Quantitative estimation of the bitter taste intensity of oxyphenonium bromide reduced by cyclodextrins from electromotive force measurements

Article abstract of DOI:10.1021/ac981286h

The bitter taste of oxyphenonium bromide, an antiacetylcholine drug, is suppressed by cyclodextrins. The extent of the suppression can be predicted from the electromotive force measurements with an oxyphenonium bromide- selective electrode. The relationship between the bitter taste intensity and the electromotive force holds true, regardless of the kind and concentration of natural and modified cyclodextrins. This result is explicable on the basis of the observation that both the bitter taste and the electric potential are determined by the concentration of free oxyphenonium bromide. Some implications and limitations of the present approach are discussed.

Full text of DOI:10.1021/ac981286h

Anal. Chem. 1999, 71, 1733-1736
Quantitative Estimation of the Bitter Taste
Intensity of Oxyphenonium Bromide Reduced by
Cyclodextrins from Electromotive Force
Measurements
Noriaki Funasaki,* Ryusaku Kawaguchi, Seiji Ishikawa, Sakae Hada, and Saburo Neya
Kyoto Pharmaceutical University, 5, Nakauchicho, Misasagi, Yamashina-ku, Kyoto, 607-8414, Japan
Takashi Katsu
Faculty of Pharmaceutical Sciences, Okayama University, 1-1-1, Tushima-naka, Okayama, 700-8530, Japan
The bitter taste of oxyphenonium bromide, an antiacetyl-
choline drug, is suppressed by cyclodextrins. The extent
of the suppression can be predicted from the electromo-
tive force measurements with an oxyphenonium bromide-
selective electrode. The relationship between the bitter
taste intensity and the electromotive force holds true,
Figure 1. Chemical structure of OB.
regardless of the kind and concentration of natural and
modified cyclodextrins. This result is explicable on the
basis of the observation that both the bitter taste and the
electric potential are determined by the concentration of
free oxyphenonium bromide. Some implications and
limitations of the present approach are discussed.
performance liquid chromatography and capillary electrophoresis.4-6
Cyclodextrins can give beneficial modifications of guest molecules
not otherwise achievable: solubility enhancement, stabilization
of labile guests, control of volatility and sublimation, and physical
isolation of incompatible compounds. Because they are practically
nontoxic, they are added into pharmaceuticals and foods for
stabilization of labile compounds and long-term protection of color,
odor, and flavor.3 Furthermore, cyclodextrins can mask bitter
tastes of drugs, e.g., propantheline bromide.7 Sensory tests
generally depend on individuals. Some instrumental methods,
therefore, are desired for such tests.
In this work, we develop an ion-selective electrode method for
the quantitative estimation of the bitter taste intensity of oxy-
phenonium bromide (Figure 1), an antiacetylcholine drug, in
aqueous solutions of cyclodextrins. Electromotive force measure-
ments have been applied to investigate the determination of
binding constants of cyclodextrins with guests, such as drugs8,9
and surfactants.10,11 It is noted that the electromotive force solely
depends on the concentration of free guest, regardless of the
concentrations of cyclodextrins and their complexes. Bitter
The developments and various applications of new electro-
chemical sensors continue to be a rapidly growing area of
analytical chemistry. Many researchers are currently working on
constructing new drug-sensitive membrane sensors to monitor
certain drugs in pure form, complex pharmaceutical formulations,
and biological materials. For analytical control of pharmaceuticals,
membrane sensor techniques offer several advantages in terms
of simplicity, rapidity, specificity, and accuracy over many known
methods.1,2
Cyclodextrins have homogeneous toroidal structures of dif-
ferent sizes. One side of the torus contains primary hydroxyl
groups, whereas the secondary groups are located on the other
side. The toroidal structure has a hydrophilic surface resulting
from the 2-, 4-, and 6-position hydroxyls, making them water
soluble. The cavity is composed of the glucoside oxygens and
methylene hydrogens, giving it a hydrophobic character.3,4 Cy-
clodextrins are used for separations and quantitative analysis of
very similar compounds, including optical isomers, by high-
(5) Vindevogel, J.; Sandra, P. Introduction to Micellar Electrokinetic Chroma-
tography; Hu¨ thig: Heidelberg, 1992; Chapter 12.
