Water accessibility to the binding cleft as a major switching factor from entropy-driven to enthalpy-driven binding of an alkyl group by synthetic receptors

Article abstract of DOI:10.1002/asia.200900679

Free energy, enthalpy, and entropy changes in the binding of alkyl pyridines to water-soluble zinc porphyrin receptors with varying accessibility of water to the binding cleft were determined to explain why the driving force of hydrophobic effects is enthalp-ic in some occasions and entropic in others. Zinc porphyrins bearing four alkyl pillars with terminal solubilizing poly(oxyethylene) (POE) chains of molecular weight of 750 (1), with eight alkyl pillars with terminal solubilizing POE chains of molecular weight of 350 (3), and with eight alkyl pillars with POE of molecular weight of 750 (4) had a binding cleft with decreasing water accessibility in this order as revealed by binding selectivity of imidazole/pyridine. Although all these porphyrins showed that the free energy of binding (-ΔG°) increases linearly as the alkyl group of the guest is lengthened (-ΔG° per CH₂ was 2.6, 2.8, and 2.6kJmol^-1 for 1, 3, and 4, respectively), the origin of the free energy gain was much different. Receptor 1 with the most hydrophilic binding site bound the alkyl group by an enthalpic driving force (4-pentylpyridine favored over 4-methylpyridine by ΔΔH° = -16.4 kJmol^-1), while receptor 4 with the most hydrophobic binding site by an entropic driving force (4-pentylpyridine favored over 4-methylpyridine by ΔΔ5° = 39.6 JK^-1mol^-1). Receptor 3 showed intermediate behavior: both enthalpic and entropic terms drove the binding of the alkyl group with the en-thalpic driving force being dominant. The binding site of the four-pillared receptor (1) is open and accessible to water molecules, and is more hydro-philic than that of the eight-pillared receptor (4). We propose that the alkyl chains of 1 are exposed to water to produce a room to accommodate the guest to result in enthalpy-driven hy-drophobic binding, whereas 4 can accommodate the guest without such structural changes to lead to entropy-driven hydrophobic binding. Therefore, accessibility of water or exposure of the binding site to the water phase switches the driving force of hydropho-bic effects from an entropic force to an enthalpic force. 2010 Wiley-VCH Verlag GmbH Co. KGaA, Weinheim.

Full text of DOI:10.1002/asia.200900679

DOI: 10.1002/asia.200900679
Water Accessibility to the Binding Cleft as a Major Switching Factor from
Entropy-Driven to Enthalpy-Driven Binding of an Alkyl Group by Synthetic
Receptors
Sayaka Matsumoto, Hiroya Iwamoto, and Tadashi Mizutani*[a]
Abstract: Free energy, enthalpy, and
entropy changes in the binding of alkyl
pyridines to water-soluble zinc porphy-
rin receptors with varying accessibility
of water to the binding cleft were de-
termined to explain why the driving
force of hydrophobic effects is enthalp-
ic in some occasions and entropic in
others. Zinc porphyrins bearing four
alkyl pillars with terminal solubilizing
poly(oxyethylene) (POE) chains of
molecular weight of 750 (1), with eight
alkyl pillars with terminal solubilizing
POE chains of molecular weight of 350
(3), and with eight alkyl pillars with
POE of molecular weight of 750 (4)
had a binding cleft with decreasing
water accessibility in this order as re-
vealed by binding selectivity of imida-
zole/pyridine. Although all these por-
phyrins showed that the free energy of
binding (ÀDGo) increases linearly as
the alkyl group of the guest is length-
ened (ÀDGo per CH2 was 2.6, 2.8, and
2.6 kJmolÀ1 for 1, 3, and 4, respective-
ly), the origin of the free energy gain
was much different. Receptor 1 with
the most hydrophilic binding site
bound the alkyl group by an enthalpic
driving force (4-pentylpyridine favored
over 4-methylpyridine by DDHo =
À16.4 kJmolÀ1), while receptor 4 with
the most hydrophobic binding site by
an entropic driving force (4-pentylpyri-
dine favored over 4-methylpyridine by
enthalpic and entropic terms drove the
binding of the alkyl group with the en-
thalpic driving force being dominant.
The binding site of the four-pillared re-
ceptor (1) is open and accessible to
water molecules, and is more hydro-
philic than that of the eight-pillared re-
ceptor (4). We propose that the alkyl
chains of 1 are exposed to water to
produce a room to accommodate the
guest to result in enthalpy-driven hy-
drophobic binding, whereas 4 can ac-
commodate the guest without such
structural changes to lead to entropy-
driven hydrophobic binding. Therefore,
accessibility of water or exposure of
the binding site to the water phase
switches the driving force of hydropho-
bic effects from an entropic force to an
enthalpic force.
DDSo =39.6 JKÀ1 molÀ1). Receptor
showed intermediate behavior: both
3
Keywords: hydrophobic effect
·
molecular recognition · porphyri-
noids · receptors · thermodynamics
Introduction
as well as chromatographic separation of biomolecules.[3]
However, the driving force of hydrophobic effects has been
ascribed to an enthalpic force[4,5] or to an entropic force[6,7]
depending on the supramolecular systems. It thus is an open
question: what is the real mechanism of hydrophobic ef-
fects?
