Binding of sodium salicylate by β-cyclodextrin or 2,6-di-o-methyl-β- cyclodextrin in aqueous solution

Article abstract of DOI:10.1021/js970117u

Speed of sound and conductivity experiments have been done at 298.15 K to study the encapsulation process of sodium salicylate (NASA) by β- cyclodextrin (β-CD) and 2,6-di-O-methyl-β-cyclodextrin (DIMEB) in aqueous solutions. Since the concentration of the salicyclic form (HSA), coming from the hydrolysis of SA^-, is negligible at biological pH, the binding process studied in this work is that of the SA^- species. The stoichiometries of the complexes DIMEB: SA^- and β-CD:SA^- have been found to be 1:1, as usually determined for most CD:drug complexes. Their association constants and their ionic molar conductivities at infinite dilution have been obtained by fitting the experimental conductivity data with a nonlinear regression method (NLR). For that purpose, a model based on that of Gelb and co-workers has been used. From the values of K(β-CD:SA^-) = (105 ± 15) M^-1 and K(DIMEB:SA^-) = (140 ± 20) M^-1 obtained, the bioavailability of the salicylate drug in the complexed form has been discussed.

Full text of DOI:10.1021/js970117u

Binding of Sodium Salicylate by
â-Cyclodextrin or
2
,6-Di-O-methyl- -cyclodextrin in Aqueous Solution
â
E
LENA
J
UNQUERA, LOURDES
P
EN˜
A
,
AND
EMILIO
A
ICART
*
Contribution from the Departamento de Qu ´ı mica F ´ı sica I, Facultad de Ciencias Qu ´ı micas, Universidad Complutense,
8040-Madrid, Spain.
2
Received March 19, 1997.
Final revised manuscript received June 4, 1997.
X
Accepted for publication September 22, 1997.
which is particularly significant in the area of antiviral and
antisense drugs, has been studied by using model systems
which can mimic the cellular membranes, such as lipo-
somes and micelles.9 Moreover, the encapsulation of these
Abstract
done at 298.15 K to study the encapsulation process of sodium
salicylate (NaSA) by -cyclodextrin ( -CD) and 2,6-di-O-methyl-
0 Speed of sound and conductivity experiments have been
â
â
â-
cyclodextrin (DIMEB) in aqueous solutions. Since the concentration
soluble drugs by cyclodextrins can be useful since the host
molecules may act as concentration-regulating agents,
decreasing the adverse side effects at the organ or cellular
level and improving the optimal dose of the drug and its
-
of the salicyclic form (HSA), coming from the hydrolysis of SA , is
negligible at biological pH, the binding process studied in this work is
-
that of the SA species. The stoichiometries of the complexes DIMEB:
-
-
SA and
â
-CD:SA have been found to be 1:1, as usually determined
2,3,6,7
frequency of administration.
for most CD:drug complexes. Their association constants and their
ionic molar conductivities at infinite dilution have been obtained by
fitting the experimental conductivity data with a nonlinear regression
method (NLR). For that purpose, a model based on that of Gelb and
The numerous therapeutic effects of salicylates, as well
as their possible adverse effects, have been known a long
time. They have been used, for example, as diuretics
(lithium salicylate), intestinal antiseptics (bismuth salicy-
co-workers has been used. From the values of K
1
â-CD:SA
-
)
(105
±
-1
-1
late), and anti-inflammatory, analgesic, and antipyretic
agents (lysine acetylsalicylate, sodium salicylate). In this
work, the encapsulation processes of sodium salicylate
5) M and KDIMEB:SA ) (140 ± 20) M obtained, the bioavailability
of the salicylate drug in the complexed form has been discussed.
