Fine tuning of the cation affinity of artificial receptors based on cyclic peptides by intramolecular conformational control

Article abstract of DOI:10.1002/1099-0690(200101)2001:2<311::AID-EJOC311>3.0.CO;2-M

A series of cyclic hexapeptides consisting of alternating 4-substituted 3-aminobenzoic acid units (R = CH₃, Cl, CH₂OCH₃, OCH₃, COOCH₃) and residues of the natural amino acid proline has been prepared and their ion affinities have been investigated. Whereas the unsubstituted parent compound (R = H) is able to bind cations through cation-π interactions with the aromatic subunits, as well as anions through hydrogen bonding with the peptide NH groups, the introduction of substituents at the 4-positions of the aromatic rings results in complete loss of the anion affinity. The cation complex stabilities depend on the substituents and cover a wide range from K_a = 140 M^-1 for R = CH₃ to K_a = 10800 M^-1 for R = COOCH₃ (K_a = 1260 M^-1 for R = H) with n-butyltrimethylammonium picrate. The conformations of the peptides in solution have been determined by one-and two-dimensional NMR techniques and FT-IR spectroscopy. It was found that all the substituents prevent the peptides from adopting the necessary conformation for anion binding. For one receptor (R = OCH₃), the results have been corroborated by a crystal structure determination. AM1 calculations have been used to estimate the electrostatic potential surfaces of the substituted aromatic subunits. The variation in the cation complex stabilities can be mainly attributed to the effects of the substituents on the solution conformations of the peptides. The influence of the substituents on the electrostatic potentials of the aromatic peptide subunits appears to be less important.

Full text of DOI:10.1002/1099-0690(200101)2001:2<311::AID-EJOC311>3.0.CO;2-M

FULL PAPER
Fine Tuning of the Cation Affinity of Artificial Receptors Based on Cyclic
Peptides by Intramolecular Conformational Control
Stefan Kubik*[a] and Richard Goddard[b]
Keywords: Supramolecular chemistry / Peptides / Receptors / CationϪπ interactions / Conformational analysis
A series of cyclic hexapeptides consisting of alternating 4-
substituted 3-aminobenzoic acid units (R CH3, Cl,
thylammonium picrate. The conformations of the peptides in
solution have been determined by one- and two-dimensional
=
CH2OCH3, OCH3, COOCH3) and residues of the natural am- NMR techniques and FT-IR spectroscopy. It was found that
ino acid proline has been prepared and their ion affinities
all the substituents prevent the peptides from adopting the
have been investigated. Whereas the unsubstituted parent necessary conformation for anion binding. For one receptor
compound (R = H) is able to bind cations through cation−π
(R = OCH3), the results have been corroborated by a crystal
interactions with the aromatic subunits, as well as anions structure determination. AM1 calculations have been used to
through hydrogen bonding with the peptide NH groups, the
introduction of substituents at the 4-positions of the aromatic
estimate the electrostatic potential surfaces of the substituted
aromatic subunits. The variation in the cation complex
rings results in complete loss of the anion affinity. The cation stabilities can be mainly attributed to the effects of the sub-
complex stabilities depend on the substituents and cover a
stituents on the solution conformations of the peptides. The
influence of the substituents on the electrostatic potentials of
the aromatic peptide subunits appears to be less important.
−1
−1
wide range from Ka = 140
M
for R = CH3 to Ka = 10800
M
for R = COOCH3 (Ka = 1260
M
−1 for R = H) with n-butyltrime-
formation with a well-defined cavity.[8,9] These receptors
form kinetically very stable complexes with suitable guests,
and it has been shown that complex formation is even pos-
sible in water.[10] Another receptor in which conformational
control is based on intramolecular hydrogen bonds has
been introduced by ourselves.[11]
Introduction
The biological activity of a protein is critically dependent
on the folding of the peptide chain. The chain folding pro-
cess itself is highly complex and is controlled by a large
number of cooperative interactions between the individual
In 1995, Ishida et al. described a cyclic peptide that acted
protein subunits,[1] the most important factors being elec- as a receptor for phosphomonoesters in [D6]DMSO,[12] but
trostatic or hydrophobic interactions, hydrogen bonding, in the same solvent the complexation of cations was not
and the steric demand of the various amino acid residues. observed. We recently described a similar cyclic peptide
In supramolecular chemistry, such interactions are utilized made up of -glutamic acid and 3-aminobenzoic acid res-
to control the conformations and binding properties of arti- idues, for which an association constant of 300 Ϫ1 with n-
ficial receptors.[2,3] For example, calixarenes are relatively butylammonium iodide in chloroform was determined.[13]
flexible and adopt a number of different conformations in Subsequently, we prepared the receptor 1d, based on a cyc-
solution, but by introducing bulky substituents on the lower lic peptide composed of alternating proline and 3-aminob-
rim of these macrocycles it is possible to stabilize a symmet- enzoic acid subunits.[14] Peptides of this general structure
rical, so-called cone conformation, which is best suited for bind cations, such as quaternary ammonium ions, through
substrate binding.[4Ϫ7] In this case, the conformational sta- cationϪπ interactions.[15,16] The cations are held in the shal-
bilization stems from the steric bulk of the substituents, low dish-shaped cavity of the peptide, the conformation of
which prevents ring inversion of the aromatic subunits. In which resembles the cone conformation of calixarenes. In-
the self-folding cavitands described by Rebek and co- terestingly, anions can also interact with these peptides, i.e.
workers, a seam of hydrogen bonds at the open end of these when they are able to form hydrogen bonds to the peptide
container molecules is used to stabilize a vase-like con- amide groups. Anion complexation results in the stabiliza-
tion of a receptor conformation where all the NH groups
point towards the center of the cavity. In the absence of
[a]
Institut für Organische Chemie und Makromolekulare Chemie,
suitable anions, the peptides are more flexible due to the
large number of rotatable bonds between their aromatic su-
bunits. Although the flexibility can be somewhat reduced
by using proline subunits as the natural amino acid, cyclo-
peptide 1d is not completely rigid as there is still conforma-
tional flexibility of the bonds around the secondary amide
groups.[14]
Heinrich-Heine-Universität,
Universitätsstrasse 1, 40225 Düsseldorf, Germany
Fax: (internat.) ϩ 49-(0)211/81-14788
E-mail: kubik@uni-duesseldorf.de
Max-Planck-Institut für Kohlenforschung,
Kaiser-Wilhelm-Platz 1, 45470 Mülheim/Ruhr, Germany
Supporting information for this article is available on the
WWW under http://www.wiley-vch.de/home/eurjoc or from
the author.
[b]
Eur. J. Org. Chem. 2001, 311Ϫ322
WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ193X/01/0102Ϫ0311 $ 17.50ϩ.50/0
311

S. Kubik, R. Goddard
FULL PAPER
We chose to examine five derivatives, specifically those
with methyl (1a), chloro (1b), methoxymethyl (1c), methoxy
(1e), and methoxycarbonyl substituents (1f). The latter four
contain an electronegative atom, i.e. oxygen or chlorine,
which can interact with an NH group through hydrogen
bonding. The conformational stabilization of 1f by hydro-
gen bonding between the methoxycarbonyl substituents and
the amide groups has already been demonstrated.[11] We ex-
pected similar, albeit probably weaker effects, in all the
other cyclopeptides except 1a, where only steric effects are
expected to play a role. We include the unsubstituted pep-
tide 1d[14] in this discussion for comparison purposes.
The syntheses of the peptides 1a, 1b, 1c, 1e, and 1f are
straightforward because, in most cases, the corresponding
3-aminobenzoic acid derivatives are commercially available.
1-Methyl aminoterephthalate and 3-amino-4-chlorobenzoic
acid were directly converted into the corresponding benzyl
esters by reaction with benzyl bromide/NaHCO3 in DMF
(Scheme 1). 3-Amino-4-methylbenzoic acid and 3-amino-4-
methoxybenzoic acid were converted into the benzyl esters
via their Boc-protected derivatives (Scheme 2). Benzyl 3-
amino-4-(methoxymethyl)benzoate was obtained from the
Boc-protected methyl 3-amino-4-methylbenzoate by a se-
quence of NBS bromination, nucleophilic substitution of
the bromine by methylate, and transesterification, followed
by cleavage of the Boc group (Scheme 3).
More recently, we have succeeded in suppressing amide
rotation through the introduction of methoxycarbonyl
groups at the 4-positions of the aromatic peptide subunits
(1f).[11] These groups form strong intramolecular hydrogen
bonds to neighboring NH groups, and lock the amides in a
fixed conformation.[17] The introduction of such substitu-
ents into the cyclopeptide leads to a significant reduction in
its flexibility, thereby resulting in improved cation binding
ability. An interesting consequence of this is that, because
the NH groups are locked in an orientation unsuitable for
interaction with anions, no anion complexation can be de-
tected.[11]
In this paper, we extend the series of new substituted cyc-
lic peptides, and study the influence of substituents on the
conformations and complexing abilities of these com-
pounds.
Scheme 1
Results and Discussion
Scheme 2
Preparation of Cyclopeptides 1a؊1f
Following our findings that methoxycarbonyl substitu-
All the benzyl esters were coupled with Boc-()-proline
ents have a significant effect on the conformation and ion using PyCloP as the coupling reagent. The dipeptides were
binding properties of the cyclic hexapeptide 1f,[11] we de- chain elongated to give the linear hexapeptides according to
cided to prepare and study a series of these cyclopeptides standard peptide synthesis protocol. The fully deprotected
with various substituents in the 4-position with the aim of hexapeptides were cyclized under high dilution conditions
gaining some insight into the nature of the influence of the and were typically obtained in yields in the range 20Ϫ50%
substituent type on the properties.
(Scheme 4).
Scheme 3
312
Eur. J. Org. Chem. 2001, 311Ϫ322

