Molecular recognition of aqueous dipeptides at multiple hydrogen-bonding sites of mixed peptide monolayers

Article abstract of DOI:10.1021/ja961526f

Oligopeptide amphiphiles with different dipeptide moieties of -XYNH₂ (X = Gly and Ala, Y = GLy, Ala, Val, Leu, and Phe) were synthesized. Binding of aqueous dipeptides onto monolayers of equimolar mixtures of these amphiphiles with a benzoic ac

Full text of DOI:10.1021/ja961526f

J. Am. Chem. Soc. 1996, 118, 9545-9551
9545
Molecular Recognition of Aqueous Dipeptides at Multiple
Hydrogen-Bonding Sites of Mixed Peptide Monolayers
Xiao Cha,† Katsuhiko Ariga, and Toyoki Kunitake*,‡
Contribution from the Supermolecules Project, JRDC, 2432 Aikawa, Kurume Research Park,
Kurume, 839 Japan
ReceiVed May 7, 1996X
Abstract: Oligopeptide amphiphiles with different dipeptide moieties of -XYNH2 (X ) Gly and Ala, Y ) Gly,
Ala, Val, Leu, and Phe) were synthesized. Binding of aqueous dipeptides onto monolayers of equimolar mixtures
of these amphiphiles with a benzoic acid amphiphile (2C18BCOOH) was investigated by π-A isotherm measurement,
FT-IR spectroscopy, and XPS elemental analysis. For given GlyX dipeptides (X ) neutral and hydrophobic residues),
the binding ratio was lessened with increasing sizes of the side chain of the Y residue in the GlyY dipeptide moiety
of the host amphiphiles. The Langmuir-type saturation behavior was observed for binding of GlyLeu to an equimolar
monolayer of 2C18BGly2NH2 and 2C18BCOOH. Its binding constant of 475 M-1 was 10 times larger than that
observed for a single-component monolayer of 2C18BGly2NH2 (K ) 35 M-1). The saturation guest/host ratio was
0.47. The mode of substrate insertion into the binding site was examined by FT-IR spectroscopy. When the
hydrophobic residue was on the C-terminal of a guest dipeptide (GlyX), the C-terminal insertion was selected with
accompanying formation of cyclic carboxylic acid dimers at the interface. In the case of XGly guests, the N-terminal
insertion with salt bridge formation with the host was observed. When the two residues of a dipeptide had close
hydrophobicities, both C- and N-terminal insertions were observed. Formation of these binding sites is apparently
induced by dipeptide binding.
Introduction
Scheme 1
Molecular recognition between signal peptides and receptor
proteins is a basic feature of many biological processes.1-3 These
receptors are usually located on the biomembrane surface. How
to mimic these processes and design artificial peptide receptors
has intrigued many chemists,4-6 because of their practical
applications in addition to their use as a tool to study natural
receptor processes.
It has been reported that an ordered array of functional groups
formed at the interface controls binding of amino acids and
subsequent crystal growth.7 More recently, Higashi et al.
reported enantioselective binding of R-amino acids by a poly-
(L-glutamic acid)-functionalized monolayer.8 These examples
suggest that functional arrays formed at the air-water interface
are useful for selective peptide binding. In order to develop
peptide receptors at the artificial interface, we have been
investigating specific binding of aqueous dipeptides onto
† Permanent address: Department of Biology, Sichuan Tumor Institute
and Hospital Southern Train Station, Chengdu 610041, The People’s
Republic of China.
‡ Permanent address: Faculty of Engineering, Kyushu University,
Fukuoka 812, Japan.
X Abstract published in AdVance ACS Abstracts, September 15, 1996.
(1) See, e.g.: Germain, R. N.; Margulies, D. H. Annu. ReV. Immunol.
1993, 11, 403.
(2) Stern, L. J.; Brown, J. H.; Jardetzky, T. S.; Gorga, J. C.; Urban, R.
G.; Strominger, J. L.; Wiley, D. C. Nature 1994, 368, 215.
(3) Tame, J. R. H.; Murshudov, G. N.; Dodson, E. J.; Neil, T. K.; Dodson,
G. G.; Higgins, C. F.; Wilkinson, A. J. Science 1994, 264, 1578.
(4) (a) Borchart, A.; Still, W. C. J. Am. Chem. Soc. 1994, 116, 373. (b)
Torneiro, M.; Still, W. C. J. Am. Chem. Soc. 1995, 117, 5887.
(5) Albert, J. S.; Goodman, M. S.; Hamilton, A. D. J. Am. Chem. Soc.
1995, 117, 1143.
(6) Labrenz, S. R.; Kelly, J. W. J. Am. Chem. Soc. 1995, 117, 1655.
(7) (a) Gidalevitz, D.; Weissbuch, I.; Kjaer, K.; Als-Nielson, J.; Leis-
erowitz, J. Am. Chem. Soc. 1994, 116, 3271. (b) Landau, E. M.; Grayer
Wolf, S.; Levanon, M.; Leiserowitz, L.; Lahav, M.; Sagiv, J. J. Am. Chem.
Soc. 1989, 111, 1436.
peptide-functionalized monolayers. Peptide binding was not
detectable in the case of monolayers of single-chain derivatives
of oligoglycines.9 In this case, strong inter-peptide hydrogen
bonding prevented binding of guest peptides. Very recently,
we have shown that aqueous dipeptides such as GlyX can be
selectively bound to the monolayer of an amphiphile in which
the dioctadecylamine moiety was connected with the glycyl-
glycinamide head group via the terephthaloyl unit (2C18BGly2-
NH2; see Scheme 1).10 In this system, selective binding of
dipeptides was promoted by antiparallel hydrogen bonding and
(9) Cha, X.; Ariga, K.; Kunitake, T. Bull. Chem. Soc. Jpn. 1996, 69,
163.
(8) Higashi, N.; Saitou, M.; Mihara, T.; Niwa, M. J. Chem. Soc., Chem.
Commun. 1995, 2119.
(10) Cha, X.; Ariga, K.; Onda, M.; Kunitake, T. J. Am. Chem. Soc. 1995,
117, 11833.
S0002-7863(96)01526-0 CCC: $12.00 © 1996 American Chemical Society

9546 J. Am. Chem. Soc., Vol. 118, No. 40, 1996
Cha et al.