-
(NaSA), a soluble anti-inflammatory drug, by â-CD or by
its methylated derivative, DIMEB, are studied by means
of very precise speed of sound and conductimetric measure-
ments. Speed of sound has been widely proved to be a
physicochemical property capable of detecting any struc-
Introduction
Poor bioavailability is, unfortunately, a frequent problem
for drug delivery. Many molecules that are biologically
active in vitro are inactive in vivo due to a variety of
possible problems such as limited solubility or stability,
adverse side effects, unfavorable tastes and odors, and
limited transport across cell membranes.
tural change occurring in the solution. Particularly, it is
a well-known method5
,10,11
to study the formation of su-
pramolecular aggregates, i.e. micelles and inclusion com-
plexes. On the other hand, conductivity has been widely
used to characterize a large amount of different host-guest
systems,5
,12-23
many of them with a drug molecule
13-17
as
Many works can be found in the literature regarding the
low bioavailability of insoluble and unstable drugs. Among
the different methods proposed to overcome this problem,
the molecular encapsulation of these drugs by cyclodextrins
the guest and a cyclodextrin as the host. In this work, the
speed of sound measurements allowed us to determine the
stoichiometry of the complexes and to analyze qualitatively
the affinity of the binding process, while the conductivity
data provide us not only with the stoichiometries but also
with the association constants of the CD:NaSA inclusion
complexes, obtained from the NLR fitting of the conductiv-
ity data. These association constants are, probably, the
most important thermodynamic parameters regarding the
bioavailability of the drug. It is well-known that, if reliable
binding constants values are required, accurate experi-
mental data and nonlinear regression (NLR) methods to
fit these data are necessary. We have recently demon-
CD’s) is probably the most widely used.1
-7
(
Nowdays, CD’s,
which are well-known nontoxic macrocyclic sugars of
natural origin, are considered a whole family of pharma-
ceutical excipients and drug carriers, and many patents
and even commercial products containing CD’s are, actu-
ally, in course. Their doughnut-shaped structure, with
hydrophilic external faces and hydrophobic inner surface,
makes them the most important simple organic compounds
capable of forming noncovalently bonded inclusion com-
plexes with a wide variety of drug molecules in aqueous
solution. The resulting host-guest systems become more
soluble and stable, the stability being a result of the
concurrence of different contributions, such as van der
Waals interactions, hydrophobic effect, solvent reorganiza-
tion, and hydrogen bonding.8
24
strated that the binding constant determination can be
problematic in the case of acid-base conjugated systems
being the substrates, in the sense that the pH value must
guarantee the presence of only the base or the acid species.
On the contrary, the result obtained is just an apparent
constant, averaged over the encapsulation processes of both
species. Since the concentration of the salicyclic acid form,
coming from the neutralization of salicylate in aqueous
medium, is almost negligible at biological pH, the encap-
sulation processes of NaSA by â-CD and/or DIMEB at this
pH or higher (as is the case of the present work) are
On the other hand, drugs which are highly polar and
soluble in aqueous medium can also be poorly bioavailable
due to inefficient transport across the hydrophobic lipid
bilayer constituting the cellular membranes. This problem,
*
To whom correspondence should be addressed. Fax: 34-1-
944135. E-mail: aicart@eucmax.sim.ucm.es.
3
X
-
Abstract published in Advance ACS Abstracts, November 1, 1997.
reduced to the inclusion of the SA species.
8
6 / Journal of Pharmaceutical Sciences
S0022-3549(97)00117-2 CCC: $15.00
Published on Web 01/02/1998
© 1998, American Chemical Society and
American Pharmaceutical Association
Vol. 87, No. 1, January 1998

The u measurements were made (i) as a function of concentration
for the aqueous solutions of the pure substances â-CD, DIMEB,
and NaSA and (ii) as a function of the drug concentration at
different constant values of cyclodextrin concentration, for the
systems â-CD + NaSA and DIMEB + NaSA.
Conductivity data were collected at 298.15 K with a Hewlett-
Packard 4263A LCR meter, using a Metrohm electrode with a cell
-
1
constant of 0.8129 cm
. The experimental procedure, fully
2
7
computerized, was widely described elsewhere. The accuracy on
the specific conductivity, κ, obtained as an average of 2400
measurements for each concentration, is believed to be better than
0
.03%. The conductivity measurements were made for the pure
NaSA and for the systems â-CD + NaSA and DIMEB + NaSA (i)
as a function of NaSA concentration at different constant values
of [CD], (ii) as a function of cyclodextrin concentration at several
constant [NaSA], and (iii) as a function of [CD] and [NaSA] at a
constant stoichiometric ratio [CD]/[NaSA] equal to 1.
