Metallopeptide design: Tuning the metal cation affinities with unnatural amino acids and peptide secondary structure

Article abstract of DOI:10.1021/ja9619723

The ability to tune the metal binding affinity of small peptides through the incorporation of unnatural multidentate α-amino acids and the preorganization of peptide structure is illustrated. Herein, we describe the exploitation of a family of α-amino acids that incorporate powerful bidentate ligands (bipyridyl and phenanthrolyl groups) as integral constituents of the residues' side chains. The residues involved are the 6-, 5-, and 4-substituted (S)-2-amino-3-(2,2'-bipyridyl)propanoic acids (1, 6Bpa;2, 5Bpa; 3, 4Bpa), (S)-2-amino-3-(1,10-phenanthrol-2-yl)propanoic acid (4, Fen), and a novel neocuproine-containing α-amino acid, (S)-2-amino-3-(9-methyl-1,10-phenanthrol-2-yl)propanoic acid (5, Neo). Within this family of amino acids, variations in metal binding due to the nature of the ring system (2,2'-bipyridyl or 1,10-phenanthrolyl) and the point of attachment to the amino acid β-carbon are observed. Additionally, the underlying peptide architecture significantly influences binding for peptides that include multiple metal-ligating residues. These differences in affinity arise from the interplay of ligand type and structural preorganization afforded by the peptide sequence, resulting in dissociation constants ranging from 10^-3 to < 10^-6 M for Zn(II). These studies illustrate that significant control of metal cation binding affinity, preference, and stoichiometry may be achieved through the use of a wide variety of native and unnatural metal-coordinating amino acids incorporated into a polypeptide architecture.

Full text of DOI:10.1021/ja9619723

J. Am. Chem. Soc. 1996, 118, 11349-11356
11349
Metallopeptide Design: Tuning the Metal Cation Affinities with
Unnatural Amino Acids and Peptide Secondary Structure
Richard P. Cheng, Stewart L. Fisher, and Barbara Imperiali*
DiVision of Chemistry and Chemical Engineering, California Institute of Technology,
Pasadena, California 91125
ReceiVed June 11, 1996X
Abstract: The ability to tune the metal binding affinity of small peptides through the incorporation of unnatural
multidentate R-amino acids and the preorganization of peptide structure is illustrated. Herein, we describe the
exploitation of a family of R-amino acids that incorporate powerful bidentate ligands (bipyridyl and phenanthrolyl
groups) as integral constituents of the residues’ side chains. The residues involved are the 6-, 5-, and 4-substituted
(S)-2-amino-3-(2,2′-bipyridyl)propanoic acids (1, 6Bpa; 2, 5Bpa; 3, 4Bpa), (S)-2-amino-3-(1,10-phenanthrol-2-yl)-
propanoic acid (4, Fen), and a novel neocuproine-containing R-amino acid, (S)-2-amino-3-(9-methyl-1,10-phenanthrol-
2-yl)propanoic acid (5, Neo). Within this family of amino acids, variations in metal binding due to the nature of the
ring system (2,2′-bipyridyl or 1,10-phenanthrolyl) and the point of attachment to the amino acid â-carbon are observed.
Additionally, the underlying peptide architecture significantly influences binding for peptides that include multiple
metal-ligating residues. These differences in affinity arise from the interplay of ligand type and structural
preorganization afforded by the peptide sequence, resulting in dissociation constants ranging from 10-3 to <10-6
M
for ZnII. These studies illustrate that significant control of metal cation binding affinity, preference, and stoichiometry
may be achieved through the use of a wide variety of native and unnatural metal-coordinating amino acids incorporated
into a polypeptide architecture.
Introduction
ligating atom. Additionally, the chemical synthesis of small
peptides generally affords milligram quantities of material and
allows the incorporation of the unnatural amino acids at any
position in the primary sequence.
Recent efforts in the de noVo design of metalloproteins have
focused on the assembly of polypeptides with defined structural
and potentially functional properties.1,2 The production of
peptidyl motifs designed to detect ZnII in solution emphasizes
the importance of tuning the metal binding affinity of such
constructs.3 The ability to modulate metal binding affinity is
essential in order to operate such peptide-based sensors over a
wide metal cation concentration range. Furthermore, it is
necessary for the peptide to preferentially bind the metal cation
of interest to exploit such peptides in sensor applications. One
approach to the construction of these devices involves the use
of novel multidentate metal-binding residues.4-9 An advantage
to the utilization of such residues in small peptidyl constructs
is that enhanced metal binding affinities may be achieved.
Whereas nature exploits the protein architecture to carefully
position the metal-binding side chain functionalities within
metalloprotein systems, unnatural metal-chelating amino acids
simplify the task by simultaneously orienting more than one
The bidentate heteroaromatic ligands 2,2′-bipyridine and 1,-
10-phenanthroline are among the best known coordination
agents.10 These multidentate nitrogen ligands generally afford
high binding affinities for metal cations due to the chelation
effect and the π-accepting ability of these moieties.11 While
alkyl substitution at the 6,6′-positions of 2,2′-bipyridine and the
2,9-positions of 1,10-phenanthroline is expected to raise the
basicity of the donor nitrogens, substituted ligands generally
exhibit weaker metal binding affinities than the parent hetero-
cycle due to steric effects. Additionally, alkyl substitution can
affect the metal binding preference and complex stoichiometry.
R-Amino acids featuring these bidentate ligands as side chains
can therefore be exploited to control the metal cation binding
affinities and preferences of peptides incorporating these
residues.
Herein, we describe the metal-binding properties of several
peptides that incorporate 2,2′-bipyridine, 1,10-phenanthroline,
or 2,9-dimethyl-1,10-phenanthroline (neocuproine) functional-
ities. The syntheses and peptide incorporation of the 6-, 5-,
and 4-substituted (S)-2-amino-3-(2,2′-bipyridyl)propanoic acid
(1, 6Bpa; 2, 5Bpa; 3, 4Bpa) have been previously reported.6
The utility of these residues has been highlighted in the
semisynthesis of cytochrome c mutants with enhanced electron
transfer properties.12 Additionally, the residue (S)-2-amino-3-
(1,10-phenanthrol-2-yl)propanoic acid (4, Fen) has been used
X Abstract published in AdVance ACS Abstracts, November 1, 1996.
(1) Rabanal, F.; DeGrado, W. F.; Dutton, P. L. J. Am. Chem. Soc. 1996,
118, 473-474.
(2) Regan, L. Trends Biochem. Sci. 1995, 20, 280-285 and references
therein.
