Enhancement of Drug Solubility in Supramolecular

and Colloidal Systems

ANDRÁS SÜLE,1 LAJOS SZENTE,2 FERENC CSEMPESZ1

1
Laboratory of Colloid and Supramolecular Systems, Eötvös University, Institute of Chemistry, P.O. Box 32,
H-1518 Budapest 112, Hungary
2
Cyclolab Cyclodextrin Research & Development Ltd, P.O. Box 435, H-1525 Budapest, Hungary

Received 20 December 2007; revised 31 March 2008; accepted 13 April 2008

Published online 27 May 2008 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.21437

ABSTRACT: Statins, as efficient HMG-CoA (3-hydroxy-3-methylglutaryl-coenzyme-A)

reductase inhibitors are widely used in the management of cardiovascular diseases.
Interactions in aqueous solutions between highly lipophilic statins and cyclodextrins
(CDs) in the absence and the presence of a dissolved polymer or its monomeric
compound, respectively, were studied. The solubility of lovastatin and simvastatin at
various temperatures and pHs were investigated by phase-solubility measurements.
Surface activity of solutes in binary (CD–statin) and ternary (CD–statin–polymer)
systems was studied by determining the surface tension of the solutions. For the
characterization of the CD–statin inclusion complexes, stability constants for associates
of different molar ratios have been calculated. It was shown that complexation may lead
to improvement of the aqueous solubilities of both statins by 1–2 orders of magnitude.
Especially, randomly methylated b-cyclodextrin (RAMEB) showed outstanding solubi-
lizing effects. In binary systems dominantly CD–statin associates of 1:1 molar ratios
form, which exhibit considerable surface activity. RAMEB forms more stable complexes
with these drugs than the native b-CD, and also the surface activity of the former solutes
is higher. In polymer-containing ternary systems the solubility of both statins could be
further improved. The enhanced drug solubilities can be ascribed to the formation of
CD–statin–polymer associates with supramolecular structure. A portion of the surface
active CD–statin complexes are very likely anchored at the macromolecular chains. In
these solutions, the total amounts of solutes are composed of the sum of the ‘‘free’’ binary
and the supramolecular ternary associates. ß 2008 Wiley-Liss, Inc. and the American
Pharmacists Association J Pharm Sci 98:484–494, 2009
Keywords: cyclodextrins; dissolution; inclusion compounds; supramolecular associ-
ates; nanocapsules; macromolecular drug delivery

INTRODUCTION growing amount of papers pointed out that statins

Improvement of the efficacy and safety of existing
drugs is a key issue in pharmaceutical sciences
and the related fields. Over the past few years, a
Correspondence to: Ferenc Csempesz (Telephone: þ36-1-
372-2544; Fax: þ36-1-372-2592; E-mail: csf@chem.elte.hu)
Journal of Pharmaceutical Sciences, Vol. 98, 484–494 (2009)
ß 2008 Wiley-Liss, Inc. and the American Pharmacists Association
provide crucial benefits when used in either lipid
lowering therapy or coronary heart disease (CHD)
therapy. It is well documented that statins inhibit
the rate-limiting steps in cholesterol synthesis.
The 3-hydroxy-3-methylglutaryl-like domain of
statins may attach to the appropriate binding site
of the reductase enzyme and therefore, sterically
prevent the natural substrates to bind.1 Sche-
matic molecular structure of two statins is
illustrated in Figure 1.