Thermodynamics of hydrophobic effects can be rational-
ized by considering the solubilization of hydrocarbons in
water. Solubilization of gaseous hydrocarbons in water is
characterized by negative entropy changes,[8] which can be
explained by a mechanism through which water molecules
on the surface of a hydrocarbon are strongly hydrogen
bonded and lose motional freedom compared with those in
the bulk phase. Enthalpy changes of solubilization of gas-
eous hydrocarbons are positive or negative depending on
the solute molecule, but close to zero, and the entropy term
is dominant in determining the free-energy changes.[9] For
Hydrophobic interactions[1] are key to a number of biologi-
cal functions such as binding of organic molecules to pro-
teins or DNAs, and structural stabilization of the cell mem-
brane. An understanding of hydrophobic effects is essential
in diverse applications such as the rational design of drugs[2]
[a] S. Matsumoto, H. Iwamoto, Dr. T. Mizutani
Department of Molecular Chemistry and Biochemistry
Faculty of Science and Engineering, and Center for Nanoscience
Research
Doshisha University
Kyotanabe, Kyoto 610-0321 (Japan)
Fax : (+81)774-65-6794
E-mail: tmizutan@mail.doshisha.ac.jp
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.200900679.
Chem. Asian J. 2010, 5, 1163 – 1170
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1163

FULL PAPERS
an incremental increase in the number of methylene groups
in the dissolution of gaseous hydrocarbons in water, the
free-energy change (DGo), the enthalpy change (DHo), and
the value of ÀTDSo were +0.75 kJmolÀ1, À2.8 kJmolÀ1, and
3.56 kJmolÀ1, respectively, at 298.15 K.[1c,8b] These values in-
dicate that the unfavorable free-energy change was deter-
mined by the cancellation of larger contributions from fa-
vorable enthalpy and unfavorable entropy, and the origin of
the lower solubility of alkanes in water as a result of the ad-
ditional methylene group was ascribed to the fact that the
unfavorable entropic term dominates the favorable enthalp-
ic term. Association of hydrophobic molecules in water in-
volves the reverse process of solubilization of hydrocarbons
in water, since the solvent-accessible surface area is reduced
upon association of hydrophobic moieties. This mechanism
can account for positive entropy changes in binding by hy-
drophobic effects. In addition to the entropic term, solute–
solute attractive interactions such as attractive van der
Waals interactions, including Londonꢁs dispersion interac-
tions, make a favorable enthalpic contribution. Homans and
co-workers demonstrated that attractive van der Waals in-
teractions between the hydrophobic moieties of protein and
ligand made a significant contribution to the binding ther-
modynamics: the binding of alcohols to the major urinary
protein becomes enthalpically favorable as the alkyl chain
of the guest is longer.[10] Because the enthalpy and entropy
changes of binding of a signal molecule to a protein or of a
guest molecule to a synthetic receptor involve not only hy-
drophobic effects but also hydrogen bonding and conforma-
tional and translational entropy,[11] it is generally difficult to
extract the contribution from hydrophobic effects.[12]
Jencks[13] has made an interesting proposal that hydrophobic
bonds can be classified into two categories: a “classical hy-
drophobic bond” forms between nonpolar molecules and a
“nonclassical hydrophobic bond” between partially polar
molecules. The former hydrophobic bond is characterized by
negative entropy changes, whereas the latter hydrophobic
bond is characterized by negative enthalpy changes.
Schneider et al. suggested that discrimination of solvophobic
forces and solute–solute intermolecular forces, such as Lon-
donꢁs dispersion forces, is important to elucidate the binding
mechanism of nonpolar molecules in water, particularly to
explain the enthalpic and entropic contributions to the free-
energy changes in the binding.[14] If we determine differen-
tial enthalpy and entropy changes for a series of guests with
systematic variation in hydrophobicity, they should be a
good estimate of enthalpy and entropy owing to the hydro-
phobic effects. Similar perturbation studies to elucidate the
binding mechanism of organic molecules to carbonic anhy-
drase have been reported by Whiteside and co-workers.[15]
In previous papers,[7,16] we reported that the entropy term
is favorable and the enthalpy term unfavorable as the alkyl
group of 4-alkylpyridines is longer for binding of 4-alkylpyri-
dines to water-soluble zinc porphyrin 4. The entropic driving
Abstract in Japanese:
force is consistent with the picture of hydrophobic effects
described above, whereby the water molecules around the
hydrophobic alkyl chain obtain motional freedom upon con-
tact of the hydrophobic alkyl chain of the guest with those
of the receptors. We report herein that a similar receptor (1)
having a less hydrophobic binding site, which allows more
water molecules to penetrate into the binding cleft, showed
enthalpically driven binding of the alkyl group of guest. Re-
1164
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1163 – 1170

Water Accessibility to the Binding Cleft
ceptor 3, which has a binding site of intermediate hydropho-
bicity, showed both enthalpically and entropically driven hy-
drophobic effects, intermediate behavior between 1 and 4.
Thus, the hydrophobicity of the binding cleft—the extent
that water molecules are excluded from the binding cleft—is
strongly correlated with the driving force of hydrophobic ef-
fects.[17]
a measure of hydrophobicity of the binding site. These data
are compared with those determined for a similar receptor
4,[7] in which eight alkyl pillars with terminal POE of aver-
age molecular weight of 750 are used to construct a hydro-
phobic binding site.