Fine Tuning of the Cation Affinity of Artificial Receptors Based on Cyclic Peptides
FULL PAPER
Scheme 4
Complexation of Various n-Butyltrimethylammonium Salts
by 1a؊1f
Complex formation between the cyclopeptides and cat-
ionic guests was followed both qualitatively and quantita-
tively by NMR spectroscopy. As representative guest mole-
cules, the picrate, iodide, and tosylate salts of the n-butyltri-
methylammonium cation (BTMAϩ) were chosen. The same
salts were also used in our investigations on the complexing
behavior of 1d[14] and 1f[11] and hence comparisons with
previous results can easily be made.
As in the case of cyclopeptide 1d,[14] the addition of
BTMAϩ salts to solutions of the substituted peptides in
CDCl3 resulted in significant upfield shifts of all the
BTMAϩ proton signals of up to ca. 0.6 ppm. This effect
has been explained in terms of the inclusion of the cationic
guest in the receptor cavity, which brings the guest protons
in close proximity to the aromatic subunits, where they ex-
perience a shielding.[15,16]
The stabilities of the BTMAϩ complexes of the various
peptides were quantitatively determined by means of NMR
titrations.[18,19] Increasing amounts of the host were added
to a solution of the BTMAϩ salt in CDCl3 and the shifts of
1
the guest proton resonances were followed in the H NMR
spectra. Stability constants were calculated from the re-
sulting saturation curves using a nonlinear least-squares fit-
ting method for 1:1 complexes. Representative titration
curves are depicted in Figure 1a. The 1:1 complex stoichi-
ometry for the complex between 1f and BTMAϩ picrate
was confirmed by using Job’s method of continuous vari-
ations (Figure 1b).[18,20] In view of the excellent fits between
the calculated and observed saturation curves in our titra-
tions, it seems reasonable to assume that the same stoichi-
Figure 1. (a) 1H NMR titration curves of complexes of peptides 1a
(stars), 1b (triangles), 1c (diamonds), 1e (circles), and 1f (squares)
with BTMAϩ picrate; (b) Job plot of the complex between 1f and
BTMAϩ picrate
ometry holds for all the complexes. The titrations were usu- could not be determined accurately by our method because
ally repeated several times, and the error limit of the of the low solubility of the cyclopeptides in CDCl3 at higher
1
method is well below 20%. Since the H NMR spectra of concentrations, so that in the case of BTMAϩ iodide with
the peptides are not significantly affected by varying the 1a only the general trend is indicated in the table.
concentration in the range 2.0Ϫ0.2 m, intermolecular as-
It is evident from these results that the various substitu-
sociation of the receptors under these conditions can be ents on the investigated cyclopeptides have a marked influ-
ruled out. For all the substituted cyclopeptides, the calcu- ence on cation complex formation. The stability constants
lated stability constants (Ka) proved to be independent of of the BTMAϩ picrate complexes vary over two orders of
the choice of the BTMAϩ proton which was followed dur- magnitude from ca. 100 Ϫ1 for 1a to ca. 10000 Ϫ1 for 1f.
ing the titrations. Consequently, only the results derived Furthermore, it is clear that whereas the stability constants
from the shift of the BTMAϩ N-methyl proton resonances for unsubstituted cyclopeptide 1d increase on going from
are summarized in Table 1. Stability constants below 50 Ϫ1 the picrate to the iodide and to the tosylate, for all the other
Eur. J. Org. Chem. 2001, 311Ϫ322
313

S. Kubik, R. Goddard
FULL PAPER
Table 1. BTMAϩ complex stabilities in CDCl3 at 298 K (Ka Bartsch et al. have reported that the extraction efficiency of
stability constant in Ϫ1, error limits for Ka Ͻ 20%; ∆δmax max-
certain crown ether salt complexes shows an inverse correla-
imum chemical shift; ∆GH ϭ Gibbs free energy of hydration of the
tion with the hydration enthalpy of the anion.[21] In accord-
ance with these findings, the stabilities of the BTMAϩ com-
plexes of the substituted cyclopeptides decrease with in-
creasing Gibbs free energy of hydration of the anion.[22] Ro-
elens et al. recently described a similar anion dependence
with regard to cationϪπ interactions between salts of qua-
ternary ammonium ions and cyclophane hosts.[23] They
concluded that the higher the charge density on the anion,
the stronger the charge polarization of the quaternary am-
monium cation, and the less available the cation becomes
for interactions with the host. Ion-pair aggregation thus
strongly influences the cation complex stabilities of artificial
anions in kJ·molϪ1
)
BTMAϩ picrate
BTMAϩ iodide
BTMAϩ tosylate
Ka
Ϫ∆δmax Ka
Ϫ∆δmax Ka
Ϫ∆δmax
1a
140
0.51
0.42
0.51
0.70
0.68
Ͻ 50
100
60
0.54
0.30
0.52
1b
290
110
230
0.34
0.60
1c
550
1260
2700
140
1d[14]
1e
21100 1.11
5050000 1.16
850
3310
283
0.73
0.59
340
740
318
0.56
0.54
1f[11]
10800 0.54
Ϫ∆GH 197
cyclic peptides they decrease. Only in the case of 1a (R ϭ receptors in non-polar solvents. In addition, the high
Me) is a slight deviation from this sequence observed. In stability of the cyclopeptide complexes with BTMAϩ pic-
the case of the unsubstituted cyclopeptide 1d, the increase rate may be a consequence of the so-called ‘‘picrate ef-
in cation complex stability with iodide as the counterion, fect’’.[24] This effect is due to πϪπ interactions between the
and even more so with tosylate, has been attributed to a picrate anions and aromatic subunits of an artificial recep-
simultaneous complexation of the anion and cation, which tor, which occur when cations are incorporated into the re-
is possible with these anions but not with picrate.[14] Since ceptor cavity. In agreement with these observations, we
none of the other cyclopeptides exhibit a similar Ka depend- found a small but characteristic upfield shift of the picrate
1
ence, we assume that these receptors are not capable of an- proton signals in the H NMR spectra of the BTMAϩ pic-
ion complexation. FT-IR spectroscopic investigations sup- rate complexes of our cyclopeptides.
port this assumption, since no dependence of the NH vibra-
tional frequencies of the various substituted cyclopeptides complexes of the substituted cyclopeptides on the nature of
on the nature of the anion could be detected. the anion is mainly influenced by ion-pair aggregation and
In summary, the dependence of the stabilities of cation
In the case of 1f (R ϭ CO2Me), a reduced anion affinity to a lesser extent by specific interactions of the anions with
has been attributed to hydrogen-bonding interactions be- the receptors. In the following discussion, we therefore fo-
tween the peptide NH groups and the adjacent methoxycar- cus on the effects of the aromatic substituents on the
bonyl substituents.[11] Such interactions are certainly not BTMAϩ picrate complex stabilities. Two major factors may
possible in 1a (R ϭ Me), and therefore other factors must be responsible for the observed dependence of complex
be invoked to account for the lack of anion affinity of this stability on the type of substituent. Firstly, there is the effect
cyclopeptide. The crystal structure of the iodide complex of of the substituents on the solution conformation of the pep-
the unsubstituted 1d shows that for anion complexation the tides, and secondly, the different electrostatic potentials of
aromatic subunits and the secondary amide groups have to the aromatic peptide subunits. Both factors will affect the
adopt an almost coplanar conformation, with the amide cationϪπ interactions with the guests.
carbonyl group pointing towards the proton in the 4-posi-
tion.[14] We assume that unfavorable steric interactions be-
tween the aromatic methyl substituent and the neighboring
Conformational Studies of 1a؊1f
The solution conformations of the substituted cyclopep-
carbonyl groups prevent 1a from adopting such a structure. tides were studied by NMR and FT-IR spectroscopy. The
This assumption is supported by the results of molecular 1H NMR spectra of 1c and 1f in CDCl3 are rather complex,
modeling calculations. When methyl groups are introduced which indicates that interconversion of the various cyclo-
into the structure of the anion complex of parent com- peptide conformers is slow on the NMR time scale, and
pound 1d, the secondary amide groups tilt away from the hence that these peptides are conformationally relatively ri-
center of the cavity upon energy minimization and possible gid. The reduced flexibility is probably due to hydrogen
anion interactions are lost. Clearly, the steric effects of suit- bonds between the amide NH groups and the oxygen atoms
ably positioned substituents on the aromatic rings of 1d are of the corresponding substituents, which would prevent
sufficient to eliminate its anion affinity. Only the slight devi- amide rotation. This assumption is corroborated by the FT-
ation of the dependence of cation complex stability on the IR spectroscopic investigations described below.
type of anion from the order picrate Ͼ iodide Ͼ tosylate
The remaining substituted cyclopeptides exhibit NMR
may account for the weak anion complexation by 1a. With spectra consistent with averaged C3-symmetric conforma-
all the other substituted cyclopeptides, steric effects of the tions, and hence their solution conformations can easily be
substituents certainly also contribute to the loss of anion investigated by NOESY NMR. The most important indica-
affinity of these receptors.
tions of the preferred conformation are the NOEs involving
The fact that the stability constants of the complexes of the secondary amide NH groups. In peptide 1a (R ϭ Me),
the substituted cyclopeptides decrease in the order picrate the NH protons exhibit NOEs with H(α) of the adjacent
Ͼ iodide Ͼ tosylate probably stems from various factors. proline residues, as well as with H(2) of the aromatic rings
314
Eur. J. Org. Chem. 2001, 311Ϫ322

