Molecular Recognition of Amines and Amino Esters by Zinc Porphyrin Receptors: Binding Mechanisms and Solvent Effects

Article abstract of DOI:10.1021/jo000557x

Zinc porphyrin receptors bearing 12 ester groups in the meso phenyl groups (1-3) were prepared, and binding of amines and α-amino esters was studied with emphasis on the binding mechanisms. The X-ray crystallographic analysis of 5,10,15,20-tetrakis(2,6-bis(carbomethoxymethoxy)-4-carbomethoxyphenyl)porphyrin (free base of 1) showed that the receptor has a binding pocket above the porphyrin plane. UV-visible titration experiments revealed that the zinc porphyrin receptors bound amines and α-amino esters with binding constants (K_a) ranging from 0.5 to 52 700 M^-1 in CH₂Cl₂ at 25°C. The ester functional groups of 1 assisted the binding of aromatic α-amino esters (K_a = 8 000-23 000 M^-1 in CH₂Cl₂ at 25°C) and inhibited the binding of bulky aliphatic α-amino esters (K_a = 460 M^-1 for Leu-OMe in CH₂Cl₂ at 25°C), ndicating that CH-π type interactions and steric repulsions control the selectivity. The binding of amines and α-amino esters was tight both in a nonpolar solvent (CH₂Cl₂) and in a polar solvent (water) but loose in a solvent of intermediate polarity (H₂O-MeOH (1:1)), demonstrating that two competitive driving forces are operating: (1) attractive electrostatic forces between host and guest such as coordination of the amino group to the zinc atom, and (2) entropic forces stemming from desolvation as well as enthalpic forces due to the host-guest dispersion forces. The former forces drive the binding in CH₂Cl₂ while the latter forces drive the binding in water. The enthalpy changes in the binding in CH₂Cl₂ and those in water range from -50 to -30 kJ mol^-1 and from -35 to 0 kJ mol^-1, respectively. The entropy changes in CH₂Cl₂ and those in water range from -120 to -60 J K^-1 mol^-1 and from -50 to +60 J K^-1 mol^-1, respectively. Thus the binding in CH₂Cl₂ is characterized by large negative enthalpy changes, while that in water by less negative entropy changes. These thermodynamic parameters also indicate that host-guest polar interactions (enthalpic forces) drive the binding in CH₂Cl₂ while both host-guest dispersion interactions (an enthalpic force) and desolvation (an entropic force) drive the binding in water. Enthalpy-entropy compensation observed for the binding in water indicates that the binding of amines and amino esters in water by zinc porphyrins is associated with conformational changes as well as a high degree of dehydration. In CH₂Cl₂, no clear compensation was observed, consistent with the mechanism that neither desolvation processes nor conformational changes contribute significantly to the binding energetics.

Full text of DOI:10.1021/jo000557x

J . Org. Chem. 2000, 65, 6097-6106
6097
Molecu la r Recogn ition of Am in es a n d Am in o Ester s by Zin c
P or p h yr in Recep tor s: Bin d in g Mech a n ism s a n d Solven t Effects
Tadashi Mizutani,* Kenji Wada, and Susumu Kitagawa
Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering,
Kyoto University, Yoshida, Sakyo-ku, Kyoto 606-8501 J apan
mizutani@sbchem.kyoto-u.ac.jp.
Received April 12, 2000
Zinc porphyrin receptors bearing 12 ester groups in the meso phenyl groups (1-3) were prepared,
and binding of amines and R-amino esters was studied with emphasis on the binding mechanisms.
The X-ray crystallographic analysis of 5,10,15,20-tetrakis(2,6-bis(carbomethoxymethoxy)-4-car-
bomethoxyphenyl)porphyrin (free base of 1) showed that the receptor has a binding pocket above
the porphyrin plane. UV-visible titration experiments revealed that the zinc porphyrin receptors
bound amines and R-amino esters with binding constants (Ka) ranging from 0.5 to 52 700 M-1 in
CH2Cl2 at 25 °C. The ester functional groups of 1 assisted the binding of aromatic R-amino esters
(Ka ) 8 000-23 000 M-1 in CH2Cl2 at 25 °C) and inhibited the binding of bulky aliphatic R-amino
esters (Ka ) 460 M-1 for Leu-OMe in CH2Cl2 at 25 °C), indicating that CH-π type interactions and
steric repulsions control the selectivity. The binding of amines and R-amino esters was tight both
in a nonpolar solvent (CH2Cl2) and in a polar solvent (water) but loose in a solvent of intermediate
polarity (H2O-MeOH (1:1)), demonstrating that two competitive driving forces are operating: (1)
attractive electrostatic forces between host and guest such as coordination of the amino group to
the zinc atom, and (2) entropic forces stemming from desolvation as well as enthalpic forces due to
the host-guest dispersion forces. The former forces drive the binding in CH2Cl2 while the latter
forces drive the binding in water. The enthalpy changes in the binding in CH2Cl2 and those in
water range from -50 to -30 kJ mol-1 and from -35 to 0 kJ mol-1, respectively. The entropy
changes in CH2Cl2 and those in water range from -120 to -60 J K-1 mol-1 and from -50 to +60
J K-1 mol-1, respectively. Thus the binding in CH2Cl2 is characterized by large negative enthalpy
changes, while that in water by less negative entropy changes. These thermodynamic parameters
also indicate that host-guest polar interactions (enthalpic forces) drive the binding in CH2Cl2 while
both host-guest dispersion interactions (an enthalpic force) and desolvation (an entropic force)
drive the binding in water. Enthalpy-entropy compensation observed for the binding in water
indicates that the binding of amines and amino esters in water by zinc porphyrins is associated
with conformational changes as well as a high degree of dehydration. In CH2Cl2, no clear
compensation was observed, consistent with the mechanism that neither desolvation processes nor
conformational changes contribute significantly to the binding energetics.
In tr od u ction
of synthetic receptors are significantly influenced by
changing solvent polarity.3,4 Therefore to design a recep-
tor, we need to know the media or microenvironment
where the binding site exists. In some instances, we even
need to design the microenvironment around the binding
site.5,6 The molecular recognition in the protein inner
cavity or on the surface of the bilayer membranes
involves interfacial properties between organic solvent
and water, and the microscopic environment is not so
simple such as in pure organic solvents or in water.
Therefore more elaborate understanding of the driving
forces of binding in various microenvironments should
be helpful to design a receptor for a compound of
biological interest under relevant circumstances.
Elucidation of fundamental forces driving host-guest
complexation1 is important in a variety of areas such as
drug design, protein engineering, and gene technology.
For instance, rational design of drugs is reviewed,2 and
it is emphasized that the binding mode of the drug to
the receptor is quite different from expectation based on,
for example, molecular modeling. The difficulties in the
rational design are ascribed to (1) lack of understanding
of relative importance of solvation energy in the binding
energetics (hydrophobic interactions) and (2) lack of
understanding of conformational flexibility of receptor
proteins (induced-fit binding). For the role of solvation,
it is well-known that both binding affinity and selectivity
(3) Smithrud, D. B.; Diederich, F. J . Am. Chem. Soc. 1990, 112, 339.
(4) For examples of solvent effects on binding kinetics, see Haino,
T.; Rudkevich, D. M.; Rebek, J ., J r. J . Am. Chem. Soc. 1999, 121,
11253-11254.
* Tel: +81-75-753-5662. Fax: +81-75-753-4979.
(1) For recent reviews, see: (a) Schneider, H.-J . Angew. Chem., Int.
Ed. Engl. 1991, 30, 1417. (b) Diederich, F.; Smithrud, D. B.; Sanford,
E. M.; Wyman, T. B.; Ferguson, S. B.; Carcanague, D. R.; Chao, I.;
Houk, K. N. Acta Chem. Scand. 1992, 46, 205. (c) Lemieux, R. U. Acc.
Chem. Res. 1996, 29, 373.
(5) Tabushi, I.; Kuroda, Y.; Mizutani, T. J . Am. Chem. Soc. 1986,
108, 4514.
(6) Adrian, J . C., J r.; Wilcox, C. S. J . Am. Chem. Soc. 1992, 114,
1398.
(2) Davis, A. M.; Teague, S. J . Angew. Chem. Int. Ed. 1999, 38, 736.
10.1021/jo000557x CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/22/2000

6098 J . Org. Chem., Vol. 65, No. 19, 2000
Mizutani et al.