484 JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009


Figure 1. Chemical structure of lovastatin and sim-
vastatin.
Chemically different statins (such as lovastatin,
simvastatin, pravastatin, atorvastatin) differ
mostly in their absorption parameters, plasma
protein binding and solubility, therefore they may
exhibit varied efficacy in reducing the low-density
lipoprotein (LDL) cholesterol levels even at the
same dosages. In general, by doubling the statin
dose, LDL-cholesterol level is reduced by more
than 7%.2 Nevertheless, 22–45% reduction in the
blood triglyceride levels of hypertriglyceridemic
patients and a minor elevation in high-density
lipoprotein (HDL) cholesterol levels could also be
observed.
A hard obstacle in statin therapy is the very low
aqueous solubility of the pharmacons. Their
bioavailability is, therefore, also low and may
exhibit high variability in individuals.3,4 In
addition, the HMG-CoA reductase inhibitors are
associated with two uncommon, but important
side effects: asymptomatic elevation of liver
enzymes and skeletal muscle abnormalities.
These latter can range from benign muscle pain
to myopathy (severe muscle pain and weakness)
and life-threatening rhabdomyolysis.5,6 Despite of
these shortcomings, monotherapy with either
lovastatin or simvastatin is proven to be safe,
but concerning their interactions in complex
delivery systems, many questions may arise.7,8
Both statin molecules are known to have a ring-
closed lactone, and ring-opened hydroxyacid form,
from which the latter is the active substance.9,10
The ring-closed prodrugs are metabolized to their
respective acidic form in liver cells. It is possible
however that hydrolysis might occur in aqueous
media that affects their biopharmacological prop-
erties too. During drug formulation it is, there-
fore, crucial to ensure the appropriate chemical
stability of statins.
Cyclodextrins (CDs) are peculiar cyclic oligo-
mers of a-D-glucose and are extensively used as
complexing agents for lipophilic compounds. Their
shapes resemble truncated cones with primary
DOI 10.1002/jps
ENHANCEMENT OF DRUG SOLUBILITY 485
Figure 2. (a) Chemical structure of alpha-cyclodex-
trin. (b) Computer-simulated surface model of alpha-
cyclodextrin.12
and secondary hydroxyl groups located around
their narrower and wider rim, respectively.11 The
most widely used natural cyclodextrins consist of
6, 7, and 8 glucose units and according to the
number of monomers in the macrocylce, they are
named as a-, b-, and g-cyclodextrin. It is well
known that CD molecules have a hydrophobic
inner cavity, while the large number of hydroxyl
groups on the outer surface makes them water-
soluble (Fig. 2).
Due to this special molecular structure, lipo-
philic guest molecules (or molecule parts) of the
appropriate size can be encapsulated in the
cavities of CD molecules, hereby forming inclusion
complexes.13 The chemical structure of CDs can be
expediently modified by esterification of the 2-, 3-,
and 6-hydroxyl groups and also, other functional
groups can be introduced at these sites. For
instance, these structural changes perturb the
internal hydrogen-bond network of b-CDs, result-
ing in an elevated water-solubility.14
Extensive studies showed that more than 40%
of developed drug molecules exhibit poor bio-
pharmacological properties, partly because of
their diminished dissolution and/or inadequate
permeability.15 A great challenge for the present
pharmaceutical research is, therefore, to formu-
late orally administrable dosage forms with
decent bioavailability of pharmacons. To achieve
this, the drug must be dissolved in the gastro-
intestinal tract and it must be absorbed in a way
that sufficient drug levels arise at the pharmaco-
logically active sites.
To date, only little information is available on
the relevance of colloids in controlling physico-
chemical and pharmacological behavior of drugs
in complex delivery systems. Much less is known
about how colloidal components may affect non-
covalent interactions of pharmacons in aqueous
media. Possible use of water-soluble polymers to
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
486 SÜLE, SZENTE, AND CSEMPESZ
improve inclusion properties of CDs is well
documented by Loftsson et al.16 Nevertheless,
the effect of colloidal additives on the complex
formation between cyclodextrins and pharma-
cons, and interactions of colloids with the formed
supramolecules in complex drug delivery systems
has not been revealed in detail.
Complexation of lovastatin and simvastatin
with cyclodextrins, and the effect of water-soluble
macromolecules and their noncolloidal chemical
analogues on the formation of supramolecular
complexes are in the focus of the present study.
Solubilization of statins in aqueous cyclodextrin
solutions and its control in the absence and the
presence of a macromolecular colloid and its
monomeric compound, respectively, have been
investigated. Possible ways of enhancing the
aqueous solubility of the drugs in both binary
and ternary systems have also been discussed.
MATERIALS AND METHODS
Materials
Lovastatin and simvastatin of USP23-grade
(Chiesi Farmaceutici SpA, Parma, Italy), a-, b-,
g-cyclodextrin and RAMEB [randomly methylated
b-cyclodextrin, average degree of methylation: 1.8
methyl groups/glucose monomers], as well as PVP
K30 polyvinyl pyrrolidone (Fluka AG, Buchs SG,
Switzerland) and NEP [1-Ethyl-2-pyrrolidinone]
(Sigma-Aldrich Chemie GmbH, Munich, Germany)
were used in the experiments. The cyclodextrins
were of analytical grade and manufactured by
Cyclolab R&D Ltd (Budapest, Hungary).
Solubility Measurements
Phase-solubility studies of the statins in aqueous
solutions of several cyclodextrins of different
chemical structures were carried out according
to the Higuchi–Connors procedure.17 Various
amounts of CD were generally dissolved in dis-
tilled water and lovastatin or simvastatin was
added to the solutions in vast excess (see Tab. 1).
Twenty-four hours of temperature-programmed
incubation period in a refrigerated bath (HAAKE
PhoenixII C35P instrument) was ensured for the
dissolution of the pharmacon at 14, 25, 36, and
478C, respectively. After longer equilibration
times (3–14 days), no definite increase in drug
solubilities could be detected. After removing the
nondissolved statin from the solutions with
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
Table 1. Ranges of Additive Concentrations in
Different Measurements
Concentration
Measurement Component Range (mg/mL)
Phase solubility a-, b-, g-CD, 0–50
(binary systems) RAMEB
Phase solubility RAMEB 0–12
(ternary systems) PVP 0–0.20
NEP 0–0.10
Surface tension b-CD, RAMEB 0–10
PVP 0.20
Millipore type GS 0.22 mm membrane filter, the
absorption spectra of the dissolved pharmacon
were then taken in the range of 190–400 nm
wavelengths. The highly lipophilic character of
the molecules of both drugs might raise the ques-
tion of possible drug adsorption to the membrane
filters. Measuring the statin content of both
ethanolic and aqueous solutions throughout 10
consecutive filtration cycles proved that no sig-
nificant adsorption occurs during the separation
procedures.
The statin content of the solutions was deter-
mined with a UV-spectrophotometric assay based
on the PhEur directives.18 The absorbance of
lovastatin and simvastatin was measured at 236
and 240 nm, respectively. For these measure-
ments a computer-controlled spectrophotometer
(Perkin-Elmer Lambda Series 2S instrument) was
used and the raw spectra were evaluated by
OriginLab Origin 7.0 software suite. The solubi-
lity isotherms were calculated by using calibra-
tion curves determined with ethanolic statin
solutions of known concentrations. It is worth-
while to mention that compared to aqueous and
CD-containing matrices, the ethanol did not
distort either the UV-spectra of the drugs, or
the response factor. Therefore, throughout these
measurements no use of correction factors was
necessary.19 As noted in the Introduction Section,
preserving the chemical stability of statins is a
crucial factor in these experiments. In aqueous
solutions, no changes in the molecular structures
of the drugs could be detected either by pre-
liminary conductometric, circular dichroism, or by
NMR measurements.
The solubility of statins was also investigated
at two moderate pH values, relevant to that of
the gastro-intestinal tract. To adjust the pH of
the solutions, HCl (pH ¼ 1.2, cHCl ¼ 0.08 M,
cNaCl ¼ 0.034 M) solution and phosphate buffer
(pH ¼ 7.4, c ¼ 0.25 mM), was used, respectively.
DOI 10.1002/jps