UV/Visible Absorption Spectroscopic Studies on the
Aggregation of 1 in Water
Results and Discussion
A solution of 1 in dichloromethane showed a sharp Soret
band at 413 nm. A freshly prepared solution of 1 in a pH 7
potassium phosphate buffer at 258C showed splitting of the
Soret band at 421 and 416 nm, as shown in Figure 1. The
We prepared water-soluble synthetic receptors 1 and 3. The
receptors consist of three components, a zinc porphyrin,
alkyl pillars, and polyoxyethylene (POE) with an average
molecular weight of either 750 or 350 at each terminal of
the alkyl pillars. The zinc porphyrin moiety binds a nitroge-
neous base through Lewis acid–Lewis base interactions,[18]
and alkyl pillars provide a hydrophobic environment to the
binding site.[19] The POE groups were attached to the termi-
nal of alkyl pillars to solubilize the whole molecule in water
and to protect the binding cleft from water.[20] Because the
coordinating interaction between the pyridyl nitrogen atom
and the zinc atom is relatively strong, the orientation of the
guest in the host–guest complex is well-defined. The alkyl
group of 4-alkylpyridines should be forced to align with the
alkyl pillars when bound to the receptor (Scheme 1). By sys-
Figure 1. Variable temperature UV/Vis spectra of a solution of 1 in phos-
phate buffer at pH 7. The solution was heated from 258C to 708C.
splitting of the Soret band was not observed for a similar re-
ceptor 4. We ascribe the splitting of the Soret band to for-
mation of aggregates of 1 as a result of the intermolecular
hydrophobic effects of the porphyrin core and the alkyl
chains. When the solution was warmed up to 708C, the peak
at 421 nm decreased and disappeared, leaving a sharp band
at 416 nm. The Soret band remained sharp for a while after
the solution was cooled to room temperature. When the so-
lution was left overnight at room temperature, however, the
splitting of the Soret band was again observed. A plot of the
absorbance at 421 nm against temperature is shown in
Figure 2. The heating and cooling curve showed hysteresis,
revealing that aggregate formation is a kinetically slow pro-
cess. To avoid any errors in binding constants owing to ag-
gregation, we started the titration of the receptor solution
with guest just after the host solution was annealed at 708C
and cooled to the designated temperature.
Scheme 1. Schematic representation of orientation of alkylpyridine in the
binding cleft of receptor 3.
tematic variation of the alkyl chain length in 4-alkylpyri-
dines, we can estimate the hydrophobic interaction between
the alkyl chain and the hydrophobic wall of the receptor.
We determined the free-energy changes, the enthalpy
changes, and the entropy changes in the binding of 4-alkyl-
pyridines in water to evaluate the mechanism of hydropho-
bic effects, by using UV/Vis spectrophotometry and isother-
mal titration calorimetry.[21] We also determined the binding
constants for imidazoles and hydroxyalkylpyridines that
favor a hydrophilic binding site because of hydration of the
nitrogen atom and the hydroxy group, respectively. Compar-
ison of the binding affinity of imidazoles and pyridines gives
UV/Visible Absorption Spectroscopic Studies on the
Aggregation of 3 in Water
A solution of 3 in water (0.1m potassium phosphate buffer
at pH 7.0) showed a sharp Soret band at 258C. Upon heat-
ing the solution to 458C, the Soret peak was red-shifted
from 425 to 428 nm. When the solution was heated above
Chem. Asian J. 2010, 5, 1163 – 1170
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1165

T. Mizutani et al.
FULL PAPERS
Table 1. Binding constants (K) of alkylpyridines, w-hydroxyalkylpyri-
dines, and imidazoles to receptors 1–4 in water at 258C, 0.1m potassium
phosphate buffer pH 7.0.[a]
Host
Guest
K [mÀ1
]
1
1
1
1
1
1
1
1
1
1
2
2
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4-methylpyridine
4-ethylpyridine
4-propylpyridine
4-pentylpyridine
4-hydroxypyridine
4-hydroxymethylpyridine
4-(2-hydroxyethyl)pyridine
4-(3-hydroxypropyl)pyridine
N-methylimidazole
N-ethylimidazole
4-methylpyridine
4-ethylpyridine
4-methylpyridine
4-ethylpyridine
4-propylpyridine
4-pentylpyridine
N-methylimidazole
N-ethylimidazole
4-methylpyridine
14300
46000
130000
960000
11
11000
2500
6000
730
920
38400[b]
103000[b]
7700
Figure 2. Plot of the absorbance at 421 nm against the temperature of the
solution of 1 in pH 7 phosphate buffer.
21000
64000
640000
80
99
558C, the Soret band became broadened, and on further
heating to 858C the Soret peak became diminished, indicat-
ing that some of 3 was phase-separated. Therefore, 3 forms
aggregates at a higher temperature, which is contrasting be-
havior to that of 1, which aggregates at lower temperature.
The aggregation of 3 at higher temperature is similar to
lower critical solution temperature behavior observed for
neutral surfactant containing POE groups such as Triton-
X[22] and water-soluble polymers with a hydrophobic side
chain such as poly(N-isopropylacrylamide).[23]
19000[c]
54200[c]
148000[c]
13
4-ethylpyridine
4-propylpyridine
4-hydroxypyridine
4-hydroxymethylpyridine
4-(2-hydroxyethyl)pyridine
4-(3-hydroxypropyl)pyridine
N-methylimidazole
N-ethylimidazole
5200
2000
6300
150
260
[a] [1]=2.8ꢃ10À6 m. Binding constants are averages of three to four inde-
pendent determinations, and errors of the mean are estimated to be 5%.
[b] Taken from Ref. [16]. [c] Taken from Ref. [7].
Hydrophobic Interaction Free Energy
The binding constants of 4-alkylpyridines to receptors 1 and
3 were determined by UV/Vis spectroscopic titration experi-
ments. Addition of stock solutions of 4-alkylpyridines to an
aqueous solution of 1 in pH 7.0 phosphate buffer caused a
red shift in the Soret band. In the difference spectra, the ab-
sorbance at 427 nm increased with decreasing absorbance at
410 nm. Similarly, the addition of 4-methylpyridine to 3 in
pH 7.0 phosphate buffer resulted in a difference spectrum
with decreased absorbance at 425 nm and increased absorb-
ance at 438 nm. The spectral changes were similar to those
observed for coordination of amines to zinc porphyrins.[24]
Curve fitting to the 1:1 binding isotherm was performed
using the least-squares method to obtain binding constants
(Table 1). The values of the binding free energy of receptors
1 and 3 are plotted against the number of methylene groups
(n) in the guest in Figure 3. From the least-squares fit to the
data, we obtained the slope of the line, which gives the free-
energy stabilization for an additional CH2 group on the
guest. For receptors 1 and 3, the free-energy stabilization for
Figure 3. Plot of the binding free energy of 4-alkylpyridines to 1 (*) and
3 (&) against the number of CH2 groups (n) in the guest. The lines ob-
tained by the least-squares fitting is shown: ÀDGo/kJmolÀ1 =2.6 n+21.2
for 1 and ÀDGo/kJmolÀ1 =2.8 n+19.3 for 3.
