Porphyrin receptors for amines, amino acids, and oligopeptides in water

Article abstract of DOI:10.1021/ja9922126

A series of water-soluble porphyrin receptors having a hydrophobic binding pocket, a Lewis acidic site (Zn), and an electrostatic recognition site (COO^- groups) were prepared. All the porphyrin receptors have a [meso- tetrakis(p-carboxyphenyl)porphyrinato]zinc (Zn·TCPP) as a common structural unit, and eight ω-carboxyalkyloxy groups (alkyloxy = methoxy (1), butoxy (2), decyloxy (3)) at the ortho positions of the phenyl groups. These receptors bind amines, α-amino acid esters, and oligopeptides in water with significant selectivity. For binding of hydrophilic guests, 1, 2, and 3 bind histamine in pH 8 buffer at 25 °C with binding constants of 157000, 31000, and 18200 M^-1, respectively, where the coordination (Imd-Zn) and the salt bridge (NH₃⁺-^-OOC) stabilized the complex. The large dependence of the binding constants on the ionic strength indicated that the electrostatic interaction between the ammonium group of histamine and the carboxylate groups of receptor contributes significantly to tight binding in water. Receptors 1-3 also bind a cationic guest, Arg-OMe, with a binding constant of 440-11000 M^-1. The effective charge of the receptors for electrostatic recognition of Arg-OMe in pH 9 Borax (1 = 0.1 M) at 25 °C was evaluated by the Debye-Huckel limiting law as 4.2, 4.3, and 3.0 for 1, 2, and 3, respectively. These observations indicate that a salt bridge, which is exposed to water and involves hydrogen bonding, as seen in the 1-histamine and 1-Arg-OMe complexes, can be used as a significant recognition force. Binding of Arg-OMe by 2 and 3 was entropically driven, and binding of Arg-OMe by 1 was enthalpically driven. Therefore, the driving force of binding is desolvation from the ionic groups in the former case and the electrostatic attraction in the latter case. For binding of hydrophobic guests, 3 binds Trp-OMe or pyridine in water with binding constants of 7000-8000 M^-1, while 1 and Zn·TCPP bind these guests less tightly with binding constants of 20- 500 M^-1, indicating the importance of the long alkyl chains to provide a hydrophobic binding pocket above the porphyrin plane.

Full text of DOI:10.1021/ja9922126

J. Am. Chem. Soc. 1999, 121, 11425-11431
11425
Porphyrin Receptors for Amines, Amino Acids, and Oligopeptides in
Water
Tadashi Mizutani,* Kenji Wada, and Susumu Kitagawa
Contribution from the Department of Synthetic Chemistry and Biological Chemistry, Graduate School of
Engineering, Kyoto UniVersity, Sakyo-ku, Kyoto 606-8501, Japan
ReceiVed June 28, 1999
Abstract: A series of water-soluble porphyrin receptors having a hydrophobic binding pocket, a Lewis acidic
site (Zn), and an electrostatic recognition site (COO- groups) were prepared. All the porphyrin receptors have
a [meso-tetrakis(p-carboxyphenyl)porphyrinato]zinc (Zn‚TCPP) as a common structural unit, and eight
ω-carboxyalkyloxy groups (alkyloxy ) methoxy (1), butoxy (2), decyloxy (3)) at the ortho positions of the
phenyl groups. These receptors bind amines, R-amino acid esters, and oligopeptides in water with significant
selectivity. For binding of hydrophilic guests, 1, 2, and 3 bind histamine in pH 8 buffer at 25 °C with binding
constants of 157000, 31000, and 18200 M-1, respectively, where the coordination (Imd-Zn) and the salt
bridge (NH3+--OOC) stabilized the complex. The large dependence of the binding constants on the ionic
strength indicated that the electrostatic interaction between the ammonium group of histamine and the carboxylate
groups of receptor contributes significantly to tight binding in water. Receptors 1-3 also bind a cationic guest,
Arg-OMe, with a binding constant of 440-11000 M-1. The effective charge of the receptors for electrostatic
recognition of Arg-OMe in pH 9 Borax (I ) 0.1 M) at 25 °C was evaluated by the Debye-Hu¨ckel limiting
law as 4.2, 4.3, and 3.0 for 1, 2, and 3, respectively. These observations indicate that a salt bridge, which is
exposed to water and involves hydrogen bonding, as seen in the 1-histamine and 1-Arg-OMe complexes,
can be used as a significant recognition force. Binding of Arg-OMe by 2 and 3 was entropically driven, and
binding of Arg-OMe by 1 was enthalpically driven. Therefore, the driving force of binding is desolvation
from the ionic groups in the former case and the electrostatic attraction in the latter case. For binding of
hydrophobic guests, 3 binds Trp-OMe or pyridine in water with binding constants of 7000-8000 M-1, while
1 and Zn‚TCPP bind these guests less tightly with binding constants of 20-500 M-1, indicating the importance
of the long alkyl chains to provide a hydrophobic binding pocket above the porphyrin plane.
Introduction
interactions are the major driving force of binding. For instance,
cyclodextrin binds hydrophobic guests such as substituted
benzenes with the association constant ranging from 102 to 104
M-1.3 Because most biomolecules have polar functional groups,
a general strategy to recognize such polar functional groups in
water needs to be established. Recognition of polar molecules
in water has two intrinsic difficulties: (1) polar functional groups
are better hydrated and diminish hydrophobic interactions and
(2) hydrogen bonding or electrostatic interactions between polar
functional groups are weakened by hydration when compared
to the interactions in nonpolar solvents; the interaction energy
is sometimes comparable to changes in solvation energy and is
no longer a major driving force of binding. This is seen in the
weak binding of hydrophilic guests by cyclodextrins. Although
modification of cyclodextrin with polar substituents improves
the binding affinity compared to the parent cyclodextrin, the
overall binding affinity is still moderate.4
Recognition of biomolecules in water is a challenging subject
in host-guest chemistry. Continuing efforts have been made
to understand the mechanism of recognition and to develop an
artificial receptor comparable to proteins.1 Design of a molecule
that recognizes a target molecule in nonpolar solvents is rather
straightforward because contribution from solvation energy is
less important than the polar interactions between receptor and
guest, so that, as a first approximation, only interactions between
host and guest should be considered to design the system.
However, strategies successful in receptor design in nonpolar
organic solvents are not always applicable to receptor design
in water, since polar interactions are considerably hindered by
hydration.2 Thus, a target of water-soluble synthetic receptors
is most frequently a nonpolar molecule, where hydrophobic
(1) For recent studies on molecular recognition in water, see: (a)
Schneider, H.-J.; Kramer, R.; Simova, S.; Schneider, U. J. Am. Chem. Soc.
1988, 110, 6442. (b) Kral, V.; Furuta, H.; Shreder, K.; Lynch, V.; Sessler,
J. L. J. Am. Chem. Soc. 1996, 118, 1595. (c) Chen, H.; Ogo, S.; Fish, R. H.
J. Am. Chem. Soc. 1996, 118, 4993. (d) Dandliker, P. J.; Diederich, F.;
Zingg, A.; Gisselbrecht, J. P.; Gross, M.; Louati, A.; Sanford, E. HelV. Chim.
Acta 1997, 80, 1773. (e) Nelen, M. I.; Eliseev, A. V. J. Chem. Soc., Perkin
Trans. 2 1997, 1359. (f) Metzger, A.; Anslyn, E. V. Angew. Chem., Int.