The buffers were prepared on the basis of the
directives of the European Pharmacopoeia’s
5th Edition and the Hungarian Pharmacopoeia’s
8th Edition.18,20
For the preparation of statin–CD–polymer
ternary systems, a specially controlled thermal
program was used in a similar way described by
Loftsson and Masson.21 The appropriate mixtures
(listed in Tab. 1) were put in a computer-driven
refrigerated bath, where the samples were heated
up to 708C and held at this elevated temperature
for 2 h. After that, the solutions were cooled down
to 258C and kept at this temperature for 22 h. UV-
spectra of solutions indicated that such a rapid
thermal treatment does not cause alterations in
the molecular structure of the statins. Seventy
degree Celsius was found as an optimum tem-
perature where the thermal degradation of
components could be avoided and also, after the
incubation period at this temperature sufficient
solubility enhancing effect could be detected (for
more details see Tab. 4).
Surface Tension Determination
Surface tension of aqueous solutions was mea-
sured by computer controlled Wilhelmy-plate
b11 or b12 ¼ K12
K11).
method, using a KSV Sigma 70 instrument.
In each measurement, the shown data are
the averages of at least three independent
experiments. In addition, the spectrophotometric
assays have been validated by measurements
of 15 samples of the same composition, and
standard deviation of the experimental data has
been accordingly calculated at several additive
concentrations.
Theoretical Approach to the
CD–Statin Association
Interactions that may lead to formation of
inclusion complexes between guest (G) and host
(H) molecules can be represented as
pG þ qH , Gp Hq (1)
where p and q indicate the stoichiometric factors.
The stability constant (b) for any associate can be
defined in the following way:22,23
½Gp Hq 
bpq ¼ (2)
ð½Gp  ½Hq Þ
A computer program based on the rule of mass
balance provides evaluation of different bpq
DOI 10.1002/jps
ENHANCEMENT OF DRUG SOLUBILITY 487
constants, using the total concentrations of the
reactants:
cG ¼ Spbpq ½Gp ½Hq (3)
cH ¼ Sqbpq ½Gp ½Hq (4)
where b10 and b01 values are equal to 1.
Knowing the total concentrations (cG, cH) and
assuming realistic bpq values, [G] and [H] can be
determined in an iterative procedure. Recalculat-
ing the values of total concentrations (c), the
minimum of the next equation represents the best
fit for the iterative process:
U ¼ Sðccalculated cmeasured Þ2 (5)
Assuming possible molecular structures of com-
plexes that may really form in statin–CD binary
systems, two combinations may have relevance:
(a) 1:1 complex is formed, (b) 1:1 and 1:2 (statin/
CD) complexes are both formed. The stability
constants of 1:1 statin–CD complexes are denoted
by b11 (usually denoted by K11). The cumulative
stability constants of 1:2 stain–CD complexes are
denoted by b12. K12 —is the stepwise stability
constant for the complexation step where a second
CD molecule is bound to a 1:1 statin–CD complex
(b12 ¼ K12
Accordingly, Eqs. (3) and (4) can be simplified as
follows:
cG ¼ ½G þ b11 ½G½H (3a)
cH ¼ ½H þ b11 ½G½H (4a)
cG ¼ ½G þ b11 ½G½H þ b12 ½G½H2 (3b)
cH ¼ ½H þ b11 ½G½H þ 2b12 ½G½H2 (4b)
RESULTS AND DISCUSSION
Solubility of Pharmacons in Aqueous Media
Aqueous solubility of pharmacologically active
substances is a key parameter regarding their
biological availability. The solubility of both
lovastatin and simvastatin in water is extremely
poor (Slovastatin ¼ 1.5 mg/mL, Ssimvastatin ¼ 2.0 mg/
mL). Improving the aqueous solubility of these
pharmacons might offer, therefore, considerable
technological benefits without altering their
pharmacodynamic and metabolic (first-pass meta-
bolism above all) behavior. In Figures 3a and b
and 4 typical solubility isotherms determined for
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
488 SÜLE, SZENTE, AND CSEMPESZ
Figure 3. (a) Phase-solubility curves for lovastatin in
cyclodextrin solutions, T ¼ 258C, pH ffi 6.5. (b) Phase-
solubility curves for simvastatin in cyclodextrin solu-
tions, T ¼ 258C, pH ffi 6.5.
lovastatin and simvastatin in several CD solu-
tions at 258C are shown. The marked points
and the corresponding error bars in the plots
indicate the mean and the standard deviation,
respectively, of the experimental data.
These results well demonstrate that in solutions
of the cyclodextrins the solubility of both statins is
significantly increased. At higher CD concentra-
tions more than 10-fold increase in the solubility
of both pharmacons could be detected. Especially,
Figure 4. Phase-solubility curves for lovastatin and
simvastatin in RAMEB solutions, T ¼ 258C, pH ffi 6.5.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
RAMEB proved to be effective solubilizer for these
drugs, enhancing their solubilities by two orders
of magnitude. The observed improvement of drug
solubilities can presumably be attributed to the
formation of inclusion complexes between the CD
and the statin molecules.14 The linear sections of
the isotherms support this notion. In the range of
CD concentrations up to about 20 mg/mL, a 1:1
complexation is sufficient to explain the observed
enhancement in solubility. It should be noted here
that b-CD itself has a rather low aqueous solu-
bility and therefore, the solubility of the assumed
b-CD–statin complexes can also be limited.24
Effect of Temperature and pH on Drug Solubilities
Complexation of statins with CD in aqueous
media was investigated at 14, 25, 36, and 478C.
At higher temperatures and neutral pH, the
aqueous solubilities of the natural statins only
slightly increased, but in cyclodextrin solutions
these were considerably enhanced. Data in
Figure 5a–d show that raising the dissolution
temperature from 14 to 478C results in more than
threefold increase in the solubility of both statins.
Somewhat less enhancement of the lovastatin
solubility could be observed. The solubility of
simvastatin in several CD-solutions could also be
improved.25
The apparent increase in the solubilities of
complexed drugs can be ascribed to the improved
intrinsic solubilities of the uncomplexed pharma-
cons, as well. While the complexation constants
tend to decrease with increasing temperature
(Tab. 2), the increased intrinsic drug solubilities
can outweigh the modest decrease of the corre-
sponding stability constants (see also Eq. 2).26,27
At physiologically relevant acidic and basic
pH-values, both statin-derivatives exhibited some-
what higher solubilities than in distilled water. In
CD-solutions the drug solubilities could be further
enhanced. Figure 6a and b illustrates that the
solubility of both lovastatin and simvastatin in
acidic solutions is increased by 40–100% in a wide
concentration range of b-CD, relative to the
corresponding values obtained in distilled water.
To evaluate these results, some structural
change of the dissolved statin molecules that
may take place in acidic media should be taken
into account. It is reasonable to assume that at
low pH the inactive lactone form of statins can
be transformed into active acidic forms.9 Such
a structural alteration may lead to a less
DOI 10.1002/jps