2C18BGlyAlaNH2. 2C18BGlyOH (0.157 g, 0.228 mmol) was
dissolved in CH2Cl2 (100 mL), and DEPC (0.070 mL, 0.46 mmol) was
added at 0 °C. After stirring for 15 min, L-alaninamide-HBr salt
(HBr-AlaNH2; 0.0533 g, 0.315 mmol) and triethylamine (0.150 mL,
1.08 mmol) dissolved in CH2Cl2 (50 mL) were added at 0 °C. The
mixture was allowed to react at room temperature for 45 h followed
by removal of the solvent under reduced pressure. The residue was
chromatographed on SiO2 (1:1 CHCl3/acetone) to give 2C18BGlyAlaNH2
as a white solid (0.081 g, 44.8%): mp 135.5-136.0 °C; TLC Rf 0.2
hydrophobic interaction between host peptide units and guest
dipeptides. However, the binding of XGly or XX′ dipeptides
(X, X′ ) neutral amino acid residues other than Gly) onto the
2C18BGly2NH2 monolayer is not detectable, probably because
of steric crowding (for XX′ dipeptides) or weak parallel
hydrogen bonding (for XGly dipeptides).10 We need to develop
new monolayer systems in order to prepare binding sites that
are selective for these dipeptides. The nature of the binding
cavity would be readily modified by using mixed monolayers.
Short acidic or basic groups should be appropriate for forming
new cavities upon mixing with oligopeptide polar groups, as
we indicated in a preliminary publication.11
In this paper we present a full account of dipeptide recognition
by mixed monolayers of oligopeptide amphiphiles (2C18-
BXYNH2) and benzoic acid amphiphiles (2C18BCOOH). The
details of binding selectivity and the structure of the binding
site are elucidated with the help of XPS analysis and FT-IR
spectroscopy. The recognition site is self-assembled on the
surface of the monolayer via the interaction with guest dipep-
tides. This is analogous to induced-fit phenomena at the active
site of enzymes.
1
(1:1 CHCl3/acetone); H NMR (CDCl3) δ 0.88 (t, J ) 6.6 Hz, 6H, 2
CH3), 1.25 (br, 60H, 30 CH2), 1.37 (d, J ) 7.0 Hz, 3H, alanine CH3),
1.47-1.65 (br, 4H, 2 CH2CH2N), 3.13 (br, 2H, CH2N), 3.47 (br, 2H,
CH2N), 4.08 (m, 2H, glycine CH2), 4.35 (m, 1H, alanine R-CH), 5.67
(br, 1H, amide), 6.60 (br, 1H, amide), 7.00 (d, J ) 6.9 Hz, 1H, amide),
7.38 (d, J ) 8.0 Hz, 2H, aromatic), 7.62 (br, 1H, amide), 7.84 (d, J )
8.1 Hz, 2H, aromatic). Anal. Calcd for C49H88N4O4‚1/3H2O: C, 73.27;
H, 11.13; N, 6.97. Found: C, 73.24; H, 11.06; N, 6.92.
2C18BGlyValNH2. 2C18BGlyOH (0.099 g, 0.14 mmol) was dis-
solved in CH2Cl2 (100 mL), and DEPC (0.035 mL, 0.23 mmol) was
added at 0 °C. After stirring for 20 min, L-valinamide-HCl salt (HCl-
ValNH2; 0.0255 g, 0.167 mmol) and triethylamine (0.060 mL, 0.43
mmol) dissolved in CH2Cl2 (30 mL) were added at 0 °C. The mixture
was allowed to react at room temperature for 77 h followed by removal
of the solvent under reduced pressure. The residue was chromato-
graphed on SiO2 (1:1 CHCl3/acetone) to give 2C18BGlyValNH2 as a
white solid (0.085 g, 75.9%): mp 206.0-207.0 °C; TLC Rf 0.3 (1:1
CHCl3/acetone); 1H NMR (CDCl3) δ 0.88 (t, J ) 6.6 Hz, 6H, 2 CH3),
0.94 (d, J ) 6.9 Hz, 3H, valine CH3), 0.97 (d, J ) 7.0 Hz, 3H, valine
CH3), 1.25 (br, 60H, 30 CH2), 1.47-1.65 (br, 4H, 2 CH2CH2N), 2.18
(m, 1H, valine CH of the side chain), 3.13 (br, 2H, CH2N), 3.47 (br,
2H, CH2N), 4.14 (m, 2H, glycine CH2), 4.30 (m, 1H, valine R-CH),
5.66 (br, 1H, amide), 6.30 (br, 1H, amide), 6.85 (d, J ) 8.6 Hz, 1H,
amide), 7.40 (d, J ) 8.1 Hz, 2H, aromatic), 7.41 (br, 1H, amide), 7.84
(d, J ) 8.2 Hz, 2H, aromatic). Anal. Calcd for C51H92N4O4‚1/2H2O:
C, 73.42; H, 11.24; N, 6.72. Found: C, 73.42; H, 11.20; N, 6.61.
2C18BGlyLeuNH2. 2C18BGlyOH (0.152 g, 0.209 mmol) was
dissolved in CH2Cl2 (150 mL), and DEPC (0.050 mL, 0.33 mmol) was
added at 0 °C. After stirring for 15 min, L-leucinamide-HCl salt
(HCl-LeuNH2; 0.0421 g, 0.253 mmol) and triethylamine (0.100 mL,
0.717 mmol) dissolved in CH2Cl2 (50 mL) were added at 0 °C. The
mixture was allowed to react at room temperature for 44 h followed
by removal of the solvent under reduced pressure. The residue was
chromatographed on SiO2 (3:1 and 1:1 CHCl3/acetone) to give 2C18-
BGlyLeuNH2 as a white solid (0.133 g, 76.0%): mp 42.5-43.5 °C;
TLC Rf 0.2 (3:1 CHCl3/acetone); 1H NMR (CDCl3) δ 0.88 (t, J ) 6.6
Hz, 6H, 2 CH3), 0.93 (d, J ) 6.0 Hz, 3H, leucine CH3), 0.94 (d, J )
5.8 Hz, 3H, leucine CH3), 1.25 (br, 60H, 30 CH2), 1.47-1.74 (br, 7H,
2 CH2CH2N + CH and CH2 in the leucine side chain), 3.13 (t, J ) 8.0
Hz, 2H, CH2N), 3.49 (t, J ) 8.0 Hz, 2H, CH2N), 4.11 (m, 2H, glycine
CH2), 4.47 (m, 1H, leucine R-CH), 5.51 (br, 1H, amide), 6.31 (br, 1H,
amide), 6.68 (d, J ) 7.7 Hz, 1H, amide), 7.28 (br, 1H, amide), 7.40
(d, J ) 8.0 Hz, 2H, aromatic), 7.82 (d, J ) 8.1 Hz, 2H, aromatic).
