Preparation and characterization of biodegradable or enteric-coated microspheres containing the protease inhibitor camostat

Article abstract of DOI:10.1211/0022357011775280

We have produced biodegradable or enteric-coated microspheres containing camostat mesylate, a protease inhibitor, using a water-oil-water emulsion solvent evaporation method. The characteristics of the microspheres were determined. When polylactic acid, a biodegradable polymer, was used as a wall material, the optimized microsphere obtained showed a loading efficiency of almost 95% and had a mean diameter of 30 μm. This microsphere showed a sustained-release profile, with nearly 25% of drug being released at seven days in a dissolution test. When hypromellose acetate succinate (AS-HG type, with a high content of succinyl group) was used as an enteric wall material, optimized microspheres showed a loading efficiency of almost 80%. In this case, pH 3.0 citrate buffer was used as an internal aqueous phase, and citrate buffer containing 0.5% polyvinylalcohol was used as the external aqueous phase. These microspheres showed a rapid release profile in pH 6.8 buffer, whereas the release was extremely slow in pH 1.2 buffer. Hypromellose acetate succinate microspheres were also prepared containing 10% (w/w) N-benzoyl-dl-arg-4-nitroanilide as a model substrate for trypsin, with or without 5% (w/w) camostat. These microspheres were incubated in pH 6 or 7 buffer containing trypsin at 37°C. When camostat was included in the microspheres, the substrate was protected from attack by trypsin, while in the absence of camostat, the released substrate was immediately attacked by trypsin to produce the degradation product N-benzoyl-dl-arginine.

Full text of DOI:10.1211/0022357011775280

Communications
Journal of
Pharmacy and Pharmacology
JPP 2001, 53: 255–261
# 2001 The Authors
Received August 9, 2000
Accepted October 2, 2000
ISSN 0022-3573
Preparation and characterization of biodegradable or
enteric-coated microspheres containing the protease
inhibitor camostat
Takahiro Uchida, Noriko Yasuda and Kenji Matsuyama
Abstract
We have produced biodegradable or enteric-coated microspheres containing camostat
mesylate, a protease inhibitor, using a water–oil–water emulsion solvent evaporation method.
The characteristics of the microspheres were determined. When polylactic acid, a biodegradable
polymer, was used as a wall material, the optimized microsphere obtained showed a loading
efficiency of almost 95% and had a mean diameter of 30 µm. This microsphere showed a
sustained-release profile, with nearly 25% of drug being released at seven days in a dissolution
test. When hypromellose acetate succinate (AS-HG type, with a high content of succinyl group)
was used as an enteric wall material, optimized microspheres showed a loading efficiency of
almost 80%. In this case, pH 3.0 citrate buffer was used as an internal aqueous phase, and
citrate buffer containing 0.5% polyvinylalcohol was used as the external aqueous phase. These
microspheres showed a rapid release profile in pH 6.8 buffer, whereas the release was
extremely slow in pH 1.2 buffer. Hypromellose acetate succinate microspheres were also
prepared containing 10% (w/w) N-benzoyl-dl-arg-4-nitroanilide as a model substrate for
trypsin, with or without 5% (w/w) camostat. These microspheres were incubated in pH 6 or 7
buffer containing trypsin at 37mC. When camostat was included in the microspheres, the
substrate was protected from attack by trypsin, while in the absence of camostat, the released
substrate was immediately attacked by trypsin to produce the degradation product N-benzoyl-
dl-arginine.
Faculty of Pharmaceutical
Sciences, Mukogawa Women’s
University, 11-68, Koshien 9-
Bancho, Nishinomiya City, 663-
8179, Japan
Introduction
Takahiro Uchida, Noriko Yasuda,
Kenji Matsuyama
Camostat mesylate is a protease inhibitor that has been developed as a treatment
for acute pancreatitis (Taniguchi et al 1997) or disseminated intravascular co-
agulation (Naito et al 1989). As the elimination rate of camostat from the body is
high, the development of a sustained-release formulation is desirable. In this study,
two different materials have been used to produce a microencapsulated formulation
of camostat, the biodegradable polymer polylactic acid (PLA) and the enteric
polymer hypromellose acetate succinate. The drug release profiles of the prepared
microspheres were evaluated.
Correspondence: T. Uchida,
School of Pharmaceutical
Sciences, Mukogawa Women’s
University, 11–68 Koshien 9-
Bancho, Nishinomiya City 663-
8179, Japan. E-Mail:
takahiro!mwu.mukogawa-
u.ac.jp
Acknowledgement and
Funding: We thank Miss
Tomoko Ninomiya for her
assistance in preparation and
evaluation of the microspheres.