The temperature control in both speed of sound and conducti-
metric experiments, as previously reported,2
mK.
5,27
is better than (1
Results and Discussion
Figure 1
s
Plot of
∆
u (
)
u
−
u
o
) versus [NaSA], at several constant
-CD/NaSA/H O.
â-CD
concentrations for the ternary system
â
2
Sp eed of Sou n d Stu d ysIn Figure 1, the experimental
values of ∆u ()u - u , where u is the speed of sound for
o
o
the CD initial solution) are plotted as a function of
concentration for the system â-CD + NaSA at two different
cyclodextrin concentrations, together with the ∆u values
for the pure NaSA. As might be expected, no change in
the slope of ∆u vs concentration has been found for the
pure NaSA water solution, since the concentration at which
NaSA form aggregates in water (cmc), behaving as a
classical surfactant,28 is around 0.67 M, completely out of
the concentration range studied in this work. When the
â-CD is added, a change of the slope of ∆u vs [NaSA],
attributed to the inclusion complex formation, might be
expected. This change, which has been observed in most
of the systems previously studied,1
1,19-22
does not seem to
appear in Figure 1 for the system â-CD + NaSA. A first
interpretation of this fact could be that in this case the
-
inclusion complex â-CD:SA is not formed. Nevertheless,
it can be observed as well as in Figure 1 that at a given
[NaSA] the ∆u value for the system is always lower than
that for the pure NaSA, and the higher the â-CD concen-
tration is, the wider this difference becomes. This feature
must be attributed only to the formation of the complex
-
â-CD:SA , although the stoichiometry cannot be deter-
mined from the speed of sound experiments. The most
plausible explanation of this fact is that, although the
inclusion complex CD:SA is formed, its association is not
strong enough to provoke a detectable change in the slope
of ∆u vs [NaSA] in this system.
Figure 2
s
Plot of
∆
u (
)
u
−
u
o
) versus [NaSA], at several constant DIMEB
O. The inset in the
concentrations for the ternary system DIMEB/NaSA/H
2
-
bottom indicates, as an example, the method to obtain the stoichiometry, A,
of the inclusion complex for the concentration 0.012 M of DIMEB.
In Figure 2, the experimental values of ∆u are plotted
for the system DIMEB + NaSA at three constant concen-
trations of DIMEB. The same behavior found in Figure 1
can be observed in Figure 2: as long as the constant
DIMEB concentration increases, the curve is further away
Experimental Section
Sodium salicylate (NaSA) was purchased from Aldrich Co., with
purities of 99% or greater. â-Cyclodextrin (â-CD) and 2,6-di-O-
methyl-â-cyclodextrin (DIMEB) were obtained from Aldrich Co.
and Cyclolab (Budapest, Hungary), respectively. All of them were
used without purification. Two thermogravimetric analyses (TG)
were done for both cyclodextrins to check the amount of water.
The results obtained, which were considered to calculate solute
concentrations, were 13.5% of water in â-CD and 1.7% of water in
DIMEB. All the solutions were prepared with distilled, deionized
-
than that of pure NaSA, indicating that the DIMEB:SA
complex is formed as well. Nevertheless, in this case, as
can be seen in the zoomed window for one of the DIMEB
concentrations, a slight change in the slope due to the
complex formation can be noticed, in contrast with the
previous system, â-CD + NaSA. This fact, which is a
qualitative indication of a higher binding constant, makes
possible the determination of the complex stoichiometry
from the [NaSA] value at which a change in the slope
of ∆u is observed. The stoichiometry is thus obtained as
the ratio [DIMEB]/[NaSA], where [DIMEB] is the constant
cyclodextrin concentration and [NaSA] is the intersec-
tion point of the two straight lines at which the experi-
mental ∆u values are fitted below and above the change
in the slope. The average value, Ah , for the three experi-
(
taken from a Millipore Super-Q System, with a conductivity lower
-1
-1
than 18 µΩ cm ), and also degassed water.