Fine Tuning of the Cation Affinity of Artificial Receptors Based on Cyclic Peptides
FULL PAPER
1e crystallizes from methanol with seven solvent molecules
per peptide unit. In spite of the appreciable polarity of this
solvent, no interactions of the NH protons with solvent mo-
lecules are apparent in the crystal. Indeed, all three peptide
NH groups are oriented towards the adjacent methoxy sub-
stituents and are only involved in hydrogen bonding to the
oxygen atoms of these substituents. As a result, the overall
peptide conformation is highly symmetrical. Similar con-
formational stabilization of peptides by aromatic methoxy
substituents has recently been demonstrated by Nowick et
al.[25] as well as by Gong et al.[26,27]
Hydrogen-bonding interactions in solution between the
Cl and NH groups in 1b and between the OCH3 and NH
groups in 1e are also evident from the FT-IR spectra of
these peptides. The results of the FT-IR spectroscopic in-
vestigations on all the cyclopeptides are summarized in
Table 2. In the spectra of the unsubstituted cyclopeptide 1d,
the weak bands at 3425 cmϪ1 and 3408 cmϪ1 have been
assigned to the vibrations of free NH groups.[14] Another
weak band appears in the same spectral region at 3268
cmϪ1, which can be attributed to NH groups involved in
intramolecular hydrogen bonds with the carbonyl groups of
the adjacent tertiary amides.[28] This interaction does not
appear to be strong. No band due to free NH groups is
visible in the FT-IR spectra of 1b or 1e and the sharp,
strong bands at 3410 cmϪ1 for 1b and 3414 cmϪ1 for 1e
can be attributed to the respective NH···Cl and NH···OCH3
interactions. The second band at lower frequency is much
stronger in the FT-IR spectra of these cyclopeptides than in
the case of 1d. Clearly, other hydrogen-bonding interactions
besides those involving the aromatic substituents also occur.
Figure 2. Intramolecular NOEs in the NOESY NMR spectra of
1a (a), 1b (b), and 1e (c) in CDCl3
(Figure 2a). A weaker NOE between NH and the aromatic
methyl group is also in evidence. Such multiple effects can
be accounted for in terms of conformational flexibility of
the secondary amides that results from the NH groups be-
ing oriented either towards or away from the aromatic
methyl group. In contrast, in the NOESY NMR spectra of
peptides 1b (R ϭ Cl) and 1e (R ϭ OCH3), cross-peaks are
only visible between the NH groups and the proline H(α)
(Figures 2b and 2c).
In these peptides, the secondary amide moieties are thus
preferentially oriented towards the corresponding aromatic
substituents, possibly because of hydrogen-bonding interac-
tions with the oxygen or chlorine atom. In the case of 1e
(R ϭ OCH3), this observation could be confirmed by a
crystal structure determination of the compound (Figure 3).
Table 2. NH and CϭO vibration bands in the FT-IR spectra of
cyclopeptides 1aϪ1f (c ϭ 2 m in 1% [D6]DMSO/CDCl3) (s ϭ
strong, w ϭ weak, b ϭ broad)
ν(NϪH)
ν(CϭO)tert. amide
1a
1b
1c
1d
1e
1f
3430 (w)/3276 (b, s)
3410 (s)/3271 (b)
1620/1608 (shoulder)
1618 (b)
3340 (b)/3282 (b)
(3425/3408) (w)/3268 (w)
3414 (s)/3314 (b)
1624/1617 (shoulder)
1622
1619/1613 (shoulder)
1635/1619
3305 (b)/3260 (b)
Since interaction of the NH protons with the methyl sub-
stituent in 1a is not possible, a weak band due to the free
NH groups is visible at 3430 cmϪ1 in the spectrum of
this compound. Again, the second band at 3276 cmϪ1
is much more intense than in the spectrum of the parent
compound 1d and presumably can also be assigned to
NH···(OϭC)tert. amide interactions (Figure 4).
Figure 3. Molecular structure of cyclopeptide 1e·7MeOH viewed
perpendicular to a plane through the amide N atoms N1, N3, and
N5 showing the near C3 point symmetry of the cyclopeptide; the
following groups of atoms are approximately coplanar: O1, C1, N1,
˚
C35, C34, O9, C39 (r.m.s. deviation 0.04 A); O3, C13, N3, C11,
˚
C10, O7, C37 (r.m.s. deviation 0.07 A); O5, C25, N5, C23, C22,
Figure 4. Possible hydrogen bonds between two aromatic subunits
of the cyclopeptides in a peptide chain fragment
˚
O8, C38 (r.m.s. deviation 0.07 A)
Eur. J. Org. Chem. 2001, 311Ϫ322
315

S. Kubik, R. Goddard
FULL PAPER
Because of the steric effects of the methyl groups, the
The FT-IR spectroscopic results clearly indicate a shift
conformational equilibrium of 1a seems to be significantly of the conformational equilibria of the substituted cyclo-
shifted towards conformers in which the NH groups are peptides towards conformations with NH···(OϭC)tert. amide
involved in hydrogen bonds. Peptides 1c (R ϭ CH2OCH3) hydrogen bonds. The conformational shift is, however, most
and 1f (R ϭ CO2CH3) possess two strong, broad bands in marked in the case of 1a, where only steric effects of the
the NH vibration region of their FT-IR spectra. These may substituent are important, and it is not so pronounced
be attributed to both possible interactions of the NH when the substituents are themselves able to form hydrogen
groups, i.e. one with the aromatic substituent and the other bonds with the secondary amides as in 1c, 1e, and 1f.
The addition of 1 equiv. of BTMAϩ picrate to a solution
of 1e (R ϭ OCH3) in CDCl3 was found to have an interest-
ing effect on the FT-IR spectrum. Thus, the NH band at
3314 cmϪ1 almost completely disappeared, while the CO
band at 1619 cmϪ1 intensified at the expense of that at
1613 cmϪ1 (Figure 5).
with the carbonyl groups on the adjacent proline subunits.
If the bands at ca. 3270 cmϪ1 in the spectra of the substi-
tuted peptides do indeed represent NH groups that are in-
volved in intramolecular hydrogen bonds with the tertiary
amides, then an effect on the vibrational frequencies of the
corresponding carbonyl groups should also be visible.
Table 2 shows that this is indeed the case. In the unsubsti-
Complex formation with the cation and anion clearly re-
tuted cyclopeptide 1d, the carbonyl group of the tertiary sults in a loss of the NH···(OϭC)tert. amide hydrogen bonds.
amide gives rise to a single band at 1622 cmϪ1, which indic- The resulting conformational reorganization may slightly
ates that intramolecular hydrogen bonds between adjacent increase the size of the receptor cavity so that it can better
amino acid residues are either absent or at least very weak. accommodate the guests. The NH groups are nevertheless
In contrast, peptides 1a, 1c, 1e, and 1f exhibit split bands of still held in the previous orientation through the interaction
their tertiary amide carbonyl vibrations, with a maximum at with the methoxy groups. A similar effect was also observed
ca. 1620 cmϪ1 (1635 cmϪ1 in the case of 1f) and another upon addition of BTMAϩ picrate to solutions of 1c and 1f.
one significantly shifted to lower frequencies. This second The requirement for intramolecular NH···(OϭC)tert. amide
band can be attributed to carbonyl groups that are involved hydrogen bond breaking for optimal cation complexation
in hydrogen-bonding interactions. In the case of peptide 1b, would explain the low affinity of 1a towards this guest.
the carbonyl band is rather broad, hence an assignment of Table 2 shows that in 1a these hydrogen bonds are the
individual bands is difficult.
strongest among all the cyclopeptides. It would appear that
Figure 5. NH and CϭO vibration regions of the FT-IR spectrum of 1e in 1% [D6]DMSO/CDCl3 (c ϭ 2 m) (a) before and (b) after the
addition of 1 equiv. of BTMAϩ picrate
316
Eur. J. Org. Chem. 2001, 311Ϫ322

Fine Tuning of the Cation Affinity of Artificial Receptors Based on Cyclic Peptides
FULL PAPER
complex formation between 1a and BTMAϩ cannot com- been explained in detail by Klärner et al. using the example
pensate for the energy needed to cleave these bonds.
of their pincette receptors.[30] In the cyclopeptides, however,
the cavity walls are lined by at least three, and, in the case
of 1f, six carbonyl groups, which further contribute to the
negative potential inside the cavities. Interestingly, we have
Calculations of the Electrostatic Potential Surfaces of
1a؊1f
The results described above demonstrate the pronounced not observed any effect of cation complexation on the vi-
influence of the substituents on the conformational equilib- brational frequencies of these carbonyl groups. We therefore
rium of the cyclopeptides. Interaction of the substituent believe the interactions responsible for cation binding do
with the NH groups provides an adequate explanation for not involve hydrogen bonding but rather are purely electro-
the loss of the anion affinity of all the substituted cyclopep- static.[31] The more positive EPS (blue) on the outside of
tides, the low cation affinity of 1a, and also the high affinity 1d, as compared with 1e and 1f, is due to the different ori-
of 1f. We have assumed all along, however, that cationϪπ entations of the secondary amide groups in this peptide.
interactions between the cations and the aromatic peptide
The AM1 calculations thus provide us with valuable in-
subunits are not affected by the substituents. This is cer- formation concerning the interaction of the cyclopeptides
tainly not the case since a substituent may increase or de- with the cations. They explain the electronic effects of the
crease the electrostatic potentials of the aromatic systems aromatic substituents on cation complexation, and they
and thereby influence cation affinity. It is directly evident also show where the majority of the negative potential re-
from Table 1, however, that the dependence of cation com- sides. Although the substituents have some influence on the
plex stability on the type of aromatic substituent cannot be electrostatic potential inside the cyclopeptides, their most
adequately explained in terms of electronic effects alone. If important contribution to cation affinity is the effect they
this were to be the case, one would expect, for example, a have on the peptide conformation in solution.
much lower affinity of peptide 1f towards cations because a
methoxycarbonyl substituent should reduce the electrostatic
potential of the aromatic ring more than, say, a methoxy
group.
Conclusions
Dougherty et al. have shown that the electronic effects of
The nature of the substituent in the 4-position of the aro-
aromatic systems on cationϪπ interactions can be estimated matic peptide subunits has been shown to have a significant
by AM1 calculations of the electrostatic potential surfaces effect on the conformational and receptor properties of the
(EPS) of the aromatics.[29] We carried out similar calcula- cyclic hexapeptides 1aϪ1f. Whereas the parent compound
tions on 3-(acetylamino)-N-methylbenzamides with the can bind both anions and cations, the introduction of a
same substituents in the 4-position as in the cyclopeptides substituent results in a complete loss of anion affinity since
1aϪ1f. The results show that the incorporation of the the NH groups are prevented from pointing towards the
methyl substituent has little effect on the EPS of the aro- center of the cavity, a necessary prerequisite for anion bind-
matic rings as compared with the unsubstituted parent ing. In the case of a methyl substituent, the reason is steric,
compound. Chloro and methoxycarbonyl substituents de- while for the other compounds the NH groups are addition-
crease the EPS of the aromatic rings somewhat. Methoxy ally engaged in hydrogen-bonding interactions with the sub-
and, to an even greater extent, methoxymethyl groups in- stituents. In contrast, the cation affinities show a large sub-
crease the EPS of the aromatic rings. Although the poten- stituent dependence. The cation complex stability is lowest
tial surfaces alone cannot account for the sequence of cat- for the methyl substituent. This is because the conforma-
ion complex stabilities of the peptides 1aϪ1f, they do ex- tional change required for cation binding requires the rup-
plain the much lower cation affinity of 1b in comparison ture of intramolecular NH···(OϭC)tert. amide hydrogen
with 1e, since the solution conformations of both of these bonds. In the other compounds, such prior bond breaking
peptides are quite similar. Analogously, the higher cation is less important since the NH groups are more involved in
affinity of 1e as compared with 1d may be partly due to the hydrogen-bonding interactions with the substituents. Sub-
electronic effects of the methoxy substituents.
stituents that interact most strongly with the NH groups
In order to see whether these effects might be mirrored stabilize a peptide conformation with a high cation affinity.
in the cyclopeptides themselves, we carried out similar AM1 Although the substituents also affect the electrostatic po-
calculations on the cyclopeptides for which crystal struc- tential inside the cyclopeptides, their most important con-
tures have been determined. For 1d, the crystal structure of tribution to cation affinity is the effect they have on the
the N-quinuclidinium iodide complex[14] was used as a peptide conformation in solution. The ease with which
model, and for 1f that of the acetone complex.[11] In both structural variation of the aromatic subunits of the peptide
cases, the guest molecules were removed. Similarly, the can be introduced, coupled with the effects of substituents
structure depicted in Figure 3 was used for the calculations as investigated herein, enables a facile fine tuning of the
on peptide 1e. Figure 6 shows the results.
affinity of receptors of this type towards cationic guests.
The views clearly illustrate that the majority of the nega- 1e and 1f are particularly promising candidates for further
tive potential (red) resides on the inside of the cavities as investigation as they are not only suitably pre-organized for
the concave surface curvature inside the receptor cavity of guest binding, but additional binding sites can be intro-
the cyclopeptides strongly enhances the EPS. This effect has duced on the methoxy and methoxycarbonyl groups
Eur. J. Org. Chem. 2001, 311Ϫ322
317