This study was supported by
Magnetic Health Science
Foundation.
PLA microspheres have been used in the formulation of long-acting, injectable,
drug delivery systems for the past 15 years (Jalil & Nixon 1989; Edman et al 1993;
Larionova et al 1999). A wide range of synthetic drugs, as well as biologicals such
as enzymes, hormones and proteins, have been microencapsulated by this polyester
polymer (Ogawa et al 1988; Kentepozidou et al 1995; Uchida et al 1998).
255

256
Takahiro Uchida et al
Preparation of PLA or hypromellose acetate
succinate microspheres containing camostat
Even though commercial camostat products are now
available in oral dosage forms, the design of appropriate
release rates of camostat and substrate such as peptides
seems to be essential for appropriate gastrointestinal
delivery. Therefore, enteric-coated microspheres con-
taining camostat would be an attractive proposition.
We have used the enteric polymer hypromellose acetate
succinate to encapsulate camostat because the polymer
dissolves easily at pH over 6. In this case, camostat and
N-benzoyl-dl-arg-4-nitroanilide, a substrate of protease,
were encapsulated in the polymer. Model substrate-
loaded microspheres, in the presence or absence of
camostat, were soaked and incubated at pH 6 or 7 in
buffered solution containing trypsin as a protease, and
N-benzoyl-dl-arginine. HPLC was used to periodically
screen for degradation products of the model substrate
for trypsin.
For the PLA microspheres, a water–oil–water (w\o\w)
emulsion solvent evaporation method (Heya et al 1991;
Uchida et al 1995, 1996) was adopted. Firstly, 20 mg
camostat (corresponding to a theoretical loading of
10%) was suspended in purified water. This medium, as
the internal aqueous phase, was emulsified with 5 mL
methylene chloride containing 180 mg polylactic acid
for 1 min using an ultrasonic disrupter (UD-200, Tomy
Seiko Co., Ltd., Tokyo, Japan). This w\o emulsion was
poured into 200 mL 0.5% (w\v) PVA solution (dis-
solved in buffers of various pH) as the external aqueous
phase. Emulsification was continued using a homo-
genizer (Physcotron, Nichion Irikakikai Co., Ltd.,
1
−
Tokyo) at 3000 rev min for 1 min. This dispersion
was gently agitated in a 500-mL beaker on a stirring
plate with a 3.9-cm stirring bar for 4 h at room tem-
perature. The microspheres were collected by centri-
fugation at 3000 rev min 1 for 15 min (1580 g), washed
−
with purified water and freeze-dried (FD-5N, Tokyo
Rikakikai Co., Ltd., Tokyo) for at least 4 h. Changing
the volume of methylene chloride varied polymer con-
centration.
Materials and Methods
Materials
The preparation of hypromellose acetate succinate
enteric microspheres was essentially the same except
that pH 3.0 citrate buffer was used as an internal or
external aqueous phase. The preparation of micro-
spheres containing both camostat and N-benzoyl-dl-
arg-4-nitroanilide as a model substrate of trypsin was
also essentially the same, except that the theoretical
camostat and substrate loading was 5 and 10%, re-
spectively.
Camostat as mesylate was kindly donated by Ono Co.
Ltd. (Osaka, Japan). Polylactic acid (MW 18000) was
purchased from Wako Chemical Co. Ltd. (Osaka,
Japan). Hypromellose acetate succinate, AS-HG grade,
containing 10–14% (w\w) succinyl group, was kindly
donated by Shin-Etsu Chemical Industry Co., Ltd.
(Tokyo, Japan). Polyvinyl alcohol (PVA) (MW 85000–
146000, 87–89% hydrolysed) was supplied by Aldrich
Chemical Co., Ltd. (Milwaukee, WI). Model substrate,
N-benzoyl-dl-arginine, was purchased from Sigma Co.,
Ltd. (St Louis, MO) andtrypsin wassupplied byNakalai
Tesque Co., Ltd. (Kyoto, Japan). Other reagents were
all of special grade.
For all batches, yields were determined as the per-
centage of weight of the recovered microspheres after
drying divided by the initial amount of enteric polymer
and the drug employed.
A morphological study was performed using a bio-
logical microscope (Aliphaphot-2 YS2-H, Nikon Co.,
Ltd., Tokyo). A Microtrac Particle Size Analyzer
(Model 7995-30, Leeds and Northrup, North Wales,
PA) was utilized to determine the microsphere average
volume diameters and distribution.