Speeds of sound, u, were measured at 298.15 K by using a
pulse-echo-overlap technique of fixed path type at a frequency
of 2.25 Mhz. This technique operates in a multiple-echo mode with
broadband pulses. The equipment used, as well as the experi-
mental method, have been fully described previously.25 The
transducer-reflector distance of the measuring cell was calibrated
-
1
from the speed of sound of pure water (1496.74 m s ) reported
2
6
-1
by Kroebel and Mahrt. The precision of u data is (0.1 m s
.
Journal of Pharmaceutical Sciences / 87
Vol. 87, No. 1, January 1998

Figure 3
s
Plot of the molar conductance,
Λ
, versus [NaSA], at several
O.
Figure 5
s
Plot of the specific conductivity,
κ, versus [CD] and [NaSA], the
constant DIMEB concentrations for the ternary system DIMEB/NaSA/H
2
concentration ratio being A
)
[CD]/[NaSA] equal to 1.
In Figure 4, the stoichiometry, A, of the complex can be
estimated following the same procedure previously ex-
plained for the speed of sound experiment (Figure 2). It
means that A is determined as the ratio [DIMEB]/[NaSA],
where [NaSA] is the constant drug concentration and
[DIMEB] is the cyclodextrin concentration at which a
change of the slope of Λ vs [DIMEB] is observed (see
zoomed window). It is noteworthy that this change is not
as sharp as that one found for the CD:guest complexes with
higher association constants.2
0-22
The resulting average
value, Ah , is in agreement with that obtained from the speed
of sound data, confirming the 1:1 stoichiometry for the
-
-
DIMEB:SA complex. In the case of the â-CD:SA com-
plex, whose stoichiometry could not be determined from
the speed of sound experiments, a value of 1.0 ( 0.05 has
been found as well.
In a third study, the conductivity has been measured as
a function of [CD] and [NaSA] at stoichiometric ratio. It
means that both [CD] and [NaSA] increase simultaneously,
the 1:1 ratio being kept constant. Figure 5 shows the
specific conductivity κ vs [CD] and [NaSA], together with
the values obtained for the pure NaSA. Again, it can be
observed that the curves for the CD + NaSA systems run
below that of NaSA in the absence of cyclodextrin, cor-
roborating the features observed in the two previous
studies. Moreover, the curve for the system DIMEB +
NaSA runs slightly below that for the â-CD + NaSA
system. This behavior can be considered again a qualita-
tive indication of a tighter affinity of DIMEB upon binding
NaSA, with respect to that of â-CD.
Associa tion Con sta n t Deter m in a tion sThe encapsu-
lation of the drug molecule into the CD cavity is an
equilibrium process governed by the binding constant. In
the present case, since the salicylate species is the conju-
gated base of an acid/base system, the equilibria involved
in this encapsulation process are shown in the following
scheme:
Figure 4
s
Plot of the molar conductance,
Λ
, versus [DIMEB], at several
O. The
constant NaSA concentrations for the ternary system DIMEB/NaSA/H
2
inset in the top indicates, as an example, the method to obtain the
stoichiometry, A, of the inclusion complex for the 0.012 M of NaSA.
ments is 1.0 ( 0.05, as usually found for most CD:guest
systems.20-22,29-34
Con d u ctivity Stu d ysIn Figures 3 and 4, the values
of the molar conductance (Λ ) 1000κ/[NaSA]) of the drug
solutions are plotted as a function of sodium salicylate
concentration (at various constant [DIMEB]) and as a
function of DIMEB concentration (at various constant
[NaSA]), respectively. Similar plots have been obtained
for the system â-CD + NaSA.
As can be seen in Figure 3, the molar conductivity Λ of
pure NaSA decreases appreciably in the presence of
DIMEB, the magnitude of this decrease being higher as
long as the constant DIMEB concentration increases. This
-
fact can be attributed to the formation of the DIMEB:SA
complex. The SA species loose mobility as long as it is
included in the CD cavity, and the mobility of the CD:SA
complex is lower than that of the SA alone. This loss of
-
-
-
mobility justifies the decrease in Λ observed in the figure.