S. Kubik, R. Goddard
FULL PAPER
Figure 6. AM1 electrostatic potential surfaces of cyclopeptides 1d (a), 1e (b), and 1f (c)
(Molecular Simulations Inc.). For molecular modeling, the Drei-
ding force field 2.21 was used.[32] Geometry optimization of the 3-
(acetylamino)-N-methylbenzamide derivatives was carried out at
the AM1 level. Electrostatic potential surfaces of the aromatic rings
were generated with the program Spartan (Wavefunction, Inc.) by
mapping AM1 electrostatic potentials onto surfaces of molecular
through ether or ester linkages. We will report on our pro-
gress in this area in due course.
Experimental Section
˚
electron density (0.002 electron/A) followed by color-coding. Over
General Methods: Analyses were carried out as follows: Melting
points: Büchi 510 apparatus. Ϫ Optical rotations: PerkinϪElmer
241 MC digital polarimeter (d ϭ 10 cm). Ϫ NMR: Varian VXR
300 or Bruker DRX 500 equipped with an automatic sampler (the
chemical shifts in the NMR titrations were referenced to the CHCl3
signal at δ ϭ 7.27 as an internal standard). Ϫ FT-IR: Bruker Vector
22 FT-IR spectrometer; cuvette d ϭ 5 mm (NaCl). Ϫ Elemental
analysis: Pharmaceutical Institute of the Heinrich-Heine-Univer-
sity, Düsseldorf. Ϫ Mass spectrometry: Finnigan INCOS 50. Ϫ
Chromatography: ICN silica gel 32Ϫ63 µm (ICN Biomedicals),
Merck LiChroprep RP-8 (40Ϫ63 µm) prepacked column, size B
(310Ϫ25). The following abbreviations are used: TsOH ϭ 4-tolu-
enesulfonic acid, Bn ϭ benzyl, Boc ϭ tert-butyloxycarbonyl,
DIEA ϭ N-ethyldiisopropylamine, TFA ϭ trifluoroacetic acid, Py-
CloP ϭ chlorotripyrrolidinophosphonium hexafluorophosphate,
TBTU ϭ O-(1H-benzotriazol-1-yl)-N,N,NЈ,NЈ-tetramethyluronium
tetrafluoroborate, Pro ϭ proline, AB ϭ 3-aminobenzoic acid,
NBS ϭ N-bromosuccinimide.
all the surfaces, the potential energy values range from ϩ40 to Ϫ20
kcal/mol, with red signifying a value greater than or equal to the
maximum in negative potential and blue signifying a value greater
than or equal to the maximum in positive potential.
Materials: All solvents were dried according to standard proce-
dures prior to use. DMF p.A. was purchased from Fluka and was
used without further purification. PyCloP was prepared according
to the literature procedure.[33] TBTU, 3-amino-4-chlorobenzoic
acid, methyl 3-amino-4-methylbenzoate, 3-amino-4-methoxyben-
zoic acid, and 1-methyl aminoterephthalate are commercially avail-
able. The synthesis of cyclo-[()-Pro-AB]3 (1d) has been described
elsewhere.[14]
Job Plot: Equimolar solutions (1 m) of n-butyltrimethylammon-
ium picrate (BTMAϩ picrate) and cyclopeptide 1f in 1%
1
[D6]DMSO/CDCl3 were prepared and mixed in various ratios. H
NMR spectra of the resulting solutions were recorded, and the
change in chemical shift of the N-methyl proton signals of the guest
was analyzed.[18,20]
Computational Methods: Most of the calculations were performed
with a Silicon Graphics workstation using the program CERIUS2
318
Eur. J. Org. Chem. 2001, 311Ϫ322