Solubilities of camostat in buffers of various pH
Buffers of various pH (pH 1.5, 3, 5, 7, 10, 11.5) were
1
used as solvent. Excess camostat (1 g dL ) was put into
−
a 30-mL beaker, buffer was added, and the beaker was
shaken at 80 rev min 1 at 25mC for 10 h. In a pilot study
we had confirmed that equilibrium was reached within
−
Drug loading
10 h. The suspension was filtered using a Millipore filter For the PLA microspheres, 200 µL methylene chloride
(0.45-µm pore size) and the filtered solution was diluted was added to 5 mg microspheres, and 2 mL pH 3.0
with buffer. The camostat concentration of the diluted citrate buffer was added to the solution, which was then
sample was determined using an HPLC method as agitated adequately for 5 min. After centrifugation at
−
described below.
3000 rev min 1 for 15 min (1580 g), the water phase was

257
Microspheres containing camostat
Table 1 The effect of preparative conditions on camostat loading efficiency in polylactic acid microspheres
theoretically loaded with 10% camostat.
Formulation
number
Internal aqueous
phase (w)
Oily
External aqueous
phase (w)
Loading
efficiency
(%)
s.d.
phase (o)
methylene
chloride
(Camostat 20 mg)
PVA concn Buffer
Solvent
Volume
volume (mL) (%)
pH
(µL)
1
2
Water
Water
Water
Water
Water
Water
Water
Water
Water
Water
Water
60
60
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
1.5
8.3
5.9
3.2
3
5
2.4
3.8
6.8
12.4
6.9
4.6
5.3
3.3
4.3
3.9
3
60
15.2
23.0
72.0
67.8
34.2
28.5
75.1
84.5
95.3
4
60
7
5
60
10
11.5
10
10
10
10
10
6
60
7
200
400
600
60
8
9
10
11
60
0.75
Each value represents the mean of four to six experiments.
collected and the methylene chloride phase extracted
For enteric microspheres containing 10% of sub-
with 2 mL pH 3.0 buffer. The water phases were strate, N-benzoyl-dl-arg-4-nitroanilide, and 5% camo-
combined, and a 50-µL sample was injected onto stat, the camostat and substrate were simultaneously
a chromatograph (Shimadzu LC-10A, Kyoto, Japan) extracted by the method described above. The con-
equipped with a UV detector (Shimadzu SPD-10AV), centration of substrate was determined using a reversed-
an integrator (Shimadzu C-R4A) and a reversed phase phase column (Cosmosil 5C18-AR, 4.6i150 mm,
column (Cosmosil 5C18-AR, 4.6i150 mm, Nakalai Nakalai Tesque Co., Ltd., Kyoto) with acetonitrile:
Tesque Co., Ltd, Kyoto, Japan). The following mobile phosphoric acid (10 m) containing 50 m sodium sul-
phase systems were used: A, methanol; B, 0.2% sodium fate (33:67) as the mobile phase. The flow rate was
heptane sulfonate; C, 0.1% sodium lauryl sulfate; D, 1.0 mL min 1 and the wavelength was set at 265 nm.
−
glacial acetic acid (A:B:C:D, 200:100:50:1). The flow
1
rate was 1.0 mL min and the wavelength was set at
−
In-vitro camostat release from PLA or
hypromellose acetate succinate microspheres
265 nm. Loading was calculated from the weight of the
initial microspheres and the amount of camostat in-
corporated.
The in-vitro release profile of camostat from PLA or
For the enteric hypromellose acetate succinate micro- hypromellose acetate succinate microspheres was ex-
spheres, the actual camostat loading percentage was amined as follows. Microspheres corresponding to 1 mg
determined as follows. A quantity of approximately camostat were suspended in 50 mL buffer solution and
1
−
5 mg microspheres was precisely weighed and dissolved shaken horizontally at 60 rev min at 37mC. For PLA
in 2 mL methanol in a glass vial for further dissolution microspheres, 0.02% Tween 80 and sodium azide were
of the enteric polymer. Two millilitres pH 3.0 citrate added to prevent microspheres from coagulating, and
buffer was added to the sample solution to precipitate bacterialgrowthin the dissolutionmedium, respectively,
the enteric polymer. The camostat solution containing since the release test was performed for a week. At
1
−
precipitated polymer was centrifuged at 3000 rev min
predetermined intervals, 150-µL suspension was taken
for 15 min. The camostat concentration of the super- as a sample, centrifuged (3000 rev min 1 for 5 min), and
−
natant was determined using the HPLC method de- the concentration of the substrate in the supernatant
scribed above.
was analysed by the HPLC method described above.

