research, 20, 283–291.
Cho, C. S., Seo, S. J., Park, I. K., Kim, S. H., Kim, T. H., Hoshiba, T., et al. (2006).
Galactose-carrying polymers as extracellular matrices for liver tissue engineer-
ing. Biomaterials, 27, 576–585.
Rosas-Flores, W., Ramos-Ramírez, E. G., & Salazar-Montoya, J. A. (2013). Microen-
capsulation of Lactobacillus helveticus and Lactobacillus delbrueckii using alginate
and gellan gum. Carbohydrate Polymers, 98, 1011–1017.
Chu, X., Shi, X., Feng, Z., Gu, J., Xu, H., Zhang, Y., et al. (2009). In vitro evaluation of
a multi-layer radial-flow bioreactor based on galactosylated chitosan nanofiber
scaffolds. Biomaterials, 30, 4533–4538.
Seo, S., Akaike, T., Choi, Y., Shirakawa, M., Kang, I., & Cho, C. (2005). Alginate
microcapsules prepared with xyloglucan as a synthetic extracellular matrix for
hepatocyte attachment. Biomaterials, 26, 3607–3615.
Chung, T. W., Yang, J., Akaike, T., Cho, K. Y., Nah, J. W., Kim, S. I., et al. (2002). Prepa-
ration of alginate/galactosylated chitosan scaffold for hepatocyte attachment.
Biomaterials, 23, 2827–2834.
Seo, S., Choi, Y., Akaike, T., Higuchi, A., & Cho, C. (2006). Alginate/galactosylated chi-
tosan/heparin scaffold as a new synthetic extracellular matrix for hepatocytes.
Tissue Engineering, 12, 33–44.
Craparo, E. F., Triolo, D., Pitarresi, G., Giammona, G.,
&
Cavallaro, G. (2013).
Seo, S., Kim, I., Choi, Y., Akaike, T., & Cho, C. (2006). Enhanced liver functions of
hepatocytes cocultured with NIH 3T3 in the alginate/galactosylated chitosan
scaffold. Biomaterials, 27, 1487–1495.
Galactosylated micelles for ribavirin prodrug targeting to hepatocytes.
a
Biomacromolecules, 14, 1838–1849.
Davidovich-Pinhas, M., Harari, O., & Bianco-Peled, H. (2009). Evaluating the mucoad-
hesive properties of drug delivery systems based on hydrated thiolated alginate.
Journal of Controlled Release, 136, 38–44.
Strand, B. L., Morch, Y. A., Espevik, T., & Skjak-Braek, G. (2003). Visualization
of alginate-poly-l-lysine-alginate microcapsules by confocal laser scanning
microscopy. Biotechnology and Bioengineering, 82, 386–394.
Domard, A. (2011). A perspective on 30 years research on chitin and chitosan. Car-
bohydrate Polymers, 84, 696–703.
Tian, M., Chen, F., Ren, D., Yu, X., Zhang, X., Zhong, R., et al. (2010). Preparation of a
series of chitooligomers and their effect on hepatocytes. Carbohydrate Polymers,
79, 137–144.
Tian, M., Yang, Z., Kuwahara, K., Nimni, M. E., Wan, C., & Han, B. (2012). Delivery of
demineralized bone matrix powder using a thermogelling chitosan carrier. Acta
Biomaterialia, 8, 753–762.
Tran, N. M., Dufresne, M., Duverlie, G., Castelain, S., Défarge, C., Paullier, P., et al.
(2013). An appropriate selection of a 3D alginate culture model for hepatic Huh-
7 cell line encapsulation intended for viral studies. Tissue Engineering, A, 19,
103–113.
Villa, R., Cerroni, B., Viganò, L., Margheritelli, S., Abolafio, G., Oddo, L., et al. (2013).
Targeted doxorubicin delivery by chitosan-galactosylated modified polymer
microbubbles to hepatocarcinoma cells. Colloids and Surfaces B: Biointerfaces,
110, 434–442.
Wang, J., Xiao, F., Zhao, Y., Chen, L., Zhang, R., & Guo, G. (2010). Cell proliferation
and thermally induced cell detachment of galactosylated thermo-responsive
hydrogels. Carbohydrate Polymers, 82, 578–584.
Xu, Y., Li, L., Wang, H., Yu, X., Gu, Z., Huang, C., et al. (2013). In vitro cytocompatibil-
ity evaluation of alginate dialdehyde for biological tissue fixation. Carbohydrate
Polymers, 92, 448–454.
Fan, J., Shang, Y., Yuan, Y., & Yang, J. (2010). Preparation and characterization of chi-
tosan/galactosylated hyaluronic acid scaffolds for primary hepatocytes culture.
Journal of Materials Science: Materials in Medicine, 21, 319–327.
Goh, C. H., Heng, P. W. S., & Chan, L. W. (2012). Alginates as a useful natural polymer
for microencapsulation and therapeutic applications. Carbohydrate Polymers, 88,
1–12.
Gotoh, Y., Ishizuka, Y., Matsuura, T., & Niimi, S. (2011). Spheroid formation and
expression of liver-specific functions of human hepatocellular carcinoma-
derived FLC-4 cells cultured in lactose–silk fibroin conjugate sponges.
Biomacromolecules, 12, 1532–1539.