(6) Krstulovic, A. M. Chiral Separations by HPLC; Ellis Horwood: Chichester,
1989; Chapter 10.
* Corresponding author: (fax) +81-75-595-4762; (e-mail) funasaki@
mb.kyoto-phu.ac.jp.
(7) Funasaki, N.; Uemura, Y.; Hada, S.; Neya, S. J. Phys. Chem. 1 9 9 6 , 100,
16298-301.
(1) Cosofret, V. V.; Buck, R. P. Pharmaceutical Applications of Membrane Sensors;
(8) M-Papazoglou, A.; Christopoulos, T. K.; Diamandis, E. P.; Hadjiioannou, T.
CRC Press: Boca Raton, 1992; pp 1-4.
P. Analyst 1 9 8 5 , 110, 1091-4.
(2) Katsu, T.; Mari, Y. Anal. Chim. Acta 1 9 9 7 , 343, 79-83.
(3) Bender, M. L.; Komiyama, M. Cyclodextrin Chemistry; Springer-Verlag:
Berlin: 1978; Chapters 2 and 3.
(9) Valsami, G. N.; Macheras, P. E.; Koupparis, M. A. J. Pharm. Sci. 1 9 9 0 , 79,
1087-94.
(10) WanYanus, W. M. Z.; Taylor, J.; Bloor, D. M.; Hall, D. G.; Wyn-Jones, E. J.
(4) Szejtli, J. Cyclodextrin Technology; Kluwer Academic Publishers: Dor-
drecht: 1988; Chapters 3 and 7.
Phys. Chem. 1 9 9 2 , 96, 8979-82.
(11) Tominaga, T.; Hachisu, D.; Kamado, M. Langmuir 1 9 9 4 , 10, 4676-80.
10.1021/ac981286h CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/19/1999
Analytical Chemistry, Vol. 71, No. 9, May 1, 1999 1733

Table 1. Degree of Substitution and Binding Constants
of Oxyphenonium Bromide for Cyclodextrins at 309.7
K
CD
deg sub
K1 (M-1
)
n
s (mV)
R-CD
58
8500
96
6660
4290
1460
1920
10010
7510
12
11
12
11
11
11
11
10
11
0.54
1.02
0.38
2.89
3.08
2.84
5.17
3.80
4.06
â-CD
γ-CD
DΜâ-CD
Mâ-CD
ΗPâ-CD
GLâ-CD
CMâ-CD
MLâ-CD
∼2
0.5-0.7
0.6
∼1
?
∼1
compounds are generally hydrophobic,12 but their cyclodextrin
complexes are rather hydrophilic, because of the hydrophilicity
of cyclodextrins.7 Thus we can expect that the bitter taste of
oxyphenonium bromide in a cyclodextrin solution is estimated
from the value of electromotive force of an appropriate ion-
selective electrode.
Figure 2. Electromotive force plotted against the logarithm of the
OB concentration in the absence of CD. The solid line is calculated
from eq 1.
Measurements of Electromotive Forces. Potentiometric
measurements were carried out with a DKK model IOL-40 digital
pH/ mV meter. The electrochemical cell was constructed as
follows: Ag/ AgCl|KCl solution|PVC membrane|sample solution|1
mM OB, 10 mM NaBr|AgBr/ Ag. The Ag/ AgBr electrode was
kindly supplied by DKK. The electromotive force was referred to
a DKK 4083-0.65C double-junction reference electrode. The vessel
containing the sample solution was jacketed to maintain a constant
temperature of 309.7 ( 0.1 K. The temperature was monitored
continuously with a thermometer. The electromotive force of a
fresh aqueous solution reached an equilibrium value typically
within 2 min. The response was faster as the OB concentration
was increased. The calibration curve for OB was determined as
follows: 25 cm3 of a solution containing 0.02 mM OB and 154
mM NaBr was titrated successively by a 20 mM OB solution and
the equilibrium potential was measured digitally. The results of
three runs are reported herein. The effect of the CD concentration
on the potential of a 4 mM OB solution was investigated as
follows: 15 cm3 of a solution containing 4 mM OB, CD, and 154
mM NaBr was titrated stepwise by a solution of 4 mM OB and
154 mM NaBr. For â-CD, a solution of 154 mM NaBr and 4 mM
OB was titrated by a solution containing 154 mM NaBr, 4 mM
OB, and 12 mM â-CD. The two sets of potentiometric titration
data agreed with one another, indicating the reliability of our
method.