258
Takahiro Uchida et al
60% (20 min). The flow rate was 1.0 mL min 1 and the
−
100
wavelength was set at 410 nm.
80
60
40
20
0
Results and Discussion
Optimization of camostat-loaded PLA
microspheres and their characterization
Table 1 shows the effect of different preparative con-
ditions on the loading efficiency of camostat in PLA
microspheres, all of which were theoretically loaded
with 10% camostat. Camostat is a basic compound and
so its solubility was expected to be low under alkaline
conditions. This expectation was confirmed experimen-
tally (Figure 1). Therefore, the pH of the external
aqueous phase was varied in formulations 1–6. In these
formulations, the internal aqueous phase volume was
fixed at 60 µL, and polymer phase (methylene chloride)
volume was fixed at 1.25 mL. As expected, the drug-
loading efficiency under alkaline conditions (pH 10, 11)
was comparatively high (72.8 and 67.8%, respectively).
The volumes of the internal phase, oily phase, and
external aqueous phase were changed in formulations
7–11. When the internal aqueous phase volume was
increased from 60 µL to 200, 400, or 600 µL, the loading
efficiency was dramatically decreased, as shown in form-
ulations 7–9. This suggested that the smaller volume of
the internal aqueous phase was an important factor for
high loading. Entrapment efficiency seemed to depend
on the phase volume of oil (methylene chloride) relative
to the internal aqueous phase, since the smaller the
oil phase volume\internal aqueous phase volume, the
smaller the exposure probability of the internal aqueous
phase.
0
2
4
6
pH
8
10
12
Figure 1 The influence of pH on the solubility of camostat at 37mC.
Values are the means of three determinationsps.d.
30
20
10
0
0
2
4
6
8
Time (day)
Figure 2 Camostat release from PLA microspheres in pH 6.8 buffer
at 37mC. Values are the means of three determinationsps.d.
In-vitro substrate degradation study of
substrate-loaded hypromellose acetate
succinate microspheres in the absence or
presence of camostat
Finally, when the volume of the oily phase was
Microspheres containing approximately 1 mg substrate decreased to 0.75 mL, the highest loading efficiency
and approximately 0.5 mg camostat, and control micro- (95%) was achieved (formulation 11), which suggested
spheres containing only 1 mg substrate, were used in that a high concentration of methylene chloride might
this study. The two kinds of microspheres were sus- beasignificantbarriertodrugdiffusionfromtheinternal
pended in 50 mL of pH 6 or 7 phosphate-buffered to the external aqueous phase. For the three batches, the
solution containing 2.5 mg trypsin, and shaken hori- yield was almost 80% and the mean average diameters
1
−
zontally at 60 rev min at 37mC. At predetermined of the microspheres were 30p4 µm.
intervals, 150-µL suspension was taken as a sample,
The most favourable sustained release profile was
1
−
centrifuged (3000 rev min for 5 min) and the con- obtained with formulation 11. Almost 10% of the
centration of the substrate, N-benzoyl-dl-arg-4-nitro- camostat was released within 4 h, and this was followed
anilide, in the supernatant was analysed by HPLC. The by a secondary slow-release phase (Figure 2). Approxi-
determination of the degradation product, N-benzoyl- mately 75% of the drug remained unreleased one week
dl-arginine, was performed simultaneously using a mo- later. The remaining camostat was extracted from the
bile phase consisting of: A, 10 m phosphate buffer PLA matrix using the same method as described under
containing 50 m sodium sulfate (pH 2.6); B, aceto- Drug loading and a good mass balance was reached
nitrile. A linear gradient was used: phase B from 5 to (data not shown).

259
Microspheres containing camostat
Table 2 The effect of preparative conditions on camostat loading efficiency in hypromellose acetate
succinate microspheres theoretically loaded with 10% camostat.
Formulation
number
Internal aqueous
phase (w)
Oily
External aqueous
phase (w)
Loading
efficiency
(%)
s.d.
phase (o)
methylene
chloride
(Camostat 20 mg)
PVA concn Buffer
Solvent
Volume
volume (mL) (%)
pH
3
(µL)
1
2
3
4
5
Citrate
buffer
(pH 3)
Citrate
buffer
(pH 3)
Citrate
buffer
(pH 3)
Citrate
buffer
(pH 3)
Citrate
buffer
(pH 3)
60
5.0
5.0
5.0
2.5
1.25
0.5
0.5
0.5
0.5
0.5
14.3
3.5
3.1
1.2
0.8
4.1
5.2
200
400
60
3
3
2.7
3
69.0
81.0
60
3
Each value represents the mean of four to six experiments.