Anal. Calcd for C52H94N4O4‚1/2H2O: C, 73.62; H, 11.29; N, 6.60.
Found: C, 73.69; H, 11.21; N, 6.54.
2C18BGlyPheNH2. 2C18BGlyOH (0.204 g, 0.280 mmol) was
dissolved in CH2Cl2 (150 mL), and DEPC (0.065 mL, 0.43 mmol) was
added at 0 °C. After stirring for 15 min, L-phenylalaninamide-HCl
salt (HCl-PheNH2; 0.0627 g, 0.312 mmol) and triethylamine (0.130
mL, 0.932 mmol) dissolved in CH2Cl2 (50 mL) were added at 0 °C.
The mixture was allowed to react at room temperature for 45 h followed
by removal of the solvent under reduced pressure. The residue was
chromatographed on SiO2 (3:1 and 1:1 CHCl3/acetone) to give 2C18-
BGlyPheNH2 as a white solid (0.202 g, 84.0%): mp 91.8-92.3 °C;
TLC Rf 0.2 (3:1 CHCl3/acetone); 1H NMR (CDCl3) δ 0.88 (t, J ) 6.6
Hz, 6H, 2 CH3), 1.26 (br, 60H, 30 CH2), 1.43-1.53 (br, 4H, 2 CH2-
CH2N), 3.16 (br, 4H, 2 CH2N), 3.48 (t, J ) 6.2 Hz, 2H, Ar-CH2 in
phenylalanine), 4.05 (d, J ) 4.8 Hz, 2H, glycine CH2), 4.67 (m, 1H,
R-CH in phenylalanine), 5.46 (br, 1H, amide), 6.03 (br, 1H, amide),
6.75 (br, 1H, amide), 7.13 (br, 1H, amide), 7.23 (m, 5H, aromatic in
phenylalanine), 7.40 (d, J ) 8.1 Hz, 2H, COPhCO), 7.79 (d, J ) 8.4
Hz, 2H, COPhCO). Anal. Calcd for C55H92N4O4‚1/2H2O: C, 74.87;
H, 10.62; N, 6.35. Found: C, 74.96; H, 10.66; N, 6.08.
Experimental Section
Synthesis of Amphiphiles. Amphiphiles 2C18BGlyAlaNH2, 2C18-
BGlyValNH2, 2C18BGlyLeuNH2, 2C18BGlyPheNH2, 2C18BAlaGlyNH2,
and 2C18BAla2NH2 were synthesized by the pathway given in Scheme
1. Syntheses of 2C18BGly2NH2,10 dioctadecylamine,12 and N,N-
dioctadecylterephthalamic acid (2C18BCOOH)10 are described else-
where. The other chemicals were commercially available. Melting
points were recorded on a Yanaco micro melting point apparatus and
1
uncorrected. Chemical shifts of H NMR spectra were recorded on a
Bruker ARX-300 (300 MHz) spectrometer and are given relative to
chloroform (δ 7.26) or tetramethylsilane (δ 0.00). Elemental analyses
(C, H, and N) were performed at the Faculty of Science, Kyushu
University.
2C18BGlyOBn. N,N-Dioctadecylterephthalamic acid (2C18BCOOH;
2.15 g, 3.21 mmol) was dissolved in CH2Cl2 (150 mL), and diethyl
phosphorocyanidate (DEPC; 0.850 mL, 5.60 mmol) was added to the
solution at 0 °C. After stirring for 20 min, glycine benzyl ester-p-
toluenesulfonic acid salt (TsOH-GlyOBn; 1.31 g, 3.87 mmol) and
triethylamine (1.50 mL, 10.8 mmol) dissolved in CH2Cl2 (100 mL)
were added at 0 °C. The mixture was allowed to react at room
temperature for 64.5 h followed by removal of the solvent under reduced
pressure. The residue was chromatographed on SiO2 (3:1 and 1:1
n-hexane/EtOAc) to give 2C18BGlyOBn as a white solid (2.47 g,
94.1%): mp 59.0-59.5 °C; TLC Rf 0.2 (3:1 n-hexane/EtOAc); 1H NMR
(CDCl3) δ 0.88 (t, J ) 6.6 Hz, 6H, 2 CH3), 1.26 (br, 60H, 30 CH2),
1.46-1.65 (br, 4H, 2 CH2CH2N), 3.12 (t, J ) 7.7 Hz, 2H, CH2N),
3.47 (t, J ) 7.4 Hz, 2H, CH2N), 4.30 (d, J ) 5.0 Hz, 2H, glycine
CH2), 5.24 (s, 2H, COOCH2), 6.72 (br, 1H, amide), 7.35 (s, 5H,
COOCH2Ph), 7.41 (d, J ) 8.3 Hz, 2H, COPhCO), 7.82 (d, J ) 8.2
Hz, 2H, COPhCO). Anal. Calcd for C53H88N2O4: C, 77.89; H, 10.85;
N, 3.43. Found: C, 77.90; H, 10.85; N, 3.46.
2C18BGlyOH. Pd/C (Pd 5%, 0.259 g) and 2C18BGlyOBn (2.42 g,
2.74 mmol) were dispersed in THF (20 mL) and ethanol (20 mL). The
reaction mixture was kept under a H2 gas atmosphere at room
temperature for 7 h. After filtration, the solvents were removed in
vacuo. The title compound was obtained as a white solid (1.82 g,
1
91.7%): mp 111.0-111.5 °C; H NMR (CDCl3) δ 0.88 (t, J ) 6.6
Hz, 6H, 2 CH3), 1.25 (br, 60H, 30 CH2), 1.48-1.66 (br, 4H, 2 CH2-
CH2N), 3.14 (t, J ) 7.2 Hz, 2H, CH2N), 3.48 (t, J ) 7.2 Hz, 2H, CH2N),
4.11 (d, J ) 4.5 Hz, 2H, glycine CH2), 7.16 (br, 1H, amide), 7.38 (d,
J ) 8.2 Hz, 2H, aromatic), 7.81 (d, J ) 8.1 Hz, 2H, aromatic). Anal.
Calcd for C46H82N2O4: C, 75.98; H, 11.37; N, 3.85. Found: C, 75.80;
H, 11.29; N, 3.81.
(11) Cha, X.; Ariga, K.; Kunitake, T. Chem. Lett. 1996, 73.