Fine Tuning of the Cation Affinity of Artificial Receptors Based on Cyclic Peptides
FULL PAPER
Host؊Guest Titrations: Stock solutions of the guest (0.2 µmol/
100 µL) in 1% [D6]DMSO/CDCl3 and the cyclopeptide (1a, 1b:
4 µmol/800 µL; 1c: 3 µmol/800 µL; 1e, 1f: 2 µmol/800 µL) in CDCl3
were prepared. In total, 11 NMR tubes were set up by adding in-
creasing amounts of the host solution (0Ϫ800 µL) to 100 µL ali-
quots of the guest solution. All samples were made up to a volume
was suspended in DMF (25 mL). Benzyl bromide (2.37 mL,
20 mmol) was then added and the reaction mixture was stirred for
48 h at room temperature. After the addition of 10% aqueous
Na2CO3 (75 mL), the mixture was extracted three times with ethyl
acetate. The combined organic layers were washed with water and
dried. The solvent was removed in vacuo and the residue was
recrystallized from hexane/ethyl acetate. Yield 2.90 g (85%); m.p.
136Ϫ138 °C. Ϫ 1H NMR (300 MHz, CDCl3, 25 °C, TMS): δ ϭ
1.51 (s, 9 H, tBuCH3), 2.28 (s, 3 H, PhCH3), 5.35 (s, 2 H, PhCH2),
6.32 (br. s, 1 H, NH), 7.20 [d, 3J ϭ 8.1 Hz, 1 H, ABH(5)], 7.37 (br.
1
of 1 mL with CDCl3 and the respective H NMR spectra were re-
corded. The chemical shifts of prominent guest protons were plot-
ted against the host concentration. From the resulting saturation
curves, Ka and ∆δmax were calculated by a nonlinear least-squares
fitting method for 1:1 complexes using the SIGMA Plot 3.0 (Jandel
Scientific) software package.[18,19]
3
4
m, 5 H, PhH), 7.71 [dd, J ϭ 7.9 Hz, J ϭ 1.8 Hz, 1 H, ABH(6)],
8.43 [s, 1 H, ABH(2)]. Ϫ C20H23NO4 (341.4): calcd. C 70.36, H
6.79, N 4.10; found C 70.36, H 6.81, N 4.17.
General Procedure for the Cleavage of Benzyl Esters: The ester was
dissolved in methanol (50 mL/mmol). After the addition of 10%
Pd/C (100 mg) [in the case of compounds with a chloro substituent
on the aromatic moieties, Raney nickel (1 g) was used], the reaction
mixture was subjected to hydrogenation at 1 atm for about 2 h.
That the reaction had reached completion was checked by TLC.
The catalyst was then filtered off by passage through a layer of
Celite and washed with methanol. The combined filtrate and wash-
ings were concentrated to dryness in vacuo.
Benzyl 3-Amino-4-chlorobenzoate (3b): A mixture of 3-amino-4-
chlorobenzoic acid (8.58 g, 50 mmol) and NaHCO3 (5.04 g,
60 mmol) was suspended in DMF (125 mL). Benzyl bromide
(7.12 mL, 60 mmol) was added and the reaction mixture was stirred
for 72 h at room temperature. After the addition of 10% aqueous
Na2CO3 (375 mL), the resulting mixture was extracted three times
with ethyl acetate. The combined organic layers were washed with
water and dried. The solvent was removed in vacuo and the product
was isolated from the residue by chromatographic workup (hexane/
ethyl acetate, 6:1). It was finally recrystallized from hexane/ethyl
acetate. Yield 2.87 g (22%); m.p. 74Ϫ75 °C. Ϫ 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ ϭ 4.10 (br. s, 2 H, NH), 5.32 (s, 2 H,
PhCH2), 7.28 [d, 3J ϭ 8.3 Hz, 1 H, ABH(5)], 7.37 [br. m, 6 H, PhH
General Procedure for the Cleavage of tert-Butyloxycarbonyl
Groups. ؊ Method A: The carbamate was dissolved in CH2Cl2
(5 mL). The resulting solution was cooled in an ice bath, and then
trifluoroacetic acid (5 mL) was added dropwise. The reaction mix-
ture was stirred for 1.5 h at 0Ϫ5 °C and then the solvent was evap-
orated in vacuo. The residue was redissolved in ethyl acetate and
this solution was extracted twice with 10% Na2CO3 solution and
three times with water. The organic layer was dried, 1  HCl was
added (1 mL/mmol), and then it was concentrated to dryness in
vacuo. The resulting crude product was redissolved in CH2Cl2, the
solvent was evaporated, and the residue was dried in vacuo. It was
used without further purification. Ϫ Method B: The carbamate was
suspended in 1,4-dioxane (20 mL). This suspension was cooled in
4
ϩ ABH(6)], 7.46 [d, J ϭ 2.0 Hz, 1 H, ABH(2)]. Ϫ C14H12ClNO2
(261.7): calcd. C 64.25, H 4.62, N 5.35; found C 64.42, H 4.71,
N 5.34.
Methyl 3-[(tert-Butyloxycarbonyl)amino]-4-methylbenzoate (6c):
Methyl 3-amino-4-methylbenzoate (5.28 g, 32 mmol) and di-tert-
butyl dicarbonate (7.41 g, 34 mmol) were dissolved in dry THF
(100 mL). The resulting solution was heated to reflux for 48 h.
After cooling, it was concentrated to dryness in vacuo. The residue
an ice bath and a 6  solution of HCl in 1,4-dioxane (40 mL) was was stirred with hexane. The undissolved material was collected by
added dropwise. The reaction mixture was stirred for 2 h at 0Ϫ5 °C
and then concentrated to dryness in vacuo.
filtration and recrystallized from hexane/ethyl acetate. Yield 7.41 g
(87%); m.p. 130Ϫ132 °C. Ϫ 1H NMR (300 MHz, CDCl3, 25 °C,
TMS): δ ϭ 1.53 (s, 9 H, tBuCH3), 2.29 (s, 3 H, PhCH3), 3.89 (s, 3
H, OCH3), 6.45 (br. s, 1 H, NH), 7.20 [d, 3J ϭ 7.9 Hz, 1 H,
ABH(5)], 7.68 [dd, J ϭ 7.9 Hz, J ϭ 1.7 Hz, 1 H, ABH(6)], 8.44
[s, 1 H, ABH(2)]. Ϫ C14H19NO4 (265.3): calcd. C 63.38, H 7.22, N
5.28; found C 63.50, H 7.26, N 5.47.
3-[(tert-Butyloxycarbonyl)amino]-4-methylbenzoic Acid (4a): 3-Am-
ino-4-methylbenzoic acid (7.55 g, 50 mmol) was suspended in 1,4-
dioxane/water (2:1; 150 mL). Triethylamine (6.96 mL, 50 mmol)
was added and the mixture was cooled in an ice bath. A solution of
di-tert-butyl dicarbonate (11.99 g, 55 mmol) in 1,4-dioxane (50 mL)
was then added dropwise over a period of 30 min under stirring.
3
4
Methyl 4-(Bromomethyl)-3-[(tert-butyloxycarbonyl)amino]benzoate
Stirring was continued for 2 h at ca. 5 °C and then overnight at (5c): Compound 6c (6.63 g, 25 mmol) was dissolved in dry CCl4
room temperature. Thereafter, the dioxane was removed in vacuo,
and the remaining aqueous solution was diluted with water and
10% aqueous Na2CO3 solution (50 mL). The resulting mixture was
extracted three times with diethyl ether. The organic layers were
discarded. The aqueous layer was then acidified with 20% aqueous
(100 mL) under gentle heating. NBS (4.89 g, 27.5 mmol) was added
and the reaction mixture was heated to reflux for 1 h under irradi-
ation by a UV lamp (500 W). After cooling, the succinimide pro-
duced was filtered off. The filtrate was concentrated in vacuo and
the residue was recrystallized from hexane/ethyl acetate. The hot
KHSO4 and extracted three times with ethyl acetate. The combined solution was filtered through a glass frit (P4) to remove insoluble
1
organic layers were washed with water, dried, and concentrated to
dryness in vacuo. The residue was triturated with hexane, collected
material. Yield 6.39 g (74%); m.p. 114Ϫ116 °C (dec.). Ϫ H NMR
(300 MHz, CDCl3, 25 °C, TMS): δ ϭ 1.55 (s, 9 H, tBuCH3), 3.91
by filtration, and dried. It was used for the next step without fur-
(s, 3 H, OCH3), 4.50 (s, 2 H, PhCH2Br), 6.73 (br. s, 1 H, NH), 7.36
1
3
ther purification. Yield 10.92 g (87%); m.p. Ͼ 190 °C (dec.). Ϫ H [d, 3J ϭ 8.0 Hz, 1 H, ABH(5)], 7.74 [dd, J ϭ 8.0 Hz, 4J ϭ 1.7 Hz,
NMR (300 MHz, [D6]DMSO, 25 °C, TMS): δ ϭ 1.48 (s, 9 H, 1 H, ABH(6)], 8.48 [s, 1 H, ABH(2)].
tBuCH3), 2.26 (s, 3 H, PhCH3), 7.28 [d, 3J ϭ 8.1 Hz, 1 H, ABH(5)],
Methyl 3-[(tert-Butyloxycarbonyl)amino]-4-(methoxymethyl)benzo-
ate (4c): Compound 5c (5.16 g, 15 mmol) was dissolved in dry
methanol (200 mL). After the addition of sodium methoxide
(1.62 g, 30 mmol), the reaction mixture was heated to reflux for
4 h. After cooling, diethyl ether (750 mL) was added, and the mix-
7.60 [dd, 3J ϭ 7.8 Hz, 4J ϭ 1.8 Hz, 1 H, ABH(6)], 7.99 [d, 4J ϭ
1.5 Hz, 1 H, ABH(2)], 8.69 (br. s, 1 H, NH), 12.86 (br. s, 1 H,
COOH).
Benzyl 3-[(tert-Butyloxycarbonyl)amino]-4-methylbenzoate (3a): A
mixture of 4a (2.51 g, 10 mmol) and NaHCO3 (1.68 g, 20 mmol) ture was washed with water, dried, and concentrated in vacuo. The
Eur. J. Org. Chem. 2001, 311Ϫ322 319