an additional CH2 group on the guest, Àd
(DGo)/dn, was 2.6
ACHTUNGTRENNUNG
and 2.8 kJmolÀ1, respectively. The free-energy increments
per methylene group (namely, the slope of the lines) for re-
ceptors 1–5 are listed in Table 2. The values of Àd
N
of Àd
(DGo)/dn was in the range of 2.4–3.2 kJmolÀ1.[25] As-
ACHTUNGTRENNUNG
suming the solvent accessible surface area of a CH2 group is
31.8 ꢂ2, the values of 2.2 kJmolÀ1 and 3.4 kJmolÀ1 for
ÀdDGo/dn correspond to the hydrophobic interaction free
1166
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1163 – 1170

Water Accessibility to the Binding Cleft
Table 2. The binding free energy increment per CH2 group (Àd
(DGo)/dn, kJmolÀ1) and the binding free
ACHTUNGTRENNUNG
hydrophilic/hydrophobic
ance is important for the design
of synthetic receptors.
bal-
energy extrapolated to n=0 (ÀDGo (n=0)) for POE–Zn porphyrin conjugates 1–5 having either four or eight
POE alkyl chains and POE molecular weight of 350, 750, or 2000.
Receptor Number of
alkyl pillars
Mol. Wt. of each
POE group
Àd
G
ÀDGo (n=0)
Mol. Wt. POE/Mol. Wt.
whole mol.
[kJmolÀ1
[kJmolÀ1
]
Enthalpy and Entropy Changes
Associated with Hydrophobic
Effects
1
2
3
4
5
4
4
8
8
8
750
2000
350
750
2000
2.6
3.4
2.8
2.6
2.2
21.2
20.2
19.3
21.8
24.4
0.70
0.86
0.55
0.73
0.88
Although the recognition free
energy of the alkyl chain by re-
ceptor 1 is similar to that of re-
energy per ꢂ2 contact area of 0.069 and 0.107 kJmolÀ1 per
ꢂ2, respectively.
ceptor 4, the enthalpic and entropic contributions are much
different. The enthalpy changes and the entropy changes
were determined by van’t Hoff analysis of the binding con-
stants in the temperature range 15–458C as well as isother-
mal titration calorimetry. The values of the enthalpy changes
and the entropy changes determined by isothermal titration
calorimetry were close to those determined by the van’t
Hoff analysis. However, reproducibility of data obtained by
the isothermal titration calorimetry was poorer, presumably
because aggregation of the receptors occurs for higher con-
centrations of 1. Thus, we used the data obtained by the
van’t Hoff analysis, in which low concentrations of the re-
ceptors can be used to avoid any aggregation. The values of
the enthalpy changes and the entropy changes of binding of
4-alkylpyridines to 1, 3, and 4 are summarized in Table 3.
The values of TDSo are plotted against DHo for 1, 3, and 4
in Figure 4, and the arrows indicate the direction of the in-
crease in the guest alkyl chain length. Interestingly, the en-
thalpy term becomes favorable and the entropy term slightly
unfavorable as the alkyl chain of the guest is longer for re-
ceptor 1, whereas the entropy term becomes favorable and
the enthalpy term becomes unfavorable as the alkyl chain is
lengthened for receptor 4. It is noteworthy that the hydro-
phobic effects are characterized by an increase in entropy in
eight-pillared receptor 4 but by a decrease in enthalpy in
four-pillared receptor 1. Receptor 3 showed intermediate
behavior: both enthalpy and entropy terms become favora-
ble as the alkyl chain is lengthened, and the enthalpy term
changes more significantly.
The hydrophobic free energy per unit contact surface area
has been evaluated on the basis of 1) solubility of alkanes
and amino acids, 2) the binding data of ligand–protein com-
plexes, 3) the binding data of host–guest complexes, and
4) partition coefficients between organic solvent and water.
Hermann,[26] Tanford and co-workers,[27] and Sharp et al.[28]
estimated the hydrophobic surface free energy of alkanes in
water to be 0.14, 0.10, and 0.2 kJmolÀ1 per ꢂ2, respectively,
based on solubility data of hydrocarbons in water. Ri-
chards[29] stated in his review that 0.084 kJmolÀ1 per ꢂ2 has
a fair chance of being an appropriate value for the hydro-
phobic free energy. Boehm[30] reported that the lipophilic
contact between protein and a ligand results in 0.1 and
0.17 kJmolÀ1 per ꢂ2, respectively. Cohen and Connors[31] re-
ported the lipophilic free energy per ꢂ2 contact area is
0.38 kJmolÀ1 on the basis of the binding free energies of
host–guest systems, although the analysis was performed for
relatively simple organic host–guest systems such as benzene
derivative–anthracene derivative complexes, and the binding
constants were in the range 2–230mÀ1. Higuchi and co-work-
ers[32] reported that there was a constant increment of
3.8 kJmolÀ1 in the free energy for each additional increment
in methylene group for anions based on the ion pair extrac-
tion experiments. This value corresponds to 0.12 kJmolÀ1
per ꢂ2. The hydrophobic free energies per ꢂ2 are thus in a
rather broad range, depending on what host–guest complex-
ation or solubility data were employed to derive the energy.