hydrophobic form of statins, which may certainly
cause increases in the drug solubilities.
Higher enhancement in the solubilities of both
statins were detected in buffered media at
DOI 10.1002/jps
ENHANCEMENT OF DRUG SOLUBILITY 489
pH ¼ 7.4, which corresponds to the basicity of
the intestinal tract. As can be seen in
Figure 6a and b, at this pH the solubility of both
lovastatin and simvastatin is enhanced by mostly
100–150%, compared to those of their solubilities
measured in aqueous CD-solutions. Hydrolysis
and ionization of the lactone of statins at this pH,
and their enhanced complexation with CD may
alike contribute to the improvement of drug
solubilities.28
The results of phase solubility measurements
provide solid experimental evidences for the
complexation of statin molecules with cyclodex-
trins. The linear parts of the solubility isotherms
suggest that at low CD concentrations in aqueous
solutions presumably host-guest type inclusion
complexes of 1:1 molar ratios form, but consider-
ing the chemical structure of the statins and the
complexation preference of CDs,14 development of
other assemblies cannot be excluded. A possible
site of inclusion of statin with CD is schemati-
cally illustrated in Figure 7. (Formation of such
associates are supported by preliminary NMR
measurements.)
Considering the theoretical approaches22 dis-
cussed in the section of Materials and Methods,
stability constants for the statin–CD complexes
both of 1:1 and 1:2 molar ratios have been calculat-
ed. The b11 and b12 association constants calculat-
ed by iteration procedure are shown in Table 2.
These results indicate that formation of statin–
CD complexes of 1:2 molar ratios is less likely than
that of 1:1 associates. The K12 stepwise association
constants are lower than the corresponding b11
(K11) values. Also, the RAMEB-statin complexes
of 1:1 molar ratio are more stable than those of the
native b-CD. The appropriate stability constants
for the RAMEB-statin associates are about one
order of magnitude higher.
For comparison purposes, stability constants
have also been calculated directly from the solu-
bility curves using an approach offered by Iga
et al.29
m1
K11 ¼ (6)
S0 ð1 m1 Þ
Figure 5. (a) Effect of temperature on the aqueous
solubility of lovastatin in b-CD solutions, pH ffi 6.5. (b)
Effect of temperature on the aqueous solubility of lovas-
tatin in RAMEB solutions, pH ffi 6.5. (c) Effect of tem-
perature on the aqueous solubility of simvastatin in b-
CD solutions, pH ffi 6.5. (d) Effect of temperature on the
aqueous solubility of simvastatin in RAMEB solutions,
pH ffi 6.5.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
490 SÜLE, SZENTE, AND CSEMPESZ