The same trend has been found in the system â-CD +
NaSA.
where
8
8 / Journal of Pharmaceutical Sciences
Vol. 87, No. 1, January 1998

K ) (a
OH
-
aHSA)/(aSA
-
)
(1)
(2)
(3)
(4)
Table 1
Hydrodynamic Parameters, a
Inclusion Complexes
s
Molar Ionic Conductivities at Infinite Dilution, λ°
i
, Ion Size
h
n
, and Association Constants, K, of the
KCD:SA
KCD:HSA ) (aCD:HSA)/(aCDaHSA
K ′ ) (aCD:HSAaOH )/(aCD:SA )
-
) (aCD:SA
-
)/(aCDaSA )
-
o
Ω- cm mol
1
2
-1
K/M-
1
ion
λi
/
n
a /Å
)
Na+
50.141
442
-
SA
â
34
22
17
±
±
±
2
3
3
7
10
13
±
±
±
1
2
2
h
-
-
-CD:SA-
DIMEB:SA
105
140
±
±
15
20
-
h h
The constants K and K ′ are the hydrolysis constants of
-
the salicylate species SA in the free and complexed form
salicylate (free and complexed, eq 6) and the cyclodextrin
eq 7),
respectively, while KCD:SA- and KCD:HSA are the binding
(
constants of the inclusion complexes formed by the cyclo-
dextrin and the nonprotonated and protonated forms of
HSA, respectively.
These association constants, together with the stoichi-
ometry, are thermodynamic parameters necessary to char-
acterize these CD:drug inclusion complexes. Factors such
as the bioavailability and dosage of the drug, of great
importance in the pharmacological industry, can be con-
trolled with the knowledge of the binding constant.
-
-
[
drug]total ) [SA ] + [CD:SA ]
(6)
(7)
-
[CD]total ) [CD] + [CD:SA ]
and the charge balance (eq 8)
+
+
-
-
-
[H ] + [Na ] ) [SA ] + [CD:SA ] + [OH ]
(8)
It is necessary to find a model to determine the binding
constants by relating them with the experimental values
of the studied physicochemical property as a function of
concentration. Several features regarding CD/drug/water
systems need to be remarked upon in order to choose the
most suitable model: (i) Most drug molecules are known
Besides, we have considered the following particularities
in the model: (a) In spite of the low concentrations used
in this work for SA (10 -10 M), the activity coefficients
of the charged species, obtained through the extended
Debye-H u¨ ckel theory, have been taken into account. (b)
-
-4
-2
to form 1:1 inclusion complexes with CD’s.5
,6,30-34
The ionic molar conductivities of the charged species, λ
i
,
This fact,
have been related with the corresponding values at infinite
together with the well-known cavity size of the â-CD and
o
derivatives3
5-37
-
dilution, λ , through the well-known Debye-H u¨ ckel-On-
i
40
and the small size of the SA molecule,
-
sager theory. (c) The ionic molar conductivity at infinite
leads us to assume that complexes such as CD
2
:SA or CD:
-
o
SA
-
(
SA )
2
or those with even higher stoichiometries are not
dilution of SA ion, λ
-
, has been directly determined by
expected. Furthermore, this fact has been confirmed in
a NLR method, from the conductivity measurements of the
system NaSA + H O reported in this work. The obtained
value and that reported in the literature for the Na ion
are shown in Table 1. (d) The ion size hydrodynamic
parameters, a , necessary for the calculation of the activity
coefficients, have been either taken from the literature for
this work through speed of sound and conductivity mea-
2
surements. (ii) Several authors1
8,38
have found that the
+
41
counterion of small molecules are not associated to the
inclusion complex, opposite to the behavior found for CD:
n
surfactant inclusion complexes.2
2,38,39
Thus, Na ion can
be considered in this work as a mere “spectator” of the CD:
drug inclusion process. (iii) Since the pK of the salicyclic
acid is 2.9524 at 25 °C, the hydrolysis of NaSA in concen-
+
+
42
the Na ion or estimated from the ionic molar conductiv-
o
i
43
a
ity at infinite dilution, λ , using the Br u¨ ll equation for
-
the SA ion (Table 1).
