Journal of Controlled Release 66 (2000) 271–280

www.elsevier.com / locate / jconrel

Controlled release of a water-soluble drug, captopril, by a

combination of hydrophilic and hydrophobic cyclodextrin
derivatives
Yoichi Ikeda a , Kenya Kimura b , Fumitoshi Hirayama b , Hidetoshi Arima b ,
Kaneto Uekama b , *
a
Wakunaga Pharmaceutical Co., 1624 Shimokotachi, Kodacho, Takata-gun, Hiroshima 739 -1195, Japan
b
Faculty of Pharmaceutical Sciences, Kumamoto University, 5 -1 Oe-honmachi, Kumamoto 862 -0973, Japan
Received 16 October 1999; accepted 7 December 1999

Abstract

Parent b-cyclodextrin (b-CyD) and 2-hydroxypropyl-b-CyD (HP-b-CyD) form 1:1 solid complexes with an orally active
angiotensin-converting enzyme inhibitor, captopril, while hydrophobic perbutanoyl-b-CyD (TB-b-CyD) forms a solid
dispersion or solid solution with the drug. The binary system of captopril / HP-b-CyD or captopril / TB-b-CyD and the ternary
system of captopril / TB-b-CyD/ HP-b-CyD in different molar ratios were prepared by the kneading method, and the release
behavior of the drug was investigated. The release rate of captopril from the binary HP-b-CyD system was rather fast,
whereas that from the binary TB-b-CyD system was comparatively slower, the retarding effect being dependent on the
amounts of TB-b-CyD. The release rate from the ternary captopril / TB-b-CyD/ HP-b-CyD system was slowed down by the
addition of small amounts of HP-b-CyD, whereas the rate became faster as the molar ratio of HP-b-CyD further increased
(.0.25 molar ratio). Both water penetration studies and microscopic observation suggested that the retarding effect is
attributable to a gel formation of HP-b-CyD in the TB-b-CyD hydrophobic matrix. It was difficult to prolong plasma levels
of captopril by administering orally either the binary HP-b-CyD or TB-b-CyD system in dogs. On the other hand, the ternary
captopril / TB-b-CyD/ HP-b-CyD system (molar ratio of 1:0.5:0.5) gave a plasma profile comparable to that of a
commercially available sustained release preparation (Captoril R ). Therefore, a combination of HP-b-CyD and TB-b-CyD
is useful for the controlled release of water-soluble drugs such as captopril.  2000 Elsevier Science B.V. All rights
reserved.

Keywords: 2-Hydroxypropyl-b-cyclodextrin; Perbutanoyl-b-cyclodextrin; Combination of hydrophilic and hydrophobic cyclodextrins;

Captopril; Controlled release; Gel formation

1. Introduction efficacy of drug activity through the use of rationally

There is increasing interest in optimizing the
*Corresponding author. Tel.: 181-96-371-4160; fax: 181-96-
372-7023.
E-mail address: uekama@gpo.kumamoto-u.ac.jp (K. Uekama)
designed drug carrier materials. Cyclodextrin (CyD)
is a strong candidate for such a role, modifying
physical, chemical, and biological properties of drug
molecules through the formation of inclusion com-
plexes [1–3]. Recently, various kinds of CyD deriva-
tives have been prepared to improve physicochemi-