The hydrophobic free energy observed in our receptor–
ligand complexes is comparable or somewhat smaller than
these reported values.
As shown in the plots in Figure 4, the entropy changes are
relatively constant for 1 and the enthalpy changes are rela-
The ratios of the molecular
Table 3. Enthalpy changes and entropy changes of binding of 4-alkylpyridines to porphyrin receptors 1, 3, and
weight of the POE moieties to
the molecular weight of the
whole molecule are shown in
Table 2. On the basis of this
ratio, 5 has the largest propor-
tion of POE and this value de-
creases in the order 5>2>4>
1>3. The largest value of Àd-
4.[a]
[c]
Porphyrin
Guest
DHo [kJmolÀ1
]
DSo [JKÀ1 molÀ1
]
DGo [kJmolÀ1
]
1
1
1
3
3
3
4
4
4
4-methylpyridine
4-propylpyridine
4-pentylpyridine
4-methylpyridine
4-ethylpyridine
4-propylpyridine
4-methylpyridine
4-propylpyridine
4-pentylpyridine
À23.6Æ2.4[b]
À33.7Æ0.9
À40.0Æ1.7
À15.0Æ2.3
À17.5Æ2.4
À19.4Æ1.0
À24.5Æ1.2[d]
À23.3Æ0.5[d]
À22.8Æ0.2[d]
0.6Æ7.9b
À23.8
À29.2
À34.3
À22.2
À24.7
À27.4
À24.4
À29.4
À34.4
-15.1Æ2.9
-19.3Æ5.3
23.7Æ7.6
24.4Æ8.1
26.7Æ3.3
À0.6Æ4.7[d]
19.8Æ1.7[d]
39.0Æ0.3[d]
ACHTUNGTRENNUNG
(DGo)/dn for receptor 2 implies
that the receptor with an inter-
mediate ratio of POE and other
moieties had the best binding
site for an alkyl group; that is, DHo and DSo with T=298 K. [d] Taken from Ref. [7].
[a] Standard deviations of DHo and DSo were calculated according to literature.[33] [b] Isothermal calorimetry
gave the following data: DHo =À25.5 kJmolÀ1, DSo =À5.31 JKÀ1 molÀ1, [1]=4.2ꢃ10À5 m. [c] Calculated from
Chem. Asian J. 2010, 5, 1163 – 1170
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1167

T. Mizutani et al.
FULL PAPERS
cussed in the next section, that water accessibility to the
binding cleft of a-cyclodextrin could be similar to that of re-
ceptor 1. The binding site of a-cyclodextrin is rather open,
resulting in the enthalpy-driven binding.
Imidazole/Pyridine Selectivity as a Measure of
Hydrophobicity of the Binding Site
The hydrophobicity of the binding site can be estimated by
comparing the binding constants of imidazoles with those of
pyridines. Because imidazoles have two nitrogen atoms, and
the nitrogen atom not bound to the zinc center should be
strongly hydrated, the binding of imidazoles should be in-
hibited if the binding site is hydrophobic and resistant to hy-
dration. The ratios of binding constant of 4-methylpyridine
to that of N-methylimidazole and those of 4-ethylpyridine to
that of N-ethylimidazole are listed in Table 4. The ratio K(4-
Figure 4. Plot of TDSo (T=298 K) against DHo for binding of 4-alkylpyri-
dines to 1, 3, and 4. Arrows indicates the direction of changes as the
length of the alkyl group, and thus the binding free energy ÀDGo, in-
creases. Bars indicate standard deviations.
Table 4. Ratios of binding constants of pyridines to imidazoles as a
probe for the polar/nonpolar environment of the binding site.
1
3
4
5
tively constant for 4, showing that there is little enthalpy–en-
tropy compensation.[34] If enthalpy–entropy compensation
was observed, the free-energy changes should be almost
constant, even if there were large changes in enthalpy and
entropy. In the present case, the changes in enthalpy or en-
tropy were directly reflected in the free-energy changes. As
a component of the free energy, the enthalpy term is pre-
dominant in the alkyl recognition free energy for receptors
1 and 3, whereas the entropy term is predominant for recep-
tor 4. Receptor 3 showed intermediate behavior between 1
and 4, whereby recognition of the alkyl group was driven by
an enthalpic term to a lesser extent than 1. As discussed in
the next section, the binding cleft of 1 is accessible to water,
whereas that of 4 is protected from water. The binding cleft
of 3 showed intermediate accessibility of water. Therefore,
we suggest that the hydrophobicity of the binding cleft—the
extent of exclusion of water molecules from the binding
cleft—is the major factor that determines whether binding is
driven by an enthalpic term or by an entropic term. The sug-
gested mechanism is different from the classical versus non-
classical hydrophobic bond as suggested by Jencks.[13] Ac-
cording to his mechanism, the partially polar nature of the
solute molecules would result in enthalpy-driven nonclassi-
cal hydrophobic bond. In the present systems, we compared
the binding thermodynamics for the same guest molecules
and we focused on the interaction with an alkyl group, so
that the difference in the driving force should be ascribed to
the difference in solvation.
K(4-methylpyridine)/K(N-methylimidazole)
K(4-ethylpyridine)/K(N-ethylimidazole)
20
50
96
212
124
210
164
340
methylpyridine)/K(N-methylimidazole) is 124 for receptor 4
whereas it is 20 and 96 for receptors 1 and 3, respectively.