(12) Onda, M.; Yoshihara, K.; Koyano, H.; Ariga, K.; Kunitake, T. J.
Am. Chem. Soc. in press.

Molecular Recognition of Aqueous Dipeptides
J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9547
2C18BAlaOBn. 2C18BCOOH (0.457 g, 0.682 mmol) was dissolved
in CH2Cl2 (100 mL), and DEPC (0.150 mL, 0.989 mmol) was added
to the solution at 0 °C. After stirring for 20 min, L-alanine benzyl
ester-p-toluenesulfonic acid salt (TsOH-AlaOBn; 0.283 g, 0.805
mmol) and triethylamine (0.300 mL, 2.15 mmol) dissolved in CH2Cl2
(100 mL) were added at 0 °C. The mixture was allowed to react at
room temperature for 48 h followed by removal of the solvent under
reduced pressure. The residue was chromatographed on SiO2 (3:1, 1:1,
and 1:3 n-hexane/EtOAc) to give 2C18BAlaOBn as a white solid (0.464
1
g, 81.8%): mp 59.0-59.5 °C; TLC Rf 0.2 (3:1 n-hexane/EtOAc); H
NMR (CDCl3) δ 0.88 (t, J ) 6.6 Hz, 6H, 2 CH3), 1.25 (br, 60H, 30
CH2), 1.47-1.62 (br, 4H, 2 CH2CH2N), 1.54 (d, J ) 7.1 Hz, 3H, alanine
CH3), 3.12 (t, J ) 7.2 Hz, 2H, CH2N), 3.47 (t, J ) 7.2 Hz, 2H, CH2N),
4.85 (m, 1H, alanine, R-CH), 5.23 (s, 2H, COOCH2), 6.76 (d, J ) 7.2
Hz, 1H, amide), 7.34 (s, 5H, COOCH2Ph), 7.41 (d, J ) 8.2 Hz, 2H,
COPhCO), 7.82 (d, J ) 8.3 Hz, 2H, COPhCO). Anal. Calcd for
C54H90N2O4: C, 72.08; H, 10.91; N, 3.37. Found: C, 77.88; H, 10.86;
N, 3.34.
Figure 1. π-A isotherms of 2C18BGlyYNH2 monolayers at 20.0 (
0.3 °C on pure water: 1, 2C18BGly2NH2; 2, 2C18BGlyAlaNH2; 3, 2C18-
BGlyValNH2; 4, 2C18BGlyLeuNH2; 5, 2C18BGlyPheNH2.
2C18BAlaOH. Pd/C (Pd 5%, 0.057 g) and 2C18BAlaOBn (0.508 g,
0.611 mmol) were dispersed in ethanol (10 mL) and THF (10 mL).
The mixture was allowed to react under H2 gas at room temperature
for 6 h. After filtration, the solvents were removed in vacuo. The
title compound was obtained as a white solid (0.351 g, 77.5%): mp
deionized and doubly distilled. The spreading solutions of oligopeptide
amphiphiles were ca. 0.16 mg‚cm-3 in CHCl3. LB films were prepared
by using the vertical dipping method at up-stroke and down-stroke
motions of 8 and 100 mm‚min-1, respectively, from pure water and
dipeptide subphases. Monolayers were transferred onto gold-deposited
1
54.0-55.0 °C; H NMR (CDCl3) δ 0.88 (t, J ) 6.5 Hz, 6H, 2 CH3),
1.26 (br, 60H, 30 CH2), 1.43-1.66 (br, 4H, 2 CH2CH2N), 1.54 (d, J )
7.0 Hz, 3H, alanine CH3), 3.14 (br, 2H, CH2N), 3.48 (br, 2H, CH2N),
4.67 (m, 1H, alanine R-CH), 6.84 (d, J ) 5.4 Hz, 1H, amide), 7.40 (d,
J ) 8.1 Hz, 2H, aromatic), 7.80 (d, J ) 8.1 Hz, 2H, aromatic). Anal.
Calcd for C47H84N2O4: C, 76.16; H, 11.42; N, 3.78. Found: C, 76.06;
H, 11.24; N, 3.68.
glass slides at a surface pressure of 25 mN‚m-1
.
FT-IR Measurements. Infrared spectra of the LB film on a gold-
deposited glass were obtained on an FT-IR spectrometer (Nicolet 710)
equipped with a MCT detector (for RAS, reflection absorption
spectroscopy). All data were collected by the RAS method at a spectral
2C18BAlaGlyNH2. 2C18BAlaOH (0.146 g, 0.197 mmol) was
dissolved in CH2Cl2 (150 mL), and DEPC (0.050 mL, 0.33 mmol) was
added at 0 °C. After stirring for 15 min, glycinamide-HCl salt (HCl-
GlyNH2; 0.0272 g, 0.246 mmol) and triethylamine (0.100 mL, 0.717
mmol) dissolved in dry DMF (10 mL) were added at 0 °C. The mixture
was allowed to react at room temperature for 163 h. The organic layer
was washed with water and dried over Na2SO4 followed by removal
of the solvent under reduced pressure. The residue was chromato-
graphed on SiO2 (1:1 CHCl3/acetone) to give 2C18BAlaGlyNH2 as a
white solid (0.073 g, 46.5%): mp 103.5-104.5 °C; TLC Rf 0.3 (1:1
CHCl3/acetone); 1H NMR (CDCl3) δ 0.88 (t, J ) 6.6 Hz, 6H, 2 CH3),
1.26 (br, 60H, 30 CH2), 1.47-1.60 (br, 4H, 2 CH2CH2N), 1.52 (d, J )
7.1 Hz, 3H, alanine CH3), 3.13 (br, 2H, CH2N), 3.47 (br, 2H, CH2N),
3.70-3.92 (m, 2H, glycine CH2), 4.59 (m, 1H, alanine R-CH), 5.53
(br, 1H, amide), 6.61 (br, 1H, amide), 6.92 (br, 1H, amide), 7.23 (br,
1H, amide), 7.38 (d, J ) 8.2 Hz, 2H, aromatic), 7.81 (d, J ) 8.1 Hz,
2H, aromatic). Anal. Calcd for C49H88N4O4‚1/2H2O: C, 73.00; H,
11.13; N, 6.95. Found: C, 72.83; H, 10.99; N, 6.96.