S. Kubik, R. Goddard
FULL PAPER
product was isolated from the residue by chromatographic workup
(hexane/ethyl acetate, 4:1). It was finally recrystallized from hexane.
Yield 3.25 g (73%); m.p. 79Ϫ80 °C. Ϫ 1H NMR (300 MHz, CDCl3,
(75 mL), the mixture was extracted three times with ethyl acetate.
The combined organic layers were washed with water and dried.
The solvent was removed in vacuo, and the product was isolated
25 °C, TMS): δ ϭ 1.54 (s, 9 H, tBuCH3), 3.36 (s, 3 H, CH2OCH3), from the residue by chromatographic workup (hexane/ethyl acetate,
3.91 (s, 3 H, COOCH3), 4.54 (s, 2 H, CH2OCH3), 7.22 [d, 3J ϭ 3:1). It was dried in vacuo. Yield 3.32 g (93%); oil. Ϫ 1H NMR
7.8 Hz, 1 H, ABH(5)], 7.68 [dd, 3J ϭ 7.9 Hz, 4J ϭ 1.7 Hz, 1 H,
ABH(6)], 7.77 (br. s, 1 H, NH), 8.65 [s, 1 H, ABH(2)]. Ϫ
C15H21NO5 (295.3): calcd. C 61.00, H 7.17, N 4.74; found C 61.05,
H 7.29, N 4.63.
(300 MHz, CDCl3, 25 °C, TMS): δ ϭ 1.52 (s, 9 H, tBuCH3), 3.90
(s, 3 H, OCH3), 5.35 (s, 2 H, PhCH2), 6.85 [d, J ϭ 8.6 Hz, 1 H,
3
ABH(5)], 7.04 (br. s, 1 H, NH), 7.37 (br. m, 5 H, PhH), 7.75 [dd,
3
4
1 H, J ϭ 8.5 Hz, J ϭ 2.1 Hz, ABH(6)], 8.75 [s, 1 H, ABH(2)]. Ϫ
C20H23NO5 (357.4): calcd. C 67.21, H 6.49, N 3.92; found C 67.28,
H 6.54, N 3.85.
Benzyl 3-[(tert-Butyloxycarbonyl)amino]-4-(methoxymethyl)benzo-
ate (3c): Compound 4c (2.95 g, 10 mmol) was stirred in a mixture
of 1,4-dioxane/1  aqueous NaOH (1:1; 100 mL) for 4 h at room
temperature. Thereafter, the reaction mixture was concentrated to
half of its original volume in vacuo. Water was added, and the
resulting solution was extracted twice with ethyl acetate. The or-
ganic layers were discarded. The aqueous layer was acidified with
20% aqueous KHSO4 and then extracted three times with ethyl
acetate. The combined organic layers were washed with water,
dried, and concentrated to dryness in vacuo. The residue was tritur-
ated with hexane, collected by filtration, and dried. The crude pro-
duct and NaHCO3 (1.68 g, 20 mmol) were suspended in DMF
(25 mL). Benzyl bromide (2.37 mL, 20 mmol) was added, and the
reaction mixture was stirred for 48 h at room temperature. After
the addition of 10% aqueous Na2CO3 (75 mL), the mixture was
extracted three times with ethyl acetate. The combined organic
layers were washed with water and dried. The solvent was removed
in vacuo, and the product was isolated from the residue by chroma-
tographic workup (hexane/ethyl acetate, 6:1). It was finally recrys-
tallized from hexane. Yield 3.13 g (84%); m.p. 78Ϫ80 °C. Ϫ 1H
NMR (300 MHz, CDCl3, 25 °C, TMS): δ ϭ 1.52 (s, 9 H, tBuCH3),
3.34 (s, 3 H, CH2OCH3), 4.52 (s, 2 H, CH2OCH3), 5.37 (s, 2 H,
PhCH2), 7.20 [d, 3J ϭ 7.8 Hz, 1 H, ABH(5)], 7.37 (br. m, 5 H,
1-Methyl-4-benzyl 2-Aminoterephthalate (3f): A mixture of 1-
methyl 2-aminoterephthalate (3.90 g, 20 mmol) and NaHCO3
(2.02 g, 24 mmol) was suspended in DMF (50 mL). Benzyl bromide
(2.85 mL, 24 mmol) was added and the reaction mixture was stirred
for 72 h at room temperature. After the addition of 10% aqueous
Na2CO3 (150 mL), the mixture was extracted three times with ethyl
acetate. The combined organic layers were washed with water and
dried. The solvent was removed in vacuo, and the product was
isolated from the residue by chromatographic workup (hexane/ethyl
acetate, 6:1). It solidified on drying. Yield 3.58 g (63%); m.p. 79Ϫ80
°C. Ϫ 1H NMR (300 MHz, CDCl3, 25 °C, TMS): δ ϭ 3.88 (s, 3
H, OCH3), 5.34 (s, 2 H, PhCH2), 5.80 (br. s, 2 H, NH), 7.28 [dd,
4
3J ϭ 8.4 Hz, J ϭ 1.7 Hz, 1 H, ABH(6)], 7.39 [br. m, 6 H, PhH ϩ
ABH(2)], 7.89 [d, 3J ϭ 8.3 Hz, 1 H, ABH(5)]. Ϫ C16H15NO4
(285.3): calcd. C 67.36, H 5.30, N 4.91; found C 67.53, H 5.30,
N 4.71.
Dipeptides: Prior to coupling, 3a, 3c, and 3e were deprotected at
the terminal amino group according to Method A. The aromatic
amine (7.50 mmol), Boc-()-proline (1.2 equiv., 9.00 mmol, 1.94 g),
and PyCloP (1.2 equiv., 9.00 mmol, 3.79 g) were dissolved in
CH2Cl2 (150 mL). At room temperature, DIEA (2.4 equiv.,
18.0 mmol, 3.13 mL) was added dropwise, and then the reaction
mixture was stirred overnight. The solvent was subsequently evap-
orated in vacuo, and the product was isolated from the residue by
chromatographic workup. In the case of dipeptides 2b and 2f, 1.5
equiv. of Boc-()-proline, 1.5 equiv. of PyCloP, and 3.0 equiv. of
DIEA were used and the reaction time was extended to 7 d.
3
PhH), 7.70 [dd, 1 H, J ϭ 7.8 Hz, 4J ϭ 1.7 Hz, ABH(6)], 7.72 [s, 1
H, ABH(2)], 8.67 (br. s, 1 H, NH). Ϫ C21H25NO5 (371.4): calcd. C
67.91, H 6.78, N 3.77; found C 67.74, H 6.83, N 3.80.
3-[(tert-Butyloxycarbonyl)amino]-4-methoxybenzoic Acid (4e): 3-
Amino-4-methoxybenzoic acid (3.34 g, 20 mmol) was suspended in
1,4-dioxane/water (2:1; 60 mL). After the addition of triethylamine
(2.78 mL, 20 mmol), the mixture was cooled in an ice bath. A solu-
tion of di-tert-butyl dicarbonate (5.23 g, 24 mmol) in 1,4-dioxane
(20 mL) was then added dropwise over a period of 30 min under
stirring. Stirring was continued for 2 h at ca. 5 °C and then over-
night at room temperature. The dioxane was subsequently removed
in vacuo and the remaining aqueous solution was diluted with
water and 10% aqueous Na2CO3 (20 mL). The resulting mixture
was extracted three times with diethyl ether. The organic layers
were discarded. The aqueous layer was acidified with 20% aqueous
KHSO4 and extracted three times with ethyl acetate. The combined
organic layers were washed with water, dried, and concentrated to
dryness in vacuo. The residue was triturated with hexane, collected
by filtration, and dried. It was used for the next step without fur-
ther purification. Yield 4.50 g (84%); m.p. 193Ϫ194 °C (dec.). Ϫ
1H NMR (300 MHz, [D6]DMSO, 25 °C, TMS): δ ϭ 1.47 (s, 9 H,
tBuCH3), 3.88 (s, 3 H, OCH3), 7.09 [d, 3J ϭ 8.7 Hz, 1 H, ABH(5)],
Boc-(L)-Pro-AB(4-CH3)-OBn (2a): Eluent for the chromatographic
purification: hexane/ethyl acetate, 2:1. Yield 3.21 g (98%); oil;
[α]2D5 ϭ Ϫ55.1 (c ϭ 2, MeOH). Ϫ 1H NMR (300 MHz, [D6]DMSO,
100 °C, TMS): δ ϭ 1.38 (s, 9 H, tBuCH3), 1.76Ϫ2.02 [m, 3 H,
ProC(β)H ϩ ProC(γ)H2], 2.20 [m, 1 H, ProC(β)H], 2.27 (s, 3 H,
PhCH3), 3.39 [m, 2 H, ProC(δ)H2], 4.33 [dd, 3J(Hax,Hax) ϭ 8.5 Hz,
3J(Hax,Heq) ϭ 3.9 Hz, 1 H, ProC(α)H], 5.33 (s, 2 H, PhCH2), 7.37
3
4
[br. m, 6 H, PhH ϩ ABH(5)], 7.70 [dd, J ϭ 7.9 Hz, J ϭ 1.8 Hz,
1 H, ABH(6)], 8.06 [d, 4J ϭ 1.7 Hz, 1 H, ABH(2)], 9.17 (s, 1 H,
ABNH). Ϫ C25H30N2O5 (438.5): calcd. C 68.47, H 6.90, N 6.39;
found C 68.31, H 6.96, N 6.37.
Boc-(L)-Pro-AB(4-Cl)-OBn (2b): Eluent for the chromatographic
purification: hexane/ethyl acetate, 3:1. Yield 3.37 g (98%); oil;
[α]2D5 ϭ Ϫ66.7 (c ϭ 2, MeOH). Ϫ 1H NMR (300 MHz, [D6]DMSO,
100 °C, TMS): δ ϭ 1.38 (s, 9 H, tBuCH3), 1.85 [m, 2 H,
ProC(γ)H2], 2.02 [m, 1 H, ProC(β)H], 2.25 [m, 1 H, ProC(β)H],
3.40 [m, 2 H, ProC(δ)H2], 4.41 [dd, 3J(Hax,Hax) ϭ 8.4 Hz,
3J(Hax,Heq) ϭ 4.0 Hz, 1 H, ProC(α)H], 5.36 (s, 2 H, PhCH2), 7.38
3
7.67 [dd, J ϭ 8.5 Hz, 4J ϭ 2.1 Hz, 1 H, ABH(6)], 8.08 (br. s, 1 H,
NH), 8.35 [d, 4J ϭ 1.8 Hz, 1 H, ABH(2)], 12Ϫ13 (v br., 1 H,
3
COOH).
(br. m, 5 H, PhH), 7.61 [d, J ϭ 8.4 Hz, 1 H, ABH(5)], 7.76 [dd,
4
4
3J ϭ 8.4 Hz, J ϭ 2.0 Hz, 1 H, ABH(6)], 8.45 [d, J ϭ 2.0 Hz, 1
H, ABH(2)], 9.33 (s, 1 H, ABNH). Ϫ C24H27ClN2O5 (458.9): calcd.
C 62.81, H 5.93, N 6.10; found C 62.56, H 5.95, N 5.85.
Benzyl 3-[(tert-Butyloxycarbonyl)amino]-4-methoxybenzoate (3e): A
mixture of 4e (2.67 g, 10 mmol) and NaHCO3 (1.68 g, 20 mmol)
was suspended in DMF (25 mL). Benzyl bromide (2.37 mL,
20 mmol) was added and the reaction mixture was stirred for 48 h
at room temperature. After the addition of 10% aqueous Na2CO3
Boc-(L)-Pro-AB(4-CH2OCH3)-OBn (2c): Eluent for the chromato-
graphic purification: hexane/ethyl acetate, 2:1. Yield 3.29 g (94%);
320
Eur. J. Org. Chem. 2001, 311Ϫ322