These results suggest that the hydrophilicity of the binding
site of the receptors, water accessibility to the binding cleft,
decreases in the order: 1>3>4. The binding site becomes
more hydrophilic as the number of alkyl pillars decreases
and the molecular weight of the POE groups is lower. This
order of water accessibility of binding sites is not identical
to the order of affinity to alkyl groups as measured by
ÀdDGo/dn (Table 2). The imidazole to pyridine selectivity
reflects hydrophobicity near the porphyrin plane, whereas
the value of ÀdDGo/dn reflects the hydrophobicity at some
point distant from the porphyrin plane.
Binding of w-Hydroxyalkylpyridines
We reported that the presence of a polar substituent in the
alkyl chain of alkylpyridines considerably inhibits binding.[16]
The ratios of binding constants of alkylpyridines to w-hy-
droxyalkylpyridines are 1.3 (1–Me-py), 18 (1–Et-py), 22 (1–
Pr-py), 3.7 (4–Me-py), 27 (4–Pr-py), and 23 (4–Pr-py). Thus,
the binding constants of 4-(2-hydroxyethyl)pyridine and 4-
(3-hydroxypropyl)pyridine become one order of magnitude
smaller than those of nonhydroxylated guests. These ratios
amount to the free-energy penalty of binding of 7.2 to
8.2 kJmolÀ1 for ethyl and propylpyridines. A similar thermo-
dynamic penalty was reported for the binding of 1-octanol
and 1,8-octandiol to the major urinary protein: a decrease
of the binding free energy of 18 kJmolÀ1 was observed when
the polar OH group was forced to enter to a hydrophobic
binding site.[35] The effects are insignificant for 4-methylpyri-
It would be interesting to compare the thermodynamic
data obtained herein with those reported for binding of
simple aliphatic alcohols to a-cyclodextrin. Spencer et al.[5c]
reported that the binding increment for a methylene group
was DDGo =À3.0 kJmolÀ1 and DDHo =À3.83 kJmolÀ1.
These values are close to the enthalpy and entropy changes
observed for
1
(DDGo =À2.6 kJmolÀ1 and DDHo =
À4.1 kJmolÀ1). Therefore, the comparison implies, as dis-
1168
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1163 – 1170

Water Accessibility to the Binding Cleft
dine: 4-hydroxymethylpyridine binds to receptors 1 and 4
with relatively strong affinity. The high affinity of 4-hydroxy-
methylpyridine is somewhat surprising, and polar OH
groups comfortably enter the binding site. Molecular dy-
namics calculations of the receptor 1–4-hydroxymethylpyri-
dine complex suggest that the binding site of receptor 1 is
rather compact, and the hydroxy group of the 4-hydroxy-
methyl group is hydrogen bonded to the oxyethylene group
of the receptor (see Figure S1 in the Supporting Informa-
tion).
contrast, receptor 4 has enough POE moieties to protect the
binding cleft from water, and the conformational changes
have little effect on the values of the enthalpy and entropy
changes. We expect that desolvation of the guest alkyl chain
upon binding to 4 is more extensive. The hydrophobic/hy-
drophilic balance of the receptors altered the accessibility of
water to the binding site, and they significantly affected the
enthalpic and entropic contribution to the binding thermo-
dynamics.
Conclusions
Conformational Changes, Guest Desolvation, and
Receptor–Guest Interactions upon Binding
The binding of an alkyl group to the hydrophobic site in a
synthetic receptor was driven by either an enthalpic force or
an entropic force. Entropically driven binding was explained
by a classical picture of hydrophobic effects, where water
molecules around the nonpolar molecular surface are re-
leased to bulk water phase to lead to increase in entropy.
However, attractive van der Waals interactions between the
nonpolar surfaces of receptor and ligand also make a signifi-
cant contribution to the enthalpic driving force, as suggested
for nonclassical hydrophobic interactions. Therefore, the
overall thermodynamic parameters are determined by a
subtle balance of enthalpic and entropic driving forces, and
other factors such as conformational changes in the receptor
upon binding switch the driving force from being entropy-
driven to being enthalpy-driven. In our work, we compared
the binding mechanism of an alkyl group among the three
receptors. Receptor 1 has only four alkyl pillars and the
binding site is open, whereas receptor 4 has eight alkyl pil-
lars and the binding site should
In an aqueous solution, receptors 1 and 4 should have a
folded conformation whereby the alkyl chains are in contact
with each other and with the porphyrin plane to form a non-
polar core and the core is surrounded by the POE moieties
to avoid direct contact with water. To bind a guest in the
binding site of receptors 1, 3, and 4, conformational changes
in the alkyl pillar groups should occur to create a space to
accommodate the guest (Scheme 2). We propose that the
alkyl chains of receptor 1 become more exposed to water
when the guest is bound to accommodate and encompass
the alkylpyridine in water. The POE auxiliary groups of 1
could be insufficient to cover the increased hydrophobic sur-
face. Thus, the binding of hydrophobic guest caused more
water molecules to have contact with hydrophobic surfaces
and hence result in a negative entropic change. Because
there are some water molecules in the binding site of 1, only
partial desolvation of the guest alkyl chain may occur. In
be well protected from water.
Receptor 3 has also eight alkyl
pillars, as well as shorter POE
moieties than in 4. Aggregation
behavior and imidazole/pyri-
dine selectivity indicated that
receptor 4 has a hydrophobic
binding site in which the acces-
sibility of water molecules is
low, whereas receptor
1 has
some water in its binding site
and a less hydrophobic binding
site. The hydrophobicity of re-
ceptor 3 was intermediate be-
tween 1 and 4. We demonstrat-
ed that receptor 4 having a hy-
drophobic binding site recog-
nizes an alkyl group by an en-
tropic driving force, whereas
receptor 1 having a less hydro-
phobic binding site recognizes
an alkyl group by an enthalpic
driving force. Receptor
showed intermediate behavior
between 1 and 4. Therefore, the
3
Scheme 2. Schematic representation of the binding of guest to receptors 1 and 4. In receptor 1, water mole-
cules enter into the binding site to result in enthalpy-driven binding, whereas in receptor 4 complete dehydra-
tion occurs to result in entropy-driven binding.