2C18BAla2NH2. 2C18BAlaOH (0.102 g, 0.138 mmol) was dissolved
in CH2Cl2 (150 mL), and DEPC (0.030 mL, 0.20 mmol) was added at
0 °C. After stirring for 15 min, L-alaninamide-HBr salt (HBr-
AlaNH2; 0.0302 g, 0.179 mmol) and triethylamine (0.060 mL, 0.43
mmol) dissolved in CH2Cl2 (30 mL) were added at 0 °C. The mixture
was allowed to react at room temperature for 163 h. The organic layer
was washed with water and dried over Na2SO4 followed by removal
of the solvent under reduced pressure. The residue was chromato-
graphed on SiO2 (EtOAc) to give 2C18BAla2NH2 as a white solid (0.065
g, 58.0%): mp 117.5-118.5 °C; TLC Rf 0.1 (EtOAc); 1H NMR
(CDCl3) δ 0.87 (t, J ) 6.6 Hz, 6H, 2 CH3), 1.25 (br, 60H, 30 CH2),
1.31 (d, J ) 7.0 Hz, 3H, alanine CH3), 1.50 (d, J ) 7.0 Hz, 3H, alanine
CH3), 1.51-1.64 (br, 4H, 2 CH2CH2N), 3.13 (t, J ) 7.1 Hz, 2H, CH2N),
3.46 (t, J ) 7.1 Hz, 2H, CH2N), 4.06 (m, 1H, alanine R-CH), 4.27 (br,
1H, amide), 4.67 (m, 1H, alanine R-CH), 5.49 (br, 1H, amide), 6.56
(br, 1H, amide), 6.88 (br, 1H, amide), 7.39 (d, J ) 8.1 Hz, 2H,
aromatic), 7.84 (d, J ) 8.1 Hz, 2H, aromatic). Anal. Calcd for
C50H90N4O4‚H2O: C, 72.42; H, 11.18; N, 6.76. Found: C, 72.34; H,
10.94; N, 6.59.
resolution of 4 cm-1
.
XPS Measurement. X-ray photoelectron spectra of the LB films
on a gold-deposited glass were measured with a Perkin-Elmer PHI 5300
ESCA instrument (X-ray source Mg KR, 300 W, scan range 0-1000
eV, takeoff angle 45°). The elemental composition was obtained by
dividing the observed peak area by intrinsic sensitivity factors of each
element.
Results and Discussion
Monolayer Behavior and Langmuir-Blodgett Transfer.
Monolayers of the peptide amphiphiles and mixed monolayers
of peptide/benzoic acid amphiphiles (1:1 mole ratio) give
analogous surface area-pressure behaviors on pure water. They
have expanded phases at low pressures with limiting areas of
ca. 0.52-0.55 nm2 and collapse pressures of 48-58 mN‚m-1
.
Figure 1 summarizes isotherms of single-component monolayers
on pure water. All the isotherms have similar molecular areas
at the condensed phase, but show different expansion behavior
at low pressures. It can be seen that introduction of the
hydrophobic side chain in the dipeptide moiety of a host
amphiphile leads to expansion of its π-A isotherm at low
pressures. The isotherm of 2C18BGlyPheNH2 has the highest
phase transition pressure, while 2C18BGly2NH2 forms a con-
densed phase at the lowest pressure. Side chains in these
peptide-lipids clearly affect their aggregation behavior. How-
ever, the similarity of their limiting areas strongly indicates that
the molecular packing at the condensed phase is independent
of steric hindrance caused by the side chains.
Figure 2 shows π-A isotherms of mixed monolayers of 2C18-
BGly2NH2/2C18BCOOH (Figure 2a) and 2C18BGlyValNH2/
2C18BCOOH (Figure 2b) at a 1:1 molar ratio on pure water
and on aqueous dipeptides (10 mM LeuGly and GlyLeu). The
mean molecular areas are used. The mean molecular area of
the 2C18BGly2NH2/2C18BCOOH monolayer at 25 mN‚m-1 on
pure water showed 12% positive deviation from that of the ideal
mixture calculated as a simple average of the separate single-
component monolayers. Therefore, the mixed monolayer is
neither ideally mixed nor phase separated. The interlipid
Surface Pressure-Area (π-A) Isotherms and Langmuir-
Blodgett (LB) Films. A computer-controlled film balance system FSD-
110 (trough size 100 × 200 mm, USI System, Japan) was used. π-A
isotherms were taken at a compression rate of 4 mm‚min-1 and a
subphase temperature of 20.0 ( 0.3 °C. The subphase water was

9548 J. Am. Chem. Soc., Vol. 118, No. 40, 1996
Cha et al.
Table 1. Binding Ratios of Dipeptides toward 1:1 Mixed and Single-Component Monolayer as Determined by XPSa
monolayer
Guest/Host (mol/mol)b
peptide component
second component
GlyLeu
LeuGly
0.45
AlaPhe
PheAla
0.42
LeuPhe
LeuLeu
GlyGly
0.00
2C18BGly2NH2
2C18BGlyAlaNH2
2C18BGlyValNH2
2C18BGlyLeuNH2
2C18BGlyPheNH2
2C18BAlaGlyNH2
2C18BAla2NH2
2C18BGly2NH2
2C18BCOOH
2C18BCOOH
2C18BCOOH
2C18BCOOH
2C18BCOOH
2C18BCOOH
2C18BCOOH
2C18BCOOH
0.41
0.36
0.34
0.14
0.00
0.38
0.41
0.33
0.17
0.60
0.27
0.00
0.33
0.09
0.05
0.45
0.00
0.00
0.35
0.39
0.00
0.17
0.00
0.00
0.00
a LB films of 14 layers were used. b The concentration of the aqueous guest was 0.01 M.
layers of 2C18BGly2NH2/2C18BCOOH in the down-stroke mode
on the subphases of 0.01 M GlyLeu, GlyPhe, LeuGly, and
PheGly, while the ratio was (0.2-0.4) ( 0.1 for pure water.
Generally, the transfer ratio in the down-stroke mode decreased
in the presence of dipeptides.
Selectivity in Dipeptide Binding. The binding behavior of
dipeptides to the monolayers is summarized in Table 1. The
host/guest ratios are given for a 0.01 M guest concentration.
Our previous studies established that the 2C18BGly2NH2 mono-
layer can bind aqueous dipeptides of GlyX type only.10 Other
dipeptide monolayers of 2C18BGlyYNH2 (Y ) Ala, Val, Leu,
and Phe) did not bind aqueous dipeptides efficiently. Clearly,
the amino acid residue larger than Gly in 2C18BGlyYNH2
amphiphiles cannot provide proper binding cavities in their
single-component monolayers. The 2C18BCOOH monolayer is
similarly not capable of efficient binding of aqueous dipeptides.