Optimization of camostat-loaded hypromellose
acetate succinate microspheres and their
characterization
100
80
60
40
20
0
In a pilot study, we found that the critical factor for the
successful preparation of enteric microspheres was the
choice of an acidic external aqueous phase, such as pH
3.0 citrate buffer containing 0.5% PVA solution. When
0.5% PVA (pH 5.5) was used as an external aqueous
phase for microencapsulation, the prepared enteric
microspheres were very fragile and were ruptured by
centrifugation in the recovery process (data not shown).
Therefore, in this study we used pH 3.0 citrate buffer
containing 0.5% PVA.
pH 1.2
pH 6.8
0
1.0
2.0
3.0
Time (h)
Figure 3 Camostat release from hypromellose acetate succinate
microspheres in pH 1.2 or 6.8 buffer at 37mC. Values are the means of
three determinationsps.d.
Table 2 summarizes the effect of preparative con-
ditions on camostat loading efficiency in hypromellose
acetate succinate microspheres theoretically loaded with
10% camostat. Formulations 1–3 showed the effect of
internal aqueous phase volume on loading efficiency. As when 1.25 mL methylene chloride was used as an oleagi-
the internal aqueous phase volume increased from 60 µL nous phase, over 80% loading was observed. As sug-
to 200 or 400 µL, the loading efficiency dramatically gested for PLA microspheres, the high concentration of
decreased to 3.5% or 2.7%, respectively.
methylene chloride might present a significant barrier to
Secondly, the effect of oily phase volume (polymer drug diffusion from the internal to the external aqueous
concentration of methylene chloride) on loading effi- phase.
ciency was examined. In this case, the internal aqueous Figure 3 shows the camostat release profiles from
phase volume was fixed at 60 µL. In formulation 5, hypromellose acetate succinate microspheres at pH 1.2

260
Takahiro Uchida et al
the pH 1.2 buffer, only a low level of release was
observed. This information confirmed that enteric-
coated microspheres had been prepared successfully.
A
100
80
60
40
20
0
Preparation and characterization of
hypromellose acetate succinate microspheres
loaded with N-benzoyl-dl-arg-4-nitroanilide and
camostat
Batch B (pH 7)
Batch A (pH 7)
Batch B (pH 6)
Batch A (pH 6)
Two batches of hypromellose acetate succinate micro-
spheres were prepared, one containing 4.5% camostat
and 9.5% N-benzoyl-dl-arg-4-nitroanilide as a substrate
(batch A), and a second containing 9.2% substrate only
(batch B). For the two batches, the yield was almost
80% and the mean average diameters of the micro-
spheres were 34p5 and 36p7 µm, respectively.
Figure 4A shows the cumulative substrate released
from the two batches of microspheres in pH 6 or 7 buffer
medium without trypsin. As expected, at pH 7 a com-
paratively fast release was observed for both batches,
and almost all substrate was released within 6 h. In-
corporation of camostat did not affect the release rate of
the substrate significantly. The release rate at pH 6 was
slightly slower.
0
1.0
2.0
3.0
4.0
5.0
6.0
Time (h)
B
100
80
60
40
20
0
Batch A (pH 6)
Batch A (pH 7)
Figure 4B shows the release profile of camostat from
batch A at pH 6 or 7 in a medium without trypsin. Even
in this case, comparatively fast release of camostat was
observed at pH 7, while the release rate at pH 6 was
more moderate. The difference in the release rates of
camostat and N-benzoyl-dl-arg-4-nitroanilide seemed
to be due to differences in their solubilities. This ex-
periment also showed that the substrate was released
from hypromellose acetate succinate microspheres to-
gether with camostat. Once the substrate was released
from the microspheres into the medium in the presence
of trypsin, it would be degraded immediately and the
degradation product would be formed. Therefore, as
shown in Figure 4A and B, we confirmed the substrate
was released with camostat from hypromellose acetate
succinate microspheres, even though the buffer medium
did not contain trypsin.
