Journal of Controlled Release 77 (2001) 297–307

www.elsevier.com / locate / jconrel

Factors affecting protein release from alginate–chitosan

coacervate microcapsules during production and gastric /
intestinal simulation
G.W. Vandenberg, C. Drolet, S.L. Scott 1 , J. de la Noue
¨ *
´
Groupe de recherche en recyclage biologique et aquiculture. Departement des sciences animales, Universite´ Laval,
´ , Canada G1 K-7 P4
Pavillon Paul Comtois, Ste-Foy, Quebec
Received 26 March 2001; accepted 3 October 2001

Abstract

A series of experiments was performed to evaluate the influence of a number of physico–chemical factors on the diffusion
of a model protein, bovine serum albumin (BSA), from dried chitosan-coated alginate microcapsules. Diffusion of BSA was
quantified during the microcapsule manufacture processes (gelation, washing, rinsing) and during incubation in conditions
simulating the pH encountered during the gastric (0.1 N HCl; pH 1.5) and intestinal (200 mM Tris–HCl; pH 7.5) phases of
digestion. Factors tested included alginate and chitosan concentration, calcium chloride (CaCl 2 ) concentration in the gelation
medium, loading rate, chitosan molecular mass and pH of the gelation medium. Microcapsule size and gelation time were
altered in order to determine their effects on protein retention. Alginate and chitosan concentration significantly influenced
BSA retention during microcapsule manufacture and acid incubation, as did calcium chloride concentration in the gelation
medium (P,0.05). BSA retention during manufacture was not significantly altered by protein loading rate or pH of the
encapsulation medium, however, protein retention during acid incubation decreased significantly with increasing protein
loading rate and encapsulation medium pH (P,0.05). Microcapsules that were washed with acetone following manufacture
demonstrated significantly increased protein retention during acid incubation (P,0.05). In microcapsules that had been
acetone-dried to a point whereby their mass was reduced to 10% of that immediately following encapsulation, protein
retention was over 80% following 24-h acid incubation vs. only 20% protein retention from non acetone-dried
microcapsules. The presence of calcium in the neutral buffer medium significantly reduced BSA diffusion in a concentration-
dependent manner (P,0.05).  2001 Published by Elsevier Science B.V.

Keywords: Alginate; Chitosan; Ionotropic gelation; Bovine serum albumin

1. Introduction

The process of immobilization and microencapsu-

*Corresponding author. Tel.: 11-418-656-2131.
lation facilitates a wide range of biological processes
¨
E-mail address: joel.de-la-noue@san.ulaval.ca (J. de la Noue).
1
Present address: Brandon Research Centre, Agriculture and for both fundamental research and industrial applica-
Agri-Food Canada, P.O. Box 1000A, Brandon, Manitoba, Canada, tions. Due to their characteristics related to biocom-
R7A 5Y3. patibility and biodegradability, a number of natural

0168-3659 / 01 / $ – see front matter  2001 Published by Elsevier Science B.V.