Two types of forces are involved in the binding pro-
cesses: direct interactions between a host molecule and
a guest molecule and interactions between host/guest
molecules and solvent molecules. Both of these interac-
tions can be the driving force of the binding. Host-guest
complexation can be classified into three categories
depending on polarity of the host/guest molecules and
polarity of the media: (1) When host and guest have polar
functional groups, the electrostatic interaction between
the polar functional groups can be the driving force of
binding. A typical example of this class of binding is the
binding of metal cations by crown ethers through host-
guest attractive electrostatic forces. (2) When host and
guest are nonpolar, both dispersion forces between host
and guest and the desolvation of polar and cohesive
solvent molecules can drive the binding. For instance,
cyclodextrins bind apolar organic guests mainly through
desolvation-driven hydrophobic forces in water.7 (3) When
host and guest molecules possess both polar functional
groups and apolar moieties, then both forces, host-guest
interactions and desolvation, can be the driving forces
of binding.
It is known that the binding of a polar guest is favored
in a less polar solvent,8 whereas the binding of a nonpolar
guest is most favorable in a polar and cohesive solvent
such as water.3 For representative examples of solvent
effects on the binding constant (Ka) of polar guests, the
values of log Ka of sodium and potassium by 15-crown-5
are 0.7 and 0.7, respectively, in water while 3.24 and 3.43,
respectively, in MeOH. Similarly, the values of log Ka of
sodium and potassium by 18-crown-6 are 0.8 and 2.0 in
water and 4.35 and 6.08 in MeOH, respectively.9 There-
fore changing the solvent from water to MeOH gives rise
to the 102.5-fold to 104.1-fold increase in binding constants
of metal cations by the crown ethers.10 Adrian and Wilcox
reported that the binding affinity of amine by a hydrogen-
bond-based receptor is reduced by the addition of water
to chloroform.11 As examples of solvent effects on the
binding of an apolar guest, Siegel and Breslow reported
that the binding constant of anisole by â-cyclodextrin in
DMSO is 2 orders of magnitude smaller than that in
water.12 Diederich and co-workers determined the bind-
ing affinity of a cyclophane receptor for pyrene in various
solvents, and it decreases monotonically in the following
order: water > ethanol > acetone > carbon disulfide, a
result suggesting that binding of apolar guests driven by
desolvation and host-guest dispersion forces occurs in
a wide range of solvents.3,13
interactions, because a number of molecules of biological
interest have both polar and apolar groups as important
interaction groups. In the previous paper,14 we reported
that water-soluble porphyrin receptors show selective
molecular recognition of amines and amino acids in
water, particularly for a charged guest. As a complemen-
tary study, in this paper, we performed a study of the
complexation of porphyrin receptors with several amines
and R-amino esters in CH2Cl2, and the results are
interpreted by comparison with the molecular recognition
behaviors of the water-soluble receptors having a struc-
turally similar binding pocket. The objectives of the
present study are twofold: first, to clarify details of host-
guest interactions, and second, to gain deeper insight into
the roles of solvent molecules in the energetics of molec-
ular recognition.
Resu lts a n d Discu ssion
Syn th esis of Zin c P or p h yr in Recep tor s a n d De-
ter m in a tion of Bin d in g Con sta n ts. Zinc porphyrins
bearing ester groups 1-4 and a simple zinc tetraphenyl-
porphyrin 5 were employed as receptors in organic
solvent, and zinc porphyrins bearing carboxylate groups
1a -3a and 6 were employed as receptors in water. The
aromatic macrocycle of porphyrin provides a rigid frame-
work for a preorganized binding pocket, as well as
versatile spectroscopic methods for detection of binding
events.15 One disadvantage is that porphyrins have a
pronounced tendency to undergo face-to-face dimeriza-
tion, especially in polar solvents. Therefore we designed
receptors having substituents on both faces of porphyrin
to avoid the porphyrin’s aggregation. Preparation of
porphyrins bearing bulky substituents above and below
the porphyrin plane is a synthetic challenge, and Lindsey
and co-workers performed elaborate investigations into
the porphyrin synthesis from benzaldehyde bearing bulky
substituents at the ortho positions. They reported that
reaction of benzaldehydes bearing 2,6-dibenzyloxy groups
with pyrrole proceeds in CHCl3 in relatively high yields.16,17
Receptors 1-4 were thus prepared14 following Lindsey’s
method.
These receptors are soluble in a variety of solvents
including chloroform, ethyl acetate, and THF. The ester
receptors 1-4 can be readily made soluble in water by
hydrolysis, by keeping the geometry of the binding site
almost intact. Therefore we can compare the binding
affinity toward guest in various solvents. The structures
1
of these receptors were characterized by H NMR and
We focus on the binding in which both polar and
nonpolar functional groups participate in the binding
high-resolution mass spectroscopy. For the free base of
1, a crystal suitable for X-ray crystallographic study
was obtained and the structure was also confirmed by
X-ray diffraction study. Crystal data and experimental
details for the free base of 1 are sumarized in Table 1.
The molecular structure of the free base of 1 is shown
in Figure 1. The porphyrin framework was planar. The
average distance of the 24 atoms of carbon and nitro-
gen of the porphyrin framework from the least-squares
plane obtained from the 24 atoms was 0.009 Å, with
the largest distance (0.05 Å) observed for one of the
(7) (a) Tabushi, I. Acc. Chem. Res. 1982, 15, 66. (b) Schneider, H.-J .;
Kramer, R.; Simova, S.; Schneider, U. J . Am. Chem. Soc. 1988, 110,
6442. (c) Connors, K. A. Chem. Rev. 1997, 97, 1325. (d) Liu, Y.; Han,
B.-H.; Li, B.; Zhang, Y.-M.; Zhao, P.; Chen, Y.-T.; Wada, T.; Inoue, Y.
J . Org. Chem. 1998, 63, 1444. (e) Kitae, T.; Nakayama, T.; Kano, K.
J . Chem. Soc., Perkin Trans. 2 1998, 207-212.
(8) Maitra, U.; Rao, P.; Vijay Kumar, P.; Balasubramanian, R.;
Mathew, L. Tetrahedron Lett. 1998, 39, 3255.
(9) Gokel, G. W.; Schall, O. F. Comprehensive Supramolecular
Chemistry; Gokel, G. W., Ed.; Pergamon Press Ltd.: Oxford, 1996; Vol.
1, p 126.
(10) For solvent effects on the binding by crown ethers, see Solov'ev,
V. P.; Strakhova, N. N.; Raevsky, O. A.; Ruediger, V.; Schneider, H.-J .
J . Org. Chem. 1996, 61, 5221-5226.
(11) Adrian, J . C., J r.; Wilcox, C. S. J . Am. Chem. Soc. 1991, 113,
678.
(12) Siegel, B.; Breslow, R. J . Am. Chem. Soc. 1975, 97, 6869.
(13) (a) Ferguson, S. B.; Sanford, E. M.; Seward, E. M.; Diederich,
F. J . Am. Chem. Soc. 1991, 113, 5410. (b) Smithrud, D. B.; Wyman, T.
B.; Diederich, F. J . Am. Chem. Soc. 1991, 113, 5420.
(14) Mizutani, T.; Wada, K.; Kitagawa, S. J . Am. Chem. Soc. 1999,
121, 11425.
(15) Ogoshi, H.; Mizutani, T. Acc. Chem. Res. 1998, 31, 81.
(16) (a) Wagner, R. W.; Ruffing, J .; Breakwell, B. V.; Lindsey, J . S.
Tetrahedron Lett. 1991, 32, 1703. (b) Wagner, R. W.; Lindsey, J . S.;
Turowska-Tyrk, I.; Scheidt, W. R. Tetrahedron 1994, 50, 11097.
(17) Lindsey, J . S.; Wagner, R. W. J . Org. Chem. 1989, 54, 828.

Molecular Recognition of Amines and Amino Esters
J . Org. Chem., Vol. 65, No. 19, 2000 6099
F igu r e 1. Structure of the free base of 1 determined by X-ray
crystallographic study. One of the methoxy groups and the
carobonyl oxygens in the ortho position was disordered and
appeared at 60% and 40% occupancy. The inner NH hydrogens
cannot be located from the difference Fourier map.
Ta ble 1. Cr ysta l Da ta a n d Exp er im en ta l Deta ils for th e
F r ee Ba se of 1‚2CH3OH
formula
formula weight
color
C78H76O34N4
1613.46
violet
crystal system
space group
a, Å
b, Å
c, Å
monoclinic
P21/c
13.501(3)
21.378(5)
13.766(3)
98.40(1)
3930(1)
2
â, deg
V, Å3
Z
scan mode
2θ max
ω scan
54.2°
d
calc, g cm-3
1.36
λ, Å (Mo KR)
0.71069
1.08
298
µ (Mo KR), cm-1
T, K
no. of reflections observed
8132
no. of reflections used (I0 > 3.00σ(I0))
no. of variables
4315
532
R
0.063
Rw
0.073
goodness of fit indicator
2.67
due to the amino group coordination to the zinc atom.18
The spectral shift in the Soret band of a solution of 2-4
in CH2Cl2 from 423 to 434 nm and that of 1 from 420
to 431 nm was similar to that of a solution of 1a -3a in
water (from 424 to 434 nm in 2a -3a and from 421 to
431 nm in 1a ), confirming that the coordination of
the amino group of guest to the zinc atom occurs both
in CH2Cl2 and in water. The binding constants were
determined by monitoring the absorbance changes of
the Soret band as a function of guest concentrations at
25 °C and fitting the saturation plot to the 1:1 binding
isotherm: Ka ) [porphyrin‚guest]/([porphyrin][guest]).