Ed. 1998, 37, 649. (g) Hossain, M. A.; Schneider, H.-J. J. Am. Chem. Soc.
1998, 120, 11208. (h) Park, H. S.; Lin, Q.; Hamilton, A. D. J. Am. Chem.
Soc. 1999, 121, 8. (i) Ngola, S. M.; Kearney, P. C.; Mecozzi, S.; Russell,
K.; Dougherty, D. A. J. Am. Chem. Soc. 1999, 121, 1192. (j) Inouye, M.;
Fujimoto, K.; Furusyo, M.; Nakazumi, H. J. Am. Chem. Soc. 1999, 121,
1452. (k) Sirish, M.; Schneider, H.-J. Chem. Commun. 1999, 907.
(2) Adrian, J. C., Jr.; Wilcox, C. S. J. Am. Chem. Soc. 1991, 113, 678.
A new water-soluble host having both a hydrophobic binding
pocket and polar recognition groups should help in clarifying
the binding mechanism of polar molecules in water. A rigid
and hydrophobic porphyrin framework can be used as a receptor
in water if it is solubilized by peripheral substitution by polar
groups. We prepared water-soluble zinc porphyrins 1-3, which
have a Lewis acidic site (Zn), a salt bridge site (COO- groups),
(3) Rekharsky, M. V.; Inoue, Y. Chem. ReV. 1998, 98, 1875.
(4) (a) Tabushi, I.; Shimizu, N.; Sugimoto, T.; Shiozuka, M.; Yamamura,
K. J. Am. Chem. Soc. 1977, 99, 7100. (b) Tabushi, I.; Kuroda, Y.; Mizutani,
T. Tetrahedron 1984, 40, 545. (c) Tabushi, I.; Kuroda, Y.; Mizutani, T. J.
Am. Chem. Soc. 1986, 108, 4514.
10.1021/ja9922126 CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/24/1999

11426 J. Am. Chem. Soc., Vol. 121, No. 49, 1999
Mizutani et al.
Scheme 1a
a Reagents and conditions: (a) H2SO4/MeOH; (b) R-Br, K2CO3/DMF; (c) Br2/CCl4, hν; (d) DMSO/NaHCO3; (e) BF3‚OEt2, pyrrole/CH2Cl2,
then DDQ; (f) Zn(OAc)2/CHCl3; (g) 0.5 M KOH, MeOH, THF.
and a hydrophobic binding pocket (the porphyrin framework,
aryl groups, and alkyl chains). We report here that these
receptors bind both hydrophilic and hydrophobic guests such
as amines, amino acids, and oligopeptides with significant
selectivity in water, with particular affinity for histamine and a
histidine-containing oligopeptide.5
isotherm or the 1:1 and 1:2 binding isotherm for some cationic
guests: K1 ) [P‚G]/([P][G]), and K2 ) [P‚G2]/([P‚G][G]). The
binding constants K1 and K2 (where applicable) were determined
for 1-3 and Zn‚TCPP ([5,10,15,20-tetra(p-carboxyphenyl)-
porphyrinato]zinc), and are listed in Table 1. Standard deviations
for the curve fitting were less than 5%. For the binding by Zn‚
TCPP, isosbestic points were not observed. The binding isotherm
for Zn‚TCPP was analyzed based on the 1:1 complexation,
because attempts to analyze the data based on the 1:1 and 1:2
complexation failed. Nevertheless, the standard deviations for
the curve fitting were less than 5%. For other receptor-guest
combinations except for the combinations involving 1:2 com-
plexes, isosbestic points were always observed. In the following
discussions, we focus only on the magnitude of K1. The binding
constants were also determined in CH2Cl2 using 9b, 9c, and
Zn‚TPP ([5,10,15,20-tetraphenylporphyrinato]zinc), and are
listed in Table 2. Details of the binding behavior of 9a-c will
be reported elsewhere.
Results
Synthesis and Characterization of Porphyrin Receptors.
Porphyrin receptors 1-3 were prepared according to Scheme
1. Precursor 4 was prepared from 3,5-dihydroxybenzoic acid
according to the reported method.6 Aldehyde 6 was obtained
in 45-61% yield from 4. Condensation with pyrrole by
Lindsey’s method7 afforded free base porphyrins 8a-c in
15-22% yield. Zinc insertion followed by alkaline hydrolysis
gave 1-3 in 63-75% yield. Compounds 4-8 were character-
ized by 1H NMR, high-resolution mass spectroscopy, and
elemental analysis. The structure and purity of 1-3 were
confirmed by 1H NMR spectroscopy. The zinc porphyrins 1-3
are soluble in water, and they are also soluble in a MeOH-
water mixture. In Borax buffer at pH 9 (ionic strength, I ) 0.1
M), the absorbance in the Q-band followed the Lambert-Beer
law up to at least 90 µM, indicating that the porphyrin is
monomeric. At pH <7.5, a gradual decrease in the absorbance
in the Soret band was observed, suggesting that aggregation
occurred. At pH >11, the absorbance in the Soret band
increased, suggesting the deprotonation of Zn-OH2 occurred.
Thus, the binding experiments were performed at pH 8 or 9.
Dynamic light scattering experiments of a solution of 0.1 mM
of 3 in pH 9 Borax buffer showed that no micellar-like aggregate
was formed.
1H NMR Studies of Conformations of Receptors. 1H NMR
spectra of the solution of 3 showed that the chemical shifts of
the alkyl chain protons of 3 changed upon addition of guest or
inorganic salts. Addition of NaCl to the solution of 3 in a Borax
buffer caused the upfield shifts in all the alkyl methylene protons
as shown in Figure 1. The assignments of signals are based on
1
the H-1H COSY and homonuclear spin-decoupling experi-
ments. Addition of pyridine to a solution of 3 caused upfield
shifts in the H1′-H3′ resonances and downfield shifts in the
H4′-H9′ resonances (Figure 2). In Figure 3, the 1H NMR
spectra of 2 and 3 are compared with their free bases. The
resonances of all the alkyl protons are shifted downfield upon
zinc insertion except for the H-4′ protons of 2. The resonance
of H-4′ protons of 2 is shifted upfield relative to that of the
free base.
Binding of Amines, r-Amino Acid Esters, and Oligopep-
tides. Addition of guest caused a red-shift in the Soret band, a
typical spectral change due to the amino group coordination to
zinc. The binding constants were determined by monitoring the
absorbance changes of the Soret band as a function of guest
concentrations, and fitting the saturation plot to the 1:1 binding
Enthalpy and Entropy Changes in Complexation. The
enthalpy changes and the entropy changes of complexation were
determined by a van’t Hoff plot using the binding constants K1
determined by UV-vis titration experiments in the temperature
range of 2-25 °C. Figure 4 shows the plots of ∆H° against
∆S°, indicating that there is enthalpy-entropy compensation
for the binding. The values of ∆H° and ∆S° of complexation
showed a substantial variation, depending on the combination
of the host-guest complexes. For instance, although the binding
(5) For amino acids recognition in water by zinc porphyrins, see: (a)
Mikros, E.; Gaudemer, A.; Pasternack, R. Inorg. Chim. Acta 1988, 153,
199. (b) Konishi, K.; Yahara, K.; Toshishige, H.; Aida, T.; Inoue, S. J.