-
4
-2
trations around 10 -10 M is moderately basic. As a
consequence, almost all of NaSA in solution is in the
salicylate form, with a negligible contribution of the pres-
ence of HSA species. Thus, the previously presented
equilibria are reduced in this work to that for the encap-
sulation of the salicylate species by the cyclodextrin:CD +
A nonlinear regression method, based on a nonlinear
Newton-Raphson and a Marquardt algorithm, has been
used to fit the experimental conductivity data, according
with the above explained model. The values obtained for
the binding constant K
CD:SA
o
-
and the ionic molar conductiv-
ity at infinite dilution λ
, are reported in Table 1 for
CD:SA-
-
-
-
-
SA T CD:SA .
both â-CD:SA and DIMEB:SA complexes. The values in
Table 1 indicate that the mobility of the associated sali-
cylate ion decreases between 40 and 50% with respect to
the mobility of the free salicylate ion, causing a decrease
in the conductivity of the solution as long as the complex
Since all the species in the solution (i.e. H
2
O, CD, SA-,
Na , and CD:SA ) will contribute to the speed of sound
value, while in the case of conductivity only the charged
+
-
-
+
-
species (SA , Na and CD:SA ) have to be considered, a
model based on conductivity data, with a lower number of
variables, seems to be less complicated. Thus, keeping in
mind all the above mentioned premises, the model proposed
by Gelb et al.18 has been chosen to obtain the binding
constant of the inclusion complexes studied in this work,
from conductivity results. Gelb and co-workers have
studied a number of CD:substrate complexes by using their
model with satisfactory results.18 On the basis of this
model, the specific conductivity, κ, of a solution of sodium
salicylate and cyclodextrin can be expressed as a function
-
CD:SA is formed, justifying the behavior found in Figures
3 and 5. It can also be noticed that the â-CD derivative
(DIMEB) binds NaSA with a higher affinity (around 35%
higher) than the unsubstituted â-CD does, probably due
3
5-37
to the higher flexibility of the DIMEB molecule,
which
makes it easier for the CD-drug interaction to occur. This
feature, which has been noticed in the speed of sound and
conductimetric experiments, can justify why only the
-
stoichiometry of the complex DIMEB:SA could be ob-
tained. Anyway, it should be remarked that both binding
constants are moderate, in contrast with those found for
the complexes formed by CD’s and the salicylic acid, HSA,
of the ionic molar conductivities, λ
i
, of all the ionic species
at their corresponding concentration as follows,
-
1 24,44
which are in the range of 1000 M
(at pH ≈ 1), around
-
-1
+
-
[CD:SA-] (5)
CD:SA
8 times higher. A value of K ) 220 ( 40 M has been
reported for the complex HPBCD:SA (at pH ) 7) using
a fluorescence technique, and only a value of Kâ-CD:SA-
κ ) λNa
+
[Na ] + λ
SA
-
[SA ] + λ
-
24
)
-
1
where the concentrations are related to each other through
the binding constant (eq 2), the mass balances of the
(51 ( 7) M has been reported at pH ) 7.4 by Sideris et
al.,45 using a dynamic dialysis technique. The value
Journal of Pharmaceutical Sciences / 89
Vol. 87, No. 1, January 1998

reported by Sideris et al., although of the same order of
magnitude, differs from that obtained by us in this work
within the experimental uncertainties. We believe that
these differences can be attributed to the method used by
Sideris et al. to obtain the binding constant, a Scatchard
plot method which, in contrast with NLR methods, is well-
known to give unreliable values for the association constant
in most cases. In contrast, the K values obtained for the
11. J unquera, E.; Tardajos, G.; Aicart, E. Langmuir 1993, 9,
213.
1
1
2. Connors, K. A. Binding Constants: The Measurement of
Molecular Complex Stability; J ohn Wiley & Sons: New York,
1
987.
13. Cramer, F.; Saenger, W.; Spatz, H.-Ch. J . Am. Chem. Soc.
1
967, 89, 14.
1
1
4. Connors, K. A.; Lipari, J . M. J . Pharm. Sci. 1976, 65, 379.
5. Uekama, K.; Otagiri, M.; Kanie, Y.; Tanaka, S.; Ikeda, K.
Chem. Pharm. Bull. 1975, 23, 1421.