(3) (a) Walkup, G. K.; Imperiali, B. J. Am. Chem. Soc. 1996, 118, 3053-
3054. (b) Godwin, H. A.; Berg, J. M. J. Am. Chem. Soc. 1996, 118, 6514-
6515.
(4) Kazmierski, W. M. Int. J. Pept. Protein Res. 1995, 45, 241-247.
(5) Gilbertson, S. R.; Chen, G.; McLoughlin, M. J. Am. Chem. Soc. 1994,
116, 4481-4482.
(6) (a) Imperiali, B.; Prins, T. J.; Fisher, S. L. J. Org. Chem. 1993, 58,
1613-1616. (b) Imperiali, B.; Fisher, S. L. J. Org. Chem. 1992, 57, 757-
759. (c) Imperiali, B.; Fisher, S. L. J. Am. Chem. Soc. 1991, 113, 8527-
8528.
(10) (a) Summers, L. A. AdV. Heterocyclic Chem. 1978, 22, 1-69. (b)
McWhinnie, W. R.; Miller, J. D. AdV. Inorg. Chem. Radiochem. 1969, 12,
135-213. (c) Ko¨nig, E. Coord. Chem. ReV. 1968, 3, 471-495. (d) Lindoy,
L. F.; Livingstone, S. E. Coord. Chem. ReV. 1967, 2, 173-193. (e) Brandt,
W. W.; Dwyer, F. P.; Gyarfas, E. C. Chem. ReV. 1954, 54, 959-1017.
(11) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry; Jonh
Wiley & Sons, Inc: Singapore, 1988.
(7) Rana, T. M.; Ban, M.; Hearst, J. E. Tetrahedron Lett. 1992, 33, 4521-
4524.
(8) Wilson, S. R.; Yasmin, A.; Wu, Y. J. Org. Chem. 1992, 57, 6941-
6945.
(9) Ruan, F.; Chen, Y.; Hopkins, P. B. J. Am. Chem. Soc. 1990, 112,
9403-9404.
(12) Wuttke, D. S.; Gray, H. B.; Fisher, S. L.; Imperiali, B. J. Am. Chem.
Soc. 1993, 115, 8455-8456.
S0002-7863(96)01972-5 CCC: $12.00 © 1996 American Chemical Society

11350 J. Am. Chem. Soc., Vol. 118, No. 46, 1996
Cheng et al.
to provide a spectroscopic reporter group in an iterative design
process that resulted in the formation of a folded, metal-
independent ââR motif.13 The syntheses of Fen (4) and the
novel neocuproine-containing R-amino acid, (S)-2-amino-3-(9-
methyl-1,10-phenanthrol-2-yl)propanoic acid (5, Neo), are also
presented in this paper. These syntheses afford multigram
quantities of NR-Fmoc-protected R-amino acids for incorporation
into polypeptides. A key step in the syntheses of both the Fen
and Neo residues involves an enzymatic resolution with alkaline
protease.6a,14 These unnatural amino acids are completely
compatible with peptide assembly using standard Fmoc-based
solid phase methodology and do not require side chain protec-
tion.15 Therefore, these bidentate unnatural residues with
enviable inherent metal-binding properties can be readily
incorporated into small peptides.
Figure 1. (a) Schematic diagram of the â-hairpin motif designed to
bring two 6Bpa residues into close proximity (generated using the
program MOLSCRIPT17). (b) An edge-on view of the same model.
utilized to connect the two metal-binding residues. The metal-
binding affinity can therefore be varied by changing the extent
of structural preorganization of the motif required for high-
affinity and cooperative binding between two ligating residues.
Several hexapeptides incorporating the unnatural metal-
chelating amino acids described above were synthesized, and
their metal-binding properties were investigated. Peptides
containing only one bidentate residue and those with two
bidentate amino acids display different affinities for metal
cations, indicating cooperative ligation in the latter. Further-
more, constructs containing both a bidentate residue and a
naturally occurring metal-binding residue (Asp or His) were
prepared and exhibited varying metal-binding affinities. Two
sequences, -Val-Pro-D-Ser-Phe- and -Thr-Pro-D-Ala-Val- with
different type II â-turn propensities were used to connect the
metal-binding residues to probe the influence of structural
preorganization on metal binding affinity.19,21,22 In light of the
successful utilization of a type II′ turn as a structural nucleation
element in a designed ââR motif,13 an additional peptide
containing the sequence -Val-D-Pro-Ser-Phe-, which was pre-
dicted to form a type II′ â-turn, was also investigated. For all
the peptides studied, the N-termini were capped with an acetyl
group and the C-termini were capped with an amide in order to
avoid potential peptide conformational changes due to electro-
static interactions and competing metal cation coordination
effects involving the free carboxyl or amino termini.
Naturally occurring metalloproteins frequently form metal
binding sites with multiple residues to achieve the high affinity
and selectivity for their native metal cations. Therefore, the
incorporation of more than one metal-ligating amino acid in
designed metallopeptides could enhance the affinity and selec-
tivity of such constructs. In order for two residues to coop-
eratively participate in the formation of a high-affinity metal
binding site, it is advantageous to place the coordinating side
chain functionalities in close proximity. One of the most direct
methods to achieve this placement would be to situate the key
residues flanking a preorganized polypeptide motif such as a
reverse turn. The chain reversal afforded by a â-turn would
place residues at either end of the sequence proximal to each
other and promote the cooperative binding of metal cations.16
Judicious placement of the residues is essential to situate the
two metal binding side chains on the same face of the â-hairpin
motif (Figure 1).17 Molecular modeling studies reveal that two
metal-binding amino acids flanking a four-residue turn-forming
peptide sequence places the metal coordinating moieties on the
same face of the motif and in close proximity. The type II
â-turn has been studied extensively,18-20 and therefore different
peptide sequences with varying turn-forming propensities were
Results
Synthesis of Nr-Fmoc-(S)-2-amino-3-(1,10-phenanthrol-
2-yl)propanoic Acid (Fmoc-Fen) and Nr-Fmoc-(S)-2-amino-
3-(9-methyl-1,10-phenanthrol-2-yl)propanoic Acid (Fmoc-
Neo). The syntheses of the two racemic amino acids were
accomplished using a modified version of the Sørenson method23
from the corresponding chloromethyl derivatives. Esterification
of the amino acids in acidic methanol produced the amino acid
methyl esters (6 and 7)24 ready for enzymatic resolution.
Stereoselective hydrolysis of the R-amino acid methyl esters
was achieved using the enzyme alkaline protease (Scheme 1).6a,14
The hydrolysis afforded the L-enantiomer with high stereose-
lectivity (>98% ee). The absolute configuration of the resulting
(13) (a) Struthers, M. D.; Cheng, R. P.; Imperiali, B. Science 1996, 271,
342-345. (b) Struthers, M. D.; Cheng, R. P.; Imperiali, B. J. Am. Chem.