Table 2. Stability Constants of Statin–CD Complexes Determined by Iterative Computational Method

Temperature
(8C) Binary System b11 (M 1) K12 (M 1) b12 (M 2)
25 b-CD–lovastatin 1.86  102  2.2  101 1.13  102  1.5  101 2.1  104  3.74  103
36 b-CD–lovastatin 1.75  102  2.1  101 8.46  101  1.2  101 1.48  104  2.75  103
25 b-CD–simvastatin 2.88  102  3.2  101 1.61  102  2.1  101 4.63  104  7.94  103
36 b-CD–simvastatin 2.59  102  2.8  101 1.60  102  2.1  101 4.14  104  7.05  103
25 RAMEB–lovastatin 9.38  102  9.3  101 2.63  101  3.7  100 2.47  104  4.25  103
36 RAMEB–lovastatin 9.00  102  8.8  101 3.01  101  5.2  100 2.71  104  5.38  103
25 RAMEB–simvastatin 1.21  103  1.1  102 2.38  102  2.9  101 2.88  105  4.38  104
36 RAMEB–Simvastatin 7.41  102  7.3  101
where K11 is the stability constant of 1:1
assemblies, m1 is the slope of the linear section
of the solubility curve, S0 its intersection with the
ordinate axis.
The K11 constants calculated from the solubility
isotherms according to Eq. 6 are shown in Table 3.
Estimation of the association constants in such
a way provides somewhat higher values for K11,
but these data are in line with the corresponding
values listed in Table 2. No information can be
obtained, however, from these results concerning
Figure 6. (a) Aqueous solubility of lovastatin in b-CD
solutions at pH 1.2 and pH 7.4, T ¼ 258C. (b) Aqueous
solubility of simvastatin in b-CD solutions at pH 1.2 and
pH 7.4, T ¼ 258C.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
6.10  101  8.6  100 4.52  104  7.77  103
the molecular mechanism of drug inclusion in
greater details.
Complexation in Ternary Systems
In ternary systems, the effects of a water-soluble
polymer and a chemically analogous compound of
its monomer, respectively, on the solubilization of
statins by CDs were studied. By adding a third
component to a CD–statin solution, in principle a
conventional competition between the pharmacon
and the additive for the CD cavities may take
place. Such competition may result in a definite
decrease in the amount of the complexed statin.
Changes in the solubilities of statins in CD-
solutions due to the addition of 1-ethyl-2-pyrro-
lidinone (NEP) are illustrated in Figure 8. As
can be seen, the added NEP may notably reduce
drug solubility, presumably due to its preferred
complexation in CD cavities. The higher is the
concentration of the small molecular mass addi-
tive at constant CD concentration, the higher is
the reduction in the solubility of the drug.
These results confirm a distinct competitive
mechanism in the binary interactions between the
CD, the statin and the NEP molecules, res-
pectively. Complexation of the drug with CD is
Figure 7. A possible inclusion complexation of sta-
tins with cyclodextrins.
DOI 10.1002/jps
 ENHANCEMENT OF DRUG SOLUBILITY 491