-
-
complexes â-CD:SA or DIMEB:SA (this work) and for
1
6. Otagiri, M.; Miyaji, T.; Uekama, K.; Ikeda, K. Chem. Pharm.
-
24
HPBCD:SA
are all determined under analogous satura-
Bull. 1976, 24, 1146.
tion degree46 range and by using NLR methods, although
17. Miyaji, T.; Kurono, Y.; Uekama, K.; Ikeda, K. Chem. Pharm.
Bull. 1976, 24, 1155.
with different experimental techniques and mathematical
models. They reveal that the affinity of NaSA upon binding
a cyclodextrin is greater in the cases of the â-CD derivatives
than for the partner cyclodextrin (â-CD), behavior usually
found in most CD + guest systems.
1
8. (a) Gelb, R. I.; Schwartz, L. M.; Murray, C. T.; Laufer, D. A.
J . Am. Chem. Soc. 1978, 100, 3552. (b) Gelb, R. I.; Schwartz,
L. M.; Laufer, D. A. J . Am. Chem. Soc. 1978, 100, 5875.
19. (a) Satake, I.; Ikenoue, T.; Takeshita, T.; Ayakawa, K.;
Maeda, T. Bull. Chem. Soc. J pn. 1985, 58, 2746. (b) Satake,
I.; Yoshida, S.; Ayakawa, K.; Maeda, T.; Kusumoto, Y. Bull.
Chem. Soc. J pn. 1986, 59, 3991.
From simple mathematical operations with the values
-
1
-1
of Kâ-CD:SA- ) 105 M and KDIMEB:SA- ) 140 M obtained
in this work, it can be concluded that a normal dose of
sodium salicylate, if administered encapsulated by â-CD
or DIMEB at biological pH, leaves in the medium an initial
20. (a) Palepu, R.; Reinsborough, V. C. Aust. J . Chem. 1990, 43,
2119. (b) Palepu, R.; Reinsborough, V. C. Can. J . Chem.
1
988, 66, 325.
2
1. (a) Tawarah, K. M.; Wazwaz, A. A. J . Chem. Soc. Faraday
Trans. 1993, 89, 1729. (b) Tawarah, K. M.; Wazwaz, A. A.
Ber. Bunsenges. Phys. Chem. 1993, 97, 727.
-
quantity of free and available SA which ranges from 40
to 60% of the total amount of the administered drug. And,
what is more important, as long as the organism “makes
use” of the drug, the equilibrium (eq 2) is shifted by mass
action toward the release of the active principle, keeping
and regulating its presence in the medium. These effects
would be even more emphasized when the HSA species is
absolutely predominant, which occurs at a strongly acidic
pH, characteristic for example of the stomach. Thus,
repeating the same simple calculations with the binding
constants of the complexes formed by cyclodextrins and
salicylic acid, it can be concluded that with similar doses,
just 25% of the drug would be free in the medium.
22. (a) J unquera, E.; Pe n˜ a, L.; Aicart, E. Langmuir, 1995, 11,
4
685. (b) J unquera, E.; Pe n˜ a, L.; Aicart, E. J . Inclusion
Phenom. 1996, 24, 233.
2
3. Posp ´ı sil, L.; Colombini, M. P. J . Inclusion Phenom. 1993, 16,
2
55.
24. J unquera, E.; Aicart, E. J . Inclusion Phenom. 1997, 29, 119.
25. Tardajos, G.; Diaz Pe n˜ a, M.; Aicart, E. J . Chem. Thermodyn.
1
986, 18, 683.
2
2
2
6. Kroebel, W.; Marht, K. M. Acustica 1976, 35, 134.
7. J unquera, E.; Aicart, E. Rev. Sci. Instrum. 1994, 65, 2672.
8. Treiner, C.; Khodja, A. A.; Fromon, M.; Chevalet, J . J . Sol.
Chem. 1989, 18, 217.
29. Martin Davies, D.; Savage, J . J . Chem. Soc. Perkin Trans. 2
994, 1525.