For example, the binding ratios of GlyLeu, LeuGly, LeuLeu,
and GlyGly toward the 2C18BCOOH monolayer are 0.17, 0.17,
0.00, and 0.00, respectively. In contrast, the equimolar mixture
of 2C18BGly2NH2/2C18BCOOH can bind both GlyX and XGly
dipeptides, and it can even bind XX′ dipeptides which have
large side chains on both of the two amino acid residues.
The binding efficiency depends on the combination of
monolayer components. As shown in Table 1, their binding is
lessened or is lost with increasing sizes of the side chain of the
dipeptide moiety in the host molecules. The binding ratio of
GlyLeu is 0.41 to the 2C18BGly2NH2/2C18BCOOH monolayer,
while GlyLeu is hardly bound to 2C18BGlyPheNH2/2C18-
BCOOH. Most of dipeptides except for small GlyGly show
higher affinities to 2C18BGly2NH2/2C18BCOOH monolayers
than to 2C18BGlyValNH2/2C18BCOOH. The aqueous dipeptide
binding is also lessened or lost with increasing sizes of the guest
dipeptides. For example, binding ratios of LeuPhe and LeuLeu
to the 2C18BGlyAlaNH2/2C18BCOOH monolayer are 0.09 and
0.00, respectively, while those of GlyLeu and AlaPhe are around
0.3. A similar tendency was observed in the case of the 2C18-
BGlyValNH2/2C18BCOOH monolayer. In contrast, slim GlyGly
shows a higher binding affinity to 2C18BGlyValNH2/2C18-
BCOOH than to the 2C18BGly2NH2/2C18BCOOH monolayer.
These facts imply that size matching between the guest molecule
and host cavity is required for effective binding.
Figure 2. (a) π-A isotherms of a 2C18BGly2NH2/2C18BCOOH mixed
monolayer (1:1 mole ratio) at 20.0 ( 0.3 °C: 1, on pure water; 2, on
0.01 M LeuGly; 3, on 0.01 M GlyLeu. (b) π-A isotherms of a 2C18-
BGlyValNH2/2C18BCOOH mixed monolayer (1:1 mole ratio) at 20.0
( 0.3 °C: 1, on pure water; 2, on 0.01 M LeuGly; 3, on 0.01 M GlyLeu.
hydrogen bonding in the 2C18BGly2NH2 monolayer is apparently
broken by mixing with the 2C18BCOOH molecule, which causes
the positive deviation in the molecular area.
Although the structural difference in peptide side chains does
not alter the limiting area, the presence of aqueous peptides
clearly affects molecular packing in the mixed monolayer even
at the condensed phase. This is indicated by the expansion of
the π-A isotherms in Figure 2. Interestingly, the expansion is
influenced by the type of aqueous dipeptides. The π-A
behavior of the two mixed monolayers were little affected by
10 mM LeuGly relative to those on pure water, but the isotherms
show expansion on 10 mM GlyLeu. Thus, the monolayer/
dipeptide interaction is specific to the dipeptide structure.
Monolayers were transferred onto a gold-deposited glass plate
at a surface pressure of 25 mN‚m-1. Single-component peptide
monolayers showed varied transfer behavior. The monolayers
of 2C18BGly2NH2, 2C18BGlyAlaNH2, and 2C18BAla2NH2 showed
Y type transfer behavior, while the monolayers of 2C18-
BGlyValNH2, 2C18BGlyLeuNH2, and 2C18BGlyPheNH2 were
transferred in the Z mode. Only the 2C18BAlaGlyNH2 mono-
layer showed unsuccessful transfer due to return of the
transferred monolayer to water in the down-stroke motion.
All the mixed monolayers were successfully transferred onto
a gold-deposited glass plate at 25 mN‚m-1. The transfer ratio
from pure water and from the aqueous dipeptide subphase was
1.0 ( 0.1 in the up-stroke mode in all cases, but the ratio was
varied in the down-stroke motion, depending on the hydropho-
bicity of the guest peptides and polar groups of the host
molecules. For example, no transfer was observed for mono-
The binding ratios of GlyLeu to monolayers of 2C18-
BGlyAlaNH2/2C18BCOOH, 2C18BAlaGlyNH2/2C18BCOOH, and
2C18BAla2NH2/2C18BCOOH are 0.36, 0.38, and 0.41, respec-
tively. These close ratios indicate that the nature of the binding
cavity is not significantly altered by the replacement of Ala and
Gly residues in the host molecules for the GlyLeu guest.
Saturation and Stoichiometry of Dipeptide Binding. The
binding behavior of GlyLeu to 2C18BGly2NH2/2C18BCOOH
monolayer was examined more closely to determine binding
parameters. The molar ratio of the bound guest per lipid is
plotted as a function of guest concentration in Figure 3. The

Molecular Recognition of Aqueous Dipeptides
J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9549
Figure 3. Binding curve of GlyLeu to an equimolar mixed monolayer
of 2C18BGly2NH2/2C18BCOOH. The inset represents Langmuir plots
of GlyLeu as described in eq 1.
Figure 4. FT-IR-RAS spectra (1200-1800 cm-1) of LB films (15
layers): a, a 2C18BCOOH monolayer transferred from pure water; b,
a 2C18BGlyValNH2/2C18BCOOH equimolar monolayer transferred from
pure water; c, a 2C18BGlyValNH2/2C18BCOOH equimolar monolayer
transferred from 0.03 M GlyLeu; d, a 2C18BGlyValNH2/2C18BCOOH
equimolar monolayer transferred from 0.03 M LeuGly.