EXPERIMENTAL SECTION
Reagents. Oxyphenonium bromide (OB) was purchased from
Sigma Chemical Co. Because this sample was analyzed to be pure
by reversed-phase liquid chromatography, it was used without
purification. Sodium bromide of analytical grade and R-, â-, and
γ-CDs from Nacalai Tesque Co. were used as received. 2,6-O-
Dimethyl-â-CD (DMâ-CD), 2-hydroxypropyl-â-CD (HPâ-CD),
6-maltosyl-â-CD (MLâ-CD), and 6-glucosyl-â-CD (GLâ-CD) were
purchased from Nacalai Tesque Co. These modified CDs were
mixtures of CDs different in degrees of substitution as detected
by reversed-phase liquid chromatography. The degree of substitu-
tion for some cyclodextrins, provided by Nacalai Tesque Co., is
shown in Table 1. Carboxymethyl-â-CD (CMâ-CD) and methyl-
â-CD (Mâ-CD) were obtained from Cylcolab, Budapest, and Sigma
Chemical Co., respectively. Sodium tetraphenylborate, o-nitro-
phenyl octyl ether, and tetrahydrofuran (THF) were obtained from
Dojindo Laboratories. Poly(vinyl chloride) (PVC) was from Wako
Pure Chemicals Co. The ion-exchanged water was used after
double distillation.
P reparation of the P VC Membrane Electrode. The PVC
membrane was prepared according to the methods used by Zhang
et al.13 and recommended by Denki Kagaku Keiki Co. (DKK).
Sodium tetraphenylborate (5 mg) and 1.9 g of o-nitrophenyl octyl
ether (plasticizer) were dissolved into 8 cm3 of THF. Then PVC
(0.2 g) was added stepwise into the THF solution under magnetic
stirring. Immediately after the DKK membrane filter (6 mm in
diameter), previously immersed in THF, was transferred into the
THF membrane solution, it was fitted to the tip of an ion-selective
electrode body. Further, a drop of the membrane solution was
added with a micropipet to the fixed filter, followed by evaporation
of the THF in 20 min. This operation was repeated 10 times. The
resulting membrane body was soaked in a 10 mM OB solution in
3 h. Then an internal solution containing 1 mM OB and 10 mM
NaBr was filled into the body. Finally, an Ag/ AgBr electrode was
mounted to the body. The electrode was stored in 1 mM OB
solution.
Intensity of Bitter Taste. Five volunteers were involved in
the sensory test. These panelists tasted aqueous 154 mM NaBr
solutions containing OB alone and a mixture of 4 mM OB and
CD. The bitter taste intensities of these solutions were evaluated
by the following scores: 0, no bitter taste; 1, very slightly bitter
taste; 2, slightly bitter taste; 3, appreciably bitter taste; 4, very
bitter taste; 5, extremely bitter taste. The average of bitterness
scores over the five individuals was used for further analysis.
RESULTS AND DISCUSSION
Electromotive Forces and Bitter Taste Intensities of
Aqueous Solutions of Oxyphenonium Bromide. As Figure 2
shows, the equilibrium electromotive force E of an OB solution
containing 154 mM NaBr increased with increasing OB concen-
tration. Over the investigated range of the OB concentration Cp
(mM) from 0.02 to 10.6 mM, the electromotive force (mV) obeyed
(12) Kurihara, K. Taste and Smell; Kagaku Doujin: Kyoto, 1990; Chapter 6.
(13) Zhang, G. H.; Imato, T.; Asano, Y.; Sonoda, T.; Kobayashi, H.; Ishibashi, N.