Standard deviations for the curve fitting are less than
5%. For all combinations of the receptors and the guests,
isosbestic points were always observed in the UV-visible
spectral titration, suggesting that the receptor and the
guest form a 1:1 complex exclusively.19 The binding
nitrogens. The phenyl groups are not perpendicular
to the porphyrin plane, and the dihedral angle between
the porphyrin plane and the phenyl plane was 73.3°-
73.4°. In the crystal, one of the methyl groups of the
methoxycarbonylmethoxy substituents was located on
the porphyrin plane of the neighboring molecule (Figure
2). The distances between the carbon of the methyl group
and the two nitrogens of the next porphyrin were
3.42 and 3.49 Å. Thus, the methyl group was bound in
the binding pocket of the next porphyrin, forming an
inclusion polymer in the crystal.
The binding constants of amines and amino esters in
CH2Cl2 were determined by UV-visible spectroscopic
titration. Addition of guest caused a red shift in the Soret
band of the porphyrin receptors, a typical spectral change
(18) Nappa, M.; Valentine, J . S. J . Am. Chem. Soc. 1978, 100, 5075.
(19) (a) Miller, J . R.; Dorough, G. D. J . Am. Chem. Soc. 1952, 74,
3977. (b) Kirksey, C. H.; Hambright, P.; Storm, C. B. Inorg. Chem.
1969, 8, 2141. (c) Kirksey, C. H.; Hambright, P. Inorg. Chem. 1970, 9,
958.

6100 J . Org. Chem., Vol. 65, No. 19, 2000
Mizutani et al.
F igu r e 3. Free energy changes, -∆G°, of binding by zinc
porphyrin receptors in three solvents, CH2Cl2, MeOH-borax
pH 9.0 (1:1), and borax pH 9.0 at 25 °C. In CH2Cl2, ester
receptor 3 and ZnTPP 5 are used, and in MeOH-borax and
in borax, carboxylate receptors 3a and 6 were used.
Ta ble 2. Bin d in g Con sta n ts (Ka ) of Am in es a n d Am in o
Ester s by Zin c P or p h yr in Recep tor s in CH2Cl2 a n d in
Bor a x Bu ffer a t 25 °C
In CH2Cl2
Ka/M-1
a
1
2
3
4
5
pKa
pyridine
n-BuNH2
t-BuNH2
PhCH2NH2
PhCH2CH2NH2
PhCH2CH2CH2NH2 21 300 224
Ala-OMe
2 920 401
21 200 255
9 110 249
21 100 179
52 700 477
4 480 3 650 7 720 5.34
5 960 12 700 40 000 11.48
4 740 6 530 14 000 11.44
3 330 6 690 12 000 9.0
6 900 17 500 46 000 9.96
4 940 11 800 45 000 10.8
918 2 170 4 660 7.74
F igu r e 2. Packing of the free base of 1. The methoxy group
in the ortho position is in van der Waals contact (the C-N
distances, 3.42 and 3.49 Å) with the next porphyrin’s nitrogens
and carbons to form an inclusion polymer in the crystal. A
methanol molecule is weakly hydrogen bonded to the carbonyl
group of the ester group in the para position, where the O-O
distance was 3.01 Å.
4 210 75
Leu-OMe
PhGly-OMeb
Phe-OMe
Trp-OMe
461
1 600
8 560 36
22 700 110
0.5
c
18
13
580 6 860 7.62
926 5 250 6.63
364 2 160 4 910 7.0
1 090 3 940 8 730 7.3
constants Ka for 1-5 are listed in Table 2. Receptors 1-5
bound amines and amino esters with binding constants
ranging from 300 to 53 000 M-1 in CH2Cl2 at 25 °C except
for some host-guest combinations involving 2 and Leu-
OMe and PhGly-OMe, which formed very loose com-
plexes. Binding constants and thermodynamic param-
eters for the binding of amines and amino esters
by 1a -3a in water and aqueous MeOH were reported in
the previous paper.14
Solven t Effects on Bin d in g Affin ity. In Figure 3
are compared the binding free energies (-∆G°) in CH2Cl2,
aqueous MeOH, and water. The binding affinity de-
creases from CH2Cl2 to MeOH-water and then starts to
increase to water. Competitive binding of water or MeOH
to the zinc atom could, at least in part, explain the
smaller binding constants in aqueous MeOH compared
to those in CH2Cl2.20 The binding affinity of nitrogen,
oxygen, phosphorus, and sulfur donors toward ZnTPP
was compared, and binding constants of an oxygen donor
are 1 order of magnitude smaller than those of a nitrogen
donor.21 Therefore water or MeOH can competitively bind
to the zinc atom when used as a solvent, particularly in
In Borax Buffer pH 9.0
Ka/M-1
2a
1a
3a
Ala-OMed
Phe-OMed
Trp-OMed
5
30
67
13
107
670
240
1 130
7 160
a
The dissociation constants of RNH3+ in water, pKa values, are
taken from ref 29. Phenylglycine methyl ester. c No interaction
b
d
was observed in the UV-visible spectrum. Binding data for
1a -3a in water, taken from ref 14.
1a , in which the zinc atom is relatively exposed to the
solvent. Sanders and co-workers, we and others also
reported that binding of various guests by zinc porphyrins
in chloroform is weakened by adding a polar cosolvent
such as water, alcohol, and ether.6,22,23 Poorer binding
ability in aqueous MeOH than in CH2Cl2 can be ascribed
to the deactivation of the acidity of the zinc atom in a
(22) (a) Bonar-Law, R. P.; Sanders, J . K. M. J . Am. Chem. Soc. 1995,
117, 259. (b) Mizutani, T.; Murakami, T.; Matsumi, N.; Kurahashi, T.;
Ogoshi, H. J . Chem. Soc., Chem. Commun. 1995, 1257. (c) Hayashi,
T.; Miyahara, T.; Koide, N.; Ogoshi, H. Chem. Commun. 1997, 1865.
(d) Mizutani, T.; Kurahashi, T.; Murakami, T.; Matsumi, N.; Ogoshi,
H. J . Am. Chem. Soc. 1997, 119, 8991.
(20) Preliminary UV-vis titration experiments showed that the
binding constants of EtOH, acetone, THF, and AcOEt by 5 in CH2Cl2
were 8.9, 0.9, 47, and 1.1 M-1, respectively.
(23) For some host-guest pairs with incomplete complementarity,
addition of polar cosolvent makes the binding tight. See ref 22.
(21) Vogel, G. C.; Searby, L. A. Inorg. Chem. 1973, 12, 936.

Molecular Recognition of Amines and Amino Esters
J . Org. Chem., Vol. 65, No. 19, 2000 6101
polar environment. But the increased affinity by chang-
ing the media from aqueous MeOH to water should be
ascribed to another mechanism involving solvophobic
interactions.
As shown in Table 2, the binding constants of Ala-OMe,
Phe-OMe, and Trp-OMe by 1 in CH2Cl2 are larger than
the corresponding binding constants of 2 and 3, while
the binding constants of Ala-OMe, Phe-OMe, and Trp-
OMe by 1a in water are smaller than those of 2a and
3a .14 Thus, these hydrophobic guests are bound to host
with longer methylene chains more tightly in water.
Therefore, long methylene chains in host 3a provide a
hydrophobic environment to assist the binding of non-
polar guest in water, while they showed weakly inhibitory
effects on binding in CH2Cl2.
The solvent effects shown in Figure 3 suggest that
there are two competing driving forces; polar interaction
between host and guest directly drives binding in apolar
media such as CH2Cl2, and the desolvation from the
host-guest interacting surfaces drives binding in a polar
solvent such as water.
Because polar interactions such as Lewis acid/Lewis
base interactions and hydrogen bonding between host
and guest will be diminished in a polar solvent11,24 and
contribution from desolvation of host-guest contact
surface will be dominant in a polar solvent, we can expect
that importance of these two interactions will switch at
some solvent with intermediate polarity. Hunter and co-
workers25 reported that, as the solvent polarity decreases
from water to chloroform, the strength of the intramo-
lecular interaction between aromatic rings goes through
a minimum at DMSO and then starts to increase again.