plots show saturation behavior, indicating that the monolayer
provides a specific binding site. The plots were analyzed by
using the Langmuir isotherm:
[S]/y ) 1/(RK) + [S]/R
(1)
where y is the guest/host ratio, [S] is the guest concentration in
the subphase, R is the saturation binding ratio, and K is the
binding constant. The plots show binding saturation at more
than 10 mM aqueous guest, suggesting that the recognition is
site-specific. Curve fitting of the plots gives a site occupancy
(i.e., the guest/host ratio at saturation) of 0.47 and a binding
constant of 475 M-1. The R value observed indicates that one
GlyLeu molecule is bound to two monolayer molecules. This
behavior is different from the equimolar site occupancy observed
for the single-component monolayer of 2C18BGly2NH2 with the
same guest. The magnitude of K is also different. The binding
constant with the mixed monolayer is much larger than that
with the 2C18BGly2NH2 single-component monolayer (K ) 35
M-1).10
FT-IR Investigation of the Host Structure of Mixed
Monolayers. FT-IR spectra in the reflection-absorption mode
of LB films of monolayers of 2C18BCOOH and 2C18-
BGlyValNH2/2C18BCOOH, both transferred from pure water,
are shown in Figure 4 (1200-1800 cm-1 region) and Figure 5
(2450-2800 cm-1 region). The 2C18BCOOH film from pure
water (Figures 4a and 5a) shows strong peaks characteristic of
the hydrogen-bonded dimer of benzoic acid (1697 (νCdO
(dimeric COOH)), 2543 and 2666 cm-1 (νOH, (dimeric COOH))
and a very weak peak of non-hydrogen-bonded COOH (1721
cm-1, νCdO). It means that self-hydrogen-bonding is formed
in the 2C18BCOOH film. On the contrary, in the spectrum of
2C18BGlyValNH2/2C18BCOOH LB film (Figures 4b and 5b),
these characteristic peaks at 1697, 2543, and 2665 cm-1
Figure 5. FT-IR-RAS spectra (2450-2800 cm-1) of LB films (15
layers): a, a 2C18BCOOH monolayer transferred from pure water; b,
a 2C18BGlyValNH2/2C18BCOOH equimolar monolayer transferred from
pure water; c, a 2C18BGlyValNH2/2C18BCOOH equimolar monolayer
transferred from 0.03 M GlyLeu; d, a 2C18BGlyValNH2/2C18BCOOH
equimolar monolayer transferred from 0.03 M LeuGly.
lecular hydrogen bonding (dimeric COOH for the 2C18BCOOH
monolayer with νCdO at 1701 cm-1, and hydrogen-bonded
oligoglycine units for the 2C18BGly2NH2 monolayer with νNH
at 3309 cm-1). In the mixed monolayer, the νNH peak is shifted
to 3320 cm-1, and the COOH peak becomes a broadened
shoulder at 1700-1720 cm-1. These IR features indicate that
intermolecular hydrogen bonds among the individual compo-
nents are destroyed due to mixing of the two components. This
observation is consistent with the positive deviation in molecular
area upon mixing of the two components.
FT-IR Investigation of the Host-Guest Interaction. A
mixed monolayer of 2C18BGlyValNH2/2C18BCOOH shows
different IR characteristics when transferred from 0.03 M
aqueous GlyLeu (Figures 4c and 5c). We can see strong
characteristic peaks of the dimeric COOH at 1692, 2572, 2656,
and 2717 cm-1, and characteristic peaks of amide I at 1629
and amide II at 1559 cm-1 (from the guest peptide). Only a
disappear and the peak of monomeric COOH (1721 cm-1
)
becomes stronger. Therefore, the two components in the
monolayer are mixed well with each other without phase
separation. The microdomain formation of 2C18BCOOH, if any,
would produce self-association of COOH groups. The peptide
head group of amphiphile 2C18BGlyValNH2 probably hinders
formation of the COOH dimer even when the monolayer is
transferred in the Y mode. Other characteristic peaks, amide I
(1656-1678 cm-1, overlapped) and amide II (1545 cm-1) from
2C18BGlyValNH2, are also observed.
Similar spectral characteristics are found in the 2C18BGly2-
NH2/2C18BCOOH system (spectra not shown). FT-IR spectra
of the individual components show characteristics of intermo-

9550 J. Am. Chem. Soc., Vol. 118, No. 40, 1996
Cha et al.
The single-component monolayer of 2C18BGly2NH2 showed
a selective binding capability toward GlyX.10 Combination of
a bulky double alkyl chain and a smaller Gly2NH2 unit can
produce a suitable cavity for dipeptide binding at the interface.
Binding of GlyX with C-terminal insertion is promoted by
formation of antiparallel hydrogen bonds and incorporation of
the guest hydrophobic chains into the hydrophobic side of the
host monolayer (Figure 7b). Binding of XGly in N-terminal
insertion that can bring guest hydrophobic groups into the
hydrophobic region of monolayer receptors is not favored,
because this mode can only form less stable parallel hydrogen
bonds. This difference in the stability of hydrogen bonding
would result in selective binding of GlyX to the 2C18BGly2-
NH2 monolayer. These examples indicate the importance of
the mode of hydrogen bonding for the dipeptide binding.
Dipeptide binding was not observed in the case of single-
component oligopeptide monolayers of 2C18BGlyYNH2 (Y )
Ala, Val, Leu, and Phe). Since molecular areas of these
monolayers in the condensed phase are almost the same as that
of 2C18BGly2NH2 (Figure 1), lipids with a large side chain
cannot provide cavities large enough for insertion of guest
dipeptides. The nature of the binding cavity can be readily
modified by using mixed monolayers. The binding to mixed
monolayers of 2C18BXYNH2/2C18BCOOH was detected for
some dipeptides. The binding behavior depends on the size of
the side chains of the amino acid residues in both the dipeptide
guests and peptide monolayer hosts (see Table 1). Combination
of a large guest and a large host, for example, a LeuLeu guest
and a 2C18BGlyValNH2/2C18BCOOH host, and that of a small
guest and a small host, for example, a GlyGly guest and a 2C18-
BGly2NH2/2C18BCOOH host, did not produce effective binding.
In contrast, complementary combinations of large guest/small
host, e.g., LeuLeu and 2C18BGly2NH2/2C18BCOOH, and of
small guest/large host, e.g., GlyGly and 2C18BGlyValNH2/2C18-
BCOOH, showed significant binding. These facts imply that
size matching based on van der Waals contact between the
cavity and guest is essential for effective binding. A plausible
model of incorporation of GlyLeu into the receptor site of the
2C18BGlyValNH2/2C18BCOOH monolayer is depicted in Figure
7c. In this model, aqueous GlyLeu is bound to the monolayer
from its C-terminal by forming hydrogen-bonded dimeric
COOH with the host benzoic acid. The hydrophobic side chain
of GlyLeu faces the hydrophobic part of the monolayer.
Antiparallel hydrogen bonding is formed with surrounding
dipeptide moieties of the host molecules.
Figure 6. FT-IR-RAS spectrum of a 2C18BGly2NH2/2C18BCOOH
equimolar monolayer transferred from 0.03 M LeuLeu in the region
of 1000-3500 cm-1
.
small characteristic peak due to salt bridge formation between
+
host COO- and guest NH3 is detected in the region between
amide I and amide II. The free COOH peak is also not found
at 1721 cm-1. These results suggest that guest GlyLeu mainly
forms a COOH dimer at its C-terminal with the benzoic acid
group of the 2C18BCOOH amphiphile.