Soc. 1996, 118, 3073-3081.
(14) Chen, S.-T.; Wang, K.-T.; Wong, C. H. J. Chem. Soc., Chem.
Commun. 1986, 1514-1516.
(15) Fields, G. B.; Noble, R. L. Int. J. Pept. Protein Res. 1990, 35, 161-
214.
(16) Imperiali, B.; Kapoor, T. M. Tetrahedron 1993, 49, 3501-3510.
(17) Kraulis, P. J. J. Appl. Crystallogr. 1991, 24, 946-950.
(18) Hutchinson, E. G.; Thornton, J. M. Protein Sci. 1994, 3, 2207-
2216.
(19) Imperiali, B.; Fisher, S. L.; Moats, R. A.; Prins, T. J. J. Am. Chem.
Soc. 1992, 114, 3182-3188.
(20) Dyson, H. J.; Rance, M.; Houghten, R. A.; Lerner, R. A.; Wright,
P. E. J. Mol. Biol. 1988, 201, 161-200.
(21) Fisher, S. L. Ph.D. Thesis, California Institute of Technology, 1993.
(22) Prins, T. J. M.S. Thesis, California Institute of Technology, 1992.
(23) Albertson, N. F. J. Am. Chem. Soc. 1946, 68, 450-453.
(24) Experimental details for the preparation of compounds 6 and 7 are
described in the Supporting Information.

Metallopeptide Design
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11351
Scheme 1. Synthesis of Fmoc-Protected Fen and Neo
R-Amino Acids
a Alkaline protease, tBuOH, 10% NaHCO3, room temperature.
b Fmoc-OSu, dioxane, bicarbonate solution, room temperature.
amino acids was established by the specificity of the enzyme
to hydrolyze only the L-amino acid ester and the convention of
the elution profile for the CrownPak CR (+) chiral column.
Furthermore, the change of optical rotation upon lowering the
pH for Neo followed the Clough-Lutz-Jirgenson rule (pH 8.8,
[R]2D5 +80.3; pH 0, [R]D25 8.5),25 which further defines the
configuration of the stereocenter. Subsequent protection of the
L-amino acid with N-9-fluorenylmethylsuccinimidyl carbonate
(Fmoc-OSu) yielded the fully protected phenanthroline amino
acid derivatives (8 and 9), which were used directly for solid
phase peptide synthesis. All steps leading to both R-amino acids
are amenable to large-scale synthesis.
Dissociation Constants of the Peptide-MII Complexes.
The hexapeptides are designated according to the residues
present at positions 1 and 6. Standard three-letter abbreviations
are employed for natural residues, while 6Bpa, 5Bpa, and 4Bpa
are used to designate the unnatural R-amino acids 1, 2, and 3,
and Fen and Neo are used for residues 4 and 5, respectively.
The prime (′) designates that the type II′ turn sequence -Val-
D-Pro-Ser-Phe- connects the first and last residues. The
subscript w indicates the connecting sequence -Thr-Pro-D-Ala-
Val-, which displays a limited tendency to form a turn.22 Lack
of any additional notation would refer to the intervening
sequence -Val-Pro-D-Ser-Phe- which exhibits stronger turn
propensity.19,21
Bpa-Containing Peptides. The peptides 4BpaPhew (Ac-
4Bpa-Thr-Pro-D-Ala-Val-Phe-NH2), 5BpaPhew (Ac-5Bpa-Thr-
Pro-D-Ala-Val-Phe-NH2), and 6BpaPhew (Ac-6Bpa-Thr-Pro-D-
Ala-Val-Phe-NH2) were synthesized and studied. Additionally,
peptides 6Bpa6Bpa (Ac-6Bpa-Val-Pro-D-Ser-Phe-6Bpa-NH2)
and 6Bpa6Bpaw (Ac-6Bpa-Thr-Pro-D-Ala-Val-6Bpa-NH2), which-
both contain two 6Bpa residues, were also investigated. The
metal cation titrations of these Bpa-containing peptides were
monitored by absorption spectroscopy (UV-Vis) as illustrated
in Figure 2. The π-π* transition of the bipyridyl moiety
displays a distinct red-shift upon chelation of metal cations.
Clean isosbestic points were observed, supporting a two-state
model equilibrium. Binding isotherms were obtained from such
experiments. The apparent dissociation constants for the
peptide-MII complexes were determined to be the free metal
cation concentration when half of the peptide was bound. The
apparent dissociation constants of the peptide-MII complexes
for the Bpa-containing peptides are summarized in Table 1.
Fen- and Neo-Containing Peptides. Peptides FenPhew (Ac-
Fen-Thr-Pro-D-Ala-Val-Phe-NH2) and NeoPhe (Ac-Neo-Val-
Pro-D-Ser-Phe-Phe-NH2) were synthesized and evaluated to
assess the inherent chelation properties of the residues Fen and
Neo, respectively. Additionally, five hexapeptides were de-
signed to probe the effect upon metal binding by delivering an
auxiliary, natural metal-binding residue that may act coopera-
tively with the Fen or Neo bidentate residue (Table 2). The
Figure 2. (a) Absorption spectra of 52 µM 4BpaPhew upon addition
of ZnCl2 up to 178 µM at pH 6.8-7.0 in 0.2 M NaCl. (b) The binding
isotherm obtained from the data shown in panel a at 306 nm.
Table 1. Apparent Dissociation Constants (µM) for
Bpa-Containing Peptide-MII Complexesa
peptide
sequence
ZnII CoII
4BpaPhew
5BpaPhew
6BpaPhew
Ac-4Bpa-Thr-Pro-D-Ala-Val-Phe-NH2
Ac-5Bpa-Thr-Pro-D-Ala-Val-Phe-NH2
Ac-6Bpa-Thr-Pro-D-Ala-Val-Phe-NH2
22 NDb
19 NDb
2700 1100
960 110
6Bpa6Bpaw Ac-6Bpa-Thr-Pro-D-Ala-Val-6Bpa-NH2
6Bpa6Bpa Ac-6Bpa-Val-Pro-D-Ser-Phe-6Bpa-NH2
16
7
a Metal cation titrations performed at pH 7 in 0.2 M NaCl.
b Dissociation constants not determined due to stoichiometry of 2:1
peptide:metal.
metal cation titrations of the Fen- and Neo-containing peptides
were monitored by UV-Vis spectroscopy (Figures 3 and 4).