PII: S0168-3659( 01 )00517-X
298 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307
polymers have been identified as components for gel
entrapment and encapsulation strategies. Alginates
are a group of natural polysaccharides belonging to a
family of binary copolymers arranged linearly in
blocks of alternating mannuronic and guluronic sugar
residues [1]. These polymers have found utility as
immobilization matrices for bioreactor systems and
have been widely studied for biomedical applications
due to their ability to form gels under relatively mild
conditions, thus allowing retained biological activity
of the encapsulated material [2]. Encapsulation using
alginates is most often carried out by the dispersion
of the alginate / encapsulant solution onto a gelation
medium of calcium chloride. Contact between the
alginate and the calcium in solution induces immedi-
ate interfacial ionic polymerization of the alginate
via binding of calcium within the cavities of the
guluronic residues, thus forming a polyanionic mi-
crocapsule [3]. Addition of a polycation (poly-L-
lysine or chitosan) to the gelation medium induces
formation of polyanionic–polycationic complexes,
which stabilize the ionic gel network and reduce
alginate permeability [4].
More recently, membrane-coated alginate mi-
crocapsules have been suggested as candidates to
orally deliver a wide range of compounds including
drugs [5], vaccines [6,7], DNA [8], and a host of
therapeutic proteins and peptides [9] past the
stomach to intestinal sites of absorption. For many
compounds of interest, an ideal oral delivery system
would be characterized as having a high efficiency of
encapsulation, maximal stability in acidic pH and
provide rapid release at neutral pH. Several studies
have attempted to identify critical formulation pa-
rameters in order to optimize alginate–chitosan
microcapsules for oral delivery. However, several of
these studies evaluated protein diffusion in water,
saline or a variety of buffers [10], rather than in
physiologically-pertinent media. Furthermore, the
influence of formulation parameters on losses during
manufacture was not always monitored [4,11]. In
order to evaluate the suitability of coated alginate
microcapsules for oral delivery, knowledge of en-
capsulant dynamics is required under relevant phys-
iological conditions that represent the pH conditions
during the different phases of digestion.
The purpose of the current study was to examine
the influence of a number of physico–chemical
factors on the retention of a model protein, bovine
serum albumin (BSA), during microcapsule manu-
facture. Protein loss from alginate microcapsules was
subsequently determined during incubation in acidic
and neutral pH media, simulating the pH encoun-
tered during gastric and intestinal phases of diges-
tion, respectively. This was done in order to optimize
an encapsulation system as an oral delivery system
for proteins to rainbow trout.
2. Materials and methods
Commercial samples of sodium alginate were
purchased from BDH Co. (Montreal, QC, Canada).
This alginate is isolated from the stips of Laminaria
hyperborea, and contains a high level of guluronic
acid residues. Casein, BSA, a-lactalbumin and N-
acetyl-glucosamine were purchased from Sigma
Chemical Co (St. Louis, MO, USA). Chitosan,
having a degree of deacetylation of 86% and low
viscosity, was obtained from Pronova Biopolymers
(Washington, OR, USA). Samples of chitosan used
in experiments employing different molecular mass,
the properties of which are given in Table 1, was
obtained from Chitogenics Inc. (Dartmouth, NS,
Canada).
2.1. Chitosan characterization
The degree of deacetylation of chitosan was
determined by UV absorbance using N-acetyl-glucos-
amine as a standard according to Ref. [12]. Briefly,
samples of chitosan were mixed with 0.01% (v:v)
acetic acid to give a final chitosan concentration of
Table 1
Properties of low (LMM), medium (MMM), and high (HMM)
molecular mass chitosan and subsequent protein retention in
chitosan-coated alginate microcapsules during a 24-h acid incuba-
tion
Chitosan Deacetylation Molecular mass Protein retention
(%) (kDa) (%)
LMM 82 3.98310 6 46.864.9 A
MMM 73 6.61310 6 72.267.1 B
HMM 72 9.44310 6 58.166.2 A
For each item, within needle size, different superscripts A – C are
significantly different (P,0.05).
 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307
0.01% (w:v). Samples were read at 200, 201, 202,
203 and 204 nm wavelengths and the N-acetyl-
glucosamine concentration calculated using a least-
square regression equation derived from the standard
curve employing concentrations of N-acetyl-glucos-
amine ranging from 0 to 32 mg ml 21 dissolved in
0.01% (v:v) acetic acid. The intrinsic viscosity of
chitosan was measured according to standard meth-
ods [13]. Using a capillary viscometer, the residence
time of serially-diluted chitosan solutions (0.2–
0.025% (w:v)) dissolved in 0.1 M acetic acid / 0.02 M
NaCl solution was measured vs. a standard acetic
acid solution. The average molecular mass (Mw ) was
calculated using the Mark Houwink equation:
[h ] 5 K(Mw )a
where [h ] is the intrinsic viscosity in ml g 21 , K5
3.04310 25 and a51.26.
2.2. Alginate microcapsule formation
For the basal encapsulation protocol, 2% (w:v)
alginate was dissolved in water and BSA added at
25% loading rate (mass BSA: mass alginate).
Chitosan was ground finely and dissolved in 2%
(w:v) acetic acid for 4 h with gentle warming and
filtered to remove any undissolved particles. The pH
was adjusted to 5.5 with 4 M NaOH and CaCl 2 was
added to a final concentration of 1.5% (w:v). Ap-
proximately 10 g of the alginate / BSA solution was
pipetted into a syringe fitted with a 21 ga. needle and
a coaxial air jet. Alginate / BSA was extruded drop-
wise into 50 ml chitosan / calcium chloride solution
and allowed to react for 10 min, during which time
samples of the gelation medium were taken at 0, 2,
4, 6, 8 and 10 min following extrusion. Microcap-
sules were removed from the encapsulation medium
via filtration under gentle vacuum and rinsed twice
with 10 ml dilute 0.1 N HCl; a sample of the acid
rinse filtrate was taken to quantify protein loss at this
step. A sub-sample of 10–15 microcapsules was
taken to measure individual microcapsule diameter.
Microcapsules were visualized under 43 magnifica-
tion and the diameter noted using a microscope
equipped with a calibrated side-mounted tracing
device (Leitz LaborLux Model K, Toronto, ON,
Canada), permitting an accuracy of 650 mm. All
299
microcapsules were then washed twice with 5 ml
acetone under gentle vacuum, weighed, placed in an
oven overnight to dry (308C), and the final dry mass
carefully recorded. Preliminary experiments showed
that there was negligible protein loss during the
acetone washes (data not shown). Starting the above
encapsulation protocol, individual parameters were
evaluated separately, allowing a systematic evalua-
tion of factors affecting BSA retention
The factors tested included alginate concentration
(1.0, 1.5, 2.0, 2.5, 3.0% (w:v)), chitosan concen-
tration (0.0, 0.125, 0.25, 0.375, 0.5, 0.75% (w:v)),
CaCl 2 concentration in the gelation medium (0.05,
0.1, 0.5, 1.0, 1.5, 5.0% (w:v)), and loading rate (mass
protein: mass alginate (w:w); 25, 50, 75, 100%),
chitosan molecular mass (high, medium and low
molecular mass) and pH of the gelation medium (3.0,
4.0, 5.0, 6.0). Microcapsule size was altered by
changing needle size (14, 21 and 26 ga.) and / or
coaxial air volume (0, 20 or 40 l min 21 ). Gelation
time was altered in order to determine its effects on
protein retention. To determine the influence of
drying on protein retention, the volume of acetone
and contact time between microcapsules and acetone
was varied.
2.3. Microcapsule incubation and protein
determination
Approximately 25 mg dried microcapsules were
accurately weighed and incubated in 4.0 ml HCl (0.1
N; pH 1.5) for 24 h in a rotating agitator with
sampling at 0, 0.5, 1, 2, 4, 8, and 24 h. Microcap-
sules were subsequently transferred and incubated in
4.0 ml Tris (200 mM; pH 7.5) for an additional 24 h
in a rotating agitator with sampling at 0, 5, 10, 15,
30, 60, 120 min and at 24 h. The influence of
calcium in the neutral buffer solution was investi-
gated by the addition of 10 or 100 mM calcium.
Samples were analyzed for protein using a modified
Coomassie blue protein assay (Pierce Inc., New
York, NY, USA) in 96-well ELISA plates.
2.4. Statistical analyses
All presented results are mean values from four
replicate microcapsule preparations. All data were
tested for homogeneity of variance using standard
300 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307