Other studies also reported a similar behavior regarding
the solvent effects on molecular recognition, although the
range of solvent polarity studied was more limited owing
to the solubility problems.26
F igu r e 4. Plot of -∆G° of binding by zinc porphyrin receptors
in CH2Cl2 at 25 °C against pKa values of the amino group of
guest. (a) 5-guest complexation (correlation coefficient, r )
0.82) and (b) 1-guest complexation (r ) 0.43).
contributions from factors other than basicity of the
amino group to the binding energetics of 1 and 2.30
Na tu r e of Host-Gu est In ter a ction s in CH2Cl2. The
solvent effects observed in our porphyrin-guest systems
indicate that the binding in CH2Cl2 is primarily driven
by polar interactions between host and guest. The first
candidate for the polar host-guest interactions is the
coordination interaction between the zinc atom and the
amino group, which is known to be electrostatic in
origin.21,27 If this is the case, the basicity of the amino
group mainly determines the magnitude of the binding
constants.28 To evaluate the validity of this mechanism,
the plot of -∆G° of binding by 5 against pKa of the amino
group of guest is shown in Figure 4a.29 The correlation
coefficient between the two quantities is 0.82, showing
that the basicity is certainly one of important factors for
the binding mechanism of 5. A similar plot is shown for
1 in Figure 4b, where the correlation was poor. The
correlation coefficients are 0.43, 0.58, 0.72, and 0.75 for
1, 2, 3, and 4, respectively (plots not shown for 2-4). The
poor correlations for 1 and 2 indicate that there are
Role of Ester Gr ou p s of P or p h yr in Recep tor s in
Bin d in g Selectivity. As a general trend, binding con-
stants decrease in the order 5 > 1 > 4 > 3 . 2. To
examine the origin of the weak binding ability of 2, the
1H NMR spectra of the zinc complexes 1-3 are compared
with those of the free base ligands in Figure 5. For 1 and
3, the resonance of the terminal methyl groups in the
2,6-positions of the phenyl groups appeared at the same
magnetic field as that of the free base ligand, while the
resonance of the terminal methyl groups of 2 is shifted
upfield. Therefore, one of the ester groups intramolecu-
(29) Liolta, C. L.; Perdue, E. M.; Hopkins, H. P., J r. J . Am. Chem.
Soc. 1974, 96, 7981 (py). Grunen, L.; Laskowski, M.; Scheraga, H. J .
Am. Chem. Soc. 1959, 81, 3891 (BuNH2). Hetzer, H. B.; Robinson,
R. A.; Bates, R. G. J . Phys. Chem. 1962, 66, 2696 (t-BuNH2). Hine,
J .; Yeh, C. Y.; Schmalstieg, F. G. J . Org. Chem. 1970, 35, 340
(PhCH2NH2). Girault-Vexlearschi, G. Bull. Chim. Soc. Fr. 1956, 589
(PhCH2CH2NH2). Carothers, W. H.; Bickford, C. F.; Harwitz, G. J . J .
Am. Chem. Soc. 1927, 49, 2908 (PhCH2CH2CH2NH2). Hay, R. W.;
Morris, P. J . J . Chem. Soc. B 1970, 1577 (Ala-OMe). Hay, R. W.; Porter,
L. J .; Morris, P. J . Aust. J . Chem. 1966, 19, 1197 (Leu-OMe). Almond,
H. R.; Kerr, R. J .; Niemann, C. J . Am. Chem. Soc. 1959, 81, 2856
(Phe-OMe). Hay, R. W.; Porter, L. J . J . Chem. Soc. B 1907, 1261. For
PhGly-OMe, pKa was calculated using ACD/pKa version 3.0, Advanced
Chemistry Development, Inc.
(24) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry;
VCH Verlagsgesellschaft: Weinheim, 1988.
(25) Breault, G. A.; Hunter, C. A.; Mayers, P. C. J . Am. Chem. Soc.
1998, 120, 3402.
(26) Recently binding constants of amines to zinc porphyrins in
water and chloroform were reported, where binding constants in water
were similar to those in chloroform: Imai, H.; Munakata, H.; Taka-
hashi, A.; Nakagawa, S.; Ihara, Y.; Uemori, Y. J . Chem. Soc., Perkin
Trans. 2 1999, 2565.
(27) Vogel, G. C.; Beckmann, B. A. Inorg. Chem. 1976, 15, 483.
(28) Summers, J . S.; Stolzenberg, A. M. J . Am. Chem. Soc. 1993,
115, 10559.
(30) In contrast, in water, the correlation coefficients between -∆G°
and pKa were -0.04 (guest ) Ala-OMe, Phe-OMe, and Trp-OMe) and
-0.49 (guest ) pyridine, butylamine, benzylamine, Ala-OMe, Leu-OMe,
PhGly-OMe, Phe-OMe, and Trp-OMe) for 1a and 3a , respectively. For
instance, amines are bound less tightly to 1a -3a than amino esters.
These negative correlation coefficients can be rationalized by the
existence of pre-equilibrium in which more basic amines are more
easily protonated and to be deactivated as a Lewis base for zinc.

6102 J . Org. Chem., Vol. 65, No. 19, 2000
Mizutani et al.
F igu r e 6. 1H NMR spectra of 1, the 1-phenylethylamine
complex, and phenylethylamine in CD2Cl2 at 25 °C. The
concentration of 1 was 1.0 mM. The spectrum of 1 only (a),
1 + phenylethylamine (6.07 mM) (b), and phenylethylamine
only (c).
arylporphyrins having bulky substituents at the 2 and 6
positions of the aryl groups do not show weak affinity
toward Leu-OMe or PhGly-OMe.43,44,46 The origin of the
weak affinity of Leu-OMe and PhGly-OMe is not clear,
but it is probably due to a mechanism in which the bulky
side chain group of these guests forced the guest molecule
to adopt an unfavorable conformation such as the unde-
sirable orientation of the ester groups.48 This effect is
quite sensitive to small changes in the guest structure.
For instance, PhGly-OMe is bound weakly, whereas Phe-
OMe is bound as strongly as Ala-OMe.
For another example of steric effects49,50 or cavity
effects51,52 on binding selectivity, the ratio of binding
constant of butylamine to that of tert-butylamine de-
creases in the order 5 > 1 > 4 > 3 > 2. The space of the
binding sites of receptors decreases in the order 5 > 4 >
1 > 2 > 3. Thus, 5 having the largest binding site showed
the best discrimination between butylamine and tert-
butylamine. Thus there seems no steric repulsion be-
tween the -(CH2)nCOOMe groups and the tert-butyl
group. This observation implies that the steric repulsion
between the zinc porphyrins and Leu-OMe occurs at a
place distant from the Zn-N coordination site.
F igu r e 5. 1H NMR spectra of zinc complexes of 1-3 and free
bases of 1-3 in CD2Cl2 at 23 °C.
larly coordinates to the zinc in 2 and is subject to the
ring current effects of the porphyrin π electrons. This was
also supported by molecular modeling studies, showing
that the geometry for the CdO‚‚‚Zn interaction is plau-
sible. Therefore the weak binding ability of 2 should be
ascribed to the quenching of the zinc acidity by the
intramolecular ester coordination.
For guest binding selectivity, weak binding of Leu-OMe
and PhGly-OMe by 1-3 is noteworthy. Water-soluble
receptors 1a -3a also showed reduced binding affinity
toward Val-OMe, Leu-OMe, and PhGly-OMe in water.14
Therefore this weak binding should be attributable to the
host-guest interaction and not to the desolvation pro-
cesses. The ability to discriminate Leu-OMe or PhGly-
OMe from Ala-OMe is high in 2 and 3 and low in 1.
Therefore the longer flexible methylene chains of the
receptors help discriminate these amino esters. To date,
a number of zinc porphyrins were reported to bind amino
esters,31-47 but much reduced affinity toward Leu-OMe
and PhGly-OMe is unprecedented. Several zinc tetra-
CH-π In ter a ction s. The ratios of Ka of aromatic
amino esters such as Phe-OMe and Trp-OMe to that of
Ala-OMe were large for 1. The large affinity of 1 toward
aromatic amino esters suggests that there are some
attractive interactions between the CH2COOMe group of
(31) Tabushi, I.; Kugimiya, S.; Kinnaird, M. G.; Sasaki, T. J . Am.
Chem. Soc. 1985, 107, 4192.
(32) Kugimiya, S. J . Chem. Soc., Chem. Commun. 1990, 432.