0168-3659 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved.
PII: S0168-3659( 99 )00286-2
272 Y. Ikeda et al. / Journal of Controlled Release 66 (2000) 271 – 280
cal properties of parent CyD as multi-functional drug
carriers. For example, hydrophilic CyD derivatives
such as 2-hydroxypropyl-b-CyD and sulfobutylether
of b-CyD are useful for improving solubility and
dissolution rates of poorly water-soluble drugs, while
hydrophobic CyD derivatives such as ethylated and
acylated CyDs can work as slow-release carriers for
water-soluble drugs [4–6]. Furthermore, advanced
controlled releases can be achieved by a pertinent
combination of CyD derivatives and pharmaceutical
polymers [7,8]. In a previous paper [9], we reported
that a combination of short- and long-chain
peracylated b-CyDs serve as a sustained release
carrier for a water-soluble drug, diltiazem, i.e., the
oral administration of the diltiazem / peracetyl-b-
CyD/ peroctanoyl-b-CyD system in dogs gave a
constant plasma drug level for more than 48 h, with
a consequent significant increase in bioavailabiity.
Furthermore, the double-layer tablet consisting of the
nifedipine / 2-hydroxypropyl-b-CyD/ HCO-60 system
as the fast release portion and the nifedipine / hy-
droxypropylcellulose system as the slow release
portion gave prolonged plasma drug levels with a
more balanced bioavailability [10,11]. In this study,
we conducted the release control of an orally active
angiotensin-converting enzyme inhibitor, captopril,
by means of a combination of hydrophiic and
hydrophobic CyD derivatives, i.e., 2-hydroxypropyl-
b-CyD (HP-b-CyD) and perbutanoyl-b-CyD (TB-b-
CyD), respectively, and report here that HP-b-CyD
plays an important role in the release of the drug
from the hydrophobic TB-b-CyD matrix.
2. Materials and methods
2.1. Materials
HP-b-CyD (degree of substitution 4.8) was sup-
plied from Japan Maize Co. (Tokyo, Japan) and
TB-b-CyD was prepared according to the method
reported previously [12]. Captopril was purchased
from Sigma Co. (St. Louis, USA). Other chemicals
and solvents were of analytical reagent grade, and
deionized, double-distilled water was used through-
out the study.
2.2. Preparation of binary and ternary captopril /
CyD systems
The binary system of captopril with b-CyD, HP-b-
CyD or TB-b-CyD and the ternary system of captop-
ril with (HP-b-CyD/ TB-b-CyD) were prepared by
kneading appropriate amounts of these components
in water or ethanol. For example, captopril (217 mg)
and b-CyD (1135 mg) or HP-b-CyD (1413 mg) in a
molar ratio of 1:1 were triturated with a small
amount (1–2 ml) of water and kneaded thoroughly
for about 40 min and the resulting slurry was dried
under vacuum for 24 h at 408C. After sieving,
powder fractions with a particle size less than 100
mesh were used for further studies. For the prepara-
tion of binary captopri / TB-b-CyD and ternary
captopril / TB-b-CyD/ HP-b-CyD systems, ethanol
was used as a solvent, because TB-b-CyD is practi-
cally insoluble in water.
2.3. Apparatus
The differential scanning calorimetry (DSC) was
carried out using a SEIKO EXTAR 6000 thermal
analyzer (Chiba, Japan), operated with a sample
weight of 5 mg and a scanning rate of 108C / min
(Fig. 1) or 58C / min (Fig. 2). The heat of fusion was
calibrated with indium (purity 99.99%; melting
156.608C; DH 28.59 mJ / mg; heating rate 108C /
min). The powder X-ray diffraction patterns were
measured with a Rigaku Rint-2500 diffractometer
(Tokyo, Japan) under the following conditions: Ni-
filtered Cu-Ka radiation (1.542 A),˚ a voltage of 40
kV, a current of 40 mA, a divergent slit of 1.74 mm
(18), a scattering slit of 0.94 mm (18), a receiving slit
of 0.15 mm, and a goniometer angular increment of
18 / min. Appearance of solid samples was observed
with a scanning electron microscope (SEM) instru-
ment (Hitachi-Akashi 510, Tokyo, Japan), after
samples were mounted onto a SEM sample stub with
double-sided sticky tape and coated with gold by a
direct current sputter technique. Water penetration
rates into tablets were measured by means of the
apparatus reported previously [6]. The powder sam-
ple (50 mg) was compressed into a cylindrical tablet
(diameter 7 mm) at a pressure of 1000 kg / cm 2 for 1
min. The uptake volume of water was read from a
calibrated capillary tube (diameter, 0.58 mm).
 Y. Ikeda et al. / Journal of Controlled Release 66 (2000) 271 – 280 273

Fig. 1. DSC thermograms of binary and ternary captopril / CyD systems with different compositions. (A) captopril / b-CyD system, (B)
captopril / HP-b-CyD system, (C) captopril / TB-b-CyD system, (D) captopril / TB-b-CyD/ HP-b-CyD system. *Molar ratio. p.m.: Physical
mixture.