methods [14]. The effects of needle size and coaxial

air volume on microcapsule diameter and protein
loss were analysed in a 333 factorial design. For the
microcapsule incubation experiments, the percentage
protein retention for each step (manufacture, acid
incubation, neutral buffer incubation) represents the
aggregate loss for each step (manufacture, acid and
neutral buffer incubation), rather than the cumulative
loss for all steps. Data were subjected to analysis of
variance; means were compared using the Tukeys
post-hoc test using the statistical program Statistica
6.0 (StatSoft Inc., Tulsa, OK, USA). Differences
between means were considered significant at P,
0.05.

3. Results

Using the basal encapsulation conditions, the

typical release pattern of BSA from alginate–
chitosan microcapsules during manufacture and incu-
bation in acid and neutral pH media are given in Fig.
1. Under these conditions, BSA was encapsulated
with high efficiency, losing only 5% during the
gelation and washing processes (Panel A). Over the
24 h acid incubation, approximately 35% of the BSA
diffused out of the microcapsules (Panel B), with
90% of the remaining BSA being released within
120 min following transfer into a neutral buffer
(Panel C). Fig. 1. Typical release pattern of BSA from chitosan-coated
Many parameters influenced protein retention dur- alginate microcapsules as a function of time during manufacture
ing microcapsule manufacture and acid incubation. (Panel A), incubation in an acid medium (0.1 N HCl, pH 1.5;
With the exception of the presence of calcium in the Panel B) and incubation in a neutral pH medium (200 mM
Tris–HCl, pH 7.5; Panel C). Error bars represent the standard
neutral buffer (Fig. 8), the tested physico–chemical
error from four replicate microcapsule samples.
factors were without effect on the release of protein
from microcapsules during neutral buffer incubation.
In general, over 90% of the remaining protein altering microcapsule diameter on protein retention
diffused into the medium following microcapsule during acid or neutral buffer incubation (data not
transfer into the neutral buffer. shown).
Final microcapsule diameter was controlled by Fig. 2 demonstrates the effects of increasing
altering the needle diameter and coaxial air volume alginate concentration on protein retention during
used during the extrusion process (Table 2). In manufacture and acid incubation. Protein retention
general, increasing air volume had a more marked during manufacture increases significantly with algi-
effect on final microcapsule diameter than decreasing nate concentration and was maximized at 2% algi-
needle diameter. Decreasing microcapsule diameter nate (w:v) whereas protein retention during acid
by increasing coaxial air volume to 40 l min 21 incubation was maximized at 2.5% alginate (w:v).
significantly increased BSA loss during microcapsule The addition of chitosan to the encapsulation
manufacture (Table 2). There was no influence of medium has a considerable effect on the retention of
 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307 301

Table 2
Influence of altering needle size and coaxial air volume on
microcapsule diameter and protein loss during manufacture pro-
cesses
Needle size a Air volume Bead Dia.b Protein loss
(gauge) (l min 21 ) (mm) (%)
14 0 5430631 A 1.960.5 A
14 20 3600630 B 3.160.7 A
14 40 1730643 C 17.062.3 B
21 0 4102646 A 5.761.3 A
21 20 2475645 B 4.461.1 A
21 40 1250635 C 21.763.5 B
26 0 3072631 A 1.460.5 A
26 20 1450643 B 3.860.7 A
26 40 1080634 C 23.062.2 B Fig. 3. Protein retention during manufacture and 24-h acid
A–C
For each item, within needle size, different superscripts are incubation of chitosan-coated alginate microcapsules produced by
significantly different (P,0.05). varying the chitosan concentration. Error bars represent the
a
Internal diameter of 14, 21 and 26 gauge needles is 160, 50 standard error of the mean based on four replicate microcapsule
and 25 mm, respectively. samples.
b
Mean6S.E.M.; n54.
microcapsule aggregation. The influence of increas-
protein both during microcapsule manufacture and ing chitosan concentration on protein retention dur-
during acid incubation. The addition of lowest ing acid incubation reveals a different pattern of
concentration of chitosan (0.125%; w:v) significantly response (Fig. 3). Protein retention is maximized
increases protein retention during manufacture from with 0.25% (w:v) chitosan in the encapsulation
25% without chitosan to 90% (Fig. 3). Encapsulation medium, above which retention decreases signifi-
efficiency continues to increase with increasing cantly. The retention of BSA during manufacture and
chitosan concentration, reaching a maximum at acid incubation in response to increasing calcium
0.75% (w:v). Increasing the chitosan concentration in concentration in the gelation medium showed similar
the encapsulation medium above 0.75% (w:v) made tendencies (Fig. 4). In both cases, protein retention
encapsulation extremely difficult, as the viscous was maximal at lower calcium levels, and was
nature of the gelation medium induced excessive significantly lower above 0.5% CaCl 2 (w:v).