Table 3. Stability Constants for 1:1 CD–Statin

Complexes Determined from the Solubility Isotherms

Temperature
(8C) Binary System K11 (M 1)
25 b-CD–lovastatin 1.9  102  3.5  101
25 b-CD–simvastatin 2.1  102  3.8  101
25 RAMEB–lovastatin 1.5  103  2.1  102
25 RAMEB–simvastatin 1.7  103  2.4  102
hindered, and/or the pharmacon is displaced from
the cavity of cyclodextrins by the preferred NEP
molecules.
In polymer-containing systems, just an opposite
effect of the macromolecular colloid on the solu-
bility of statins could be observed. Figure 9 well
demonstrates that in solutions of CD–drug–
polyvinyl pyrrolidone, the solubility of both
statins, under the specified experimental condi-
tions, definitely increased.
At the studied CD-concentrations, the dissolved
polymer resulted in 5–50% increase in the solu-
bility of lovastatin. Based on these solubility
isotherms, stability constants for polymer con-
taining systems could be estimated according to
the method of Iga et al.29 This method was
originally proposed for host-quest type binary
systems, so estimation of cumulative stability
constants for three component systems can
only be regarded as a rough approximation.
Nevertheless, for comparison purposes, cumula-
tive stability constant for a ternary system of
lovastatin-RAMEB-PVP at 0.20 mg/mL polymer
concentration has also been estimated. Its
approximated value is as high as K ¼ 6.5  103.
As noted in the section of Materials and Methods,
these complex systems were subjected to a heat
treatment as described in the experimental
methods. Drug solubilities observed in such
Figure 8. Effect of NEP on lovastatin solubility in
RAMEB solutions, T ¼ 708C/258C, pH ffi 6.5.
DOI 10.1002/jps
Figure 9. Effect of PVP on lovastatin solubility in
RAMEB solutions, T ¼ 708C/258C, pH ffi 6.5.
systems presumably cannot be regarded as
real equilibrium solubilities, rather as solubilities
representing ‘‘frozen’’ metastable states. The
polymer-containing samples stored at 258C for
24 h showed no significant increase in the
solubility of the statins. Heat treatment proved
to be a crucial step, since the drug solubility could
be enhanced only when a short ‘‘thermal jump’’ in
the storage regimen was implied. In the studied
temperature range, the increase in the solubility
of the drug seems to be proportional to the extent
of ‘‘heat treatment’’ (Tab. 4). Without polymer,
such enhance of the drug solubilites could not be
observed.
These results clearly indicate that under
suitable conditions, the dissolved macromolecules
may significantly affect complexation of the drug
with CD, but in a different way from that as the
monomeric additives act.
Loftsson et al.30 pointed out first that water-
soluble polymers can improve the dissolution of
certain guest molecules in cyclodextrin solutions
through complex formation. The enhancement of
solubility was attributed to the formation of ter-
nary associates. Possible increase of the solubi-
lizing effect of aqueous surfactant solutions by
water-soluble polymers was also shown.31 ‘‘Simple
adsorption’’ of individual surfactant molecules on
the polymer chains, or formation of ‘‘polymer-
aggregate complexes,’’ that is, complex formation
between the micelles and the polymer chains were
supposed as likely outcomes of the surfactant-
polymer interactions.32 Detailed mechanism of
the formation and structure of the ternary
complexes has not been revealed.
For a better understanding of the relevant
interactions in ternary systems of CD, statin and
PVP, surface activities of the solutes in aqueous
CD–statin solutions with and without the polymer
have also been studied.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
492 SÜLE, SZENTE, AND CSEMPESZ

Table 4. Effect of Heat Treatment on the Solubility of Lovastatin in Ternary Systems

Change in Drug Change in Drug Change in Drug

Change in Drug Solubility, Solubility, Solubility,
PVP Solubility, 2.0 mg/mL 2.7 mg/mL 4.0 mg/mL
Peak Temp. Concentration Distilled RAMEB RAMEB RAMEB
(8C) (mg/mL) Water (%) Solution (%) Solution (%) Solution (%)
25 0.05 <1 <1 1.6  0.5 0.9  0.7
0.10 <1 <1 3.2  0.5 4.1  0.7
0.20 <1 <1 2.7  0.5 3.7  0.7
47 0.05 <1 1.6  0.3 2.2  0.7 1.1  0.9
0.10 <1 3.5  0.3 7.6  0.7 2.4  0.9
0.20 1.2  0.1 4.2  0.3 9.6  0.7 5.1  0.9
70 0.05 2.5  0.2 4.2  0.5 13.3  0.8 42.1  1.1
0.10 3.1  0.2 7.0  0.5 15.9  0.8 40.5  1.1
0.20 3.8  0.2
In Figure 10, the surface tensions of aqueous
CD solutions and CD–simvastatin solutions,
respectively are plotted as a function of the CD
concentration.
Considerable surface activities of the solutes in
cyclodextrin solutions are demonstrated by these
curves. At higher CD concentrations, about 15–
25 mJ/m2 decrease in the surface tension of water
could be detected. Without simvastatin, only
the dissolved RAMEB exhibited surface activity.
It can be concluded from these results that
formation of CD–statin complexes leads to the
development of surface-active solutes, which can
accumulate at the air/water interface. In other
words, complexation of statin can also be regarded
as a CD-induced ‘‘amphiphilization’’ of the hydro-
phobic drug. Dissolved CD–statin complexes with
an amphiphilic character may thus cause the
significant reduction in the surface tensions de-
tected in their aqueous solutions. In the polymer-
containing ternary systems, no further decrease of
surface tensions was measured.
Figure 10. Surface activity of cyclodextrins and CD–
statin complexes, T ¼ 258C, pH ffi 6.5.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
9.1  0.5 16.3  0.8 48.0  1.1
The enhanced drug solubilities in polymer solu-
tions can be likely ascribed to the formation of
ternary associates. At the elevated temperatures
used during the thermal treatments, host-guest
association is shifted towards enhanced complexa-
tion of the drug, that is, the concentration of CD–
statin complexes is increased. The surface-active
complexes may attach to the dissolved macro-
molecules in a way that they are anchored by their
hydrophobic part at the polymeric chains. Hereby,
the concentration of the free associates in the
solution may remain the same as that correspond-
ing to the equilibrium concentrations of the binary
complexes develop at 258C in polymer-free solu-
tions. The total concentration of the dissolved
CD–statin complexes is evidently the sum of that
of the free and the bound associates. Accordingly,
at low CD-concentrations where self-association
of CD-molecules can be disregarded,33 a ‘‘simple
sorption’’ mechanism similar to the development
of surfactant-polymer associates can be assumed
in these systems as well, which may lead to the
formation of ternary CD–statin–polymer associ-
ates of supramolecular structure.
A potential benefit in using macromolecular
colloids can also involve a significant improve-
ment of the in vivo bioavailability of these drugs.
Moreover, such ternary assemblies appear to be
better adsorbed to biological membranes, thus
ensuring an enhanced drug delivery and bioavail-
ability.30
CONCLUSIONS
The results of our phase-solubility studies suggest
that in aqueous media, a host-guest type inclusion
complexation of lovastatin and simvastatin with
DOI 10.1002/jps

