Chemistry Letters 2001
689
In aqueous solution a nitrogenous axial ligand (L) such as
amino carboxylates can substitute the coordinated water on a
zinc porphyrin (ZnP) and this reaction is explained as follows:
racemic mixture, chiral recognition for the enantiomers of the
amino carboxylates was expected. Our binding data on amino
acids, however, do not show chiral recognition evidence; no
difference in K is observed between DL- and L-Trp. Contrary to
this, a slight but certain difference in K for 3 was obtained
between Gly–DL-Trp and Gly–L-Trp.10 This can be reasonably
explained in terms of the relative magnitude of the two
hydrophobic interactions produced by the indole group of the
amino carboxylates with the porphyrin plane and with the
phenyl group of the porphyrin. In the bound Trp the former
interaction must be strong to eclipse the latter interaction,
whereas in the bound Gly–Trp the relatively decreased former
interaction by elongating the separation of the indole group
from the porphyrin plane might lead to the appearance of the
chiral recognition.
where the N atom of L coordinates to the zinc ion.7 Since the
pKa values of the coordinated H2O of these zinc porphyrins
were estimated to be larger than 12 and the pKa values of amino
+
acids (–NH3 ) are smaller than 9.6, the binding experiments
were carried out at a pH value between them. Binding con-
stants (K = [ZnP·L]/[ZnP·H2O][L]) for the equilibrium were
determined spectrophotometrically by a usual method.8
References and Notes
1
T. Mizutani, T. Ema, T. Tomita, Y. Kuroda, and H. Ogoshi, J. Am.
Chem. Soc., 116, 4240 (1994); Y. Kuroda, Y. Kato, T. Higashioji, J.
Hasegawa, S. Kawanami, M. Takahashi, N. Shiraishi, K. Tanabe, and
H. Ogoshi, J. Am. Chem. Soc., 117, 10950 (1995).
2
3
K. Konishi, K. Yahara, H. Toshishige, T. Aida, and S. Inoue, J. Am.
Chem. Soc., 116, 1337 (1994).
E. Mikros, A. Gaudemer, and R. Pasternack, Inorg. Chim. Acta, 153,
199 (1988); C. Verchére-Bèaur, E. Mikros, M. Perrèe-Fauvet, and A.
Gaudemer, J. Inorg. Biochem., 40, 127 (1990).
4
5
6
T. Mizutani, K. Wada, and S. Kitagawa, J. Am. Chem. Soc., 121, 11425
(1999).
T. Mizutani, K. Wada, and S. Kitagawa, J. Org. Chem., 65, 6097
(2000).
Selected data: Compound 1 (chloride): 1H NMR (400 MHz, DMSO-
d6) δ 2.25 (s, 18H, N(CH3)3), 6.61 (d, 4H, CO–ph), 6.67 (t, 4H,
CO–ph), 6.95 (t, 2H, CO–ph), 7.66 (m, 4H, ph), 7.88 (m, 6H, ph),
7.99 (d, 2H, ph), 8.02 (d, 2H, ph), 8.33 (d, 2H, ph), 8.60 (d, 2H, pyr-
role), 8.63 (d, 2H, pyrrole), 8.72 (s, 2H, pyrrole), 8.73 (s, 2H, pyrrole),
8.98 (s, 4H, NHCO), 9.17 (s, 4H, NHCO); vis (H2O/NaHCO3–Na2CO3,
pH 10.4) λmax (log ε) 424 (5.44), 557 (4.18), 596 (3.61) nm; Anal.
Table 1 lists binding data of amino carboxylates and butyl-
amine to zinc porphyrins. It is appeared that amino carboxy-
lates bind more tightly to the zinc porphyrins than butylamine
and that the K values for Asp are greater than those for Gly,
suggesting that Coulomb interaction between an –N+(CH3)3
group of the porphyrins and the –COO– group(s) of the coordi-
nated carboxylates enhances the binding. This was supported
.
Calcd for C68H60N10O4Zn. Cl2 6H2O: C, 61.61; H, 5.47; N, 10.57%.