(33) Danks, I. P.; Sutherland, I. O.; Yap, C. H. J . Chem. Soc., Perkin
Trans. 1. 1990, 421.
(34) Hunter, C. A.; Meah, M. N.; Sanders, J . K. M. J . Am. Chem.
Soc. 1990, 112, 5773.
(35) Imai, H.; Nakagawa, S.; Kyuno, E. J . Am. Chem. Soc. 1992,
114, 6719.
(45) Kuroda, Y.; Kato, Y.; Higashioji, T.; Hasegawa, J .; Kawanami,
S.; Takahashi, M.; Shiraishi, N.; Tanabe, K.; Ogoshi, H. J . Am. Chem.
Soc. 1995, 117, 10950.
(36) Imai, H.; Uemori, Y. J . Chem. Soc., Perkin Trans. 2 1994, 1793.
(37) Liang, Y.; Chang, C. K. Tetrahedron Lett. 1995, 36, 3817.
(38) Liang, Y.; Chang, C. K.; Peng, S.-M. J . Mol. Recognit. 1996, 9,
149.
(39) Hayashi, T.; Nonoguchi, M.; Aya, T.; Ogoshi, H. Tetrahedron
Lett. 1997, 38, 1603.
(40) Mikros, E.; Gaudemer, A.; Pasternack, R. Inorg. Chim. Acta
1988, 153, 199.
(41) Verchere-Beaur, C.; Mikros, E.; Perree-Fauvet, M.; Gaudemer,
A. J . Inorg. Biochem. 1990, 40, 127.
(42) Aoyama, Y.; Yamagishi, A.; Asagawa, M.; Toi, H.; Ogoshi, H.
J . Am. Chem. Soc. 1988, 110, 4076.
(43) Mizutani, T.; Ema, T.; Tomita, T.; Kuroda, Y.; Ogoshi, H. J .
Am. Chem. Soc. 1994, 116, 4240.
(44) Mizutani, T.; Ema, T.; Yoshida, T.; Kuroda, Y.; Ogoshi, H. Inorg.
Chem. 1993, 32, 2072.
(46) Mizutani, T.; Murakami, T.; Kurahashi, T.; Ogoshi, H. J . Org.
Chem. 1996, 61, 539.
(47) Crossley, M. J .; Mackay, L. G.; Try, A. C. J . Chem. Soc., Chem.
Commun. 1995, 1925.
(48) For an example of complementary electrostatic interactions
in a host-guest system, see Muehldorf, A. V.; Engen, D. V.; Warner,
J . C.; Hamilton, A. D. J . Am. Chem. Soc. 1988, 110, 6561.
(49) Ogoshi, H.; Sugimoto, H.; Yoshida, Z. Tetrahedron Lett. 1976,
4481.
(50) Imai, H.; Nakagawa, S.; Kyuno, E. J . Am. Chem. Soc. 1992,
114, 6719.
(51) Bonar-Law, R. P.; Mackay, L. G.; Sanders, J . K. M. J . Chem.
Soc., Chem. Commun. 1993, 456.
(52) Elemans, J . A. A. W.; Claase, M.; Aarts, P. P. M.; Rowan, A.
E.; Schenning, A. P. H. J .; Nolte, R. J . M. J . Org. Chem. 1999, 64,
7009.

Molecular Recognition of Amines and Amino Esters
J . Org. Chem., Vol. 65, No. 19, 2000 6103
F igu r e 7. A plausible structure of the 1-phenylethylamine
complex consistent with 1H NMR studies. The methylene
group of 1 is close to the phenyl group of the guest. The
geometry was optimized by Spartan 3.1, using a PM3 Hamil-
tonian. A model compound for 1 was used, where the meth-
oxycarbonyl groups of the other face of the porphyrin were
replaced with hydrogens.
F igu r e 8. Plots of ∆S° against ∆H° of binding of amines and
R-amino esters by 1a -3a and 6 in borax buffer at pH 9.0, I )
0.1 M (O) and of binding by porphyrin receptors 2-3 and 5 in
CH2Cl2 (b). In the borax buffer, (a) 3a -Arg-OMe, (b) 2a -Arg-
OMe, (c) 2a -pyridine, (d) 2a -Trp-OMe, (e) 6-Trp-OMe, (f)
6- pyridine, (g) 1a -Arg-OMe, (h) 3a -pyridine, (i) 3a -Trp-
OMe, (j) 6-Arg-OMe. In CH2Cl2, (k) 2-Ala-OMe; (l) 3-Ala-
OMe; (m) 5-PhCH2CH2NH2; (n) 3-Trp-OMe; (o) 2-PhCH2-
CH2NH2; (p) 2-pyridine; (q) 2-nBuNH2; (r) 5-Ala-OMe; (s)
5-pyridine; (t) 5-Trp-OMe; (u) 3-PhCH2CH2NH2; (v) 3-
pyridine; (w) 5-nBuNH2; (x) 3-nBuNH2; (y) 2-Trp-OMe. For
a-j, see ref 14. A line was drawn by least-squares line fitting
to the data in water (a-j).
1
1 and the aromatic side chain. The H NMR spectrum of
1 is compared with that of the complex between 1 and
phenylethylamine (Figure 6). The methylene protons
(OCH2COOCH3) of the complexed 1 was 0.16 ppm upfield
from those of the uncomplexed 1, showing that the
methylene protons point into the shielding region of the
phenyl group of phenylethylamine in the complex. It
should be noted that there are eight methylene groups
and the observed shifts are the average of these eight
methylene resonances. Therefore, the complexation-
induced shift of 0.16 ppm amounts to a 1.3 ppm upfield
shift for one methylene group, which is a significant shift.
The CH-π interaction53-56 between the phenyl group and
the acidic methylene protons may contribute to the
Ta ble 3. En th a lp y Ch a n ges (∆H°, k J m ol-1) a n d En tr op y
Ch a n ges (∆S°, J K-1 m ol-1) in Bin d in g of Am in es a n d
Am in o Ester s by Zin c P or p h yr in Recep tor s in CH2Cl2
2
3
5
∆H°
∆S°
∆H°
∆S°
∆H°
∆S°
pyridine
n-BuNH2
PhCH2CH2NH2 -39.3
Ala-OMe
Trp-OMe
-41.1
-40.6
-87.7 -47.5 -89.4 -45.4 -77.6
-90.1 -49.5 -93.8 -50.8 -82.2
-80.5 -46.7 -83.0 -45.4 -63.1
-66.7 -36.0 -63.7 -43.7 -76.3
-30.7
(53) For CH-π interaction in host-guest chemistry, see (a) Koba-
yashi, K.; Asakawa, Y.; Kikuchi, Y.; Toi, H.; Aoyama, Y. J . Am. Chem.
Soc. 1993, 115, 2648. (b) Ehama, R.; Tsushima, M. Y.; Tomoaki;
Suezawa, H.; Sakakibara, K.; Hirota, M. Bull. Chem. Soc. J pn. 1993,
66, 814. (c) Poh, B.-L.; Tan, C. M. Tetrahedron 1994, 50, 3453. (d) Poh,
B.-L.; Tan, C. M. J . Inclusion Phenom. Mol. Recognit. Chem. 1994,
18, 93. (e) Nishio, M.; Umezawa, Y.; Hirota, M.; Takeuchi, Y.
Tetrahedron 1995, 51, 8665. (f) Sugimoto, H.; Aida, T.; Inoue, S. J .
Chem. Soc., Chem. Commun. 1995, 1411. (g) Kikuchi, Y.; Aoyama, Y.
Bull. Chem. Soc. J pn. 1996, 69, 217. (h) Kodama, M. Bull. Chem. Soc.
J pn. 1996, 69, 3179. (i) Sugiura, K.-i.; Ushiroda, K.; Tanaka, T.;
Sawada, M.; Sakata, Y. Chem. Lett. 1997, 927. (j) Gillard, R. E.; Raymo,
F. M.; Stoddart, J . F. Chem. Eur. J . 1997, 3, 1933. (k) Kim, E.-i.;
Paliwal, S.; Wilcox, C. S. J . Am. Chem. Soc. 1998, 120, 11192. (l)
Mallinson, P. R.; Wozniak, K.; Wilson, C. C.; McCormack, K. L.; Yufit,
D. S. J . Am. Chem. Soc. 1999, 121, 4640. (m) Yu, L.; Schneider, H.-J .
Eur. J . Org. Chem. 1999, 1619.
(54) For CH- π interaction in coordination compounds, see (a)
Muller, T. E.; Mingos, D. M. P.; Williams, D. J . J . Chem. Soc., Chem.
Commun. 1994, 1787. (b) Chowdhury, S. K.; J oshi, V. S.; Samuel, A.