2.4. Release studies ternary drug / CyDs systems (equivalent to 5 mg

captopril) was measured according to the paddle
The release rate of captopril from hard gelatin method of the dissolution test in Japanese Phar-
capsules (volume 0.95 ml) containing the binary and macopoeia XIII, i.e., the capsule containing test
274 Y. Ikeda et al. / Journal of Controlled Release 66 (2000) 271 – 280
Fig. 2. Plots according to Eq. 1 for binary captopril / b-CyD and
captopril / HP-b-CyD systems. The solid lines are theoretical ones
calculated on the basis of 1:2, 1:1 and 2:1 (captopril:CyD)
stoichiometries, and the diamonds are experimental data of b-CyD
(♦) and HP-b-CyD (앳) systems.
powder was placed in a stainless steel basket and
immersed in the dissolution medium. The dissolution
medium (JP XIII second fluid (pH 6.8), 500 ml) was
continuously stirred at 100 rpm at 378C. At appro-
priate intervals, an aliquot (0.5 ml) was withdrawn
with a cotton plugged pipette and analyzed for
captopril by high performance liquid chromatog-
raphy (HPLC) under the following conditions: a
Hitachi L-655 A-11 pump equipped with a Hitachi
655A UV detector, a Hitachi D-2500 chromato-inte-
grator (Tokyo, Japan), a Shiseido C18 AG 120
column (5.0 mm, 4.63250 mm, Tokyo, Japan), a
mobile phase of 25 mM KH 2 PO 4 (pH 2.5) / acetoni-
trile (25:5 v / v), a flow rate of 1.0 ml / min, and a
detection of 220 nm. There was no degradation of
captopril such as a formation of captopril disulfide
dimer, during the dissolution test.
2.5. Absorption studies
The absorption studies were carried out using male
beagle dogs (9–11 kg) that were fasted for 24 h
before drug administration. The capsules (volume
0.95 ml) containing the binary and ternary captopril /
CyDs systems (equivalent to 18.75 mg captopril)
were administered orally with water (50 ml). Blood
samples (2.0 ml) were withdrawn from the cephalic
vein with a heparinized syringe and centrifuged at
11003g for 10 min. The total amount of captopril in
plasma was determined by treating the plasma with
tri-n-butylphosphine to reduce the oxidized form of
captopril [13] and employing N-(3-fluoranthyl)
maleimide (FAM) as a fluorescent probe to detect
the thiol of captopril [14]. To 0.5 ml of the plasma
sample were added 1.0 ml of 0.1 M phosphate buffer
(pH 6.75), 0.1 ml of 0.1 M disodium EDTA solution
and 0.1 ml of 1% tri-n-butylphosphine / methanol
solution. Captopril standard solution (50 ml) was
added to the plasma blank. After the mixture was
warmed at 508C for 1 h, it was cooled in an ice box
for a few minutes. Immediately 0.5 ml of 0.5 M
phosphate buffer (pH 7.0) was added and the mix-
ture was washed with 2.0 ml of toluene. A 75 ml of
0.2% FAM / acetonitrile solution was added to 1.0 ml
of the aqueous layer, and the mixture was allowed to
stand for 15 min at room temperature. The mixture
was acidified with 0.3 ml of 1.0 N HCl and then
washed with 6.0 ml of n-heptane. One millilitre of
the aqueous layer was shaken with 5.0 ml of toluene.
The organic layer (4.0 ml) was evaporated to dryness
under reduced pressure and the residue was dissolved
in 0.4 ml of mobile phase (A) for HPLC analysis.
The HPLC conditions were as follows: a Shimadzu
LC-10ATVP pump equipped with a Shimadzu RF-
10AXL spectrofluorometer, a Shimadzu CBM-10A
communication bus module (Kyoto, Japan), a Tosoh
TSK-GEL ODS-80TS column (5 mm, 4.63150 mm,
Tokyo, Japan), mobile phases (A) and (B) were
consisted of 5.0 mM KH 2 PO 4 (pH 3.0) solution /
acetonitrile (60 / 40 v / v) and (40 / 60 v / v), respective-
ly, a flow rate of 1.2 ml / min with a gradient elution,
and an excitation of 360 nm and a detection of 462
nm.
3. Results and discussion
3.1. Interaction of captopril with CyDs
DSC and powder X-ray diffraction studies were
carried out to gain insight into the interaction of
captopril with CyDs in the solid state. Fig. 1A,B