Fig. 2. Protein retention during manufacture and 24-h acid
incubation of chitosan-coated alginate microcapsules produced by
varying the alginate concentration. Error bars represent the
standard error of the mean based on four replicate microcapsule
samples.
Fig. 4. Protein retention during manufacture and 24-h acid
incubation of chitosan-coated alginate microcapsules produced by
varying the calcium chloride concentration. Error bars represent
the standard error of the mean based on four replicate mi-
crocapsule samples.
302 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307

Fig. 6. Protein (BSA) retention during manufacture and 24-h acid

Fig. 5. Protein retention during manufacture and 24-h acid
incubation of chitosan-coated alginate microcapsules produced by
incubation of chitosan-coated alginate microcapsules produced by
varying the encapsulation media pH. Error bars represent the
varying the percentage protein loading (mass BSA:mass alginate).
standard error of the mean based on four replicate microcapsule
Error bars represent the standard error of the mean based on four
samples.
replicate microcapsule samples.

protein retention during neutral buffer incubation as

Fig. 5 demonstrates the efficiency with which a result of microcapsule drying (data not shown).
alginate–chitosan microcapsules are able to encapsu- The addition of CaCl 2 to the neutral buffer
late relatively large quantities of protein. There was incubation medium significantly reduced the release
no decrease in protein retention during microcapsule of protein from alginate–chitosan microcapsules in a
manufacture when protein loading was increased dose-dependent manner (Fig. 8). The percentage of
from 25 to 100% (w:w). However, increased protein BSA released from microcapsules into the neutral
loading resulted in significantly reduced protein buffer containing either 10 or 100 mM calcium was
retention during acid incubation. reduced to 50 and 25%, respectively, of that of
Fig. 6 demonstrates the effect of gelation medium microcapsules in calcium-free buffer.
pH on protein retention. While increasing pH was
without effect on protein retention during manufac-
ture, retention during acid incubation was signifi-
cantly increased when pH was increased from 3.0 to
6.0. Qualitative observations during manufacture also
revealed that when a low pH gelation medium was
employed (i.e. ,4.0), microcapsules tended to ag-
gregate, and were opaque rather than translucent, as
was observed at higher pH values.
Fig. 7 demonstrates the substantial influence of
microcapsule drying (as a result of rinsing with
acetone) on the retention of BSA during acid incuba-
tion. Microcapsules that were not dried with acetone
retained only 23% of the BSA during the 24-h acid Fig. 7. Protein (BSA) retention during 24-h acid incubation of
chitosan-coated alginate microcapsules at differing degrees of
incubation, whereas those rinsed for an extended
dryness resulting from changing time of exposure to acetone.
period of time, thus reducing their mass to 10% of Individual points represent the mean of duplicate microcapsule
that immediately following gelation, retained almost samples. Regression equation is: y5 226.614 ln(x)116.865, r 2 5
80% of their protein. There was no alteration of 0.93.
 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307
Fig. 8. Release pattern of BSA from chitosan-coated alginate
microcapsules as a function of time during incubation in a neutral
pH medium (200 mM Tris–HCl, pH 7.5) containing increasing
concentration of calcium. Individual points represent the mean
values from four replicate microcapsule samples. The pooled
standard error for 0, 10 an 100 mM Ca was 8.8, 4.9 and 2.7,
respectively. The regression equations are 0 mM Ca: y5
0.7805x 3 212.932x 2 174.859x256.413, r 2 50.96; 10 mM Ca: y 5
0.7789x 2 1 2.5093x 1 1.462, r 2 50.99; 100 mM Ca: y 5
1.1005x 2 2 4.4677x 1 5.6508; r 2 50.98.
4. Discussion
Ideally, oral delivery systems designed to transport
a compound of interest past the stomach would be
characterized as having a high encapsulation ef-
ficiency, provide maximal stability and therefore
limited release in acidic pH ranges and rapid release
in neutral pH conditions. In vitro evaluation of
protein release reported herein suggests that under
optimal encapsulation conditions, chitosan-coated
alginate microcapsules fulfill these criteria; BSA is
encapsulated with an efficiency of .95%, of which
80% is retained within the microcapsules during a
24-h incubation in 0.1 N HCl. Following transfer
into a neutral buffer, over 90% of the remaining
protein is released within 120 min. The present study
identifies a number of factors that influence the
efficiency with which BSA is incorporated into
chitosan-coated alginate microcapsules, as well as its
retention under conditions simulating the pH en-
countered during gastric and intestinal phases of