Similar to the Bpa-containing peptides, a distinct red-shift of
the π-π* transition for the phenanthrolyl moiety was observed
upon binding metal cations. Thus, UV-Vis spectroscopy was
used to monitor the metal cation titrations. Clean isosbestic
points were observed for all the titrations except for experiments
performed on peptide NeoPhe, in which essentially no change
in the absorption spectrum was observed upon the addition of
ZnII or CoII. For this peptide only, metal cation titrations were
monitored by circular dichroism (CD) spectroscopy, since
distinct changes in the near-UV region of the CD spectra
(25) Greenstein, J. P.; Winitz, W. D. Chemistry of the Amino Acids; John
Wiley & Sons, New York, 1961; pp 1879-2155.

11352 J. Am. Chem. Soc., Vol. 118, No. 46, 1996
Cheng et al.
Table 2. Apparent Dissociation Constants (µM) of Fen- and
Neo-Containing Peptide-MII Complexesa
peptide
sequence
ZnII
CoII
7.91
FenPhew Ac-Fen-Thr-Pro-D-Ala-Val-Phe-NH2
FenHisw Ac-Fen-Thr-Pro-D-Ala-Val-His-NH2
FenHis
FenHis′
FenAsp
NeoPhe
NeoHis
26.8
1.50
0.65
0.96
8.28
2.48
0.60
1.57
3.91
Ac-Fen-Val-Pro-D-Ser-Phe-His-NH2
Ac-Fen-Val-D-Pro-Ser-Phe-His-NH2
Ac-Fen-Val-Pro-D-Ser-Phe-Asp-NH2
Ac-Neo-Val-Pro-D-Ser-Phe-Phe-NH2 110b
300b
Ac-Neo-Val-Pro-D-Ser-Phe-His-NH2
0.69
5.45
a Metal cation titrations performed in 0.05 M HEPES buffer at pH
8.25. b Values obtained from titrations performed in 0.01 M Tris buffer
at pH 8.25 monitored by circular dichroism (CD) spectroscopy.
Figure 4. (a) Absorption spectra of 12.8 µM NeoHis upon addition
of ZnCl2 up to 123 µM in 0.05 M HEPES buffer at pH 8.25. (b) The
binding isotherm obtained from the data shown in panel a at 280 nm.
free MII concentration in solution, where b is the path length of
cuvette and ∆A is the change in absorbance at the wavelength
of interest).
Discussion
The NR-Fmoc-protected Fen and Neo residues have been
synthesized and incorporated into several peptides. This further
expansion in the repertoire of R-amino acid building blocks
enables the construction of a wide variety of novel metal-binding
peptides. Additionally, these residues facilitate the formation
of metal-binding sites with specific ligand type and geometry.
These multidentate amino acids provide avid metal binding
within short peptides and are also useful as structural probes
due to their rich spectroscopic properties.13
Bpa-Containing Peptides. Peptides 4BpaPhew, 5BpaPhew,
and 6BpaPhew each contain only one metal-binding Bpa residue.
These peptides differ only at the point of attachment to the
â-carbon of the bipyridyl-containing amino acid (4Bpa, 5Bpa,
and 6Bpa). Since the 4 and the 6 isomers have previously
behaved differently in the construction of bipyridyl-alanine
cytochrome c mutants,12 the various isomers were investigated
in peptides with the same underlying sequence to evaluate their
individual metal-binding characteristics more carefully. The
Figure 3. (a) Absorption spectra of 7.85 µM FenHis upon addition of
ZnCl2 up to 42 µM in 0.05 M N-(2-hydroxyethyl)piperazine-N′-2-
ethanesulfonic acid (HEPES) buffer at pH 8.25. (b) The binding
isotherm obtained from the data shown in panel a at 277 nm.
attributable to the neocuproine moiety were observed (Figure
5). Calculations employing an iterative method based on the
Scott equation26 were performed to obtain the apparent dis-
sociation constants for the peptide-MII complexes (Table 2).
The final iteration from a typical analysis is shown in Figure 6,
in which [MII]free/b∆A is plotted against [MII]free and the apparent
dissociation constant is given by y-intercept/slope ([MII]free
)
(26) Connors, K. A. Binding Constants: the Measurement of Molecular
Complex Stability; Wiley: New York, 1987.

Metallopeptide Design
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11353
Figure 6. Final analysis of the data obtained from Figure 3 where
7.85 µM FenHis is titrated with ZnCl2 up to 42 µM. The ratio of the
y-intercept and slope derived from the linear fit gives the dissociation
constant of 0.55 µM.
Figure 5. (a) Circular dichroism spectra of 66 µM NeoPhe upon
addition of CoCl2 up to 1.0 mM in 0.01 M Tris buffer at pH 8.25. (b)
The binding isotherm obtained from the data shown in panel a at 271
nm.
Figure 7. Schematic diagram of the metal binding event for (a) the
hexapeptide 6Bpa6Bpa incorporating a structurally preorganized in-
tervening sequence with high turn propensity and (b) the hexapeptide
6Bpa6Bpaw with a less structured connecting sequence (generated using
the program MOLSCRIPT17).
dissociation constants of the peptide-ZnII complexes for
6BpaPhew, 5BpaPhew, and 4BpaPhew exhibit the overall trend
6BpaPhew > 5BpaPhew ∼ 4BpaPhew (Table 1). Thus, the
peptide 6BpaPhew shows the lowest affinity for ZnII among the
three peptides. The steric hindrance of the ortho substitution
presumably results in the larger dissociation constant for peptide
6BpaPhew. As the substitution of the residues 5Bpa and 4Bpa
does not interfere with the metal-binding process, the peptides
5BpaPhew and 4BpaPhew exhibit similar affinities for ZnII.
Unlike all the other peptide-MII complexes studied in this paper,
the stoichiometries of the peptide-CoII complexes for 5BpaPhew
and 4BpaPhew were determined to be 2:1, reflecting the
unencumbered chelation by these residues. Thus, a change in
the point of attachment of the bipyridyl ring to the amino acid
results in alteration of both the complex stoichiometry and the
metal binding affinity.
The affinities of peptides 6Bpa6Bpaw and 6Bpa6Bpa for ZnII
and CoII were significantly stronger than those observed for
peptide 6BpaPhew (i.e., lower dissociation constants), indicating
the cooperative participation of both bipyridyl moieties as
ligands (Table 1). This participation of a second 6Bpa residue
enhances the metal binding affinity by two orders of magnitude.