shows DSC thermograms of the binary system
 Y. Ikeda et al. / Journal of Controlled Release 66 (2000) 271 – 280
consisting of captopril with b-CyD or HP-b-CyD,
respectively, in different molar ratios. Captopril gave
an endothermic peak at about 1068C due to the
melting. This peak intensity decreased with increas-
ing amounts of b-CyD and HP-b-CyD and dis-
appeared at a molar ratio (captopril / CyDs) of about
1:1, whereas the 1:1 physical mixture of each
component gave the endothermic peak at 1068C.
Giordano et al. [15] reported that the stoichiometry
of solid CyD complexes can be determined by DSC
measurements of the complexes containing excess
amounts of guest, where the free mole fraction of
guest (FGMF) is expressed by Eq. (1):
FGMF 5 TGMF(1 1 R) 2 R (1)
where TGMF and R stand for the total mole fraction
of guest and the stoichiometry (guest / host) of the
complex, respectively. Therefore, DSC curves of the
b- and HP-b-CyD systems were analyzed by Eq. (1).
Fig. 2 shows the theoretical lines of Eq. (1), assum-
ing R52 (2:1), 1(1:1), and 0.5 (1:2), together with
FGMF values obtained experimentally from the
fusion enthalpy of captopril. The fusion enthalpy
(28.53 kJ / mol) of captopril was assumed not to be
changed by the addition of b- and HP-b-CyDs,
because there was no change in the melting tempera-
ture. It is apparent that the experimental data fitted
well to the theoretical line of R51, indicating a 1:1
stoichiometry of the solid b-CyD and HP-b-CyD
complexes. In the case of the TB-b-CyD system.
(Fig. 1C), two endothermic peaks were observed.
The peak intensity at 1068C decreased with increas-
ing amounts of TB-b-CyD and disappeared at a
molar ratio of about 1:3. The melting temperature of
TB-b-CyD gradually shifted to 1248C, the intrinsic
melting temperature of TB-b-CyD, with increasing
amounts of TB-b-CyD, and there was no change in
the temperature above the 1:3 molar ratio. Unfor-
tunately, DSC data of the TB-b-CyD system could
not be quantitatively analyzed by Eq. (1), because
the melting temperature of TB-b-CyD varied as the
composition changed and two endothermic peaks
overlapped each other at lower molar ratios. These
results suggested that captopril forms a solid disper-
sion or solid solution with TB-b-CyD. Powder X-ray
diffraction studies supported the above results, i.e.,
captopril gave higher diffraction peaks at 2u 519.78,
275
25.88 and 26.08, but these peaks were disappeared by
the complexations with of b-CyD and HP-b-CyD
above a 1:1 molar ratio and TB-b-CyD above a 1:3
molar ratio (drug:CyD), whereas simple physical
mixtures of each component gave diffraction peaks
characteristic of the drug.
Fig. 1D shows DSC curves of the ternary
captopril / TB-b-CyD/ HP-b-CyD system with molar
ratios of 1 / 0.5 /x (x50–0.5). The intensity of the
endothermic peak of captopril at 1068C was lowered
by the addition of HP-b-CyD and disappeared when
the molar ratio exceeded 0.5. Similar results were
obtained in the ternary captopril / TB-b-CyD/ HP-b-
CyD system with a molar ratio of 1 / 0.25 / 0.5. These
compositions were different from those of the binary
captopril / HP-b-CyD or captopril / TB-b-CyD system
described above. There may be additional molecular
interactions, such as HP-b-CyD/ TB-b-CyD and
complex / TB-b-CyD or high order complexation of
the drug with HP-b-CyD, although it was difficult to
describe quantitatively the exact interaction mode of
the ternary system. However, it is apparent that
captopril was monomolecularly dispersed in these
TB-b-CyD/ HP-b-CyD matrices. Unfortunately, it
was difficult to prepare the ternary system of parent
b-CyD by means of the kneading method, because of
the difficulty in dissolving three components in a
solvent, i.e., b-CyD is soluble in water but less
soluble in ethanol whereas the reverse was found for
TB-b-CyD [12].
3.2. Release behavior of captopril from binary and