Fine Tuning of the Cation Affinity of Artificial Receptors Based on Cyclic Peptides
oil; [α]2D5 ϭ Ϫ60.5 (c ϭ 2, MeOH). Ϫ 1H NMR (300 MHz,
FULL PAPER
3
PhCH3), 3.39 ϩ 3.63 [2 m, 2 ϫ 3 H, ProC(δ)H2], 4.71 [d, J(Hax,-
[D6]DMSO, 100 °C, TMS): δ ϭ 1.39 (s, 9 H, tBuCH3), 1.84Ϫ2.02 Hax) ϭ 9.1 Hz, 3 H, ProC(α)H], 7.27 [s, 6 H, ABH(5) ϩ ABH(6)],
[m, 3 H, ProC(β)H ϩ ProC(γ)H2], 2.22 [m, 1 H, ProC(β)H], 3.33 (s,
7.80 [s, 3 H, ABH(2)], 9.57 (s, 3 H, ABNH). Ϫ 13C NMR (75 MHz,
3 H, CH2OCH3), 3.42 [m, 2 H, ProC(δ)H2], 4.28 [dd, 3J(Hax,Hax) ϭ [D6]DMSO, 25 °C, TMS): δ ϭ 17.6 (PhCH3), 24.4 [ProC(γ)], 29.8
8.6 Hz, 3J(Hax,Heq) ϭ 3.9 Hz, 1 H, ProC(α)H], 4.48 ϩ 4.52 (2 d,
[ProC(β)], 49.3 [ProC(δ)], 59.9 [ProC(α)], 122.1 [ABC(6)], 124.4
2J ϭ 13.3 Hz, 2 H, CH2OCH3), 5.35 (s, 2 H, PhCH2), 7.37 [br. m, [ABC(2)], 129.4 [ABC(5)], 132.1 [ABC(4)], 134.6 [ABC(1)], 136.2
6 H, PhH ϩ ABH(5)], 7.76 [dd, 3J ϭ 7.9 Hz, 4J ϭ 1.7 Hz, 1 H,
[ABC(3)], 168.0 (ABCO), 170.5 (ProCO). Ϫ C39H42N6O6 · 2 H2O
ABH(6)], 8.34 [d, 4J ϭ 1.8 Hz, 1 H, ABH(2)], 9.25 (s, 1 H, ABNH). (726.8): calcd. C 64.45, H 6.38, N 11.56; found C 64.69, H 6.46, N
Ϫ C26H32N2O6 (468.5): calcd. C 66.65, H 6.88, N 5.98; found C 11.70. Ϫ CI-MS (NH3): m/z (%): 708 (100) [M ϩ NH4ϩ].
66.38, H 7.01, N 5.99.
cyclo-[(L)-Pro-AB(4-Cl)]3 (1b): The product eluted with 1,4-diox-
Boc-(L)-Pro-AB(4-OCH3)-OBn (2e): Eluent for the chromato-
ane/H2O, 1:1. If it was not obtained white after the RP-8 column
step, it was subjected to further column chromatography on silica
gel eluting with acetone. The product was finally triturated with
methanol. Yield 0.39 g (26%); m.p. Ͼ 250 °C; [α]2D5 ϭ ϩ43.1 (c ϭ
graphic purification: hexane/ethyl acetate, 1:1. Yield 3.28 g (96%);
oil; [α]2D5 ϭ Ϫ79.8 (c ϭ 2, MeOH). Ϫ 1H NMR (300 MHz,
[D6]DMSO, 100 °C, TMS): δ ϭ 1.37 (s, 9 H, tBuCH3), 1.86 [m, 2
H, ProC(γ)H2], 2.02 [m, 1 H, ProC(β)H], 2.17 [m, 1 H, ProC(β)H],
1
2, DMF). Ϫ H NMR (300 MHz, [D6]DMSO, 25 °C, TMS): δ ϭ
3
3.39 [m, 2 H, ProC(δ)H2], 3.91 (s, 3 H, OCH3), 4.39 [dd, J(Hax,-
1.85Ϫ1.98 [m, 9 H, ProC(β)H ϩ ProC(γ)H2], 2.26 [m, 3 H,
ProC(β)H], 3.40 [m, 3 H, ProC(δ)H], 3.61 [m, 3 H, ProC(δ)H], 4.84
Hax) ϭ 8.4 Hz, 3J(Hax,Heq) ϭ 4.0 Hz, 1 H, ProC(α)H], 5.33 (s, 2
3
3
3
H, PhCH2), 7.14 [d, J ϭ 8.7 Hz, 1 H, ABH(5)], 7.37 (br. m, 5 H,
[dd, J(Hax,Hax) ϭ 8.2 Hz, J(Hax,Heq) ϭ 2.2 Hz, 3 H, ProC(α)H],
3
4
7.39 [dd, 3J ϭ 8.3 Hz, 4J ϭ 1.9 Hz, 3 H, ABH(6)], 7.58 [d, 3J ϭ
8.2 Hz, 3 H, ABH(5)], 8.03 [d, J ϭ 2.0 Hz, 3 H, ABH(2)], 9.94 (s,
PhH), 7.75 [dd, J ϭ 8.6 Hz, J ϭ 2.2 Hz, 1 H, ABH(6)], 8.69 [d,
4J ϭ 2.1 Hz, 1 H, ABH(2)], 8.95 (s, 1 H, ABNH). Ϫ C25H30N2O6
(454.5): calcd. C 66.06, H 6.65, N 6.16; found C 65.81, H 6.59,
N 5.98.
4
3 H, ABNH). Ϫ 13C NMR (75 MHz, [D6]DMSO, 25 °C, TMS):
δ ϭ 24.4 [ProC(γ)], 29.7 [ProC(β)], 49.3 [ProC(δ)], 59.8 [ProC(α)],
123.2 [ABC(6)], 125.0 [ABC(2)], 126.4 [ABC(4)], 128.9 [ABC(5)],
134.9 [ABC(1)], 135.9 [ABC(3)], 166.9 (ABCO), 170.7 (ProCO). Ϫ
C36H33Cl3N6O6 (752.1): calcd. C 57.50, H 4.42, N 11.17; found C
57.58, H 4.63, N 11.02. Ϫ CI-MS (NH3): m/z (%): 770 (100) [M ϩ
NH4ϩ], 733 (20) [M Ϫ Cl ϩ NH4ϩ].
Boc-(L)-Pro-AB(4-COOCH3)-OBn (2f): Eluent for the chromato-
graphic purification: hexane/ethyl acetate, 2:1. Yield 3.14 g (87%);
oil; [α]2D5 ϭ Ϫ78.1 (c ϭ 2, MeOH). Ϫ 1H NMR (300 MHz,
[D6]DMSO, 100 °C, TMS): δ ϭ 1.34 (s, 9 H, tBuCH3), 1.87 [m, 2
H, ProC(γ)H2], 1.99 [m, 1 H, ProC(β)H], 2.25 [m, 1 H, ProC(β)H],
3
3.46 [m, 2 H, ProC(δ)H2], 3.90 (s, 3 H, OCH3), 4.24 [dd, J(Hax,-
cyclo-[(L)-Pro-AB(4-CH2OCH3)]3 (1c): The product eluted with
Hax) ϭ 8.7 Hz, 3J(Hax,Heq) ϭ 4.3 Hz, 1 H, ProC(α)H], 5.39 (s, 2
1,4-dioxane/H2O, 1:2. If it was not obtained white after the RP-8
column step, it was subjected to further column chromatography
on silica gel eluting with acetone. Yield 0.55 g (35%); m.p. 216 °C
(softening from 189 °C); [α]2D5 ϭ Ϫ198.6 (c ϭ 2, MeOH). Ϫ 1H
NMR (300 MHz, [D6]DMSO, 100 °C, TMS): δ ϭ 1.84Ϫ2.00 [m, 9
H, ProC(β)H ϩ ProC(γ)H2], 2.47 [m, 3 H, ProC(β)H], 3.32 (s, 9 H,
CH2OCH3), 3.55 [m, 6 H, ProC(δ)H2], 4.51 (m, 6 H, CH2OCH3),
4.75 [m, 3 H, ProC(α)H], 7.28 [dd, 3J ϭ 7.9 Hz, 3 H, ABH(6)], 7.38
[d, 3J ϭ 7.7 Hz, 3 H, ABH(5)], 7.88 [s, 3 H, ABH(2)], 9.18 (s, 3 H,
ABNH). Ϫ 13C NMR (75 MHz, [D6]DMSO, 100 °C, TMS): δ ϭ
24.3 [ProC(γ)], 29.6 [ProC(β)], 49.1 [ProC(δ)], 57.7 (CH2OCH3),
60.6 [ProC(α)], 70.6 (CH2OCH3), 122.3 [ABC(6)], 123.6 [ABC(2)],
128.0 [ABC(5)], 131.8 [ABC(4)], 136.0 [ABC(1)], 136.6 [ABC(3)],
168.5 (ABCO), 170.5 (ProCO). Ϫ C42H48N6O9 (780.9): calcd. C
64.60, H 6.20, N 10.76; found C 64.42, H 6.18, N 10.72. Ϫ CI-MS
(NH3): m/z (%): 798 (100) [M ϩ NH4ϩ], 768 (50) [M Ϫ OCH3 ϩ
H ϩ NH4ϩ], 738 (15) [M Ϫ 2 OCH3 ϩ 2 H ϩ NH4ϩ], 708 (2) [M
Ϫ 3 OCH3 ϩ 3 H ϩ NH4ϩ], 781 (3) [M ϩ Hϩ], 751 (5) [M Ϫ
OCH3 ϩ H ϩ Hϩ], 721 (4) [M Ϫ 2 OCH3 ϩ 2 H ϩ Hϩ].
4
H, PhCH2), 7.39 (br. m, 5 H, PhH), 7.74 [dd, 3J ϭ 8.3 Hz, J ϭ
1.7 Hz, 1 H, ABH(6)], 8.05 [d, 3J ϭ 8.3 Hz, 1 H, ABH(5)], 9.04 [d,
4J ϭ 1.7 Hz, 1 H, ABH(2)], 10.81 (s, 1 H, ABNH). Ϫ C26H30N2O7
(482.5): calcd. C 64.72, H 6.27, N 5.81; found C 64.57, H 6.28,
N 5.71.
Cyclopeptides: The synthesis of the linear hexapeptides was carried
out analogously to the previously described procedures starting
from the appropriate dipeptide.[13,14] The hexapeptide (2.00 mmol)
was then hydrogenated according to the general procedure. The
product was triturated with diethyl ether/hexane, 1:1, to afford a
white solid. It was deprotected at the terminal amino group accord-
ing to Method B. The product obtained was triturated with diethyl
ether. The resulting completely deprotected peptide was dissolved
in a mixture of DMF (100 mL/mmol) and DIEA (3.2 equiv.) and
the solution was heated to 80 °C. A solution of TBTU (1.1 equiv.)
in DMF (20 mL) was then slowly added dropwise. Stirring was
continued for 2 h at 80 °C and then the solvent was evaporated in
vacuo. The product was isolated from the mixture by chromato-
graphic workup. An initial purification step was carried out using a
silica gel column (CH2Cl2/MeOH, 15:1) (in the case of 1e, CH2Cl2/
MeOH, 5:1, was used). The material recovered was further purified
on an RP-8 column. For this, it was dissolved in a small amount
of DMF (in the case of 1f, DMSO was used) and applied to a
column conditioned with 1,4-dioxane/H2O, 1:10. The eluent com-
position was gradually changed until the pure product eluted.
cyclo-[(L)-Pro-AB(4-OCH3)]3 (1e): The product eluted with 1,4-di-
oxane/H2O, 1:2.5. It was finally recrystallized from methanol. Yield
0.61 g (41%); m.p. Ͼ 250 °C (dec.) (softening from 245 °C); [α]2D5
ϭ
1
ϩ31.2 (c ϭ 2, DMF). Ϫ H NMR (300 MHz, [D6]DMSO, 25 °C,
TMS): δ ϭ 1.75Ϫ1.98 [m, 9 H, ProC(β)H ϩ ProC(γ)H2], 2.27 [m,
3 H, ProC(β)H], 3.48 [m, 6 H, ProC(δ)H2], 3.90 (s, 9 H, OCH3),
4.87 [dd, 3J(Hax,Hax) ϭ 8.4 Hz, 3J(Hax,Heq) ϭ 2.1 Hz, 3 H,
ProC(α)H], 7.06 [d, 3J ϭ 8.7 Hz, 3 H, ABH(5)], 7.21 [dd, 3J ϭ
8.4 Hz, 4J ϭ 2.0 Hz, 3 H, ABH(6)], 8.29 [d, 4J ϭ 1.9 Hz, 3 H,
ABH(2)], 9.53 (s, 3 H, ABNH). Ϫ 13C NMR (75 MHz, [D6]DMSO,
25 °C, TMS): δ ϭ 24.5 [ProC(γ)], 29.7 [ProC(β)], 49.4 [ProC(δ)],
cyclo-[(L)-Pro-AB(4-CH3)]3 (1a): The product eluted with 1,4-diox-
ane/H2O, 1:2.5. If it was not obtained white after the RP-8 column
step, it was subjected to further column chromatography on silica
gel eluting with acetone. Yield 0.58 g (42%); m.p. Ͼ 250 °C (dec.)
(softening from 245 °C); [α]2D5 ϭ ϩ26.3 (c ϭ 2, DMF). Ϫ 1H NMR 55.7 (PhOCH3), 59.4 [ProC(α)], 110.2 [ABC(5)], 121.4 [ABC(2)],
(300 MHz, [D6]DMSO, 25 °C, TMS): δ ϭ 1.84Ϫ2.00 [m, 9 H, 121.9 [ABC(6)], 127.0 [ABC(1)], 129.5 [ABC(3)], 150.0 [ABC(4)],
ProC(β)H ϩ ProC(γ)H2], 2.56 [m, 3 H, ProC(β)H], 2.30 (s, 9 H,
168.1 (ABCO), 170.7 (ProCO). Ϫ C39H42N6O9·H2O (756.8): calcd.
Eur. J. Org. Chem. 2001, 311Ϫ322
321