The peptide 6Bpa6Bpa, which contains an intervening sequence
with higher type II â-turn propensity, exhibits stronger binding
for both ZnII and CoII than the weak turn-forming peptide
6Bpa6Bpaw. This result strongly suggests that the connecting
residues can serve to bring the two metal-chelating residues into
close proximity and therefore play a role in the process of
binding metal cations (Figure 7). Thus, through the control of
the peptidyl structure, the metal-binding affinity of a peptide
for a specific metal cation can be modulated by 60-fold.
Fen- and Neo-Containing Peptides. Peptides FenPhew and
NeoPhe were analyzed to determine the inherent metal-binding
properties of the bidentate residues. The peptide FenPhew has
a stronger affinity for ZnII and CoII than peptide NeoPhe (Table
2). This suggests that the steric effect of the 9-methyl group
Two bidentate metal-chelating residues were positioned at
opposite ends of peptides 6Bpa6Bpa and 6Bpa6Bpaw. The
intervening sequences were designed to adopt a â-turn structure
with different turn-forming propensities. The sequence -Val-
Pro-D-Ser-Phe- has been shown to exhibit more turn character
than -Thr-Pro-D-Ala-Val- by CD and NMR spectroscopy.19,21,22

11354 J. Am. Chem. Soc., Vol. 118, No. 46, 1996
Cheng et al.
of Neo interferes with metal chelation. This trend parallels that
observed for the parent ligands 2-methyl-1,10-phenanthroline
(ZnII complex, KD ) 11.0 µM; CoII complex, KD ) 7.94 µM)
and 2,9-dimethyl-1,10-phenanthroline (neocuproine) (ZnII com-
plex, KD ) 79.4 µM; CoII complex, KD ) 63 µM).27 However,
the peptide NeoPhe preferentially binds ZnII over CoII . This
result is unusual given the opposite trends observed for Bpa
and Fen peptides as well as the parent ligand neocuproine. This
reversal in binding specificity is most likely due to the steric
bulk introduced by the peptide. However, the participation of
functional groups other than the Neo side chain cannot be
completely ruled out.
Peptides FenHisw, FenHis, FenHis′, and FenAsp exhibit
stronger metal-binding affinities than the parent “Fen only”
peptide, FenPhew (Table 2). Thus, the auxiliary naturally
occurring coordinating residues (His or Asp) incorporated within
these peptides must be actively participating in the metal-binding
event. Furthermore, the greater metal-binding affinity of peptide
FenHis as compared to peptide FenHisw suggests that the
sequence with higher turn propensity promotes a higher metal-
binding affinity, which is consistent with the results obtained
for peptides 6Bpa6Bpa and 6Bpa6Bpaw. Additionally, FenHis′
binds CoII and ZnII with affinities similar to FenHis. In this
regard, the type II′ turn does not appear to contribute a
significant advantage over the type II turn in preorganizing the
metal binding site.
Comparison between peptides NeoPhe and NeoHis illustrates
that incorporation of the auxiliary metal-binding residue, his-
tidine, results in a 160-fold increase in affinity for ZnII (Table
2). Additionally, the peptide NeoHis was anticipated to have
diminished metal-binding properties as compared to the analo-
gous Fen-containing peptide due to the steric crowding of a
bound metal cation by the 9-methyl group near a nitrogen donor
atom. While this was demonstrated for the CoII binding
affinities of peptides NeoHis and FenHis, the KD of the ZnII
complexes for the two peptides were comparable. In addition,
there was 1 order of magnitude preference for NeoHis to bind
ZnII over CoII. This result highlights the potential for tuning
the metal-binding affinities and selectivities of short peptide
motifs for future metal cation sensor designs by judicious
selection of the unnatural bidentate residue and auxiliary
naturally occurring metal-ligating amino acid. Since CoII is
known to favor octahedral geometry due to ligand field
stabilization while ZnII has no specific geometrical preference,2,28
it is possible that the steric effect of the 9-methyl group inter-
feres with the formation of an octahedral complex and results
in the increased KD for the peptide-CoII complex. On the other
hand, ZnII does not prefer a specific geometry and therefore
exhibits similar metal-binding properties for FenHis and Neo-
His. Note that the differences in metal-binding preferences
between FenHis and NeoHis must be ascribed to a single methyl
group.
stoichiometry of the peptide-CoII complexes for peptides
4BpaPhew and 5BpaPhew are 2:1. In contrast, high affinity and
1:1 stoichiometry can be achieved through the incorporation of
the phenanthrolyl-containing amino acids (Fen, Neo). Second,
further modulation of the metal-binding affinity can be ac-
complished via the incorporation of auxiliary metal-binding
amino acids (6Bpa, Asp, or His). The participation of an
auxiliary residue is exemplified by the 160-fold enhancement
in ZnII binding affinity for NeoHis over NeoPhe. Third, control
of the peptide structure can be exploited to bring two metal-
binding residues into close proximity and facilitate the
metal-binding process. Peptide 6Bpa6Bpa, which contains a
sequence with high turn propensity connecting two 6Bpa
residues, exhibits a 80-fold higher affinity for ZnII than peptide
6Bpa6Bpaw, which has a considerably less structured intervening
sequence.
Conclusion
The ability to tune the metal-binding affinity of designed
peptides has been demonstrated in this study by exploiting three
different features: selection of unnatural bidentate residue,
incorporation of auxiliary metal-binding amino acid, and control
of structural preorganization of the peptide architecture. Dis-
sociation constants ranging from 10-3 to <10-6 M for ZnII have
been achieved by altering these three features. The combination
of a variety of bidentate amino acids with the structural
preorganization afforded by a peptidyl template, such as a
reverse turn, results in the ability to modulate the metal-binding
affinity of metallopeptides. The availability of the novel
chelating R-amino acids and the guidelines for their utilization
derived from this study can be exploited in future metalloprotein
and peptide-based sensor design.
Experimental Section
General. All reagents were obtained from Aldrich Chemical Co.
and used without further purification. Protease Type VIII (Subtilisin
Carlsberg, Bacterial, CAS Registry No. 9014-01-1, 11.6 units/mg) was
purchased from Sigma Chemical Co. and stored at 4 °C. All synthetic
intermediates were stored at 4 °C after purification. 1H and 13C NMR
spectra were recorded on a General Electric QE-300 at 300 MHz for
1H NMR, 75 MHz for 13C.