G.; Puranik, V. G.; Tavale, S. S.; Sarkar, A. Organometallics 1994,
13, 4092. (c) Yamanari, K.; Nozaki, T.; Fuyuhiro, A.; Kaizaki, S. Chem.
Lett. 1996, 35. (d) Miyamura, K.; Mihara, A.; Fujii, T.; Gohshi, Y.; Ishii,
Y. J . Am. Chem. Soc. 1995, 117, 2377. (e) Yoshida, N.; Oshio, H.; Ito,
T. Chem. Commun. 1998, 63. (f) Hancock, K. S. B. Chem. Commun.
1998, 1409. (g) Mizutani, M.; Tomosue, S.; Kinoshita, H.; J itsukawa,
K.; Masuda, H.; Einaga, H. Bull. Chem. Soc. J pn. 1999, 72, 981.
(55) For CH-π inteaction in biological molecules, see (a) Chakra-
barti, P.; Samanta, U. J . Mol. Biol. 1995, 251, 9. (b) Shimohigashi, Y.;
Maeda, I.; Nose, T.; Ikesue, K.; Sakamoto, H.; Ogawa, T.; Ide, Y.;
Kawahara, M.; Nezu, T.; Terada, Y.; Kawano, K.; Ohno, M. J . Chem.
Soc., Perkin Trans. 1 1996, 2479. (c) Kim, D. H.; Li, Z.-H.; Lee, S. S.;
Park, J .-I.; Chung, S. J . Bioorg. Med. Chem. 1998, 6, 239. (d) Umezawa,
Y.; Nishio, M. Bioorg. Med. Chem. 1998, 6, 2507.
-46.6 -116.5 -40.0 -76.5 -45.9 -78.5
stabilization of the complex.57 By molecular modeling
studies, we obtained a structure shown in Figure 7, where
the methylene protons are on the phenyl group of the
guest. For aromatic amines, very tight binding of PhCH2-
CH2NH2 by 1 is noteworthy, which may also be ascribed
to the attractive interaction between the phenyl group
and the ester group.
High affinity for aromatic guests and poor affinity for
bulky amino esters such as Leu-OMe of 1-3 demon-
strated that even flexible ester groups near the binding
site can alter the binding selectivity significantly.
En th a lp y a n d En tr op y Ch a n ges of Bin d in g. En-
thalpy and entropy changes of the binding were deter-
mined by van’t Hoff analysis (ln Ka ) -(1/RT)∆H° +
(1/R)∆S°; R is the gas constant) of binding constants in
the temperature range of 5-25 °C. The values of the
enthalpy and entropy changes are listed in Table 3. In
Figure 8 are shown the plots of ∆S° against ∆H° of
binding in CH2Cl2 and of binding in water.14
As seen in Figure 8, the values of ∆H° and ∆S° are
readily grouped into two regions depending on the media
used. Binding in CH2Cl2 is always associated with large
(56) For quantum mechanical calculations on CH-π interactions,
see Tsuzuki, S.; Honda, K.; Uchimaru, T.; Mikami, M.; Tanabe, K. J .
Phys. Chem. A 1999, 103, 8265.
(57) For interactions between ester and aromatic rings, see (a)
Kemula, A.; Iwamoto, R. T. J . Phys. Chem. 1968, 72, 2764. (b) Kikuchi,
Y.; Aoyama, Y. Supramol. Chem. 1996, 7, 147.

6104 J . Org. Chem., Vol. 65, No. 19, 2000
Mizutani et al.
negative enthalpy changes and large negative entropy
changes. The negative enthalpy changes can be ascribed
to the coordination interaction between the zinc and
amine, and the negative entropy changes can be ascribed
to the restriction of translational, rotational, and internal
rotational motions upon binding. In organic solvent
such as CH2Cl2, solvation energy is not very large and
the enthalpy changes directly reflect the attractive
interactions between host and guest. The values of ∆H°
of -30 to -50 kJ mol-1 are comparable to the reported
∆H° for zinc porphyrin-guest complexation in organic
solvents.45,46,50,58-60 In contrast, enthalpy changes and
entropy changes are close to zero in water, reflecting the
compensation effects of desolvation. One origin of the
difference in ∆H° and ∆S° between water and CH2Cl2
shown in Figure 8 is desolvation from polar functional
groups such as the zinc and the amino group. In water,
dehydration of polar substituents largely masks the
negative enthalpy changes of host-guest interactions.
The values of ∆S° of -50 to +60 J K-1 mol-1 observed
for 1a -3a and 6 in water are in the range of ∆S°
observed for the complexation by cyclodextrins61,62 and
other water-soluble receptors.26,63
Th er m od yn a m ics of Sa lt Br id ge a n d Hyd r ogen
Bon d in g In ter a ction s. Positive entropy changes were
observed for the binding of Arg-OMe in water, which is
ascribed to dehydration from the ionic groups upon salt
bridge formation.64,65 Ionic groups are most strongly
hydrated, so that salt bridge formation leads to a large
entropy gain owing to the gain in translational motional
freedom of the released water molecules. For the binding
of Arg-OMe by 1a , the salt bridge should be exposed to
water. Thus, the number of dehydrated water molecules
may be smaller and the entropic gain due to dehydration
was smaller than that for 2a (3a )-Arg-OMe complexation
as shown in a and b vs g in Figure 8.
F igu r e 9. Schematic representation of the relationship
between binding affinity and solvent polarity for host-guest
complexation driven by nonpolar and solvophobic forces (a),
electrostatic forces (c), and both of these forces (b).
enthalpic gain in the apolar solvent as seen in Figure 8
is in sharp contrast to the solvent effects on the binding
of pyrene by the cyclophane, where enthalpic gain is
larger in polar solvents than in apolar solvents.13 For the
binding of pyrene by the cyclophane, neither host nor
guest has any polar functional group. As a result, only
nonpolar interactions such as dispersion forces are
operating between host and guest. The larger enthalpic
gain in polar solvents implies that the net stabilization
due to host-guest dispersion forces will be larger with
decreasing the polarizability of solvent (that is, with
increasing solvent polarity). Similarly, for binding of
benzene derivatives by cyclodextrins in water, an enthal-
pic gain is observed and is ascribed to London’s dispersion
force.67
In Figure 9 are schematically illustrated the effects of
solvent polarity on the binding free energy. For binding
of apolar guests to an apolar binding site of receptors
such as cyclodextrins and cyclophanes (Figure 9, a),
binding is favored in polar solvents. For binding of ionic
guests to a polar binding site of receptors such as crown
ethers (Figure 9, c), the binding is favorable in apolar
solvents. For the binding of a guest by a receptor, both
of which have polar and apolar functional groups at
the contact surface, such as the combination of zinc
porphyrins and amino acid derivatives (Figure 9, b), the
binding affinity is high in both polar and apolar media
and low in a solvent with intermediate polarity.
For the binding of amine by a hydrogen-bond-based
host, Adrian and Wilcox reported that the addition of
water to chloroform has no especially large effect on the
free energy changes in the binding, but it has a signifi-
cant effect on the enthalpy and entropy changes; both
values were closer to zero when water is added.11 Release
of water molecules, which have interacted with hydrogen
bonding functional groups, upon binding thus signifi-
cantly contributes to ∆H° and ∆S°.
Com p a r ison of Bin d in g b y Zin c P or p h yr in s
w ith Bin d in g by Cyclop h a n es a n d Cr ow n Eth er s.
Thermodynamics of binding in an aqueous solution is
complicated by desolvation processes and is not well
understood.2,3,66 Solvent effect on the enthalpy changes
and the entropy changes of binding is a key to under-
standing molecular recognition in water. The larger
As an empirical rule, the magnitude of free energy
changes originated from desolvation is known to be
proportional to the contact surface area. Contribution of
desolvation in water to the binding free energy was
estimated to be 0.025 kcal/mol/Å2 from the free energy
changes of transfer of amino acids from water to ethanol
by Chothia.68 Both polar amino acids and nonpolar amino
acids show similar dependence of desolvation energy on
the accessible surface area. Importance of the solvent
cohesive interactions was also demonstrated by the
relationship between the binding energy and surface
tension of the solvents by Connors and Sun.69 All of these
(58) Miller, J . R.; Dorough, G. D. J . Am. Chem. Soc. 1952, 74, 3977.
(59) Vogel, G. C.; Stahlbush, J . R. Inorg. Chem. 1977, 16, 950.
(60) Hunter, C. A.; Meah, M. N.; Sanders, J . K. M. J . Am. Chem.
Soc. 1990, 112, 5773.
(61) (a) Kato, Y.; Conn, M. M.; Rebek, J ., J r. Proc. Natl. Acad. Sci.