digestion.
303
The extrusion of alginate into a bath containing a
multivalent ion (referred to as external gelation)
results in the production of inhomogeneous mi-
crocapsules [15]. Microcapsules formed using this
method have a high polymer gradient near the
microcapsule surface, which decreases as the core is
approached. This gradient is formed during gelation
as a result of an outward diffusion of alginate toward
an inward migrating gelling zone. Alginate con-
centrations reaching as high as 10% have been
reported near the surface of externally-gelled mi-
crocapsules produced under certain conditions [15],
thus providing a relatively non-permeable outer shell
around externally-gelled microcapsules. The polymer
gradient is governed by the rate of diffusion between
the soluble alginate molecules and the calcium ions
[16] and may be affected by a number of factors.
In the present study, the concentration of alginate
significantly influenced the retention of protein dur-
ing microcapsule manufacture and acid incubation
with protein retention being maximized at alginate
concentrations .2.5% (w:v). This is in agreement
with a number of previous studies showing a direct
relationship between alginate concentration and re-
tention in alginate microcapsules [17–21]. Martinsen
et al. [18] suggested that the effect of increasing
alginate concentration amplifies the alginate gradient
that is established in externally-gelled microcapsules
[2,15], thus providing a more compact microcapsule
cortex and decreasing permeability. Factors indepen-
dent of this phenomenon seem to be involved as
well, as increasing alginate concentration also de-
creases diffusion [18] and increases encapsulant
retention [22] in microcapsules produced using an
internal source of calcium, which are more homoge-
neous than externally-gelled microcapsules [23].
The concentration of calcium in the encapsulation
medium exerts a strong influence on the retention of
BSA during both microcapsule manufacture and
subsequent acid incubation (Fig. 4), with BSA loss
increasing with calcium concentration. As described
above, the production of externally-gelled alginate
microcapsules results in microcapsules with a rela-
tively thick outer cortex due to an outward-diffusing
gelling zone. The degree of alginate inhomogeneity
is related to a number of factors including the
calcium concentration in the gelation medium; the
increase in permeability revealed in the present study
304 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307
˚
correspond well to the observations by Skjak-Bræk
et al. [15], who reported that the degree of alginate
homogeneity increased significantly as the external
calcium chloride concentration is reduced from 0.34
to 0.01 M. The degree of alginate inhomogeneity is
governed by the rate of outward polymer diffusion,
as well as the rate of inward calcium ion diffusion
toward the gelling front [2]. Thus, as external
calcium concentration is decreased, the resulting
decreased rate of calcium diffusion towards the
gelling front promotes a higher alginate gradient and
reduced microcapsule permeability.
Previous work has revealed that the physical and
functional attributes of alginate microspheres can be
manipulated by the addition of polycationic polymers
to the gelation medium [4,24–26]. Positively charged
amino groups of poly-L-lysine and chitosan form
membranes through ionic interactions with carboxy-
lic residues of the alginate [27]; the addition of
polycationic polymers to the gelation medium results
in reduced microcapsule swelling [4], improved
stability [8,26] and reduced microcapsule permeabili-
ty [28]. Similar to the current study, the addition of
chitosan to the encapsulation medium significantly
improved the encapsulation efficiency of dextran
[19] and BSA [9]. The coating of alginate mi-
crocapsules with either 0.4% (w:v) chitosan or 0.05%
(w:v) poly-L-lysine decreased their permeability as
evidenced by increased retention of DNA residuals
following DNAase exposure [8]. The response of
protein retention to increasing chitosan concentration
(Fig. 3) during acid incubation has not been previ-
ously reported. Incubating alginate–chitosan mi-
crocapsules in a phosphate buffer, Sezer and Akbuga
[19] reported dissimilar results; the release of dextran
from alginate-chitosan microcapsules decreased with
increasing chitosan concentration between 0.25 and
0.4%, although interpretation of the data is difficult
as it is unclear whether the dextran release represents
simply an aggregate loss or incorporates manufac-
ture-related losses. Improvement of mechanical
characteristics in alginate-coated chitosan microcap-
sules has been reported when the alginate concen-
tration was reduced from 2.0 to 0.75% [29], with
reduced capsule strength noted as alginate was
further reduced below 0.5%. Polk et al. [4] reported
