Research paper

A new drug nanocarrier consisting of chitosan

and hydoxypropylcyclodextrin
Francesca Maestrelli
a,b
, Marcos Garcia-Fuentes
b
, Paola Mura
a
, Maria Jose Alonso
b,
*
a
Department of Pharmaceutical Sciences, Florence University, Polo Scientico di Sesto Fiorentino, Sesto Fiorentino (Florence), Italy
b
Department of Pharmacy and Pharmaceutical Technology, University of Santiago de Compostela, Spain
Received 31 March 2005; accepted in revised form 14 December 2005
Available online 9 March 2006
Abstract
The objective of the present work was to develop a new drug nanocarrier consisting of nanoparticles made of chitosan and cyclodex-
trins. The rationale behind the design of this new nanosystem was to simultaneously implement the cyclodextrin drug complexation
power and the inherent properties of chitosan nanoparticles, in a unique delivery system. The complexation with the cyclodextrin permits
the solubilization as well as the protection for sensitive drugs, whereas the entrapment in the chitosan network is expected to facilitate
their absorption. Chitosan nanoparticles including hydroxypropylcyclodextrins could be prepared by the ionic crosslinking of chitosan
with sodium tripolyphosphate in the presence of cyclodextrins. Two hydrophobic drugs, triclosan and furosemide, were selected as mod-
els for complexation with the cyclodextrin and further entrapment in the chitosan nanocarrier. The resulting nanosystems were thor-
oughly characterized for their size and zeta potential and also for their ability to associate and deliver the complexed drugs. The
results showed that the size of the nanoparticles was slightly aected by the incorporation of cyclodextrins, whereas the f potential
did not suer a signicant modication. Moreover, the complexation of the drugs with the cyclodextrin facilitated their entrapment into
the nanoparticles, increasing up to 4 and 10 times (for triclosan and furosemide, respectively) the nal drug loading of the nanoparticles.
These results led to the conclusion that the drugcyclodextrin complex was eciently retained in the nanoparticulate structure. Finally,
the in vitro release prole observed for these nanoparticles was characterized by an initial fast release followed by a delayed release phase.
In conclusion, this new nanosystem oers an interesting potential for the transmucosal delivery of hydrophobic compounds.
2006 Elsevier B.V. All rights reserved.
Keywords: Nanoparticle drug carriers; Chitosan; Cyclodextrins; Class IV drugs
1. Introduction
Together with a good toxicity prole, adequate biophar-
maceutical/pharmacokinetic characteristics are essential
for the clinical success of drug candidates [1]. Currently,
a high percentage of drugs with good ecacy/toxicity ratio
fail to go through the drug discovery pipeline due to their
low intestinal absorption [2]. The two main causes of low
oral bioavailability of drugs are poor solubility [3] and
low permeability through biological membranes [4]. The
Biopharmaceutics Classication System gives specic
guidelines to classify drugs by their solubility and perme-
ability characteristics into: class 1 (high solubility, high per-
meability), class 2 (low solubility, high permeability), class
3 (high solubility, low permeability) and class 4 (low solu-
bility, low permeability) [5]. Drug classes 24 are expected
to have their bioavailability constrained by their physico-
chemical characteristics [6]. Two main alternatives have
been chosen in order to solve the problem related to the
increasing number of drug candidates with low bioavail-
ability: (i) dropping drug candidates with poor biopharma-
ceutical characteristics early on the drug development
process and (ii) designing new delivery systems capable of
overcoming the bioavailability problems of drugs.
0939-6411/$ - see front matter 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejpb.2005.12.006
*
Corresponding author. Department of Pharmacy and Pharmaceutical
Technology, University of Santiago de Compostela, Campus Sur s/n,
15782 Santiago de Compostela, Spain. Tel.: +34 981 594627; fax: +34 981
547148.
E-mail address: mjalon@usc.es (M.J. Alonso).
www.elsevier.com/locate/ejpb
European Journal of Pharmaceutics and Biopharmaceutics 63 (2006) 7986
Although the rst approach has often been the rst option
for many pharmaceutical companies, the second one is
gaining increasing attention due to the diculties encoun-
tered over the last years in the area of drug discovery.
Among the drug delivery strategies intended to increase
the bioavailability of drugs, the use of polymeric nanocar-
riers has shown a signicant degree of success. More con-
cretely, in our laboratory, we have developed
nanocarriers made of the polysaccharide chitosan (CS)
which have shown a great capacity to increase the systemic
absorption of peptides [710]. The experiments aimed at
investigating the mechanism of action of these nanocarriers
have suggested that the mucoadhesive properties of CS
play a signicant role in this positive behaviour [11].
Although mucoadhesion is a property inherent to CS, its
presentation in a nanoparticulate form makes feasible the
oral administration avoiding the uncontrolled precipitation
of CS in the intestinal medium. In addition, it has been sug-
gested that CS nanoparticles interact with the mucus layer
providing the delivery of the associated drug to the under-
lying epithelium.