Found: C, 61.36; H, 5.38; N, 10.03%; MS m/z 1181 (M+), M+ calcd
for C68H60N10O4ZnCl, 1180. Compound 2 (chloride): 1H NMR (400
MHz, DMSO-d6) δ 0.07 (s, 18H, C(CH3) 3), 2.30 (s, 18H, N(CH3)3),
7.58 (t, 2H, ph), 7.61 (s, 2H, NHCO), 7.71 (t, 2H, ph), 7.81 (t, 2H, ph),
7.86 (t, 2H, ph), 7.97 (m, 6H, ph), 8.40 (d, 2H, ph), 8.58 (s, 4H, pyr-
role), 8.64 (s, 4H, pyrrole), 9.01 (s, 2H, NHCO); vis
(H2O/NaHCO3–Na2CO3, pH 10.4) λmax (log ε) 424 (5.65), 557 (4.25),
596 (3.63) nm; Anal. Calcd for C64H68N10O4Zn.Cl2 .5H2O: C, 60.71;
H, 6.02; N, 11.06%. Found: C, 60.69; H, 5.83; N, 11.01%; MS m/z
1141(M+), M+ calcd for C64H68N10O4ZnCl, 1140. Compound 3
(chloride): 1H NMR (400 MHz, DMSO-d6) δ 2.34 (s, 18H, N(CH3)3),
6.67 (m, 8H, CO–ph), 6.96 (m, 2H, CO–ph), 7.66 (m, 4H, ph), 7.88
(m, 6H, ph), 7.99 (m, 4H, ph), 8.32 (d, 2H, ph), 8.52 (s, 2H, pyrrole),
8.62 (d, 2H, pyrrole), 8.75 (d, 2H, pyrrole), 8.84 (s, 2H, pyrrole), 9.02
(s, 2H, NHCO), 9.29 (s, 2H, NHCO); vis (H2O/NaHCO3–Na2CO3,
pH 10.4) λmax (log ε) 424 (5.18), 557 (4.20), 596 (3.66) nm; Anal.
from examining the dependence of K on ionic strength I where
–
slope for plotting log K versus √I can correlate to the magnitude
of Coulomb interaction.4,9 The slope on the binding of butyl-
amine to 2 was found to be 0.09 while those of L-Trp and DL-
Asp were –1.26 and –1.63, respectively, indicating that
Coulomb interaction strengthens the binding.
On the binding of amino carboxylates to 5,10,15,20-tetra-
kis(N-methyl-4-pyridyl)porphyrinatozinc(II) in aqueous solu-
tion, the K values for Phe and Trp anions were increased in
terms of hydrophobic interactions between the hydrophobic
side chains of the carboxylates and the porphyrin plane.3 This
effect cannot be expected for the amino carboxylates with no
hydrophobic side chain such as Gly and Asp. For the zinc por-
phyrins prepared, similar increments in K are also observed for
the binding of Phe and Trp compared to Gly, suggesting the
presence of hydrophobic interaction with the porphyrin plane.
Another hydrophobic interaction is also possible between the
side chain of Phe or Trp and the t-butyl or phenyl group of the
zinc porphyrins prepared. This can be correlated to the fact that
the K values for 1 and 3 with the amines except Asp are appar-
ently larger than those for 2. This result might be reasonably
ascribed to the increased hydrophobic area of the phenyl group
compared to the t-butyl group.
.
Calcd for C68H60N10O4Zn.Cl2 3H2O.2CH3OH: C, 62.95; H, 5.58; N,
10.49%. Found: C, 63.23; H, 5.21; N, 10.06%; MS m/z 1181(M+), M+
calcd for C68H60N10O4ZnCl, 1180.
7
8
H. Imai, H. Munakata, A. Takahashi, S. Nakagawa, and Y. Uemori,
Chem. Lett., 1997, 819; H. Imai, H. Munakata, A. Takahashi, S.
Nakagawa, Y. Ihara, and Y. Uemori, J. Chem. Soc., Perkin Trans. 2,
1999, 2565.
A plot of [L] versus [L]/∆A gave a straight line with intercept –K–1,
where ∆A is the difference between the absorbance at 424 nm of the
solution of ZnP without L and that with a given concentration of free L;
J. P. Collman, J. I. Brauman, K. M. Doxsee, T. R. Halbert, S. E. Hayes,
and K. S. Suslick, J. Am. Chem. Soc., 100, 2761 (1978).
9
M. A. Hossain and H.-J. Schneider, Chem. Eur. J., 5, 1284 (1999).
10 The K values of 3 with chiral amines were estimated from the formation
of the Zn–N bond (reference 8) where the two diastereomeric conforma-
tions of the amine adducts can not be distinguishable. Therefore chiral
recognition of the amines by 3 results a decrease in K for L- (or D-) body
of the amines compared with the DL- racemic mixture in terms of a
decreased amount of a preferred diastereomeric conformation.
Thus, compounds 1 and 3 can recognize amino carboxy-
lates on the basis of coordination, Coulomb interaction, and
two-point hydrophobic interactions. Since 3 exists as the