CrownPak CR Column Analysis. Characterization was obtained
through HPLC analysis using a CrownPak CR(+) column (J. T. Baker),
on an Isco Model 2350 HPLC equipped with an Isco Model V4
absorbance detector using the conditions as described. 2-Amino-3-
(1,10-phenanthrol-2-yl)propanoic acids and methyl esters: flow rate
1.0 mL/min; 25 °C, monitored at 278 nm; eluent 10% MeOH/0.1 M
HClO4. Methyl esters: D 11.50 min, L 12.25 min; amino acids: D
7.75 min, L 8.25 min. 2-Amino-3-(9-methyl-1,10-phenanthrol-2-yl)-
propanoic acids and methyl esters: flow rate 0.6 mL/min; 25 °C,
monitored at 280 nm; eluent 10% MeOH/0.1 M HClO4. Methyl
esters: D 15.73 min, L 18.10 min; amino acids: D 11.87 min, L 12.61
min.
Nr-(9-Fluorenylmethyloxycarbonyl)-(S)-2-amino-3-(1,10-phenan-
throl-2-yl)propanoic Acid (8). A solution of racemic 2-amino-3-(1,-
10-phenanthrol-2-yl)propanoic acid methyl ester24 (6, 3.6587 g, 16.0
mmol) was dissolved in 45 mL of tert-butanol and added to 445 mL
of 10% NaHCO3. A sample of alkaline protease (125 mg, 800-2000
units) in 120 mL of 0.2 M NaHCO3 was added to the amino acid methyl
ester (8 mmol) in a 1-L Erlenmeyer flask. The mixture was rotated at
120-200 rpm and monitored by HPLC at 20-min intervals. When
the ratio of the D-amino acid methyl ester to the L-amino acid methyl
ester was >50:1 (about 40 min), the mixture was extracted with CHCl3
(6 × 150 mL). The aqueous phase was reduced in volume to remove
any chloroform and then lyophilized yielding a mixture of amino acid
and carbonate salts. The combined organic layers were dried over Na2-
SO4, and the solvent was evaporated. HPLC analysis of the aqueous
phase showed a 92-94% ee of the L-amino acid based on conventions
Summary. Several guidelines for the design of short
metallopeptides with high metal-binding affinity can be derived
from this study. First, the novel chelating amino acids presented
generally bind metal cations with higher affinity than naturally
occurring metal-binding residues due to the bidentate nature of
the side chain functionality. The selection of unnatural chelating
amino acid can affect both affinity and stoichiometry of the
metal complex for the designed peptide. The metal-binding
affinity for the hexapeptides with bipyridyl-containing amino
acids follows the trend 4Bpa ∼ 5Bpa > 6Bpa. However, the
(27) Smith, R. M.; Martell, A. E. Critical Stability Constants, Volume
2, Amines; Plenum Press: New York, 1975; pp 235-262.
(28) Lippard, S. J.; Berg, J. M. Principles of Bioinorganic Chemistry;
University Science Books: Mill Valley, 1994 .

Metallopeptide Design
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11355
of the elution profiles on the CrownPak(+) column and the specificity
of alkaline protease. The amino acid/carbonate salt mixture was
dissolved in 25 mL of 10% Na2CO3. The 9-fluorenylmethylsuccin-
imidyl carbonate (1.2 equiv) was dissolved in 15 mL of dioxane and
added dropwise to the amino acid solution. The reaction mixture was
shaken periodically for 1.5 h, then transferred to a separatory funnel,
and diluted with 100 mL of water. This mixture was washed with
ether (4 × 60 mL) and transferred to a 250-mL Erlenmeyer flask. After
being cooled to 0 °C and adjusted to pH <2 with concentrated HCl,
the precipitate was isolated by centrifugation and washed with water
(3 × 100 mL). The residue was transferred to a round bottom flask
with methanol (200 mL). The methanol was removed under reduced
pressure and replaced with toluene (80 mL). The resulting toluene/
water mixture was removed under reduced pressure. The residue was
rubbed with ether to obtain a solid that was dried in vacuo to yield the
Fmoc-protected amino acid (85% based on the amount of L-amino acid
methyl ester). mp ) 187 °C (dec); HRMS: calculated [MH]+ for
C30H24N3O4 [490.1767]; observed [490.1749]; TLC (SiO2; 4:1 CHCl3:
Table 3. Calculated and Observed High-Resolution Mass Values
for the Synthesized Hexapeptides
peptide
formula
calcd [MH+]
obsd [MH+]
a
a
a
4BpaPhew
5BpaPhew
6BpaPhew
C41H53N9O8
C41H53N9O8
C41H53N9O8
C45H55N11O8
C50H57N11O8
C43H53N9O8
C40H51N11O8
C45H53N11O8
C43H51N9O10
C45H53N11O8
C49H57N9O8
C46H55N11O8
800.4095
800.4095
800.4095
878.4313
940.4470
824.4095
814.4000
876.4156
854.3837
898.3976c
900.4408
890.4302
800.4055
800.4057
800.4153
878.4285
940.4459
824.4117
814.3983
876.4140
854.3841
898.4031d
900.4422
890.4313
a
6Bpa6Bpaw
6Bpa6Bpaa
a
FenPhew
FenHisw
a
FenHisa
FenAspa
FenHis’b
NeoPheb
NeoHisb
a Peptides synthesized on the automated peptide synthesizer. b Pep-
tides synthesized on macrocrown supports. c Calculated [MNa+]. d Ob-
served [MNa+].
1
MeOH; UV); Rf ) 0.55; H NMR (DMSO-d6) δ: 3.45 (m, 1H), 3.65
(m, 1H), 4.10 (m, 3H), 4.40 (m, 1H), 4.80 (d, 2H, J ) 5.8 Hz), 7.05
(m, 2H), 7.35 (m, 2H), 7.50 (m, 2H), 7.68 (d, 1H, J ) 7.2Hz), 7.80 (d,
1H, J ) 7.6 Hz), 7.93 (m, 2H), 8.10 (m, 2H), 8.64 (d, 1H, J ) 8.3Hz),
9.00 (d, 1H, J ) 4.6 Hz), 9.25 (d, 1H, J ) 4.85 Hz); 13C NMR (DMSO-
d6) δ: 47.5, 55.1, 66.6, 121.0, 124.3, 125.0, 126.0, 126.1, 127.2, 127.3,
127.4, 128.0, 128.6, 128.8, 129.6, 137.2, 137.6, 141.6, 144.7, 144.8,
150.9, 159.4, 174.2; IR (mineral oil mull) cm-1 : 3307, 1690, 1601,
1537, 1454, 1372, 1337, 1261, 1150, 1102, 1078, 1037, 850, 761, 732;
[R]2D5 -67.9° (c ) 0.35, DMSO); UV (MeOH, nm) λmax ) 299
(11 000), 268 (32 000), 224 (38 900).