On the other hand, cyclodextrins are well-known cyclic
oligosaccharides with a lipophilic central cavity and a
hydrophilic outer surface, able to form inclusion complexes
with hydrophobic molecules. Cyclodextrin complexation
has been thoroughly investigated for bettering the unfa-
vourable biopharmaceutical properties of drugs, such as
poor solubility and/or stability [12,13]. In practice, cyclo-
dextrin derivatives such as hydroxypropylcyclodextrin
(HPCD) are preferred to natural cyclodextrins for drug
formulation because they have higher water-solubility
and a better biocompatibility prole.
Based on this previous information, our purpose in the
present work was to develop a new nanoparticulate drug
carrier that combines the benets of CS nanoparticles
and cyclodextrins with regard to their potential for enhanc-
ing the bioavailability of drugs. In our understanding, such
a system would be of special interest for the formulation of
class 2 and 4 drugs (low solubility and low permeability).
With this objective in mind, we chose two model drugs, tri-
closan and furosemide, and evaluated the feasibility of
their incorporation and delivery from the new nanosystem.
2. Experimental section
2.1. Materials
Triclosan (TRI), 5-chloro-2(2,4-dichlorophenoxy)phe-
nol, was kindly donated by Carlo Erba, Italy; Furose-
mide (FUR) was from Sigma (Spain); Chitosanase-RD
(from Bacillus sp.) was from Pias Co. (Japan); hydroxy-
propyl-a (HPaCD), hydroxypropyl-b (HPbCD) cyclodex-
trins with an average molar substitution degree (MS, i.e.,
the mole number of substituents per mole of glucose in
the CD) of 0.6 were a gift of Wacker-Chemie GmbH
(Mu nchen, Germany). Tripolyphosphate (TPP) was from
Sigma (Spain). Chitosan hydrochloride (Seacure Cl113,
M
w
110 kDa; Seacure Cl213, M
w
272 kDa) and gluta-
mate (Seacure G113, M
w
128 kDa; Seacure G213, M
w
260 kDa) were supplied by Pronova Biopolymer (Dram-
men, Norway). According to the relevant certicates of
analysis, the degree of deacetylation was 87% for Seacure
Cl113, 84% for Seacure Cl213, 85% for Seacure G113
and 86% for Seacure G213. All other reagents were of
analytical grade or better.
2.2. Methods
2.2.1. Preparation of cyclodextrin-containing CS
nanoparticles
Nanoparticles (NPs) were prepared using the ionotropic
gelation method [14]. Briey, aqueous solutions containing
dierent concentrations of HPCD (from 0 to 25 mM) were
incubated with CS (0.2% w/v) for 24 h under magnetic stir-
ring and then ltered through 0.45 lm pore-size cellulose
lters. The cross-linking agent TPP was added to this solu-
tion (TPP/CS w/w ratios of 1:3 to 1:6) leading to the con-
trolled gelation of CS in the form of NPs. Finally, the NPs
were isolated by centrifugation on a glycerol bed at 16,000g
for 30 min and resuspended in water. The yield of the NPs
fabrication process was calculated by weighting dried sam-
ples of isolated NPs and referring it to the initial amount of
CS and TPP.
2.2.2. Phase-solubility studies
Before the preparation of the drugHPCDCS NPs, we
studied the kinetics and dynamics of the solubilization pro-
cess of the selected drugs (TRI or FUR) in the presence of
CS and HPbCD. We selected HPbCD over HPaCD for
these analyses, because previous studies had indicated the
greater solubilizing power of the former CD derivative
towards both TRI and FUR. Phase-solubility studies were
performed by adding an excess of drug (50 mg) to 10 ml of
pH 4.6 acetate buer solution containing increasing
amounts of HPbCD (from 0 to 25 mM) in the presence
or the absence of a xed amount of CS (0.2% (w/v)), in
sealed glass containers stirred at 37 0.5 C until equilibri-
um (3 days). An aliquot was then withdrawn, ltered (pore
size 0.45 lm) and spectrophotometrically assayed for drug
concentration (k = 280 nm for TRI and k = 230 nm for
FUR) (Agilent 8453 spectrophotometer, Agilent Technolo-
gies S.L., Spain). The apparent 1:1 stability constants were
calculated from the straight line portion of the phase-solu-
bility diagrams, according to Higuchi and Connors [15].
Each test was repeated at least three times (coecient of
variation C.V.<2%). These solubility studies were per-
formed in acetate buer with a pH similar to that of CS
aqueous solution (pH 4.6) due to the fact that FUR has
a pH-dependent solubility.
2.2.3. Preparation of drug-loaded cyclodextrin-containing
CS nanoparticles
To load the selected drugs (TRI or FUR) in CS NPs, an
excess of drug was incubated under magnetic stirring for
80 F. Maestrelli et al. / European Journal of Pharmaceutics and Biopharmaceutics 63 (2006) 7986
24 h with a 0.2% (w/v) CS water solution containing dier-
ent amounts of HPCD (from 0 to 25 mM). After incuba-
tion, the drug suspensions were ltered and the resulting
solutions were assayed spectrophotometrically for drug
content. This solution of drugHPCD complexes and CS
was then used for NP formation by the ionotropic cross-
linking method as described above.