an increased diffusion of BSA in a neutral pH
medium from alginate microcapsules extruded onto
0.2% chitosan versus those produced using 0.1%
chitosan. These authors suggested that the lower
concentration of chitosan provided a less viscous
medium, thus permitting an improved capacity for
interaction with the alginate microcapsule, resulting
in a thicker, less permeable membrane.
Cation attributes such as polymer structure (mo-
lecular mass, chemical composition) and process
factors (reaction time, pH, additives) have been
studied in attempts to control permeability from
cation-coated alginate microcapsules [30]. The in-
fluence of polycation molecular mass on alginate–
chitosan microcapsule permeability is not clear.
Okhamafe and Goosen [30] suggested that, theoret-
ically, membrane permeability of coacervate mi-
crocapsules would be lower with increasing molecu-
lar mass due to increased chain packing and rigidity,
as well as increased inter-chain bonding. Quong and
Neufeld [8] reported that DNA encapsulated in
alginate microcapsules coated with high molecular
mass poly-L-lysine showed a lower degree of hy-
drolysis, suggesting reduced membrane permeability
and diffusion of extracapsular DNAse. The same
study reported decreased polycation uptake and
membrane thickness with increasing molecular mass
for both chitosan and poly-L-lysine, suggesting that
membrane permeability is independent of membrane
thickness. Similar relationships between polycation
molecular mass and microcapsule permeability have
been observed in other studies reporting the encapsu-
lation of hemoglobin [31], BSA [4,7] and lipophilic
drugs [22] in chitosan coated alginate microcapsules.
However, several studies have also reported data to
the contrary, with membrane permeability being
related directly to polycation molecular mass. Using
poly-L-lysine for membrane formation, increased
diffusion of BSA into the microcapsules with in-
creasing cation molecular mass has been reported
[24,28]. Other reports [32,33], reveal no influence of
polycationic molecular mass on membrane per-
meability. Interactions between additional factors
including the type of encapsulant, microcapsule type
(liquid vs. solid core), the degree of polycationic
polymer deacetylation and the type of study (i.e.
diffusion in vs. diffusion out of the microcapsule)
may explain the differing observations between
studies.
Although the encapsulation pH had no effect on
encapsulation efficiency in the current study, it did
significantly affect protein retention during the 24-h
 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307
acid incubation, with increasing BSA retention as pH
of the gelation medium is increased from 3 to 5 (Fig.
6). Differing results have been reported for both
BSA and ovalbumin [34] and for hemoglobin [31],
despite the fact that Polk et al. [7] reported that
capsule strength and flexibility significantly de-
creased with decreasing pH. In these reports, the
decreased permeability of alginate–chitosan capsules
in low-pH encapsulation medium has been attributed
to a number of factors, including the alteration of
encapsulant solubility and membrane thickness / den-
sity at different pH values. Huguet et al. [31]
suggested that the decreased permeability of algi-
nate–chitosan microcapsules at lower pH values was
due to thicker, stronger membranes. In contrast,
membrane thickness has been reported to increase as
encapsulation medium pH was increased from 6.0 to
7.0 [32]. When 0.9% NaCl (w:v) was added to the
encapsulation medium the opposite was observed,
with decreasing membrane thickness related to in-
creasing pH. However, no relationship was found
between membrane thickness and permeability of
chitosan-coated alginate microcapsules produced in
media varying in pH [32]. Obviously, there may be
other unrecognized factors interacting with encapsu-
lation pH that play important roles in regulating
microcapsule permeability.
Following microcapsule gelation, exposure to
acetone significantly increased protein retention dur-
ing acid incubation (Fig. 7). Quong et al. [8]
reported that lyophilization of alginate microcapsules
containing DNA resulted in reduced DNA hydrolysis
following incubation with extracapsular DNAse,
indicating reduced microcapsule permeability. It has
been suggested that the partial drying of alginate
microcapsules effectively increases alginate concen-
tration. Assuming a uniform alginate distribution
throughout the microcapsule, the dimensions of the
pores within the alginate network is estimated to be
proportional to the inverse cube root of the alginate
concentration [26]. Thus, as water is removed from
the gel network, alginate concentration increases and
average pore size is reduced. This phenomenon may
be more applicable to microcapsules produced using
alginate rich in guluronic acid, as they will reswell