several cyclodextrins takes place. Dominantly
CD–statin associates of 1:1 molar ratios form,
development of 1:2 complexes seems to be negli-
gible. More stable complexes are formed with the
randomly methylated b-cyclodextrin than with
the native b-CD. Complexation leads to definite
increase in the aqueous solubility of these phar-
macons. In distilled water and at moderate
pH values, improvement of drug solubilities by
1–2 orders of magnitude could be observed.
RAMEB showed outstanding solubilizing effect
for both statins. Also, the dissolved CD–statin
complexes exhibited considerable surface activity,
especially those of the RAMEB.
By adding a macromolecular colloid to the
binary CD–statin systems, the drug solubilities,
after a certain temperature treatment, could
be further improved. The enhancement of drug
solubilities can be attributed to the development
of CD–statin–polymer associates of supramolecu-
lar structure. In these ternary complexes a portion
of the surface-active solutes are attached to the
macromolecular chains. The enhanced drug com-
plexation due to the polymeric additive could
possibly result in improved dissolution of statins
in the gastrointestinal tract. Inclusion complexa-
tion might level off interindividual differences in
bioavailability of these pharmacons, preserve the
liver preference of such lipophilic compounds9 and
also, may protect against some types of food
interactions. Cyclodextrin-based ‘‘nanoencapsula-
tion’’ of statins might, therefore, offer benefits for
the standard antihiperlipidemic therapy.
ACKNOWLEDGMENTS
The authors are highly indebted to Dr. Á. Buvári-
Barcza and late Prof. L. Barcza for their valuable
suggestions and useful discussions.
REFERENCES
1. Istvan ES, Deisenhofer J. 2001. Structural mech-
anism for statin inhibition of HMG-CoA reductase.
Science 292:1160–1164.
2. Roberts WC. 1997. The rule of 5 and the rule of 7 in
lipid-lowering by statin drugs. Am J Cardiol 80:
106–107.
3. Maron DJ, Fazio S, Linton MF. 2000. Current
perspectives on statins. Circulation 101:207–213.
4. Kang BK, Lee JS, Chon SK, Jeong SY, Yuk SH,
Khang G, Lee HB, Cho SH. 2004. Development of
DOI 10.1002/jps
ENHANCEMENT OF DRUG SOLUBILITY 493
self-microemulsifying drug delivery systems
(SMEDDS) for oral bioavailability enhancement
of simvastatin in beagle dogs. Int J Pharm 274:
65–73.
5. Grundy SM. 1988. HMG-CoA reductase inhibitors
for treatment of hypercholesterolemia. N Engl J
Med 319:24–33.
6. Tobert JA. 1988. Efficacy and long-term adverse
effect pattern of lovastatin. Am J Cardiol 62:28J–
34J.
7. Bradford RH, Shear CL, Chremos AN, Dujovne CA,
Franklin FA, Grillo RB, Higgins J, Langendorfer A,
Nash DT, Pool JL. 1994. Expanded Clinical Evalua-
tion of Lovastatin (EXCEL) study results: Two-year
efficacy and safety follow-up. Am J Cardiol 74:667–
673.
8. Haria M, McTavish D. 1997. Pravastatin: A re-
appraisal of its pharmacological properties and
clinical effectiveness in the management of coron-
ary heart disease. Drugs 53:299–336.
9. Hamelin BA, Turgeon J. 1998. Hydrophilicty/lipo-
philicity: Relevance for the pharma-cology and clin-
ical effects of HMG-CoA reductase inhibitors.
Trends Pharmacol Sci 19:26–36.
10. Kearney AS, Crawford LF, Mehta SC, Radebaugh
GW. 1993. The interconversion kinetics, equili-
brium, and solubilities of the lactone and hydro-
xyacid forms of the HMG-CoA Reductase inhibitor,
CI-981. Pharm Res 10:1461–1465.
11. Brewster ME, Loftsson T. 2007. Cyclodextrins as
pharmaceutical solubilizers. Adv Drug Deliv Rev