S. Kubik, R. Goddard
FULL PAPER
[1]
ˇ
C. M. Dobson, A. Sali, M. Karplus, Angew. Chem. 1998, 110,
908Ϫ935; Angew. Chem. Int. Ed. 1998, 37, 868Ϫ893.
J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, 1995.
F. Vögtle, Supramolekulare Chemie, B. G. Teubner, Stuttgart,
1992.
C. D. Gutsche, Calixarenes, RSC, London, 1998.
C. D. Gutsche, in: Inclusion Compounds, vol. 4 (Eds.: J. L. At-
wood, J. E. D. Davies, D. D. MacNicol), Oxford University
Press, Oxford, 1991, p. 27Ϫ63.
C 61.89, H 5.86, N 11.10; found C 62.18, H 5.98, N 11.40. Ϫ CI-
MS (NH3): m/z (%): 756 (100) [M ϩ NH4ϩ], 739 (5) [M ϩ Hϩ].
[2]
[3]
cyclo-[(L)-Pro-AB(4-COOCH3)]3 (1f): The product eluted with 1,4-
dioxane/H2O, 1:1. It was finally recrystallized from acetone. Yield
[4]
[5]
0.43 g (26%); m.p. Ͼ 250 °C (dec.) (softening from 234 °C); [α]2D5
ϭ
1
Ϫ139.5 (c ϭ 2, CHCl3); [α]2D5 ϭ ϩ19.9 (c ϭ 2, DMF). Ϫ H NMR
(300 MHz, [D6]DMSO, 25 °C, TMS):
δ ϭ 1.92 [m, 6 H,
[6]
[7]
ProC(γ)H2], 2.08 ϩ 2.35 [2 m, 2 ϫ 3 H, ProC(β)H2], 2.09 (s, 6 H,
acetone), 3.44 [m, 6 H, ProC(δ)H2], 3.92 (s, 9 H, OCH3), 4.76 [dd,
3J(Hax,Hax) ϭ 8.7 Hz, 3J(Hax,Heq) ϭ 3.3 Hz, 3 H, ProC(α)H], 7.33
S. Shinkai, Tetrahedron 1993, 49, 8933Ϫ8968.
V. Böhmer, Angew. Chem. 1995, 107, 785Ϫ818; Angew. Chem.
Int. Ed. Engl. 1995, 34, 713Ϫ745.
[8]
[9]
D. M. Rudkevich, G. Hilmersson, J. Rebek, Jr., J. Am. Chem.
3
[dd, J ϭ 8.1 Hz, 4J ϭ 1.5 Hz, 3 H, ABH(6)], 8.00 [d, 3J ϭ 8.1 Hz,
Soc. 1998, 120, 12216Ϫ12225.
4
3 H, ABH(5)], 8.54 [d, J ϭ 1.4 Hz, 3 H, ABH(2)], 10.71 (s, 3 H,
D. M. Rudkevich, J. Rebek, Jr., Eur. J. Org. Chem. 1999,
1991Ϫ2005.
T. Haino, D. M. Rudkevich, J. Rebek, Jr., J. Am. Chem. Soc.
ABNH). Ϫ 13C NMR (75 MHz, [D6]DMSO, 25 °C, TMS): δ ϭ
24.4 [ProC(γ)], 29.1 [ProC(β)], 30.6 (acetone-CH3), 49.0 [ProC(δ)],
52.6 (OCH3), 60.4 [ProC(α)], 118.5 [ABC(4)], 120.4 [ABC(2)], 120.5
[ABC(6)], 130.7 [ABC(5)], 139.3 [ABC(1)], 142.0 [ABC(3)], 166.9
[10]
1999, 121, 11253Ϫ11254.
[11]
[12]
S. Kubik, R. Goddard, Chem. Commun. 2000, 633Ϫ634.
H. Ishida, M. Suga, K. Donowaki, K. Ohkubo, J. Org. Chem.
1995, 60, 5374Ϫ5375.
S. Kubik, J. Am. Chem. Soc. 1999, 121, 5846Ϫ5855.
S. Kubik, R. Goddard, J. Org. Chem. 1999, 64, 9475Ϫ9486.
J. C. Ma, D. A. Dougherty, Chem. Rev. 1997, 97, 1303Ϫ1324.
ϩ
167.3 (ABCO), 169.9 (ProCO), 206.4 (acetone-CO). Ϫ
C42H42N6O12 · C3H6O · 0.5 H2O (889.9): calcd. C 60.74, H 5.55, N
9.44; found C 60.72, H 5.53, N 9.18. Ϫ CI-MS (NH3): m/z (%):
841 (100) [M ϩ H ϩ NH4ϩ], 808 (5) [M Ϫ OCH3 ϩ NH2 ϩ Hϩ].
[13]
[14]
[15]
[16]
[17]
´
P. Lhotak, S. Shinkai, J. Phys. Org. Chem. 1997, 10, 273Ϫ285.
Y. Hamuro, S. J. Geib, A. D. Hamilton, J. Am. Chem. Soc.
X-ray Crystallographic Structure Determination of 1e · 7 MeOH:
Crystal data: C39H42N6O9 · 7 (CH4O), Mw ϭ 963.1, colorless plate
0.40 ϫ 0.22 ϫ 0.08 mm, orthorhombic P212121 (No. 19), a ϭ
1996, 118, 7529Ϫ7541.
[18]
[19]
[20]
[21]
K. A. Connors, Binding Constants, Wiley, New York, 1987.
R. S. Macomber, J. Chem. Educ. 1992, 69, 375Ϫ378.
V. M. S. Gil, N. C. Oliveira, J. Chem. Educ. 1990, 67, 473Ϫ478.
3
˚
˚
13.9313(8), b ϭ 16.0836(8), c ϭ 21.118(1) A, V ϭ 4731.8(4) A ,
T ϭ 100 K, DX ϭ 1.352 g cmϪ3, λ ϭ 0.71073 A, µ ϭ 1.02 cm
,
Ϫ1
˚
U. Olsher, M. G. Hankins, Y. D. Kim, R. A. Bartsch, J. Am.
Nonius Kappa CCD diffractometer, 2.16° Ͻ θ Ͻ 30.97°, 22862
measured reflections, 13045 independent, 6052 with I Ͼ 2σ(I). The
structure was solved by direct methods (SHELXS-97) and refined
by least-squares (SHELXL-97) using Chebyshev weights on F2o to
R1 ϭ 0.089 [I Ͼ 2σ(I)], wR2 ϭ 0.267, 546 parameters, H atoms on
Chem. Soc. 1993, 115, 3370Ϫ3371.
[22]
[23]
Y. Marcus, Ion Properties, Marcel Dekker, New York, 1997.
S. Bartoli, S. Roelens, J. Am. Chem. Soc. 1999, 121,
11908Ϫ11909.
[24]
G. G. Talanova, N. S. A. Elkarim, V. S. Talanov, R. E. Hanes,
Jr., H.-S. Hwang, R. A. Bartsch, R. D. Rogers, J. Am. Chem.
Soc. 1999, 121, 11281Ϫ11290.
peptide constrained, H atoms on methanol solvate molecules not
Ϫ3
˚
˚
located, S ϭ 0.965, residual electron density 0.87 e A (1.65 A
from C42). Crystallographic data (excluding structure factors) for
the structure reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary pub-
lication no. CCDC-144565. Copies of the data can be obtained free
of charge on application to the CCDC, 12 Union Road, Cambridge
CB2 1EZ, U.K. [Fax: (internat.) ϩ 44-1223/336-033; E-mail:
deposit@ccdc.cam.ac.uk].
[25]
[26]
[27]
J. S. Nowick, J. H. Tsai, Q.-C. D. Bui, S. Maitra, J. Am. Chem.
Soc. 1999, 121, 8409Ϫ8410.
B. Gong, Y. Yan, H. Zeng, E. Skrzypczak-Jankunn, Y. W. Kim,
J. Zhu, H. Ickes, J. Am. Chem. Soc. 1999, 121, 5607Ϫ5608.
H. Zeng, R. S. Miller, R. A. Flowers II, B. Gong, J. Am. Chem.
Soc. 2000, 122, 2635Ϫ2644.
[28]
[29]
´
J. Neel, Pure Appl. Chem. 1972, 31, 201Ϫ225.
S. Mecozzi, A. P. West Jr., D. A. Dougherty, Proc. Natl. Acad.
Sci. USA 1996, 93, 10566Ϫ10571.
[30]
[31]
[32]
[33]
M. Kamieth, F.-G. Klärner, F. Diederich, Angew. Chem. 1998,
110, 3497Ϫ3500; Angew. Chem. Int. Ed. 1998, 37, 3303Ϫ3306.
S. Roelens, R. Torriti, J. Am. Chem. Soc. 1998, 120,
Acknowledgments
S. K. thanks Prof. G. Wulff for his generous support, Mrs. D. Ku-
bik for her invaluable assistance in performing the syntheses, and
Mr. V. Schmid for his help with the AM1 calculations. Thanks
are also due to the Deutsche Forschungsgemeinschaft for funding
this research.
12443Ϫ12452.
S. L. Mayo, B. D. Olafson, W. A. Goddard III, J. Phys. Chem.
1990, 94, 8897Ϫ8909.
J. Coste, E. Frerot, P. Jouin, J. Org. Chem. 1994, 59,
´
2437Ϫ2446.
Received August 16, 2000
[O00425]
322
Eur. J. Org. Chem. 2001, 311Ϫ322