2.2.4. Characterization of cyclodextrin-containing CS
nanoparticles
Particle size and f potential of colloidal systems were
determined, respectively, by photon correlation spectrosco-
py and laser doppler anemometry, using a Zetasizer Nano
ZS (Malvern Instruments, UK).
The morphological examination of the NPs was per-
formed by transmission electron microscopy (TEM)
(CM12 Philips, Eindhoven, Netherlands) and by scanning
electron microscopy (SEM) (JEOL JSM-6400, Tokio,
Japan). The NP samples for TEM analysis were stained
with 2% (w/v) phosphotungstic acid and placed on copper
grids with Formvar

lms (SPI supplies, Inc., West Ches-

ter, PA, USA) for viewing. Nanoparticle samples for
SEM analysis were coated with gold palladium to achieve
a lm of 70-nm thickness.
The determination of the drug loading of the NPs was
made indirectly from the calculation of the non-encapsulat-
ed drug which remained dissolved in the NPs suspension
medium. The concentration of this free drug was assayed
spectrophotometrically and the dierence between this val-
ue and the theoretical drug content (total amount of drug
added for the encapsulation) was taken as the quantity
loaded into the NPs. The parameter drug loading is
shown in the form of percentage that refers to the amount
of drug entrapped in 100 mg of NPs. On the other hand,
the parameter encapsulation eciency refers to the per-
centage of drug that is entrapped with respect to the total
amount of drug added in the nanoparticle preparation
process.
Alternatively, in some cases, the amount of drug associ-
ated to the NPs after the complexation with CD (25 mM of
HPCD) was determined after enzymatic degradation of CS
with chitosanase-RD (1:2 chitosanase-RD/NPs w/w ratio,
1 h, 37 C). Following this process, the enzyme was de-ac-
tivated and precipitated by the addition of a large volume
of ethanol. Samples were then centrifuged (20,000g,
30 min) to discard the enzymatic precipitate and the
amounts of extracted drugs were determined by the stan-
dard UV method.
2.2.5. Drug release studies
For the release experiments, isolated NPs were resus-
pended in acetate buer (pH 6.0) (1.6 mg/ml of NPs) and
maintained under agitation, at 37 C. This release medium,
acetate buer at pH 6 with a low ionic strength, was select-
ed because it allows chitosan NP suspensions to remain sta-
ble for prolonged times while at the same time having an
acidity value similar to that of the intestinal tract. Samples
of the release medium were withdrawn at dierent times
(0.5, 1.5 and 4.5 h) and centrifuged at 20,000g for 30 min.
The drug released from the NPs, present in the superna-
tant, was determined by UV spectrophotometry. These
experiments were performed under sink conditions accord-
ing to the solubility of the drug in the presence of HPCD.
3. Results
3.1. Preparation and characterization of cyclodextrin-
containing CS nanoparticles
3.1.1. Preparation of blank cyclodextrin-containing CS
nanoparticles
Preliminary studies were intended to determine the
range of conditions suitable for the formation of NPs in
the presence of cyclodextrins. As a rst step, we studied
the possibility to form NPs with dierent TPP/CS ratios,
in the presence of HPCD. More specically, the volume
of the TPP solution (2 mg/ml) added to the 0.2% w/v CS
solution was changed in order to obtain TPP/CS w/w
ratios from 1:3 to 1:6 (the ratios typically used for the for-
mation of CS NPs in the absence of cyclodextrin). In the
presence of HPCD, the 1:6 TPP/CS ratio did not lead to
the formation of NPs, whereas those of 1:3 and 1:3.5 w/
w ratios gave rise to unstable particles with a marked ten-
dency to agglomerate. The optimal TPP/CS w/w ratios for
the formation of NPs in the presence of HPCD were the 1:4
and 1:5. The 1:4 w/w TPP/CS ratio was nally chosen for
further studies since it corresponded to the highest process
yield.
On the other hand, in a second step, we evaluated the
eect of the type of CS salt (hydrochloride, glutamate)
on the formation of NPs in the presence of HPCD and
using the 1:4 w/w TPP/CS ratio. Under the conditions
assayed, glutamate salt produced NPs with a poor stability,
consequently, the hydrochloride CS salt was nally selected
for further experiments.
In a third step, the inuence of CS M
w
(Table 1) on the
characteristics of the NPs, made of CS hydrochloride and
with a 1:4 w/w TPP/CS ratio, was analysed. The results
showed that the lowest molecular weight polymer led to
Table 1
Eect of the chitosan molecular weight (CS M
w
), the presence of HPbCD
and the concentration of the crosslinking agent (TPP) on the mean particle
size polydispersity (mean SD, n = 3)
CS M
w
(kDa)
HPbCD
concentration
(mM)
TPP
concentration
(mg/ml)
Size
(nm)
Polydispersity
index (PI)
110 0 1.25 484 32 0.3
110 6.29 1.25 454 19 0.3
110 0 2.0 578 1 0.3
110 6.29 2.0 590 1 0.2
272 0 2.0 887 5 0.5
272 6.29 2.0 808 0 0.5
The ratio TPP/CS was 1:4 w/w.