only slightly upon rehydration [2]. The observed
decrease in protein diffusion during acid incubation
following exposure to acetone observed in the pres-
ent study may also be due to acetone-induced protein
305
precipitation. Both acetone and lyophilization are
used in a number of histological applications in order
to stabilize and localize otherwise water-soluble,
protein-based cellular components [35]. However,
the ability of proteins to rapidly diffuse from
acetone-treated microcapsules into the neutral pH
incubation medium indicates that this is unlikely.
One of the recognized drawbacks of utilizing
alginate in any encapsulation strategy is its sensitivi-
ty to a number of ions. Chelating and anti-gelling
compounds are known to destabilize Ca-alginate
gels, whereas the addition of free Ca to the storage
medium increases gel stability [2]. Thu et al. [27]
observed that the addition of 5 mM calcium to a
physiological (0.9% NaCl (w:v)) incubation medium
containing alginate microcapsules significantly re-
duces microcapsule swelling. Similarly, Quong and
Neufeld [8] observed limited bead swelling of
chitosan and poly-L-lysine-coated alginate microcap-
sules incubated in a neutral pH Tris–HCl buffer in
the presence of 10 mM Ca addition. The same study
showed that microcapsules co-incubated with 10 mM
Mg nearly doubled in size during the 2 h incubation
period, indicating capsule instability. We report that
the diffusion of BSA from of alginate–chitosan
microcapsules during neutral buffer incubation was
significantly reduced with increasing calcium con-
centration in the incubation medium (Fig. 8). Addi-
tion of 10 and 100 mM Ca to the incubation buffer
reduced BSA diffusion by 50 and 75%, respectively.
This presents a number of problems related to the
potential application of alginate–chitosan microcap-
sules for oral delivery of bioactive proteins. For
example, domestic monogastric animal diets are
supplemented with calcium of various forms The
presence of calcium in the digesta would signifi-
cantly reduce protein diffusion out of alginate–
chitosan microcapsules, thus limiting the application
of this approach in these species. In vivo studies are
required in order to confirm the utility of alginate–
chitosan microcapsules to provide oral routes of
bioactive protein delivery to domestic animals.
5. Conclusions
Monitoring the retention of BSA in chitosan-
coated alginate microcapsules during manufacture
and gastric / intestinal simulation permitted the identi-
306 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307
fication of a number of physico–chemical factors
that influence protein retention at each step. From
these results, optimal formulation conditions can be
determined to permit the efficient encapsulation of
protein with minimal losses during gastric phases of
digestion, and rapid protein release once in the small
intestine.
Acknowledgements
This work was supported by the Natural Sciences
and Engineering Research Council (NSERC
Strategic Projects Program) and BASF Canada Inc.
GWV is the recipient of an NSERC Postgraduate
Scholarship. The authors wish to thank Drs R.
Neufeld and D. Quong for their constructive com-
ments related to the manuscript.
References
[1] E. Murano, Use of natural polysaccharides in the microen-
capsulation technique, J. Appl. Ichthyol. 14 (3–4) (1998)
245–249.
˚
[2] O. Smidsrød, G. Skjak-Bræk, Alginate as immobilization
matrix for cells, Trends Biotechnol. 8 (3) (1990) 71–78.
˚
[3] A. Martinsen, G. Skjak-Bræk, O. Smidsrød, Alginate as
immobilization material: I Correlation between chemical and
physical properties of alginate gel beads, Biotechnol. Bioeng.
33 (1) (1989) 79–89.
[4] A. Polk, B. Amsden, K. De Yao, T. Peng, M.F. Goosen,
Controlled release of albumin from chitosan–alginate mi-
crocapsules, J. Pharm. Sci. 83 (2) (1994) 178–185.
[5] S. Takka, O.H. Ocak, F. Acarturk, ¨ Formulation and in-
vestigation of nicardipine HCl-alginate gel beads with fac-
torial design-based studies, Eu. J. Pharmaceut. Sci. 6 (3)
(1998) 241–246.
[6] T.L. Bowersock, H. HogenEsch, M. Suckow, P. Guimond, S.
Martin, D. Borie, S. Torregrosa, H. Park, K. Park, Oral
vaccination of animals with antigens encapsulated in alginate
microspheres, Vaccine 17 (13–14) (1999) 1804–1811.
[7] A.E. Polk, B. Amsden, D.J. Scarratt, A. Gonzal, A.O.
Okhamafe, M.F.A. Goosen, Oral delivery in aquaculture:
controlled release of proteins from chitosan–alginate mi-
crocapsules, Aquacult. Eng. 13 (4) (1994) 311–323.
[8] D. Quong, R.J. Neufeld, DNA protection from extracapsular
nucleases, within chitosan- or poly-L-lysine-coated alginate
beads, Biotechnol. Bioeng. 60 (1) (1998) 124–134.
[9] P.R. Hari, T. Chandy, C.P. Sharma, Chitosan / calcium algi-
nate microcapsules for intestinal delivery of nitrofurantoin, J.
Microencapsul. 13 (3) (1996) 319–329.
[10] T. Ostberg, L. Vesterhus, C. Graffner, Calcium alginate