Chem. Asian J. 2010, 5, 1163 – 1170
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1169

T. Mizutani et al.
FULL PAPERS
[15] a) J. Gao, S. Qiao, G. M. Whitesides, J. Med. Chem. 1995, 38, 2292;
b) V. M. Krishnamurthy, B. R. Bohall, V. Semetey, G. M. Whitesides,
J. Am. Chem. Soc. 2006, 128, 5802; c) V. M. Krishnamurthy, B. R.
Bohall, C.-Y. Kim, D. T. Moustakas, D. W. Christianson, G. M.
Whitesides, Chem. Asian J. 2007, 2, 94; d) V. M. Krishnamurthy,
G. K. Kaufman, A. R. Urbach, I. Gitlin, K. L. Gudiksen, D. B.
Weibel, G. M. Whitesides, Chem. Rev. 2008, 108, 946.
accessibility of water to the binding site switches the driving
force from entropy to enthalpy.
Experimental Section
Experimental details are reported in the Supporting Information.
[16] T. Mizutani, K. Kozake, K. Wada, S. Kitagawa, Chem. Commun.
2003, 2918.
[17] a) G. A. Weiland, K. P. Minneman, P. B. Molinoff, Nature 1979, 281,
114; b) P. Gilli, V. Ferretti, G. Gilli, P. A. Borea, J. Phys. Chem.
1994, 98, 1515; c) K. N. Houk, A. G. Leach, S. P. Kim, X. Zhang,
Angew. Chem. 2003, 115, 5020; Angew. Chem. Int. Ed. 2003, 42,
4872.
[18] a) C. Verchere-Beaur, E. Mikros, M. Perree-Fauvet, A. Gaudemer,
J. Inorg. Biochem. 1990, 40, 127; b) T. Mizutani, K. Wada, S. Kitaga-
wa, J. Am. Chem. Soc. 1999, 121, 11425; c) H. Imai, K. Misawa, H.
Munakata, Y. Uemori, Chem. Pharm. Bull. 2008, 56, 1470.
[19] T. Mizutani, K. Wada, S. Kitagawa, Chem. Commun. 2002, 1626.
[20] a) J. L. Sessler, N. A. Tvermoes, D. M. Guldi, T. D. Mody, W. E.
Allen, J. Phys. Chem. A 1999, 103, 787; b) V. Krꢄl, J. Davis, A. An-
drievsky, J. Kralova, A. Synytsya, P. Pouckova, J. L. Sessler, J. Med.
Chem. 2002, 45, 1073.
[21] For recent studies on the binding of an alkyl group by a synthetic re-
ceptor, see: a) G. Castronuovo, M. Niccoli, J. Inclusion Phenom.
Macrocyclic Chem. 2005, 53, 69; b) Y. H. Ko, H. Kim, Y. Kim, K.
Kim, Angew. Chem. 2008, 120, 4174; Angew. Chem. Int. Ed. 2008,
47, 4106.
[22] a) T. Nakagawa in Nonionic Surfactants (Ed.: M. J. Schick), Marcel
Dekker, New York, 1966, pp. 558–603; b) R. A. Mackay in Non-
ionic Surfactants.Physical Chemistry, Vol. 23 (Ed.: M. J. Schick),
Marcel Dekker, New York, 1987, pp. 297; c) M. J. Rosen, Surfactants
and Interfacial Phenomena, 2nd ed., Wiley, New York, 1989.
[23] C. de Las Heras Alarcon, S. Pennadam, C. Alexander, Chem. Soc.
Rev. 2005, 34, 276.
[24] T. Mizutani, K. Wada, S. Kitagawa, J. Org. Chem. 2000, 65, 6097.
[25] R. Murakami, A. Minami, T. Mizutani, Org. Biomol. Chem. 2009, 7,
1437.
[26] R. B. Hermann, J. Phys. Chem. 1972, 76, 2754.
[27] J. A. Reynolds, D. B. Gilbert, C. Tanford, Proc. Natl. Acad. Sci. USA
1974, 71, 2925.
[28] a) K. A. Sharp, A. Nicholls, R. Friedman, B. Honig, Biochemistry
1991, 30, 9686; b) K. A. Sharp, A. Nicholls, R. F. Fine, B. Honig, Sci-
ence 1991, 252, 106.
[29] F. M. Richards, Annu. Rev. Biophys. Bioeng. 1977, 6, 151.
[30] H.-J. Bçhm, J. Comput.-Aided Mol. Des. 1994, 8, 243.
[31] J. L. Cohen, K. A. Connors, J. Pharm. Sci. 1970, 59, 1271.
[32] S. M. J. Harris, T. Higuchi, J. H. Rytting, J. Phys. Chem. 1973, 77,
2694.
Acknowledgements
This work was supported by “Creating Research Center for Advanced
Molecular Biochemistry”, Strategic Development of Research Infrastruc-
ture for Private Universities, the Ministry of Education, Culture, Sports,
Science and Technology (MEXT), Japan.
[1] a) H. S. Frank, M. W. Evans, J. Chem. Phys. 1945, 13, 507; b) W.
Kauzmann, Adv. Protein Chem. 1959, 14, 1; c) C. Tanford, The Hy-
drophobic Effect, Wiley, New York, 1973; d) W. Blokzijl, J. B. F. N.
Engberts, Angew. Chem. 1993, 105, 1610; Angew. Chem. Int. Ed.
Engl. 1993, 32, 1545; e) J. B. F. N. Engberts, M. J. Blandamer, Chem.