F. Maestrelli et al. / European Journal of Pharmaceutics and Biopharmaceutics 63 (2006) 7986 81
the formation of the smallest NPs. In addition, the incor-
poration of HPbCD (used in this rst series of experiments
at a constant concentration of 10 mg/ml, i.e., 6.29 mM) to
the aqueous CS solution did not aect markedly the size
and the f potential of the NPs. In contrast, the concentra-
tion at which TPP was added to the CS solution had an
impact on the size of the NPs. Indeed, when the concentra-
tion of the TPP solution went from 1.25 to 2 mg/ml, the
size of the NPs increased from 450 to 600700 nm.
Taking into account these preliminary results, we select-
ed the 1:4 w/w TPP/CS ratio, the low molecular weight
chitosan hydrochloride (Seacure Cl113) and the two con-
centrations of TPP (1.25 and 2 mg/ml) for further studies
aimed at investigating the eect of the HPCD type (a or
b) and concentration (from 3.14 to 25 mM) on the physico-
chemical characteristics (size, polydispersity index and f
potential) and yield of the NP formation process. The
results showed that the inclusion of the selected cyclodex-
trins inuenced the size of the resulting NPs (Tables 2
and 3). Indeed, in the case of NPs prepared with HPaCD
and a TPP concentration of 1.25 mg/ml TPP, we observed
a reduction of the size along with an increase in the cyclo-
dextrin concentration. However, this eect was not detect-
ed for the NPs prepared with the 2 mg/ml TPP; a result
that could be simply due to the broad size distribution
observed for these NPs. With regard to other physicochem-
ical parameters investigated, CS NPs prepared with HPCD
displayed f potentials in the same range as the control for-
mulations (Tables 2 and 3). Finally, it was found that the
yield of NPs production increased as the amount of HPCD
added to the chitosan solution increased.
Morphological studies of the NPs prepared in the pres-
ence of cyclodextrins were conducted by SEM and TEM
(Figs. 1 and 2). Both imaging techniques conrmed the for-
mation of spherical NPs. As is typical for CS NP, sizes
observed by electronic microscopy were far smaller than
those obtained from photon correlation spectroscopy, a
result that has been attributed to the dehydration of the
NPs in the process of sample preparation required for elec-
tron microscopy.
3.1.2. Preparation and characterization of drug-loaded
cyclodextrin-containing CS nanoparticles
As a last step, NPs loaded with drugHPCD complexes
were prepared. Two poorly soluble drugs (TRI and FUR)
were selected for these studies. These molecules were con-
sidered as prototypes of drugs whose bioavailability is
compromised by their low solubility in the GI tract
[16,17] .
In order to get a comprehensive picture of the encapsu-
lation process of these drugs in the new NPs, we considered
it important to perform phase-solubility studies with
increasing HPCD concentrations, in the presence or
absence of CS. In these studies, CS concentration was that
used for NP preparation (0.2% w/v). Additionally, because
FUR has a pH-dependent solubility, we performed all sol-
ubility studies in acetate buer, pH 4.6, which is the pH of
CS solutions used for the nanoparticle formation. For
these studies we selected HPbCD because of its favourable
solubilizing properties as compared to those of HPaCD
[18,24].
The solubility of TRI and FUR as a function of the
amount of HPbCD added to their respective dissolution
media is displayed in Figs. 3A and B. As expected, both
molecules showed a marked increase in their solubility as
the HPbCD concentration increased. However, the solubi-
Table 2
Eect of HPCD type and concentration on the characteristics of CD-containing CS nanoparticles (size, polydispersity index, f potential and NP
production yield) (concentration, 1.25 mg/ml; ratio TPP:CS, 1:4) (mean SD, n = 3)
HPCD concentration (mM) HPCD type Size (nm) Polydispersity index (PI) f potential (mV) Production yield (%)
0 484 32 0.3 +37.6 0.9 42 7
3.14 a 410 29 0.2 +36.9 0.6 45 6
b 456 37 0.3 +34.2 1.0 51 6
6.29 a 398 14 0.2 +35.9 3.8 48 7
b 454 19 0.3 +34.8 3.2 54 4
25 a 361 18 0.2 +35.8 1.7 65 9
b 443 27 0.5 +29.8 2.9 74 3
Table 3
Eect of HPCD type and concentration on the characteristics of CD-containing CS nanoparticles (size, polydispersity index, f potential and NP
production yield) (concentration of TPP, 2 mg/ml; ratio TPP:CS, 1:4) (mean SD, n = 3)
HPCD concentration (mM) HPCD type Size (nm) Polydispersity index (PI) f potential (mV) Production yield (%)
0 686 1 0.5 +33.8 3.4 36 7
3.14 a 625 4 0.6 +34.7 2.3 40 9
b 590 1 0.3 +35.3 3.8 39 5
6.29 a 645 7 0.6 +33.8 0.5 44 7
b 624 0 0.3 +36.2 0.5 41 3
25 a 690 3 0.4 +35.3 3.8 84 2
b 670 4 0.6 +33.1 3.3 71 8
82 F. Maestrelli et al. / European Journal of Pharmaceutics and Biopharmaceutics 63 (2006) 7986
lization capacity was much more important for TRI than
for FUR. Indeed, up to a 125-fold increase in TRI solubil-
ity and a 27-fold increase in FUR solubility could be
achieved using 25 mM HPbCD solutions. This solubiliza-
tion capacity of HPbCD for TRI was signicantly reduced
by the presence of CS (0.2% w/v). Contrarily, in the case of
FUR, there is an apparent increase in the solubility due to
the presence of CS. Additional dissolution kinetic studies
performed for both drugs revealed that at least 99% of
the maximum drug solubility was already reached in 24 h
(data not shown). This result was important as it allowed
us to reduce drug/HPbCD incubation time to 24 h for
the solutions intended for loaded NP preparation.