Anal. Chem. 1 9 9 0 , 62, 1644-8.
1734 Analytical Chemistry, Vol. 71, No. 9, May 1, 1999

Figure 3. Score of bitter taste of aqueous OB solutions plotted
against the OB concentration.
Figure 5. Effects of CDs on the electromotive force of a 4 mM OB
solution: R-CD (b), â-CD (O), and γ-CD (9), DMâ-CD (0), HPâ-CD
(4), MLâ-CD (1), GLâ-CD ([), CMâ-CD (3), and Mâ-CD (2). The
solid lines are calculated from eq 6 using the equilibrium binding
constants shown in Table 1.
Effects of Cyclodextrins on the Electromotive Force of a
4 mM OB Solution. As Figure 5 shows, the electromotive force
of a 4 mM OB solution is decreased by the addition of CD. This
decrease is ascribed to the reduction of free OB concentration
[P], caused by the entrapment of OB into the CD cavity. The
extent of decrease is in the following order: CMâ-CD > â-CD >
MLâ-CD > DMâ-CD > Mâ-CD > GLâ-CD > HPâ-CD > γ-CD >
R-CD.
Binding Constants of Oxyphenonium Bromide for Cyclo-
dextrins. The electromotive force data in Figure 5 were analyzed
on the basis of the 1:1 complexation of oxyphenonium ion (P)
and CD (D). The equilibrium constant of this complexation is
defined as
Figure 4. Relationship between the electromotive force and the
bitter taste intensity of OB in the absence (solid line) and the presence
of R-CD (b), Β-CD (O), and γ-CD (9), DMâ-CD (0), HPâ-CD (4),
MLâ-CD (1), GLâ-CD ([), CMâ-CD (3), and Mâ-CD (2).
K1 ) [PD]/ [P][D]
(2)
eq 1. Because OB forms the micelle above a critical micelle
E ) -22.902 + 61.35 log Cp
(1)
The total concentrations of OB and CD are written as
CP ) [P] + [PD]
concentration (cmc) of 0.108 M,14 it might form oligomers even
at concentrations below this cmc. However, the slope of the
electromotive force vs log OB concentration plot does not depend
on the concentration in the concentration range investigated. This
concentration independence of the slope shows that the self-
association of OB is negligible. The slope in eq 1 is very close to
a theoretical value of 61.45 mV at 309.7 K. This fact indicates that
our electrode responds to OB normally. Furthermore, it is noted
that the activity coefficient of OB is almost independent from Cp,
because of the presence of excess NaBr.
(3)
(4)
CD ) [D] + [PD]
From eqs 2-4, we can obtain the concentration of free oxy-
phenonium ion:
[P] ) {K1CP - K1CD - 1+ [(K1CP - K1CD - 1)2 +
4K1CP]1/ 2}/ 2K1 (5)
The intensity of bitter taste of an OB solution is shown as a
function of OB concentration in Figure 3. A combination of the
results of Figures 2 and 3 yields the relationship between the bitter
taste intensity and the electromotive force for aqueous solutions
of OB alone. As the solid line in Figure 4 shows, the bitter taste
intensity of OB solutions increases with elevation in electromotive
force.
Because the CDs used are hydrophilic, their complexes with
the oxyphenonium ion would be also hydrophilic. As the result,
these complexes would not form any salt with tetraphenylborate
ion and consequently would not interfere with the electromotive
force of the present electrode. Under these conditions, the
electromotive force of any aqueous solution of OB and CD is
determined by the concentration of the uncomplexed oxyphe-
(14) Attwood, D. J. Phys. Chem. 1 9 7 6 , 80, 1984-7.
Analytical Chemistry, Vol. 71, No. 9, May 1, 1999 1735

Figure 7. Schematic relationship between the equilibria of CD
complexation and receptor binding of OB.
reduction by the kind of CD is almost consistent with the result
of Figure 5 and that of the magnitude of the equimolar binding
constant.