Am. Chem. Soc. 1994, 116, 1337.
(6) Borchardt, R. T.; Sinhababu, A. K. J. Org. Chem. 1981, 46, 5021.
(7) Lindsey, J. S.; Wagner, R. W. J. Org. Chem. 1989, 54, 828.

Preparation of Water-Soluble Porphyrin Receptors
J. Am. Chem. Soc., Vol. 121, No. 49, 1999 11427
Table 1. Binding Constants, K1, K2/M-1, between Porphyrin
Receptors and Amines, R-Amino Esters, or Oligopeptides in
Aqueous Buffer (Borax at pH 9 and I ) 0.1 M) at 25 °Ca
K1 (K2)b
entry
guest
Gly-OMe
1
2
3
Zn‚TCPP
1
2
3
4
5
6
7
8
9
15
e
5
e
f
f
f
30
e
67
e
43
15
13
13
2
460
240
240
340
16
16
14
14
14
43
59
27
89
300
370
520
240
9
Ala-OMec
Ala-OMe
Ala-OMed
Val-OMe
Leu-OMe
PhGly-OMe
Phe-OMe
Trp-OMec
1
2
15
12
107
550
670
680
310
8
1130
7000
7160
7600
2750
46
10 Trp-OMe
11 Trp-OMed
12 Tyr-OMe
13 Ser-OMe
38
5
14 His-OMe
15 Lys-OMe
16 Arg-OMeh
17 Arg-OMec
18 Arg-OMeg
19 Arg-OMe
20 pyridine
71
400 (100)
410
140
1210
690
200
57
67
150
e
105
21
e
11100 (2690) 4500 (830) 5700 (350)
7430 (1670)
2780 (700)
1400 (270)
14
e
e
7
20 (1)
47 (8)
190 (21)
2690 (100)
13 900 (1930) 30 100
157 000 (6100) 31 000
2540 (580) 4310 (470)
1050 (280) 2250 (410)
440 (150) 1260 (90)
1920
19
45
340
2970
2950
2960
9500
8200
15
21 butylamine
22 benzylamine
23 nicotinamide
24 imidazole
25 imidazoleh
26 imidazolei
27 histamine
28 histamineh
29 histaminei
30 Trp-Arg-OH
31 Gly-His-Lys-OH
430
1500
2610
2660
2600
5170
17 400
18 200
360 (28)
480
850
2850
7
e
26
69
e
68
90
150
150
e
150
e
270
6
8
2
11
13
16
Figure 1. 1H NMR spectra of a solution of 3 (1 mM) in (a) MeOH-
d4, (b) a Borax buffer at pD ) 8.6 and I ) 0.1 M, (c) a Borax buffer
at pD ) 8.6 and I ) 0.2 M, and (d) a Borax buffer at pD ) 8.6 and
I ) 0.5 M. The ionic strength was adjusted by adding NaCl.
83 (12)
1
73 (0.3)
170
660
32 Gly-His-Lys-OHh 790 (2)
of histamine. The following two observations revealed the
importance of the electrostatic interactions in the binding: (1)
the binding constant of histamine was much larger than that of
imidazole, and (2) the binding constant of histamine at pH 9
was increased when the ionic strength was reduced from 0.1 to
0.01 M (Table 1, entries 27-28) and it was also increased when
the pH was changed from 9 to 8 (Table 1, entries 28-29). Since
the pKa values of histamine are 6.1 (imidazole) and 9.9 (amino),8
histamine with the protonated NH2 group, which is the major
species at pH 8, has much higher affinity for 1 than neutral
histamine. These results indicate that the contribution of the
electrostatic interaction between the COO- groups of 1 and the
33 Gly-His-Lys-OHi 22 300 (1070) 3180
34 Ala-NH2
35 Lys-OH
e
e
e
e
e
e
f
f
13
6
13
6
7
6
f
36
36 Ala-OMej
37 Leu-OMej
38 Phe-OMej
39 Trp-OMej
94
a Averages of 2-3 independent determinations. The estimated errors
in K1 and K2 were less than 5%. b K1 and K2 are the binding constants
for 1:1 and 1:2 complexes, respectively. c I ) 0.02 M (KCl). d I ) 0.5
f
M (KCl). e Not determined. Not bound. g I ) 0.05 M (KCl). h I ) 0.01
M (KCl). i Borax at pH 8 and I ) 0.01 M. j MeOH-Borax (pH 9.0, I
) 0.1 M) ) 10:1 (v/v).
NH3 group of histamine was quite large even in water.9 A
+
Table 2. Binding Constants, K1/M-1, between Porphyrin Receptors
9 and R-Amino Acid Esters or Amines in CH2Cl2 at 25 °C
molecular modeling study showed that the ammonium group
of histamine can interact with two carboxylate groups of receptor
1 via hydrogen bonding (Figure 5). Receptors 1-3 also bind
tripeptide, Gly-His-Lys-OH, with a binding constant of 2850-
22 300 M-1 in 0.01 M Borax at pH 8. A large dependence on
pH and ionic strength (Table 1, entries 31-33) suggests that
the complexation is also dominated by salt bridge formation
between the N-terminal NH3+ group of the guest and the COO-
groups of 1-3.
The binding affinity for cationic amino esters, Lys-OMe and
Arg-OMe, is moderate compared to that of histamine. The
binding constants of Arg-OMe were decreased with an increase
in the ionic strength. In contrast, the binding constants of Ala-
guest
9b 9c Zn•TPPa
guest
9b
9c Zn•TPPa
Ala-OMe
Leu-OMe
PhGly-OMe
75 920
0.5 18
4660
6860
5250
Phe-OMe 36
Trp-OMe 110 1090
pyridine 401 4480
360
4910
8730
7720
b
13
a [5,10,15,20-Tetra(phenyl)porphyrinato]zinc. b Not bound.
constants of Arg-OMe by 1 and 3 were similar in magnitude
(1400 and 1260 M-1, respectively), the different signs and values
of ∆H° and ∆S° (Figure 4, a vs. g) indicate that the origin of
the driving force of binding is different.
Discussion
(8) Rabenstein, D. L.; Bratt, P.; Peng, J. Biochemistry 1998, 37, 14121.
(9) Hosseini et al. reported that the binding of ATP by a protonated
macrocyclic polyamine was driven by hydrogen bonding between the
ammonium group and the phosphate group, with the binding constant of
1011 M-1. See: Hosseini, M. W.; Lehn, J.-M.; Mertes, M. P. HelV. Chim.
Acta 1983, 66, 2454. See also: Gokel, G. W.; Abel, E. ComprehensiVe
Supramolecular Chemistry; Gokel, G. W., Ed.; Pergamon Press Ltd.:
Oxford, 1996; Vol. 1, p 511. Mock, W. L.; Shih, N.-Y. J. Org. Chem. 1986,
51, 4440.
Binding of Hydrophilic Guests. Receptors 1-3 showed tight
binding to cationic amino esters (Lys-OMe and Arg-OMe) and
histamine in water (Borax, pH 9, I ) 0.1 M). The large binding
constant for histamine by 1 is noteworthy. This high affinity
can be ascribed to the coordination interaction between zinc
and the imidazolyl group as well as the electrostatic interaction
between the carboxylate groups of 1 and the ammonium group

11428 J. Am. Chem. Soc., Vol. 121, No. 49, 1999
Mizutani et al.