In a second step, the study was addressed to investigate
how the solubilization step into the cyclodextrin cavity
could facilitate the association of the complexed drug to
CS NPs. The results of the loading studies of TRI can be
seen in Table 4 and the loading studies of FUR in Table
5. The poor solubility of TRI in the presence of low
HPbCD concentrations (3.14 mM) limited its association
to the NPs. As the amount of HPbCD increased, the
amount of drug complexed with the cyclodextrin and the
nal loading of the NPs increased. This increase in the nal
loading was, however, accompanied by a decrease in the
drug entrapment eciency. On the other hand, it was
observed that the TRI association to the NPs in the pres-
ence of HPaCD (25 mM) was lower than that achieved
with HPbCD. This result was related to the lower complex-
ing and solubilizing abilities of HPaCD toward TRI as
compared to HPbCD [18].
Fig. 1. Transmission electron micrographs of CS NPs prepared with 25 mM of HPbCD (2 mg/ml TPP formulation, top left picture; 1.25 mg/ml TPP
formulation top right picture).
Fig. 2. Scanning electron micrographs of CS NPs prepared with 25 mM
HPbCD (2 mg/ml TPP formulation).
Fig. 3. (A) Phase solubility of Triclosan (TRI) in the presence of
increasing concentration of HPbCD with or without 0.2% w/v of chitosan
(CS). (B) Phase solubility of Furosemide (FUR) in the presence of
increasing concentration of HPbCD with or without 0.2% w/v of chitosan
(CS).
F. Maestrelli et al. / European Journal of Pharmaceutics and Biopharmaceutics 63 (2006) 7986 83
In the case of FUR, the lower solubility of the com-
plexed drug led to slightly lower drug loadings than
those observed for TRI. In addition, as noted for TRI,
the loading values obtained in the presence of HPaCD
were lower than those corresponding to HPbCD. As in
the case of TRI, this behaviour was attributed to the
reduced solubilizing properties of this cyclodextrin deriv-
ative for FUR.
The above-indicated values of loading were calculated
taking into account the amount of drug that remained free
(non-encapsulated) in the NPs suspending medium. The
accuracy of these values was conrmed by the direct deter-
mination of the loading upon enzymatic degradation of the
NPs. Indeed, this calculation, performed in the 25 mM
HPCD (a- and b-) formulations, led us to accept the accu-
racy of the values determined indirectly.
3.1.3. In vitro release experiments
The results of the drug release experiments from the
25 mM HPCD CS NPs are shown in Fig. 4. All the formu-
lations tested showed similar release proles in acetate
buer, pH 6.0. These proles are characterized by an initial
fast release phase followed by a delayed release one can
reach the plateau level (between 40 and 60% released) after
1.5 h, being this level maintained for at least 24 h. Despite
an apparent reduction in the total amount of drug released
detected for the latest data points in some formulations, the
statistical analysis conrmed that no statistical dierences
were found between 1.5, 4 and 24 h.
4. Discussion
The recent methods used in the drug discovery process
have led to an increasing number of drug candidates which
display poor solubility and, frequently, poor permeability
characteristics [4]. These inadequate biopharmaceutical
properties have often been the reason for failure of many
drug candidates [2]. In this work, our aim was to design
a new delivery approach that could, in principle, combine
the advantage of CS NPs, in terms of increasing the drug
permeability through epithelia, with the solubilizing prop-
erties of cyclodextrins. In fact, previous results from our
group with CS nanocarriers have underlined their capacity
to enhance the transmucosal transport of poorly absorb-
able drugs (e.g., peptides, proteins, etc.). This positive
behaviour has been attributed to the colloidal size of the
delivery system and also to the mucoadhesive and absorp-
tion enhancing properties of CS. An additional advantage
inherent to these NPs is their production technology. This
is an extremely mild and rapid process that permits a very
ecient encapsulation of water-soluble macromolecules
[19,14]. Hydrophobic compounds can also be entrapped
within the nanomatrix [20], however, in this case, the drug
has to be nanoprecipitated from an organic solution in the
presence of CS. As a dierent strategy for the encapsula-
tion of hydrophobic compounds, in the present work, we
attempted to nanoencapsulate the drug previously solubi-
lized in the form of drugcyclodextrin complexes.