59:645–666.
12. Loftsson T, Brewster M. 1996. Pharmaceutical
applications of cyclodextrins. I. Drug solubiliza-
tion and stabilization. J Pharm Sci 85:1017–
1025.
13. Liu L, Guo QX. 2002. The driving forces in the
inclusion complexation of cyclodextrins. J Incl
Phenom Macro 42:1–14.
14. Szejtli J. 1982. Cyclodextrins and their inclusion
complexes. Budapest, Hungary: Akadémiai Kiadó.
15. Prentis RA, Lis Y, Walker SR. 1988. Pharmaceu-
tical innovation by seven UK-owned pharmaceuti-
cal companies (1964–1985). Br J Clin Pharmacol
25:387–396.
16. Loftsson T, Friðriksdóttir H, Sigurðardóttir AM,
Ueda H. 1994. The effect of watersoluble polymers
on drug-cyclodextrin complexation. Int J Pharm
110:69–177.
17. Higuchi T, Connors KA. 1965. Phase-solubility
techniques. Adv Anal Chem Instrum 4:117.
18. European Directorate for the Quality of Medicines
& HealthCare: European Pharmacopeia 5th
Edition 5.8 Supplement.
19. Validation of analytical procedures: methodology
(CPMP/ICH/281/95).
20. Országos Gyógyszerészeti Intézet-Gyógyszerkö-
nyvi Osztály—VIII. Magyar Gyógyszerkönyv
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
494 SÜLE, SZENTE, AND CSEMPESZ
(Pharmacopoeia Hungarica VIII.) I./237–239,
II./839.
21. Loftsson T, Masson M. 2004. The effects of water-
soluble polymers on cyclodextrins and cyclodextrin
solubilization of drugs. J Drug Deliv Sci Tec 14:35–
343.
22. Buvári-Barcza Á, Barcza L. 2000. Changes in the
solubility of b-cyclodextrin on complex formation:
Guest enforced solubility of b-cyclodextrin inclusion
complexes. J Incl Phenom Macro 36:355–380.
23. Puskás I, Barcza L, Szente L, Csempesz F. 2006.
Features of the interaction between cyclodextrins
and colloidal liposomes. J Incl Phenom Macro
54:89–93.
24. Redenti E, Szente L, Szejtli J. 2000. Drug/cyclodex-
trin/hydroxy acid multicomponent systems. Proper-
ties and pharmaceutical applications. J Pharm Sci
89:1–8.
25. Süle A, Csempesz F. 2005. Hatóanyag-oldékonyság
szabályozása ciklodextrinekkel és kolloidokkal.
Abst. Semmelweis Egyetem PhD Tudományos
Napok, Budapest, Hungary, p 64.
26. Zia V, Rajewski RA, Stella VJ. 2000. Thermody-
namics of binding of neutral molecules to sulfobutyl
ether b-Cyclodextrins (SBE-b-CDs): The effect of
total degree of substitution. Pharm Res 17:936–941.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 2, FEBRUARY 2009
27. Junquera E, Aicart E. 1997. Potentiometric study of
the encapsulation of ketoprophen by hydroxypro-
pyl-beta-cyclodextrin. Temperature, solvent and
salt effects. J Phys Chem B 101:7163–7171.
28. Marcin C, White R, Hirsch C, Ferris F, Sykes R,
Houck D, Greasham R, Chartrain M. 1991. Biocon-
version of the sodium salt of Simvastatin (MK-733)
to 6-desmethyl-6-a-hydroxymethyl Simvastatin.
J Ind Microbiol Biotechnol 8:157–164.
29. Iga K, Hussain A, Kashihara T. 1981. New direct
calculation of K1: 1 and K1: 2 complexation con-
stants using solubility method. J Pharm Sci 70:108–
109.
30. Loftsson T, Stefánsson E, Friðriksdóttir H,
Kristinsson JK. 1996. Novel CD-based Drug
Delivery System. Proceedings of the Eighth Inter-
national Symposium on Cyclodextrons. p 407–
412.
31. Attwood F, Florence AT. 1983. Surfactant systems.
Their chemistry, pharmacy and biology. London,
UK: Chapman and Hall. pp 361–365.
32. Myers D. 1988. Surfactant science and technology.
New York, USA: VCH Publishers. pp 142–145.
33. Loftsson T, Másson M, Brewster ME. 2004. Self-
association of cyclodextrins and cyclodextrin com-
plexes. J Pharm Sci 93:1091–1099.
DOI 10.1002/jps