0
1.0
2.0
3.0
4.0
5.0
6.0
Time (h)
C
100
80
60
40
20
0
Batch A (pH 7)
Batch B (pH 7)
0
1.0
2.0
3.0
4.0
5.0
6.0
Time (h)
Finally, Figure 4C shows the appearance of degra-
dation product when batch A or B microspheres were
incubated in pH 7 buffer containing trypsin as a pro-
tease. For hypromellose acetate succinate microspheres
containing substrate only (batch B), the substrate degra-
dation product N-benzoyl-dl-arginine was produced
almost immediately at pH 7 in a trypsin-containing
medium, in contrast with the hypromellose acetate
Figure 4 A. Release profile of N-benzoyl-dl-arg-4-nitroanilide from
hypromellose acetate succinate microspheres with and without camo-
stat at pH 6 or 7 at 37mC. Batch A contained 4.5% camostat and
9.5% N-benzoyl-dl-arg-4-nitroanilide; batch B contained 9.2% N-
benzoyl-dl-arg-4-nitroanilide. B. Release profile of camostat from
batch A at pH 6 or 7. C. Appearance of the degradation product N-
benzoyl-dl-arginine from batch A or B at pH 7 in the presence of
trypsin. Each value represents the mean of two or three experiments.
and 6.8. In pH 6.8 buffer, comparatively fast release was succinate microspheres containing camostat and sub-
observed, suggesting the enteric characteristics of the strate (batch A), in which the released camostat almost
microspheres; all drug was released within 45 min. In completely prevented the degradation of the substrate.

261
Microspheres containing camostat
capsules containing proteinase inhibitor for oral protein
administration. Int. J. Pharm. 189: 171–178
In the formulation design, the release rates of sub-
strate and camostat seemed to be important criteria.
Not only the substrate release rate but also the release
rate of camostat seems to be important in drug delivery.
Naito, K., Morioka, S., Nakajima, S., Ogawa, H (1989)
Proteinase inhibitors block formation of pemphigus
acantholysis in experimental models of neonatal mice and
skin explants: effects of synthetic and plasma proteinase
inhibitors on pemphigus acantholysis. J. Invest. Dermatol.
93: 173–177
Ogawa, Y., Yamamoto, M., Okada, H., Yashiki, H.,
Shimamoto, T. (1988) Controlled-release of leuprolide acet-
ate from polylactic acid or copoly(lactic\lycolic) acid micro-
spheres; influence of molecular weight and copolymer ratio
of polymer. Chem. Pharm. Bull. 36: 1095–1103
References
Edman, P., Bjork, E., Ryden, L. (1993) Characterization of
drug loaded poly(dl-lactide) microspheres. J. Contr. Rel.
21: 165–172
Taniguchi, H., Yomota, E., Kume, E., Shikano, T., Endo, T.,
Nagasaki, M. (1997) Effect of T-0632, a cholecystokininA
receptor antagonist, on experimental acute pancreatitis.
Jpn. J. Pharmacol. 73: 105–112
Heya, T., Okada, H., Tanigawara, Y., Ogawa, Y., Toguchi,
H. (1991) Effects of counteranion of TRH and loading
amount on control of TRH release from copoly(dl-lactic\
glycolic acid) microspheres prepared by an in-water drying
method. Int. J. Pharm. 69: 69–75
Uchida, T., Goto, S., Foster, T. P. (1995) Particle sizes studies
for subcutaneous delivery of poly(lactide-co-glycolide)
microspheres containing ovalbumin as vaccine formulation.
J. Pharm. Pharmacol. 46: 1214–1221
Jalil, R., Nixon, J. R. (1989) Microencapsulation using poly
(L-lactic acid). I: Microcapsule properties affected by the
preparative technique. J. Microencapsul. 6: 473–484
Kentepozidou, A., Kiparissides, A. (1995) Production of
water-containing polymer microcapsules by the complex
emulsion\solvent evaporation technique. Effect of process
variables on the microcapsule size distribution. J. Micro-
encapsul. 12: 627–638
Uchida, T., Yoshida, K., Goto, S. (1996) Preparation and
characterization of polylactic acid microspheres containing
water-soluble dyes using a novel w\o\w emulsion solvent
evaporation method. J. Microencapsul. 13: 219–228
Uchida, T., Shiosaki, K., Nagareya, N., Fukuda, K., Eda, Y.,
Tokiyosi, S., Matsuyama, K. (1998) In vivo evaluation of
poly(lactide-co-glycolide) microspheres containing Hepa-
titis B core antigen. Pharm. Res. 15: 1708–1713
Larionova, N. V., Ponchel, G., Duchene, D., Larionova, N. I.
(1999) Biodegradable cross-linked starch\protein micro-