In the case of 2C18BGlyValNH2/2C18BCOOH transferred
from 0.03 M aqueous LeuGly, a characteristic peak for
monomeric COOH at 1721 cm-1 disappears, and peaks for
dimeric COOH are absent in the regions of 1680-1700 and
2500-2700 cm-1 (Figures 4d and 5d). These facts strongly
indicate that the interaction between the host monolayer and
guest LeuGly do not contain formation of the COOH dimer. In
contrast, a large overlapped shoulder occurs at around 1600
cm-1 between the amide I and amide II peaks of the host, which
usually represents existence of COO- and NH3+. Therefore,
the salt bridge may be formed between the carboxylate of the
host monolayer and the N-terminal NH3+ of the guest dipeptide
upon N-terminal guest insertion. In addition, hydrogen bonding
between amide groups is implied by the absence of the free
νNH peak (data not shown).
On the other hand, LeuLeu appears to be inserted into the
monolayer of 2C18BGly2NH2/2C18BCOOH at both of the C-
and N-terminals. As shown in Figure 6, a characteristic peak
of amide II from guest LeuLeu is found at 1532 cm-1. A peak
corresponding to formation of a salt bridge of COO- (from the
monolayer) and NH3+ (from LeuLeu) is observed at about 1600
cm-1. At the same time, a broad peak corresponding to dimeric
COOH is found at 2680-2703 cm-1, and suggests interaction
of the monomeric benzoic acid with the C-terminal of LeuLeu.
A strong peak at 1684 cm-1 may be attributed to overlapped
peaks of dimeric COOH (it is usually located at 1690-1705
cm-1) in the C-terminal insertion mode and of hydrogen-bonded
In contrast, aqueous LeuGly is bound to the monolayer of
2C18BGlyValNH2/2C18BCOOH from its N-terminal (Figure 7d).
FT-IR spectra indicated formation of a carboxylate/ammonium
salt bridge at about 1600 cm-1
. We suspected that XGly
dipeptides could not bind to the single-component monolayer
of 2C18BGly2NH2 because the anticipated parallel hydrogen
bonding is not sufficiently strong. However, the formation of
the salt bridge at the N-terminal of the guest can supply
additional host-guest interaction and probably compensates the
disadvantage of the parallel hydrogen bonding. The hydropho-
bic interaction that is expected in this model between the
hydrophobic side chain of LeuGly and the hydrophobic part of
the monolayer would also contribute to effective binding.
COOH with the host amide (it is usually at 1676-1650 cm-1
in the N-terminal insertion mode.
)
Binding Mechanism of Dipeptides. Schematic representa-
tions of the receptor sites are summarized in Figure 7. Key
factors for dipeptide binding are cavity size, mode of hydrogen
bonding, and disposition of the guest hydrophobic group.
Monolayers of single-chain derivatives of oligoglycines
formed strong inter-peptide hydrogen bonding among compo-
nent amphiphiles, and binding of guest peptides was not
detectable. Formation of the inter-peptide hydrogen bonds
probably destroys molecular space needed for peptide insertion,
and free amide groups are not available for guest binding (Figure
7a).9
Induction of a Recognition Site by Guest Binding. The
experimental results presented here have an important implica-
tion for the formation of receptor sites. Two monolayer
components, 2C18BGly2NH2 and 2C18BCOOH, are mixed well
on pure water. This is clear from the positive deviation in the
molecular area and the IR spectral data of the mixed monolayer.
Specific interaction of these two components is not supported

Molecular Recognition of Aqueous Dipeptides
J. Am. Chem. Soc., Vol. 118, No. 40, 1996 9551
Figure 7. Schematic representation of binding modes of aqueous peptides to peptide-functionalized monolayers: a, a monolayer of a single-chain
oligoglycine amphiphile; b, a monolayer of a double-chain oligoglycine amphiphile (2C18BGly2NH2) with GlyX by C-terminal insertion: c, a 1:1
mixed monolayer of 2C18BGlyValNH2/2C18BCOOH with GlyX by C-terminal insertion; d, a 1:1 mixed monolayer of 2C18BGlyValNH2/2C18-
BCOOH with XGly by N-terminal insertion. The dotted lines represent hydrogen bonds.
by the available data, and therefore, they must be mixed
randomly on pure water. In contrast, a specific 2:1 interaction
(two host and one guest molecules) between the host monolayer
of 2C18BGly2NH2/2C18BCOOH and the guest GlyLeu is ob-
served, as suggested by the Langmuir adsorption isotherm. The
guest binding must induce redistribution of monolayer compo-
nents so as to produce a specific binding site. Thus, the binding
site is created through the “induced-fit” mechanism. The
induced-fit concept was first proposed by Koshland: binding
of substrates to an enzyme active site causes conformational
changes that align the catalytic groups in their correct orienta-
tion.13 A similar situation appears to exist for the mixed
monolayer. This conclusion points to an interesting possibility.
Mixed monolayers with suitable lipid combinations would create
receptor sites appropriate for different guest molecules through
the induced-fit mechanism. This combinatorial recognition site
is crudely analogous to the hypervariable region of antibodies.14
We believe that the existence of flexible recognition sites is
characteristic of mixed monolayers. As we reported already,
three functional components in mixed monolayers can bind
specifically to one molecule of flavin adenine dinucleotide
(FAD) at the air-water interface.15-17 The three monolayer
components appear to be statistically mixed in the absence of
aqueous FAD, but are organized regularly via specific interac-
tions with FAD molecules.
JA961526F
(14) Capra, J. D. Sci. Am. 1977, 236, 50.
(15) Taguchi, K.; Ariga, K.; Kunitake, T. Chem. Lett. 1995, 701.
(16) Oishi, Y.; Torii, Y.; Kuramori, M.; Suehiro, K.; Ariga, K.; Taguchi,
K.; Kamino, A.; Kunitake, T. Chem. Lett. 1996, 411.
(17) Oishi, Y.; Torii, Y.; Kato, T.; Kuramori, M.; Suehiro, K.; Ariga,
K.; Taguchi, K.; Kamino, A.; Koyano, H.; Kunitake, T. Langmuir, submitted
for publication.
(13) Koshland, D. E., Jr.; Neme´thy, G.; Filmer, D. Biochemistry 1971,
5, 365.