matrices for oral multiple unit administration: II Effect of
process and formulation factors on matrix properties, Int. J.
Pharm. 97 (1–3) (1993) 183–193.
[11] ¨
F. Acarturk, S. Takka, Calcium alginate microparticles for
oral administration: II Effect of formulation factors on drug
release and drug entrapment efficiency, J. Microencapsul. 16
(3) (1999) 291–301.
[12] R.A.A. Muzzarelli, R. Rocchetti, The determination of the
degree of deacetylation of chitosans by spectrophotometry,
in: R.A.A. Muzzarelli, C. Jeuniaux, G.W. Gooday (Eds.),
Chitin in Nature and Technology, Plenum Press, New York,
1986, pp. 385–388.
[13] G.A. Roberts, J.G. Domszay, Determination of the vis-
cometric constants for chitosan, Int. J. Biol. Macromol. 4
(1982) 374–377.
[14] G.W. Snedecor, W.G. Cochran, Statistical Methods, 7th
Edition, The Iowa State University Press, Ames, IA, 1980.
[15] ˚
G. Skjak-Bræk, H. Grasdalen, O. Smidsrød, Inhomogeneous
polysaccharide ionic gels, Carbohyd. Polym. 10 (1989) 31–
54.
[16] A. Mikkelsen, A. Elgsaeter, Density distribution of calcium-
induced alginate gels. A numerical study, Biopolymers 36
(1995) 17–41.
[17] A. Bartkowiak, D. Hunkeler, Alginate-oligochitosan mi-
crocapsules. II. Control of mechanical resistance and per-
meability of the membrane, Chem. Mater. 12 (1) (2000)
206–212.
[18] ˚
A. Martinsen, I. Storro, G. Skjak-Bræk, Alginate as im-
mobilization material: III. Diffusional properties, Biotechnol.
Bioeng. 39 (2) (1992) 186–194.
[19] A.D. Sezer, J. Akbuga, Release characteristics of chitosan
treated alginate beads: II. Sustained release of a low molecu-
lar drug from chitosan treated alginate beads, J. Microencap-
sul. 16 (6) (1999) 687–696.
[20] H. Tanaka, M. Matsumura, I.A. Veliky, Diffusion characteris-
tics of substrates in Ca-alginate gel beads, Biotechnol.
Bioeng. 26 (1984) 53–58.
[21] K. Yamagiwa, Y. Shimizu, T. Kozawa, M. Onodera, A.
Ohkawa, Ethanol production by encapsulated and immobil-
ized yeast, Biotechnol. Tech. 8 (4) (1994) 271–274.
[22] A.J. Ribeiro, R.J. Neufeld, P. Arnaud, J.C. Chaumeil, Mi-
croencapsulation of lipophilic drugs in chitosan-coated algi-
nate microspheres, Int. J. Pharm. 187 (1) (1999) 115–123.
[23] D. Quong, R.J. Neufeld, G. Skjak-Bræk, ˚ D. Poncelet,
External versus internal source of calcium during the gela-
tion of alginate beads for DNA encapsulation, Biotechnol.
Bioeng. 57 (4) (1998) 438–446.
[24] M.F. Goosen, G.M. O’Shea, H.M. Gharapetian, S. Chou,
A.M. Sun, Optimization of microencapsulation parameters:
Semipermeable microcapsules as a bioartificial pancreas,
Biotechnol. Bioeng. 27 (2) (1985) 146–150.
[25] D. Quong, J.N. Yeo, R.J. Neufeld, Stability of chitosan and
poly-L-lysine membranes coating DNA-alginate beads when
exposed to hydrolytic enzymes, J. Microencapsul. 16 (1)
(1999) 73–82.
[26] B. Thu, P. Bruheim, T. Espevik, O. Smidsrød, P. Soon-
˚
Shiong, G. Skjak-Bræk, Alginate polycation microcapsules.
 G.W. Vandenberg et al. / Journal of Controlled Release 77 (2001) 297 – 307
II. Some functional properties, Biomaterials 17 (11) (1996)
1069–1079.
[27] B. Thu, P. Bruheim, T. Espevik, O. Smidsrød, P. Soon-
˚
Shiong, G. Skjak-Bræk, Alginate polycation microcapsules.
I. Interaction between alginate and polycation, Biomaterials
17 (10) (1996) 1031–1040.
[28] G.A. King, A.J. Daugulis, P. Faulkner, M.F.A. Goosen,
Alginate–polylysine microcapsules of controlled membrane
molecular weight cutoff for mammalian cell culture en-
gineering, Biotechnol. Prog. 3 (4) (1987) 231–240.
[29] M.M. Daly, D. Knorr, Chitosan–alginate complex coacervate
capsules: Effects of calcium chloride, plasticizers, and
polyelectrolytes on mechanical stability, Biotechnol. Prog. 4
(2) (1988) 76–81.
[30] A.O. Okhamafe, M.F.A. Goosen, Control of membrane
permeability in microcapsules, in: M.A.F. Goosen (Ed.),
Fundamentals of Animal Cell Encapsulation and Immobiliza-
tion, CRC Press, Boca Raton, FL, 1993, pp. 55–78.
307
[31] M.L. Huguet, A. Groboillot, R.J. Neufeld, D. Poncelet, E.
Dellacherie, Hemoglobin encapsulation in chitosan / calcium
alginate beads, J. Appl. Polym. Sci. 51 (1994) 1427–1432.
[32] A. Bartkowiak, D. Hunkeler, Alginate–oligochitosan mi-
crocapsules: a mechanistic study relating membrane and
capsule properties to reaction conditions, Chem. Mater. 11
(9) (1999) 2486–2492.
[33] C.A. McKnight, A. Ku, M.A.F. Goosen, D. Sun, C. Penney,
Synthesis of chitosan–alginate microcapsule membranes, J.
Bioact. Compat. Polym. 3 (1988) 334–355.
[34] S.K. Kim, C. Rha, Transmembrane permeation of proteins in
˚
chitosan capsules, in: G. Skjak-Bræk, T. Anthonsen, P.A.
Sandford (Eds.), Chitin and Chitosan: Sources, Chemistry,
Biochemistry, Physical Properties, And Applications,
Elsevier Applied Science, London, 1989, pp. 635–642.
[35] G.I. Murray, S.W. Ewen, Enzyme histochemistry on freeze-
substituted glycol methacrylate-embedded tissue, J. Histoch-
em. Cytochem. 38 (1) (1990) 95–101.