In order to investigate the possibility of associating
cyclodextrins to CS NPs, we rst analysed the inuence
of some formulation parameters in the physicochemical
characteristics of the NPs. The results indicated that the
dierences in size and f potential of the NPs prepared in
the presence or the absence of HPCD are minor (Tables
2 and 3). From these results, one could simply deduce that
cyclodextrins do not interfere with the NPs formation pro-
cess and, also, that they are not necessarily associated to
the NPs. Nevertheless, the remarkable increase in the
NPs production yield with increasing amounts of HPCD
clearly supports the entrapment of HPCD into CS nanom-
Table 4
The eect of CD type and concentration on the solubilization of triclosan,
encapsulation eciency (EE) and the nal loading of triclosan into CS
nanoparticles (TPP concentration, 2 mg/ml; ratio TPP:CS, 1:4)
(mean SD, n = 3)
HPCD
type
HPCD
concentration
(mM)
Triclosan
solubility
(mg/L)
Triclosan
EE (%)
Triclosan
loading (%)
0 68 17 12.5 8 0.8 0.3
b 3.14 211 24 5.2 7 1.1 0.2
b 6.29 588 18 4.8 3 2.2 0.1
b 25 1120 13 5.5 3 3.1 0.1
a 25 870 59 4.6 2 2.0 0.1
Table 5
The eect of CD type and concentration on the solubilization of
furosemide, encapsulation eciency (EE) and the nal loading of
furosemide into CS nanoparticles (TPP concentration, 1.25 mg/ml; ratio
TPP:CS, 1:4) (mean SD, n = 3)
HPCD
type
HPCD
concentration
(mM)
Furosemide
solubility
(mg/L)
Furosemide
EE (%)
Furosemide
loading (%)
0 7.8 1.3 22.3 1.4 0.23 0.07
b 3.14 42.3 2.4 17.1 3.0 0.89 0.04
b 6.29 95.4 10.1 12.1 1.3 1.43 0.24
b 25 387.3 10.3 7.2 3.1 2.39 0.72
a 25 253.5 9.8 8.8 2.7 1.92 0.55
0%
10%
20%
30%
40%
50%
60%
70%
80%
0 1 2 3 4 5
Time (h)
d
e
s
a
e
l
e
R
g
u
r
D
FUR HP CD FUR HP CD
TRI HP CD TRI HP CD
Fig. 4. Release of triclosan (TRI) or furosemide (FUR) from CS
nanoparticles prepared with 25 mM of HPaCD or 25 mM of HPbCD in
acetate buer (pH 6.0) (mean SD, n = 3).
84 F. Maestrelli et al. / European Journal of Pharmaceutics and Biopharmaceutics 63 (2006) 7986
atrices (Tables 2 and 3). In addition, the fact that the size of
the NPs was hardly modied leads to the conclusion that
the presence of HPCD results in either the formation of a
greater number of particles or the formation of more com-
pact nanoparticles.
Besides the minor eect of the presence of cyclodextrins
in the size of the resulting NPs, the TEM and SEM images
do not show an apparent change in the nanoparticles
appearance (Figs. 1 and 2). This could be due to the fact
that CS NPs have, per se, a sponge-like structure, which
is justied by their formation by an ionic gelation process.
Consequently, small changes related to the entrapment of
cyclodextrins would be dicult to identify.
In order to achieve an adequate incorporation of the drug
into CS NPs, using the drugcyclodextrin complexation
approach, it was necessary to, rst, solubilize a reasonable
amount of drug due to its complexation with the cyclodex-
trin and, second, to entrap a sucient amount of complexes
into the NPs structure. The stability constant of the TRI-
HPbCD complex was quite high (K
1:1
= 8094 M
1
). Inter-
estingly, despite the inherent CS solubilizing eect towards
TRI (TRI solubility increased up to 4 times for a CS 0.2%
(w/v) solution), a decrease in both drug solubility and com-
plex stability constant (K
1:1
= 480 M
1
) was observed when
the three components (drug, HPbCD and CS) were together
in solution (Fig. 3A). Asimilar result was obtained for FUR.