Commun. 2001, 1701; f) L. R. Pratt, A. Pohorille, Chem. Rev. 2002,
102, 2671; g) N. T. Southall, K. A. Dill, A. D. J. Haymet, J. Phys.
Chem. B 2002, 106, 521; h) B. Widom, P. Bhimalapuram, K. Koga,
Phys. Chem. Chem. Phys. 2003, 5, 3085; i) K. A. Dill, T. M. Truskett,
V. Vlachy, B. Hribar-Lee, Annu. Rev. Biophys. Biomol. Struct. 2005,
34, 173.
[2] G. M. Whitesides, V. M. Krishnamurthy, Q. Rev. Biophys. 2005, 38,
385.
[3] M. Turowski, N. Yamakawa, J. Meller, K. Kimata, T. Ikegami, K.
Hosoya, N. Tanaka, E. R. Thornton, J. Am. Chem. Soc. 2003, 125,
13836.
[4] For comprehensive reviews on the mechanism of hydrophobic inter-
actions in host–guest systems, see: a) M. V. Rekharsky, Y. Inoue,
Chem. Rev. 1998, 98, 1875; b) H.-J. Schneider, A. Yatsimirsky, Prin-
ciples and Methods in Supramolecular Chemistry, Wiley, Ltd., Chi-
chester, 2000.
[5] a) R. I. Gelb, L. M. Schwartz, B. Cardelino, H. S. Fuhrman, R. F.
Johnson, D. A. Laufer, J. Am. Chem. Soc. 1981, 103, 1750; b) D. B.
Smithrud, T. B. Wyman, F. Diederich, J. Am. Chem. Soc. 1991, 113,
5420, and references therein; c) J. N. Spencer, J. DeGarmo, I. M.
Paul, Q. He, X. Ke, Z. Wu, C. H. Yoder, S. Chen, J. E. Mihalick, J.
Solution Chem. 1995, 24, 601; d) G. Arena, A. Contino, T. Fujimoto,
D. Sciotto, Y. Aoyama, Supramol. Chem. 2000, 11, 279.
[6] R. J. Hooley, H. J. V. Anda, J. Rebek, Jr., J. Am. Chem. Soc. 2007,
129, 13464.
[33] J. N. Miller, J. C. Miller, Statistics and Chemometrics for Analytical
Chemistry, Prentice-Hall Publishing Inc., Englewood Cliffs, 2000;
see also Ref. [25].
[7] H. Iwamoto, T. Mizutani, K. Kano, Chem. Asian J. 2007, 2, 1267.
[8] a) R. A. Pierotti, J. Phys. Chem. 1965, 69, 281; b) M. H. Abraham, J.
Am. Chem. Soc. 1982, 104, 2085.
[34] For enthalpy–entropy compensation, see: a) R. Lumry, S. Rajender,
Biopolymers 1970, 9, 1125; b) M. S. Searle, D. H. Williams, J. Am.
Chem. Soc. 1992, 114, 10690; c) Y. Inoue, T. Hakushi, Y. Liu, L.
Tong, B. Shen, D. Jin, J. Am. Chem. Soc. 1993, 115, 475; d) W.
Linert, Chem. Soc. Rev. 1994, 23, 429; e) J. D. Dunitz, Chem. Biol.
1995, 2, 709; f) D. H. Williams, M. S. Westwell, Chem. Soc. Rev.
1998, 27, 57; g) O. Exner, Chem. Commun. 2000, 1655; h) K. Sharp,
Protein Sci. 2001, 10, 661; i) T. Mizutani, K. Wada, S. Kitagawa, J.
Am. Chem. Soc. 2001, 123, 6459; j) A. Cornish-Bowden, J. Biosci.
2002, 27, 121; k) K. Wada, T. Mizutani, H. Matsuoka, S. Kitagawa,
Chem. Eur. J. 2003, 9, 2368; l) D. M. Ford, J. Am. Chem. Soc. 2005,
127, 16167.
[9] a) C. McAuliffe, J. Phys. Chem. 1966, 70, 1267; b) F. Franks, M.
Gent, H. H. Johnson, J. Chem. Soc. 1963, 2716.
[10] R. Malham, S. Johnstone, R. J. Bingham, E. Barratt, S. E. V. Phillips,
C. A. Laughton, S. W. Homans, J. Am. Chem. Soc. 2005, 127, 17061.
[11] a) W. P. Jencks, Proc. Natl. Acad. Sci. USA 1981, 78, 4046; b) M. I.
Page, W. P. Jencks, Proc. Natl. Acad. Sci. USA 1971, 68, 1678;
c) M. I. Page, Angew. Chem. 1977, 89, 456; Angew. Chem. Int. Ed.
Engl. 1977, 16, 449; d) D. H. Williams, E. Stephens, D. P. OꢁBrien,
M. Zhou, Angew. Chem. 2004, 116, 6760; Angew. Chem. Int. Ed.
2004, 43, 6596.
[12] M. A. Cheema, P. Taboada, S. Barbosa, Emilio Castro, M. Siddiq, V.
Mosquera, Biomacromolecules 2007, 8, 2576.
[13] W. P. Jencks, Catalysis in Chemistry and Enzymology, McGraw-Hill
Book Company, New York, 1969.
[35] E. Barratt, A. Bronowska, J. Vondrasek, J. Cerny, R. Bingham, S.
Phillips, S. W. Homans, J. Mol. Biol. 2006, 362, 994.
[14] a) H.-J. Schneider, L. Tianjun, M. Sirish, V. Malinovski, Tetrahedron
2002, 58, 779; b) H.-J. Schneider, Angew. Chem. 2009, 121, 3982;
Angew. Chem. Int. Ed. 2009, 48, 3924.
Received: November 28, 2009
Published online: April 8, 2010
1170
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1163 – 1170