IR (KBr) cm-1 : 3316, 3063, 3043, 2946, 1728, 1694, 1538, 1446,
1372, 1261, 1222, 1149, 1105, 1080, 1046, 857, 759, 740; [R]D25
-71.4°.
Peptide Synthesis. All peptides were synthesized on a 0.084-0.168
mmol scale by solid phase methods using NR-9-fluorenylmethyloxy-
carbonyl (Fmoc) amino acid with BOP/HOBT-activated ester chemistry
on a Milligen 9050 automated peptide synthesizer, except for the Neo-
containing peptides and FenHis′. These peptides were synthesized on
a macrocrown support purchased from Chiron Mimotopes. Com-
mercially available starting materials and reagents were purchased from
MilligenBiosearch, EM Science, NovaBiosearch, or Aldrich Chemical
Co.
N-(9-Fluorenylmethoxycarbonyl)-(S)-2-amino-3-(9-methyl-1,10-
phenanthrol-2-yl)propanoic Acid (9). A solution of racemic 2-amino-
3-(9-methyl-1,10-phenanthrol-2-yl)propanoic acid methyl ester24 (7,
0.5733 g, 2.04 mmol) was dissolved in 20 mL of tert-butanol and added
to 200 mL of 10% NaHCO3. A sample of alkaline protease (55.4 mg,
400-1000 units) was added to the amino acid methyl ester in a 1-L
Erlenmeyer flask. The mixture was rotated at 150 rpm and monitored
by HPLC at 20-min intervals. When the ratio of the D-amino acid
methyl ester to the L-amino acid methyl ester was >50:1 (about 40
min), the mixture was extracted with CHCl3 (6 × 150 mL). The
aqueous phase was reduced in volume to remove any CHCl3 and then
lyophilized to yield a mixture of amino acid and carbonate salts. HPLC
analysis of the aqueous phase showed a 98% ee of the L-amino acid
based on conventions of the elution profiles on the CrownPak(+)
column and the specificity of alkaline protease. The resulting product
L-Neo had [R]2D5 +80.3 at pH 8.8 and [R]2D5 +8.5 at pH 0, which is
consistent with the Clough-Lutz-Jirgenson rule, confirming the
configuration of the stereocenter. The amino acid/carbonate salt mixture
was suspended in 200 mL of water. The 9-fluorenylmethylsuccinimidyl
carbonate (0.5508 g, 1.63 mmol) was dissolved in 20 mL of dioxane
and added dropwise to the amino acid solution. The reaction mixture
was stirred at room temperature for 1.5 h. To the reaction, a solution
of 9-fluorenylmethylsuccinimidyl carbonate (0.3431 g, 1.02 mmol)
dissolved in 15 mL of dioxane was added dropwise. After being stirred
overnight, this mixture was washed with ether (4 × 100 mL). After
being cooled to 0 °C and adjusted to pH < 2 with concentrated HCl,
the precipitate was isolated by centrifugation and washed with water
(3 × 100 mL). The residue was transferred to a round bottom flask
with methanol (150 mL). The methanol was removed under reduced
pressure and replaced with toluene (80 mL). The resulting toluene/
water mixture was removed under reduced pressure. The residue was
rubbed with ether to yield a solid that was dried in vacuo (0.4726 g,
92% yield based on the amount of L-amino acid methyl ester). mp )
131.4-132.7 °C (dec.); HRMS: calculated for [MH]+ C31H26N3O4
[504.1923]; observed [504.1931]; TLC (SiO2; 4:1 CHCl3:MeOH; UV);
Rf ) 0.36; 1H NMR (DMSO-d6) δ: 22.76 (s, 3H), 3.45 (m, 1H), 3.55
(m, 1H), 4.18 (m, 3H), 4.61 (m, 1H), 7.15 (t, 1H, J ) 7.4 Hz), 7.21 (t,
1H, J ) 7.4 Hz), 7.35 (m, 3H), 7.61 (m, 2H), 7.70 (d, 1H, J ) 8.2
Hz), 7.84 (d, 2H, J ) 7.6 Hz), 7.90 (s, 2H), 8.17 (d, 1H, J ) 8.3 Hz),
8.36 (d, 1H, J ) 8.2 Hz), 8.40 (d, 1H, J ) 8.3 Hz); 13C NMR (DMSO-
d6) δ: 40.8, 47.0, 47.2, 54.2, 66.2, 120.5, 120.6, 125.6, 126.3, 127.3,
127.5, 127.6, 128.0, 128.1, 137.5, 141.2, 144.4, 156.6, 158.7, 173.7;
Solid Phase Peptide Synthesis. Fmoc-PAL-PEG-PS resin (0.21
mmol/g) was used to afford carboxyl terminus primary amides. Typical
protocols for coupling a residue involved 30-90 min coupling cycles
with 3 or 4 equiv of amino acid. However, for the bipyridyl and
phenanthrolyl amino acids, a double coupling protocol was employed,
where 2.5 equiv was used for the first coupling followed by a similar
acylation with 1.5 equiv of the residue with 60-min coupling times for
each. Activated esters were formed in situ using benzotriazol-1-yloxy-
tris(dimethylamino)-phosphoniumhexafluorophosphate (BOP), 1-hy-
droxybenzotriazole (HOBt), and N-methylmorpholine or diisopropyl-
carbodiimide (DIPCDI), HOBt, and diisopropylethylamine (DIEA). Due
to poor solubility of the bipyridyl and phenanthrolyl amino acid-
activated esters, the amino acid ester solutions were filtered manually
through a 0.45-µm nylon syringe filter (Gelman Scientific) prior to
manual injection onto the column. Deprotection of FMOC-protected
amine groups were performed using either a 7-min 20% piperidine/
dimethylformamide (DMF) wash or a 5-min 2% 1,7-diazabicyclo[5,4,0]-
undec-7-ene (DBU)/DMF wash. Peptides were N-acylated on the resin
using either 7 equiv of acetic anhydride and 7 equiv of triethylamine
(TEA) or 21 equiv of acetic anhydride and 5 equiv of triethylamine in
3 mL of DMF for 2 h. The resin was then washed with DMF and
dichloromethane and lyophilized overnight.
The peptides were then deprotected and cleaved from the resin by
treatment with trifluoroacetic acid (TFA)/phenol/H2O/thioanisole/
ethanedithiol (82.5:5:5:5:2.5) or TFA/thioanisole/ethanedithiol/anisole
(90:5:3:2) for 2 h.29 The resin was filtered and washed with TFA, and
the combined filtrates were concentrated to 2 mL and precipitated with
ether/hexane 1:1 or 2:1 (20 mL). The supernatant was decanted, and
the peptides were triturated with ether/hexane 1:1 (5 × 20 mL) or ether/
hexane 2:1 (3 × 20 mL). The peptides were lyophilized overnight
and purified by reverse-phase C18 HPLC. All peptides were confirmed
by mass spectrometry (Table 3).