U.S.A. 1995, 92, 1208. (b) Stauffer, D. A.; Barrans, R. E., J r.;
Dougherty, D. A. J . Org. Chem. 1990, 55, 2762.
(62) Inoue, Y.; Hakushi, T.; Liu, Y.; Tong, L.; Shen, B.; J in, D. J .
Am. Chem. Soc. 1993, 115, 475.
(63) Inoue, Y.; Liu, Y.; Tong, L.-H.; Shen, B.-J .; J in, D.-S. J . Am.
Chem. Soc. 1993, 115, 10637.
(64) Mizutani, T.; Horiguchi, T.; Koyama, H.; Uratani, I.; Ogoshi,
H. Bull. Chem. Soc. J pn. 1998, 71, 413.
(67) Tabushi, I.; Kiyosuke, Y.; Sugimoto, T.; Yamamura, K. J . Am.
Chem. Soc. 1978, 100, 916. Tabushi, I.; Mizutani, T. Tetrahedron 1987,
43, 1439.
(68) Chothia, C. Nature 1974, 248, 338. See also Nozaki, Y.; Tanford,
C. J . Biol. Chem. 1971, 246, 2211. Schultz, G. E.; Schirmer, R. H.
Principles of Protein Structure; Springer-Verlag: New York, 1979.
Matouschek, A.; Serrano, L.; Fersht, A. R. Analysis of Protein
Folding by Protein Engineering; Oxford University Press: Oxford,
1994; pp 137. DeLano, W. L.; Ultsch, M. H.; de Vos, A. M.; Wells, J . A.
Science 2000, 287, 1279.
(65) Kano, K.; Kamo, H.; Negi, S.; Kitae, T.; Takaoka, R.; Yama-
guchi, M.; Okubo, H.; Hirama, M. J . Chem. Soc., Perkin Trans. 2 1999,
15.
(66) Lipscomb, W. N. Acc. Chem. Res. 1982, 15, 232.

Molecular Recognition of Amines and Amino Esters
J . Org. Chem., Vol. 65, No. 19, 2000 6105
studies revealed that desolvation-driven binding should
be important if host-guest contact surface area is large
and the solvent is polar. In contrast, for the binding of
cations by crown ethers, the contact surface area is
obviously smaller, and the desolvation energy may not
be very significant as to overcome the host-guest inter-
action energies. Thus, it is reasonable that the binding
mechanism of crown ethers is dominated by host-guest
electrostatic interactions even in polar solvents, whereas
that by cyclodextrins and porphyrin receptors is governed
by both solvation interactions and host-guest polar and
nonpolar interactions.
If host and guest have polar functional groups, the
complementary arrangement of polar functional groups
of host and guest is important to avoid electrostatic
repulsion at the contact surface. We found that there is
no spectral change in the solutions of the free bases of
1a -3a upon addition of pyridine or histamine in water
(borax buffer, pH 9.0), although the zinc complexes (1a -
3a ) showed tight binding. Therefore complementary
electrostatic potential near the zinc atom of host and the
amino group of guest is necessary for tight binding to
occur.
As shown in Figure 8, plots of ∆S° against ∆H° of binding
in water (O) showed an enthalpy-entropy compensation,
i.e., the negative enthalpy change causes the negative
entropy change. However, plots of ∆S° against ∆H° of
binding in CH2Cl2 (b) showed no clear compensation
effects. We suggest that the enthalpy-entropy com-
pensation is originated from two different sources:
(1) restriction of motion of host and guest molecules
due to mutual attractive forces and (2) enthalpic loss
due to desolvation and entropic gain due to motional
freedoms of desolvated molecules. In water, both of these
processes contribute to the observed enthalpy-entropy
compensation, so that we observed clear compensation.
In CH2Cl2, the former process will contribute to the
compensation effect but contribution from the latter
process might be minor and the observed compensation
was not clear.
In the compensation process originated from host-
guest motional restriction, the flexibility of host and guest
molecules is important. Searle and Williams reported
that entropic cost of a restriction of a single bond rotation
owing to some attractive forces between host and guest
is around 11 J K-1 mol-1 79
. We also estimated the internal
rotational entropy to be 20 J K-1 mol-1 from the binding
experiment using zinc porphyrin receptors.46 Restriction
of the internal single bond rotation by host-guest
attractive forces (∆H° < 0) will cause this entropy loss
(∆S° < 0), leading to the enthalpy-entropy compensa-
tion. If both of host and guest are rigid, we should observe
a minimal compensation. The compensation effect will
be maximum if flexible host and flexible guest form an
induced-fit type complex.80
As to the second source of compensation, the enthalpy-
entropy compensation originated from desolvation should
be distinct if the number of desolvated solvent molecules
increases as the host-guest attractive interactions are
larger, which in turn is proportional to the contact surface
area of host and guest. It should be noted that this
compensation will be best observed in a strongly solvating
polar solvent such as water. Both the conformational
restriction owing to the host-guest multipoint interac-
tions as well as the desolvation from the contact surface
contribute to the slope of the correlation line in the plot
of ∆S° against ∆H°.
Inoue et al. proposed that the two quantities, the slope
and the intercept of the plot of T∆S° against ∆H°, can
be used to estimate the conformational changes and the
degree of desolvation during host-guest complexation,
respectively.62 From the data shown in Figure 8, the slope
R is 0.81 and the intercept T∆S° is 3.1 kcal/mol in the
complexation between zinc porphyrin receptors and
amines or amino esters in water (correlation coefficient
r ) 0.914). The compensation relationship observed
for R-, â-, and γ-cyclodextrins gives the slope R of 0.9
and the intercept T∆S° of 3.1 kcal/mol,62 being very
close to the values observed in our water-soluble
zinc porphyrin receptor system. These values suggest
that both induced-fit type conformational changes and
desolvation contribute considerably to the overall
If the polar groups of host and guest do not participate
in the host-guest interactions, it is important that these
polar groups retain the favorable solvation in the com-
plex, as demonstrated by Diederich and co-workers.13a,70
We can summarize the binding mechanism as follows.
If guest is apolar and the media is water, London’s
dispersion forces between host and guest is one of the
driving forces owing to the low polarizability of water.
This process leads to the enthalpic gain. Desolvation from
the contact surfaces also contributes to stabilization of
the complex.71 This leads to the entropic gain. The net
result is that a negative enthalpy change and a negative
entropy change are observed. The negative entropy
change does not necessarily imply that desolvation-driven
entropic gain plays a minor role, because the intrinsic
entropy loss upon bimolecular association is large. The
negative enthalpy changes and the negative entropy
changes observed for the binding by cyclodextrins and
zinc porphyrin receptors in water suggest a significant
contribution of London’s dispersion forces to the overall
enthalpy changes, except for the binding of ionic guests.
If host and guest have ionic functional groups and
interact via a salt bridge, there is a considerable contri-
bution from desolvation from the ionic groups to the
entropic gain.
The classical picture of hydrophobic interactions stresses
the importance of entropic gain from the desolvated
water. However, the host-guest studies and protein
denaturation studies72 revealed that the enthalpic gain
due to van der Waals interactions are significant in
describing the hydrophobic effects.24,73-78
In d u ced -F it a n d Desolva tion -Dr iven Bin d in g As
Evid en ced by En th a lp y-En tr op y Com p en sa tion .
(69) Connors, K. A.; Sun, S. J . Am. Chem. Soc. 1971, 93, 7239.
(70) Carcanague, D. R.; Diederich, F. Angew. Chem., Int. Ed. Engl.
1990, 29, 769.
(71) For discussion of van der Waals contribution and solvophobic
contribution to the binding, see Schneider, H.-J .; Wang, M. J . Org.
Chem. 1994, 59, 7464-7472.
(72) Privalov, P. L.; Gill, S. J . Pure Appl. Chem. 1989, 61, 1097.
(73) Cramer, R. D. J . Am. Chem. Soc. 1977, 99, 5408.
(74) Evans, D. F.; Chen, S.-H.; Schriver, G. W.; Arnett, E. M. J . Am.
Chem. Soc. 1981, 103, 481.
(76) Hildebrand, J . H. Proc. Natl. Acad. Sci. U.S.A. 1979, 76, 194.
(77) Abraham, M. H. J . Am. Chem. Soc. 1980, 102, 5910.
(78) Hvidt, A. Acta Chem. Scand., Part A 1983, 37, 99.
(79) Searle, M. S.; Williams, D. H. J . Am. Chem. Soc. 1992, 114,
10690.
(75) Mirejovsky, D.; Arnett, E. M. J . Am. Chem. Soc. 1983, 105,
1112.