All the data of bitter taste intensities shown in Figures 3 and
6 and electromotive forces shown in Figures 2 and 5 are included
in Figure 4. These data on solutions containing both OB and CD
are close to the data on solutions containing OB alone. The
electromotive force and the bitter taste intensity of an aqueous
solution of OB and CD will be determined by the concentration
([P]) of uncomplexed OB in the solution, irrespective of the kind
and concentration of CD. That is,
Figure 6. Reduction of the bitter taste intensity of a 4 mM OB
solution by the addition of CDs: R-CD (b), â-CD (O), and γ-CD (9),
DMâ-CD (0), HPâ-CD (4), MLâ-CD (1), GLâ-CD ([), CMâ-CD (3),
and Mâ-CD (2). The solid lines are calculated from eq 9.
E ) f ([P])
(8)
(9)
bitter taste intensity ) g{[P]} ) g{f -1(E)}
nonium ion. Thus, CP in eq 1 is replaced by [P] for aqueous
solutions containing both OB and CD:
These equilibria are illustrated in Figure 7. The receptor of bitter
taste binds hydrophobic compounds. Although OB is a hydro-
phobic compound, its complex with CD is not hydrophobic.
Therefore, CDs and their complex do not taste bitter. Thus, eq 9
holds true for all solutions containing OB and CDs. This important
result enables us to predict the intensity of bitter taste of an OB
and CD solution from the observed electromotive force of their
mixed solution and the relation between the bitter taste intensity
and the electromotive force of an aqueous OB solution.
Implications and Limitations of the P resent Work. The
present approach will apply to other ionic bitter compounds but
does not apply to nonionic compounds and other tastes, such as
sweetness, acidity, and saltiness. Many drugs self-associate in
aqueous media by hydrophobic interactions.16 The present ap-
proach will apply to such drugs, though OB does not self-associate
in the concentration range investigated. The binding of OB and
CD is explicable in term of the 1:1 stoichiometry, though other
stoichiometries are observed for CD inclusion systems.15,17 The
present approach will apply to such multiple-complexing systems.
Because the relation of eq 9 depends on each compound, it does
not apply to other bitter compounds. It is desired to develop a
universal electrode that determines the absolute intensity of bitter
taste for all compounds.
E ) -22.902 + 61.35 log{K1CP - K1CD - 1 +
[(K1CP - K1CD - 1)2 - 4K1CP]1/ 2}/ 2K1 (6)
This equation was applied to the observed electromotive force
data shown in Figure 5. Thus, we determined the best-fit binding
constants by a nonlinear least-squares method.14 These values are
shown in Table 1, together with the number of data, n, and the
standard deviation, s (mV):
n
s ) {( (Eobsd - Ecalcd)2/ (n - 1)}1/ 2
(7)
∑
The fitting procedure has already been reported in detail else-
where.15 The standard deviations for three native CDs are very
small, so that we need not take into consideration the other
stoichiometries. The standard deviations for the six modified CDs
are beyond experimental errors. This will be explicable by taking
into consideration that these modified CDs contain many homo-
logues different in binding constant. The binding constants for
most of the modified â-cyclodextrins are smaller than that for
â-CD. The reason for this decrease with a modification would be
the steric hindrance of the substituted group. The only exception
is CMâ-CD. Because this CD has a negative charge, its binding
ability with a positive OB ion will be enhanced by electrostatic
attraction.
ACKNOWLEDGMENT
We thank Dr. Satoshi Ito of DKK for kindly supplying an Ag/
AgBr electrode.
Received for review November 18, 1998. Accepted
February 10, 1999.
Effects of Cyclodextrins on the Bitter Taste Intensity of
OB. As Figure 6 shows, the bitter taste intensity of a 4 mM OB
solution is reduced by the addition of CD. The order of this
AC981286H
(15) Funasaki, N.; Yodo, H.; Hada, S.; Neya, S. Bull. Chem. Soc. Jpn. 1 9 9 2 , 65,
1323-30.
(16) Funasaki, N. Adv. Colloid Interface Sci. 1 9 9 3 , 43, 87-136.
(17) Connors, K. A. Chem. Rev. 1 9 9 7 , 97, 1325-57.
1736 Analytical Chemistry, Vol. 71, No. 9, May 1, 1999