ternary systems
Fig. 3A,B shows release profiles of captopril from
capsules containing the binary and ternary systems,
respectively, in JP XIII second fluid (pH 6.8) at
378C. Captopril dissolved rapidly in the medium,
reflecting its high aqueous solubility (160 mg / ml).
The release of the drug from the binary b-CyD and
HP-b-CyD systems was completed within about 15
min. On the other hand, hydrophobic TB-b-CyD
retarded the release rate, the retarding effect being
dependent on amounts of TB-b-CyD in the matrices.
It is of interest to note that in the ternary captopril /
TB-b-CyD/ HP-b-CyD system (Fig. 3B), the release
rate was slowed down by the addition of small
amounts HP-b-CyD, in spite of the formation of a
276 Y. Ikeda et al. / Journal of Controlled Release 66 (2000) 271 – 280
Fig. 3. Release profiles of captopril from capsules containing
binary captopril / CyD systems (A) and ternary captopril / TB-b-
CyD/ HP-b-CyD (B) systems in JP XIII second fluid (pH 6.8) at
378C, measured by paddle method. (A): captopril alone (d);
captopril / b-CyD (1:1, ♦); captopril / HP-b-CyD (1:1, 앳);
captopril / TB-b-CyD (1:0.5, s); (1:1, n); (1:2, m); (1:3, h). (B):
captopril / TB-b-CyD/ HP-b-CyD (1:0.5:0, s); (1:0.5:0.25, d);
(1:0.5:0.5, n); (1:0.5:1, m); (1:0.5:1.5, h); (1:0.5:2, j).
fast dissolving complex with HP-b-CyD. However,
the rate became faster as the molar ratio of HP-b-
CyD further increased (.0.25 molar ratio).
In order to gain insight into the release mechanism
of the ternary system, these profiles were analyzed
according to the Korsmeyer equation (Eq. 2, [16]):
Mt 5 Ca ? k ? t n (2)
where Mt and Ca are the amounts of released drug at
time t and the total amount of drug in the matrix,
respectively, and k and n are the release rate constant
and an exponent which characterizes the release
mechanism, respectively. The analysis of release
profiles gave n values different between the early
(0.0–0.5 h) and later (0.5–4.0 h) stages of the
release, as shown in Table 1. The n values were in
the range of 0.8–1.0 for the initial release, except of
the 1:0.5:0.25 system. On the other hand, the later
stage gave an n value 0.4–0.5, indicating that the
initial stage is controlled by a dissolution mechanism
whereas the later stage is controlled by a diffusion
mechanism. Fig. 4 shows water-penetration profiles
of tablets consisting of the captopril / TB-b-CyD/ HP-
b-CyD (1:0.5:x) system. The water penetration rate
into the binary captopril / TB-b-CyD system (x50)
was very fast, but it was markedly decreased by the
addition of HP-b-CyD (x50.5) to the TB-b-CyD
matrix. Further additions of HP-b-CyD (x51 and 3)
again accelerated the penetration rate. In the case of
the binary HP-b-CyD system (h in Fig. 4), only a
small amount of water (about 3 ml) was taken at the
initial stage, but further measurements were difficult
because of dissolution of the tablet or formation of a
highly viscous layer on the tablet surface. This water
uptake behavior was very similar to the release
behavior of captopril from the ternary system (Fig.
3B), suggesting an important role of water-penetra-
tion into tablets in the release of captopril from the
ternary system. When the surface of the ternary
tablets was wetted by a drop of water and stood for
30 min, a transparent gel layer was clearly observed.
Fig. 5 shows SEM pictures of the tablet surface of
captopril / TB-b-CyD/ HP-b-CyD systems before and
after the water-penetration experiment. The tablet of
the binary captopril / TB-b-CyD (1:0.5) system had a
slightly coarse surface with tiny pores before the
experiment, while after the experiment it became
rougher with many pores, through which water may
penetrate and the drug may be released. The ternary
captopril / TB-b-CyD/ HP-b-CyD system before the
 Y. Ikeda et al. / Journal of Controlled Release 66 (2000) 271 – 280 277