(80) Thoma, J . A.; D. E. Koshland, J . J . Am. Chem. Soc. 1960, 82,
3329. Koshland, D. E., J r. Angew. Chem., Int. Ed. Engl. 1994, 33, 2375.

6106 J . Org. Chem., Vol. 65, No. 19, 2000
Mizutani et al.
and dried over MgSO4. Evaporation of the solvent and recrys-
tallization from EtOH afforded a white solid of 7 (1.4 g, 87%):
1H NMR (CDCl3) δ 2.11 (s, 3H), 3.85 (s, 6H), 3.90 (s, 3H), 7.21
(s, 2H); HRMS (FAB) calcd for C11H14O4 (M+) 210.0892, found
210.0884. Anal. Calcd for C11H14O4: C, 62.85; H, 6.71. Found:
C, 62.68; H, 6.66.
energetics of binding by zinc porphyrins in water. As
supporting evidence for the induced-fit binding, in the
1H NMR spectra there are characteristic shifts in the
alkyl protons of 3a upon complexation, showing that
the alkyl groups at the 2,6-positions modify their con-
formation to accommodate a guest.14
Meth yl 4-Br om om eth yl-3,5-d im eth oxyben zoa te (8). A
solution of 7 (1.0 g, 4.8 mmol) in CCl4 (28 mL) was irradiated
with a 500-W lamp while Br2 (525 mg, 3.28 mmol) in CCl4
(10 mL) was added dropwise over 30 min. The progress of
the reaction was monitored by TLC (SiO2, AcOEt/hexane )
1/4). After the reaction was completed, AcOEt (50 mL) was
added. The organic layer was washed with saturated aqueous
NaHCO3 and saturated aqueous NaCl and dried over MgSO4.
Evaporation of the solvent and recrystallization from ether
afforded a white solid of 8 (0.97 g, 70%): 1H NMR (CDCl3) δ
3.91 (s, 3H), 3.92 (s, 6H), 4.63 (s, 2H), 7.20 (s, 2H); HRMS
(FAB) calcd for C11H13BrO4 (MH+) 287.9997, found 287.9989.
Meth yl 4-F or m yl-3,5-d im eth oxyben zoa te (9). A solution
of powdered 8 (730 mg, 2.52 mmol) in DMSO (16 mL) and
NaHCO3 (1.85 g) was heated at 70 °C under N2 with vigorous
stirring for 25 min. The progress of the reaction was monitored
by TLC (SiO2, AcOEt/hexane ) 1/1). The reaction mixture was
then immediately cooled in an ice bath, poured into saturated
NaCl (40 mL), and extracted with AcOEt. The organic layers
were combined and dried over MgSO4. Evaporation of the
solvent and recrystallization from benzene/hexane afforded a
white solid of 9 (425 mg, 75%): 1H NMR (CDCl3) δ 3.94 (s,
9H), 7.23 (s, 2H), 10.50 (s, 1H); HRMS (FAB) calcd for C11H12O5
(MH+) 224.0684, found 224.0677. Anal. Calcd for C11H12O5: C,
58.93; H, 5.39. Found: C, 58.64; H, 5.36.
The host-guest polar interactions, the contact surface
area, flexibility of host and guest molecules, and polarity
of solvent are important parameters determining the
enthalpy and entropy changes in binding. If the contact
surface area is small, then the host-guest polar interac-
tions should be the main driving force. If the contact
surface area is large and the solvent is polar and
cohesive, all of desolvation, host-guest polar interactions,
and host-guest nonpolar (dispersion) interactions play
a significant role in the thermodynamics of binding.
Recognition of molecules of biological interest such as
protein or DNA is associated with a large contact area,
and regulation of the desolvation should be important
for the rational design of the system.
Con clu sion s
The present study clarified the relative importance of
intermolecular interactions in molecular recognition of
amines and R-amino esters by zinc porphyrin receptors.
The driving force of binding was discussed in terms of
two classes of interactions, the host-guest interactions
and solvation interactions. The binding affinity was high
both in CH2Cl2 and in water and low in water-MeOH.
Thus we suggest that two driving forces competitively
contribute to the overall binding free energy. The host-
guest electrostatic interactions are important for the
binding selectivity in organic solvents, while desolvation-
driven binding is dominant in water. For the binding in
CH2Cl2, the ester groups of the zinc porphyrin receptors
contributed to the tight binding of aromatic guests, and
the long alkyl chains of the zinc porphyrins contributed
to the loose binding of Leu-OMe and other bulky amino
acid esters. Relatively flexible carbomethoxyalkyl groups
can dictate the binding selectivity of the receptors. The
difference in solvent polarity manifests itself by compen-
sation effects of ∆H° and ∆S°. Clear compensation was
observed in water, reflecting the importance of induced-
fit conformational changes and desolvation processes for
binding in water.
5,10,15,20-Tetr a k is(4-m eth oxyca r bon yl-2,6-d im eth oxy-
p h en yl)p or p h yr in (10). Aldehyde 9 (224 mg, 1 mmol) and
pyrrole (69 µL, 1 mmol) were dissolved in CH2Cl2 (100 mL)
under N2 and then BF3‚OEt2 (42 µL, 0.33 mmol) was added.
After the reaction mixture was stirred at room temperature
for 80 min, 2,3-dichloro-5,6-dicyanobenzoquinone (170 mg, 0.75
mmol) was added, and the mixture was refluxed for 2 h. The
solution was then neutralized by addition of triethylamine (46
µL) and evaporated. The mixture was separated by column
chromatography (SiO2, CHCl3/AcOEt ) 100/1 to 10/1), and the
crude product was washed with methanol thoroughly to afford
a purple solid of 10 (87 mg, yield 32%): 1H NMR (CDCl3) δ
-2.57 (s, 2H), 3.55 (s, 24 H), 4.08 (s, 12H), 7.66 (s, 8H), 8.60
(s, 8H); UV-vis (CH2Cl2) λmax (log ꢀ) 419 (5.56), 513 (4.27), 545
(3.68), 592 (3.72); HRMS (FAB) calcd for C60H54N4O16 (M+)
1086.3534, found 1086.3536. Anal. Calcd for C60H54N4O16: C,
66.29; H, 5.01; N, 5.15. Found: C, 66.02; H, 5.29; N, 4.90.
[5,10,15,20-Tetr akis(4-m eth oxycar bon yl-2,6-dim eth oxy-
p h en yl)p or p h yr in a to]zin c(II) (4). A solution of 10 (52 mg,
48 µmol) and Zn(OAc)2-saturated methanol (9 mL) in CHCl3
(50 mL) was refluxed for 1.5 h. After cooling, the solution was
washed with saturated aqueous NaHCO3 and saturated aque-
ous NaCl and dried over Na2SO4. Evaporation of the solvent
and washing with methanol thoroughly afforded a pink solid
of 4 (52 mg, 95%): 1H NMR (CDCl3) δ 3.55 (s, 24H), 4.08 (s,
12H), 7.67 (s, 8H), 8.68 (s, 8H); UV-vis (CH2Cl2) λmax (log ꢀ)
421 (5.61), 548 (4.29), 586 (3.36); HRMS (FAB) calcd for
Exp er im en ta l Section
1
Gen er a l Meth od s. H NMR spectra were recorded at 500
MHz, and chemical shifts are reported relative to Me4Si or
residual protons of deuterated solvents. High-resolution mass
spectra were obtained using 3-nitrobenzyl alcohol as a matrix.
The binding constants were determined as previously de-
scribed.14 X-ray diffraction data were collected at 295 K with
a Rigaku CCD diffractometer (Mercury). Crystals for the X-ray
study were obtained by recrystallization of the free base of 1
from methanol.
C
60H52N4O16Zn (M+) 1148.2668, found 1148.2714.
Ack n ow led gm en t. This work was supported by a
Grant-in Aid for Scientific Research from the Ministry
of Education, Science, Sports, and Culture, J apan. We
thank T. Kobatake for help in mass spectroscopic
studies. We also thank H.-C. Chang and M. Kondo for
kind help in X-ray diffraction studies.
Ma ter ia ls. Porphyrins 1-3 and 1a -3a were prepared
according to the published procedures.14
Meth yl 3,5-Dim eth oxy-4-m eth ylben zoa te (7). A mixture
of 3,5-dihydroxy-4-methylbenzoic acid81 (1.3 g, 7.7 mmol),
K2CO3 (2.5 g), and dimethyl sulfate (2.5 mL) in acetone (15
mL) was refluxed for 4 h. After AcOEt (50 mL) was added,
the organic layer was washed with saturated aqueous NaCl
Su p p or tin g In for m a tion Ava ila ble: Tables of structure
factors and positional parameters for the free base of 1. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(81) Borchardt, R. T.; Sinhababu, A. K. J . Org. Chem. 1981, 46, 5021.
J O000557X