Peptide Synthesis on Macrocrowns. A typical full coupling cycle
involves the following steps; a 30 min soak in 20% (v/v) piperidine/
DMF suspended in DMF followed by a 5 min wash with DMF and
three rinses with methanol. After air drying for 10 min and soaking
in DMF 5 min, the pin is submitted to the coupling cocktail. Typical
couplings were performed for 2 h with Fmoc-protected amino acid
(29) King, D. S.; Fields, C. G.; Fields, G. B. Int. J. Pept. Protein Res.
1990, 36, 255-266.

11356 J. Am. Chem. Soc., Vol. 118, No. 46, 1996
Cheng et al.
pentafluorophenyl esters and HOBt at 80 mM concentration each with
450 µL of activated amino acid ester solution per pin. After
coupling,the pins were then washed once with DMF and three times
with methanol. The pins were then air dried for 10 min to complete
a full cycle. The residues Neo and Fen were coupled with 3 equiv of
amino acid and O-(7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate (HATU) /1-hydroxy-7-azabenzotriazole (HOAt)/
DIEA mediated chemistry in N-methylpyrrolidinone (NMP).30-32 These
cycles were repeated until the peptide was complete. Peptides were
N-acetylated on the solid support using 450 µL of DMF/Ac2O/TEA
(43:1:1) in place of a typical coupling solution. The peptides
synthesized were deprotected and removed from the solid support by
treatment with trifluoroacetic acid (TFA)/phenol/H2O/thioanisole/
ethanedithiol (82.5:5:5:5:2.5) for 4 h.29 The resulting solution was
concentrated to 2 mL volume and precipitated with ether/hexane 1:1.
The supernatant was decanted, and the peptides were triturated with
ether/hexane 1:1 (5 × 10 mL). The peptides were lyophilized overnight
and purified by reverse-phase C18 HPLC. The peptides were confirmed
by mass spectrometry (Table 3).
iterative process based on the Scott equation:26
KD
b[M]
∆A
[M]
)
+
[Pep]t [Pep]t∆ꢀ
where b is the cell path length, [M] is the free metal cation
concentration, ∆A is the change in absorbance due to complex
formation, [Pep]t is the total peptide in solution, and ∆ꢀ is the change
in molar absorptivity due to complex formation. Initial estimations of
[M] were achieved by using data where [M]total . [Pep]t. An analysis
was performed where b[M]/∆A was plotted against [M]. The ∆ꢀ thus
generated was used to calculate [M] more accurately ([M] ) [M]total
-
(∆A/∆ꢀ)) for every iteration until ∆ꢀ converges. When the ∆ꢀ
converged, the dissociation constants (KD) were calculated from the
last iterations as
y-intercept
KD )
slope
Absorption Spectroscopy (UV-Vis) Studies. The absorption
spectra were obtained on a Shimadzu UV-160U in 10-mm path length
cuvettes. The concentrations of the peptide solutions were determined
spectrophotometrically using the characteristic absorption band of the
bipyridyl (λmax ) 285 nm, ꢀmax ) 12800) or phenanthrolyl (Fen, λmax
) 268 nm, ꢀmax ) 20890; Neo, λmax ) 274 nm, ꢀmax ) 24150) moiety.
The aqueous solutions of known peptide concentrations were then
titrated with standard metal cation solutions at room temperature. For
the metal cation titration performed on bipyridyl-containing peptides,
the pH of the solution was held within the range pH ) 7.0-7.5 through
small additions of 0.1 N NaOH or 0.1 N HCl in the presence of 200
mM NaCl. For the peptides containing Fen or Neo, the titrations were
performed in 50 mM N-(2-hydroxyethyl)piperazine-N′-(2-ethanesulfonic
acid) (HEPES) buffer, pH 8.25. The distinct red-shift of the bipyridyl
or phenanthrolyl π-π* transition upon metal-binding cations was
monitored by absorption (UV-Vis) spectroscopy. With peptide
concentrations ranging from 1 to 500 µM, the metal-binding constants
were determined by calculations carried out using KaleidaGraph version
3.0 (Abelbeck Software, CA). A minimum of three titrations were
performed for each peptide-metal cation complex. The binding
isotherms were assumed to follow a two-state equilibrium when well-
defined isosbestic points associated with the addition of metal cations
were observed. The dissociation constants (KD) were obtained using a
Circular Dichroism Spectroscopy Studies. Circular dichroism
spectra were recorded on a Jasco J600 circular dichroism spectrometer.
The concentration of the peptide solutions was determined spectro-
photometrically using the characteristic absorption band of the phenan-
throlyl (Neo, λmax ) 274 nm, ꢀmax ) 24 150) moiety. The spectra
werebaseline corrected and noise reduced using the Jasco J600 System
software and were then further analyzed using KaleidaGraph version
3.0 (Abelbeck Software). Metal cation titrations were performed with
peptide concentrations ranging from 65 to 70 µM in 0.01M Tris buffer
(pH 8.25) in a 1-mm path length cuvette. The change in CD signature
where the phenanthrolyl moiety absorbs was monitored, and the binding
isotherms were thus obtained. The dissociation constants were
calculated as described in the UV-Vis experiments.
Acknowledgment. Financial support from the National
Science Foundation is gratefully acknowledged. R.P.C. is the
recipient of an NIH traineeship in Bioorganic and Bioinorganic
Chemistry. The authors would like to thank Thomas Prins for
his assistance in the initial phase of this research and Grant
Walkup and Dr. Mary Struthers for their help in the preparation
of this manuscript.
Supporting Information Available: Text description of the
experimental details for the syntheses of compounds 6 and 7
with the 1H NMR spectra of the new compounds 6-9, 12, and
13 (11 pages). See any current masthead page for ordering
information and Internet access instructions.
(30) (a) Carpino, L. A.; El-Faham, A.; Minor, C. A.; Albericio, F. J.
Chem. Soc., Chem. Commun. 1994, 201-203. (b) Carpino, L. A. J. Am.
Chem. Soc. 1993, 115, 4397-4398.
(31) Kuroda, H.; Chen, Y.-N.; Kimura, T.; Sakakibara, S. Int. J. Pept.
Protein Res. 1992, 40, 294-299.
(32) Fields, G. B.; Fields, C. G. J. Am. Chem. Soc. 1991, 113, 4202-
4207.
JA9619723