Indeed, CS alone had a solubilizing eect towards FUR
(FUR solubility increased up to ve times for a CS 0.2%
(w/v) solution). In the absence of CS, the HPbCDFUR
complex displayed a reasonably high stability constant
(K
1:1
= 1641 M
1
). However, upon mixing the three compo-
nents (CS, HPbCD and FUR) we observed an apparent
slight increase in the solubility of FUR, but a dramatic
decrease in the complex stability constant (K
1:1
=
172 M
1
) (Fig. 3B). The apparent increase in drug solubility
is due to the fact that the negative eect of CS in the drug
HPbCDcomplexation is counteracted by its inherent solubi-
lizing eect. These results agree with those previously report-
ed for the complex of gliburide with HPbCD in the presence
of chitosan [25]. Nevertheless, this outcome was rather unex-
pected since previous reports have claimed the synergistic
solubilizing and complexing eects of cyclodextrins in the
presence of hydrophilic polymers [21] . At this stage, the rea-
son behind this decrease in the complex stability constant in
the presence of CS is not clear, but our hypothesis is that
there might be a competition between drug and polymer
for the cavity of the HPbCD.
Taking into account the suggested interaction between
CS and HPCDs and the higher TRI and FUR concentra-
tions in solutions achieved in the presence of cyclodextrins,
we then studied the drug encapsulation capacity of CS NPs
prepared in the presence of drugcyclodextrin complexes.
The results in Tables 4 and 5 show that as the amount of
drug increased in parallel with the amount of cyclodextrins,
the encapsulation eciency decreased (from 12% to 4.5%
for TRI and from 22% to 7% for FUR). However, the nal
loading increased up to the values of 3.1 and 2.4, respec-
tively, for TRI and FUR complexed to HPbCD. In addi-
tion, it can be noted that the drug loading of the NPs
was signicantly increased (up to 10 times in the case of
FUR and up to four times in the case of TRI) in the pres-
ence of cyclodextrins due to their solubilizing properties.
Therefore, from these results we could conclude that, the
use of cyclodextrins permits to enhance the encapsulation
eciency of poorly soluble drugs to CS NPs and that, as
the solubility of the complexed drug increases, the loading
of HPCD-containing CS NPs raises.
Finally, all the formulations tested displayed very simi-
lar release proles independent of the drug loaded (FUR
or TRI) and also irrespective of the type of HPCD used
(a or b). The typical prole is characterized by an initial
rapid release phase that reached a plateau in a relatively
short time (1.5 h). These results suggest that a certain
amount of drugcyclodextrin complexes can be easily
released by simple diusion through the polymer network
as it has been the case for peptide molecules [22]. However,
there is an additional amount of complex that cannot be
released from the particles unless the polymer matrix is
degraded. In fact, as noted in Section 3, the total amount
of the encapsulated drug could be recovered upon incuba-
tion of the particles in the presence of enzyme chitosanase.
From these results we could speculate that, following oral
administration, some of the encapsulated drug molecules
will be rapidly released from the NPs whereas some will
remain associated to the NPs and interact with the mucosa
[22]. At this level, the degradation of the particles, probably
driven by the enzyme lysozyme [23], could facilitate the
delivery of the associated drug at the absorption site.
5. Conclusions
This paper reports, for the rst time, the possibility to
entrap cyclodextrins within CS nanoparticles using a very
simple ionic gelation technique. This new approach permits
to enhance the entrapment of hydrophobic drugs by form-
ing molecular inclusion complexes with cyclodextrins in
aqueous media. Such a device could be of interest for
increasing the absorption of poorly soluble and poorly per-
meable drugs and also for conferring a protection to some
specic drug molecules through the complexation followed
by entrapment in the polymer matrix.
Acknowledgment
This work was supported by the Spanish Government
(CICYT: SAF2003-08765-C03-03).
References
[1] G. Klopman, L.R. Stefan, R.D. Saiakhov, ADME evaluation 2. A
computer model for the prediction of intestinal absorption in humans,
Eur. J. Pharm. Sci. 17 (2002) 253263.
[2] A.M. Palmer, New horizons in drug metabolism, pharmacokinetics
and drug discovery, Drug News Perspect. 16 (2003) 5762.
F. Maestrelli et al. / European Journal of Pharmaceutics and Biopharmaceutics 63 (2006) 7986 85
[3] T.L. Rogers, I.B. Gillespie, J.E. Hitt, K.L. Fransen, C.A. Crowl, C.J.
Tucker, G.B. Kupperblatt, J.N. Becker, D.L. Wilson, C. Todd, C.F.
Broomall, J.C. Evans, E.J. Elder, Development and characterization
of a scalable controlled precipitation process to enhance the disso-
lution of poorly water-soluble drugs, Pharm. Res. 21 (2004) 2048
2057.
[4] A. Malkia, L. Murtomaki, A. Urtti, K. Kontturi, Drug perme-
ation in biomembranes In vitro and in silico prediction and
inuence of physicochemical properties, Eur. J. Pharm. Sci. 23
(2004) 1347.
[5] FDA, Guidance for Industry, Waiver of in vivo bioavailability and
bioequivalence studies for immediate-release solid oral dosage forms
based on a biopharmaceutics classication system, Food & Drug
Administration, Rockville, MD, 2000.
[6] R. Lobenberg, G.L. Amidon, Modern bioavailability, bioequivalence
and biopharmaceutics classication system. New scientic approaches
to international regulatory standards, Eur. J. Pharm. Biopharm. 50
(2000) 312.