Figure 3. 1H NMR spectra of a solution of 2 or 3 and the corresponding
free bases in D2O (1 mM, Borax, pD 8.6): (a) free base of 2, (b) 2, (c)
free base of 3, and (d) 3.
Figure 2. Plots of the complexation-induced chemical shifts (∆δ/ppm)
of 3 (1.0 mM) against the concentration of pyridine in Borax buffer at
pD ) 8.6 and I ) 0.1 M at 298 K. Curves were calculated by a least-
squares method based on 1:1 complex formation. The binding constant
determined from these chemical shift displacements for all the signals
except for H-4′ and H-10′ was 5900 ( 700 M-1
.
OMe and Trp-OMe were constant in the range of ionic strength
from 0.02 to 0.5 M (Table 1, entries 2-4 and 9-11). The
Debye-Hu¨ckel limiting law gives the following relationship,1a,10
x
log K ) log K′ - 1.018|zazb| I (in water, 25 °C), where K′ is
the binding constant at I ) 0, and za and zb are the effective
charges of receptor and guest, respectively. The plots of log K
x
against I afforded a straight line, and the effective charge (za)
of 1, 2, and 3 for binding of Arg-OMe was determined to be
4.2, 4.3, and 3.0, respectively, if we assume that zb is 1. These
results support the view that the carboxylate groups of 1-3 serve
as the electrostatic recognition site. The effective charge in the
range of 3 to 4 suggests that only a portion of the COO- groups
contribute electrostatic interactions.
Figure 4. Plot of the enthalpy changes of complexation against the
entropy changes in Borax buffer at pH 9.0 and I ) 0.1 M: (a) 3-Arg-
OMe, (b) 2-Arg-OMe, (c) 2-py, (d) 2-Trp-OMe, (e) Zn‚TCPP-Trp-
OMe, (f) Zn‚TCPP-py, (g) 1-Arg-OMe, (h) 3-py, (i) 3-Trp-OMe,
(j) Zn‚TCPP-Arg-OMe.
Binding of Hydrophobic Guests. The hydrophobic interac-
tion is also important as a driving force of recognition. The
binding by hydrophobic interactions becomes very strong if the
contact surface area between host and guest is high. Breslow et
al. reported that the cyclodextrin dimer binds a hydrophobic
R-amino acids and R-amino amides showed very weak affinity
(Table 1, entries 34 and 35). This is partly because the basicity
of the amino group of R-amino acids and R-amino amides is
higher than that of R-amino esters. For instance, the values of
pKa of Ala-OH, Ala-OMe, and Ala-NH2 are 9.7, 7.7, and 8.2,
respectively.12 Therefore, at pH 9, most of the NH2 group of
Ala-OMe is not protonated and is able to interact with zinc,
while a major part of the NH2 group of Ala-OH is protonated
and is unable to interact with zinc. The small binding constants
guest with a binding constant of 1010 M-1 11
If one compares
.
the binding constants for a given hydrophobic guest such as
Leu-OMe, Phe-OMe, and pyridine among 1, 2, and 3, then 3
showed the largest binding constants (Table 1, entries 6, 8, and
20). This is characteristic of the receptor-guest complex
formation driven by hydrophobic interactions. Although methyl
esters of R-amino acids showed moderate affinity for 1-3,
(10) Hossain, M. A.; Schneider, H.-J. Chem. Eur. J. 1999, 5, 1284.
(11) Breslow, R.; Chung, S. J. Am. Chem. Soc. 1990, 112, 9659.
(12) Hay, R. W.; Morris, P. J. J. Chem. Soc. B 1970, 1577.

Preparation of Water-Soluble Porphyrin Receptors
J. Am. Chem. Soc., Vol. 121, No. 49, 1999 11429
Another interesting feature is that, in CH2Cl2, 9b and 9c
showed lower affinity for a given guest than Zn‚TPP, while 2
and 3 showed higher affinity than 1 and Zn‚TCPP for aliphatic
and aromatic amino esters, suggesting that the alkyl groups of
2 and 3 stabilize the complex by strengthening hydrophobic
effects and/or the zinc of 2 and 3 is reactive due to the
hydrophobic environment provided by the alkyl groups. In
methanol-pH 9 Borax (10:1 (v/v)), the binding constants of
Ala-OMe, Leu-OMe, Phe-OMe, and Trp-OMe by 2, 3, and Zn‚
TCPP are decreased to 2-94 M-1 (Table 1, entries 36-39).
These observations also support the important role of water in
selective binding. Thus, the hydrophobic binding pocket pro-
duced in water should be an important strategy for the design
of receptors in water.
Figure 5. A possible binding geometry of a complex between
+
protonated histamine and 1. The distances between N of NH3 and O
Conformations of Receptors. Figure 1 shows that the
chemical shifts of the alkyl groups of 3 change as the ionic
strength changes or guest is added. The addition of salt will
reduce the electrostatic repulsion between the COO- groups so
that the ω-carboxyalkyl groups tend to aggregate above the
porphyrin plane. This will explain the upfield shifts of the alkyl
protons at higher ionic strengths.
of COO- are 2.72 and 2.70 Å. For the modeling study, the substituents
at 4′- and 6′-positions of 1 were replaced by hydrogens and methoxy
groups, respectively.
of simple amines such as butylamine or benzylamine (Table 1,
entries 21 and 22) are also attributable to the higher pKa values
of the amino groups, which are mostly protonated in pH 9 buffer.
For R-amino acids, electrostatic repulsion between the COO-
group of R-amino acid and the COO- groups of 1-3 also
disfavors the binding. Verche´re-Be`aur et al. reported that there
is no interaction between Zn‚TCPP and C-terminal free R-amino
1
Comparison of the H NMR chemical shifts of the alkyl
protons between free base and the zinc complexes 2 and 3
(Figure 3) showed that only the resonance of the R-protons (H-
4′) of the zinc complex of 2 shifted upfield relative to the free
base, while the resonances of other protons are shifted down-
field. This implies that there is a strong tendency for the COO-
group of 2 to intramolecularly coordinate to the zinc.
1
acids in water on the basis of H NMR studies.13
1H NMR studies (Figure 2) showed that the alkyl protons of
3 underwent a complexation-induced upfield or downfield shift
upon binding pyridine, indicating that the guest binding induces
conformational changes in the receptor.14 Upfield shifts of H1′-
H3′ resonances and downfield shifts of H4′-H9′ resonances
suggest the following conformational changes. The C1′-C3′
carbons move toward the porphyrin framework to accommodate
a hydrophobic environment for the pyridine binding, this
conformational change then gives rise to the strong repulsion
between the COO- groups, and C4′-C9′ carbons will be forced
to move away from the porphyrin framework. The binding
constant determined from the NMR titration was 5900 ( 700
M-1 for the 3-pyridine complex. Anomalous shifts were
observed for the H-4′ resonance and it is difficult to explain
this behavior. Zn‚TCPP having no carboxyalkyl groups showed
only weak affinity for most of the guests studied. These
observations suggest that the ω-carboxyalkyl chains of 1-3 form
a hydrophobic binding pocket with the porphyrin framework,
and this binding pocket favors the binding of hydrophobic guest.