Table 1
Parameters a of release process of captopril from capusules containing ternary captopril / TB-b-CyD/ HP-b-CyD systems with different
compositions b
Molar ratio of Early stage (0.0–0.5 h) Later stage (0.5–4.0 h)
captopril / TB-b-CyD/ HP-b-CyD 2n
k (h ) n k (h 2n) n
1 / 0.5 / 0.25 0.3660.15 0.4960.11 0.2660.05 0.2860.07
1 / 0.5 / 0.5 0.3860.04 0.8360.23 0.2960.01 0.3960.04
1 / 0.5 / 1 0.4760.12 0.8460.00 0.3360.03 0.4360.06
1 / 0.5 / 1.5 0.5060.15 0.9060.04 0.3060.08 0.4560.04
1 / 0.5 / 2 0.5860.14 0.9760.02 0.3660.06 0.4860.02
a
In Korsmeyer Eq. (2) (see text).
b
Each value represents the mean6S.E. of 3–4 experiments.

experiment showed a coarse surface with many void
spaces between aggregated particles. This coarse
surface changed to a smooth surface with pores of
different sizes, at the end of the experiment. The
pore size was dependent on the composition of
HP-b-CyD, i.e., small pores in lower HP-b-CyD
contents (x50.25 and 0.5), whereas large pores in
larger HP-b-CyD contents (x52). These results
suggested that HP-b-CyD forms a gel on the surface
or inside the tablets after contact with water, re-
sulting in a smooth surface with tiny pores in the
SEM observation. This gelation of HP-b-CyD may
prevent the penetration of water into tablets or the
release of captopril from tablets. In the case of large
HP-b-CyD contents, captopril, HP-b-CyD or its
complex dissolved or eroded from the surface,
resulting in larger pores on a smooth surface. There-
fore, the release of captopril from the ternary TB-b-
CyD/ HP-b-CyD system may proceed in two mecha-
nisms, i.e., captopril or HP-b-CyD complex dis-
solves from the surfaces of tablets at an early stage
of the release, while a viscous gel layer of HP-b-
CyD gradually forms on the surfaces or inside the
tablets. At a later stage of the release, captopril may
be released through a diffusion of the gel layer,
together with the dissolution of the drug or the
complex. In the system of the lowest HP-b-CyD
content (1:0.5:0.25, Table 1), the diffusion mecha-
nism may be in operation from the initial release
stage, because of the decreased hydrophilic surface
area or smaller pore size. The recovery in release
rate of the ternary system with higher molar fractions
of HP-b-CyD (x52) may be due to a fast dissolution
or erosion of the complex or HP-b-CyD itself.

3.3. Absorption behavior of captopril in dogs

Fig. 6 shows plasma levels of captopril after the

Fig. 4. Penetration of water into tablets (diameter 7 mm, weight
50 mg) of binary and ternary captopril / CyD systems with
different compositions at 378C. Captopril / TB-b-CyD/ HP-b-CyD
(1:0.5:0, s); (1:0.5:0.5, n); (1:0.5:1, m); (1:0.5:3, ♦); (1:0:0.5,
h).
oral administration of capsules containing the binary
HP-b-CyD (1:0.5) and TB-b-CyD (1:0.5) systems
and the ternary HP-b-CyD/ TB-b-CyD (1:0.5:0.5)
system in dogs. The absorption and disappearance of
captopril were fast for the drug alone system, giving
pharmacokinetic parameters as follows: t max (time
278 Y. Ikeda et al. / Journal of Controlled Release 66 (2000) 271 – 280