0. Fronza, G.; Mele, A.; Redenti, E.; Ventura, P. J . Pharm. Sci.
992, 81, 1162.
1
3
3
These estimations rebound the advantages of the use of
cyclodextrins as drug carriers, widely commented on in the
literature:1 (i) the adverse side effects can be substan-
tially reduced since the quantity of drug which is really
free and available in the medium may be much lower than
the administered dose; (ii) the time of action of the drug
can be prolonged, which in the case of the NSAID therapy
1
1. Senel, S.; C¸ akoglu, O¨ .; Sumnu, M.; Duch eˆ ne, D.; Hincal, A.
-6
A. J . Inclusion Phenom. 1992, 14, 171.
32. Puglisi, G.; Ventura, C. A.; Fresta, M.; Vandelli, M. A.;
Cavallaro, G.; Zappala, M. J . Inclusion Phenom. 1996, 24,
1
93.
3
3. Esclusa-D ´ı az, M. T.; Guimaraens-M e´ ndez, M.; P e´ rez-Marcos,
M. B.; Vila-J ato, J . L.; Torres-Labandeira, J . J . Int. J . Pharm.
1996, 143, 203.
(salicyclic and related drugs) is particularly important,
3
3
4. Loukas, Y. L.; Vraka, V.; Gregoriadis, G. Int. J . Pharm. 1996,
given its well-known fast onset of action and subsequent
short elimination half-life; and (iii) as a consequence, the
number of doses and their frequency can be reduced with
the subsequent decrease of the side effects as well. The
confirmation of all these estimations in vivo, although out
of the scope of this work, would be welcomed.
1
44, 225.
5. Bender, M.; Komiyama, M. Cyclodextrin Chemistry; Springer-
Verlag: Berl ´ı n, 1978.
36. Saenger, W. Inclusion Compounds; Atwood, J . L., Davies, J .
E. D., MacNicol, D. D., Eds.; Academic Press: London, 1984.
3
3
7. Clarke, R. J .; Coates, J . H.; Lincoln, S. F. Adv. Carbohydr.
Chem. Biochem. 1988, 46, 205.
8. MacPherson, Y. E.; Palepu, R.; Reinsborough, V. C. J .
Inclusion Phenom. 1990, 9, 137.
References and Notes
. Luzzi, L.; Palmieri, A. Biomedical Applications of Microen-
capsulation; F. Lim Ed.: CRC Press, Boca Rat o´ n, FL, 1984.
. Duch eˆ ne, D.; Wouessidjewe, D. Drug Dev. Ind. Pharm. 1990,
1
2
3
1
6, 2487.
. (a) Szejtli, J . Med. Res. Rev. 1994, 14, 353. (b) Szejtli, J . J .
Inclusion Phenom. 1992, 14, 25.
4
5
. Loftsson, T.; Brewster, M. E. J . Pharm. Sci. 1996, 85, 1017.
. Szejtli, J .; Osa, T. Comprehensive Supramolecular Chemistry.
Vol. 3. Cyclodextrins; J . M., Lehn, Ed.; Pergamon Press:
Oxford, UK, 1996, and references therein.
6
. Thompson, D. O. Crit. Rev. Ther. Drug Carrier Syst. 1997,
1
4, 1.
7
8
9
. Rajewski, R. A.; Stella, V. J . J . Pharm. Sci. 1996, 85, 1142.
. Schneider, H.-J . Angew. Chem. Int. Ed. Engl. 1991, 30, 1417.
. (a) Nowick, J . S.; Chen, J . S.; Noronha, G. J . Am. Chem. Soc.
Acknowledgments
1
993, 115, 7636. (b) Nowick, J . S.; Cao, T.; Noronha, G. J .
Am. Chem. Soc. 1994, 116, 3285.
0. J unquera, E.; Aicart, E.; Tardajos, G. J . Phys. Chem. 1992,
6, 4533.
Authors are grateful to the Ministerio de Educaci o´ n y Cultura
of Spain for funding through the DGES PB95-0356.
1
9
J S970117U
9
0 / Journal of Pharmaceutical Sciences
Vol. 87, No. 1, January 1998