Surprisingly, 1-3 showed very weak affinity for hydrophobic
bulky amino acid esters such as Val-OMe, Leu-OMe, and
PhGly-OMe, although Zn‚TCPP showed moderate affinity for
these guests. This seems contradictory to the role of carboxy-
alkyl groups described above that the carboxyalkyl chains help
bind hydrophobic guest. However, this behavior can be under-
stood if one sees the same trend that 9b and 9c have weak
affinity toward Leu-OMe and PhGly-OMe in CH2Cl2 (Table
2).15 Therefore, the weak affinity of 2 and 3 for Val-OMe, Leu-
OMe, and PhGly-OMe can be attributable to some steric
repulsion in the small binding pocket of 2 and 3, not to
solvation-desolvation processes.
Thermodynamic Parameters of Complex Formation. The
values of enthalpy changes and entropy changes of complexation
were smaller in magnitude than those reported for the binding
of amines or R-amino esters by zinc porphyrins in organic
solvents. Typically, ∆H° ranges from -30 to -100 kJ/mol and
∆S° from -30 to -220 J/(K mol) in organic solvents.16 The
smaller values of ∆H° and ∆S° in water reported here are
attributable to the enthalpy-entropy compensation effects,
which are partially owing to solvation-desolvation processes.17
The compensation temperature (Tc) determined from the slope
of the plot (Figure 4) was 81 °C. Thus, at 81 °C, the free energy
changes of the binding are the same among the complexes if
the compensation relationship is strictly followed, leading to
diminished selectivity of recognition.
Figure 4 shows that the binding of hydrophobic guest was
driven enthalpically. The binding of Arg-OMe by 1 was also
enthalpically driven. In contrast, the binding of cationic guest
by hydrophobic receptors, 2 and 3, is entropically driven (Figure
4, parts a and b). The origin of the entropic driving force for
the binding of Arg-OMe by 2 and 3 seems to be the desolvation
of the ionic groups of both receptor and guest upon salt bridge
formation in a relatively hydrophobic environment. We reported
that the binding of 1,2,4,5-benzenetetracarboxylate by
[5,10,15,20-tetrakis(1-pentyl-3-pyridinio)porphyrinato]zinc(II) in
water was entropically driven.18 Kano et al. also reported
entropically driven complexation between cationic cyclodextrin
and N-acetylamino acids in water.19 Thus entropically driven
(16) Imai, H.; Nakagawa, S.; Kyuno, E. J. Am. Chem. Soc. 1992, 114,
6719. 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. Mizutani, T.; Murakami, T.; Kurahashi, T.; Ogoshi, H.
J. Org. Chem. 1996, 61, 539.
(17) Lamry, R.; Rajender, S. Biopolymers 1970, 9, 1125-1227.
(18) Mizutani, T.; Horiguchi, T.; Koyama, H.; Uratani, I.; Ogoshi, H.
Bull. Chem. Soc. Jpn. 1998, 71, 413.
(13) Verche´re-Be`aur, C.; Mikros, E.; Perree-Fauvet, M.; Gaudemer, A.
J. Inorg. Biochem. 1990, 40, 127.
(14) Similar complexation-induced shifts were also observed for the
complex between 3 and Ala-OMe, so that the shifts observed for the pyridine
complex are not ascribed to the ring current effects of pyridine.
(15) Profound steric effects on binding by a zinc porphyrin were
reported: Bhyrappa, P.; Vaijayanthimala, G.; Suslick, K. S. J. Am. Chem.
Soc. 1999, 121, 262.
(19) Kitae, T.; Nakayama, T.; Kano, K. J. Chem. Soc., Perkin Trans. 2
1998, 207-212

11430 J. Am. Chem. Soc., Vol. 121, No. 49, 1999
Mizutani et al.
182.0579, found 182.0581. Anal. Calcd for C9H10O4: C, 59.34; H, 5.53.
Found: C, 59.19; H, 5.56.
binding seems to be a general mechanism of binding of ionic
molecules having hydrophobic moieties in water.20
Methyl 3,5-Bis(methoxycarbonylmethoxy)-4-methylbenzoate (5a).
This compound was prepared from 4 (5.0 g) in a manner similar to
that for 5c: yield 84% (7.54 g); 1H NMR (CDCl3) δ 2.26 (s, 3H, CH3),
3.79 (s, 6H, CO2Me), 3.87 (s, 3H, CO2Me), 4.69 (s, 4H, CH2), 7.24 (s,
2H, phenyl-H). Anal. Calcd for C15H18O8: C, 55.21; H, 5.56. Found:
C, 54.99; H, 5.52.
Conclusions
The present study demonstrated that anionic zinc porphyrins
bind histamine and a histidine-containing oligopeptide tightly.
Strong dependence of the binding affinity for these guests on
ionic strength and pH revealed that electrostatic interactions
between charged functional groups are an important driving
force for recognition of hydrophilic guest molecules in water.
Binding of ionic guest by salt-bridge formation in a hydrophobic
environment was driven by entropically favorable desolvation
as seen in the positive ∆S° for the 3-Arg-OMe complex. Lower
affinity of receptors in MeOH-water than in water indicated
that water plays a significant role in binding energetics.
Comparisons of binding affinity between hydrophilic receptor
1 and hydrophobic receptor 3 revealed that the hydrophobic
binding pocket of 3 constructed in water enhances the affinity
toward hydrophobic guests.
Methyl 3,5-Bis(4-methoxycarbonylbutoxy)-4-methylbenzoate (5b).
This compound was prepared from 4 (3.82 g) in a manner similar to
1
that for 5c: yield 91% (7.85 g); H NMR (CDCl3) δ 1.83 (m, 8H,
CH2), 2.11 (s, 3H, CH3), 2.39 (t, J ) 7.0 Hz, 4H, CH2), 3.66 (s, 6H,
CO2Me), 3.88 (s, 3H, CO2Me), 4.00 (t, J ) 5.5 Hz, 4H, CH2), 7.15 (s,
2H, phenyl-H). Anal. Calcd for C21H30O8: C, 61.45; H, 7.37. Found:
C, 61.26; H, 7.33.
Methyl 3,5-Bis(10-methoxycarbonyldecyloxy)-4-methylbenzoate
(5c). A mixture of 4 (3.58 g, 21.2 mmol), methyl 11-bromoundecanoate
(14.2 g, 50.8 mmol), and K2CO3 (15.4 g) in DMF (55 mL) was stirred
at 40 °C for 24 h under N2. After AcOEt (300 mL) was added, the
organic layer was washed with saturated aqueous NaCl (50 mL × 3)
and dried over MgSO4. Evaporation of the solvent and washing with
hexane (200 mL) afforded a white solid of 5c (11.3 g, 95%): 1H NMR
(CDCl3) δ 1.28 (m, 20H, CH2), 1.45 (quintet, J ) 7.5 Hz, 4H, CH2),
1.60 (quintet, J ) 7.0 Hz, 4H, CH2), 1.78 (quintet, J ) 7.1 Hz, 4H,
CH2), 2.12 (s, 3H, CH3), 2.28 (t, J ) 7.5 Hz, 4H, CH2), 3.65 (s, 6H,
CO2Me), 3.88 (s, 3H, CO2Me), 3.98 (t, J ) 6.5 Hz, 4H, CH2), 7.16 (s,
2H, phenyl-H); HRMS (FAB) calcd for C33H54O8 (M+) 578.3818, found
578.3809. Anal. Calcd for C33H54O8: C, 68.48; H, 9.40. Found: C,
68.36; H, 9.60.