Fig. 5. Scanning electron micrographs of binary and ternary captopril / CyD systems before and after water penetration experiment.
Captopril / TB-b-CyD/ HP-b-CyD (1:0.5:0, A); (1:0.5:0.25, B); (1:0.5:0.5, C); (1:0.5:2, D).
 Y. Ikeda et al. / Journal of Controlled Release 66 (2000) 271 – 280
Fig. 6. Plasma levels of captopril after oral administration of
capsules containing binary captopril / CyD systems, ternary
captopril / TB-b-CyD/ HP-b-CyD system and Captoril-R (equiva-
lent to 18.75 mg captopril) in dogs. d, captopril alone; 앳,
captopril / HP-b-CyD (1:0.5); s, captopril / TB-b-CyD (1:0.5); n,
captopril / TB-b-CyD/ HP-b-CyD (1:0.5:0.5); m, Captoril R .
Each point represents the mean6S.E. of 3–4 experiments.
279
required to reach the maximum plasma concentration
(Cmax ))50.860.2 h, Cmax 52.660.0 mg / ml, MRT
(mean residence time)53.660.1 h, AUC (area under
the plasma concentration–time curve up to 12 h)5
7.360.0 h(mg / ml) and F (extent of bioavailability
compared with the AUC value of intraveneously
administered captopril)538.4%. These parameters
were insignificantly changed by the oral administra-
tion of the binary captopril / HP-b-CyD and
captopril / TB-b-CyD systems, i.e., t max 51.260.3 h,
Cmax 52.160.5 mg / ml, MRT53.860.2 h, AUC5
7.760.5 h(mg / ml) and F540.5% and t max 50.860.1
h, Cmax 52.560.0 mg / ml, MRT53.460.1 h, AUC5
7.060.5 h(mg / ml) and F 536.8%, respectively. The
binary captopril / TB-b-CyD system with higher
molar ratios of 1:2 and 1:3 decreased the plasma
levels by lowering the bioavailability and no sus-
tained-release pattern was observed. Therefore, it is
difficult to prolong plasma levels of captopril by
administering singly either HP-b-CyD or TB-b-CyD
systems. On the other hand, a sustained plasma level
of the drug was observed when the ternary captopril /
TB-b-CyD/ HP-b-CyD with a molar ratio of
1:0.5:0.5, which showed a slow release pattern in
vitro (Fig. 3B), was administered. This system gave
the plasma profile comparable to that of a commer-
cially available sustained release preparation (Cap-
toril R [17]), although many other factors such as
effects of foods, gastrointestinal pH, gastric empty-
ing rate, and first-pass effect etc. should be taken into
consideration to compare quantitatively two prepara-
tions in dogs [18,19].
4. Conclusions
The present results suggested that a combination
of HP-b-CyD and TB-b-CyD is useful for the
controlled release of a water-soluble drug, captopril,
and the release rate may be controlled by adjusting
the molar ratio of the components. The gel formation
of HP-b-CyD plays an important role in the sus-
tained release. It is reported that a blend of water-
insoluble ethylcellulose and water-soluble hydroxy-
propylcellulose is useful for release control of water-
soluble drugs, where the latter excipient controls the
release rate through a gel formation [20,21]. The
present hydrophilic / hydrophobic CyD system seems
280 Y. Ikeda et al. / Journal of Controlled Release 66 (2000) 271 – 280
to resemble this polymer blending system. However,
HP-b-CyD has higher stabilizing and solubilizing
abilities over hydroxypropylcelluloses. Therefore,
HP-b-CyD can work as a gel-forming agent with an
inclusion ability and may be used as a substitute of
hydroxypropylcellulose in release controls. The gel
forming mechanism of HP-b-CyD and the blending
effect of other hydrophobic matrices are now under
investigation and will be reported elsewhere.
Acknowledgements
We thank S. Motoune, N. Ono, M. Yamamoto, F.
Kihara, T. Yamano and M. Wakasugi for their
technical assistance.
References
[1] W. Saenger, Cyclodextrin inclusion compounds in research
and industry, Angew. Chem. Int. Ed. Engl. 19 (1980) 344–
362.
[2] D. Duchene,ˆ C. Vaution, F. Glomot, Cyclodextrins: their
value in pharmaceutical technology, Drug Dev. Ind. Pharm.