[7] R. Fernandez-Urrusuno, D. Romani, P. Calvo, J.L. Vila-Jato, M.J.
Alonso, Development of a freeze-dried formulation of insulin-loaded
chitosan nanoparticles intended for nasal administration, S.T.P.
Pharma Sci. 9 (1999) 429436.
[8] M. Garcia-Fuentes, C. Prego, D. Torres, M.J. Alonso, A comparative
study of the potential of solid triglyceride nanostructures coated with
chitosan or poly(ethylene glycol) as carriers for oral calcitonin
delivery, Eur. J. Pharm. Sci. 25 (2005) 133143.
[9] A. Vila, A. Sanchez, M. Tob o, P. Calvo, M.J. Alonso, Design of
biodedegradable particles for protein delivery, J. Control. Release 78
(2002) 1524.
[10] P. Yan, L. Ying-jian, Z. Hiu-ying, Z. Jun-min, X. Hui, W. Gang, H.
Hin-song, C. Fu-de, Bioadhesive polysaccharide in protein delivery
system: chitosan nanoparticles improve the intestinal absorption of
insulin in vivo, Int. J. Pharm. 249 (2002) 139147.
[11] H. Takeuchi, H. Yamamoto, Y. Kawashima, Mucoadhesive nano-
particulate systems for peptide drug delivery, Adv. Drug Deliv. Rev.
47 (2001) 3954.
[12] E.M. Martin Del Valle, Cyclodextrins and their uses: a review,
Process. Biochem. 39 (2004) 10331046.
[13] V.J. Stella, R.A. Rajewski, Cyclodextrins: their future in drug
formulation and delivery, Pharm. Res. 14 (1997) 556567.
[14] P. Calvo, C. Remun an-Lo pez, J.L. Vila-Jato, M.J. Alonso, Novel
hydrophilic chitosanpolyethylene oxide nanoparticles as protein
carriers, J. Appl. Polym. Sci. 63 (1997) 125132.
[15] T. Higuchi, K.A. Connors, Phase-solubility techniques, Adv. Anal.
Chem. Instr. 4 (1965) 117212.
[16] T. Loftsson, N. Leeves, B. Bjornsdottir, L. Duy, M. Masson, Eect
of cyclodextrins and polymers on triclosan availability and substan-
tivity in toothpastes in vivo, J. Pharm. Sci. 88 (1999) 12541258.
[17] J.V. Turner, D.J. Maddalena, S. Agatonovic-Kustrin, Bioavailability
prediction based on molecular structure for a diverse series of drugs,
Pharm. Res. 21 (2004) 6882.
[18] M.D. Veiga, M. Merino, M. Cirri, F. Maestrelli, P. Mura, Compar-
ative study on triclosan interactions in solution and in the solid state
with natural and chemically modied cyclodextrin, J. Incl. Phenom.
Macrocycl. Chem. 53 (2005) 7783.
[19] I. Behrens, A.I. Vila Pena, M.J. Alonso, T. Kissel, Comparative
uptake studies of bioadhesive and non-bioadhesive nanoparticles in
human intestinal cell lines and rats: the eect of mucus on particle
adsorption and transport, Pharm. Res. 19 (2002) 11851193.
[20] A.M. De Campos, A. Sanchez, M.J. Alonso, Chitosan nanoparticles:
a new vehicle for the improvement of the delivery of drugs to the
ocular surface. Application to cyclosporin A, Int. J. Pharm. 224
(2001) 159168.
[21] T. Loftsson, H. Frioriksdottir, A.M. Siguroardottir, H. Ueda, The
eect of water-soluble polymers on drugcyclodextrin complexation,
Int. J. Pharm. 110 (1994) 169177.
[22] R. Fernandez-Urrusuno, P. Calvo, C. Remun an-Lo pez, J.L. Vila-
Jato, M.J. Alonso, Enhancement of nasal absorption of insulin using
chitosan nanoparticles, Pharm. Res. 16 (1999) 15761581.
[23] H. Sashiwa, H. Saimoto, Y. Shigemasa, Lysozyme susceptibility of
partially deacetylated chitin, Int. J. Biol. Macromol. 12 (1990) 295
296.
[24] E. Spamer, D.G. Mu ller, P.L. Wessels, J.P. Venter, Characterization
of the complexes of furosemide with 2-hydroxypropyl-b-cyclodextrin
and sulfobutyl ether-7-b-cyclodextrin, Eur. J. Pharm. Sci. 16 (2002)
247253.
[25] N. Zerrouk, G. Corti, S. Ancillotti, F. Maestrelli, M. Cirri, P. Mura,
Inuence of cyclodextrins and chitosan, separately or in combination,
on glyburide solubility and permeability, Eur. J. Pharm. Biopharm.
(in press, available on-line: doi:10.1016/j.ejpb.2005.08.010).
86 F. Maestrelli et al. / European Journal of Pharmaceutics and Biopharmaceutics 63 (2006) 7986