Experimental Section
1
General Methods. H NMR spectra were obtained using a JEOL
A-500 spectrometer and chemical shifts are reported relative to Me4Si
or residual protons of deuterated solvents. UV-vis spectra were
recorded on a Hewlett-Packard 8452 diode array spectrophotometer
with a thermostated cell compartment. Dynamic light scattering was
performed with a 10 mW He-Ne laser. High-resolution mass spectra
were obtained with a JEOL JMS-HX110A mass spectrometer using
3-nitrobenzyl alcohol as a matrix.
Materials. 3,5-Dihydroxy-4-methylbenzoic acid was prepared ac-
cording to the literature procedure.6 Methyl 11-bromoundecanoate and
methyl 5-bromovalerate were prepared by esterification of the corre-
sponding acid in a manner similar to that described for 4.
Methyl 4-(Bromomethyl)-3,5-bis(methoxycarbonylmethoxy)ben-
zoate (6a). A solution of 5a (2.0 g, 6.13 mmol) in CCl4 (30 mL) and
benzene (15 mL) was irradiated with a 500-W lamp while Br2 (0.99 g,
6.2 mmol) in CCl4 (10 mL) was added dropwise over 30 min. The
same workup described for 6c gave 6a: yield 88% (2.19 g); 1H NMR
(CDCl3) δ 3.79 (s, 6H, CO2Me), 3.88 (s, 3H, CO2Me), 4.75 (s, 2H,
CH2Br), 4.78 (s, 4H, CH2), 7.12 (s, 2H, phenyl-H). Anal. Calcd for
C15H17BrO8: C, 44.46; H, 4.23. Found: C, 44.65; H, 4.13.
Methyl 4-(Bromomethyl)-3,5-bis(4-methoxycarbonylbutoxy)ben-
zoate (6b). This compound was prepared from 5b (2.0 g) in a manner
UV-Vis Titrations. Binding constants were determined by UV-
vis titrations. The details of the determination of the binding constant
are as follows. Borax buffer solution was prepared by dissolving 1.53
g of H3BO3 in 50 mL of distilled water. The pH of the solution was
adjusted to 9 by adding KOH. KCl was added to make the ionic strength
0.1 M. To ca. 2 × 10-6 M of 1, 2, 3, or ZnTCPP in Borax buffer at pH
9.0 and I ) 0.1 M was added a stock solution of guest in the same
buffer at 25 °C. The decreases in absorbance at 424 nm and the increase
at 434 nm in the Soret band were monitored at different concentrations
of guest, with volume changes due to guest addition being taken into
account during analysis. The titration was completed within 15 min
after the amino ester solution was prepared to avoid any hydrolysis of
the ester. Isosbestic points were observed except for the binding by
Zn‚TCPP and the binding involving 1:2 complexes. Assuming 1:1 and
1:2 complexation (where applicable), the binding constants were
evaluated by a nonlinear least-squares parameter estimation based on
the Damping Gauss-Newton method.21
Methyl 3,5-Dihydroxy-4-methylbenzoate (4). A solution of 3,5-
dihydroxy-4-methylbenzoic acid (19.5 g, 116 mmol) in absolute
methanol (400 mL) and concentrated sulfuric acid (3.5 mL) was
refluxed for 12 h. After the solution was concentrated to about 100
mL, saturated aqueous NaHCO3 (200 mL) was added. The aqueous
layer was extracted with AcOEt (60 mL × 4) and the combined organic
layers were washed with saturated aqueous NaCl (50 mL × 2) and
dried over Na2SO4. Evaporation of the solvent and recrystallization from
AcOEt/CHCl3 afforded a white solid of 4 (17.1 g, 81%): 1H NMR
(acetone-d6) δ 2.11 (s, 3H, Me), 3.79 (s, 3H, CO2Me), 7.06 (s, 2H,
phenyl-H), 8.48 (s, 2H, OH); HRMS (FAB) calcd for C9H10O4 (M+)
1
similar to that for 6c: yield 79% (1.88 g); H NMR (CDCl3) δ 1.87
(m, 8H, CH2), 2.41 (t, J ) 7.0 Hz, 4H, CH2), 3.66 (s, 6H, CO2Me),
3.89 (s, 3H, CO2Me), 4.08 (t, J ) 5.5 Hz, 4H, CH2), 4.62 (s, 2H, CH2-
Br), 7.15 (s, 2H, phenyl-H); HRMS (FAB) calcd for C21H29BrO8 (M+)
488.1045, found 488.1061 Anal. Calcd for C21H29BrO8: C, 51.54; H,
5.97. Found: C, 51.02; H, 6.01.
Methyl 4-(Bromomethyl)-3,5-bis(10-methoxycarbonyldecyloxy)-
benzoate (6c). A solution of 5c (3.96 g, 6.84 mmol) in CCl4 (40 mL)
was irradiated with a 500-W lamp while Br2 (1.1 g, 6.9 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 (100 mL) was added. The organic layer was
washed with saturated aqueous NaHCO3 (50 mL × 2) and saturated
aqueous NaCl (50 mL × 2) and dried over MgSO4. Evaporation of the
solvent and recrystallization from ether afforded a white solid of 6c
(4.28 g, 95%): 1H NMR (CDCl3) δ 1.28 (m, 20H, CH2), 1.49 (quintet,
J ) 7.5 Hz, 4H, CH2), 1.60 (quintet, J ) 7.5 Hz, 4H, CH2), 1.82
(quintet, J ) 7.5 Hz, 4H, CH2), 2.28 (t, J ) 7.5 Hz, 4H, CH2), 3.64 (s,
6H, CO2Me), 3.89 (s, 3H, CO2Me), 4.05 (t, J ) 6.5 Hz, 4H, CH2),
4.63 (s, 2H, CH2Br), 7.15 (s, 2H, phenyl-H); HRMS (FAB) calcd for
C33H54BrO8 (MH+) 657.3002, found 657.2998. Anal. Calcd for C33H53-
BrO8: C, 60.27; H, 8.12. Found: C, 60.09; H, 7.88.
Methyl 4-Formyl-3,5-bis(methoxycarbonylmethoxy)benzoate (7a).
A solution of powdered 6a (2.0 g, 4.93 mmol) in DMSO (25 mL) and
solid NaHCO3 (3.0 g) was heated at 70 °C under N2 with vigorous
stirring for 30 min. The reaction mixture was then immediately cooled
in an ice bath, poured into saturated aqueous NaCl (100 mL), and
extracted with AcOEt (100 mL × 2). The organic layers were combined
and dried over MgSO4. Evaporation of the solvent and purification by
flash column chromatography (SiO2, AcOEt/hexane ) 1/1) afforded a
white solid of 7a (1.43 g, 85%): 1H NMR (CDCl3) δ 3.78 (s, 6H,
CO2Me), 3.90 (s, 3H, CO2Me), 4.77 (s, 4H, CH2), 7.14 (s, 2H, phenyl-
(20) Formation of some crown ether-metal cation complexes is entropi-
cally driven, see: Bradshaw, J. S.; Izatt, R. M.; Bordunov, A. V.; Zhu, C.
Y.; Hathaway, J. K. ComprehensiVe Supramolecular Chemistry; Gokel, G.
W., Ed.; Pergamon Press Ltd.: Oxford, 1996; Vol. 1, p 35.