12 (1986) 2193–2215.
[3] K. Uekama, M. Otagiri, Cyclodextrins in drug carrier
system, CRC Crit. Rev. Ther. Drug Carrier Systems 3 (1987)
1–40.
[4] K. Uekama, F. Hirayama, T. Irie, Cyclodextrin drug carrier
systems, Chem. Rev. 98 (1998) 2045–2076.
¨
[5] J. Savolainen, K. Jarvinen, ¨
L. Matilainen, T. Jarvinen,
Improved dissolution and bioavailability of phenytoin by
sulfobutylether-b-cyclodextrin ((SBE) 7m -b-CD) and
hydroxypropyl-b-cyclodextrin (HP-b-CD) complexation, Int.
J. Pharm. 165 (1998) 69–78.
[6] Y. Horiuchi, F. Hirayama, K. Uekama, Slow-release charac-
teristics of diltiazem from ethylated b-cyclodextrin complex-
es, J. Pharm. Sci. 79 (1990) 128–132.
´
[7] M. Masson, ´
T. Loftsson, G. Masson, ´
E. Stefansson, Cyclo-
dextrins as permeation enhancers: some theoretical evalua-
tions and in vitro testing, J. Control. Release 59 (1999)
107–118.
´
[8] H. Fridriksdottir, ´
T. Loftsson, E. Stefansson, Formulation
and testing of methazolamide cyclodextrin eye drop solu-
tions, J. Control. Release 44 (1997) 95–99.
[9] O.A. Soliman, K. Kimura, F. Hirayama et al., Controlled
release of diltiazem by a combination of short- and long-
chain peracylated b-cyclodextrin in dogs, Pharm. Sci. 2
(1996) 533–536.
[10] Z. Wang, F. Hirayama, K. Ikegami, K. Uekama, Release
characteristics of nifedipine from 2-hydroxypropyl-b-cyclo-
dextrin complex during storage and its modification of
hybridizing polyvinylpyrrolidone K-30, Chem. Pharm. Bull.
41 (1993) 1822–1826.
[11] Z. Wang, T. Horikawa, F. Hirayama, K. Uekama, Design and
in-vitro evaluation of a modified-release oral dosage form of
nifedipine by hybridization of hydroxypropyl-b-cyclodextrin
and hydroxypropylcellulose, J. Pharm. Pharmacol. 45 (1993)
942–946.
[12] F. Hirayama, M. Yamanaka, T. Horikawa, K. Uekama,
Characterization of peracylated b-cyclodextrins with differ-
ent chain lengths as a novel sustained release carrier for
water-soluble drugs, Chem. Pharm. Bull. 43 (1995) 130–
136.
[13] K. Hayashi, M. Miyamoto, Y. Sekine, Determination of
captopril and its mixed disulfides in plasma and urine by
high-performance liquid chromatography, J. Chromatogr.
338 (1985) 161–169.
[14] Y. Kanaoka, T. Takahashi, M. Machida, K. Yamamoto, T.
Sekine, N-(3-Fluoranthyl)maleimide (FAM): A medium
lifetime fluorescent thiol reagent, Chem. Pharm. Bull. 24
(1976) 1417–1419.
[15] F. Giordano, G. Bruni, G.P. Bettinetti, Solid-state mi-
crocalorimetry on drug-cyclodextrin binary systems, J. Ther-
mal Anal. 38 (1992) 2683–2691.
[16] R.W. Korsmeyer, R. Gurny, E. Doelker, P. Buri, N.A.
Peppas, Mechanisms of solute release from porous hydro-
philic polymers, Int. J. Pharm. 15 (1983) 25–35.
[17] Y. Seta, F. Higuchi, T. Otsuka et al., Preparation and
pharmacological evaluation of captopril sustained-release
dosage forms using oily semisolid matrix, Int. J. Pharm. 41
(1988) 255–262.
[18] H. Ogata, N. Aoyagi, N. Kaniwa et al., Evaluation of beagle
dogs as an animal model for bioavailability testing of
cinnarizine capsules, Int. J. Pharm. 29 (1986) 121–126.
[19] H. Mizuta, Y. Kawazoe, K. Haga, K. Ogawa, Effect of meals
on gastric emptying and small intestinal transit times of a
suspension in the beagle dog assessed using acetaminophen
and salicylazosulfapyridine as markers, Chem. Pharm. Bull.
38 (1990) 2224–2227.
[20] H. Yuasa, T. Ozeki, Y. Kanaya, K. Oishi, T. Oyake, Applica-
tion of the solid dispersion method to controlled release of
medicine. I. Controlled release of water soluble medicine by
using solid dispersion, Chem. Pharm. Bull. 39 (1991) 465–
467.
[21] T. Ozeki, H. Yuasa, Y. Kanaya, K. Oishi, Application of the
solid dispersion method to the controlled release of medi-
cine. V. Suppression mechanism of the medicine release rate
in the three-component solid dispersion system, Chem.
Pharm. Bull. 42 (1994) 337–343.