Hernández, R. M., Orive, G., Murua, A., & Pedraz, J. L. (2010). Microcapsules and micro-
carriers for in situ cell delivery. Advanced Drug Delivery Reviews, 62, 711–730.
Hersel, U., Dahmen, C., & Kessler, H. (2003). RGD modified polymers: Biomaterials
for stimulated cell adhesion and beyond. Biomaterials, 24, 4385–4415.
Huang, X., Zhang, X., Wang, X., Wang, C., & Tang, B. (2012). Microenvironment of
alginate-based microcapsules for cell culture and tissue engineering. Journal of
Bioscience and Bioengineering, 114, 1–8.
Kang, I. K., Moon, J. S., Jeon, H. M., Meng, W., Kim, Y. I., Hwang, Y. J., et al. (2005).
Morphology and metabolism of Ba–alginate encapsulated hepatocytes with
galactosylated poly(allyl amine) and poly (vinyl alcohol) as extracellular matri-
ces. Journal of Materials Science: Materials in Medicine, 16, 533–539.
Lee, K. Y., & Mooney, D. J. (2012). Alginate: Properties and biomedical applications.
Progress in Polymer Science, 37, 106–126.
Li, C., Zhang, D., Guo, H., Hao, L., Zheng, D., Liu, G., et al. (2013). Preparation and
characterization of galactosylated bovine serum albumin nanoparticles for liver-
targeted delivery of oridonin. International Journal of Pharmaceutics, 448, 79–86.
Lim, F., & Sun, A. M. (1980). Microencapsulated islets as bioartificial endocrine pan-
creas. Science, 210, 908–910.
Ma, Y., Zhang, Y., Wang, Y., Wang, Q., Tan, M., Liu, Y., et al. (2013). Study of the
effect of membrane thickness on microcapsule strength, permeability, and cell
proliferation. Journal of Biomedical Material Research, A, 101A, 1007–1015.
Matute, A., Cardelle-Cobas, A., García-Bermejo, A., Montilla, A., Olano, A., & Corzo,
N. (2013). Synthesis, characterization and functional properties of galactosy-
lated derivatives of chitosan through amide formation. Food Hydrocolloids, 33,
245–255.
Murua, A., Portero, A., Orive, G., Hernández, R. M., Castro, M., & Pedraz, J. L. (2008).
Cell microencapsulation technology: Towards clinical application. Journal of
Controlled Release, 132, 76–83.
Muzzarelli, R. A. A. (2009a). Genipin-crosslinked chitosan hydrogels as biomedical
and pharmaceutical aids. Carbohydrate Polymers, 77, 1–9.
Yang, J., Gotob, M., Ise, H., Cho, C. S.,
&
Akaike, T. (2002). Galactosy-
lated alginate as
471–479.
a scaffold for hepatocytes entrapment. Biomaterials, 23,
Yang, N., Chen, L., Yang, M., Bi, S., He, X., Zhu, Z., et al. (2012). In vitro study of the inter-
actions of galactosylated thermo-responsive hydrogels with cells. Carbohydrate
Polymers, 88, 509–516.
Yin, C., Chia, S. M., Quek, H. C., Yu, H., Zhuo, R., Leong, K., et al. (2003). Microcap-
sules with improved mechanical stability for hepatocyte culture. Biomaterials,
24, 1771–1780.
Zhang, J., Tang, C., & Yin, C. (2013). Galactosylated trimethyl chitosanecysteine
nanoparticles loaded with Map4k4 siRNA for targeting activated macrophages.
Biomaterials, 34, 3667–3677.
Zhang, W., Zhao, S., Rao, W., Snyder, J., Choi, J. K., Wang, J., et al. (2013). A novel
core–shell microcapsule for encapsulation and 3D culture of embryonic stem
cells. Journal of Materials Chemistry B, 1, 1002–1009.
Zhang, Y., Shi, X., Han, B., Gu, J., Chu, X., Xiao, J., et al. (2012). The influence of mem-
brane molecular weight cutoff on a novel bioartificial liver. Artificial Organs, 36,
86–93.
Zheng, D., Duan, C., Zhang, D., Jia, L., Liu, G., Liu, Y., et al. (2012). Galactosylated chi-
tosan nanoparticles for hepatocyte-targeted delivery of oridonin. International
Journal of Pharmaceutics, 436, 379–386.
Zhou, N., Zan, X., Wang, Z., Wu, H., Yin, D., Liao, C., et al. (2013). Galactosylated
chitosan-polycaprolactone nanoparticles for hepatocyte-targeted delivery of
curcumin. Carbohydrate Polymers, 94, 420–429.
Zhu, J. H., Wang, X. W., Ng, S., Quek, C. H., Ho, H. T., Lao, X. J., et al. (2005). Encapsu-
lating live cells with water-soluble chitosan in physiological conditions. Journal
of Biotechnology, 117, 355–365.
Muzzarelli, R. A. A. (2009b). Chitins and chitosans for the repair of wounded skin,
nerve, cartilage and bone. Carbohydrate Polymers, 76, 167–182.
Nugraha, B., Hong, X., Mo, X., Tan, L., Zhang, W., Chan, P. M., et al. (2011). Galac-
tosylated cellulosic sponge for multi-well drug safety testing. Biomaterials, 32,
6982–6994.