(21) The least-squares fitting was performed by the computer program
SPANA, kindly provided by Prof. Y. Kuroda, Department of Polymer
Science, Kyoto Institute of Technology, Matsugasaki, Kyoto 606-8585,
Japan. e-mail: ykuroda@ipc.kit.ac.jp.

Preparation of Water-Soluble Porphyrin Receptors
J. Am. Chem. Soc., Vol. 121, No. 49, 1999 11431
H), 10.57 (s, 1H, CHO). Anal. Calcd for C15H16O9: C, 52.95; H, 4.74.
Found: C, 52.66; H, 4.82.
24H, CO2Me), 4.05 (s, 12H, CO2Me), 4.33 (s, 16H, CH2), 7.57 (s, 8H,
phenyl-H), 8.84 (s, 8H, â-pyrrole); UV-vis (CH2Cl2) λmax (log ꢀ) 423
(5.71), 555 (4.29), 592 (3.51); HRMS (FAB) calcd for C76H68N4O32Zn
(M+) 1612.3106, found 1612.3073.
Methyl 4-Formyl-3,5-bis(4-methoxycarbonylbutoxy)benzoate (7b).
A solution of powdered 6b (1.35 g, 2.76 mmol) in DMSO (30 mL)
and NaHCO3 (3.9 g) was heated at 90 °C under N2 with vigorous stirring
for 10 min. The same workup described for 7c gave 7b in 75% yield
(0.88 g): 1H NMR (CDCl3) δ 1.85 (m, 8H, CH2), 2.39 (t, J ) 7.0 Hz,
4H, CH2), 3.66 (s, 6H, CO2Me), 3.92 (s, 3H, CO2Me), 4.09 (t, J ) 5.8
Hz, 4H, CH2), 7.18 (s, 2H, phenyl-H), 10.50 (s, 1H, CHO); HRMS
(FAB) calcd for C21H29O9 (MH+) 425.1829, found 425.1812. Anal.
Calcd for C21H28O9: C, 59.43; H, 6.65. Found: C, 59.17; H, 6.54.
Methyl 4-Formyl-3,5-bis(10-methoxycarbonyldecyloxy)benzoate
(7c). A solution of powdered 6c (3.0 g, 4.6 mmol) in DMSO (60 mL)
and NaHCO3 (6.0 g) was heated at 120 °C under N2 with vigorous
stirring for 10 min. The reaction mixture was then immediately cooled
in an ice bath, poured into saturated aqueous NaCl (100 mL), and
extracted with AcOEt (100 mL × 2). The organic layers were combined
and dried over MgSO4. Evaporation of the solvent and recrystallization
from benzene/hexane afforded a white solid of 7c (2.26 g, 84%): 1H
NMR (CDCl3) δ 1.27 (m, 20H, CH2), 1.45 (quintet, J ) 7.4 Hz, 4H,
CH2), 1.60 (quintet, J ) 7.5 Hz, 4H, CH2), 1.81 (quintet, J ) 7.0 Hz,
4H, CH2), 2.28 (t, J ) 7.5 Hz, 4H, CH2), 3.65 (s, 6H, CO2Me), 3.92
(s, 3H, CO2Me), 4.06 (t, J ) 6.8 Hz, 4H, CH2), 7.18 (s, 2H, phenyl-
H), 10.51 (s, 1H, CHO); HRMS (FAB) (MH+) calcd for C33H53O9
593.3689, found 593.3702. Anal. Calcd for C33H52O9: C, 66.87; H,
8.84. Found: C, 66.21; H, 8.89.
[5,10,15,20-Tetrakis(4-methoxycarbonyl-3,5-bis(4-methoxycarbo-
nylbutoxy)phenyl)porphyrinato]zinc(II) (9b). This compound was
prepared from 8b (91 mg, 48 µmol) in a manner similar to that described
for 9c: yield 83% (78 mg); 1H NMR (CDCl3) δ 0.64 (quintet, J ) 7.5
Hz, 16H, CH2), 0.92 (quintet, J ) 7.5 Hz, 16H, CH2), 1.01 (t, J ) 7.0
Hz, 16H, CH2), 2.61 (s, 24H, CO2Me), 3.88 (t, J ) 6.5 Hz, 16H, CH2),
4.08 (s, 12H, CO2Me), 7.64 (s, 8H, phenyl-H), 8.66 (s, 8H, â-pyrrole);
UV-vis (CH2Cl2) λmax (log ꢀ) 425 (5.64), 552 (4.32), 592 (3.23); HRMS
(FAB) calcd for C100H116N4O32Zn (M+) 1948.6862, found 1948.6614.
[5,10,15,20-Tetrakis(4-methoxycarbonyl-3,5-bis(10-methoxycar-
bonyldecyloxy)phenyl)porphyrinato]zinc(II) (9c). A solution of 8c
(97 mg, 38 µmol) and Zn(OAc)2-saturated methanol (15 mL) in CHCl3
(110 mL) was refluxed for 3 h. After cooling, the solution was washed
with saturated aqueous NaHCO3 (100 mL × 2) and saturated aqueous
NaCl (100 mL × 2) and dried over Na2SO4. Evaporation of the solvent
and purification by flash column chromatography (SiO2, CHCl3/AcOEt
) 1/1) and recrystallization from CH2Cl2/hexane afforded a pink solid
of 9c (80 mg, 80%): 1H NMR (CDCl3) δ 0.55 (quintet, J ) 7.5 Hz,
16H, CH2), 0.71 (m, 32H, CH2), 0.82 (quintet, J ) 7.5 Hz, 16H, CH2),
0.93 (m, 32H, CH2), 1.03 (quintet, J ) 7.5 Hz, 16H, CH2), 1.43 (quintet,
J ) 7.5 Hz, 16H, CH2), 2.17 (t, J ) 7.5 Hz, 16H, CH2), 3.60 (s, 24H,
CO2Me), 3.82 (t, J ) 6.5 Hz, 16H, CH2), 4.07 (s, 12H, CO2Me), 7.63
(s, 8H, phenyl-H), 8.66 (s, 8H, â-pyrrole); UV-vis (CH2Cl2) λmax (log
ꢀ) 424 (5.63), 551 (4.26), 592 (3.45); HRMS (FAB) calcd for
5,10,15,20-Tetrakis(4-methoxycarbonyl-3,5-bis(methoxycarbon-
ylmethoxy)phenyl)porphyrin (8a). This compound was prepared from
7a (680 mg, 2 mmol) in a manner similar to that described for 8c:
C
148H212N4O32Zn (M+) 2621.4374, found 2621.4663.
1
yield 15% (120 mg); H NMR (CDCl3, 500 MHz) δ -2.62 (s, 2H,
[5,10,15,20-Tetrakis(4-carboxy-3,5-bis(carboxymethoxy)phenyl)-
NH), 3.42 (s, 24H, CO2Me), 4.05 (s, 12H, CO2Me), 4.34 (s, 16H, CH2),
7.57 (s, 8H, phenyl-H), 8.78 (s, 8H, â-pyrrole); UV-vis (CH2Cl2) λmax
(log ꢀ) 419 (5.66), 512 (4.43), 544 (3.83), 587 (4.00). Anal. Calcd for
C76H70N4O32: C, 58.84; H, 4.55; N, 3.61. Found: C, 58.15; H, 4.52;
N, 3.29.
porphyrinato]zinc(II) Potassium Salt (1). This compound was
prepared from 9a (87 mg, 54 µmol) in a manner similar to that described
for 3: yield 89% (92 mg); 1H NMR (deuterated Borax buffer, pD 8.6,
I ) 0.1 M) δ 4.27 (s, 16H, CH2), 7.53 (s, 8H, phenyl-H), 9.07 (s, 8H,
â-pyrrole); UV-vis (Borax buffer, pH 9.0, I ) 0.1 M) λmax (log ꢀ)
423 (5.41), 555 (3.98), 592 (3.18).
5,10,15,20-Tetrakis(4-methoxycarbonyl-3,5-bis(4-methoxycar-
bonylbutoxy)phenyl)porphyrin (8b). This compound was prepared
from 7b (2.122 g, 5 mmol) in a manner similar to that described for
8c: yield 22% (527 mg); 1H NMR (CDCl3) δ -2.62 (s, 2H, NH), 0.83
(quintet, J ) 7.5 Hz, 16H, CH2), 0.98 (quintet, J ) 7.5 Hz, 16H, CH2),
1.51 (t, J ) 7.0 Hz, 16H, CH2), 3.06 (s, 24H, CO2Me), 3.88 (t, J ) 6.5
Hz, 16H, CH2), 4.07 (s, 12H, CO2Me), 7.63 (s, 8H, phenyl-H), 8.58
(s, 8H, â-pyrrole); UV-vis (CH2Cl2) λmax (log ꢀ) 422 (5.62), 515
(4.38), 548 (3.95), 590 (4.00); HRMS (FAB) calcd for C100H118N4O32
(M+) 1886.7728, found 1886.7751. Anal. Calcd for C100H118N4O32: C,
63.62; H, 6.30; N, 2.97. Found: C, 63.37; H, 6.32; N, 2.92.
[5,10,15,20-Tetrakis(4-carboxy-3,5-bis(4-carboxybutoxy)phenyl)-
porphyrinato]zinc(II) Potassium Salt (2). This compound was
prepared from 9b (51 mg, 26 µmol) in a manner similar to that described
for 3: yield 84% (49 mg); 1H NMR (deuterated Borax buffer, pD 8.6,
I ) 0.1 M) δ 0.89 (quintet, J ) 7.5 Hz, 16H, CH2), 1.19 (quintet, J )
7.5 Hz, 16H, CH2), 1.66 (t, J ) 7.5 Hz, 16H, CH2), 3.99 (t, J ) 7.0
Hz, 16H, CH2), 7.76 (s, 8H, phenyl-H), 8.91 (s, 8H, â-pyrrole); UV-
vis (Borax buffer, pH 9.0, I ) 0.1 M) λmax (log ꢀ) 426 (5.42), 557
(4.03), 596 (3.38).
5,10,15,20-Tetrakis(4-methoxycarbonyl-3,5-bis(10-methoxycar-
bonyldecyloxy)phenyl)porphyrin (8c). Aldehyde 7c (1.78 g, 3 mmol)
and pyrrole (208 µL, 3 mmol) were dissolved in CH2Cl2 (300 mL)
under N2 and then BF3‚OEt2 (125 µL, 0.99 mmol) was added. After
the reaction mixture was stirred at room temperature for 100 min, 2,3-
dichloro-5,6-dicyanobenzoquinone (510 mg, 2.2 mmol) was added and
the mixture was refluxed for 2 h. The solution was then neutralized by
addition of triethylamine (138 µL, 1 mmol) and evaporated. The mixture
was separated by column chromatography (SiO2, CHCl3/AcOEt ) 100/
1-10/1) and crude product was washed with methanol thoroughly to
afford a purple solid of 8c (296 mg, yield 16%): 1H NMR (CDCl3) δ
-2.64 (s, 2H, NH), 0.61 (quintet, J ) 7.5 Hz, 16H, CH2), 0.76 (m,
32H, CH2), 0.88 (quintet, J ) 7.5 Hz, 16H, CH2), 0.94 (quintet, J )
7.5 Hz, 16H, CH2), 0.97 (quintet, J ) 7.5 Hz, 16H, CH2), 1.06 (quintet,
J ) 7.5 Hz, 16H, CH2), 1.45 (quintet, J ) 7.5 Hz, 16H, CH2), 2.17 (t,
J ) 7.5 Hz, 16H, CH2), 3.60 (s, 24H, CO2Me), 3.82 (t, J ) 6.5 Hz,
16H, CH2), 4.07 (s, 12H, CO2Me), 7.62 (s, 8H, phenyl-H), 8.57 (s,
8H, â-pyrrole); UV-vis (CH2Cl2) λmax (log ꢀ) 422 (5.61), 515 (4.36),
549 (3.93), 584 (4.01); HRMS (FAB) calcd for C148H214N4O32 (M+)
2559.5239, found 2559.6562. Anal. Calcd for C148H214N4O32: C, 69.40;
H, 8.42; N, 2.19. Found: C, 69.28; H, 8.32; N, 2.15.
[5,10,15,20-Tetrakis(4-carboxy-3,5-bis(10-carboxydecyloxy)phe-
nyl)porphyrinato]zinc(II) Potassium Salt (3). Zinc porphyrin 9c (48
mg, 19 µmol) was dissolved in a solution prepared by mixing THF
(36 mL), methanol (15 mL), and 0.5 M KOH (16 mL). After being
stirred at room temperature for 24 h, the solution was concentrated
and passed through Sephadex G-15 followed by lyophilization to afford
3 (43 mg, yield 79%): 1H NMR (deuterated Borax buffer, pD 8.6, I )
0.1 M) δ 0.38 (quintet, J ) 7.5 Hz, 16H, CH2), 0.51 (quintet, J ) 7.5
Hz, 16H, CH2), 0.62 (quintet, J ) 7.5 Hz, 16H, CH2), 0.71 (quintet, J
) 7.5 Hz, 16H, CH2), 0.86 (quintet, J ) 7.5 Hz, 16H, CH2), 1.00
(quintet, J ) 7.5 Hz, 16H, CH2), 1.08 (quintet, J ) 7.5 Hz, 16H, CH2),
1.40 (quintet, J ) 7.5 Hz, 16H, CH2), 2.10 (t, J ) 7.5 Hz, 16H, CH2),
3.93 (t, J ) 7.0 Hz, 16H, CH2), 7.76 (s, 8H, phenyl-H), 8.84 (s, 8H,
â-pyrrole); UV-vis (Borax buffer, pH 9.0, I ) 0.1 M) λmax (log ꢀ)
426 (5.37), 555 (3.98), 593 (2.75).
Acknowledgment. We thank K. Akiyoshi for his help in
dynamic light scattering studies. The kind help of T. Kobatake
in mass spectroscopic studies is also acknowledged. This work
was supported by a Grant-in Aid for Scientific Research on
Priority Areas No. 11132233 from the Ministry of Education,
Science, Sports and Culture, Japan.
[5,10,15,20-Tetrakis(4-methoxycarbonyl-3,5-bis(methoxycarbo-
nylmethoxy)phenyl)porphyrinato]zinc(II) (9a). This compound was
prepared from 8a (100 mg, 65 µmol) in a manner similar to that
1
described for 9c: yield 84% (87 mg); H NMR (CDCl3) δ 3.38 (s,
JA9922126